SARJ Vol 23, No 3 (2017)

South African

Respiratory Journal VOLUME 23

|

NUMBER 3

|

SEPTEMBER 2017

OFFICIAL JOURNAL OF THE S.A. THORACIC SOCIETY ISSN 2304-0017

FOXAIR

everyone and it’s yours to give

Air is for

Why wait to

prescribe?

S4 FOXAIR® 50/250 and 50/500 ACCUHALER® - 42/21.5.4/0582; 0583. Each blister contains a mixture of salmeterol xinafoate equivalent to 50 µg of salmeterol and microfine fluticasone propionate (250 µg or 500 µg). Applicant: GlaxoSmithKline South Africa (Pty) Ltd. (Co. Reg. No. 1948/030135/07). 39 Hawkins Avenue, Epping Industria 1, Cape Town, 7460. For full prescribing information plese refer to the package insert approved by the Medicines Regulatory Authority. All adverse events should be reported by calling the Aspen Medical Hotline number or directly to GlaxoSmithKline on +27117456000. FO/0713/933 A16772 08/13

THE SOUTH AFRICAN

RESPIRATORY JOURNAL VOLUME 23 | NUMBER 3 | SEPTEMBER 2017

CONTENTS EDITORIAL 46

Still dancing to the tune of the invisible piper: Failures in the PMTCT programme are costly R Masekela

ORIGINAL RESEARCH

49

Outcomes of HIV-1-positive children with pneumonia admitted to the paediatric intensive care unit: A retrospective review P M Jeena, L Githinji Pulmonary nodules in patients with drug-resistant tuberculosis who have undergone adjuvant lung resection in a high-HIV-burden setting K Maharaj, R Perumal, G Alexander

CASE REPORT

61

Cystic lung disease in a young female patient M S Abdool-Gaffar

55

REVIEW 63

An approach to pulmonary haemorrhage in children D Parris, A van Niekerk, A C Jeevarathnum, R J Green

SARJ EDITOR-IN-CHIEF Prof. K Dheda DEPUTY EDITOR Prof. C Koegelenberg SECTION EDITOR Breath-taking News: Prof. E Irusen EDITORIAL BOARD Prof. G Ainslie, Prof. E Bateman, Prof. R Green, Prof. E Irusen, Prof. M Jeebhay, Prof. P Jeena, Prof. U Lalloo, Prof. A Linegar, Prof. R Masekela, Dr K Nyamande, Dr J O’Brien, Dr R Raine, Prof. G Richards, Dr R van Zyl Smit, Prof. M Wong, Prof. H Zar INTERNATIONAL EDITORIAL BOARD Prof. A Cattamanchi - USA Prof. F Chung - UK Prof. G B Migliori - Italy Prof. S Sharma - India Prof. W W Yew - China PRESIDENT SA THORACIC SOCIETY Prof. U Lalloo

HMPG

CEO AND PUBLISHER Hannah Kikaya Email: hannahk@hmpg.co.za EXECUTIVE EDITOR Bridget Farham MANAGING EDITORS Naadia van der Bergh, Claudia Naidu TECHNICAL EDITORS Kirsten Morreira, Christelle Cronje

71

BREATH-TAKING NEWS

72

PRODUCT NEWS

73 ABSTRACTS

PRODUCTION MANAGER Emma Jane Couzens DTP AND DESIGN Clinton Griffin, Travis Arendse CHIEF OPERATING OFFICER Diane Smith | Tel. 012 481 2069 Email: dianes@hmpg.co.za JOURNAL ADVERTISING Charles Duke, Reneé Hinze, Ladine van Heerden ONLINE SUPPORT Gertrude Fani | Tel. 021 532 1281 Email: publishing@hmpg.co.za

The Editor The South African Respiratory Journal PO Box 13725, Mowbray, 7705 Tel. 021 650 3050 | Fax: 021 650 2610 | Email: sarj@iafrica.com The views expressed in individual articles and advertising material are the personal views of the authors and are not necessarily shared by the editors, the advertisers or the publishers. No articles may be reproduced without the written consent of the publishers.

The SARJ is published by the Health and Medical Publishing Group (Pty) Ltd. Co. registration 2004/0220 32/07, a subsidiary of SAMA. HEAD OFFICE: Block F, Castle Walk Corporate Park, Nossob Street, Erasmuskloof Ext. 3, Pretoria, 0181 EDITORIAL OFFICE: Suite 11, Lonsdale Building, Lonsdale Way, Pinelands, 7405 | Tel. 021 532 1281 All letters and articles for publication must be submitted online at www.sarj.org.za Email: publishing@hmpg.co.za

FINANCE Tshepiso Mokoena HMPG BOARD OF DIRECTORS Prof. M Lukhele (Chair), Dr M R Abbas, Mrs H Kikaya, Dr M Mbokota, Dr G Wolvaardt PRINTED BY TANDYM PRINT

EDITORIAL

Still dancing to the tune of the invisible piper: Failures in the PMTCT programme are costly In the current economic climate, with cost-containment being key, the World Health Organization (WHO) recommends that resource allocation be guided by the three principles: efficiency (to maximise population health), fairness (to minimise the health differences) and utility (to provide the greatest good for the greatest number).[1] It is well-known that the paediatric intensive care unit (PICU) environment is highly costly and ethically challenging, particularly in the resource-constrained environment.

The original paper by Jeena PM et al.[2] addresses this very important area in the management of critically ill children with lower respiratory tract infections, who are admitted to PICU. They explored the impact of antiretroviral treatment (ART) and cytomegalovirus (CMV) infection on children requiring respiratory support. Interestingly, 19% of children were HIV-positive and of these, half had CMV co-infection.

CMV infection was associated with a higher need for high-frequency oscillation, which in their study was used as rescue therapy for those with more severe acute respiratory distress syndrome. The mean duration prior to commencement of ganciclovir was 4 days after confirmation of a positive CMV viral load. This duration, however, did not impact the survival. The differences in survival rates between the groups (HIVpositive and -negative and -exposed), were not statistically significant.

This could perhaps partly be explained by the fact that 42% of HIVpositive children were already on ART prior to admission, although the viral loads reported were still high. They also elegantly showed that a combination of high frequency oscillatory ventilation (HFOV), ganciclovir and ART improved survival rates to 90%. Although these are excellent results and a far cry from the dismal survival rates in the early 2000s, HFOV availability is still limited to tertiary centres in South Africa.[3,4]

46 SARJ VOL. 23 NO. 3 2017

Therefore, as the authors propose, we should possibly be more concerned that almost 60% of children in this study had no prevention of mother-to-child transmission (PMTCT) measures in place. Preventive measures would logically be more cost-effective and prevent long-term morbidity, which may be associated with being a PICU graduate. As critical care specialists, we need to stop dancing to the tune of the invisible piper, and assume our role in advocating for PMTCT. R Masekela Head of Paediatrics and Child Health, Nelson R Mandela School of Medicine, University of KwaZulu-Natal, Durban, South Africa masekelar@ukzn.ac.za

1. Jeena PM, McNally LM, Stobie M, Coovaia HM, Ahikari MA, Petros AJ. Challenges in the provision of ICY services to HIV infected children in resource poor settings: A South African case study. J Med Ethics 2005;31(4):226-230. https://doi.org/10.1136/ jme.2003.004010 2. Jeena PM, Githinji L. Outcomes of HIV-infected children with pneumonia admitted to the PICU: A retrospective study. S Afr Respir J 2017;23(3):49-53. https://doi.org/10 3. Argent AC. Managing HIV in the PICU – the experience at the Red Cross War Memorial Children’s Hospital in Cape Town. Indian J Pediatr 2008;75(6):615-620. https://doi.org/10.1007/s12098-008-0118-2 4. Kitchin OP, Masekela R, Becker P, Moodley T, Risenga SM, Green RJ. Outcome of human immunodeficiency virus exposed and infected children admitted to a pediatric intensive care unit for respiratory failure. Pediatr Crit Care Med 2012;3(5):516-520. https://doi.org/10.1097/pcc.0b013e31824ea143

S Afr Respir J 2017;23(3):46. DOI:10.7196/SARJ.2017.v23i3.177

ORIGINAL RESEARCH

Outcomes of HIV-1-positive children with pneumonia admitted to the paediatric intensive care unit: A retrospective review P M Jeena, MB ChB, FCP (Paeds), Cert Paed Pulm, PhD; L Githinji, MD, Dip Allergol, Cert Paed Pulm Department of Paediatric and Child Health, University of KwaZulu-Natal, Inkosi Albert Luthuli Central Hospital, Durban, South Africa Corresponding author: P M Jeena (jeena@ukzn.ac.za)

Introduction. The outcomes of HIV-positive children with pneumonia, who require intensive care, have been poor. Advances in intensive care support for end-organ dysfunction, treatment of the primary disease and opportunistic infections could improve this outcome.

Objectives. To evaluate the incidence and outcome according to HIV status.

Methods. A retrospective chart review of the electronic dataset of children with pneumonia admitted to the paediatric intensive care unit (PICU) at Inkosi Albert Luthuli Central Hospital, Durban, South Africa was undertaken. Data on the use of ganciclovir, combination antiretroviral therapy (cART) and high frequency oscillatory ventilation (HFOV) on outcome were also evaluated.

Results. Of the 405 children with pneumonia admitted to the PICU during 2010, 77 (19%) were HIV-1-positive, and 261 (64.4%) were HIV-negative. The mortality rate among the two groups was similar (22.1% v. 15.3% (p=0.27), respectively). Among the HIV-positive cases, cytomegalovirus (CMV) was isolated in 39 (50.7%) cases, of which 18 (46.2%) required HFOV. Among the children who were treated with ganciclovir and cART, the survival rate was 90%. In HIV-positive children with CMV-associated pneumonia who received cART with ganciclovir therapy, the survival rate was 92.3%.

Conclusion. HIV-1-positive children with pneumonia requiring intensive care had a similar outcome to HIV-negative children with pneumonia. HIV-1-positive children with CMV-associated pneumonia on ganciclovir, cART, and HFOV have improved outcomes in comparison to previous studies.

S Afr Respir J 2017;23(3):49-53. DOI:10.7196/SARJ.2017.v23i3.166

In South Africa (SA), the roll-out of combined antiretroviral therapy (cART) to all HIV-infected children under the age of 5, regardless of CD4 counts or HIV viral loads, has resulted in HIV-positive children being treated as chronic disease patients.[1] This change has resulted in a new outlook for HIV-positive children with acute hypoxaemic pneumonia requiring mechanical ventilation, who have had dismal outcomes over the last three decades.[2,3] Paediatric intensive care unit (PICU) survival of these cases could now result in long-term survival, as HIV-1-positive children receiving cART have similar long-term outcomes to HIV-negative children.[4] The rational allocation of scarce PICU resources for HIV-1-positive children in terms of short-term outcomes remains controversial, and has created the ethical dilemma of whether or not these patients should be admitted.[5,6] Advanced strategies have been developed to treat both end-organ dysfunction and opportunistic infections; this warrants an updated review of the short-term outcomes of HIV-positive children in PICU. Adequate organ support during the acute phase of the illness appears to be the primary strategy to improve PICU survival among these cases. Ventilation strategies are important in supporting lung function in HIV-positive children with very severe pneumonia and acute respiratory distress syndrome (ARDS).[7] The early institution of high frequency oscillatory ventilation (HFOV) has been shown to improve the outcome among adults with pneumonia and ARDS.[8] The early recognition and treatment of opportunistic infections such as Pneumocystis jirovecii and cytomegalovirus (CMV) has also been advocated as a means to improve outcome.[9-11] Although intravenous

ganciclovir therapy has been advocated in many centres, its impact has not been adequately investigated.[12] The impact of cART during the early stages of recovery from acute disease within the complex milieu of the PICU, where drug-drug interactions and adverse drug reactions frequently occur, requires further evaluation. HIV viral load reduction and CD4 improvement on cART usually takes weeks to months to occur. Neglecting or delaying cART during the immediate recovery stage has been associated with notably poorer outcomes.[13] In this retrospective chart review we explored the impact of mechanical ventilation practices, such as intermittent positive pressure ventilation (IPPV) and HFOV, combined with ganciclovir and early cART, on the short-term outcome of HIV-positive children with CMV-associated severe pneumonia who had been admitted to the PICU.

Methods

A review of the electronic database of all cases admitted to the PICU at Inkosi Albert Luthuli Central Hospital (IALCH), a tertiary level hospital in Durban, KwaZulu-Natal Province, SA during 2010, was undertaken. Cases were categorised according to their HIV and CMV status (exposure, infection and disease). Among HIVpositive children with pneumonia, where CMV was isolated, the use of ganciclovir, antiretroviral treatment, and the need for the use of HFOV were evaluated. The use of nevirapine for the prevention of mother-to-child transmission was also recorded.

SARJ VOL. 23 NO. 3 2017

49

ORIGINAL RESEARCH Table 1. Baseline characteristics of children admitted to the PICU according to their HIV status (N=405) Characteristics

HIV-positive (n=77)

HIV-exposed negative (n=48)

HIV-negative (n=261)

HIV status unknown (n=19)

Total (N=405)

Males, n (%)

39 (50.6)

24 (50)

134 (51.3)

11 (57.9)

208 (51.4)

Age (months), median (IQR)

3.8 (1 - 14)

4.4 (2 - 10)

8.9 (1 - 24)

6.7 (1 - 32)

4.9 (1 - 32)

Age <12 months, n (%)

60 (77.9)

37 (77.1)

160 (61.3)

12 (63.2)

269 (66.4)

9 (18.8)

61 (23.4)

7 (36.8)

96 (23.7)

Malnourished, >3 SD from the mean, n (%) 19 (24.7) Diagnosis, n (%) LRTI

57 (74)

28 (58)

167 (64)

7 (37)

259 (64)

Sepsis

16 (21)

12 (25)

53 (20)

12 (63)

93 (23)

Gram-negative

34 (44)

21 (44)

132 (51)

5 (26)

192 (47)

Gram-positive

11 (14)

6 (14)

36 (14)

0

13 (3)

All viruses

41 (53)

3 (6)

40 (15)

1 (5)

85 (21)

CMV

39 (50.6)

0

20 (7.7)

1 (5)

60 (15)

All fungi

30 (39)

5 (10)

34 (13)

0

69 (17)

PCP

9 (11.7)

0

0

0

9 (2)

M. tuberculosis

5 (6)

1 (2)

2 (1)

0

8 (2)

Pathogens, n (%)

PICU = paediatric intensive care unit; LRTI = lower respiratory tract infection; CMV = cytomegalovirus; PCP = Pneumocystis jirovecii pneumonia; M. tuberculosis = Mycobacterium tuberculosis.

Standard management protocol The standard management of cases admitted to the PICU included a full blood count, urea and electrolyte measurements, liver function tests and C-reactive protein assays. The HIV status was determined with informed consent by an HIV DNA PCR (COBAS AmpliPrep/ COBAS TaqMan HIV-1 Qualitative Test (Roche Molecular Systems, Inc., USA). Endotracheal aspirates were sent for bacteriological and respiratory viral screens. Blood CMV DNA PCR (Roche Diagnostics GmbH, Germany) was performed in children who were HIV-exposed, or suspected of being HIV-positive. The standard antimicrobial policy for all cases included the use of co-amoxiclav and gentamicin as first-line therapy for community-acquired infections, and empirical changes to piperacillin/tazobactam/ amikacin in non-responsive cases, or those with hospital-acquired infections. Culture- and sensitivity-driven antibiotic changes were made once microbiological isolates was established. All infants who were clinically suspected of being HIV-positive or HIV-exposed, were treated empirically with trimethoprim-sulphamethoxazole (120 mg/kg/day). Fluconazole was added for clinical cases of candidiasis secondary to prolonged antibiotic treatment (>7 days), with fungal growth, or a positive blood (1,3)-β-D-glucan test. For the first 21 days, ganciclovir was administered intravenously at a 12-hourly dose of 5 mg/kg. For the next 21 days, it was administered daily at 5 mg/kg/day to patients who were CMV DNA-positive by PCR and had interstitial pneumonia confirmed by chest radiography.

Caregivers and/or parents of HIV-positive children were counselled and cART was commenced as soon as patients were stable, usually during the second week of illness. Patients with a partial pressure arterial oxygen (PaO 2)/FiO 2 ratio of <200, or an oxygen index of >25, on a peak inspiratory pressure of

50 SARJ VOL. 23 NO. 3 2017

>30 cmH2O and a fraction of inspired oxygen (FiO2) of >0.6, were placed on HFOV. Outcomes were documented upon discharge from the PICU or failure of treatment. Consent for this review was obtained from the Biomedical Ethics Research Committee at the University of KwaZulu-Natal (ref. no. BE151/11).

Results

Of the 405 patients admitted to the PICU at IALCH in 2010, 82 (20.2%) died. The baseline characteristics of admissions to the PICU in 2010 are shown in Table 1: 77 (19.0%) were HIV-positive; 48 (11.9%) were HIV-exposed; 261 (64.4%) were HIV-negative, and in 19 (4.7%) the HIV status was unknown. Among the HIV-positive children, 34 (44.2%) had a detectable HIV viral load of 2 335 963 copies (760 464 - >10 million). Two patients had an undetectable viral load owing to early commencement of cART. CD4% was performed in 24 (31.2%) patients with a median (interquartile range) value of 34.9% (21.3 - 44.5) recorded.

The outcome of children with pneumonia, who were HIV-positive and required mechanical ventilation, was similar to children who were HIV-exposed, but uninfected, and slightly higher than HIVnegative children; however, this finding was not statistically significant (p=0.27) (Table 2).

Impact of ganciclovir therapy in CMV-positive patients admitted to the PICU Of the 77 HIV-positive children, 50.6% were CMV DNA-positive

by PCR. Ganciclovir treatment was commenced in 79.5% of CMVpositive patients, of whom 83.9% survived to PICU discharge (p=0.56) (Fig. 1). Ganciclovir was commenced at a mean (IQR) of 4 (3 - 7) days after PICU admission.

ORIGINAL RESEARCH Table 2. Outcomes of patients admitted to the PICU at IALCH in 2010 according to HIV status (N=405)

HIV-infected n=77

Survivors, Deaths, n (%) n (%) Total admissions

405

323

82 (20.2)

HIV status HIV PCR +

77 (19.0)

60 (77.9)

17 (22.1)*

HIV-exposed, PCR –

48 (11.9)

34 (70.8)

10 (20.8)

HIV-negative, PCR –

261 (64.4)

216 (82.8)

40 (15.3)*

Unknown

19 (4.7)

10 (52.6)

9 (47.3)

CMV PCR – n=19 (24.7)

Ganciclovir n=31 (79.5)

PICU = paediatric intensive care unit; IALCH = Inkosi Albert Luthuli Central Hospital; PCR = polymerase chain reaction. *p=0.27.

Table 3. Survival of HIV-positive infants with pneumonia who received cART (n=33) according to CMV status CMV status

Received cART, n (%)

Survival, n (%)

CMV + PCR +

15 (45.)

13 (86.7)

CMV – PCR –

12 (36.4)

8 (66.7)

Not tested

6 (18.2)

5 (83.3)

The impact of the use of cART alone, and in combination with ganciclovir, among HIV-positive children with CMV-associated pneumonia Of the 77 HIV-positive children, 42.9% were on cART. Patients receiving cART had a 79% survival rate upon discharge. Fifteen cART patients were CMV PCR-positive, with a survival rate of 86.7% (Table 3). The 13 HIV-positive children with CMV disease who were on cART and ganciclovir therapy had a survival rate of 92.3% on discharge from the PICU. The 14 HIV-positive children with CMVassociated pneumonia treated with ganciclovir alone had a survival rate of 78.6% (Table 4).

Usefulness of HFOV in the management of HIV-positive children with CMV-associated pneumonia and ARDS Among the 77 HIV-positive children with pneumonia and ARDS, 24 children received HFOV following an inadequate response to conventional IPPV, while 53 patients received IPPV alone. The isolation of CMV was significantly higher among children requiring HFOV than those that did not (75% v. 39.6%; p=0.04), but the outcomes for the two groups were similar (Table 5). Eighteen HIV-1-positive children with CMV-associated pneumonia required HFOV; 94.4% of these were on ganciclovir and 55.5% were on cART and ganciclovir.

Patients receiving a combination of HFOV, ganciclovir and cART had a survival rate of 90% (Table 5).

Discussion

The main finding of this retrospective review was the 90% survival among critically ill HIV-positive children with CMV-associated severe pneumonia, who developed ARDS and who were placed on cART, ganciclovir and HFOV. This intervention improved the short-

CMV not tested n=19 (24.7)

CMV PCR + n=39 (50.7)

No ganciclovir n=8 (20.5)

Survivors n=6 (75)

Survivors

n=26 (83.9)

Fig. 1. Impact of ganciclovir in HIV- and CMV-infected children, n (%).

Table 4. Impact of ganciclovir and cART in HIV-positive children with CMV-associated pneumonia (N=39) Characteristics

cART+ (n=15)

PMTCT+* (n=4)

cART– (n=20)

Ganciclovir

13

3

14

Survival, n (%)

12 (92.3)

2 (66.7)

11 (78.6)

No ganciclovir Survival, n (%)

2

1

6

1 (50)

1 (100)

5 (83.3)

†

†

CMV = cytomegalovirus; cART = combined antiretroviral therapy; PMTCT = prevention of mother-to-child transmission. *PMTCT of HIV, which included the administration of nevirapine to all HIV-exposed infants for 6 weeks. Nevirapine was also administered for the duration of breastfeeding and 1 week beyond cessation of breastfeeding. † p=0.28.

Table 5. Impact of combination of the mode of ventilation, use of ganciclovir and cART on survival among HIV-positive children with CMV-associated pneumonia (N=77) Treatment

n (%)

CMV+, n (%)

Survival, n (%)

IPPV only

53 (68.8)

21 (39.6)

43 (81.1)

IPPV, HFOV

24 (31.2)

18 (75)

14 (77.8)

HFOV, ganciclovir

17 (22.1)

17 (100)

14 (82.3)

HFOV, ganciclovir, cART

10 (13.0)

10 (100)

9 (90)

cART = combined antiretroviral therapy; CMV = cytomegalovirus; IPPV = intermittent positive pressure ventilation; HFOV = high frequency oscillatory ventilation.

term outcomes in the patients, in comparison with similar historical controls with survival rates of 0 - 12%. The better outcomes may be attributed to better organ support, rapid responses to the diagnoses and management of primary and opportunistic infections.[14,15] In HIV-positive children with CMV-associated pneumonia who developed ARDS and required HFOV, ganciclovir treatment and cART resulted in excellent outcomes. The overall outcome of all

SARJ VOL. 23 NO. 3 2017

51

ORIGINAL RESEARCH HIV-positive children with CMV-associated pneumonia was also favourable on ganciclovir treatment alone, or with cART (83.9% and 92.3%, respectively). This improvement contributed to the overall reduced all-cause mortality rates among HIV-positive children with pneumonia admitted to the PICU, and provides support for the continued access of these patients to scarce and expensive resources in developing countries. These findings were dissimilar to studies in the pre-cART and pre-ganciclovir era in 2004/5 by Goussard et al.,[16] from the PICU at Tygerberg Academic Hospital (TBH), Cape Town, SA, and by Zampoli et al.,[17] from paediatric wards at the Red Cross War Memorial Childrenâ&#x20AC;&#x2122;s Hospital (RCWMCH), Cape Town, SA. At TBH, the mortality rate of CMV-associated pneumonia in HIV-positive children was 72%, while at RCWMCH, it was 55%.[16,17] A second major finding of this retrospective review was the significant burden of disease that CMV has among HIV-positive children admitted to the PICU with pneumonia. CMV was isolated from just over half of the HIV-positive children with pneumonia, similar to the prevalence rate of 22 - 78% from necropsy studies and 36 - 75% among antemortem studies.[12,16-20] The debate on the true pathogenic role of CMV in blood in molecular testing (CMV DNA PCR) of children is ongoing; however, both the related organ involvement (interstitial lung infiltrates, occasional hepatitis) and the clinical improvement in outcome on ganciclovir compared with historical controls suggest its pathogenic role.[21] Histological diagnoses of CMV pneumonitis by lung biopsy could not be performed on these children given their level of critical illness. While the impact of utilising other treatments like HFOV and cART could also have contributed to the improved outcomes, the management of the CMV appears particularly important. HFOV is merely supportive and the benefits of cART are likely to occur over several weeks to months post commencement and may therefore have little impact on short-term recovery.

Thirdly, 19% of children admitted to the PICU were HIV-positive, reduced from ~30% in earlier studies from the same PICU during the 1990s.[22,23] The implementation of the PMTCT programme has reduced the transmission rate to <5% in this population.[24,25] However, given the nature of patients admitted to PICUs, the finding of a 19% HIV prevalence rate among PICU admissions is likely to stabilise for as long as PICUs attract cases where the PMTCT programme has failed or has not been implemented effectively. Failure of PMTCT is indicative of poor access to health services, which often results in delayed presentation and a subsequent increase in the severity of the disease.

Limitations The few limitations of this review were the retrospective nature, the admission criteria, the small numbers and not determining the severity of disease on admission were concerns that could limit interpretation of the findings. Furthermore, the CMV viral load assessment and tissue diagnoses were not performed as part of the routine standard of care and so the diagnosis of CMV disease and the response to treatment could not be evaluated. The impact of other comorbidities on outcome was not analysed and medium-term outcomes of these cases were not evaluated, therefore the true value of improved short-term PICU outcomes could not be translated into overall survival benefits.

52 SARJ VOL. 23 NO. 3 2017

Conclusion

The findings of this retrospective review are important to support continued access of HIV-1-positive patients with pneumonia to scarce resources; the outcomes of these cases are now similar to those of other children with the same severity of illness requiring such services. In addition, the early introduction of cART should occur once a patient is stable, as delays in treatment have been associated with adverse outcomes.[13,26-29] While this retrospective review clearly shows an improved survival rate among HIV-positive children with CMV-associated pneumonia which was managed with ganciclovir, cART and HFOV, further prospective studies are required to clearly define CMV disease and the benefits of these interventions.

Acknowledgements. The authors would like to thank the Chief Executive of Inkosi Albert Luthuli Central Hospital, the staff, parents and children enrolled in the study for their participation.

Author contributions. PMJ was responsible for the conceptualisation of the study, assistance with data analysis and writing of the manuscript.

LG was responsible for the data collection and analysis and review of the manuscript.

Funding. None.

Conflicts of interest. None 1. Pillay Y. South Africa innovates to scale up and sustain its HIV response. Report presented at the UNAIDS Programme Coordinating Board (PCB). Geneva 11 - 13 December 2012.

2. Thirsk ER, Kapongo C, Jeena PM, et al. HIV exposed infants with acute lower respiratory infections managed with or without mechanical ventilation. S Afr Med J 2003;93(8):617-621.

