SARJ Vol 22, No 1 (2016)

SouthAfrican African South

Respiratory Respiratory Journal Journal VOLUME 22

|

NUMBER 1

South African

Respiratory

Journal

OFFICIAL JOURNAL OF THE S.A. THORACIC SOCIETY

|

MARCH 2016

FOXAIR

everyone and it’s yours to give

Air is for

Why wait to raise expectations in your patients?

S4 FOXAIR® 50/100, 50/250 and 50/500 ACCUHALER® - 42/21.5.4/0581; 0582; 0583. Each blister contains a mixture of salmeterol xinafoate equivalent to 50 µg of salmeterol and microfine fluticasone propionate (100 µg, 250 µg or 500 µg). S4 FOXAIR® 25/50, 25/125 and 25/250 INHALER - 42/21.5.4/0244; 0245; 0246. Each single actuation provides salmeterol xinafoate equivalent to 25 µg of salmeterol and fluticasone propionate (50, 125 or 250 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 A16773 08/13

THE SOUTH AFRICAN

RESPIRATORY JOURNAL VOLUME 22 | NUMBER 1 | MARCH 2016

CONTENTS EDITORIAL 2

Difficulties with interpreting cytokine/chemokine values in disease states Prakash M Jeena

ORIGINAL RESEARCH

3

Cytokine profile and clinical correlates in HIV-exposed infants with severe (hypoxic) pneumonia R J Green, A Terblanche, P Becker, P Rheeder, D F Wittenberg, R Anderson, R Masekela

REVIEWS 7 Tumours of the chest in children: A review A C Jeevarathnum, A van Niekerk, D Parris, K De Campos, W Wijnant, X Deadren, A Büchner, F Omar, D Reynders, R J Green 12 Eosinophilic lung diseases: A review

K Dudgeon

CASE REPORTS

Pneumocystis jiroveci and cytomegalovirus co-infection in an immunocompromised patient

D Simon, K E Greyling, E M Irusen, J Rigby, J J Taljaard, C F N Koegelenberg 22 A paradoxical cause of hypoxia and orthodeoxia in a stroke patient

P Soma, D Joseph, S Ahmad, S Ellemdin 23 Superior mediastinal masses in children: Two cases of lymphoma A C Jeevarathnum, A van Niekerk, D Parris, K De Campos, W Wijnant, X Deadren, A Büchner, F Omar, D Reynders, R J Green

20

26

BREATH-TAKING NEWS

29

WHO’S WHO

30

PRODUCT NEWS

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. Adithya Cattamanchi - USA Prof. Fan Chung - UK Prof. GB Migliori - Italy Prof. Surendra Sharma - India Prof. Wing Wai Yew - China PRESIDENT SA THORACIC SOCIETY Prof. U Lalloo

HMPG

CEO AND PUBLISHER Hannah Kikaya Email: hannah.kikaya@hmpg.co.za EXECUTIVE EDITOR Bridget Farham MANAGING EDITOR Ingrid Nye SCIENTIFIC EDITOR Simon Nye PRODUCTION MANAGER Emma Jane Couzens DTP AND DESIGN Carl Sampson HEAD OF SALES AND MARKETING Diane Smith | Tel. 012 481 2069 Email: dianes@hmpg.co.za JOURNAL ADVERTISING Charles Duke, Benru de Jager, Reneé van der Ryst, Ladine van Heerden ONLINE SUPPORT Gertrude Fani | Tel. 072 463 2159 Email: publishing@hmpg.co.za FINANCE Tshepiso Mokoena

The Editor The South African Respiratory Journal PO Box 13725, Mowbray, 7705 Telephone: 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: Suites 9 & 10, Lonsdale Building, Gardener Way, Pinelands, 7405 | 021 532 1281 All letters and articles for publication must be submitted online at www.sarj.org.za E-mail: publishing@hmpg.co.za

HMPG BOARD OF DIRECTORS Prof. M Lukhele (Chair), Dr M R Abbas, Dr M J Grootboom, Mrs H Kikaya, Prof. E L Mazwai, Dr M Mbokota, Dr G Wolvaardt PRINTED BY TANDYM PRINT

EDITORIAL

Difficulties with interpreting cytokine/chemokine values in disease states This edition of the journal features an interesting article on cytokine/chemokine profiles in HIV-exposed infants with hypoxic pneumonia. Whilst there are several interesting aspects, this paper highlights several aspects of biomarker measurements relevant to translational research. Thus, a closer look at the biomarker measurements is warranted.

Cytokines, chemokines and adhesion molecules are host-specific biomarkers that are present in health in humans. They are present at baseline concentrations (mean and range) and have natural variability. In disease states, there is either a premorbid change in biomarker levels that predisposes a person to acquisition of an illness or a responsive variation in concentrations with development of the illness. Biomarkers can therefore be utilised as diagnostic tools, and/or surrogate markers for monitoring the natural progression or response to an intervention. The measurement of biomarkers at different time intervals is likely to yield a pattern of response to the natural disease or to the intervention. Consequently, the measurement of cytokine and chemokine values in a cross-sectional manner without defining exactly the reason for the time point makes interpretation difficult.

Another concern with the measurement of biomarkers relates to the robustness of the technical analysis. While the Luminex multiplex technology by Bio-Rad Laboratory has been validated and accredited, there is a need to perform quality assurance of the results. Samples should be processed in duplicate and verified against those with predefined concentrations of cytokines/chemokines. This exercise is expensive.

A further consideration is the quality of the sample obtained and the preservation processes for storage of samples prior to testing.

Biomarker levels in different bodily fluids are likely to be dependent on the site, nature and severity of the disease/illness at that location.

In addition, biomarker values in different fluids may be impacted by the quality of the sample obtained. Sufficient good-quality sample is

2 SARJ VOL. 22 NO. 1 2016

necessary to ensure accurate readings. Measurement of whole blood and plasma protein biomarkers (proteomics) can vary with the degree of anaemia or viscosity of the sample. Sputum biomarkers can vary due to the quality of the sample obtained. Breath volatile organic compounds (metabolites) measured by gas chromatography and mass spectrometry can vary with changes in flow volumes, while urine biomarkers could be affected by its composition and the presence of inhibitors. In order to adjust for these variables, a clear understanding about the degradability of biomarkers is necessary.

Disease states often induce multiple rather than single changes in biomarkers. Co-infections or co-diseases are likely to have an agonistic or antagonistic impact on levels of biomarkers measured.