3. Rabie H, De Boer A, Van den Bos S, et al. Children with human immunodeficiency virus infection admitted to a paediatric intensive care unit in South Africa. J Trop Pediatr 2007;53(4):270-273. https://doi.org/10.1093/tropej/fmm036 4. Palella Jr FJ, Delaney KM, Moorman AC, et al. Declining morbidity and mortality among patients with advanced human immunodeficiency virus infection. HIV Outpatient Study Investigators. N Engl J Med 1998;338(13):853-860. https://doi. org/10.1056/nejm199803263381301 5. Jeena P, McNally L, Strobie M, et al. Challenges in the provision of intensive care for HIV-infected children from developing countries. J Med Ethics 2005;31(4):226-230. https://doi.org/10.1136/jme.2003.004010 6. Jensen AM, Lundgren JD, Benfield T, Nielsen TL, Vestbo J. Does cytomegalovirus predict a poor prognosis in Pneumocystis carinii pneumonia treated with corticosteroids? A note for caution. Chest 1995;108(2):411-414. https://doi. org/10.1378/chest.108.2.411 7. Puthanakit T, Aurpibul L, Oberdorfer P, et al. Sustained immunologic and virologic efficacy after four years of highly active antiretroviral therapy in human immunodeficiency virus infected children in Thailand. Pediatr Infect Dis J 2007;26(10):953-956. https://doi.org/10.1097/inf.0b013e318125720a 8. Sud S, Sud M, Fredrich JO, et al. High frequency oscillation ventilation in patients with acute injury and acute respiratory distress syndrome (ARDS): Systematic review and meta-analysis. BMJ 2010;340:c2327. https://doi.org/10.1136/bmj.c2327 9. Bates M, Monze M, Bima H, et al. High human cytomegalovirus loads and diverse linked variable genotypes in both HIV-1 infected and exposed, but uninfected, children in Africa. Virology 2008;382(1):28-36. https://doi.org/10.1016/j.virol.2008.09.001 10. Punpanich W, Groome M, Muhe L, Qazi SA, Madhi SA. Systematic review on the etiology and antibiotic treatment of pneumonia in human immunodeficiency virus infected children. Pediatr Infect Dis J 2011;30(10):192-202. https://doi.org/10.1097/ inf.0b013e31822d989c 11. Slyker JA, Lohman-Payne BL, John-Stewart GC, et al. Acute cytomegalovirus infection in Kenyan HIV-infected infants. AIDS 2009;23(1):2173-2181. https://doi.org/10.1097/ qad.0b013e32833016e8 12. Kitchin OP, Masekela R, Becker P, et al. Outcome of human immunodeficiency virus-exposed and -infected children admitted to a paediatric intensive care unit for respiratory failure. Pediatr Crit Care Med 2012;13(5):516-519. https://doi. org/10.1097/pcc.0b013e31824ea143

ORIGINAL RESEARCH 13. Cowburn C, Hatherill M, Eley B, et al. Short-term mortality and implementation of antiretroviral treatment for critical ill HIV-infected children in a developing country.

Arch Dis Child 2007;92(3):234-241. https://doi.org/10.1136/adc.2005.074856 14. Mathivha LR, Luyt D, Hon H, et al. Outcome of mechanical ventilation in children infected with the human immunodeficiency virus. S. Afr Med J 1998;88(5):1447-1451 15. Jeena PM, Coovadia HM, Bhagwanjee S. Prospective, controlled study of outcome of human immunodeficiency virus-1 antibody positive children admitted to an intensive care unit. Crit Care Med 1996;24(6):963-967. https://doi.org/10.1097/00003246199606000-00015 16. Goussard P, Kling S, Gie R, et al. CMV pneumonia in HIV-infected ventilated infants.

Pediatr Pulmonol 2010;45(7):650-655. https://doi.org/10.1002/ppul.21228 17. Zampoli M, Morrow B, Hsiao NY, Whitelaw A, Zar HJ. Prevalence and outcome of cytomegalovirus- associated pneumonia in relation to HIV. Pediatr Infect Dis 2011;30(5):413-417. https://doi.org/10.1097/inf.0b013e3182065197 18. Jeena PM, Coovadia HM, Chrystal V. Pneumocystis carinii and cytomegalovirus infections in severely ill, HIV-infected African infants. Ann Trop Paediatr 1996;16(4):361-368. https://doi.org/10.1080/02724936.1996.11747852 19. Chintu C, Mudenda V, Lucas S, et al. Lung diseases at necropsy in African children dying from respiratory illnesses: A descriptive necropsy study. Lancet 2002;360(9338):985-990. https://doi.org/10.1016/s0140-6736(02)11082-8 20. Pena A, Lorranaga C, Lushsinger V, van Harroel J, Chaven A, Wu E. Cytomegalovirus disease in HIV infection Chilean children. Rev Chilena Infecto 2007;24(6):477-484.

21. Ljungman P, Griffiths P, Paya C. Definitions of cytomegalovirus infection and disease in transplant recipients. Clin Infect Dis 2002;34(8):1094-1097. https://doi. org/10.1086/339329

22. Morrow BM, Hsiao NY, Zampoli M, Whitelaw A, Zar HJ. Pneumocystis pneumonia in South African children with and without human immunodeficiency virus infection in the era of highly active antiretroviral therapy. Pediatr Infect Dis J 2010;29(6):535-539. https://doi.org/10.1097/inf.0b013e3181ce871e 23. Coovadia HM, McNally LM, Jeena PM. The etiology and outcome of pneumonia in HIV-infected children admitted to intensive care in a developing country. Pediatr Crit Care Med 2001;2(4):280-281. https://doi.org/10.1097/00130478-200107000-00017 24. Barron P, Pillay Y, Doherty T, et al. Eliminating mother-to-child transmission in South Africa. Bull World Health Organ 2013;91(1):70-74. https://doi.org/10.2471/blt.12.106807 25. Rollins NC, Little K, Mzolo S, Horwood C, Newell ML. Surveillance of mother-to-child transmission prevention programmes at immunisation clinics: the case for universal screening. AIDS 2007;21(8):134-141. https://doi.org/10.1097/qad.0b013e32814db7d4 26. Steininger C, Puchhammer-Stockl E, Popow-Kraupp T. Cytomegalovirus disease in the era of highly active antiretroviral therapy (HAART). J Clin Virol 2006;37(1):1-9. https://doi.org/10.1016/j.jcv.2006.03.005 27. Nesheim SR, Kapogiannis BG, Soe MM, et al. Trends in opportunistic infections in the pre- and post-highly active antiretroviral therapy eras among HIV-infected children in the Perinatal AIDS Collaborative Transmission Study, 1986 - 2004. Pediatrics 2007;120(1):100-109. https://doi.org/10.1542/peds.2006-2052 28. Puthanakit T, Aurpibul L, Oberdorfer P, et al. Hospitalisation and mortality among HIV-infected children after receiving highly active antiretroviral therapy. Clin Infect Dis 2007;44(4):599-604. https://doi.org/10.1086/510489 29. MihÄ&#x192;ilescu R, Arama V, Paraschiv S, et al. Impact of highly active antiretroviral therapy on cytomegalovirus viraemia in the absence of specific anti-cytomegalovirus therapy.

Rom J Intern Med 2008;46(4):305-311. https://doi.org/10.1186/1758-2652-11-s1-p255

SARJ VOL. 23 NO. 3 2017

53

ORIGINAL RESEARCH

Pulmonary nodules in patients with drug-resistant tuberculosis who have undergone adjuvant lung resection in a high-HIV-burden setting K Maharaj,1 MB ChB, FC Cardio; R Perumal,2,3,4 MB ChB, MPH, MMed, FCP; G Alexander,1 MB ChB, MMed, FC Cardio Department of Cardiothoracic Surgery, School of Clinical Medicine, Nelson R Mandela School of Medicine, University of KwaZulu-Natal, Durban, South Africa Centre for the AIDS Programme of Research in South Africa, Nelson R Mandela School of Medicine, University of KwaZulu-Natal, Durban, South Africa 3 Department of Pulmonology and Critical Care, Inkosi Albert Luthuli Central Hospital, Durban, South Africa 4 MRC HIV-TB Pathogenesis and Treatment Research Unit at CAPRISA, Doris Duke Medical Research Institute, Nelson R Mandela School of Medicine, University of KwaZulu-Natal, Durban, South Africa 1 2

Corresponding author: K Maharaj (drkmaharaj@outlook.com)

Introduction. Radiological findings of drug-resistant tuberculosis (DR-TB) have been described; however, results were focused on HIVnegative patients.

Objectives. To describe the radiological distribution of pulmonary nodules in patients who have undergone lung resection. Furthermore, the study establishes the association between perioperative nodules and sputum culture conversion in HIV-positive and -negative patients.

Methods. A retrospective review was conducted of clinical and radiological records of adult patients who had undergone lung resections for DR-TB, between 1 January 2011 and 31 December 2013, at a hospital in Durban, South Africa.

Results. Lung resections were undertaken in 47 patients with drug-resistant tuberculosis (DR-TB.) Among patient infections, 74.5% were multidrug-resistant TB (MDR-TB) and 25.5% were extensively drug-resistant TB (XDR-TB). The prevalence of HIV was 54.3% in patients with MDR-TB and 66.7% in XDR-TB. Pre-operative radiological studies showed a similar distribution of pulmonary nodules in HIV-positive and -negative patients. Positive pre-operative sputum cultures were noted in 34% of patients, but were not associated with pulmonary nodules on pre-operative radiological evaluation (p=0.81). At 1 month postoperatively, 81.3% of patients with a positive pre-operative sputum culture converted to negative sputum culture. The cumulative culture conversion rate was 87.5% at 2 months postoperatively. Using composite sputum culture outcomes, a good outcome was defined as consistently negative postoperative sputum cultures.

There was no association between poor outcomes and the presence of nodules, either pre-operatively (p=0.73) or postoperatively (p=0.52), irrespective of HIV status.

Conclusion. The presence of pulmonary nodules on pre-operative images was not associated with perioperative positive sputum results or poor outcomes. The clinical significance of residual pulmonary nodules in DR-TB, following lung resection, remains uncertain.

S Afr Respir J 2017;23(3):55-60. DOI:10.7196/SARJ.2017.v23i3.174

The epidemic of drug-resistant tuberculosis (DR-TB) remains a global challenge. Even though the incidence and mortality of tuberculosis (TB) has decreased worldwide, the overall prevalence of multidrugresistant tuberculosis (MDR-TB) is increasing alarmingly, especially in sub-Saharan Africa. This global distribution and low cure rates with medical therapy has resulted in renewed interest in adjuvant surgical therapy, especially in XDR-TB. The role of adjuvant surgery in these patients is still being refined, but has been beneficial in patients with severe radiological disease symptoms, and those with delayed microbiological response to medical therapy.[1,2] Indications for lung resection have remained largely unchanged since they were developed by Iseman et al.[2,3] in 1990. In medical therapy, drugs do not adequately penetrate lung cavities and pulmonary nodules – focal areas of TB. Adjuvant surgery is gainfully directed at sources of ongoing lung disease where pharmacological therapy is suboptimal, but the physiological cost of lung resection may be significant. In

addition, resection is usually limited to areas of severe cavitation.

Due to their distribution and multitude, pulmonary nodules are not usually amenable to surgical resection.

Pulmonary nodules may, however represent active TB infection with a lower mycobacterial burden than cavities. It stands to reason that these high-disease-burden nodules should be resected with the intention of decreasing the bacillary load, yet sometimes, resecting these may not be feasible. The aim of surgery in patients with DRTB is to excise pulmonary cavities with the intention of reducing the bacteriological burden of disease and potentiating cure. Patients with scattered nodules or bilateral reticular infiltrates have been selected, with the premise that surgery would be a ‘debulking’ procedure and that remaining lesions could be cleared with pharmacological therapy.[4] Nodules are sometimes left unresected when it is not feasible, e.g. when nodules are surrounded by otherwise normal lung tissue, or when there is inadequate pulmonary reserve to justify additional lung

SARJ VOL. 23 NO. 3 2017

55

ORIGINAL RESEARCH resection outside of the cavitary lobe. To our knowledge, no studies have investigated the microbiological profiles of resected nodules, or the significance of nodules on the postoperative chest radiographs in DR-TB patients. This study describes the presence of pulmonary nodules in patients with DR-TB who have undergone adjuvant lung resection surgery. This study evaluated the relationship between perioperative pulmonary nodules and the intermediate treatment outcome of culture conversion.

Methods

A retrospective review of adult patients with DR-TB was done between 1 January 2011 and 31 December 2013. Data were extracted from inpatient, outpatient and surgical records. Radiological images were accessed via a picture-archiving and communication system.

Alternatively, plain films were used when electronic records were not available.

Adjuvant lung resection was performed if disease was localised to one lung (absolute criterion) and if any of the following were present: persistently positive sputum smear or culture despite optimal antituberculous chemotherapy; extensive isolate drug-resistance pattern with a high probability of failure or relapse; severe cavitary disease; two or more relapses while on treatment; intolerance to medical therapy; haemoptysis; or bronchiectasis.

The extent of lung resection was guided by radiological (chest X-ray (CXR) and computed tomography (CT) chest scans) investigations.

The presence of cavities in the lobes identified for resection was considered mandatory. Nodules were resected only when feasible.

Lobes that contained nodules alone were not resected.

After surgery, all patients were maintained on their pre-operative DR-TB regimens based on the resistance profile of the last positive sputum culture. After recovery, patients were followed up at a postoperative clinic initially at 6 weeks after surgery, and then monthly for 6 months.

CT scans were done pre-operatively, and chest radiographs were done both pre- and postoperatively. Postoperative chest radiographs were performed 6 weeks after surgery. Chest radiographs and CT scans were blind-reviewed in a random order, and independently, by two thoracic surgeons blinded to case details. Images were reported for the presence, pattern and distribution of cavitary disease and pulmonary nodules. Discrepant findings were resolved by consensus or through the opinion of a third reviewer.

For the chest radiographic analyses, each lung was divided according to its lobar anatomy using posterior-anterior and lateral views. The observers assessed the extent and presence of lung parenchymal abnormalities with particular attention to cavities and presence of nodules.

Data analyses was undertaken using SPSS software (SPSS 23.0, IBM Corp., USA). For all statistical comparisons, a 5% level of significance was used (p<0.05); correspondingly, 95% confidence intervals were used to describe effect size. All data were assessed for normality, and non-parametric tests were used where necessary. Median values and interquartile ranges (IQRs) were used for data not amenable to parametric description. A two-tailed c2 test was used to assess the association between categorical variables of interest.

The study was conducted under the oversight of the Biomedical Research Ethics Committee at the Nelson R Mandela School of

56 SARJ VOL. 23 NO. 3 2017

Medicine, University of KwaZulu-Natal, South Africa (SA) (ref. no.

BE511/14).

Results

The median (IQR) age of patients was 31 (26 - 41) years and the majority (68%) of patients were female. The prevalence of HIV in this study was 57.4%, with the majority of these infections occurring in women (21/27, 77.8%) (Table 1).

Lung resection was performed in 47 patients with DR-TB infection.

MDR-TB constituted 74.5% of the cases, while the remainder were XDR-TB infections. The prevalence of HIV was 54.3% in the MDR-TB cohort and 66.7% in the XDR-TB cohort.

All of the HIV-positive patients were either already on antiretroviral therapy (ART) and had achieved a supressed viral load (85.2%), or were not yet on ART as their CD4 counts were above the threshold for initiation (14.8%) at the time of the study.

All of the patients studied were on appropriate drug-resistant treatment prior to surgery, according to national treatment guidelines and each patientâ&#x20AC;&#x2122;s individual resistance profile at the time of surgery.

Specific information on the number of effective drugs in each regimen that was used is unknown. At the time of surgery, 38% of patients had been on treatment for more than 24 months; there was no difference in the duration of medical therapy between patients with and without HIV infection (p=0.427).

Pre-operative chest radiography performed a week before surgery indicated that the upper lobes were the most common site of pulmonary nodules, with 49.8% occurring in either the right or left upper lobe (RUL; LUL). The rest of the lobar distribution pattern were middle lobe (RML) 17%, left lower lobe (LLL) 12.8%, and right lower lobe (RLL) 6.4% (Table 1). The distribution of pulmonary nodules on CXR was similar for both HIV-positive and -negative individuals.

Pre-operative, high-resolution computed tomography (HRCT) scans, performed 2 months pre-surgery, also revealed pulmonary nodules predominantly in the upper lobes (61.3%), with 25.8% of nodules in the LLL, 22.6% in the RML, and 12.9% in the RLL. The distribution of pulmonary nodules on HRCT was similar for HIVpositive and -negative patients. Despite medical therapy during the pre-operative evaluation, 34% patients were sputum culture positive, only two of whom had had less than 12 months of treatment. There was no association between pre-operative sputum culture status and HIV infection.

The majority of patients (66%) in this study were sputum negative at the time of surgery. There was no association between pre-operative sputum culture status and the finding of pulmonary nodules on preoperative radiological evaluation (p=0.81).

The most commonly performed surgical procedure was a left pneumonectomy (48.9%), with no difference in the distribution of sites of resection between HIV-positive and -negative patients. Right upper lobectomies were performed in 25.5% of patients.

At the first postoperative assessment (6 weeks after surgery), both radiographs and sputum analyses were evaluated. A total of 81.3% of patients with a positive pre-operative sputum culture had converted to sputum culture-negative; cumulative culture conversion rates were 87.5% at 2 months. Similar culture conversion rates were achieved in HIV-positive and -negative patients (88.9% v. 85.7%; p=0.853).

ORIGINAL RESEARCH Table 1. A comparison of HIV-positive and -negative patients with drug-resistant tuberculosis (DR-TB) who had undergone adjuvant lung resection Total DR-TB HIV-positive HIV-negative Age (years), median (IQR) Females, n (%) Drug-resistant profile Patients with MDR-TB, n (%) Patients with XDR-TB, n (%) Duration of treatment prior to surgery, n (%) 0 - 12 months 13 - 18 months 19 - 24 months >24 months Pre-operative location of nodules on CXR, n (%) (n=47) RUL RML RLL LUL LLL Pre-operative location of nodules on CT scan, n (%) (n=31) RUL RML RLL LUL LLL Pre-operative sputum culture of M. tb, n (%) Positive (DR) Not on ART (CD4 >200) Patients on ART, n (%) Patients with undetectable HIV viral load, n (%) Site of resection, n (%) Left pneumonectomy Right pneumonectomy RUL RML RLL LUL LLL Cumulative sputum culture conversion following surgery, n (%) At month 1 (first post-op assessment) At month 2 Relapse Persistent positive sputum culture Postoperative location of nodules on CXR, n (%) (n=35) RUL RML RLL LUL LLL

p-value 0.628 0.098 0.454

(n=47) 31 (26 - 41) 32 (68.1)

(n=27) 30 (23 - 44) 21 (77.8)

(n=20) 31.5 (28.25 - 35.75) 11 (55)

35 (74.5) 12 (25.5)

19 (70.4) 8 (29.6)

16 (80) 4 (20)

6 (12.8) 12 (25.5) 11 (23.4) 18 (38.3)

4 (14.8) 9 (33.3) 5 (18.5) 9 (33.3)

2 (10) 3 (15) 6 (30) 9 (45)

14 (29.8) 8 (17) 3 (6.4) 9 (19.1) 6 (12.8)

8 (29.6) 3 (11.1) 1 (3.7) 5 (18.5) 2 (7.4)

6 (30) 5 (25) 2 (10) 4 (20) 4 (20)

0.978 0.210 0.383 0.898 0.201

12 (38.7) 7 (22.6) 4 (12.9) 7 (22.6) 8 (25.8)

7 (25.9) 2 (7.4) 2 (7.4) 4 (14.8) 4 (14.8)

5 (25) 5 (25) 2 (10) 3 (15) 4 (20)

0.733 0.105 0.674 0.736 0.573

16 (34) 4 (8.5) 23 (48.9) 23 (48.9)

9 (33.3) 4 (14.8) 23 (85.2) 23 (85.2)

7 (35) n/a n/a n/a

0.905

23 (48.9) 6 (12.8) 12 (25.5) 0 (0) 2 (4.3) 4 (8.5) 0 (0)

14 (51.9) 3 (11.1) 6 (22.2) 0 (0) 2 (7.4) 2 (7.4) 0 (0)

9 (45) 3 (15) 6 (30) 0 (0) 0 (0) 2 (10) 0 (0)

13 (81.3) 14 (87.5) 2 (12.5) 2 (12.5)

8 (88.9) 8 (88.9) 1 (11.1) 1 (11.1)

5 (71.4) 6 (85.7) 1 (14.3) 1 (14.3)

0.389 0.853 0.853 0.853

8 (17) 5 (10.6) 5 (10.6) 0 (0) 3 (6.4)

5 (18.5) 2 (7.4) 3 (11.1) 0 (0) 2 (7.4)

3 (15) 3 (15) 2 (10) 0 (0) 1 (5)

0.535 0.356 0.644 0.613

0.427

0.719

IQR = interquartile range; M. tb = Mycobacterium tuberculosis; n/a = not applicable; ART = antiretroviral therapy; XDR = extensively drug-resistant; CXR = chest X-ray; CT = computed tomography; RUL = right upper lobe; RML = right middle lobe; RLL = right lower lobe; LUL = left upper lobe; LLL = left lower lobe.

SARJ VOL. 23 NO. 3 2017

57

ORIGINAL RESEARCH At the end of the study, three sputum culture relapses were noted in patients with pre-operative positive sputum cultures. Only one of these relapses remained culture-positive at the end of the 6-month postsurgical follow-up period. In total, three patients had a positive sputum culture during the surgical treatment follow-up period: two with persistently positive sputum cultures, and one patient with microbiological relapse in the postoperative period.

Pulmonary nodules on the postoperative chest radiograph were either new or nodules that were not resected during the lung resection (Table 2). Of the 47 patients in the study, 28 had pulmonary nodules on the pre-operative chest radiograph. These nodules were resected in six patients, while spontaneous clearance of nodules occurred in the unresected lungs of an additional seven patients following surgery. Three patients developed new nodules following surgery, while residual nodules accounted for 83.3% (15/18) of all nodules seen on the postoperative evaluation. There was no association between the finding of nodules postoperatively and pre-operative sputum culture status (p=0.44). Since most patients were managed by left pneumonectomy, there were fewer left-sided residual nodules.

No residual nodules were noted in the LUL, and three patients were noted to have residual nodules in the LLL. It was noted that most postoperative nodules were located in the right lung, with a distribution pattern that showed RUL (n=8), RML (n=5) and RLL (n=5) nodules. Findings were similar in patients with and without HIV infection.

A positive outcome was defined as persistently negative sputum cultures following surgery. This composite sputum culture outcome could not be defined upon inspection. There was no association between this composite sputum culture outcome and the presence of nodules, either pre-operatively (p=0.73) or postoperatively (p=0.52).

In addition, neither HIV status (p=0.7) nor the site of resection (p=0.37) were associated with a poor sputum culture outcome.

Discussion

According to the World Health Organization in 2015, there were an estimated 10.4 million new TB cases worldwide, with 11% of patients HIV-positive.[5] TB, and especially DR-TB, remains a global health challenge, with high treatment costs. Due to the global burden of disease and the low cure rates of DR-TB with medical therapy alone (especially XDR-TB), there is renewed interest in adjuvant surgical therapy.

Previous studies have analysed the outcomes of adjuvant lung resection in patients with DR-TB and HIV co-infection. The

perioperative radiological distribution of disease and the significance of the unresected pulmonary nodules have not been previously investigated.[6] Cha et al.[7] found that the predominant radiological features in HIV-negative XDR-TB patients were cavities and nodules.

This study aimed to determine if the perioperative presence of pulmonary nodules had any clinical or microbiological significance.

In this study, patients were mostly young, with more than half of the patients living with HIV. The prevalence of HIV in this study was 57.4%, with the majority of these infections occurring in women, which was similar to findings in a previous study by Lomtazde et al.[8] A study of XDR-TB in SA found that 73% of the cohort was HIV coinfected, and 61% received combined ART, with a reported treatment success rate of 22%.[9] All patients in this study had severe cavitary disease, which served as the primary indication for surgery in this cohort. Resection was considered for patients with cavitary disease owing to the difficulty of antibiotics penetrating the cavity and the high number of organisms that are contained within the cavity.[10] Cavitary disease (100%) and nodules (28/47 = 59.6%) were the predominant radiological features found on radiological examination. The most frequent radiographic abnormalities in the study by Cha et al.[7] in patients with XDR-TB were nodules (100%), reticulonodular densities (73%) and cavities (47%). The same study showed no significant differences in imaging findings between patients with XDR-TB and MDR-TB. Compared with previous studies of adjuvant surgery for DR-TB, this study revealed similar chest radiographic findings and a similar duration of therapy perioperatively.[11] The upper lobes were the predominant radiographic location of nodules, with 49.8% of all nodules occurring here. Similar distribution patterns were observed in other studies.[7] The distribution of pulmonary nodules on chest radiographs was also similar for both HIV-positive and -negative individuals. Badie et al.[12] compared the radiological appearance of pulmonary TB in groups of patients with and without HIV infection.

In that study, HIV-positive patients, compared with HIV-negative patients, had more atypical features, evidenced by consolidation and military nodules with bilateral lung involvement, diffuse involvement, and middle- and lower-zone predominance(as opposed to upper lobe). The same study by Badie et al.[12] found CXR findings were less characteristic in HIV-positive patients compared with HIV-negative patients.

Pre-operative HRCT findings were similar to CXR findings in relation to the distribution of disease. Distribution of pulmonary nodules on HRCT was similar for HIV-positive and -negative patients.

Table 2. The number of surgically treated patients with pulmonary nodules by anatomical site (N=40) Anatomical site

Preoperative, n

Postoperative, n

Resected, n

Cleared, nonresected, n

New, n

RUL

14

8

2

6

2

RML

8

5

1

2

0

RLL

3

5

1

0

3

LUL

9

0

4

5

0

LLL

6

3

3

1

1

Multiple sites*

28

18

6

7

3

RUL = right upper lobe; RML = right middle lobe; RLL = right lower lobe; LUL = left upper lobe; LLL = left lower lobe. * Figures are lower than the column totals as some patients had nodules in more than one lobe.

58 SARJ VOL. 23 NO. 3 2017

ORIGINAL RESEARCH As the patients in this study were required to have a suppressed HIV viral load on ART or a CD4 cell count of greater than 200 cells/ÎźL, it is possible that this group of patients with HIV do not reflect the clinical, radiological or microbiological profiles of patients with advanced or untreated HIV. Geng et al.[13] found the most important determinant of radiologic patterns of parenchymal abnormalities to be immunological status based on HIV infection, rather than the TB drug-resistance pattern. Fishman et al.[14] found that cavitation was less frequent in patients with MDR-TB with advanced HIV, compared with HIV-negative or -positive immunocompetent patients.

We found that the majority of patients in this study were sputum culture negative at the time of surgery. This is in keeping with the emerging role of adjuvant surgery in reducing the risk of relapse in these patients, who were likely selected on the basis of their increased risk for relapse. There was no association between pre-operative sputum culture status and the presence of pulmonary nodules on pre-operative radiological evaluation. Our findings differed from those of other studies where the presence of pulmonary nodules was a reliable marker of active disease, and was the most frequent finding in sputum positive patients, whereas traction bronchiectasis and atelectasis were more frequent in the sputum negative patients.[15] Nonetheless, there is a paucity of data on the microbiological or histological profile of pulmonary nodules, and this remains an important area for future research. It was difficult to determine if the presence of nodules represented old fibrotic disease, or if they were indeed active lesions from the current DR-TB infection. The use of metabolic imaging, such as a positron emission tomographycomputed tomography (PET-CT), would have been beneficial to discriminate between these possibilities, but was unfortunately impractical in our setting.[16] Most patients with positive pre-operative sputum cultures had converted to sputum culture negative within a month of surgery, and at 2 months, cumulative culture conversion rates were 87.5%.

These outcomes were similar to those of other studies of adjuvant lung surgery for DR-TB in low-TB/HIV-burden settings.[3,11] In appropriately selected patients who received early adjuvant lung resection surgery, Xie et al.[17] showed that high conversion rates can be achieved. The finding of higher conversion rates in surgically treated patients may be confounded by the strict inclusion criteria for surgery, and the overlap between criteria for fitness for surgery and predictors of treatment success. Indeed, it is possible that patients who are surgical candidates are likely to be in better general condition than those excluded from such interventions, therefore creating a bias for the true effectiveness of the surgery. It is unclear whether surgically treated patients are systematically different to patients within the general DR-TB cohort. However, this seems unlikely based on the comparison of baseline covariates such as median CD4 cell counts in patients managed with medical therapy alone. A study by Oâ&#x20AC;&#x2122;Donnell et al.[18] that evaluated the non-surgical treatment of XDR-TB in patients with HIV infection in SA, found a median CD4 count of 200.5 cells/ÎźL In that cohort, only 20% experienced sputum culture conversion, while 42% of patients died.

The patients in our cohort had a similar degree of immunological suppression, with superior clinical outcomes.

It remains clear that there is a significant association between surgical intervention and successful outcome when compared with

non-surgical treatment alone.[19] Similar culture conversion rates were achieved in patients with and without HIV infection (88.9% v. 85.7%; p=0.853). Brust et al.[20] similarly found that MDR-TB/ HIV co-infected patients can achieve culture conversion outcomes similar to those reported in HIV-negative cohorts, albeit lower rates than those achieved in patients treated with adjuvant surgical lung resection.

Despite the negative intermediate treatment outcome based on sputum culture status, two out of the three patients with poor outcomes in this study went on to culture convert within 6 months of surgery. It is important to note that both had pulmonary nodules that remained postoperatively. Indeed, the majority of patients who were culture negative at the time of surgery, and those who achieved satisfactory outcomes post-surgery, had residual nodules on postoperative imaging. It is unlikely that pulmonary nodules should serve as entities of surgical interest in adjuvant surgery for DR-TB. Although this was a retrospective study of a small cohort of patients with DR-TB, it serves as one of the largest described cohorts of surgically treated patients with DR-TB. We did not evaluate excised nodules microbiologically or histologically. This remains a future area of interest. Residual nodules following lung surgery were not evaluated further. The use of 2-deoxy-2-(18F)fluoro-D-glucose (FDG)-PET scans to determine activity of these nodules may be useful going forward. FDG-PET may play a role in the diagnosis of active TB infection in settings where conventional microbiological methods are unavailable and holds particular promise for monitoring response to therapy, particularly in MDR-TB and extra-pulmonary TB. [16] The next priority should be to determine the impact of residual pulmonary nodules on final treatment outcomes, and on the risk for disease relapse.

Limitations

Although this is currently the largest study of the impact of pulmonary nodules in patients with DR-TB treated by adjuvant lung resection, we recognise several noteworthy limitations. The retrospective design of this study meant that all patient data originated from programmatic medical records, and that imaging was neither protocoled, nor standardised. Nonetheless, a unit protocol existed to ensure the timing of images during the postoperative period, and all images were performed at a single radiology service with standard operating procedures for all radiological examinations. Similarly, sputum culture assessments were not performed under research conditions, but were conducted under programmatic conditions within an accredited medical laboratory. Studies of surgery for DR-TB all invariably suffer unavoidably from selection bias, as only those patients with appropriate cardiopulmonary reserve and adequate nutritional status are selected for surgery. It may be that patients undergoing resectional lung surgery are more likely to have a better pre-operative functional baseline and physiological status, and therefore experience superior treatment outcomes, regardless of any surgical intervention, than those declined for surgery based on pre-operative workup. However, studies in medically treated patients with similar baseline characteristics as patients in this study have demonstrated poorer outcomes.[9] This study evaluated only intermediate outcome using sputum culture status during a 6-month post-surgical follow-up period. Larger scale, longitudinal studies with longer follow-up, evaluating the clinical

SARJ VOL. 23 NO. 3 2017

59

ORIGINAL RESEARCH significance of pulmonary nodules on final treatment outcomes, are crucial to informing the role of adjuvant lung surgery in the treatment of DR-TB.

Conclusion

There is a paucity of evidence of the clinical significance of pulmonary nodules and cavities in DR-TB. Adjuvant surgery is emerging as a potential modality for improving treatment outcomes in these vulnerable patients. This study found no association between the presence of residual pulmonary nodules and the intermediate treatment outcome of culture conversion. It also demonstrated similarly good outcomes for HIV-positive and -negative patients. As the role of adjuvant surgery becomes more clearly defined, it will be of greater importance to determine the necessary extent of such an invasive strategy. The targeting of pulmonary nodules, for example, needs to be investigated further. However, present evidence supports the current practice of preserving non-cavitary lungs that contain pulmonary nodules Acknowledgements. Many thanks to all of the authors for their

significant contributions to the study.

Author contributions. RP contributed greatly to the statistical analyses and discussion. GA gave significant perspectives regarding the surgical procedures used and contributed notably to the discussion.

Funding. None Conflicts of interest. None 1. Kempker RR, Barth AB, Vashakidze S, et al. Cavitary penetration of levofloxacin among patients with multidrug-resistant tuberculosis. Antimicrob Agents Chemother 2015;59(6):3149-3155. https://doi.org/10.1128/AAC.00379-15 2. Calligaro GL, Moodley L, Symons G, Dheda K. The medical and surgical treatment of drug-resistant tuberculosis. J Thorac Dis 2014;6(3):186-195. https://doi.org/10.3978/j. issn.2072-1439.2013.11.11 3. Iseman MD, Madsen L, Goble M, Pomerantz M. Surgical intervention in the treatment of pulmonary disease caused by drug-resistant Mycobacterium tuberculosis. Am Rev Respir Dis 1990;141(3):623-625. https://doi.org/10.1164/ajrccm/141.3.623 4. Kempker RR, Vashakidze S, Solomonia N, Dzidzikashvili N, Blumberg HM. Surgical treatment of drug-resistant tuberculosis. Lancet Infect Dis 2012;12(2):157-166. https:// doi.org/10.1016/s1473-3099(11)70244-4

60 SARJ VOL. 23 NO. 3 2017

5. World Health Organization. Global Tuberculosis Report 2016. Geneva: WHO, 2016.

6. Alexander GR, Biccard B. A retrospective review comparing treatment outcomes of adjuvant lung resection for drug-resistant tuberculosis in patients with and without human immunodeficiency virus co-infection. Eur J Cardiothorac Surg 2016;49(3):823828. https://doi.org/10.1093/ejcts/ezv228 7. Cha J, Lee HY, Lee KS, et al. Radiological findings of extensively drug-resistant pulmonary tuberculosis in non-AIDS adults: Comparisons with findings of multidrugresistant and drug-sensitive tuberculosis. Korean J Radiol 2009;10(3):207-216. https:// doi.org/10.3348%2Fkjr.2009.10.3.207 8. Lomtadze N, Aspindzelashvili R, Janjgava M, et al. Prevalance and risk factors for multidrug-resistant tuberculosis in the Republic of Georgia: A population-based study.

Int J Tuberc Lung Dis 2009;13(1):68-73.

9. Oâ&#x20AC;&#x2122;Donnell MR, Padayatchi N, Kvasnovsky C, Werner L, Master I, Horsburgh Jr CR. Treatment outcomes for extensively drug-resistant tuberculosis and HIV co-infection.

Emerg Infect Dis 2013;19(3):416-424. https://doi.org/10.3201%2Feid1903.120998 10. Man MA, Nicolau D. Surgical treatment to increase the success rate of multidrugresistant tuberculosis. Eur J Cardiothorac Surg 2012;42(1):e9-e12. https://doi. org/10.1093/ejcts/ezs215 11. Dravniece G, Cain K, Holtz T, Riekstina V, Leimane V, Zaleskis R. Adjunctive resectional lung surgery for extensively drug-resistant tuberculosis. Eur Respir J 2009;34(1):180183. https://doi.org/10.1183/09031936.00047208 12. Badie B, Mostaan M, Izadi M, Alijani M, Rasoolinejad M. Comparing radiological features of pulmonary tuberculosis with and without HIV infection. J AIDS Clin Res 2012;3(10). https://doi.org/10.4172/2155-6113.1000188 13. Geng E, Kreiswirth B, Burzynski J, Schluger NW. Clinical and radiographic correlates of primary and reactivation tuberculosis: A molecular epidemiology study. J Am Med Assoc 2005;293(22):2740-2745. https://doi.org/10.1001/jama.293.22.2740 14. Fishman JE, Sais GJ, Schwartz DS, Otten J. Radiographic findings and patterns in multidrug-resistant tuberculosis. J Thorac Imaging 1998;13(1):65-71.

15. Bolla DS, Bhatt DC, Shah D. Role of HRCT in predicting disease activity of pulmonary tuberculosis. Glob Media J 2014;69(2):91-95.

16. Heysell SK, Thomas TA, Sifri CD, Rehm PK, Houpt ER. 18-Fluorodeoxyglucose positron emission tomography for tuberculosis diagnosis and management: A case series. BMC Pulm Med 2013;13(1):14. https://doi.org/10.1186/1471-2466-13-14 17. Xie B, Yang Y, He W, Xie D, Jiang G. Pulmonary resection in the treatment of 43 patients with well-localized, cavitary pulmonary multidrug-resistant tuberculosis in Shanghai.

Interact Cardiovasc Thorac Surg 2013;17(3):455-459. https://doi.org/10.1093/icvts/ivt251 18. Oâ&#x20AC;&#x2122;Donnell MR, Padayatchi N, Master I, Osburn G, Horsburgh CR. Improved early results for patients with extensively drug-resistant tuberculosis and HIV in South Africa. Inter J Tuber Lung Dis 2009;13(7):855-861. https://doi.org/10.1164/ajrccmconference.2009.179.1_meetingabstracts.a4089 19. Marrone M, Venkataramanan V, Goodman M, Hill A, Jereb J, Mase S. Surgical interventions for drug-resistant tuberculosis: A systematic review and meta-analysis.

Inter J Tuber Lung Dis 2013;17(1):6-16. https://doi.org/10.5588/ijtld.12.0198 20. Brust JCM, Lygizos M, Chaiyachati K, et al. Culture conversion among HIV co-infected multidrug-resistant tuberculosis patients in Tugela Ferry, South Africa. PLoS One 2011;6(1):e15841. https://doi.org/10.1371/journal.pone.0015841

CASE REPORT

Cystic lung disease in a young female patient M S Abdool-Gaffar, MB ChB, FCP (SA), FCCP, CIME Kingsway Hospital, Amanzimtoti, KwaZulu-Natal, South Africa Corresponding author: M S Abdool-Gaffar (sabs@docgaffar.co.za)

A case report of a female patient with Birt-Hogg-Dubé syndrome is presented to illustrate the salient features of this condition. The 45-yearold woman was relatively asymptomatic when the chest X-ray was done. She was found to have bilateral pulmonary cysts without any other significant abnormalities. A differential diagnosis is presented with a final diagnosis of the folliculin gene-associated, Birt-Hogg-Dubé syndrome.

S Afr Respir J 2017;23(3):61-62. DOI:10.7196/SARJ.2017.v23i3.165

Case report

A 45-year-old woman presented to her general practitioner (GP) with an upper respiratory tract infection, which had eventually settled on treatment, but her chest radiograph was abnormal. She had also had an abnormal chest radiograph 2 years prior to presentation at the GP, which had not been investigated, and she was subsequently referred to our unit. She was asymptomatic from a respiratory point of view, and denied any history of chest problems. As a food technologist, she travelled at least twice a month by air, both locally and internationally.

Her history was negative for any medical problems and she was a lifetime non-smoker.

The patient’s clinical examination was normal. Her chest radiograph (Figs 1A and B) demonstrated well-circumscribed, thin-walled cysts in both lower lobes, with a right-sided predominance.

A high-resolution computed tomography (HRCT) scan of the chest (Figs 2A - C) confirmed multiple thin-walled cysts involving both lower lobes, and the medial aspect of the right middle lobe. Localised left posterior basal atelectasis was evident, but there were no other parenchymal lesions. In particular, there were no features of groundglass opacification, reticulation, or nodularity. At the time of publication of this manuscript, the patient was still being managed conservatively.

A Fig. 1. (A) Posterior-anterior view of the chest radiograph showing cysts (yellow arrow) in the lower lobes.

Discussion

The patient presented with multiple pulmonary cysts and the differential diagnosis was made based on the patient’s history, clinical examination and radiological findings. The differential diagnosis in the index case included lymphangioleiomyomatosis (LAM), pulmonary Langerhans cell histiocytosis (PLCH), folliculin gene-associated syndrome (FLCN-S), which is also known as Birt-Hogg-Dubé syndrome, lymphocytic interstitial pneumonia (LIP), Pneumocystis jiroveci pneumonia (PJP), amyloidosis, light-chain deposition disease (LCDD), neoplastic disease, desquamative interstitial pneumonia, hypersensitivity pneumonitis, bronchiolitis, and a few other rare causes.[1] Bullae in patients with emphysema may mimic cysts, but this was not applicable to our patient. Other conditions that may produce cyst-like lesions include pneumatoceles secondary to an infective process, cystic bronchiectasis, and honeycombing, but these can be distinguished by associated primary aetiologies.

B Fig. 1. (B) Lateral view chest X-ray showing cysts (yellow arrow) in the lower zones.

SARJ VOL. 23 NO. 3 2017

61

CASE REPORT

A Fig. 2. (C) CT showing subpleural cysts (yellow arrow).

B Fig. 2. (A) Computed tomography (CT) scan showing cysts in a paracardiac distribution (yellow arrowhead); (B) CT scan showing cystic changes (yellow arrows).

The patient was a lifetime non-smoker, was not immunocompromised and had no history of any pulmonary pathology. The chest X-ray and CT scan did not show any intervening pulmonary parenchymal changes and in the context of the clinical presentation, the differential diagnosis in this patient was narrowed down to LAM and FLCN-S. The CT scan in patients with LAM usually demonstrates symmetrical, bilateral, multiple thin-walled cysts throughout both lungs, which were not evident in this patient. Lymphatic abnormalities are identified radiologically and may also occur in patients with LAM. Furthermore, Raoof et al.[2] provide a very good diagnostic approach to the management of a patient with cystic lung disease.

FLCN-S is an autosomal dominant genodermatosis that manifests as a multisystem disorder. The clinical findings are often evident in the

62 SARJ VOL. 23 NO. 3 2017

lungs, skin and kidneys;[1] however, patients may manifest with only pulmonary cysts. Affected patients may be asymptomatic, but the most common symptom is skin papules, especially on the face.[1] The most common pulmonary presentation is a spontaneous pneumothorax, which occurs in ~35% of cases.[3] The pulmonary cysts are usually bilateral, multiple, and vary in size. The intervening pulmonary parenchyma is normal. Cysts are further characterised by lowerzone predominant distribution; they may be lobulated and multiseptated.[1] Renal involvement includes cysts as well as benign and/ or malignant tumours. The diagnosis is confirmed by genetic testing, which confirms a mutation in the FLCN gene. The genetic test may also be used to screen family members; however, no specific therapy for this condition is currently available. The most likely diagnosis in the index patient was FLCN-S (Birt-Hogg-Dubé syndrome). She had no extra-pulmonary manifestations of the condition; however, the patient remains at risk of a spontaneous pneumothorax that would be exacerbated by continued air travel. This may influence the patient’s career path.[3] Genetic tests had not yet confirmed the diagnosis in this patient.

Acknowledgements. None Author contributions. MSA-G wrote the manuscript.

Funding. None Conflicts of interest. None 1. Francisco FAF, Souza AS Jr, Zanetti G, Marchiori, E. Multiple cystic lung disease.

Eur Respir Rev 2015;24(138):552-564. https://doi.org/10.1183/16000617.0046-2015 2. Raoof S, Bondalapati P, Vydyula R, et al. Cystic lung diseases: Algorithmic approach.

Chest 2016;150(4):945-965. https://doi.org/10.1016/j.chest.2016.04.026 3. Gupta N, Kopras EJ, Henske EP, et al. Spontaneous pneumothoraces in patients with Birt-Hogg-Dubé syndrome. Ann Am Thorac Soc 2017;14(5):706-713. https://doi. org/10.1513/annalsats.201611-886oc

REVIEW

An approach to pulmonary haemorrhage in children D Parris, BSc, MB BCh, FCP (Paeds), Dip Allergy; A van Niekerk, MB ChB, MMed; A C Jeevarathnum, MB BCh, FCP (Paeds), Dip Allergy, MMed, Cert Paed Pulm; R J Green, PhD, DSc Department of Paediatrics and Child Health, School of Medicine, Faculty of Health Sciences, University of Pretoria, South Africa, and Steve Biko Academic Hospital, Pretoria, South Africa Corresponding author: D Parris (drdeniseparris@gmail.com)

Pulmonary haemorrhage in children is very rare and the exact incidence remains unknown. Children usually swallow blood that has been coughed up and unless the haemoptysis is substantial, it can go unnoticed. Pulmonary haemorrhage can occur as a result of diffuse or local pathology. Diagnosis is based on a thorough history and examination, with laboratory, radiological, bronchoscopic findings, and lung biopsies in some cases. Treatment is directed at the underlying condition. On review of the available literature, the authors provide an approach to investigating pulmonary haemorrhage in children.

S Afr Respir J 2017;23(3):63-70. DOI:10.7196/SARJ.2017.v23.i3.152

Pulmonary haemorrhage (PH) can affect children of all ages. It can present surreptitiously, which delays the diagnosis, or as an acute lifethreatening event requiring resuscitation. Bleeding may be diffuse or localised.[1] Diffuse alveolar haemorrhage (DAH) occurs as a result of the disruption of the alveolar-capillary basement membrane in the lung, with resultant bleeding into the alveolar spaces. Injury can be immune- or non-immune mediated, with subsequent involvement of the blood vessels and the alveolar septae.[2] DAH can sometimes present as a life-threatening event, with haemoptysis and hypoxaemic respiratory failure, as well as falling haematocrit measurements and diffuse pulmonary infiltrates. A triad of haemoptysis, anaemia and airspace opacities on chest radiographs is suggestive of pulmonary haemorrhage.[3] Focal pulmonary haemorrhage is restricted to the lung, and is subsegmental, segmental or lobar.[4] It can occur due to undiagnosed congenital lung malformation, or secondary to infection, trauma, vascular disorders, underlying lung disease, coagulopathies or neoplasms of the lung. When assessing pulmonary haemorrhage, it is important to determine the origin of the bleeding. It is important to assess the upper airways and the gastrointestinal tract as possible sources of bleeding. Furthermore, as the diagnostic workup varies for local and diffuse DAH, it is necessary to distinguish between the two entities.[1]

Pathophysiology

The lungs receive blood from two separate circulations. The bronchial circulation, which is a high-pressure, low-volume circuit, supplied by the bronchial arteries arising from the aorta and the pulmonary circulation, which is a low-pressure, high-capacitance circuit arising from the right ventricle. Blood flows to the acinar units and is involved with gas exchange.[5] Bleeding from the bronchial circulation is profuse, with massive haemoptysis and the likelihood of death. Bleeding from the pulmonary circulation is low grade, chronic and diffuse.

Table 1. Causes of focal pulmonary haemorrhage in children Condition

Precipitating pathology

Infection

Pneumonia (Staphylococcus aureus) Lung abscess Viral infections, e.g. influenza A H1N1 Chronic bronchitis

Bronchiectasis

Cystic fibrosis Primary ciliary dyskinesia Primary immunodeficiency Secondary immunodeficiency

Congenital lung malformations

Sequestration Bronchogenic cyst Congenital pulmonary airway malformations

Trauma

Foreign-body inhalation Lung contusion Lung laceration Artificial airway suction catheters Inhalation injury

Vascular disorders

Pulmonary haemangioma Arteriovenous malformations Pulmonary emboli

Coagulopathy

Von Willebrandâ&#x20AC;&#x2122;s disease Thrombocytopenia Anticoagulants

Neoplasms

â&#x20AC;&#x201C;

SARJ VOL. 23 NO. 3 2017

63

REVIEW

Aetiology

Focal pulmonary haemorrhage Focal pulmonary haemorrhage is generally restricted to a focal region in the lung, but can involve an entire lobe (Table 1).[1] In some situations, the bronchial circulation may give rise to bleeding, which causes massive haemoptysis, but the low-pressure, high-capacitance circuit that supplies the alveoli is spared, with no consequent alveolar bleeding.

Diffuse alveolar haemorrhage DAH arises from the pulmonary circulation, which is a low-pressure, high-capacitance circuit providing blood to the acinar units. DAH can be immune- or non-immune mediated (Table 2).[1] Immune-mediated alveolar haemorrhage Understanding the immune-mediated mechanism of DAH requires a basic understanding of the anatomy of blood vessels and the role of neutrophils in immune-mediated pathogenesis. Blood vessels consist of arteries, arterioles, capillaries, venules, and veins. Arteries and veins are composed of three tissue layers.

The inner layer (tunica intima) is composed of endothelial cells and a layer of elastic lamina, forming the basement membrane of the blood vessel. The bulky middle layer (intima media) contains elastin and smooth muscle, which regulates the diameter of the blood vessel.

Table 2. Causes of DAH in children

The outermost layer (tunica externa) consists of areolar or fibrous connective tissue that provides structural support and protection (Fig. 1).[6] Neutrophils are white blood cells that play an important role in innate immunity. They circulate in the bloodstream and are the first line of defense against invading organisms. Neutrophils defend by phagocytosis, antigen presentation, and activating T-cells, which stimulate the humoral immune response.[4,8] Neutrophils either circulate, or bind to the vascular endothelium as marginating neutrophils. Damaged neutrophils leak enzymes, which damage vessels as they migrate toward the site of injury or infection – a process called diapedesis.

Pathophysiology of pulmonary capillaritis

Pulmonary capillaritis (PC) is an immune-mediated disorder that targets the cytoplasmic components of neutrophils, subsequently developing antineutrophil cytoplasmic antibodies (ANCA).

Two neutrophil proteins are involved in the immune response: the myeloperoxidase (MPO), and proteinase 3 (PR3) granules.

Autoantibodies are generated against the neutrophil components.

The damaged neutrophils leak enzymes as they migrate through the vascular system, which damages the endothelium of the blood vessels, with subsequent exposure of the sub-endothelial structures.

Damaged endothelial cells elicit an antibody response, which develops anti-endothelial cell antibodies (AECA). AECAs bind to

Immune-mediated Idiopathic pulmonary capillaritis DAH Granulomatosis with polyangiitis[6] Microscopic polyangiitis Goodpasture's syndrome Systemic lupus erythematosus

Tunica externa – external elastic membrane Tunica media – smooth muscle

Henoch-Schönlein purpura

Internal elastic membrane

Behçet's disease

Endothelium

Tunica intima

Cryoglobulinaemic vasculitis Juvenile idiopathic arthritis Non-immunemediated

Idiopathic pulmonary haemosiderosis Acute idiopathic pulmonary haemorrhage of infancy: • Coagulation disorders • Asphyxiation

Fig. 1. Structure of blood vessels. (Courtesy of Dr D Parris, Department of Paediatrics and Child Health, Steve Biko Academic Hospital)

• Stachybotrys chartarum Heiner syndrome Cardiovascular causes: • Pulmonary vein atresia/stenosis • Total anomalous pulmonary venous return • Pulmonary veno-occlusive disease • Pulmonary capillary haemangiomatosis • Pulmonary telangiectasia • Mitral stenosis DAH = diffuse alveolar haemorrhage; GPA = granulomatosis with polyangiitis (Wegener's granulomatosis).

64 SARJ VOL. 23 NO. 3 2017

Fig. 2. H&E lung biopsy specimen showing increased numbers of polymorphonuclear leukocytes (yellow arrows) and fibrinoid necrosis in the interstitium, with erosion of the alveolar epithelium in a child with Behçet's disease. (Courtesy of Dr Dinkel, Department of Histopathology, University of Pretoria)

REVIEW the vessel, progressively exposing its basal components, to which antibodies are also synthesised. These include anti-endothelial cell antibodies (AECA), antiglomerular basement membrane antibodies (anti-GBM), antibasal membrane laminin antibodies (ABLA), and antiphospholipid antibodies (APLA).[4,8]

Aetiology of PC

PC is an inflammatory process involving capillaries, with subsequent inflammation and necrosis of these vessels. It involves a histological diagnosis without identifying the disorder.[9] It may present in isolation or as part of a systemic disorder (Fig. 2).

The most frequently isolated forms of PC are ANCA-associated vasculitides and systemic lupus erythematosus. Other diseases associated with PC include microscopic polyangiitis (MPA), granulomatosis with polyangiitis (GPA; Wegener's granulomatosis), Henoch-Schönlein purpura, Behçet's disease, cryoglobulinaemic vasculitis, and juvenile idiopathic arthritis.

Histological assessment can distinguish between underlying systemic vasculitis and immune-mediated vasculitis.[1] Cytoplasmic ANCA-associated disorders include GPA, MPA and eosinophilic granulomatosis with polyangiitis (EPGA; Churg-Strauss syndrome).[6]

Fig. 3. Frontal chest X-ray showing alveolar infiltrates (white arrows) bilaterally consistent with alveolar haemorrhage. (Courtesy of Dr D Parris, Department of Paediatrics and Child Health, Steve Biko Academic Hospital)

Fig. 4. High-resolution contrast scan of the chest, axial views at the level of the heart (yellow arrows) depicting ground-glass appearance, septal thickening and inhomogeneous opacification. (Courtesy of Dr D Parris, Department of Paediatrics and Child Health, Steve Biko Academic Hospital)

Although there is some overlap between disease entities, proteinase 3 (PR-3) is associated with GPA, and MPO is associated with microscopic angiitis. Goodpasture’s disease (GPD) differs from the other immunemediated disorders, as the antibodies are directed against the glomerular basement membrane.[10] Granulomatosis with polyangiitis (GPA; Wegener's granulomatosis) GPA is a systemic vasculitic disease. It is distinguished by its necrotising granulomatous inflammation, affecting both upper and lower respiratory tracts, the kidneys, and small and medium-sized vessels to varying degrees. The antibodies are directed against the PR-3 neutrophil granulocytes.

Upper respiratory symptoms include otitis media, sinusitis, chronic rhinitis and nasal cartilage destruction, which results in a saddle nose deformity. Salivary glands are often swollen and painful; subglottic stenosis and tracheobronchial ulceration are associated complications.

Parenchymal lung nodules coalesce and form cavities, which erode blood vessels, causing DAH. Clinically, patients present with a cough, dyspnoea, haemoptysis and hypoxaemia. Other organ systems, particularly the kidneys, may be involved, causing haematuria.

Chest radiographs show nodules, with or without cavitation, pulmonary haemorrhage, a reticulonodular pattern, and wedge-like consolidations (Fig. 3). Computed tomography (CT) scans are better at delineating the nodules and cavities, and characterising groundglass opacities, densities and consolidations (Fig. 4).[11] Pulmonary involvement MPA MPA is a rare pauci-immune, small-vessel vasculitis, involving the skin, joints, kidneys and lungs. It may present in childhood with haemoptysis, anaemia, hypoxaemia, pulmonary haemorrhage, or severe respiratory disease requiring ventilation. MPA is distinguished from GPA by the presence of high titres of antimyeloperoxidase (MPO)-ANCA, and the absence of granulomatous inflammation. [11,12] Lung histology demonstrates infiltration of neutrophils in small blood vessels, with associated fibrinoid necrosis. Imaging studies show diffuse alveolar infiltrates, tree-inbud patterns, and septal thickening.

Goodpasture’s disease (GPD) GPD is a rare, systemic autoimmune disease that causes DAH and a progressive crescentic glomerulonephritis. It is characterised by antiGBM antibody-mediated damage, and the antibody response is IgG1 subtype. The autoantibodies are directed at the a3 and a5 domains of the type IV collagen fibres present in the basement of blood vessels and glomeruli. Approximately one-third of patients with GPD have perinuclear antineutrophil cytoplasmic autoantibodies, directed against either the MPO or the PR3 protein of the neutrophil (Fig. 5).[10] Coughing, haemoptysis, dyspnoea, and fatigue are common presenting features. Chest X-rays (CXR) and CT scans show features of nonspecific DAH. A definitive diagnosis is made by immunofluorescent staining, demonstrating the presence of antiGBM antibodies. Early diagnosis is important, as plasmapheresis, immunosuppression with corticosteroids, and the addition of cyclophosphamide can successfully induce remission in 90% of patients who survive acute presentation.[10]

SARJ VOL. 23 NO. 3 2017

65

REVIEW syndrome children are more prone to developing celiac disease. This autoimmune disease appears to be more prominent in children with underlying genetic disorders.[14,17] However, lung biopsies show bland alveolar haemorrhage, with large amounts of haemosiderin-laden macrophages in the alveoli (Figs 8 A and B). There is a distinct absence of any inflammation, capillaritis and vasculitis on histology.[13,18] IPH is essentially a diagnosis of exclusion in a patient with DAH.