Acute-on-chronic diseases, such as exacerbations in bronchiectasis, could alter values of cytokines/chemokines obtained as compared with stable chronic disease. Subclinical infection will impede the correct interpretation of values obtained. The cost of measuring an individual specific biomarker is high, so most laboratory platforms utilise multiplex cytokine technology. Obtaining results about several cytokines/chemokines often confuses interpretation as the theoretical explanations are not congruent. The delay in obtaining the result often mitigates against its value as the patient’s condition could have changed from the time of sampling. Point-of-care measurement of cytokines/chemokines would be useful in defining a response to the result but much more unequivocal data to guide clinical practice are required. Finally, cytokine responses of the host are modified by the levels of the stimulus provided by the pathogen, which is dependent on their load, the rate of reproduction and the rate of transcription.

Prakash M Jeena Department of Paediatrics and Child Health, University of KwaZuluNatal, Durban, South Africa S Afr Respir J 2016;22(1):2. DOI: 10.7196/SARJ.2016.v22i1.65

ORIGINAL RESEARCH

Cytokine profile and clinical correlates in HIV-exposed infants with severe (hypoxic) pneumonia R J Green,1 PhD, DSc; A Terblanche,1 FCPaed (SA), Cert Gastroenterol (SA)(Paed); P Becker,2 PhD; P Rheeder,3 PhD, MMed, FCP (SA); D F Wittenberg,1 MD; R Anderson,4 PhD; R Masekela,1 PhD Department of Paediatrics and Child Health, School of Medicine, Faculty of Health Sciences, University of Pretoria, South Africa Department of Biostatistics, Medical Research Council of South Africa, Pretoria, South Africa 3 Division of Clinical Epidemiology, School of Health Systems and Public Health, Faculty of Health Sciences, University of Pretoria, South Africa 4 Medical Research Council Unit for Inflammation and Immunity, Department of Immunology, School of Medicine, Faculty of Health Sciences, University of Pretoria and Tshwane Academic Division of National Health Laboratory Service, Pretoria, South Africa 1 2

Corresponding author: R J Green (robin.green@up.ac.za)

Background. Severe pneumonia in infants who are HIV-infected is a common problem in many parts of the developing world, especially sub-Saharan Africa. What has been missing from previous studies of severe pneumonia in HIV-infected infants, however, is a description of the host inflammatory response and cytokine/chemokine profile that accompanies this disease.

Objective. To describe the cytokine profiles associated with severe hypoxic pneumonia in HIV-infected and -exposed infants.

Methods. In a cohort of HIV-exposed children diagnosed clinically with severe hypoxic pneumonia, paired serum and sputum cytokines were tested. A control group of HIV-infected children with bronchiectasis contributed matching controls.

Results. A total of 100 infants (mean age 2.8 months) with a clinical diagnosis of severe hypoxic pneumonia were included in this study.

IP-10 was markedly elevated in both sputum (mean 560.77 pg/mL) and serum (mean 9 091.14 pg/mL), while IL-10 was elevated in serum (mean 39.55 pg/mL), with both these cytokines being significantly higher than in stable children with HIV-associated bronchiectasis.

Conclusion. This study of HIV-exposed infants with severe hypoxic pneumonia suggests that IP-10 and IL-10 are associated with hypoxic lung disease in infants. However, further investigation of this association is required.

S Afr Respir J 2016;22(1):3-6. DOI: 10.7196/SARJ.2016.v22i1.60

Severe pneumonia in infants who are HIV-infected is a common problem in many parts of the developing world, especially sub-Saharan Africa. It has emerged that the condition of severe hypoxic pneumonia in early infancy is a disease of many causes, most occurring together in the individual patient.[1-3] A frequent cause of severe pneumonia in infants is Pneumocystis jiroveci. This condition is usually diagnosed clinically and managed as pneumocystis pneumonia (PCP) in the regions of the world where HIV-infected children live. Today it has become possible to make a microbiological diagnosis of P. jiroveci based on polymerase chain reaction (PCR) testing of airway secretions. However, in the developing world such testing is largely unavailable and the clinical condition still poses an enormous problem.

P. jiroveci is a fungal organism that has a predilection for the immune-compromised host and is a common pathogen in HIVinfected infants. The term PCP was retained after Pneumocystis carinii was taxonomically renamed jiroveci.[4] Since the earliest reports of HIV infection, PCP has been recognised as a severe form of acute pneumonia. The disease may occur at any age but is particularly common in early infancy.[5] PCP is recognised clinically by a distinct set of common criteria: hypoxic pneumonia, few pulmonary crackles, an interstitial appearance on chest radiographs and an elevated lactate dehydrogenase (>500 U/L).[6,7] The case fatality rate from PCP is 100% if not treated with trimethoprim-sulphamethoxazole (TMP-SMX).[8]

However, where TMP-SMX prophylaxis is employed alone, mortality is not significantly reduced.[9] Because the disease often causes severe hypoxia, these children would benefit from paediatric intensive care admission. Admitting infants with PCP to an intensive care unit, in a resource-limited setting, has created a number of ethical dilemmas for pediatricians including the historically poor outcome for these patients and the pressure on scarce resources.[10] Cytomegalovirus (CMV) is now recognised as an important copathogen of severe pneumonia in infants and may be the organism driving mortality in this form of pneumonia.[1-3] Treatment of this form of severe pneumonia with a combination of TMP-SMX and antiviral agents has had mixed success.[1-3] Some studies report improved survival with use of the antiviral agent ganciclovir. [2,3] Despite the presence of P. jiroveci and CMV a number of other pathogens also cause and contribute to severe hypoxic pneumonia in infants. What has been missing from previous studies of severe pneumonia in HIV-infected infants, however, is a description of the host inflammatory response and cytokine/chemokine profile that accompanies this disease. It is hoped that a better understanding of the host response and associated clinical correlates may aid in seeking better therapeutic options for these very ill children who frequently die.

The objective of this study was to document the cytokine/ chemokine profile of a group of young infants with all-cause, severe hypoxic pneumonia.

SARJ VOL. 22 NO. 1 2016

3

ORIGINAL RESEARCH

Methods

As part of a double-blind randomised controlled trial to assess the value of systemic steroids, conducted among infants l <18 months of age admitted with severe hypoxic pneumonia at Pretoria Academic, Kalafong and Witbank hospitals, paired sputum and serum samples were collected for cytokine analysis. The only criterion was infants with hypoxic pneumonia.

Patients were enrolled according to study protocol by the admitting doctor.

A clinical diagnosis of severe hypoxic pneumonia was made in patients with the following features: • cough • tachypnoea • hypoxia (peripheral oxygen saturation <90% in room air) out of proportion to the clinical findings on auscultation • positive HIV-1 enzyme-linked immunosorbent assay test (ELISA) (HIV-exposed).