Fig. 5. H&E stain of renal biopsy. Histology depicts diffuse, high-intensity, linear staining of the glomerular basement membrane (yellow arrow) in a patient with antiglomerular basement membrane disease. (Courtesy of the Department of Nephrology Steve Biko Academic Hospital) Isolated PC Isolated pauci-immune PC is a rare disorder with a poorly understood aetiology. Clinically, patients present with signs and symptoms of alveolar haemorrhage, but lack the renal and systemic manifestations. The histopathologic pattern shows disruption of the integrity of the alveolarcapillary basement membranes, with blood-filled alveoli. Haematological testing demonstrates low haemoglobin and elevated erythrocyte sedimentation rate (ESR) and C-reactive protein. The renal function is completely normal. Bronchoalveolar lavage (BAL) specimens contain haemosiderin-laden macrophages. If ANCA is isolated, it is usually of the MPO type, as the PR-3 variation is rarely identified. Diagnostic imaging demonstrates diffuse alveolar opacities. In the absence of ANCA, a lung biopsy is imperative, as PC can present with fulminant haemorrhage, where implementing early therapy is crucial.[12] Idiopathic pulmonary haemosiderosis Idiopathic pulmonary haemosiderosis (IPH) is a lung disease with unknown aetiology. It is characterised by alveolar capillary bleeding, with a subsequent accumulation of haemosiderin in the lungs. It is a diagnosis of exclusion, and appears to have a better prognosis than PC. Clinically the symptoms are nonspecific, with malaise, cough and tachypnoea, with or without haemoptysis. The diagnostic criteria for IPH is an iron deficiency anaemia and diffuse alveolar infiltrates on imaging studies (Figs 6 and 7). Haemosiderin-laden macrophages are isolated from sputum, gastric aspirates or BAL specimens.[13] The aetiopathogenesis of IPH remains contentious. Environmental causes in children with IPH who developed pulmonary haemorrhage following exposure to Stachybotrys chartarum. This theory has subsequently been disproved. The allergic theory, based on the correlation between pulmonary haemorrhage and cow’s milk protein (Heiner syndrome) remains contentious. Evidence for an autoimmune aetiology is surfacing.[14-16] Autoimmune antibodies cited to have a role include smooth-muscle antibodies, antinuclear antibodies, and ANCA. The antibodies are thought to elicit a primitive vasculitis, as well as systemic disease. A strong correlation exists between celiac disease and IPH. These patients have antibodies to endomysium, transglutaminase and gliadin. Genetic testing demonstrates a positive HLA-DQ2 association in patients with IPH and celiac disease. Down

66 SARJ VOL. 23 NO. 3 2017

Distinguishing PC from IPH Treatment modalities for PC and IPH differ, and therefore it is imperative to distinguish between the two disease entities. PC is immune-mediated, more difficult to treat, and often requires cytotoxic therapy to induce remission and circumvent recurrence.

IPH is classically treated with steroids, with hydroxychloroquine or azathioprine used as adjuncts for their steroid-sparing effect.[14,15] Acute idiopathic pulmonary haemorrhage of infancy (AIPH) AIPH is a disease that occurs in previously healthy infants.[4,19] Clinical presentation is severe, with acute respiratory failure, diffuse infiltrates on imaging, and marked pulmonary haemorrhage. This disease was previously ascribed to the mycotoxin-producing mould, Stachybotrys chartarum, found in patients’ dwellings. This theory has subsequently been disproved and coagulopathies, such as Von Willebrand’s disease, appear to have a stronger correlation with the disease. Thus, coagulation screening is essential.

A

B

Fig. 6. High-resolution computed tomography (HRCT) chest axial views of the patient. A fine reticular pattern and consolidation (yellow arrows) were noted. (Courtesy of Dr D Parris, Department of Paediatrics and Child Health, Steve Biko Academic Hospital)

A

B

Fig. 7. Frontal chest X-rays of a patient with idiopathic pulmonary haemosiderosis (A and B). X-rays depict bilateral diffuse reticular infiltrates (white arrows) that involve both lung fields. (Courtesy of Dr D Parris, Department of Paediatrics and Child Health, Steve Biko Academic Hospital)

REVIEW Table 3. List of investigations to elicit the aetiology directed by the history and examination Aetiology Investigation Tuberculosis/HIV-mediated Infective markers underlying lung disease Cystic fibrosis (CF) Sweat test, faecal elastase Primary ciliary dyskinesia Light phase microscopy, mucociliary clearance, nasal biopsy and electron microscopy Non-CF mediated Sputum, laboratory studies bronchiectasis Primary immunodeficiency Immune screen, disorder immunoglobulins, lymphocytes, antibody responses to capsulated and non-capsulated vaccines Bleeding diathesis Coagulation screening Cardiac disease Echocardiography Vascular malformations Angiography

A

B

Cardiovascular abnormalities Pulmonary vein atresia and stenosis, total anomalous pulmonary venous return and pulmonary veno-occlusive disease have been associated with DAH. Left ventricular failure and mitral stenosis cause pulmonary venous hypertension, and an increased risk of bleeding. Vascular malformations, either congenital, as those seen in hereditary haemorrhagic telangiectasia, or acquired post Glen or Fontan procedure for cyanotic cardiac lesions, can present with catastrophic pulmonary haemorrhage, or septic emboli and abscesses.

Approach to pulmonary haemorrhage

Fig. 8. Lung biopsy specimens. (A) H&E staining; (B) Prussian blue staining (white arrows). A large number of haemosiderin-laden macrophages and foci of fresh haemorrhage were observed. (Courtesy of Dr Dinkel, Department of Histopathology, University of Pretoria)

A

frequently absent, and the patchy infiltrates on CXR are often misdiagnosed as pneumonia. HRCT is required to make a definitive diagnosis (Figs 9A and B).

Diagnosis is based on a high index of suspicion. Laboratory findings may demonstrate elevated IgE levels, and high titres of serum IgG precipitins to various milk proteins may be present. However, a definitive diagnosis is based on the withdrawal of cow’s milk with a clinical improvement. Early re-introduction of cow’s milk is subsequently associated with clinical deterioration.[17,18]

B

Fig. 9. Diffuse alveolar infiltrates, with ground-glass attenuation in both lung fields (yellow arrows) (A and B).

Other important considerations are underlying cardiac disease, with mandatory echocardiography. Potential child abuse is excluded with skeletal surveys and a retinal assessment.

Cow’s milk sensitivity Cow’s milk sensitivity (Heiner syndrome) is a rare disorder in infants and young children. The disease is considered to be a non-IgE-mediated allergy to cow’s milk protein. The disease manifestation is variable, resulting in diagnostic delay. Infants and young children present with a history of chronic upper and lower respiratory tract infections. The classic triad of anaemia, haemoptysis, and diffuse alveolar infiltrates on imaging is

History, examination and laboratory evaluation A detailed history and examination is necessary as this may influence the investigations required to confirm a diagnosis. Full blood count, serum creatinine levels, and urinalysis are basic laboratory investigations that should be ordered on presentation.

Pulmonary-renal syndrome can be identified by elevated ESR and creatinine levels, as well as microscopic haematuria. Other important tests to consider are the autoimmune markers, including serum ANCA, anti-GBM, and antinuclear antibodies for collagen vascular disease. The tests are based on patient history.

Other investigations (Table 3) to consider based on the patient’s history include a diagnostic workup for infective causes of pulmonary haemorrhage (TB, HIV), underlying structural lung damage occurring in cystic fibrosis, and non-immunodeficiencymediated bronchiectasis.

Chest X-ray CXR is imperative in trying to elicit whether the pulmonary haemorrhage is as a result of a diffuse or localised aetiology.

Bronchoscopy Flexible bronchoscopy and consecutive lavage specimens are hallmarks of DAH diagnosis. Radiographic evaluation is useful in assessing the areas for lavage. Once identified, the bronchoscope is wedged into the associated subsegmental bronchus. Three separate 50 - 60 mL aliquots of sterile saline are instilled into the segments, and recouped. The diagnosis of DAH is confirmed by increasing haemorrhagic aliquots being retrieved. Samples should be sent for routine bacterial, fungal and viral studies to rule out infection. Cytology samples should also be sent, as haemosiderinladen macrophages can be demonstrated on BAL specimens by staining with Prussian blue. A diagnosis of DAH is confirmed if >20% of 200 macrophages stain positive for haemosiderin.[2]

SARJ VOL. 23 NO. 3 2017

67

REVIEW Biopsy Transbronchial biopsies are controversial as the patient may be too ill, and the sample size is frequently too small to make a diagnosis. Open lung biopsies may be warranted, but the role in diagnosing DAH is not well established. Renal biopsy may be more valuable in collagen vascular disorders, like granulomatosis with polyangiitis.[19] Histology Histology identifies three different patterns. Fig. 10A demonstrates the blood present both in the alveoli and the interstitium. Haemosiderinladen macrophages are visible in the parenchyma and alveolar septae with a Prussian blue stain.

DAH

Histology identifies the specific histological patterns in DAH. In 1985, Mark and Ramirez[20] identified specific histological features associated with pulmonary capillaritis, which is the most common cause of DAH: (i) fibrin thrombi occlude the capillaries in the interalveolar septae; (ii) fibrin clots adhere to the inter-alveolar septae in a pedunculated manner; (iii) fibrinoid necrosis of the capillaries; (iv) neutrophils and nuclear dust are found in the fibrin, the interstitium, and in the alveolar blood; (v) red blood cells, and haemosiderin-laden macrophages are present in the alveoli. Not all features are present in every patient (Fig. 10B).

A

B

Fig. 10. Pulmonary capillaritis. (A) A pleural biopsy of a patient with Behรงet's disease. Pulmonary capillaritis of the lung shows diffuse acute and organising haemorrhage. Neutrophils and necrotic debris are within inter-alveolar septa supportive of necrotising capillaritis (yellow arrow); (B) The ongoing haemorrhage is supported by intra-alveolar fibrin associated with plugs and clusters of intra-alveolar haemosiderophages and extensive fibrosis of the interalveolar septae (yellow arrow).

A

B

Fig. 11. Acute stage diffuse alveolar damage (DAD). (A) Hyaline membranes are plastered against alveolar septae thickened by oedema, scattered mononuclear inflammatory cells, and a few fibroblasts (yellow arrow); (B) Extensive thrombus formation, fibrinoid necrosis and interstitial fibrosis (yellow arrow).

68 SARJ VOL. 23 NO. 3 2017

Diffuse alveolar damage

Diffuse alveolar damage (DAD) resembles the histopathological findings seen in acute respiratory distress syndrome, with oedema, capillary congestion and microthrombi present in the alveolar septae (Fig. 11).

Bland pulmonary haemorrhage

The last histological pattern is bland PH. This pattern demonstrates haemorrhage in the alveolar spaces, without alveolar damage, and is found in idiopathic pulmonary hypertension, congestive cardiac failure, anticoagulant therapy, and GP. Treatment Treatment instituted for PH is based on the clinical presentation and the associated underlying aetiology. Despite the underlying aetiology, massive haemoptysis and respiratory failure require intubation, with high positive end-expiratory pressure utilised to recruit alveoli. Patients often require blood transfusions and stabilisation of haemodynamic status as a priority. All known or suspected coagulopathies should be corrected. Platelet counts should be >50 000/uL, and an acceptable prothrombin time <1.5 s.

Bronchial artery embolisation may be an option in patients with localised focal PH. Unfortunately, the literature offers no standard guidelines to approaching patients with DAH. To arrest the bleeding there are a few options, which include pulsed methylprednisone. Corticosteroids are the initial modality of therapy, especially in patients with either immune- or inflammatory-mediated DAH. They act in therapeutic doses to suppress the inflammatory process by inhibiting migration of inflammatory mediators to the site of bleeding, which would otherwise exacerbate the disease process. Corticosteroids also act by suppression of the immunological process. This process involves direct inhibition of antigen-antibody complex formation that, once formed, damages the vascular endothelium and aggravates bleeding. High-dose methylprednisone is initiated at a dose of 30 mg/kg/day for 3 - 5 days. The dose is gradually tapered.

Corticosteroids on their own usually fail to control the bleeding, with consequently high fatalities. Complementary therapies are thus required to control bleeding.[20,21] Immunosuppressant therapies that have been successfully used include agents like cyclophosphamide and hydroxychloroquine, among others, over a 3 - 6-month period.

Rituximab has been used with some success. This is a monoclonal antibody that binds CD20, which is expressed on beta-cells. It acts in three very precise ways: antibody-dependent cytotoxicity, complement-mediated cytotoxicity, and apoptosis of beta-cells.

This results in a decrease in the number of beta-cells, and less immunoglobulin formation (IgG), which binds the antigens on capillaries in the lungs and results in pulmonary haemorrhage.[22-24] If this fails, plasmapharesis and intravenous immunoglobulins may play a role in removing the circulating antigens from the circulation, or binding of the antigens to the exogenously administered immunoglobulins and prevent further binding to the exposed antigens on the capillaries.

Newer treatment modalities Other novel therapies attempted include the use of antifibrinolytics.

The two most commonly used agents are tranexamic acid (TXA)

REVIEW and epsilon-aminocaproic acid (EACA). TXA acts by binding to plasminogen, which in turn inhibits its binding to fibrin. Activation to plasmin is thus impaired. Both systemic and local administration of TXA have been used in the prophylaxis and treatment of bleeding diatheses, whether the bleeding is congenital or acquired.[25,26] TXA has been used successfully in the intravenous, aerosolised and intrapulmonary form to treat pulmonary haemorrhage. However, side-effects like seizures have rendered it less favourable for treatment. EACA in conjunction with corticosteroids has demonstrated variable results, and more studies are required to assess its value in DAH.[27] There is, however, increasing evidence for the use of recombinant factor VIIa (rFVIIa).[28,29] This drug was initially developed to control bleeding in haemophiliac patients, who were either FVII deficient, or had inhibitors to FVII. rFVIIa exerts its action via two mechanisms. The first involves activating factors X and IX at the sites of injury. Factors X and IX bind to tissue factor (TF) and activated platelets, and this promotes thrombin generation. The second method is through a TFindependent pathway, where rFVIIa directly activates factor X on the surface of activated platelets. rFVIIa is used to achieve haemostasis in a number of life-threatening bleeding situations. The drug has been successfully used in patients with bleeding diatheses secondary to platelet dysfunction or thrombocytopenias and in bleeding following haematopoeitic stem cell transplantation (HSCT), coagulopathies associated with liver failure, as well as trauma-mediated haemorrhage.

There are insufficient data in the literature to support the use of rFVIIa for DAH. However, it has been used successfully in DAH patients with incalcitrant pulmonary haemorrhage, both in immune and non-immune cases of DAH associated with connective tissue disorders, vasculitides with associated pulmonary haemorrhage, post HSCT, and pulmonary haemorrhage associated with infections.

The optimal dose for therapy is not clearly established, and the ideal route of administration is not clearly defined. However, favourable results have been achieved both with intravenous and bronchoscopic administration. Intravenous administration with dosages of between 35 and 200 µg/kg, either as a single bolus dose or repeated 2 - 4-hourly.

Intrapulmonary administration, achieved bronchoscopically, requires a total dose of 50 - 90 µg/kg of rFVIIa. This dose of rFVIIa requires dilution in normal saline, which can either be dispensed as a single dose, or if bleeding fails to abate, as repeated doses over a 24-hour period.[28] In children, the evidence currently seems to favour intrapulmonary administration to achieve effective haemostasis.[30] Using TXA, and rVIIa in a two-step approach in children with unremitting DAH Based on the mechanism of action of both drugs discussed above, a combination of the two would yield the best results in intractable DAH; studies are currently underway to assess this theory. To date, administering TXA, followed by rFVIIa, has demonstrated improved clot stability, resilience to fibrinolysis, cessation of pulmonary bleeding, and improved outcomes. This theory requires further testing.[29]

Conclusion

PH is a potentially life-threatening condition that could present as fulminant haemoptysis from the high-pressure bronchial circulation, or as an insidious bleed from the low-pressure pulmonary circulation.

Both pathologies are associated with significant morbidity and

mortality and treatment requires a thorough patient history, examination, and goal-directed investigations to make the diagnosis.

For patients with focal PH, it is imperative to rule out structural lung disorders that occur from underlying disorders, like bronchiectasis, CF, and primary immunodeficiency disorders. DAH requires a systemic workup for autoimmune disease. A bronchoscopy should be performed early and, where definitive diagnosis remains a problem, renal and lung biopsies may be imperative to establish a diagnosis and implement therapy.

Acknowledgements. Dr Dinkel, Department of Histopathology, University of Pretoria for the histology slides.

Author contributions. AvN, ACJ and RJG all proofread and edited the manuscript.

Funding. None Conflicts of interest. None 1. Vece TJ, de Guzman MM, Langston C, Fan LL. Diffuse alveolar hemorrhage in children. In: Wilmott RW, Boat TF, Bush A, Chernick V, Deterding RR, Ratjen F. eds.

Kendig and Chernick's disorders of the respiratory tract in children. Philadelphia: Elsevier, 2012:848-857. https://doi.org/10.1016/b978-1-4377-1984-0.00058-9 2. Schwarz MI. Diffuse alveolar haemorrhage syndrome. http://www.uptodate.com/ contents/the-diffuse-alveolar-hemorrhage-syndromes (accessed 21 December 2016).

3. Collard HR, Schwarz MI. Diffuse alveolar haemorrhage. Clin Chest Med 2004;25(3):583-592. https://doi.org/10.1016/j.ccm.2004.04.007 4. Avital A, Springer C, Godfrey S. Pulmonary haemorrhagic syndromes in children.

Pediatr Resp Rev 2000;1(3):266-273. https://doi.org/10.1053/prrv.2000.0058 5. Editors of Encyclopaedia Britannica. Pulmonary circulation. https://www.britannica. com/science/pulmonary-circulation (accessed 21 December 2016).

6. Khan I, Watts RA. Classification of ANCA-associated vasculitis. Curr Rheumatol Rep 2013;15(12):383. https://doi.org/10.1007/s11926-013-0383-6 7. Antranik. Blood Vessels. 2016. http://antranik.org/blood-vessels (accessed 21 December 2016).

8. Németh T, Mócsai A. The role of neutrophils in autoimmune diseases. Immunol Let 2012;143(1):9-19. https://doi.org/10.1016/j.imlet.2012.01.013 9. Franks TJ, Koss MN. Pulmonary capillaritis. Curr Opin Pulm Med 2000;6(5):430-435. https://doi.org/10.1097/00063198-200009000-00008 10. Chan AL, Louie S, Leslie KO, Juarez MM, Albertson TE. Cutting edge issues in Goodpasture’s disease. Clin Rev Allergy Immunol 2011;41(2):151-162. https://doi. org/10.1007/s12016-010-8222-2 11. Ludici M, Quartier P, Terrier B, Mouton L, Guillevin L, Puéchal X. Childhood-onset granulomatosis with polyangiitis and microscopic polyangiitis: Systematic review and meta-analysis. Orphan J Rare Dis 2016;11:141. https://doi.org/10.1186/s13023-0160523-y 12. Mehorotra AK, Gupta PR, Khublani TK, Anupam, Soni S, Feroz A. Isolated pauci-immune pulmonary capillaritis. Lung India 2015;32(1):56-59. https://doi. org/10.4103/0970-2113.148453 13. Castellazzi L, Patria MF, Frati G, Esposito AA, Esposito S. Idiopathic pulmonary haemosiderosis in paediatric patients: How to make an early diagnosis. Ital J Paediatr 2016;42(1):86. https://doi.org/10.1186/s13052-016-0296-x 14. Taytard J, Nathan N, de Blic J, et al. New insights into pediatric idiopathic pulmonary hemosiderosis: The French RespiRare cohort. Orphan J Rare Dis 2013;8(1):161. https://doi.org/10.1186/1750-1172-8-161 15. Sigua JA, Zacharisen M. Heiner syndrome mimicking an immune deficiency. Wiscon Med J 2013;112(5):215-217.

16. Newsome BR, Morales JE. Diffuse alveolar haemorrhage. South Med J 2011;104(4):269274. https://doi.org/10.1097/SMJ.0b013e3182126d3b 17. Bakalli I, Kota L, Sala D, et al. Idiopathic pulmonary haemosiderosis – a diagnostic challenge. Ital J Paediatr 2014;40:35. https://doi.org/10.1186/1824-7288-40-35 18. Poggi V, Lo Vecchio A, Menna F, Menna G. Idiopathic pulmonary haemosiderosis - a rare cause of iron-deficiency anaemia in childhood. J Pediatr Haematol Oncol 2011;33(4):160162. https://doi.org/10.1097/MPH.0b013e3182a6df 19. Moo SP. Diffuse alveolar hemorrhage. Tuberc Res Dis 2013;74(4):151-162. https://doi. org/10.4046/trd.2013.74.4.151 20. Mark EJ, Ramirez JF. Pulmonary capillaritis and hemorrhage in patients with systemic vasculitis. Arch Pathol Lab Med 1985;109(5):413-418.

21. Susarla SC, Fan LL. Diffuse alveolar hemorrhage syndromes in children. Curr Opin Pediatr 2007;19(8):314-320. https://doi.org/10.1097/MOP.0b013e3280dd8c4a 22. Alberici F, Jayne DR. Impact of rituximab trials on the treatment of ANCA-associated vasculitis. Nephrol Dial Transplant 2014;29(6):1151-1159. https://doi.org/10.1093/ ndt/gft318

SARJ VOL. 23 NO. 3 2017

69

REVIEW 23. Kallenberg CG, Hauser T. B-cell therapy in antineutrophil cytoplasmic antibody associated vasculitis. Nephrol Dial Transplant 2015;30(1):119-122. https://doi.org/10.1093/ndt/ gfv056 24. Yates M, Watts RA, Bajema IM, et al. Eular/ERA_EDTA recommendations for the management of ANCA-associated vasculitis. Ann Rheum Dis 2016;75(9):1583-1594. https://doi.org/10.1136/annrheumdis-2016-209133 25. Solomonov A, Fruchter O, Zuckerman T, et al. Pulmonary haemorrhage: A novel mode of therapy. Respir Med 2009;103(8):1196-1200. https://doi.org/10.1016/j.rmed.2009.02.004 26. Wanko SO, Broadwater G, Folz RJ, Chao NJ. Diffuse alveolar haemorrhage: Retrospective review of clinical outcomes in allogeneic transplant recipients treated with aminocaproic acid. Biol Blood Marrow Transplant 2006;12(9):949-953. https://doi.org/10.1016/j. bbmt.2006.05.012

70 SARJ VOL. 23 NO. 3 2017

27. Rathi NK, Tanner AR, Dinh A, et al. Low-, medium- and high-dose steroids with or without aminocaproic acid in adult hematopoietic patients with diffuse alveolar haemorrhage. Bone Marrow Transplant 2015;50(3):420-426. https://doi.org/10.1038/ bmt.2014.287 28. Park JA. Diffuse alveolar haemorrhage and recombinant factor VIIa treatment in pediatric patients. Korean J Pediatr 2016;59(3):105-113. https://doi.org/10.3345/ kjp.2016.59.3.105 29. Bafaqih H, Chehab M, Almohaimeed S, et al. Pilot trial of a novel two-step therapy protocol using nebulized tranexamic acid and recombinant factor VIIa in children with intractable diffuse alveolar hemorrhage. Ann Saudi Med 2015;35(3):231-239. https://doi.org/10.5144/0256-4947.2015.231

BREATH-TAKING NEWS

A novel smartphone-based point-of-care diagnosis of cystic fibrosis Cystic fibrosis (CF) is caused by a defect in the cystic fibrosis transmembrane conductance regulator (CFTR) gene of cells in the lungs, pancreas and sweat glands. The gold standard for the diagnosis is a sweat chloride concentration of >60 mmol/L. Current diagnostic markers are expensive, and low-cost chloridometers are required for diagnostic and monitoring purposes.

A team in Pennsylvania have developed a low-cost smartphone chloridometer equipped with an ultraviolet LED to generate fluorescence signals that are captured by the smartphone camera.

This can detect chloride concentrations of 0.8 - 200 mmol/L. This test has a high sensitivity, and has undergone both analytical and clinical validation with both CF and non-CF individuals, albeit in small numbers of participants. Its clinical value will be its use as a predictive pharmacological biomarker of pulmonary improvement

and compliance in CF patients. This unique testing platform has the potential to offer real access to point-of-care diagnosis and management of CF patients in resource-limited settings in developing countries.

O Kitchin Netcare Waterfall City Hospital, Midrand, Johannesburg, South Africa 1. Zhang C, Kim JP, Creer M, Yang J, Liu Z. A smartphone-based chloridometer for point-of-care diagnosis of cystic fibrosis. Biosens Bioelectron 2017;197:164-168. https://doi.org/10.1016/j.bios.2017.05.048

S Afr Respir J 2017;23(3):71. DOI:10.7196/SARJ.2017.v23i3.180

The relationship between low vitamin D levels and asthma is still an enigma The interaction between vitamin D and clinical course in asthma pathogenesis is an active area of research presently. A previous systematic review on maternal vitamin D and childhood asthma, wheeze and eczema showed no correlation between low maternal vitamin D levels and childhood wheeze and asthma.[1] A new systematic review and meta-analysis of observational studies was performed to evaluate the prevalence of vitamin D deficiency in asthmatic children.[2] A total of 23 studies (with 13Â 160 participants) were analysed. The association with asthma incidence, asthma control, and lung function was determined. The mean vitamin D levels were lower in asthmatics v. non-asthmatics; however, there was a poor correlation between vitamin D levels and incidence of asthma, with one study showing an inverse association at 4 years, and no association with asthma severity at 8 years. There was also a poor correlation with lung function tests between low levels of vitamin D and asthma control.

There is a need for further well-randomised, controlled trials in children before we decide whether asthmatic children can benefit from vitamin D supplementation, or not.

O Kitchin Netcare Waterfall City Hospital, Midrand, Johannesburg, South Africa 1. Wei Z, Zhang J, Yu X. Maternal vitamin D status and childhood asthma, wheeze, and eczema: A systematic review and meta-analysis. Pediatr Allergy Immunol 2016;27(6):612-619. https://doi.org/10.1111/pai.12593 2. Jat K, Khairwa A. Vitamin D and asthma in children: A systematic review and meta-analysis of observational studies. Lung India 2017;34(4):355-363. https://doi. org/10.4103/0970-2113.209227

S Afr Respir J 2017;23(3):71. DOI:10.7196/SARJ.2017.v23i3.181

SARJ VOL. 23 NO. 3 2017

71

PRODUCT NEWS

Concepts and misconceptions in oral anticoagulation Anticoagulants like warfarin (vitamin K antagonists (VKAs)) still have their place, however, direct oral anticoagulants (DOACs) offer unlimited advantages. This was the take-home message by Prof. Jan Beyer-Westendorf, head of the Thrombosis Research Unit in the Centre for Vascular Medicine at the Carl Gustav Carus University Hospital in Dresden, Germany. He was in South Africa (SA) from 19 to 23 September 2016, as a guest of Bayer, stating categorically that he only uses VKAs when it is unavoidable.

Anticoagulation treatment has evolved rapidly; in ~5 years, newer agents have become available, with dramatically increased use, ‘In 2010 2011, only 4.2% of patients in the GARFIELD AF registry received DOACs. By 2014 - 2015 this figure had increased to 37%.’

DOACs offer safer alternatives to warfarin, and have been shown to facilitate patient persistence with and adherence to treatment. ‘Sideeffects, treatment failure and bleeding all affect compliance. The one-year discontinuation rate for warfarin is in the region of 50% v. only 15% with factor Xa inhibitors such as rivaroxaban, suggesting that the DOACs are better accepted for long-term treatment by both prescribers and patients.’

An adherence study in Canada suggests rivaroxaban’s once-daily dosing regimen may offer a slight advantage over twice-daily regimens.

However, it demonstrated that 25 - 30% of twice-daily prescribed DOACs are taken once daily. Regardless, dosing is a challenge. In atrial fibrillation (AF) cohorts, ~20% of patients have dose reduction owing to moderate to severe renal impairment. Prescription data suggest that 35 - 50% of DOAC prescriptions indicate insufficient dosage, so a large proportion of AF patients may be undertreated.

Real-world studies are biased by patient selection, differing designs, baseline characteristics, and treatments. Prof. Beyer-Westendorf ’s advice was to ‘Never look at effectiveness and safety alone – look rather for net clinical benefit.’ Comparing DOACs is not possible in these studies.