Ethical approval was obtained from the Research Ethics committee of the University of Pretoria, Faculty of Health Sciences (Protocol number 80/2004).

The parent(s) or guardian gave informed written consent for participation in the study, as well as for HIV testing.

Patients received either prednisone 2 mg/kg/day or placebo for 7 days. Standard antibiotic regimens including co-trimoxazole were carried out in accordance with usual practice.

The following baseline investigations were performed on admission: • chest X-ray (CXR) • full blood count, differential count and platelets, C-reactive protein (CRP), lactate dehydrogenase (LDH), total protein and albumin, aspartate aminotransferase (AST) and alanine aminotransferase (ALT), CD4 count, blood culture and HIV ELISA • nasopharyngeal aspirates for respiratory viruses including CMV and PCP immunofluorescence were collected. Secretions were also tested for bacterial and tuberculosis (TB) culture.

Daily monitoring included clinical examination, recording of temperatures, oxygen saturation and oxygen requirement, and development of new features. The number of days to achieve mean oxygen saturation >90% in room air was recorded using standard pulse oximetry.

The primary study endpoint was in-hospital survival. Secondary outcome was time from admission to the first day of mean oxygen saturation >90% in room air.

Serum and sputum cytokines (interleukin-1beta (IL-1β), interleukin-2 (IL-2), interleukin-4 (IL-4), interleukin-10 (IL-10), interleukin-12 (IL-12), interleukin-13 (IL-13), tumour necrosis factor alpha (TNF-α) and interferon gamma-inducible protein-10 (IP-10)) were measured, on admission, using the Bio-Plex suspension bead array system (Bio-Rad Laboratories Inc., USA) which utilises luminex Xmap multiplex technology to enable simultaneous detection and quantitation of multiple different analytes in a single sample. A group of children with HIV-associated bronchiectasis provided serum and sputum cytokine samples and these were used as control samples since this is another disease entity where HIV infection is present.

Data analysis Statistical analysis was performed using Stata Release 11.0 (Statacorp LP, USA). Descriptive statistics were used to describe mean/median

4 SARJ VOL. 22 NO. 1 2016

and standard deviation (SD) of clinical, laboratory and cytokine variables. The Fisher exact test was used for categorical variables and the Mann-Whitney U-test for non-parametric variables. Due to the skewed distribution of the cytokine values in both groups, the groups were compared on the logarithmic scale using the Student’s t-test. If the assumptions for the t-test were not met, Wilcoxon’s rank sum test was applied. The geometric mean and its 95% confidence interval (CI) was reported as summary statistics again because of ‘skewed data’.

Before transformation to the logarithmic scale 1 was added to the observed cytokine value so as not to lose the zero values that were observed. In reporting the summary statistics the adjustment was corrected. Statistical testing was at p<0.05 level of significance.

Results

A total of 100 patients with a clinical diagnosis of severe hypoxic pneumonia were included in this study. There were 31 control patients (mean age 8 years) with HIV-associated bronchiectasis. The outcome of patients who received steroids compared with those who did not has been reported previously.[4] The mean (range) age of this group of infants admitted to hospital with pneumonia is 2.8 (1 - 12) months. With regard to the clinical presentation and laboratory findings of infants with pneumonia the major findings are reflected in Table 1.

Table 1. Clinical and laboratory parameters of HIV-exposed infants admitted with severe pneumonia Variable

Mean

Range

SD

Cough duration (days)

8.7

0 - 60

10.23

Oxygen saturation (%)

70.9

20 - 92

14.52

Length (cm)

56.3

43 - 72

5.09

Weight (kg)

4.80

2.7 - 8.5

1.10

Axillary temperature (°C)

37.4

36 - 40

0.957

CD4 % of total lymphocytes

18.7

0.19 - 64

12.43

CD4 absolute (cells/mL)

979.5

20 - 5 247

908.75

CRP (μg/mL)

17.8

1 - 602

61.82

LDH (IU/L)

916.1

329 - 6 249

671.86

The clinical chest findings in this cohort of infants with pneumonia were tachypnoea in 100 (100%), respiratory recessions in 100 (100%), subcostal recession in 94 (94%), absence of pulmonary crackles in 71 (71%), clinical hyperinflation in 22 (22%) and wheezing in 1 (1%).

The radiological findings in this cohort of infants with pneumonia were pulmonary infiltrates in 85 (85%), interstitial pattern in 86 (86%) and lymphadenopathy in the hilar region in 25 (25%).

The identified pathogens were P. jiroveci, CMV, respiratory sunctial virus (RSV), adenovirus, parainfluenza virus and influenza virus in 5, 2, 8, 1, 0 and 2 infants, respectively. Bacterial pathogens and TB were not identified in any infants.

The mean (range, SD) for the measured cytokines in sputum were IL-10: 4.01 (0 - 32.68, 5.84), IL-1β: 278.94 (0.1 - 1 902.14, 482.34), IL-2: 13.01 (0 - 199.33, 30.19), IL-4: 1.32 (0 - 20.76, 3.16), IL-12: 7.60 (0.19 - 5 054, 7.31), IL-13: 7.13 (0 - 60.19, 8.71), IP-10: 560.77 (0 - 6 975.45, 1 024.43), TNFα: 29.44 (0 - 663.02, 89.28).

ORIGINAL RESEARCH The mean (range, SD) for the measured cytokines in serum were IL-10: 39.55 (3.15 - 308.480, 47.12 ), IL-1β: 25.32 (0 - 743.5, 98.91), IL-2: 146.24 (0 - 1 070.04, 259.83), IL-4: 1.58 (0 - 15.83, 3.71), IL-12: 39.75 (0 1 112.57, 39.75), IL-13: 5.14 (0 - 64.85, 12.51), IP-10: 9 091.14 (0 - 78 172.52, 11 381.98), TNFα: 104.06 (0 - 1 201.12, 213.47).

For the four cytokines that appeared to be consistently elevated in either sputum or serum in infants with severe pneumonia, namely IL10, IL-Iß, IP-10 and TNFα, comparison with children with HIV-associated bronchiectasis (mean age 8 years) is reflected in Table 2.

Discussion

The age of this cohort of infants with severe hypoxic pneumonia is in keeping with previous publications suggesting that severe hypoxic pneumonia occurs in very young infants (age 2 - 3 months).[9] This fact is of critical importance to preventive strategies and policies. Only an effective prevention of mother-to-child (PMTC) programme will solve this problem. Early HIV detection in childhood and vaccine programmes are not a solution, as the disease occurs so early in life.