Collectively, however, they consistently help to prevent strokes in correctly-dosed patients. They have also shown consistent safety: ‘Compared to real-life studies with VKAs, not a single current study has shown higher bleeding rates with DOACs.’

Dispelling misconceptions

Prof. Beyer-Westendorf debunked four myths regarding DOACs: Misconception one: Stable VKA-treated patients will not benefit from a switch to a DOAC. ‘Guidelines actively discourage switching. But is there such a thing as a stable VKA-treated patient – and is there any evidence that they can safely continue warfarin treatment? The evidence suggests otherwise. One could argue that patients with a very stable international normalised ratio (INR) (defined as a time-in-therapeutic range or TTR of at least 70%, preferably 75%) may be regarded as stable VKA patients. However, none of the phase III DOAC studies, which evaluated only carefully selected patients treated in dedicated trial units, achieved an INR-TTR of 70% with warfarin. Furthermore, data from the ORBIT-AF study from the USA indicated that only a few patients are truly stable on warfarin and a past record of stability only weakly predicts future results’ Misconception two: With renal impairment, VKAs should be preferred. However, post hoc and meta-analyses of phase-III DOAC trials showed DOACs being as effective and safe, and in the case of factor Xa inhibitors, superior to VKAs. ‘While not all thromboembolic events can be prevented, one out of five clots that would have developed with warfarin in renally impaired patients will not occur if a DOAC is used.

Using a DOAC can also prevent bleeding. Pooled EINSTEIN deep vein thrombosis and pulmonary embolism data showed a 4-fold increase in major bleeding with warfarin, but not with rivaroxaban.’

Misconception three: DOACs cause more gastrointestinal (GI) bleeding than VKAs. This perception was propagated by a meta-analysis in 2013.[1] Prof. Beyer-Westendorflamented it as ‘poor science’, leading to gross overestimation of DOAC-related GI bleeding.[2] A subsequent study debunked this.[3] Furthermore, in GI bleeding the focus on bleeding rates may be misleading because the GI bleeding site, management and outcome need to be considered. Aspirin and VKA bleeds usually occur in the upper GI tract, while most DOAC bleeds occur in the lower GI tract. Most lower GI bleeds are haemorrhoid bleeds, which are easily managed compared with gastric ulcer bleeds. The mortality of different types of GI bleeding varies significantly and data indicate that DOACs may offer advantages in GI bleeding.

Misconception four: Managing major bleeding is simpler and better for VKAs v. DOACs. The DOAC AF studies showed the opposite.

VKA major bleeding is associated with increased mortality, with worse patients experiencing less bleeding on DOACs. Prothrombin complex concentrate is praised in VKA treatment, but is often ineffective in obese patients, dosed incorrectly, and associated with thromboembolic complications. It’s needed less often in DOAC patients, where it is effective. Regardless, the new agents have, or will soon have reversal options: idarucizumab for dabigatran reversal, while andexanet alfa, with anti-factor Xa activity, is an excellent alternative for rivaroxaban, apixaban and edoxaban patients. Both antidotes have yet to be registered in SA. Prof. Beyer-Westendorf recapitulated: In real-world settings, compared with warfarin, the DOACs are associated with high persistence and adherence, as well as acceptably low stroke and major bleeding rates. VKA-associated major bleeding rates are much higher than in DOAC cohorts but since we cannot prevent major bleeding in every patient, the survival benefit in cases of major bleeding with DOACs is the main advantage over VKAs. 1. Holster IL, Valkhoff VE, Kuipers EJ, Tjwa ETTL. New oral anticoagulants increase risk for gastrointestinal bleeding: A systematic review and meta-analysis. Gastroenterology 2013;145(1):105-112. https://doi.org/10.1053/j.gastro.2013.02.041 2. Beyer-Westendorf J, Pannach S. Increase of gastrointestinal bleeding with new oral anticoagulants: Problems of a meta-analysis.

Gastroenterology 2013;145(5):1162-1163. https://doi.org/10.1053/j.gastro.2013.09.041 3. Caldeira D, Barra M, Ferreira A, et al. Systematic review with meta-analysis: The risk of major gastrointestinal bleeding with nonvitamin K antagonist oral anticoagulants. Aliment Pharmacol Ther 2015;42(11-12):1239-1249. https://doi.org/10.1111/apt.13412

Approval Number L.ZA.COM.GM.11.2016.1398

72 SARJ VOL. 23 NO. 3 2017

ABSTRACTS

Abstracts of presentations at the Congress of the South African Thoracic Society in Cape Town, 22 - 25 August 2017 Childhood asthma: A clinical study in southern Nigeria A Frank-Briggs

University of Port Harcourt, Port Harcourt, Nigeria bomadatown@yahoo.com

Introduction. Childhood bronchial asthma is a major cause of chronic respiratory morbidity and mortality and its worldwide prevalence is notably on the rise.

Objectives. To determine the prevalence of asthmatic children in hospital, analyse the severity of their asthma, and identify associated disease risk factors.

Methods. A questionnaire-based, cross-sectional, hospital-based study carried out on paediatric patients attending the respiratory clinic of the University Teaching Hospital from 1 July to 31 December 2013.

Results. The hospital prevalence of asthmatic children was 1.2% (N=40), with a male to female ratio of 3.4:1. The mean (standard deviation) age at onset of asthma symptoms, was 4.24 (3.36) years. Extreme cold was found to exacerbate asthma symptoms in 72.5% (n=29) of patients, while a positive family history of asthma was the most commonly identified childhood risk factor for the development of asthma (52.5%, n=21). Thirty-five of the subjects (87.5%) had mild, intermittent asthma. More males than females (p=0.689), and more adolescents than children <10 years old (p=0.117) had persistent asthma (p=0.117). Males had a significantly higher frequency of asthma-related hospital admissions (p=0.023).

Conclusion. The low prevalence of asthma (1.2%) seen in this study, may be attributed to the hospital-based nature of the study, which does not provide an accurate representation of the prevalence of childhood asthma in the community. The male preponderance and high frequency of hospital admissions among the males is similar to the findings in other studies.

An audit of the common aeroallergens in children with asthma and allergic rhinitis in a regional hospital in KwaZulu-Natal Province, South Africa K Coopasamy,1 E I Goldstone,1,2 R Masekela1

Department of Maternal and Child Health, Nelson R Mandela School of Medicine, School of Clinical Medicine, University of KwaZulu-Natal, Durban, South Africa 2 Department of Paediatrics and Child Health, R K Khan Hospital, Durban, South Africa karunamayidd@gmail.com 1

Introduction. Asthma is the most common chronic childhood disease. It affects 10 - 20% of South African children, with allergic rhinitis (AR) being comorbid in up to 80% of children. Researchers lack data regarding the role of allergens in low- and middle-income countries, particularly in tropical regions.

Objectives. To identify the common allergens in children with asthma and/or AR. To quantify the prevalence of atopy, and draw a correlation between disease severity and atopy.

Methods. A retrospective chart review of children with asthma and AR was compiled over a 1-year period. The GINA 2015 and ARIA guidelines were used for disease severity grading. Laboratory tests included Phadiatop and fx5, as well as RAST for the evaluation of the different allergens. The Fisherâ&#x20AC;&#x2122;s exact tests and Ď&#x2021;2 tests were used to test for associations between disease severity and atopy. The Kruskal-Wallis rank test was further used to compare allergen load and disease severity.

Results. The study analysed 100 children, with a (standard deviation) age 5.5 (1.5) years; 60% of the children were male.

The most common aeroallergens were Dermatophagoides pteronyssinus and Dermatophagoides farina, with 66% of children being atopic. There was a significant correlation between AR severity and the presence of D. pteronyssinus (p=0.012), and D. farinae (p=0.013), respectively. There was also a significant correlation between the persistence of AR and the presence of D. pteronyssinus (p=0.007) and D. farinae (p=0.009). There was no association between the severity or persistence of asthma and the presence of allergens (p>0.05).

Conclusion. House dust mites were the most common aeroallergens and resulted in more severe and persistent symptoms.

Deep-vein thrombosis in a surgical intensive care unit: Prevalence and risk factors C Wilasrusmee, N Poprom

Department of Surgery, Faculty of Medicine, Ramathibodi Hospital, Mahidol University, Thailand chumpon.wil@mahidol.ac.th

Introduction. Critically ill patients are at a high risk of developing venous thromboembolism.

Objectives. The objective of this study was to determine the prevalence of, and risk factors for, lower extremity deep-vein thrombosis (DVT) among critically ill intensive care unit (ICU) patients Thailand.

Methods. Patients >15 years, who were admitted to a surgical ICU of a tertiary care hospital, were enrolled. Bilateral lower extremity compression Doppler ultrasonographic examination was performed to detect DVT within 14 days of ICU admission. Demographic data, primary disease, operative intervention, comorbidities, acute physiology and Acute Physiology and Chronic Health Evaluation health evaluation (APACHE) II score, and the length of ICU stay were evaluated for association with the presence of DVT. Results. Among the 190 first-time-admitted ICU patients with a mean (standard deviation) (interquartile range) APACHE II score of 9.2 (6.0) (0 - 29), 20 patients (10.5%) had DVT. Thromboprophylaxis was not administered to any patients. The only independent and significant risk factor for DVT, was a longer ICU stay. Age,

SARJ VOL. 23 NO. 3 2017

73

ABSTRACTS sex, APACHE II score, presence of comorbidities, and operative intervention were not associated with the presence of DVT. Conclusion. The prevalence of DVT in critically-ill patients in a Thai surgical ICU, was ~10.5%. Further research is needed to evaluate the risks and benefits of venous thromboprophylaxis in Thai patients.

The evaluation and surgical management of tracheal strictures following intubation at a thoracic surgery referral centre in South Africa S Ramghulam

Inkosi Albert Luthuli Central Hospital, Durban, South Africa docsherwinr@gmail.com

Introduction. The Department of Cardiothoracic Surgery at Inkosi Albert Luthuli Central Hospital (IALCH) is the sole provider of cardiothoracic surgical care for the KwaZulu-Natal Province and the eastern seaboard of South Africa, encompassing ~14 million patients. Investigating the aetiology, prevalence, and most importantly the outcomes of surgical intervention, will assist us to identify risk factors for surgical complications and improve patient outcomes.

Objectives. The surgical treatment of tracheal stenosis following endotracheal intubation, or tracheostomy, is well-described in the developed world. We present our surgical experience, and highlight nuances in the diagnosis and management of tracheal stenosis, in subSaharan Africa.

Methods. We reviewed the clinical records and archived images of patients who had undergone tracheal resection and reconstruction, for post-intubation tracheal stenosis, between 1 July 2003 and 31 July 2014, in the Department of Cardiothoracic Surgery at IALCH. Results. During the study period, 42 patients underwent tracheal resection. We evaluated the preoperative bronchoscopic characteristics of the tracheal stricture in all patients, and computed tomography (CT) was used as an adjunct in 28 (66%) patients. Most of the patients (85.7%, n=36) underwent surgery via a cervical approach, and 14.3% (n=6) of patients via a right thoracotomy.

There was no early mortality, but surgery was complicated by vocal cord palsy in 4 cases, restenosis in 2 cases, infection in 1 case, and paraparesis in 1 case.

Conclusion. Tracheal resection for the treatment of post-intubation tracheal stenosis can be performed safely, with minimal complications, in the developing world. A preoperative evaluation of the stricture through a combination of bronchoscopy and CT scans remains the most accurate technique to plan tracheal resection and reconstruction, while most lesions are removed via a cervical approach.

The epidemiology of RSV bronchiolitis – a retrospective review from Steve Biko Academic Hospital: 2013 - 2016 C X Dearden,1 A C Jeevarathnum,1 J Havinga,2 R J Green1

Department of Paediatrics and Child Health, Faculty of Health Sciences, University of Pretoria and Steve Biko Academic Hospital, Pretoria, South Africa 2 Department of Virology, University of Pretoria and Steve Biko Academic Hospital, Pretoria, South Africa xandrevdh@gmail.com 1

74 SARJ VOL. 23 NO. 3 2017

Introduction. Respiratory syncytial virus (RSV) causes seasonal morbidity and is an enormous burden on health systems across the world. High-risk groups, including premature infants and infants with underlying medical conditions, present significantly higher morbidity and mortality rates.

Objectives. The aim of this study is to describe the incidence of RSV bronchiolitis amongst patients ≤24 months of age who presented to a tertiary institution with a diagnosed viral bronchiolitis over a 4-year period.

Methods. A retrospective, chart-based analysis of laboratoryconfirmed RSV cases was conducted in children (≤24 months) who presented to the Steve Biko Academic Hospital between January 2013 and December 2016.

Results. A total of 1 127 nasopharyngeal aspirates were collected, with RSV isolated from 162. The median age (interquartile range) was 3.7 (9 days - 2 years) months, with the majority (63.4%) being <6 months old. Of the patients with known HIV status, only two were HIV-positive. A total of 49 (30.2%) patients with RSV were admitted to the intensive care unit (ICU), of whom 34 (69.4%) were <6 months old. There were 8 (4.9%) confirmed deaths. Premature birth, followed by cardiac lesions, were the most common risk factors for RSV bronchiolitis, a condition occurring predominantly during autumn and winter.

Conclusion. RSV is commonly detected among infants who are admitted for bronchiolitis. Significant risk factors were premature birth, age <6 months, and congenital cardiac disease. Gender and HIV status did not appear to increase the risk of RSV bronchiolitis.

Young babies, especially premature infants with RSV bronchiolitis, are at considerable risk and likely to be admitted to the ICU. The case fatality rate was 4.9% in children <6 months old, with premature birth being the main risk factors.

The use of airway clearance therapy in children hospitalised with acute lower respiratory tract infections in a South African tertiary hospital L Corten,1 J Jelsma,1 B M Morrow2

Department of Health and Rehabilitation Sciences (Physiotherapy), University of Cape Town, South Africa 2 Department of Paediatrics and Child Health, University of Cape Town, South Africa l.corten@uct.ac.za 1

Introduction. Airway clearance therapy (ACT) is often advocated for the clinical management of children with lower respiratory tract infection (LRTI). However, there is minimal documentation on the prescription, frequency, nature and adverse events of ACT, with no South African data.

Objectives. To describe the characteristics, management and outcomes of children with acute LRTI who were admitted to a tertiary paediatric hospital, and to describe subsequent ACT prescription and practice.

Methods. A retrospective descriptive study was conducted using routinely collected data over a 6-month period.

Results. A total of 1 357 folders were screened, of which 1 208 were eligible for inclusion, accounting for 1 440 hospitalisations.

ABSTRACTS The median (interquartile range (IQR)) age of children included in the study was 7.6 (2.8 - 19.0) months. The most common primary diagnoses were bronchiolitis (46.0%), pneumonia (36.5%), unspecified LRTI (10.0%), and other diseases (0.2%). Comorbid conditions were common and nosocomial LRTI was considered likely in 106 included patients (7.4%). The median (IQR) duration of hospitalisation was 2.3 (1.5 - 5.0) days. The mortality rate was 0.7% (n=10). Children with presumed nosocomial infections (n=6) or pneumonia (n=3) were at higher risk of death. ACT was performed by physiotherapists during 83 hospitalisations (5.8%) - most commonly for children who acquired nosocomial LRTI in hospital, followed by those admitted with communityacquired pneumonia; and least commonly for children with bronchiolitis. Manual chest wall vibrations (83.1%), modified postural drainage (55.4%), and percussions/clapping (38.6%) were the most common ACTs applied. Treatments were performed daily or bi-daily for a median (IQR) period of 3.0 (1.0 - 6.0) days.

Transient desaturation occurred in 6 children during treatment and lung collapse in 1 child an hour post treatment. No other adverse events were reported.

Conclusion. ACT appears to be safe in this population of children with LRTI. Standard practice conforms to national guidelines for bronchiolitis management, which do not support the use of ACT in this population. Research is warranted to determine the indications and contraindications of ACT, and to establish safe and effective ACT procedures.

The impact of assisted autogenic drainage in children with cystic fibrosis – a pilot study L Corten,1 J Jelsma,1 B M Morrow2

epartment of Health and Rehabilitation Sciences (Physiotherapy), University D of Cape Town, South Africa 2 Department of Paediatrics and Child Health, University of Cape Town, South Africa l.corten@uct.ac.za 1

Introduction. Airway clearance therapy (ACT) is standard physiotherapy practice to promote pulmonary mucus clearance in children with cystic fibrosis (CF). The relative utility and superiority of individual ACTs in children <8 years old is unknown.

Objectives. To evaluate the feasibility of conducting a home-based, randomised, controlled trial to determine the effects of assisted autogenic drainage (AAD) compared with standard ACT, including conventional chest physiotherapy and breathing exercises in children with CF. Methods. Children with CF, aged 1 - 8 years, were randomly assigned into intervention (AAD), or control (standard ACT) groups. Parents were taught these techniques and instructed to perform them bi-daily for a year. An intention-to-treat analysis was conducted and outcome assessors were blinded to allocation. Primary outcome measures were number of hospitalisations and exacerbations during 1 year. Secondary outcome measures included spirometry, health-related quality of life, CF clinical score, preference, neurodevelopmental screening, body mass index, weight and height for age, adherence, and mortality.

Results. Of the 36 children screened for inclusion, 16 were enrolled in

the study. The median interquartile range age of the study participants was 5.75 (4.27 - 6.28) years. There were no significant inter-group differences for any of the outcome measures; however, the number of exacerbations and the number of days on antibiotic therapy during 1 year revealed medium (Cohen’s d=0.55) and small-to-medium effect sizes (Cohen’s d=0.48), respectively, in favour of the intervention group.

Although no significant changes were found within either group, there were trends towards improvement in CF subjective and total clinical scores, as well as health-related quality of life (HRQOL) with large effect sizes (Cohen’s d=1.07, 0.87, 0.86, respectively) in the intervention group.

Adherence to ACT was poor and no participant in the intervention group performed AAD solely as per the pre-set methodology.

Conclusion. The results of this pilot study are promising in terms of the observed reductions in exacerbations and antibiotic usage, the lack of adverse events, as well as the improved CF and HRQOL scores in children receiving AAD. The lack of adherence to ACT identifies the need for research to optimise compliance with essential CF management strategies. Although research is warranted to determine the usefulness of AAD in this population, the current research model should be reconsidered prior to implementation, particularly the selection of appropriate outcome measures and methods to improve adherence to therapy.

The aetiology of malignant pleural effusion in South Africa J Shaw,1 C Koegelenberg,1 S Bennji,2 E M Irusen,1 P T Schubert3

Division of Pulmonology, Department of Medicine, Faculty of Medicine and Health Sciences, Stellenbosch University, South Africa 2 Division of General Internal Medicine, Department of Medicine, Faculty of Medicine and Health Sciences, Stellenbosch University, South Africa 3 Division of Anatomical Pathology, Tygerberg Hospital, National Health Laboratory Service, Faculty of Medicine and Health Sciences, Stellenbosch University, South Africa janeshaw@sun.ac.za 1

Introduction. Malignant pleural effusion (MPE) represents a very common cause of pleural exudates, and is one of the most challenging pleural disorders to manage. This could be attributed to the paucity of high-quality experimental evidence, and inconsistent practice worldwide. South Africa (SA) currently has no data regarding the aetiology of MPE. Objectives. The primary aim of this study was to identify the most common malignancies that cause MPE in the demographic served by our hospital, and specifically, the relative contribution of mesothelioma. The secondary aim was to evaluate the efficacy of chemical pleurodesis in these patients.

Methods. This was a 3-year retrospective analytical study on records of patients with MPE, who were treated by the hospital’s combined oncology and pulmonology services, from January 2013 to December 2015.

Results. A total of 194 patients with MPE were included in the analysis.

The aetiology of the MPE was: lung cancer in 139 patients (71.6%); breast cancer in 26 patients (13.4%); mesothelioma in 21 patients (10.8%); and MPE with unknown primary in 5 patients (2.6%).

Among patients with lung cancer, 63.3% had adenocarcinoma. Talc pleurodesis was performed in 81 patients, of which 25 were followed

SARJ VOL. 23 NO. 3 2017

75

ABSTRACTS up for up to 3 months. The success rate for talc pleurodesis among these patients was 88%.

Conclusion. In our hospital, the main cause of MPE was lung cancer, followed by breast cancer and mesothelioma. Unknown primary still represents a small proportion of MPE patients. Chemical pleurodesis is a viable palliative measure for MPE in this population.

A new technique for repair of a dislocated sternoclavicular joint using a sternal tension cable system J Janson

Division of Cardiothoracic Surgery, Stellenbosch University, Tygerberg Hospital, Cape Town, South Africa jansonj@telkomsa.net

Introduction. An unstable anterior or posterior sternoclavicular joint (SCJ) dislocation can cause severe morbidity with poor shoulder movement and strength. These dislocations need to be repaired, which can be challenging. Many different procedures have been described to obtain a stable joint fixation, with varying results.

Objectives. To report on a new technique for repairing a SCJ dislocation, by using a figure-of-eight sternal cable system.

Methods. Description of the surgical technique and a review of 15 patients with SCJ dislocations treated with a figure-of-eight sternal tension cable.

Results. There were eight anterior and seven posterior dislocations.

Ten dislocations were on the left, and five on the right. A stable reduction and fixation could be obtained on all patients. The patients were followed up for 1 year and had good function and stability of the SCJ. Four patients experienced minor episodic discomfort over the joint, but the discomfort did not impair their work or daily activities.

Conclusion. This procedure was relatively simple and reproducible, and resulted in stable and functional sternoclavicular joints.

Epidermal growth factor receptor and anaplastic lymphoma kinase mutations detected by immunohistochemistry in lung adenocarcinoma in patients from Johannesburg N Vorajee,1,2 J Murray,1,3 J I Phillips1,4

National Institute for Occupational Health, National Health Laboratory Service Department of Anatomical Pathology, School of Pathology, Faculty of Health Sciences, University of the Witwatersrand, Johannesburg, South Africa 3 School of Public Health, University of the Witwatersrand, Johannesburg, South Africa 4 Faculty of Health Sciences, University of Johannesburg, South Africa naseema.vorajee@nhls.ac.za 1

2

Introduction. Lung cancer is the leading cause of cancer mortality.

Tyrosine kinase inhibitors that target the epidermal growth factor receptor (EGFR) and anaplastic lymphoma kinase (ALK) mutations are effective in specific lung cancer subtypes. EGFR mutations are present in 18 - 25% of lung cancers and are commonly represented by the E746_A750 deletion on exon 19 and the L858R point mutation on exon 21. The ALK translocation which is present in

76 SARJ VOL. 23 NO. 3 2017

2 - 7% of lung cancers involves the EML4-ALK fusion gene product.

Immunohistochemistry (IHC) to detect these mutations has the potential of being used as an initial screening tool to facilitate shorter diagnostic time and fast track treatment options.

Objectives. To test EGFR and ALK mutational status in lung cancer using IHC. Methods. Biopsies of patients from Charlotte Maxeke Johannesburg Academic Hospital and Helen Joseph Hospital were sent to the National Institute for Occupational Health. Biopsies taken between 1 January 2008 and 30 June 2014 were reviewed and IHC analysis performed on 111 of these samples. Mutationspecific antibodies, EGFR SP111 and EGFR SP125 and the highaffinity ALK D5F3 antibody were used.

Results. Most patients were black males (61%) and the median (standard deviation) age was 58 (11.5) years. EGFR IHC was positive in 10 (9%) patients. There was no statistically significant association between age, sex, smoking history, and EGFR IHC status. ALK IHC was positive in 8 (7%) patients. ALK IHC status was statistically significantly associated with race (p=0.03), and age (p=0.081).

Conclusion. The low EGFR IHC rate may be due to antigen degradation, intratumoural heterogeneity, and/or low EGFR IHC sensitivity. The percentage of ALK IHC-positive biopsies of 7% was at the upper limit of the 2 - 7% described in international literature, raising the possibility that the ALK mutation in South African black patients may be particularly high. We propose an evidence-based diagnostic algorithm, using both EGFR and ALK IHC as rapid initial screening tests to facilitate targeted therapy.

Autologous pleural blood patch as a remedy for postoperative air leaks M G Ismail

Stellenbosch University, Tygerberg Hospital, Cape Town, South Africa ismlzane@gmail.com

Introduction. Although relatively uncommon, prolonged postoperative air leaks after lung resection or complicated thoracic surgery can be challenging. Current methods include closed lowpressure suction systems, Heimlich valves, or to ultimately redo the surgery. The instillation of autologous blood into the pleural space to occlude air leaks in selected patients has been reported in literature, but is not generally acknowledged or practised. We report our experience with a small case series, successfully employing the use of the autologous blood patch to resolve postoperative air leaks.

Objectives. To assess the efficacy of autologous blood instillation into the pleural space to resolve postoperative air leaks.

Methods. A case series of 6 patients between 2013 and 2017, who underwent thoracic surgical procedures, which were complicated by prolonged air leaks postoperatively. Autologous blood was introduced into patientsâ&#x20AC;&#x2122; affected pleural spaces 5 - 10 days postoperatively, and their chest drains were monitored for resolution.

Results. All 6 patients had resolution of their air leaks within 6 hours of autologous blood instillation. One patient required a second administration, with subsequent resolution.

Conclusion. Modern surgical techniques and adherence to good surgical principles have made postoperative air leaks very

ABSTRACTS uncommon. Even so, postoperative air leaks increase patient discomfort and morbidity, and pleural autologous blood patches appear to be a feasible method to eliminate air leaks.

A case of catamenial dyspnoea E M Taban1, A K Graham,2 N Vorajee3

Department of Pulmonology, Charlotte Maxeke Johannesburg Academic Hospital, University of the Witwatersrand, Johannesburg, South Africa 2 Department of Pulmonology, Helen Joseph Academic Hospital, University of the Witwatersrand, Johannesburg, South Africa 3 Department of Histopathology, National Institute for Occupational Health, National Health Laboratory Service, University of the Witwatersrand, Johannesburg, South Africa malish46@hotmail.com

Diagnosis and outcome of primary solid thoracic tumours in a high tuberculosis-prevalent setting C Smit, P Goussard, A van Zyl

Tygerberg Academic Hospital, Cape Town, South Africa carinesmit85@gmail.com

1

Introduction. The true prevalence of endometriosis in African women is unknown due to lack of adequate diagnostic capacity.

Historically it was thought to be a rare disease in African women, however more recent data now shows endometriosis may in fact be commoner in African indigenous women. Endometriosis typically involves the pelvis, particularly the ovaries, cul-de-sac, broad ligament, and uterosacral ligaments. However, it may occur remotely with unusual manifestations. Rare examples include thoracic endometrial syndrome, which includes catamenial pneumothorax, catamenial haemothorax, catamenial haemoptysis, and pulmonology nodules. These syndromes represent diagnostic and therapeutic dilemmas, particularly in areas where limited pleural biopsies are performed.

Methods. An unmarried, 27-year-old, nulligravida presented to our sister obstetrics and gynaecology hospital with a presumptive diagnosis of Meigs’ syndrome. Her presenting complaint was that of primary infertility. On enquiry, she reported mild abdominal distension during menstrual periods and dyspnoea, worse in the supine position and with associated chest discomfort. She was examined on the third day of her menses and displayed features consistent with a large right-side pleural effusion. She had mild pelvic ascites and a bulky ovary. A chest X-ray (CXR) showed a white out of the right hemithorax. Serum CA-125 was elevated at a level of 124 units/mL (reference range 0 - 35 units/mL). Blood-stained fluid was aspirated during diagnostic thoracentesis. The patient presented to our clinic 6 weeks later with a CXR that showed a moderate, small-sized pleural effusion. A computed tomography scan revealed a simple large right pleural effusion, with no pleural nodules or thickening and normal lung parenchyma. An intercostal drain was inserted a week later to relieve tension which had subsequently developed.

Shortly thereafter, a diagnostic video-assisted thoracoscopic surgery pleural biopsy was performed.

Results. The histology revealed benign endometrial tissue.

Mechanical pleurodesis was unsuccessful with a residual hydropneumothorax. She was placed on hormonal replacement therapy (danazol). A repeat pleurodesis is planned for the future.