The clinical findings support previous research indicating that this form of pneumonia in young HIV-exposed children is associated with severe hypoxia and relatively few pulmonary adventitious sounds.[11] LDH has been suggested as a diagnostic test for PCP, with levels >500 IU/L taken as positive evidence.[6] Whilst not all the infants in this study had a LDH >500 IU/L, most did. This result does not support a specific aetiological agent in the cause of this form of pneumonia but rather supports the severe nature of the pulmonary insult. The clinical profile of this disease is distinct in the nature of its severity.

The radiological findings, whilst most usually reflecting an interstitial (or alveolar) pattern, are not universally so. This too has been previously documented.[12] The lack of identifiable organisms causing this disease probably reflects the insensitive nature of the testing methods. This is a weakness of the study, and this study would have been significantly strengthened by newer PCR tests, which unfortunately were not available at the time of the study. This lack of a definitive causative agent responsible for the pneumonia is, however, not a detraction from the major aim of the study, that being to document the cytokine profile of

Table 2. Comparison of cytokine values for HIV-infected infants with severe pneumonia and HIV-related bronchiectasis. Geometric mean (95% CI) Cytokine / Group

Geometric mean

95% CI

Severe pneumonia

1.93

1.30; 2.74

Bronchiectasis

0.78

0.51; 1.10

Severe pneumonia

40.12

22.63; 70.56

Bronchiectasis

692.07

372.36; 1 285.54

Severe pneumonia

154.33

90.24; 263.45

Bronchiectasis

12.76

6.09; 25.71

Severe pneumonia

5.30

3.33; 8.16

Bronchiectasis

11.57

7.74; 179.89

Severe pneumonia

24.86

19.78; 31.19

Bronchiectasis

4.48

3.54; 5.63

Severe pneumonia

2.97

1.69; 4.86

Bronchiectasis

4.38

2.17; 8.15

Severe pneumonia

5 372.59

3 877.70; 7 443.63

Bronchiectasis

4 144.67

292.90; 5 856.94

Severe pneumonia

14.39

8.06; 25.15

Bronchiectasis

5.89

2.05; 4.56

p-value*

Sputum IL-10 0.0196

Sputum IL-1β <0.001

Sputum IP-10 <0.001

Sputum TNFα 0.0402

Serum IL-10 <0.001

Serum IL-1β 0.3815

Serum IP-10 0.0078

Serum TNFα 0.1009

*p-value for the most appropriate statistic of t-test or Wilcoxon’s rank sum test.

HIV-exposed infants with severe hypoxic pneumonia.

Normal values for cytokines and chemokines in health and disease are not readily available, and this is a limitation of this study. There is evidence that in healthy children and infants the cytokines measured in this study should be zero or at most very low values.[13] The lack of a definitive control group of normal children against which the cytokines are measured is a weakness of this study. However, in order to nullify the effect of HIV infection per se on the cytokine changes, it was decided to use a group of children with another form of HIVrelated pathology, but one in which acute infection was not present. Hence the value of children with HIV-associated bronchiectasis as a control group. Clearly this is not an ideal comparator group because of differences in age at presentation and degrees of disease severity.

This limitation is noted. Subclinical infection is, however, a possibility and might influence the results. The cytokine results suggest that the major cytokines associated with severe hypoxic pneumonia in very young, HIV-infected, infants are IL-10 and IP-10. Neither cytokine was significantly elevated in the control group, suggesting that the elevated values are due to the effect of acute and severe pneumonia in infants.

IL-10 is a cytokine that has important anti-inflammatory properties.[14] Coded for by the IL-10 gene, this cytokine is produced mainly by monocytes and to some extent by lymphocytes.[15] It has a major function in down-regulating the expression of Th1 cytokines.[14] There is a paucity of data on the presence of IL-10 in paediatric lung disease, especially pneumonia. However, in a study of children with severe sepsis or pneumonia, IL-10 was found to be elevated in the serum

SARJ VOL. 22 NO. 1 2016

5

ORIGINAL RESEARCH of the children with severe sepsis but not pneumonia.[16] In other paediatric pulmonary conditions there is evidence that IL-10 is elevated in RSV infection,[17] bronchopulmonary dysplasia[18] and Mycoplasma pneumoniae pneumonia.[19] An adult study of patients with communityacquired pneumonia suggests that IL-10 functions as an acute-phase reactant.[20] The finding of elevated sputum and serum IL-10 in the current study of infants with severe pneumonia is a new finding and suggests that the anti-inflammatory defences of the HIV-infected infant are mobilised early after the onset of severe pneumonic pathology. It is also possible that this cytokine may function in a pro-inflammatory way in this condition.

IP-10 is a chemokine that is secreted by several cell types, including monocytes, endothelial cells and fibroblasts, in response to INF-y.[21] It functions as a chemoattractant for macrophages, T-cells, NK-cells and dendritic cells and also has a number of newly identified functions including promotion of T-cell adhesion to endothelial cells, antitumour activity and inhibition of angiogenesis.[22,23] It has not previously been associated with a specific form of pneumonia in children. High levels of this chemokine have been shown to be associated with a poorer outcome in HIV-infected individuals with hepatitis C viral (HCV) co-infection.[24] IP-10 has been documented as a better test than both interferon gamma-based QuantiFERON TB Gold assays and tuberculin skin tests for diagnosing TB in HIV-infected individuals.[25] HCV was not measured in our study but all children had normal levels of liver enzymes. TB was not seen in our children. Elevation of this chemokine in infants with severe pneumonia may reflect significant stimulation of monocytes, in keeping with the elevated values of IL-10. It may have pro- or anti-inflammatory activity in this disease state. These functions, however, require more extensive study.

Previous studies have attempted to characterise the cytokine profile of P. jiroveci-infected individuals. These studies suggest that the actual cause of the immunosuppression predisposing to the infection may have as much impact on the cytokine profile as the organism itself.[26,27] This latter study suggests that P. jiroveci infection is associated with reduced macrophages in alveoli and elevated IL-6. However, IP-10 was not measured in that study.

IL-1β and TNFα are found to be lower in infants with severe pneumonia than the comparator group of children with bronchiectasis. The reason for this finding is unknown and should be investigated further.

Study limitations This study was not without significant limitations. Some of these include lack of definition of severity groups within both sample populations, failure to standardise sample collection, timing to disease onset and most importantly that the age of the two sample populations are markedly different.

Conclusion

This study of HIV-exposed infants with severe hypoxic pneumonia suggests that IL-10 and IP-10 are associated with hypoxic pneumonia in infants. However, further investigation of this association is required.

References

1. Morrow BM, Hsaio 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):535539. [http://dx.doi.org/10.1097/INF.0b013e3181ce871e]

6 SARJ VOL. 22 NO. 1 2016

2. Zampoli M, Morrow B, Hsiao NY, Whitelaw A, Zar HJ. Prevalence and outcome of cytomegalovirus associated pneumonia in relation to human immunodeficiency virus infection. Pediatr Infect Dis J 2011;30(5):413-417. [http://dx.doi.org/10.1097/ INF.0b013e3182065197] 3. Goussard P, Kling S, Gie RP, et al. CMV pneumonia in HIV-infected ventilated infants.

Pediatr Pulmonol 2010;45(7):650-655. [http://dx.doi.org/10.1002/ppul.21228] 4. Stringer JR, Beard CB, Miller RF, Wakefield AE. A new name (Pneumocystis jiroveci) for pneumocystis from humans. Emerg Infect Dis 2002;8(9):891-896. [http://dx.doi. org/10.3201/eid0809.020096] 5. Jeena P. The role of HIV on respiratory tract infection in sub-Saharan Africa. Int J Tuberc Lung Dis 2005;9(7):708-715.

6. Fatti GL, Zar HJ, Swingler GH. Clinical indicators of Pneumocystis jiroveci pneumonia (PCP) in South African children infected with human immunodeficiency virus. Int J Infect Dis 2006;10(4):282-285. [http://dx.doi.org/10.1016/j.ijid.2005.06.007] 7. Terblanche AJ, Green RJ, Rheerder P, Wittenberg DF. Adjunctive corticosteroid treatment of clinical Pneumocystis jiroveci pneumonia in infants less than 18 months of age - a randomized controlled trial. S Afr Med J 2008;98(4):287-290.

8. Madhi SA, Cutland C, Ismail K, O’ Reilly C, Mancha A, Klugman KP. Ineffectiveness of trimethoprim-sulfamethoxazole prophylaxis and the importance of bacterial and viral coinfections in African children with Pneumocystis carinii pneumonia. Clin Infect Dis 2002;35(9):1120-1126. [http://dx.doi.org/10.1086/343049] 9. Kitchin O, Masekela R, Becker P, Moodley T, Risenga SM, Green RJ. Outcome of HIV exposed and infected children admitted to a Pediatric Intensive Care Unit for respiratory failure. Pediatr Crit Care Med 2012;13(5):516-519. [http://dx.doi. org/10.1097/PCC.0b013e31824ea143] 10. Jeena MP, McNally LM, Stobie M, Coovadia HM, Adhikari MA, Petros AJ. Challenges in the provision of ICU services to HIV infected children in resource poor settings: A South African case study. J Med Ethics 2005;31(4):226-230. [http://dx.doi.org/10.1136/jme.2003.004010] 11. Zar HJ, Dechaboon A, Hanslo D, Apolles P, Magnus KG, Hussey G. Pneumocystis carinii pneumonia in South African children infected with human immunodeficiency virus. Pediatr Infect Dis J 2000;19(7):603-607.

12. Pitcher RD, Daya R, Beningfield SJ, Zar HJ. Chest radiographic presenting features and radiographic progression of pneumocystis pneumonia in South African children.

Pediatr Pulmonol 2011;46(10):1015-1022. [http://dx.doi.org/10.1002/ppul.21465] 13. Okazaki K, Kondo M, Kato M, et al. Serum cytokine and chemokine profiles in neonates with meconium aspiration syndrome. Pediatrics 2008;121(4):748-753.

14. Said EA, Dupuy FP, Trautmann L, et al. Programmed death-1-induced interleukin-10 production by monocytes impairs CD4+ T cell activation during HIV infection. Nat Med 2010;16(4):452-459. [http://dx.doi.org/10.1038/nm.2106] 15. Eskdale J, Kube D, Tesch H, Gallagher G. Mapping the human IL10 gene and further characterization of the 5’ flanking sequence. Immunogenetics 1997;46(2):120-128.

16. Chen DH, Li YM, Lan SL, et al. The level and clinical significance of Toll-like receptor 4 in children with severe sepsis. Zhongguo Wei Zhong Bing Ji Jiu Yi Xue 2011;23(8):475-477.

17. Midulla F, Tromba V, Lo Russo L, et al. Cytokines in the nasal washes of children with respiratory syncytial virus bronchiolitis. Int J Immunopath Pharmacol 2006;19(1):231-235.

18. Garingo A, Tesoriero L, Cayabyab R, et al. Constitutive IL-10 expression by lung inflammatory cells and risk for bronchopulmonary dysplasia. Pediatr Res 2007;61(2):197-202. [http://dx.doi.org/10.1203/pdr.0b013e31802d8a1c] 19. Pang HX, Qiao HM, Cheng HJ, Zhang YF, Liu XJ, Li JZ. Levels of TNF-α, IL-6 and IL-10 in bronchoalveolar fluid in children with Mycoplasma pneumoniae pneumonia.

Zhongguo Dang Dai Er Ke Za Zhi 2011;13(10):808-810.

20. Endeman H, Meijvis SC, Rijkers GT, et al. Systemic cytokine response in patients with community-acquired pneumonia. Eur Respir J 2011;37(6):1431-1438.

21. Luster AD, Unkeless JC, Ravetch JV. Gamma-interferon transcriptionally regulates an earlyresponse gene containing homology to platelet proteins. Nature 1985;315(6021):672-676.

22. Dufour JH, Dxiejman M, Liu MT, Leung JH, Lane TE, Luster AD. IFN-gammainducible protein 10 (IP-10; CXCL 10)-deficient mice reveal a role for IP-10 in effector T cell generation and trafficking. J Immunol 2002;168(7):3195-3204.

23. Angiolillo AL, Sgadari C, Taub DD, et al. Human interferon-inducible protein 10 is a potent inhibitor of angiogenesis in vivo. J Exp Med 1995;182:155-162.

24. Falconer K, Askarieh G, Weis N, Hellstrand K, Alaeus A, Lagging M. IP-10 predicts the first phase decline of HCV RNA and overall viral response to therapy in patients co-infected with chronic hepatitis C virus infection and HIV. Scand J Infect Dis 2010;42:11-12.

25. Syed Ahamed Kabeer B, Sikhamani R, Raja A. Comparison of interferon gamma-inducible protein-10 and interferon gamma-based QuantiFERON TB Gold assays with tuberculin skin test in HIV-infected subjects. Diagn Microbiol Infect Dis 2011;71(3):236-243.

26. Tasaka S, Kobayashi S, Kamata H, et al. Cytokine profiles of bronchoalveolar lavage fluid in patients with pneumocystis pneumonia. Microbiol Immunol 2010;54(7):425-433.