Conclusion. Our case emphasises the importance of considering a broad differential diagnosis for females presenting with pleural effusion and pelvic pathology. Of particular importance in our patient is the long-term preservation of fertility together with resolution of thoracic disease.

Introduction. Primary solid thoracic tumours (PSTTs) are rare, with diverse pathological spectrums, varying prognoses and survival rates.

Early recognition, diagnosis and treatment is key to good outcome.

Objectives. We described the incidence, diagnostic challenges and outcomes of PSTT, determined patient demographics, histological spectrums and treatment methods.

Methods. A 32-year retrospective review, including children <16 years old, treated for malignant thoracic masses at Tygerberg Children’s Hospital (TCH), from 1983 to 2015. Pulmonary metastases, benign tumours and cystic lesions were excluded.

Results. We documented 59 patients (22 with PSST), presenting with thoracic masses in TCH – a high HIV and tuberculosis (TB)prevalent tertiary hospital. The incidence of PSTT was 0.09 per 100 000 children per year, with a mean (standard deviation) age of 4.74 (4.22) years old and a 55% female predominance. Complaints and symptomatology were nonspecific, and in 36% of cases a diagnosis was delayed due to initial incorrect treatment for TB (23%) or pneumonia (13%). Chest X-rays (CXRs) suggested pulmonary abnormalities in all cases, but accurate diagnoses required invasive testing. The histological spectrum included 6 neuroblastomas (27%), 3 rhabdomyosarcomas (13%), 3 ganglioneuroblastomas (13%), 2 each of pleuropulmonary blastoma, Kaposi sarcoma and Ewing’s sarcomas (9%), 1 each of infantile fibrosarcoma, myoepithelioma, undifferentiated and osteogenic sarcoma (5%). At diagnosis, 69% were at early stage, and 31% at advanced disease stage. Management included chemotherapy (82%), radiotherapy (23%), and surgery (64%), with an overall survival rate of 64%. The outcome was better with surgical (83%) compared with non-surgical management (57%).

Conclusion. In this first study reporting on PSTT from a middleincome country, we identified numerous diagnostic challenges.

However, despite these, the outcomes of children with PSTT remains comparable between under-developed and highly-developed countries.

As these patients present with nonspecific symptomatology, children with pneumonia and TB who are not responsive to treatment need further evaluation to exclude PSTT. All children with PSTT had abnormal, but not diagnostic CXRs and diagnoses were confirmed by invasive testing.

Establishment and evaluation of a smoking cessation clinic in South Africa G Tadzimirwa,1 C Day,1 C Cooper,1 A Esmail,2 M Kamkuemah,3 K Dheda,2 R van Zyl-Smit2 Groote Schuur Hospital and University of Cape Town, South Africa University of Cape Town Lung Institute and Groote Schuur hospital, South Africa 3 Division of Public Health Medicine, School of Public Health and Family Medicine, Faculty of Health Sciences, University of Cape Town gtadzimirwa@gmail.com 1

2

SARJ VOL. 23 NO. 3 2017

77

ABSTRACTS Introduction. There are an estimated 7 million smokers in South Africa (SA), with high smoking-related mortality. SA is ranked as the second most stressful country to live in globally. A dedicated smoking cessation clinic was established at Groote Schuur Hospital in 2014. This facility was the first in SA to provide a clinical service and training centre.

Objectives. To motivate the health authorities to fund psychological support services, counselling, and access to nicotine replacement therapy, we created profiles for all patients with the goal to quit smoking at this facility.

Methods. Data on all patients who visited the clinic between 2014 and 2016 were captured. These included demographics, smoking history, nicotine dependence, CO levels and depression scores. All patients provided consent for data collection and audit. The UCT Faculty of Health Sciences Research Ethics Committee and hospital administration approved the review.

Results. Over a 2.5-year-period, 97 patients were seen. The mean (standard deviation) age of patients was 51.1 (10.9) years, with 59.8% male patients. The median (interquartile range (IQR)) age of smoking onset was 16 (8 - 28) years, with a median (IQR) cigarette consumption of 18 (2 - 80) per day. Men smoked more than women at 21 v. 14 cigarettes per day (p=0.002), resulting in total packyears smoked: 34 v. 22 (p=0.001). The level of nicotine dependence was moderate: the mean Fagerström test score was 5.3: men 6 v. women 5 (p=0.06). Half of the patients had a Fagerström score ≥6, and 22% ≥8.

The median (IQR) PHQ-9 depression score was 8 (4 - 11), with 49% of patients displaying symptoms of at least minor depression (score ≥10), similar in both men and women. At baseline, >60% had high (≥2/4) Wisconsin Smoking Withdrawal Scale anxiety and anger scores.

Conclusion. In this group seeking help to quit smoking, moderate levels of nicotine addiction were observed. Additionally, moderate depression and anxiety symptoms co-existed. These data support the need for pharmacotherapy in some patients, but additional intensive psychological support is urgently required.

Epidemiology, risk factors and outcome for fungal infection in a paediatric intensive care unit S T Hlope

Inkosi Albert Luthuli Central Hospital, Durban, South Africa sbehlophe@gmail.com

Introduction. Fungal infections and colonisation are common in hospital paediatric intensive care units (PICUs). Systemic antifungal agents have, however, been used successfully for the prevention of invasive fungal infections such as candidiasis.

Objectives. The aim of the study was to determine the incidence and identify the predictors of fungal infections. Furthermore, we describe the in-hospital mortality rate and how it is affected by fungal infections.

Methods. A retrospective case-control study conducted at the Inkosi Albert Luthuli Central Hospital (IALCH) PICU, between January 2015 and December 2016. Cases included confirmed invasive fungal infection (positive fungal blood cultures), as well as probable infections (elevated (1,3)-β-D-glucan) and/or urine/endotracheal aspirate).

Controls comprised negative fungal cultures, negative bacterial cultures, and positive, combined bacterial and fungal cultures.

78 SARJ VOL. 23 NO. 3 2017

Results. Invasive fungal infections were found in 19 of 1 029 patients with an incidence rate of 1.85 per 100 admissions. Predisposing risk factors were present in the majority of patients with central venous catheters (84%), urinary catheters (100%), previous surgeries (63%), total parenteral nutrition (63%) and previous broad-spectrum antibiotics (vancomycin; 47%). The in-hospital mortality rate was 16.5% (170/1 029), in which 36.8% (7/19) in the cases v. 16.1% (163/1 010) in controls. The average length of stay for the 19 cases was 20.5 days, compared with 8.3 days for the 19 who were culture-negative.

Conclusion. Fungal infections contribute to increased mortality and morbidity in the PICU at IALCH. Risk factors identified include catheter placements, previous surgery and use of parenteral nutrition and broad spectrum antibiotics.

Is adrenal suppression in asthmatic children reversible?

E W Zöllner

Department of Paediatrics and Child Health, Stellenbosch University, South Africa zollner@sun.ac.za

Introduction. Six hypocortisolaemic asthmatic children on physiological-dose steroid therapy were identified during a previous study.

Objectives. To establish whether hypothalamic-pituitary-adrenal axis suppression (HPAS) could be reversed in hypocortisolaemic asthmatic children treated with steroids, without jeopardising asthma control.

Methods. On diagnosis, six hypocortisolaemic asthmatic children were started on hydrocortisone (HC). Asthma treatment was modified by introducing steroid-sparing medications. Serum cortisol and repeat overnight metyrapone tests (ONMTPTs) were done until HPAS had recovered in all patients. A retrospective folder review was performed and the following extracted: body mass index (BMI) standard deviation score (SDS), asthma score, forced expiratory volume (FEV1), adherence, daily steroid type and dose, treatment modification, serum cortisol, final ONMTPT, and time taken for normalisation.

Results. Median serum cortisol recovered from 50 - 311 nmol after median (interquartile range (IQR)) period of 0.9 (0.4 - 2.2) years.

The ONMTPT had normalised by median (IQR) 3.3. (7 - 7.1) years.

Steroid load decreased from median 9.2 - 5.0 mg/m2/day (HC equivalent), while asthma score improved from median 1.42 to 0.85.

The number of prednisone courses decreased from median (IQR) 5 (0 - 7) to 3. FEV1 before and after intervention were 79% and 82% (median values), respectively. The median BMI SDS decreased from -0.08 to -0.16. Poor adherence to therapy was noted in four children, subsequently affecting modification and recovery time of the HPAS. Inhaled corticosteroid dose could be reduced in all but one patient, who was not adhering to therapy. Nasal steroids were discontinued in one patient and reduced in another, but as a group, their dosage increased. Steroid-sparing medication included salmeterol, formoterol, montelukast and long-acting theophyllines.

Conclusion. Hypocortisolaemia developed while asthmatic children received physiological-dose steroid therapy. By reducing steroid load by 40%, and supplementing therapy with steroid-sparing medication, hypocortisolaemia and HPAS were reversed, while asthma control

ABSTRACTS improved. Poor adherence to therapy may have retarded axis recovery, while BMI increase did not affect it.

Stool culture has limited diagnostic value in children with suspected pulmonary tuberculosis

E Walters,1 A M Demers,1 M M van der Zalm,1 A Whitelaw,2 M Palmer,1 C Bosch,1 H R Draper,1 R P Gie,1 A C Hesseling1 esmond Tutu Tuberculosis Centre, Department of Paediatrics and Child D Health, Stellenbosch University, South Africa 2 Department of Medical Microbiology, National Health Laboratory Service, Tygerberg Hospital, Stellenbosch University, South Africa ewal@sun.ac.za 1

Introduction. Bacteriological confirmation of Mycobacterium tuberculosis is infrequently established in young children with tuberculosis (TB); specimen collection is resource-intensive and respiratory secretions are paucibacillary, limiting the sensitivity of available diagnostic tests. Although molecular tests are becoming increasingly available globally, mycobacterial culture remains the gold standard for diagnosis and determination of drug susceptibility, and is more sensitive than molecular methods for paucibacillary TB. Objectives. We evaluated stool cultures as an alternative to respiratory specimens for the diagnosis of suspected intrathoracic TB. Methods. Children <13 years of age who presented with suspected intrathoracic TB were enrolled from Tygerberg and Karl Bremer hospitals in Cape Town, South Africa. The culture of one stool specimen was compared with Xpert MTB/RIF results and cultures of up to four respiratory specimens. Stool specimens were homogenised with phosphate-buffered saline solution. Stool and respiratory specimens were digested and decontaminated with 1.25% N-acetyl-L-cysteine/NaOH, followed by concentrated fluorescent smear microscopy, Xpert MTB/ RIF, and liquid culture. TB diagnoses were confirmed and classified using international consensus case definitions.

Results. A cohort of 188 children with a median (interquartile range) age of 14.4 (7.2 - 25.6) months; 15.4% were HIV-infected.

Cultures were contaminated in 78/188 (41.5%) stool specimens. Of the 110 children with evaluable results, stool cultures detected 7/38 (18.4%) children with confirmed TB, and 7/90 (7.8%) children were initiated on TB treatment. The sensitivity and specificity of stool cultures, compared with culture and Xpert MTB/RIF of 4 respiratory specimens were 28.6% (95% confidence interval (CI) 11.3 - 52.2), and 98.9% (95% CI 93.9 - 100.0), respectively.

Conclusion. Stool culture should not be recommended for the diagnosis of intrathoracic TB in children, until laboratory protocols are developed to reduce contamination and validate its true diagnostic value.

An approach to congenital lung disease L Sidali

University of KwaZulu-Natal, Durban, South Africa kiddo83sidali@gmail.com

Introduction. Congenital lung malformations are rare and vary widely in their clinical presentation. The incidence of congenital lung disease is between 1 in 25 000 and 1 in 35 000 live births. Despite

this low prevalence, the disorders lead to considerable morbidity and mortality when diagnosis is delayed. In addition, failure to recognise a malformation may lead to inappropriate intervention. There have also been controversies regarding the correct nomenclature regarding these anomalies. We describe our experience regarding these entities, in Durban.

Objectives. To describe our experience in Durban in order to provide a concise approach to congenital lung anomalies.

Methods. Case series literature review.

Results. Describe our Durban experience.

Conclusion. An approach for each lung anomaly is proposed.

Management options, including surgery, are discussed.

The detection of respiratory viruses in South African children with suspected pulmonary tuberculosis M M van der Zalm,1 E Walters,1 A M Demers,1 M Palmer,1 M Claassen,2 G van Zyl,2 A C Hesseling1

Department of Paediatrics and Child Health, Faculty of Medicine and Health Sciences, Stellenbosch University, South Africa 2 Department of Medical Virology, Faculty of Medicine and Health Sciences, Stellenbosch University, South Africa mariekevdzalm@sun.ac.za 1

Introduction. Tuberculosis (TB) and acute pneumonia are major causes of morbidity and mortality in children globally. Although the importance of viral and bacterial infections in the aetiology of respiratory illness is well established, key questions regarding the prevalence of specific pathogens, and the association between viruses and TB, remain unanswered.

Objectives. To investigate the prevalence of respiratory viruses in children with suspected pulmonary TB (PTB).

Methods. Analyses were nested in a prospective hospital-based cohort study in children aged 0 - 14 years, routinely investigated for suspected PTB in Cape Town, South Africa (SA). At enrolment, investigations included collecting at least three respiratory samples for smear microscopy, Xpert MTB/RIF, and liquid culture. Nasopharyngeal aspirates were collected for viral respiratory investigation using a commercially available multiplex PCR (Anyplex II, RV16, Seegene, South Korea), including 16 viruses. Children started on TB treatment were classified as cases, while other children were classified as symptomatic controls.

Results. A total of 73 children with a median (interquartile range) age of 22 (10 - 48) months were enrolled; 41/73 (56.2%) were male and 13/73 (18.6%) were HIV-infected. TB treatment was initiated in 42/73 (57.5%), and 19/42 (45.2%) were bacteriologically confirmed (on Xpert or culture). In 70/73 (95.9%) children, one or more viruses were detected: 39/42 (92.9%) were cases and 31/31 (100%) controls.

Human rhinovirus (HRV), the most prevalent virus, was detected in 53/73 (72.6%) children: 35/42 (83.3%) cases and in 18/31 (58.1%) of the controls; this difference was significant (odds ratio 3.61, 95% confidence interval (1.23 - 10.64; p=0.02). Adenovirus was the second most prevalent virus, detected in 41/73 (56.2%) children: 24/42 (57.1%) cases and 17/31 (54.8%) controls.

Conclusion. Respiratory viruses were frequently detected in children with suspected PTB in SA. HRV and adenovirus were the

SARJ VOL. 23 NO. 3 2017

79

ABSTRACTS most common viruses, with HRV being detected more frequently in children diagnosed with TB, suggesting that HRV infections may unmask underlying airway diseases due to TB. More studies are needed to understand the role of respiratory viruses in children with suspected PTB.

A clinical study report and evaluation of the ability of Strannik Virtual Scanning to screen the health of a randomly selected cohort of 50 patients G W Ewing,1 R Mohanlall,2 J K Adam3

Mimex Montague Healthcare Limited, Nottingham, England South African External Counterpulsation Clinic, Umhlanga, Durban, South Africa 3 Institutional Research Ethics Committee, Durban University of Technology, South Africa adamjk@dut.ac.za 1

2

Introduction. It is increasingly being recognised that political and financial factors limit the use of contemporary biomedicine – a challenge that affects the characterisation and treatment of serious medical conditions. The current diagnostic paradigm comprises numerous consultations, tests, pharmaceuticals, and other therapies, making it complex, time-consuming and expensive. Alternative technologies are clearly needed to simplify the healthcare process.

Objectives. To determine the efficacy of Strannik Virtual Scanning (SVS) compared with conventional diagnosis.

Methods. Fifty-eight patients were screened with SVS between March and September 2016. Patients reported ailments including cognitive, sleep, stress and neurological problems, as well as diabetes and cardiovascular issues.

Results. Of the 58 patients who participated in the study, 8 individuals did not confirm their known health status in a signed report, and their results were excluded from the study. Of the remaining 50 patients, SVS determined 271 medical conditions, of which 237 were known to the patients through previous relevant diagnostic procedures.

Conclusion. Strannik technology, particularly SVS, holds the potential to support doctors by making immediate and reasonably accurate assessments of patients’ health statuses, thereby minimising the need for additional tests in secondary care environments. This information can be used to identify therapeutic interventions that could improve patients’ quality of life.

Adolescent use of electronic cigarette and tobacco products in Johannesburg – results from a survey E Shaddock

University of the Witwatersrand, Johannesburg, South Africa ericashaddock@gmail.com

Introduction. Individuals who start smoking during adolescence are more likely to continue smoking into adulthood. Electronic cigarettes (e-cigarettes) have been shown to encourage progression to conventional tobacco product smoking. Advertising e-cigarettes is currently permitted and the wide variety of flavours are attractive to adolescents.

80 SARJ VOL. 23 NO. 3 2017

Objectives. To investigate the use of e-cigarettes and other tobacco products among schooled adolescents.

Methods. A self-administered questionnaire to 14 - 18-year-old high school students in two different income schools.

Results. A total of 229 students were recruited, of whom 79% were female. In the group, 11% admitted to smoking cigarettes. The median (interquartile range) age at which students started smoking was 15 (12 - 17) years old; 23% had tried e-cigarettes, 9% of whom were using the device at the time of the study. In the lower-income school 25% of students had tried e-cigarettes, compared with 20% in the higherincome school (p=0.43). In the grade 9 group (14 - 16-year-olds), 25% had tried e-cigarettes, compared with 20% in grade 11 (16 - 18-yearolds) (p=0.87). In the lower-income school, 50% of students had tried hookah, while in the higher-income school, 17% had tried hookah (p=0.0001).

Conclusion. Secondary school adolescents, especially from lowerincome schools, commonly experiment with tobacco products.

E-cigarette use is as common as conventional tobacco use. The prevalent use of hookahs is a major concern regarding lung damage and the spread of communicable diseases such as tuberculosis. The small sample size is, however, an important limitation. Another important consideration is the predominant female population, which occurred due to schools sampled. This has likely resulted in a biased view, as males generally engage in riskier behaviour, leading to an underestimation of the use of tobacco and e-cigarettes.

The outcome of thoracic surgery in HIVpositive and HIV-negative children: A comparative retrospective study J Morrison

Stellenbosch University, South Africa morrison@sun.ac.za

Introduction. South Africa has a high burden of complicated chronic lung disease in children, who often requiring thoracic surgery. Outcomes are rarely reported in the literature, especially in HIV-positive children.

Objectives. The aim of this study was to describe the indications, surgical outcomes and mortality of children undergoing thoracic surgery in a middle-income country, and to compare the outcomes of HIV-positive and -negative children.

Methods. We conducted a retrospective cohort study of children admitted to Tygerberg Children’s Hospital between 2010 and 2013, who underwent thoracic surgery. The study included all children from 29 days to 13 years of age with pulmonary disease who required thoracic surgery.

Results. A total of 107 children were included, 21% were HIV-exposed and -positive, 71% were HIV-negative. The mean age was 19 months.

In total, 45 (42%) children were <12 months and 62 (58%) were >12 months old. Indications for surgery in both groups were lymph gland compression caused by Mycobacterium tuberculosis (29%), thoracic empyema (16%), and diagnostic lung biopsies for undiagnosed diffuse lung disease (17%). There were 28 (26%) complications: most frequent were prolonged ventilation (>24 hours), persistent air leak (>24 hours), and anaemia requiring a blood transfusion post operatively. The only statistically significant difference in the complication rate was that the HIV-positive and HIV-exposed children received more blood transfusions than the HIV-negative

ABSTRACTS children (p=0.04). Of the HIV-positive children (n=14), 50% were receiving highly active antiretroviral therapy (HAART). There was no statistically significant difference between children receiving HAART and those who were not receiving HAART regarding indications for surgery, or complication rates.

Conclusion. There was no difference in thoracic surgery outcomes in HIV-positive and HIV-negative children. Further studies, with larger numbers of HIV-exposed and HIV-positive children, are required for statistically verifiable results.

An unusual presentation of angiosarcoma K Dudgeon

University of the Witwatersrand, Johannesburg, South Africa and Charlotte Maxeke, Johannesburg, Academic Hospital, South Africa kadet130@hotmail.com

Introduction. Angiosarcoma is a rare, malignant tumour of the vascular endothelium, accounting for 2% of all sarcomas. Pulmonary involvement is usually metastatic. Affected patients typically present at the late stage of the disease, and the clinical presentation is usually related to the amount of normal tissue replaced by tumour. Angiosarcomas are known to constitutively produce active vascular endothelial growth factor (VEGF), a molecule that has been implicated in the pathogenesis of hypertrophic osteoarthropathy (HOA). To date, there is a paucity of data in the literature describing a clinical association between angiosarcoma and HOA. This is a case report of a 50-year-old female who presented with HOA, weight loss, and lung nodules.

Objectives. To describe an unusual presentation of angiosarcoma.

Methods. A 50-year-old female was sent for a thoracoscopic lung biopsy, technetium-labelled bone scan, wrist X-rays, and a positron emission tomography (PET) scan. An echocardiogram was finally used to confirm the absence of an intracardiac tumour.

Results. The lung biopsy revealed angiosarcoma to be the cause of her symptoms. A technetium-labelled bone scan was consistent with HOA, as was an X-ray of her wrist. A PET scan showed areas of increased 2-deoxy-2-(18F)fluoro-D-glucose uptake in her liver, suggesting a possible hepatic primary. Echocardiography was negative for an intracardiac tumour.

Conclusion. We describe an unusual presentation of metastatic angiosarcoma, that of HOA. We postulate that the HOA is caused by increased levels of VEGF, a molecule that is constitutively produced and active in angiosarcomas. At the time of submission of this manuscript, attempts were being made to measure VEGF levels in the patient’s serum.

Rare cause of negative pressure pulmonary oedema – Presenting as “canon-ball” like lesions on imaging a case presentation A Idris

University of the Witwatersrand, Johannesburg, South Africa gulbub@gmail.com

Introduction. Negative pressure pulmonary oedema (NPPE) is a form of non-cardiogenic pulmonary oedema, that results from the

generation of high negative intrathoracic pressure in an attempt to overcome upper airway obstruction. Causes include post-extubation laryngospasm, hanging, choking, strangulation, and laryngeal tumours. We report a case of a previously well patient who developed NPPE due to strangulation.

Objectives. To describe an interesting case of NPPE. Methods. Case report.

Results. A 28-year-old male, presented with an incidence of strangulation, which occurred during a robbery. His presenting complaints included neck pain, dyspnoea and haemoptysis. He was initially seen at the trauma department, who performed an examination, neck X-rays, and cleared any injuries. The patient was transferred to respiratory medicine after a chest X-ray (CXR) revealed a bilateral alveolar filling pattern. He was hypoxic on room air with an oxygen saturation of 87%, and diffuse bilateral crackles. A full work-up for infectious causes and possible non-benign lesions, which included HIV-testing, procalcitonin (PCT), lung function tests, β-hCG, alpha-fetoprotein, thyroglobulin and anti-thyroglobulin were negative. A non-urgent computed tomography CT scan was booked; however, symptoms started to improve on the second day following admission, with a decrease in respiratory symptoms and normal saturation on room air. A repeat CXR was performed on the third day , which confirmed complete resolution of alveolar infiltrates and a subsequent diagnosis of NPPE was made.

Conclusion. NPPE is a fair diagnosis to consider in a previously healthy, young and fit patient, who experiences acute airway obstruction, as in our case. It is characterised by rapid onset and resolution of pulmonary oedema, within 12 - 48 hours. NPPE carries a good prognosis if promptly diagnosed and appropriately treated, with or without mechanical ventilation

Childhood necrotising pneumonia in a region with high burden of tuberculosis and HIV C Jacobs

University of Stellenbosch, South Africa carmen.jacobs@live.co.za

Introduction. Necrotising pneumonia (NP) is a rare complication of community-acquired pneumonia (CAP). The most common cause of NP remains Streptococcus pneumoniae.

Objectives. To describe the clinico-radiological features and outcomes of children with NP, who were admitted to Tygerberg Children’s Hospital (TCH), a children’s hospital in South Africa, a middleincome country with a high prevalence of tuberculosis and HIV. Methods. A retrospective descriptive study performed at TCH between 2004 and 2007. NP was defined as a complicated pneumonia with typical radiological findings, which was unresponsive to standard antibiotics.

Results. Thirty-eight children were identified with NP and 32 were included in the study. The median (interquartile range) age was 16.5 (10 - 33) months and 32% were HIV-positive. All cases had complicated CAP. Blood cultures were performed in 27 samples, and 8 cultures were positive, with Streptococcus spp. isolated in 7 cultures. Seven pleural aspirates (n=18) were culture-positive of which 6 cultured Staphylococcus aureus, and 1 yielded S. pneumoniae.

SARJ VOL. 23 NO. 3 2017

81

ABSTRACTS A bacteriological cause was identified in 41% of the study population.

Tuberculosis (TB) specimens were collected and 24 (25%) tested positive for Mycobacterium tuberculosis (M.tb). Two pleural aspirates were Ziehl-Neelsen stain positive, and in the TB group (n=8), 75% were HIV-positive (p=0.007). Chest X-rays were performed on 27 children and parenchymal disease was observed in 96%, with the presence of both effusion and necrosis observed in 67%. Computed tomography (CT) scan results showed pleural effusions in 91%, necrosis in 88%, and parenchymal disease in all children. Mediastinal lymphadenopathy was observed in both the TB and non-TB groups (p=0.5), and 47% of cases had surgery.

Conclusion. A bacterial cause was identified in 66% of cases.

Streptococcus spp. and S. aureus made up 41% of organisms and M.tb the rest. To date, M.tb has not been described as a cause for NP. Chest CT scans were not diagnostic of TB. Any child with NP should therefore be actively screened. A prospective study is needed in an age with new molecular techniques (GeneXpert), the availability of pneumococcal conjugate vaccine and antiretroviral therapy.

Massive haemoptysis due to Rasmussen aneurysms: Report of two cases M Wong

Chris Hani Baragwanath Academic Hospital, Johannesburg, South Africa and University of the Witwatersrand, Johannesburg, South Africa michelle.wong@wits.ac.za

Introduction. Life-threatening haemoptysis due to tuberculosis (TB) most commonly originates from the bronchial arteries. However, haemorrhage may occur less frequently in the pulmonary arterial circulation.

Objectives. Case reports of two patients presenting with massive haemoptysis.

Methods. We describe two patients who presented with massive haemoptysis due to ruptured Rasmussen aneurysms.

Results. Active pulmonary TB was the most likely underlying aetiology of the Rasmussen aneurysms in our patients. Coil embolisation was successfully performed, resulting in immediate cessation of haemoptysis. Follow-ups of both patients confirmed good outcomes.

Conclusion. Massive haemoptysis owing to Rasmussen aneurysms, is a relatively rare occurrence despite the high prevalence of TB in our population. Coil embolisation of these aneurysms, is an effective and life-saving modality of emergency management.

A case report on AIHA as a paraneoplastic phenomenon in lung cancer M Ostrofksy

University of the Witwatersrand, Johannesburg, South Africa ostrofsky.marc@gmail.com

Introduction. Autoimmune haemolytic anaemia (AIHA) is a welldescribed paraneoplastic phenomenon in lymphoproliferative disorders. However, its occurrence in solid tumours has not been well described. In a critical analysis of AIHA as a paraneoplastic phenomenon in solid tumours, it was found to occur most commonly

82 SARJ VOL. 23 NO. 3 2017

in tumours originating from the kidney, ovary, thymus gland and Kaposi sarcoma. Nine out of the 52 cases analysed, were lung cancers, all of which were non-small-cell lung cancers (NSCLC).

In solid cancers, two thirds of paraneoplastic AIHA are warm-type, while one third is cold-type.

Objectives. To describe a case report of a 58-year-old male patient who presented to Charlotte Maxeke Johannesburg Academic Hospital (CMJAH) with AIHA as paraneoplastic phenomenon of lung cancer.

Methods. This case report describes a 58-year-old male patient who presented to CMJAH in December 2016 with severe anaemia, the aetiology of which was determined to be warm anti-body AIHA. Routine chest radiography revealed left upper lobe atelectasis and a Luftsichel sign. A staging computed tomography scan showed metastatic disease. A left upper lobe endobronchial mass was discovered at bronchoscopy. An endobronchial biopsy of the mass showed NSCLC on histopathology. Further immunohistochemical staining was inconclusive; the final diagnosis was NSCLC, not otherwise specified.