27. Iriart X, Witkowski B, Courtais C, et al. Cellular and cytokine changes in the alveolar environment among immunocompromised patients during Pneumocystis jirovecii infection.

Med Mycol 2010;48(8):1075-1087. [http://dx.doi.org/10.3109/13693786.2010.484027]

REVIEW

Tumours of the chest in children: A review A C Jeevarathnum, MB BCh, FCPaed (SA), Dip Allergy (SA), MMed, Cert Paed Pulm (SA), European Respiratory Diploma; A van Niekerk, MB BCh, MMed; D Parris, BSc, MB BCh, FCPaed (SA), Dip Allergy (SA); K De Campos, MB ChB, MMed, Dip Allergy (SA); W Wijnant, MD Paed, Dip Allergy (SA), Cert Paed Pulm (SA); X Deadren, MB ChB, FCPaed (SA), MMed; A Büchner, MB ChB, DCH (SA), FCPaed (SA), MMed, Dip Pall Med, Cert Med Oncol (Paed)(SA); F Omar, MB ChB, FCPaed (SA), Cert Paed Med, Onc Paed (SA); D Reynders, MB ChB, FCPaed (SA), MRCPCH, Cert Paed Med, Onc Paed (SA); R J Green, PhD, DSc Department of Paediatrics and Child Health, School of Medicine, Faculty of Health Sciences, University of Pretoria and Steve Biko Academic Hospital, Pretoria, South Africa Corresponding author: A C Jeevarathnum (acjeevarathnum@gmail.com)

S Afr Respir J 2016;22(1):7-11. DOI: 10.7196/SARJ.2016.v22i1.59 Tumours of the chest in children constitute an array of pathology and clinical symptomatology. These tumours can be benign or malignant, cystic or solid. If malignant they can be primary or as a result of secondary metastases.

Collectively, tumours of the chest in children, whether benign or malignant, are very rare. The exact incidence is largely unknown globally. Primary malignant tumours as an entity of their own constitute 0.2% of malignancies in the paediatric population.[1] Non-neoplastic lesions of the lung including bronchogenic cysts, sequestrations, congenital pulmonary airway malformations as well as infective and inflammatory disorders are 60 times more common than neoplastic tumours.[2] A tumour of the chest is considerably difficult to diagnose since patients can be asymptomatic for many years before symptoms evolve.

Even more so, the symptoms are nonspecific and can suggest more common and less sinister pathology.

Clinically patients present with a variety of symptoms that depend largely on the location of the tumour. Airway tumours can be symptomatic or can present with chronic cough, wheeze, haemoptysis, atelectasis or persistent

pneumonia. Secondary malignant parenchymal tumours are likely to be symptomatic from the primary lesion. Anterior mediastinal tumours can cause compression of the large airways or superior vena caval structures.

It stands to reason that the physician needs to have a very high index of suspicion when dealing with these nonspecific signs and symptoms.

This article provides an approach to tumours of the chest and reviews the common aetiology in the different compartments of the chest. The article will focus on common tumours of the airway, lung parenchyma, mediastinum, cardiac and chest wall pathology (Table 1).

Tumours of the airways Benign tumours of the airways Subglottic haemangioma

Haemangiomas of the airway are most commonly seen in the subglottic or upper tracheal region. They are a rare finding in the distal trachea or bronchial region. These patients classically present with features of upper airway obstruction at 3 - 6 months of age. The upper airway obstruction and accompanied respiratory distress can be severe enough to cause

Table 1. Aetiology of tumours of the chest and chest wall by location and differentiation into primary and malignant causes Location

Benign

Malignant

Airway

Subglottic haemangioma Papilloma of the trachea and bronchi

Bronchial adenoma Mucoepidermoid carcinoma

Lung parenchymal

Inflammatory pseudotumour Hamartoma Teratoma

Primary

Pleuropulmonary blastoma Kaposi’s Sarcoma

Secondary

Secondary metastases originating from Wilms’ tumour, osteosarcoma, rhabdomysocarcoma, neuroblastoma and hepatoblastoma

Mediastinal

Teratoma Ectopic/hyperplastic thymus Vascular malformations Bronchogenic cyst

Hodgkin’s lymphoma Non-Hodgkin’s lymphoma Neuroblastoma

Cardiac

Rhabdomyoma Myxoma

Chest wall

Haemangioma Osteoid osteoma Osteochondroma

Rhabdomyosarcoma Osteosarcoma

SARJ VOL. 22 NO. 1 2016

7

REVIEW respiratory failure. The time of presentation is very important as this is the period in which these lesions undergo vascular proliferation and enlarge considerably.[3] Spontaneous resolution is said to occur at about 12 months of age.[4] The diagnosis should be considered in any patient in this age group who presents with hoarseness of voice, recurrent croup and a barking cough.[5] The stridor is typically biphasic in nature. Upper endoscopy reveals asymmetric swelling most commonly in the subglottic region. Up to 20% of these patients may have cutaneous lesions, especially on the face in a beard distribution.[5] Biopsy of these lesions is not recommended due to the very likely complication of bleeding.[4,5] The first-line treatment of these lesions is that of beta-blockers, which may have to be continued up to 12 months of age.[4,5] Other modalities of treatment include systemic and intra-lesion steroids and laser ablation.[4]

of patients who present with croup that is non-responsive to adrenaline nebulisation.

The most important complication of laryngeal papillomatosis is airway compromise.

Certainly patients can also develop failure to thrive and recurrent pneumonia. Malignant transformation can occur although it is very rare.[6] Concomitant pulmonary parenchymal lesions are present in less than 1% of patients with airway lesions.[1] Management involves surgical ablation, CO2 and laser vaporisation; medical adjuvant therapy for recurrent and large lesions includes alpha interferon and cidofivir.[1,6] Current therapies are unfortunately not curative and periodic resections may be necessary. [1] Preventive strategies in the form of the quadrivalent vaccine against human papillomavirus hold promise to decrease the incidence of juvenile respiratory papillomatosis.[6]

Papilloma of the trachea and bronchi

Malignant tumours of the airways

Respiratory papillomatosis is caused by human papillomavirus (most importantly types 6 and 11) and it is typically acquired during delivery.[1,6] It most commonly affects the larynx and trachea but can progress to involve the bronchi as well as the oesophagus.[2] These are the most common benign tumours of the larynx affecting children in a bimodal distribution, with 25% of patients younger than 5 years and the second peak occurring in the second and third decade of life.[6] An example of papillomas of the airway is shown in Fig. 1.

Classically these patients present with features of upper airway obstruction and a triad of hoarseness of voice, stridor and respiratory distress.[6] This condition should be remembered in the differential diagnosis

Bronchial adenoma

Together with pleuropulmonary blastomas, bronchial adenomas are the most common primary malignancies of the lung in childhood.

Two histological types of bronchial adenomas are defined: the carcinoid type, which occurs more commonly (90%), and the cylindromatous type.[7] Adenomas are more common in males and usually occur in older children and adolescents, with a mean age of presentation of 12 years.[1,2] Carcinoid tumours are low-grade tumours with low potential to metastasise; they are however locally aggressive.[1,2] This should form part of the differential diagnosis of patients who present with poorly controlled asthma, recurrent pneumonia and even haemoptysis (although rare). The diagnosis is based on histology. Some authors

Fig. 1. Papilloma of the airway on bronchoscopy (A). Subsequent cavitatory lesions in the lung noted on axial CT of the chest likely from seeding of the tumour (B) (arrow). (Courtesy of Prof.

Refiloe Masekela, Department of Paediatrics and Child Health, University of KwaZulu-Natal).

8 SARJ VOL. 22 NO. 1 2016

recommend biopsy via rigid bronchoscopy, as flexible bronchoscopy-directed biopsy carries a significant risk of bleeding.[2] Management of these lesions involves resection of a segment, lobe or entire lung depending on the position of the tumour and extent of the mass. Bronchoscopy removal is not advised due to the risk of bleeding and incomplete removal of the lesion.[2,7] The overall prognosis is excellent, with a survival rate of more than 90%.[1,2] Mucoepidermoid carcinoma

These tumours are much rarer than carcinoid tumours. They typically present just like bronchial adenomas with recurrent pneumonia, endobronchial obstruction with wheeze and haemoptysis. They most commonly occur as a polypoid exophytic causing obstruction of the main stem bronchus or proximal lobar bronchus.[2] The prognosis in children is excellent, with an 88% survival rate. The treatment consists of surgical removal with adjuvant chemotherapy reserved for those with incomplete resection.[2]

Lung parenchymal tumours

Benign lung parenchymal tumours

Inflammatory pseudotumour (inflammatory myofibroblastic tumour)

This is the most commonly occurring primary benign lung parenchymal tumour in the paediatric population.[1,2] Up to 30% of lesions are asymptomatic, but the rest present with nonspecific features of cough and fever. Patients are usually above the age of 5 years with an equal sex distribution.[1,2,7] Local invasion and recurrence are rarely seen in children diagnosed with these lesions.[1] Although classically a parenchymal lesion, in 12% of cases these lesions can occur as endobronchial lesions.[1] The chest X-ray (CXR) typically reveals a well-circumscribed solitary nodule that is peripherally based, usually located in the lower lobes.[1,2] It ranges in size from 1 to 12Â cm. The diagnosis is made histologically upon surgical removal. Biopsy specimens of the lesion are sometimes inaccurate and thus complete excision is required for the treatment and confirmation of the diagnosis.[1] Hamartoma

Intraparenchymal hamartomas are rare in children, with a peak incidence in the

REVIEW fourth and sixth decades, but can also occur in the paediatric population.[1,2] They are well-encapsulated solid masses that more commonly occur as intraparenchymal lesions. However, they may also occur in endobronchial sites. Most of these lesions are incidentally found on imaging studies as these patients are usually asymptomatic.[1] The characteristic computed tomography (CT) finding of these lesions is as â&#x20AC;&#x2DC;popcorn calcificationsâ&#x20AC;&#x2122;.[2,7] Percutaneous needle aspiration and core needle biopsies are adequate to attain a histological diagnosis.[1] Surgical resection may play a role in cases with an uncertain diagnosis.[1]

Type 1 PPB is predominantly cystic as in the case described above (Figs 5-8), type 2 mixed cystic and solid lesions and type 3 predominantly solid lesions. There is an increasing risk of malignant potential from types 1 to 3.[12] Type 1r lesions (type 1 regressed) are cystic in nature and may represent a regressed type of type 1 tumour.[12] PPB type 1 lesions carry the most favourable prognosis with a long-term survival of 80%. Unfortunately, types 2 and 3 lesions are more common, are highly aggressive and

carry a poorer prognosis (long-term survival of <50%). It is thought that type 1 lesions progress to types 2 and 3 with time.[13] The most common site of metastases is the brain followed by bone.[12] Radiologically, PPB can occur as a single cystic, a multicystic structure, a cyst with a polypoid mass, a mixed solid-cystic lesion or a completely solid lesion.[12] Unfortunately fine needle aspiration and cytology is inadequate to make a histological diagnosis and the entire lesion needs to be removed for the diagnosis.[14]

Fig. 2. Frontal CXR of 3-year-old female with right-sided mass occupying most of the right hemithorax (arrow).

Fig. 3. Contrasted axial CT of the chest of the same patient in Fig. 2. The lesion is a thinwalled fluid-filled cystic mass occupying a large portion of the right hemithorax (arrow).

Teratoma

Teratomas are tumours that originate from more than one germ cell line. Intrathoracic teratomas are the fourth most common site, the most common being the gonads.

Intrathoracically, teratomas usually occur in the anterior mediastinal region.[8] Intraparenchymal teratomas are extremely rare.[8-10] These tumours are more common in the third and fourth decades of life.

Patients can be completely asymptomatic or present with nonspecific symptoms of chest pain, cough and even haemoptysis.[8] The CXR will depict a well-demarcated lesion as in Fig. 2.

The CT scan of the chest will demonstrate a well-circumscribed cystic lesion (Fig. 3), the major differential of which is a hydatid cyst.

The management rests on surgical excision of these lesions. Adjuvant chemotherapy should be employed for immature teratomas.[11] The definitive diagnosis rests on histology after the lesion is removed, demonstrating features from multiple germ cell lines as shown in Fig. 4.

Primary malignant lung parenchymal tumours

Fig. 4. H&E section of the lesion in Figs 2 and 3 demonstrating cells from multiple germ cell lines.

This is compatible with a teratoma. (Courtesy of Dr A van Rooyen, Department of Anatomical Pathology, Tshwane Academic Division, National Health Laboratory Service (NHLS)).

Fig. 5. Frontal CXR. Left lower lobe multicystic lesion (arrow). The cysts appear large. There is significant mediastinal shift to the right.

Pleuropulmonary blastoma

Pleuropulmonary blastoma (PPB) is an extremely rare malignant primary tumour of the pulmonary or pleural mesenchyme.

Interestingly, PPB is quoted to account for 15% of primary paediatric pulmonary tumours. However, it is the most common primary parenchymal lung malignancy. The age of presentation is usually <4 years of age.