Results. After exclusion of connective tissue disease and mycoplasma infection, a diagnosis of paraneoplastic AIHA was made. The patient responded poorly to systemic corticosteroid therapy, as well as chemotherapy and passed away shortly after diagnosis.

Conclusion. AIHA is a rare paraneoplastic phenomenon associated with lung cancer.

Isolated unilateral absence of left pulmonary artery: A case report and literature review A Khamkham

University of KwaZulu-Natal, Durban, South Africa almokhtar_k@yahoo.com

Introduction. Isolated agenesis of the pulmonary artery is a very rare condition that was first described by Frentzel in 1868. It has an estimated prevalence ranging from 1 in 200 000 to 1 in 300 000 adults, with a female predominance of 2:1. It frequently occurs as an isolated unilateral absence of the pulmonary artery (UAPA), but can be associated with other cardiovascular abnormalities, such as atrial or ventricular septal defect and tetralogy of Fallot. Isolated UAPA may be asymptomatic in childhood and present in adulthood, with dyspnoea as the most common symptom. It may also present with chest pain, minor or massive haemoptysis, recurrent chest infections, and symptoms of right heart failure in advanced cases.

Objectives. To present the case of a 51-year-old South African female, with UAPA. Methods. Our patient had received a ventilation perfusion scan, showing a large mismatched defect and pulmonary hypertension, presumed to be secondary to chronic thromboembolic disease.

She was therefore referred to us for further analysis. She received a chest computed tomography (CT) scan and was later treated with pulmonary arterial vasodilators and diuretics.

Results. The pulmonary angiogram from the CT scan showed typical UAPA features, confirming the diagnosis. Our patient had also developed pulmonary hypertension, which was treated with pulmonary arterial vasodilators and diuretics.

ABSTRACTS Conclusion. Diagnosing UAPA can be challenging because the presentation is nonspecific and the condition is rare. No treatment is required if patients are asymptomatic. Surgical revascularisation can be attempted in selected cases if they present in childhood. We took this opportunity to present the case because, although it has been reported in literature, no cases have been reported in South Africa.

Generational avian-induced hypersensitivity pneumonitis T M Mabaso

University of the Witwatersrand, Johannesburg, South Africa theom_m@yahoo.com

Introduction. Hypersensitivity pneumonitis (HP), is an immunologically-mediated lung disease, resulting from continued exposure of an inhaled antigen (<5 µm). It provokes a hypersensitivity reaction characterised by non-necrotising granulomatous inflammation in the distal bronchioles, lung interstitium, and alveoli of susceptible individuals. Generational cases have reported hypersensitivity pneumonitis caused by exposure to birds.

Objectives. To describe a case of avian-induced HP in a mother and daughter pair.

Methods. A 41-year-old mother and her 10-year-old daughter were diagnosed with avian-induced HP. Their diagnoses were based on their history of bird contact, presentation of dyspnoea on exertion, crepitations on auscultation, and hypoxemia on room air with saturations of 84% and 86%, respectively. Clinical processes included: pulmonary function tests, radiological imaging, bronchoalveolar lavage, lung biopsies, avian Ag-Ab tests to birds, and provocation tests.

Results. Diffuse ground glass infiltrates was observed on imaging of both subjects. The pulmonary function tests on both mother and daughter showed severe restrictive lung disease. Measurements were made for mother and daughter, respectively: forced vital capacity (FVC; 41% and 40%); forced expiratory volume (FEV1; 46% and 44%); diffusing capacity (DLCO; 67% and 15%); and total lung capacity (TLC; 60% and 36%). After removal from exposure and being on steroids, clinical improvements included: FVC (70% and 80%); FEV1 (71% and 79%); DLCO (95% and 59%); and TLC (77% and 84%), respectively. Specific avian Ag-Ab test results were negative in the daughter (4.20 mgA/L<30 mg/L regarded as negative), whereas the mother’s test was positive (54.7 mgA/L>30 mg/L regarded as positive).

Bronchoalveolar lavage of the daughter showed 39% lymphocytes, whereas the mother’s showed scattered lymphocytes. The mother’s histology report illustrated acellular chronic interstitial pneumonia with accentuated peribronchiolar parenchyma, ill-defined nonnecrotising granulomata, and multinucleate giant cells.

Conclusion. HP can present across generations in the same family and with varying chronicity.

Sardine run J Hooijer

University of the Witwatersrand, Johannesburg, South Africa jonohoya@hotmail.com

Introduction. Amoebic disease usually manifests as dysentery and is often asymptomatic. Occasionally amoebiasis, caused primarily

by Entamoeba histolytica, can have extra-intestinal presentations involving the pulmonary, cardiac, central nervous or hepatic systems.

The protozoan gains access to the human body through the gastrointestinal system following the foecal-oral route of infection.

Thus, infection is typically higher in areas with poor sanitation or poor socioeconomic conditions, with prevalence rates of up to 50%.

It is thought that, from the initial intestinal infection, amoeba ascend the portal vein to infect the liver, which is the most common extraintestinal site.

Objectives. Case report.

Methods. A 46-year-old man with no significant medical history presented to Helen Joseph Hospital, Johannesburg, South Africa, complaining of massive haemoptysis and a two-week history of weight loss and fever. An initial chest X-ray (CXR) revealed a mass in the left lower zone followed by a preliminary urgent computed tomography (CT) scan of the chest. This scan was inconclusive due to incessant coughing, but indicated a fluid-filled mass. The patient was initiated on antibiotics and antitussives, and was transferred to Chris Hani Baragwanath Academic Hospital to gain access to bronchial artery embolisation (BAE) facilities.

Results. CT scan planning for BAE showed a fluid-filled mass in the left lower lobe of the lung, which was inseparable from the diaphragm, and associated with oedema of the left lobe of the liver.

Blood tests showed raised inflammatory markers with a preserved haemoglobin level. The patient was monitored in a high-care setting.

No haemoptysis was noted; however, an anchovy paste-like fluid was subsequently drained from the mass.

Conclusion. With drainage and antibiotics, the patient’s condition improved. Serology for amoeba was positive and the patient continued on metronidazole for 2 weeks. He was well on discharge with resolution on CXR.

The combination of ABCA3 and pulmonary interstitial glycogenosis as a cause of neonatal interstitial lung disease Y Mulambia

Department of Paediatrics and Child Health, Faculty of Medicine and Health Sciences, Stellenbosch University yabwile@gmail.com

Introduction. Interstitial lung diseases are diffuse pulmonary disorders, with altered interstitial structure, resulting in abnormal gas exchange. ATP-binding cassette A3 (ABCA3) genetic mutations form part of the surfactant dysfunction mutations seen in interstitial lung disease. Most case reports of ABCA3 genetic mutations document very severe disease, while pulmonary interstitial glycogenosis (PIG) has a favourable outcome.

Objectives. We report two cases diagnosed in a middle-income setting with mutated ABCA3 and PIG. Methods. Case report.

Results. Patient A presented at 26 days old, with increased work of breathing and requiring ventilatory support. He required very high pressures on the ventilator and had multiple desaturations. Chest X-rays showed generalised alveolar disease. He also had severe pulmonary hypertension, which responded to vasodilators. Lung biopsy showed

SARJ VOL. 23 NO. 3 2017

83

ABSTRACTS glycogen in the interstitial cells – a spectrum of surfactant bodies with dense concentric lamellations, which also contain a dense body. The latter bodies bare some resemblance to ABCA3 surfactant deficiency.

The patient was pulsed with methylprednisolone with minimal response. Despite the support, the patient died.

Patient B, who presented on day 2 of life, had increased work of breathing. Multiple doses of surfactant were administered, but the patient’s condition deteriorated and he required mechanical ventilation. A chest computed tomography scan, which showed the “crazy paving” picture and lung biopsy, confirmed PIG. Genetic studies showed compounds heterozygous for the NM_001089.2(ABCA3): c.128G>A and c.316C>T variants. The patient received pulsed steroids and hydroxycholoroquine in addition to sildenafil for pulmonary hypertension. He still had very high oxygen requirements post extubation, with continuous positive airway pressure, resulting in a tracheostomy. At the time of submission of this manuscript, the patients were still on ventilators.

Conclusion. The combination of mutated ABCA3 and PIG can cause severe lung disease in neonates. When neonates do not respond to surfactant replacement therapy, surfactant deficiency must be considered. We describe a previously unknown ABCA3 mutation.

In children with PIG, who do not respond to treatment, surfactant deficiency must be considered.

The assessment of lung function abnormalities and functional outcomes in adult patients with first-time pulmonary tuberculosis in a high HIV prevalent setting S Manie

University of Cape Town, South Africa shamila.manie@uct.ac.za

Introduction. Research in the last decade has confirmed that the end of tuberculosis (TB) treatment does not in fact equate to a clean bill of health; on the contrary, it is the commencement of chronic lung function abnormalities. It is reported, that up to 60% of TB patients suffer from abnormal lung function in the form of obstructive, restrictive or mixed patterns. There is limited research on the prevalence of the pulmonary disease caused by TB in South Africa in an urban setting.

Objectives. To establish the prevalence of lung function abnormalities in adult patients being treated for pulmonary TB near completion of treatment (4 - 6 months), with or without HIV co-infection.

Methods. A cross-sectional, observational study design using a sample of convenience, was carried out among 315 patients with first-time TB, who had completed at least 4 months of treatment. Demographic data were recorded, and the following lung function parameters measured: forced expiratory volume (FEV1), forced vital capacity (FVC), the Tiffeneau-Pinelli index (FEV1/FVC), functional capacity using the six-minute-walk test, physical activity levels using the global physical activity questionnaire (GPAQ), and quality of life questionnaires using the EQ-5D and SGRQ. These measurements were used to obtain onceoff data to ascertain the prevalence of lung function abnormalities and functional outcomes. Lung functions were classified as either normal or abnormal (obstructive, restrictive or mixed).

84 SARJ VOL. 23 NO. 3 2017

Results. The sample had 173 (55%) males and 141 (45%) females. The mean age was 37 years and 162 (52%) patients were HIV-positive.

Abnormal lung functions were observed in 84 of 236 participants (35.6%) with acceptable spirometry reports. Further in-depth analyses are progressing to determine predictors of abnormal lung function.

Conclusion. The preliminary conclusion is that the prevalence of abnormal lung function, although not as high as in previous studies, is clinically significant in first-time TB patients.

Xpert MTB/RIF Ultra for diagnosis of pulmonary tuberculosis in children using induced sputum samples

H Zar,1 L Workman,1 M Prins,1 L Bateman,1 J Munro,1 S Mbhele,2 Y Ghebrekristos,2 M P Nicol2 Department of Paediatrics and Child Health, Red Cross War Memorial Children’s Hospital and Medical Research Council Unit on Child and Adolescent Health, University of Cape Town, South Africa 2 Division of Medical Microbiology and Institute for Infectious Diseases and Molecular Medicine, University of Cape Town and National Health Laboratory Service South Africa heather.zar@uct.ac.za 1

Introduction. Rapid microbiological confirmation of paediatric pulmonary tuberculosis (PTB) is necessary for diagnosis and therapy.

Xpert MTB/Rif Ultra (Ultra) can detect disease with fewer bacilli than Xpert and so may offer an improved rapid diagnostic.

Objectives. To investigate the diagnostic yield of Ultra compared to cultures from induced sputum (IS) samples in children.

Methods. Children hospitalised for suspected PTB were enrolled from 2011 to 2016. One to three IS samples were collected: the first was split for culture and Xpert the second was split for culture and Xpert or storage, and the third sample was stored. Ultra was performed in December 2016, on a single IS sample from stored second or third IS samples. The accuracy of Ultra was compared to culture as the reference standard.

Results. Culture results were available for 380 samples. The median (interquartile range) age of children was 34.0 (15.5 - 73.4) months and 71 children (18.7%) were HIV-positive. On per sample analysis, cultures were positive in 73 (19.2%) and Ultra in 65 (17.1%). Xpert results were available for 114 samples were positive in 17 (14.9%). The sensitivity and specificity of Ultra on per sample analysis (culture and Ultra on the same sample) were 75.3% and 96.7%, respectively, with similar results in HIV-positive (sensitivity 70.6%; specificity 98.1%) and HIV-negative (sensitivity 76.8%; specificity 96.4%) children. Ultra was positive in ten children with negative cultures, of whom seven were treated as ‘unconfirmed TB’ according to the National Institutes of Health (NIH) classification. Sensitivity and specificity of Ultra on per patient analysis (Ultra from one IS sample v. culture from multiple IS samples) were 67.5% and 96.6%, respectively. If the seven clinicallydiagnosed TB cases that were positive on Ultra were regarded as true positives, the sensitivity and specificity of Ultra on per sample analysis, were 77.5% and 99%, respectively.

Conclusion. Ultra provides rapid detection of Mycobacterium tuberculosis from a single IS sample in most children with cultureconfirmed TB. Ultra may detect an additional group of children who are not detected by culture.

ABSTRACTS

Experiences of non-invasive ventilation in older people with hypercapnic respiratory failure H Ngandu

School of Healthcare Sciences, College of Biomedical and Life Sciences, Cardiff University, Wales, UK nganduh@cardiff.ac.uk

Introduction. Non-invasive ventilation (NIV) is proven to be an effective way of treating hypercapnic respiratory failure (HRF) in older people, in whom endotracheal intubation may not provide longterm benefits.

Objectives. The aim of this study was to examine the experiences of older people with HRV and on NIV treatment, to understand treatment tolerance and treatment preferences.

Methods. A qualitative approach, using a semi-structured interview, was conducted with 11 individuals (5 males and 6 females), who were receiving NIV treatment for the first time. The mean (interquartile range) age of the participants was 72 (67 - 76) years. Participants were recruited from a hospital in Wales, UK. Twelve participants were interviewed and 11 were chosen to analyse the richness of the data in terms of the participants’ lived experiences, while receiving NIV treatment. One interview was excluded due to insufficient data.

Results. Five superordinate themes were identified: (i) significant moment in life, (ii) the impact of the NIV machine, (iii) understanding the benefits of the NIV machine, (iv) decisionmaking in a life-threatening experience, (v) conflict managing the NIV machine. These analyses indicated that starting NIV caused bewilderment and confusion because participants were not aware that their respiratory conditions required such intense therapy.

Health benefits, such as improved breathing, ostensibly motivated the continued use of NIV. Decision-making in patients suffering from life-threatening conditions, such as HRF, was influenced by issues like health benefits, the spouse’s voice, the desire to improve, and most importantly, the burden of treatment v. the quality of life.

Conclusion. In this study, opinions of NIV treatment, as experienced by older people with HRF, were explored. Furthermore, factors influencing the refusal or acceptance at the initiation stage of NIV treatment in older people have been identified.

Impact of tobacco smoke exposure or indoor air pollution on nasopharyngeal bacteria in African infants in a birth cohort study A Vanker, W Barnett, P M Nduru, F S Dube, R P Gie, M P Nicol, H J Zar University of Cape Town, Cape Town, South Africa aneesa.vanker@uct.ac.za

Introduction. Bacterial nasopharyngeal (NP) carriage precedes the development of lower respiratory tract infection (LRTI), a leading cause of childhood illness globally. Exposure to indoor air pollution (IAP) or environmental tobacco smoke (ETS) may influence NP bacterial carriage.

Objectives. To investigate the impact of antenatal or postnatal IAP and ETS exposure on NP bacterial carriage in infants.

Methods. Mother-infant pairs enrolled in a South African birth cohort study and were followed from birth for 1 year. Immunisation, including conjugate 13-valent pneumococcal vaccine and Haemophillus influenzae b were given. IAP exposures (particulate matter, nitrogen dioxide and volatile organic compounds) were measured at an antenatal and postnatal home visit. Maternal and infant urine cotinine levels were used to measure ETS exposure. NP swabs were taken at birth, 6 and 12 months for bacterial culture.

Associations between NP organisms and environmental exposures were investigated using multivariate logistic regression.

Results. A total of 2 596 NP swabs were collected from 986 infants.

The predominant NP bacteria were Streptococcus pneumoniae (n=1 136, 44%), Moraxella (n=893, 34%), H. influenzae (n=709, 27%) and Staphylococcus aureus (n=239, 9%). Antenatal ETS exposure was associated with S. pneumoniae at 6 months, (odds ratio (OR) 1.62 95%; CI 1.10 - 2.37). Antenatal toluene exposure was also associated with S. pneumoniae at 6 months, (OR2.51; 95% CI 1.23 - 5.13) and with H. influenzae at 12 months, (OR2.27; 95% CI 1.24 - 4.16).

Postnatal ETS exposure was associated with S. pneumoniae at 6 months, (OR1.46; 95% CI 1.01 - 2.12) and H. influenzae at 12 months (OR2.56; 95% CI 1.21 - 5.41).

Conclusion. Early-life environmental exposures increase the prevalence of specific NP bacteria during infancy, which may predispose to the development of LRTI.

Pectus carinatum: An external custommade carinatum brace I Schewitz

University of Pretoria, Pretoria, South Africa ivan@schewitz.com

Introduction. The Ravitch procedure for correction of a pectus carinatum has been the standard repair procedure for many years.

Following the Nuss repair for pectus excavatum introduced by Nuss in 1997, Abramson adjusted this operation for pectus carinatum with great success. Various external braces were subsequently developed for a non-surgical attempt to repair the deformity.

Objectives. To introduce the custom-made, adjustable brace, and to describe the preliminary results.

Methods. This is a review of the first 12 cases performed in South Africa. All have shown improvements, some within 2 months of wearing the brace. The important point is that the patients need to wear the brace for an adequate number of hours. The brace is also worn at night. It is only removed when partaking in sports and for showering.

The brace is custom-made for each patient, the size is adjustable, as is the pressure required to correct the deformity. Pressure sores develop when excessively high pressure is applied – this may be avoided by adjusting the pressure as the deformity corrects. I will compare the brace with the Abramson procedure.

Results. Preliminary results have shown improvements in all cases.

Two patients discontinued the use of the brace within 6 months as the deformity had been corrected. One patient did not wear the brace as he felt self-conscious at school. He had not experienced any improvement.

Conclusion. The brace has been shown to be an efficient non-surgical method for the correction of Pectus carinatum.

SARJ VOL. 23 NO. 3 2017

85

ABSTRACTS

Removal of Pectus Bars: How I do it!

I Schewitz

University of Pretoria, South Africa ivan@schewitz.com

Introduction. The removal of pectus bars can present life-threatening complications, as described in the literature. This minor procedure is potentially fatal when performed incorrectly.

Objectives. I will describe my personal approach to this procedure, which aims to avoid complications.

Methods. An incision is made on the side of the thorax containing the pectus bar stabiliser. The stabiliser is removed, and the bar is straightened and rotated to loosen it along its entire length. The bar is then removed without excessive pressure. Occasionally, one requires a hammer and chisel to remove bony ingrowth into the bar. Exposing both ends of the bar is rarely required.

Results. Using the abovementioned method, 60 patients have had their pectus bars removed without complications.

Discussion. During pectus bar removal, the most critical complications are bleeding from the right ventricle (massive and life threatening) or the internal thoracic (mammary) artery, erosion of the sternum, lung trauma, and very rarely, damage to the aorta.

Conclusion. The most effective way to avoid complications during pectus bar removal is to insert it correctly. The bar must be inserted under direct vision to ensure that it is extra-pericardial. Multiple bars are often required to avoid excessive pressure on both the sternum and the internal thoracic artery. Finally, it is imperative to loosen bars with lateral rotation before withdrawing them.

The injectable contraceptive medroxyprogesterone acetate decreases peripheral effector cell-mediated Mycobacterium tuberculosis containment through immunosuppression at multiple levels involving the glucocorticoid receptor

M Tomasicchio,1 M Davids,1 A Pooran,1 G Theron,1,2 L Smith,1 L Semple,1 R Meldau,1 J Hapgood,3,4 K Dheda1,4

Lung Infection and Immunity Unit, Division of Pulmonology and UCT Lung Institute, Department of Medicine, University of Cape Town, South Africa 2 Department of Science and Technology/National Research Foundation Centre of Excellence for Tuberculosis Research, South African Medical Research Council Centre for Tuberculosis Research, Division of Molecular Biology and Human Genetics, Faculty of Health Sciences, Stellenbosch University, Cape Town, South Africa 3 Department of Molecular and Cell Biology, University of Cape Town, Cape Town, South Africa 4 Institute of Infectious Diseases and Molecular Medicine, University of Cape Town, Cape Town, South Africa michele.tomasicchio@uct.ac.za 1

Introduction. The potential effects of the widely used injectable contraceptives, medroxyprogesterone acetate (MPA) and norethisterone (NET) enanthate, on Mycobacterium tuberculosis pathogenesis has not been fully investigated. Objectives. To determine if MPA or NET reduces M. tuberculosis containment

86 SARJ VOL. 23 NO. 3 2017

in monocyte-derived macrophages through immunosuppression involving the glucocorticoid receptor.

Methods. We recruited HIV-negative females, who were not on a contraceptive, in Cape Town, South Africa, to determine if MPA or NET affects the ability of effector T-cells to contain M. tuberculosis within monocyte-derived macrophages through a mechanism involving the glucocorticoid receptor. We further investigated whether MPA or NET regulated key immune function genes known to be protective against TB using flow cytometry.

Results. MPA (p<0.005), but not NET, attenuated M. tuberculosis containment through a mechanism involving the glucocorticoid receptor. Addition of MPA, but not NET, to purified proteinderivative-stimulated effector T-cells resulted in regulatory T-cell upregulation (1.8-fold; p<0.05), reduced CD4+ T-cell interferon-Îł (IFN-Îł) production (60%; p<0.05), and decreased CD8+ perforin (50%, p<0.0001) and CD4+ granzyme B production (50%; p<0.005).

We provide further evidence that MPA regulates these immune responses by a mechanism involving the glucocorticoid receptor.

Conclusion. Our results suggest that MPA downregulates protective host immune responses against M. tuberculosis via the glucocorticoid receptor, whereas NET does not have the same effect. The data presented have potential implications for the use of MPA in high TB-burden countries.

Evaluation and characterisation of interleukin-9-producing T-cell subsets in tuberculosis A Pooran, M Davids, F Thompson, K Dheda

Lung Infection and Immunity Unit, University of Cape Town Lung Institute, South Africa anil.pooran@gmail.com

Introduction. Interleukin-9 (IL-9) has traditionally been considered a Th2 cytokine, but recent evidence suggests that the majority of IL-9 is produced by a distinct Th9 T-cell lineage. However, the involvement of IL-9 and Th9 cells in tuberculosis (TB) remains unclear.

Objectives. To determine IL-9 levels in blood and site of disease in TB patients and latent tuberculosis infection (LTBI) controls, and to further clarify the existence of the novel Th9 lineage.

Methods. Bronchoalveolar lavage (BAL) and peripheral blood mononuclear cells (PBMCs) were obtained from individuals with confirmed pulmonary TB (n=15) and presumed LTBI (n=13). IL-9 and Th1 protein levels were determined in mycobacterial-antigendriven cell culture supernatants by Luminex. Flow cytometry was used to determine the frequency of IL-9-producing CD4 T-helper (Th) and CD8 T-cytotoxic (Tc) subsets, the coexpression levels of cytokines specific to other IL-9 producing T-cell lineages, and PU.1 coexpression, a proposed lineage defining transcription factors of Th9 cells.

Results. IL-9 protein levels (BAL, p=0.02) and IL-9:Th1 ratio (BAL, p=0.001; blood, p=0.049) were increased in TB patients v. LTBI controls.

CD4+IL-9+ (blood, p=0.02; BAL, p=0.03) and CD8+IL-9+ (blood, p=0.047) cells were significant higher in TB patients vs. LTBI controls.

In the CD4+IL-9+ population, few cells coexpressed IL-13 (Th2) or IL-17 (Th17) and most (94.0% in blood v. 80.4% in BAL) exhibited

ABSTRACTS a Th9 phenotype (CD4+IL-9+IL-13–IL-17–). Similarly, most CD8+IL-9+ cells exhibited a Tc9 phenotype (CD8+IL-9+L-13–IL17–; 93.7% in blood v. 84.2% in BAL). However, expression levels of PU.1 in Th9 and Tc9 cells were very low.

Conclusion. In TB patients, IL-9 producing T-cells exhibit a Th9/ Tc9 phenotype and are increased both at the site of disease and the periphery. These preliminary data suggest that a Th9 response may be involved in TB pathogenesis. Further investigations are required to elucidate the specific role of Th9 and Tc9 cells in TB.

What is the role of Th2 cytokines in the host immune response to tuberculosis?

A Pooran,1 M Davids,1 A Nel,2 A Shoko,2 J Blackburn,2 K Dheda1 ung Infection and Immunity Unit and UCT Lung Institute, University of L Cape Town, South Africa 2 Department of Medical Biochemistry, Institute of Infectious Disease and Molecular Medicine, Faculty of Health Sciences, University of Cape Town, South Africa anil.pooran@gmail.com 1

Introduction. Tuberculosis (TB) vaccine candidates have been mostly ineffective and it remains unclear as to what constitutes protective host immunity. Despite high levels of IFN-γ, the Th1 cytokine associated with protection at the site of disease, many individuals still have progressive active TB. Whether a Th2-like immune response can subvert protective host immunity requires clarification.

Objectives. To determine Th1 and Th2 cytokine levels in the lungs and blood of TB patients and LTBI controls, and how Th2 cytokines (IL-4) affect Mycobacterium tuberculosis-specific host immunity in vitro.

Methods. Blood and/or bronchoalveolar lavage fluid (BAL) were obtained from individuals with confirmed pulmonary TB (n=25) and presumed latent TB infection (LTBI; n=25). Th1 and Th2 cytokine mRNA levels were determined by qPCR. Human recombinant interleukin-4 (hrIL-4) was expressed in a baculovirus system and functionally validated using 3H-thymidine proliferation and B-cell flow cytometry. The effect of IL-4 on mycobacterial containment (colony-forming units (CFU)/mL) and on cellular and cytokine expression (flow cytometry) were evaluated in an in vitro mycobacterial containment assay.

Results. In whole blood, TB patients expressed higher IL-4 mRNA levels (p=0.02) and had lower IFN-γ/IL-4 ratios (p=0.01) compared with LTBI controls. Functionally active hrIL-4 increased T-cell proliferation and B-cell CD23 expression in a dose-dependent manner. In the mycobacterial containment model, addition of hrIL-4 was associated with a reduction in mycobacterial containment (p=0.008), increased levels of regulatory T-cells (CD4+CD25+FoxP3+; p<0.001), decreased CD4+ Th1 cytokines (CD4+IFN-γ, p<0.001; CD4+TNFα, p=0.01), and increased macrophage DC-SIGN expression (p=0.02) in a dose-dependent manner. Anti-IL-4 antibodies abrogated the effect of IL-4 on mycobacterial containment (p=0.03), as well as the expression of CD4+IFNγ+ (p=0.03) and regulatory T-cells (p=0.03).

Conclusion. TB patients have a compartmentalised and Th2skewed host immune response. IL-4 is associated with subversion

of mycobacterial containment in human macrophages through a regulatory T-cell-mediated Th1 downregulation. These data have implications for the selection of effective vaccine candidates and the design of appropriate TB-specific immunotherapeutic interventions.

Mycobacterium avium complex immune reconstitution inflammatory syndrome presenting as endobronchial disease P Maharaj

Inkosi Albert Luthuli Central Hospital, Durban, South Africa priyaloveshamlet@icloud.com

Introduction. Mycobacterium avium complex (MAC) infection is a common opportunistic infection in patients with AIDS. It is the most common disseminated bacterial infection in patients with AIDS in industrialised countries. Despite this, endobronchial disease in MAC infection has very rarely been reported, with no reported cases in South Africa. MAC is also frequently associated with the immune reconstitution inflammatory syndrome (IRIS). Immune restoration has been postulated to have a role in the development of endobronchial lesions in MAC infection.

Objectives. To report the unusual finding of an endobronchial lesion in a patient with AIDS presenting with MAC-associated IRIS, and to investigate the relationship between endobronchial lesions in MAC infection and reconstituted immunity.

Methods. A 29-year-old HIV-positive woman was referred to our centre for bronchoscopy and lavage. She reported a 1-month history of cough productive of yellow sputum, as well as fevers, weight loss of 15 kg and fatigue. Repeated sputum samples had identified acidfast bacilli, but the GeneXpert was consistently negative. Four months prior to presentation she had commenced combination antiretroviral therapy with a baseline CD4 cell count of 18 cells/µL. Results. The repeat CD4 cell count was 103 cells/µL. On bronchoscopy, a single polyp of 0.5 cm was found at the entrance to the right upper lobe bronchus. Biopsies of the lesion showed necrotising and nonnecrotising granulomatous inflammation surrounded by a chronic inflammatory cell infiltrate. Acid-fast bacilli were observed. Molecular studies confirmed the presence of Mycobacterium avium.

Conclusion. Endobronchial lesions are a rare manifestation of MAC infection. This unusual complication is typically reported in patients with AIDS who were recently initiated on antiretroviral therapy, and appears to be a feature of MAC-associated IRIS.

Bronchial thermoplasty for severe persistent asthma: Experience from Cape Town, South Africa

A T Mnguni,1 A Esmail,2 A Pooran,2 M Davids,2 G Calligaro,1 K Dheda1 1 2

Groote Schuur Hospital, University of Cape Town, Cape Town, South Africa Lung and Infection Unit, Groote Schuur Hospital, University of Cape Town, Cape Town, South Africa trevormnguni@yahoo.com

Introduction. Bronchial thermoplasty (BT) is a Global Initiative for Asthma-recommended step-5 therapy for severe persistent

SARJ VOL. 23 NO. 3 2017

87

ABSTRACTS asthma in patients who are uncontrolled despite optimal medical therapy. However, there are no outcome data for severe asthma in resource-limited settings.

Objectives. To examine the effect of BT on severe asthma.

Methods. Patients with severe persistent asthma undergoing BT (three treatments, 1 month apart) at Groote Schuur Hospital and the University of Cape Town Academic Hospital were enrolled prospectively. The primary outcome was a decrease in the number of exacerbations – defined as the need for adjunct oral corticosteroids or antibiotics – recorded in the 12 months before the procedure, compared with the number of exacerbations in the 12 months after completion of the final BT treatment. Secondary outcomes included differences in Asthma Control Test (ACT) scores, overall chronic oral corticosteroid (OCS) dose (in mg/month), number of asthma classes of medications used pre- and post BT, and the need for admission post-procedure. Patients also underwent bronchial biopsies after each procedure, but the relevant results are not reported here.

Results. Twelve patients (50% male; median age (interquartile range (IQR)) of 59 (46 - 64); 10/12 (83%) on OCS and 8/12 (66%) with previous ICU admission) underwent BT (36 procedures) and completed the 12-month post-procedure follow-up period. The median (IQR) number of exacerbations decreased significantly post-BT 12 (6 - 12) v. 2 (0.3 - 3); p=0.0002). ACT scores post-BT were also significantly higher (7.5 (6.0 - 11.8) v. 14.5 (11.25 - 17.75); p=0.012). There was a reduction in the median (IQR) monthly OCS dose before and after BT treatment, 300 mg (200 - 450) v. 240 mg (40 - 450). The number of classes of asthma medications was unchanged. BT was well tolerated, but 8% (n=3) of patients developed post-procedural bronchospasm requiring overnight admission, one of whom developed a pneumothorax that was managed conservatively.

Conclusion. In this small single-centre study in a resource-poor setting, which enrolled patients with disease more severe than in published clinical trials, BT reduced exacerbations and improved asthma control.

The utility of endobronchial ultrasound-guided transbronchial needle aspiration (EBUS-TBNA) for the diagnosis of mediastinal lesions in a resource-limited TB- and HIV-endemic setting M Shikar, A Esmail, L Mottay, S Oelofse, G Calligaro, K Dheda

Department of Pulmonology, Groote Schuur Hospital, Cape Town, South Africa keertan.dheda@uct.ac.za

Introduction. Diagnostic evaluation by mediastinoscopy is associated with increased risk, cost and hospitalisation. Although endobronchial ultrasound-guided transbronchial needle aspiration (EBUS-TBNA) may be a useful alternative, there is limited data from tuberculosis (TB)- and HIV-endemic settings with limited access to surgical facilities.

Objectives. To investigate the utility of EBUS-TBNA for the diagnosis of mediastinal lesions in a resource-limited TB- and HIVendemic setting.

88 SARJ VOL. 23 NO. 3 2017

Methods. We prospectively evaluated 154 patients who underwent EBUS-TBNA between March 2013 and July 2017 at Groote Schuur Hospital, Cape Town, South Africa. The indications for EBUSTBNA were undiagnosed mediastinal lesions and assessments for staging of lung cancer. Patients without diagnostic clarity underwent mediastinoscopy or appropriate surgical biopsy.

Results. The diagnostic accuracy of EBUS-TBNA, regardless of the indication, was 68.7% (95% confidence interval (CI) 57.7 - 75.7) with a positive predictive value of 100% (95% CI 94.7 - 100), and negative predictive value of 63.9% (95% CI 52.1 - 71.9). Overall, EBUS-TBNA diagnosed TB in 19/24 patients (79.2%). Sarcoidosis was diagnosed in 6/16 (62.5%) patients with EBUS-TBNA alone, and 11/16 patients (68.8%) when combined with transbronchial biopsy. Malignant disease was diagnosed in 39/54 (72.2%) patients. False negative results were obtained in 31 (20%) patients, of whom 15 had malignancy, 5 had TB, and 5 had sarcoidosis. On subgroup analysis, 14/154 (9.1%) referred patients were HIV-positive, of whom EBUS-TBNA diagnosed TB in 5 patients, sarcoidosis in 1 patient, and benign disease in 7 patients, with 1 false-negative diagnosis in a patient with lymphoma.

The overall sensitivity of EBUS-TBNA in HIV-positive persons was 92.9%. The procedure was well tolerated in 149/154 patients (96.7%), with reversible complications occurring in five patients.

Conclusion. EBUS-TBNA is a safe and valuable diagnostic tool for benign and malignant disease, even in HIV-positive persons. Overall, surgical intervention was avoided in ~70% of patients, thereby improving the time to treatment initiation, and reducing associated risk, cost and hospitalisation rates. These data inform clinical practice in resource-limited TB- and HIV-endemic settings.

A human lung-orientated approach to correlates of risk in tuberculosis M Davids, A Pooran, R Meldau, F Thompson, P Gina, L Mottay, A Esmail, K Dheda Lung Infection and Immunity Unit, Department of Medicine, University of Cape Town, South Africa malika.davids@uct.ac.za

Introduction. Currently tested vaccines against tuberculosis (TB) have been ineffective. Evidence suggests that a robust Th1 immune response is insufficient to prevent disease progression. Immunological correlates of risk are poorly understood within the human lung, the organ of initial contact with Mycobacterium tuberculosis.

Objectives. To investigate, in a comparative, first in man study, local in vivo pulmonary immune responses in HIV-negative individuals with different risk susceptibility profiles based on clinical, radiological and immunodiagnostic profiles, including immunodiagnostic test negative despite exposure, presumed latent tuberculosis infection (LTBI), previous active TB, recurrent TB, and self-cured TB, using a lung antigenic challenge model (purified protein derivative (PPD) and live-Bacillus Calmette-Guérin (BCG)).

Methods. PPD, live BCG, and saline (control) were instilled into different lung segments via bronchoscopy. Initial experiments were performed to determine the immunomechanical effect of saline instillation and to optimise BCG and PPD concentrations.

Bronchoalveolar lavage was performed prior to antigenic challenge, i.e. at baseline, and 72 hours’ post-challenge. Peripheral blood samples

ABSTRACTS and BCG-challenged skin biopsy samples were collected concurrently.

Flow cytometry was used to analyse BAL and peripheral blood cells for cell surface markers and cytokine/chemokine expression profiles associated with innate and cell-mediated immune pathways.

Results. The bronchoscopic instillation of saline alone in healthy controls (n=4) induced an immune response. Antigenic challenge using BCG (104 colony-forming units) and PPD (0.5 tuberculin units) was optimal in generating measurable alveolar immune responses, i.e. an increase in total cell numbers from the baseline (BCG-driven, p=0.03; PPD-driven, p=0.004). PPD challenge in those with previous TB showed significantly increased TLR2+IL6+ coexpression in macrophages (p=0.01), but decreased biomarker-specific T-cell expression (CD4+TNFÎą (p=0.05), CD8+TNFÎą (p=0.02), and Th17 homing cells (CD4+IL17+CCR6+; p=0.004). However, a high degree of inter-patient variability was observed.

Conclusion. These preliminary findings demonstrate the feasibility of using an in vivo mycobacterial-specific human lung antigenic challenge model. The emerging data will likely have implications for the design of vaccines and immunotherapeutic interventions.

Diagnosing tuberculosis in hospitalised HIV-positive individuals who cannot produce sputum: Is urinelipoarabinomannan testing the answer?

L Mottay, A Esmail

Lung Infection and Immunity Unit, Division of Pulmonology, Department of Medicine, University of Cape Town, South Africa lynelle.mottay@uct.ac.za

Introduction. Up to one-third of HIV-positive individuals with suspected tuberculosis (TB) are sputum-scarce. The Alere Determine TB LAM Ag lateral-flow strip test can be used to diagnose TB in HIVinfected patients with advanced immunosuppression. However, how urine-LAM testing should be incorporated into testing algorithms, and in the context of specific patient subgroups, remains unclear.

Objectives. To clarify the role of urinary LAM in tuberculosis diagnostics.

Methods. This study represents a post hoc subgroup analysis of data from a randomised multicentre parent study. The study population consisted of hospitalised HIV-infected patients with suspected TB who were unable to produce sputum and who underwent urine-LAM testing. The diagnostic utility of urine LAM for TB in this group was compared with the performance of urine LAM in patients who did produce a sputum sample in the parent study.

Results. There was a total of 187 and 2 341 patients in the sputumscarce and sputum-producing cohorts, respectively. In comparison with those who produced sputum, sputum-scarce patients were younger, had a lower Karnofsky performance score, and a lower weight and body mass index at admission. A greater proportion of sputum-scarce patients were urine-LAM-positive, compared with those who were able to produce sputum (31% v. 21%, respectively; p=0.04). A higher proportion of sputum-scarce patients died within 8 weeks of admission (32% v. 24%, respectively; p=0.013).

Conclusion. Urine-LAM testing can effectively identify TB in HIVinfected patients who are at a higher risk of mortality and unable to

generate a sputum sample for diagnostic testing. Our findings support the use of urine-LAM testing in sputum-scarce, hospitalised HIVinfected patients, and its incorporation into diagnostic algorithms for this patient population.

Early-morning urine collection to improve urinary lateral flow lipoarabinomannan (LAM) assay sensitivity in hospitalised patients with HIV-TB co-infection P Gina, K Dheda

Lung Infection and Immunity Unit, Division of Pulmonology, Department of Medicine, University of Cape Town, South Africa gnxnto001@myuct.ac.za

Introduction. The HIV pandemic has fuelled a resurgence of tuberculosis (TB), which is the leading cause of death in HIV-infected persons in sub-Saharan Africa. Rapid initiation of TB treatment may reduce mortality in these vulnerable patients. Urine-LAM testing has been approved by the World Health Organization for use in hospitalised patients with advanced immunosuppression. However, the sensitivity of the test remains suboptimal.

Objectives. To examine the incremental diagnostic sensitivity of early morning urine (EMU) v. random urine sampling using the Determine lateral flow lipoarabinomannan assay (LF-LAM) in HIVTB-coinfected patients.

Methods. Consenting HIV-positive inpatients, screened as part of a larger prospective, randomised controlled trial, who were treated for TB and could donate matched random and EMU samples were included. Thus, paired samples were collected from the same patient, and LF-LAM was graded using the pre-January 2014 grade 1 and 2 manufacturer-designated cut-off points (the latter was designated grade 1 after January 2014). Single sputum GeneXpert-MTB/RIF- and/ or TB-culture positivity served as the reference standard (definite TB).

Those treated for TB but not meeting this standard were designated as cases of probable TB. Results. A total of 123 HIV-infected patients commenced anti-TB treatment and provided matched random and EMU samples. A total of 33% and 67% had definite and probable TB, respectively. Among those with definite TB, LF-LAM sensitivity using the grade 2 cut-off point increased from 12% to 39% with random v. EMU respectively (p=0.005).

Similarly, among probable TB cases LF-LAM sensitivity increased from 10% to 24% (p=0.001). LF-LAM specificity was not determined.

Conclusion. This proof-of-concept study indicates that EMU could improve the sensitivity of LF-LAM in hospitalised TB-HIV coinfected patients. These data have implications for clinical practice.

A very unusual cause of haemoptysis: Unilateral pulmonary vein atresia in an adolescent M Mitha, A Mitha

Inkosi Albert Luthuli Central Hospital, Durban, South Africa mohmitha@gmail.com

Introduction. Unilateral pulmonary vein (UPV) atresia is a very uncommon cause of haemoptysis presenting in adults. It is

SARJ VOL. 23 NO. 3 2017

89

ABSTRACTS commonly diagnosed in infancy and childhood, and can present as recurrent infections and haemoptysis.

Objectives. To present an unusual case of a 14-year-old female with congenital UPV atresia and a review of the literature.

Methods. A 13-year-old HIV-negative female presented with the problem of intermittent recurrent episodes of haemoptysis.

It started at the age of 3 years old, and she was treated for pulmonary tuberculosis (TB) for 6 months. However, the Mycobacterium tuberculosis was never detected. She had another bout of haemoptysis at age 11 and was once again treated for TB, being smear- and culture-negative and completing 6 months of treatment. A computed tomography pulmonary angiogram showed the right pulmonary vein atresia.

Results. UPV atresia is a rare condition and generally presents in early childhood, with haemoptysis and recurrent respiratory infections, and in later stages with worsening dyspnoea. It is associated with congenital heart disease in approximately one-third to one-half of cases. The presence of isolated pulmonary vein atresia without any structural cardiac or developmental venous abnormality presenting in adulthood is rare, with ~50 cases being reported.

Conclusion. This case is important in our hospital setting, as we have a high endemic pulmonary TB burden where patients can present with haemoptysis. This case highlights a delay in diagnosis for several years because pulmonary vein atresia was not considered as a differential. It also highlights that UPV atresia should always be considered as a possible cause of haemoptysis with recurrent infections, and that we should not always rush to condemn patients to 6 months of TB treatment when we do not have strong evidence to support it. The case also highlights that non-invasive methods are adequate in diagnosing this entity.

Empyema in children hospitalised at the Chris Hani Baragwanath Academic Hospital: A retrospective study Z Dangor, A Ghoor, T Mabaso, K Mopeli, S Lala, C Verwey University of the Witwatersrand, Johannesburg, South Africa ziyaad.dangor@wits.ac.za

Introduction. The incidence of empyema has significantly decreased with the introduction of the 13-valent pneumococcal conjugate vaccine. Nevertheless, the management of empyema continues to be challenging in low-resourced settings.

Objectives. To describe the aetiology and management of empyema in a high-burden HIV-TB co-infection setting.

Methods. A retrospective descriptive study was undertaken between January 2012 and December 2016 in children <14 years of age at a large secondary-tertiary referral hospital. Cases of empyema were identified through pulmonology and discharge-summary databases. Clinical, laboratory and radiological data were collected from pulmonology files and patient records.

Results. Over a 5-year period, 66 cases met our case definition for empyema, of which 22 (33%) were referred from surrounding hospitals. The median (interquartile range (IQR)) age of presentation was 52.0 (18.1 - 103.6) months. A total of 13 (19.7%) were HIV-infected and six (9.1%) were HIV-exposed but uninfected. The most common causative organisms were Staphylococcus aureus (15/66; 22.7%)

90 SARJ VOL. 23 NO. 3 2017

and Streptococcus pneumoniae (5/66; 7.8%). Treatment for Mycobacterium tuberculosis was initiated in 28 (42.4%) cases and more frequently initiated in HIV-infected children (10/13; 76.9%; p=0.010). Microbiological evidence of TB was present in five (7.6%) cases. Overall, seven (10.6%) children were ventilated and one died (case fatality rate: 1.5%). Forty-three (65.2%) cases had an intercostal drain, and 16 (24.2%) had a pigtail percutaneous catheter inserted; however, fibrinolytics were documented in only six cases (10.2%).

Eight (12.1%) cases had a thoracotomy and seven (10.6%) had videoassisted thorascopic drainage, all of whom had a prior drain inserted and a median (IQR) of 20 (10 - 33) days from admission to surgery.

Conclusion. Our study showed an increased prevalence of S. aureus empyema compared with previous reports. In HIV-infected children with empyema, a large proportion were initiated on TB therapy, highlighting challenges in managing TB-HIV co-infection. Although fibrinolytics or early surgery is recommended, neither practice was common.

Diagnostic performance of lung ultrasound compared with chest X-ray for pneumonia in children in the Drakenstein Child Health Study

J A M Stadler,1 S Andronikou,2 D Le Roux,3 E Von Delft,4 H J Zar5 University of Cape Town, South Africa Department of Paediatric Radiology, Bristol Royal Hospital for Children and University of Bristol, UK; Department of Paediatrics and Child Health and Medical Research Council Unit on Child and Adoles 3 Department of Paediatrics, Somerset Hospital, University of Cape Town, South Africa 4 Department of Paediatrics, Paarl Hospital, South Africa 5 Department of Paediatrics, Red Cross War Memorial Childrenâ&#x20AC;&#x2122;s Hospital; and Department of Paediatrics and Child Health; and Medical Research Council Unit on Child and Adolescent Health, University of Cape Town, Cape Town, South Africa attiestadler@gmail.com 1

2

Introduction. Chest X-ray (CXR) is the first-line imaging modality for suspected pneumonia in children. However, CXR interpretation is limited by moderate sensitivity and specificity and poor inter-rater reliability (IRR), and it exposes children to ionising radiation. Lung ultrasound (LUS) may be a radiation-free, lower-cost alternative for the diagnosis of pneumonia.

Objectives. To compare the diagnostic performance of LUS and CXR in children with suspected pneumonia.

Methods. LUS was performed on 103 children in the Drakenstein Child Health Study who presented with World Health Organization defined pneumonia and who received a CXR as part of routine care.

IRR between a general practitioner and a paediatric radiologist for the interpretation of LUS findings for pneumonia were compared with CXR interpretation by two specialist paediatricians. Reliability of different LUS features was also compared.

Results. The study included 72 hospitalised and 31 non-hospitalised pneumonia cases with a median age of 7.3 months. The general practitioner and paediatric radiologist reported LUS findings consistent with pneumonia in 58% (n=60) v. 52% (n=54) of cases, respectively (p=0.01). The overall agreement on LUS findings was

ABSTRACTS substantial (kappa=0.60) compared with poor agreement on CXR (kappa=0.25). On LUS, there was better agreement for consolidation (kappa=0.67) and for a normal scan (kappa=0.68) than for interstitial syndrome (kappa=0.45).

Conclusion. LUS demonstrated better IRR than CXR for detecting features of pneumonia in children. Consolidation is a more reliable sonographic sign of pneumonia than interstitial syndrome. LUS may be preferable to CXR for the diagnosis of pneumonia in children.

The incremental yield of urine LAM over Xpert MTB/RIF testing in hospitalised HIV-infected sputumexpectorating patients L Mottay, A Esmail

Lung Infection and Immunity Unit, Division of Pulmonology and UCT Lung Institute, Department of Medicine, University of Cape Town, South Africa lynelle.mottay@uct.ac.za

Introduction. GeneXpert MTB/RIF and urinary lipoarabinomanan (LAM) are newer tuberculosis (TB) tests that have been shown to be effective for the diagnosis of TB in advanced HIV co-infection.

Objectives. The goal of this analysis was to assess the diagnostic utility of urine LAM for TB in HIV-infected patients admitted for suspected TB and for whom an Xpert MTB/RIF test result was available.

Methods. The analysis included patients who were able to provide a sputum sample for GeneXpert MTB/RIF, and urine for the Alere Determine TB LAM Ag lateral-flow strip test. Three diagnostic strategies were compared: using both urine-LAM- and Xpert MTB/ RIF-testing for all patients suspected of TB; using urine-LAM testing only in those found to have a negative GeneXpert MTB/RIF test; and using GeneXpert MTB/RIF-testing in only those who had negative urine-LAM results. Sputum culture was used as a reference standard for the diagnosis of TB. Results. Urine-LAM testing had a sensitivity of 38.1% (95% confidence interval (CI) 30.9 - 45.7) and a specificity of 88.1% (95% CI 84.3 - 91.1) while GeneXpert MTB/RIF testing was found to have a sensitivity of 75.0% (95% CI 67.9 - 81.2) and a specificity of 95.1% (95% CI 92.4 - 97.0). The combination of Urine LAM and GeneXpert MTB/RIF had a sensitivity of 78.4% (95% CI 71.6 - 84.2), and a specificity of 84.6% (95% CI 80.7 - 88.1). The incremental yield of adding urine-LAM testing after GeneXpert MTB/RIF testing (in those with a negative GeneXpert MTB/RIF test) was 4.5%, while the incremental yield of adding GeneXpert MTB/RIF testing after urine LAM testing (in those with a negative urine LAM test) was 106%.

Conclusion. In sputum-expectorating patients with advanced HIV, and where both molecular and antigen testing is available, urine LAM has minimal incremental yield over GeneXpert MTB/RIF.

SARJ VOL. 23 NO. 3 2017

91

NOTES

92 SARJ VOL. 23 NO. 3 2017

NEW

Antistatic Chamber

small, solid and effective

NEW aluminium Aspen is proud to announce the launch of an innovative new antistatic holding chamber, as an addition to our respiratory portfolio.

VORTEX® inhalation aid is suitable in providing: (1) • High lung deposition, low throat deposition • High dosage consistency • Disinfectable, ergonomic SmartTouch masks Description (2)

Indication (2)

VORTEX® with a mouthpiece

VORTEX® with mouthpiece and baby mask ‘Ladybug’

To be used in conjunction with medication sprays or "metered dose inhalers" in the treatment of respiratory tract diseases.

VORTEX® with mouthpiece and child mask ‘Frog’

Nappi Code

SEP (Excl VAT)

SEP (Incl VAT)

216379001

R 267,27

R 304,69

216375001

R 291,66

R 332,50

216376001

R 291,66

R 332,50

The NEW VORTEX® aluminium chamber inhalation aid!

Attach. Breathe. Relax.

S3 FLIXOTIDE® 50/125/250 INHALER CFC-FREE. Reg No.: 35/21.5.1/0377-0082/3. Delivers 50/125/250 µg of fluticasone propionate per actuation. INDICATIONS: Prophylactic management of atopic asthma in adults and children of 6 years and older. CONTRA-INDICATIONS: History of allergy to any of its components.

PREGNANCY AND LACTATION: Safety not established. DOSAGE AND DIRECTIONS FOR USE: For inhalation use only. Should be taken regularly even when asymptomatic. The onset of therapeutic effect is 4 to 7 days. Should not be used for relief in acute attacks but for routine long term management. Patients will require a fast- and short-acting inhaled bronchodilator to relieve acute symptoms. If patients find that relief with short-acting bronchodilator treatment becomes less effective or they need more inhalations than usual, medical attention must be sought. Adults and children over 16 years of age: 100-1000 µg twice daily.

Starting dose should be appropriate for severity of the disease. Dose may be adjusted until control is achieved or reduced to the minimum effective dose, according to the individual response. Children over 6 years of age: 50-100 µg twice daily. The dose may be adjusted until control is achieved and should be reduced to the minimum effective dose according to the individual response. Special patient groups: No dose adjustment in elderly patients. For the transfer of patients being treated with oral corticosteroids: Patients treated with systemic steroids for long periods of time or at a high dose may have adrenocortical suppression and adrenocortical function should be monitored regularly and their dose of systemic steroid reduced cautiously. After approximately a week, gradual withdrawal of the systemic steroid may be commenced. Decrements in dosages should be appropriate to the level of maintenance systemic steroid, and introduced at not less than weekly intervals. In some patients on oral corticosteroids the dose reduction or replacement with inhaled corticosteroids may not be possible. Some patients feel unwell in a non-specific way during the withdrawal phase despite maintenance or even improvement of the respiratory function. They should be encouraged to persevere with inhaled fluticasone propionate and to continue withdrawal of systemic steroid, unless there are objective signs of adrenal insufficiency. SIDE EFFECTS AND SPECIAL PRECAUTIONS: Treatment should not be stopped abruptly as adrenal insufficiency may be precipitated. Candidiasis of the mouth and throat (thrush) may occur. May be helpful to rinse out mouth with water after use. Symptomatic candidiasis can be treated with topical anti-fungal therapy whilst continuing treatment. Hoarsenes. Paradoxical bronchospasm with an immediate increase in wheezing. Treat immediately with a fast-acting inhaled bronchodilator.

Treatment should be discontinued immediately, the patient assessed, and if necessary alternative therapy instituted. Cutaneous hypersensitivity. Systemic corticosteroid effects may occur. Patients transferred from other inhaled steroids or oral steroids remain at risk of impaired adrenal reserve for a considerable time after transferring to inhaled fluticasone propionate. Increasing use to control symptoms indicates deterioration of asthma control and patient should be reassessed. Sudden and progressive deterioration in asthma control is potentially life-threatening and may have several causes. Consideration should be given to increasing corticosteroid dosage if not caused by otherwise treatable causes of deterioration. Severe asthma requires regular medical assessment as death may occur. Sudden worsening of symptoms may require increased corticosteroid dosage which should be administered under urgent medical supervision. Patients weaned off oral steroids whose adrenocortical function is still impaired should carry a steroid warning card indicating that they may need supplementary systemic steroid during periods of stress, e.g. worsening asthma attacks, chest infections, major intercurrent illness, surgery, trauma, etc. Inhaled therapy may unmask underlying eosinophilic conditions (e.g. Churg Strauss syndrome). These cases have usually been associated with reduction or withdrawal of oral corticosteroid therapy. Similarly replacement of systemic steroid treatment with inhaled therapy may unmask allergies such as allergic rhinitis or eczema previously controlled by the systemic drug. These allergies should be symptomatically treated with antihistamine and/or topical preparations, including topical steroids. Patients in a medical or surgical emergency, who require high doses of inhaled steroids and/or intermittent treatment with oral steroids, are at risk of impaired adrenal reserve. The extent of the adrenal impairment may require specialist advice before elective procedures. The possibility of residual impaired adrenal response and elective situations likely to produce stress and appropriate corticosteroid treatment must be considered. Lack of response or severe exacerbations of asthma should be treated by increasing the dose of inhaled fluticasone propionate or by giving a systemic steroid and/or an antibiotic if there is an infection. Special care is necessary in patients with active or quiescent pulmonary tuberculosis. Patients on corticosteroid therapy may have adrenocortical suppression. MANAGEMENT OF OVERDOSAGE: Monitoring of adrenal reserve may be indicated. Treatment with inhaled fluticasone propionate should be continued at a dose sufficient to control asthma. APPLICANT: GlaxoSmithKline South Africa (Pty) Ltd; (Co. reg. no.1948/030135/07). 39 Hawkins Avenue, Epping Industria 1, Cape Town, 7460.

Reference: 1. Laube BL, Janssens HM, de Jongh FHC, et al. What the pulmonary specialist should know about the new inhalation therapies. Eur Respir J 2011;37:1308-1331. 2. VORTEX® package insert.

For full prescribing information, please refer to the package inserts approved by the Medicines Regulatory Authority.

All adverse events should be reported by calling the Aspen Medical Hotline number or directly to GlaxoSmithKline on +27 11 745 6000.

ZAF/FP/0004/15a A19615 04/16

The South African Respiratory Journal acknowledges with thanks the invaluable sponsorship of the following companies: Aspen GSK Division Bayer Healthcare