There are essentially three types, and a fourth type (type Ir) was added in 2006. There is no gender predilection and tumours of the right side are more common.[12]

Fig. 6. Axial CT chest. Large (>2 cm) multicystic lesion. The mediastinal shift is apparent on CT scan of the chest. The cystic lesions appear fluid filled and thus complicated.

Fig. 7. Coronal CT of the chest showing extent of the lesion.

SARJ VOL. 22 NO. 1 2016

9

REVIEW

Fig. 8. H&E section demonstrating the primitive neoplastic cells characteristic of a pleuropulmonary blastoma. (Courtesy of Dr C Crause, Department of Anatomical Pathology, Tshwane Academic Division, NHLS).

The management of these patients includes surgical removal of the lesion in the form of a lobectomy and a combination of chemotherapy and radiotherapy.[12,13] The chemotherapeutic regimen suggested includes a combination of ifosfamide, vincristine, actinomycin D and doxorubicin.[12] There is much controversy in paediatric pulmonology about the indications to remove congenital pulmonary airway malformations (CPAMs).[15] Certainly in the case of symptomatic (causing compression with respiratory distress), infected CPAMs or hybrid lesions the consensus lies in removal of these lesions.[16,17] PPB can appear identical to CPAMs and the diagnosis rests solely on histology with entire removal of the lesion. More so, PPBs are highly aggressive tumours and the outcome is generally poor if not attended to.[16] The dilemma thus arises in the face of asymptomatic lesions that appear as CPAMs. The diagnosis of a PPB runs the risk of being missed if these lesions are not removed. It is important for the physician to note that even though primary CPAMs may be removed, it does not protect against the chance of malignant transformation.[18] More so, the risk of malignant transformation in a CPAM is largely unknown and has never been quantified.[17] The issue then to remove a lesion should be discussed with the parents of the child, the individual risks including short-term surgical risks and long-term risks of infection and malignant transformation explained to them and a decision made as a team.

Kaposi’s sarcoma

Kaposi sarcoma (KS) is a low-grade vascular tumour associated with human herpes virus 8.

The physician should certainly be aware of KS of the lung, especially in the South African

10 SARJ VOL. 22 NO. 1 2016

setting. Childhood KS is now quoted as the most common AIDS-defining malignancy affecting children in sub-Saharan Africa, with pulmonary KS thought to occur in 10% of these cases.[19] Cutaneous KS is the most common site.[20] The presenting features of pulmonary KS rests on a common triad of cough, dyspnoea and lymphadenopathy.[21] Patients can also present with features of upper and lower airway obstruction; haemoptysis in children is rare. Extensive pulmonary KS can precede the development of mucocutaneous lesions. The typical radiological features include air-space disease, especially in the mid and lower zones, hilar lymphadenopathy and significantly large pleural effusions that can be bilateral.[21] The overall prognosis is unfortunately poor with chemotherapy the mainstay of treatment.[19] Controlling the HIV disease is as important as chemotherapy for these patients and has a large impact on the survival rates.[22] The diagnosis of pulmonary KS rests on a high index of suspicion with typical features described above. Histology of a peripheral site including skin lesions or lymph nodes is supportive of the diagnosis.[21]

The general principles of management of secondary lesions include control of the primary site. Secondary metastases should only be considered for removal once other sites are disease free. However, soft-tissue sarcomas and osteosarcomas should be considered for removal of pulmonary disease as removal of these lesions is associated with a better prognosis.[7]

Mediastinal tumours

The mediastinum is that area that extends from the thoracic inlet superiorly to the diaphragm inferiorly; it is bound laterally by the parietal pleura, anteriorly by the sternum and posteriorly by the vertebral column. The superior mediastinum extends from the thoracic inlet to an imaginary line from the angle of the sternum to the T4/T5 intervertebral space. It is further divided into anterior, middle and posterior compartments.

It is paramount that the physician is aware of these anatomical boundaries so as to limit the differential diagnosis based on the anatomical compartment involved.[23] Common pathological causes include: T-cell lymphoma, Hodgkin’s lymphoma, germ cell

Other primary malignant parenchymal tumours

Other primary malignant tumours of the lung including bronchogenic carcinoma, leiomyosarcoma, multiple myeloma and choriepithelioma are extremely rare in the paediatric population. In terms of bronchogenic carcinoma, adenocarcinoma accounts for more cases than squamous cell carcinoma. These tumours as a group are extremely aggressive and carry an overall poor prognosis.[2] Secondary malignant lung parenchymal tumours Secondary malignant parenchymal tumours are much more common than primary parenchymal lung malignancies. In one study, secondary malignancies were 12 times more prevalent than primary parenchymal malignancies. Secondary metastases that occur in the lung in the paediatric population differ from those associated with adult disease.[2] In a 25-year cohort of pathology specimens in Texas Children’s Hospital the following secondary malignancies were noted to be common in decreasing order of incidence: Wilms’ tumour, osteosarcoma, rhabdomyosarcoma, neuroblastoma and hepatoblastoma.[2]

Fig. 9 Frontal CXR of a 3-year-old boy demonstrating an enlarged superior mediastinum (arrows).

Fig. 10. Contrasted axial CT scan at the level of the great vessels of the same patient in Fig. 9, demonstrating lymphadenopathy in the posterior mediastinum. This was histologically confirmed to be a neuroblastoma.

REVIEW tumours and neuroblastoma and occur more commonly in different compartments of the superior mediastinum. For a more in-depth approach to masses that occur in the superior mediastinum, the reader is referred to the case report of a superior mediastinal mass elsewhere in this journal. Figs 9 and 10 show an example of a widened superior mediastinal secondary to a neuroblastoma in the posterior mediastinum.

Cardiac tumours Cardiac tumours should be considered in the differential diagnosis of mediastinal masses. Cardiac tumours in the paediatric population are very rare and most lesions are benign. These tumours are usually associated with a systemic disease process such as tuberous sclerosis. In terms of intracardiac masses the most common causes are rhabdomyoma, myxoma and fibroma. Pericardial lesions include pericardial cysts and teratomas.[24] Chest wall tumours Chest wall tumours in children can also be benign or malignant.

The most common benign lesions include infantile haemangioma, osteoid osteoma and osteochondroma.

Infantile haemangiomas are vascular tumours that proliferate rapidly in the first few months of life and thereafter undergo spontaneous regression. They are well demarcated, palpable bluish-red lesions. Since they spontaneously regress, there is often no need for treatment.

Osteoid osteomas usually occur in the long bones but can occur in the thorax. They are more common in males between the ages of 7 and 25 years. These are self-limiting lesions and usually require no further treatment, other than analgesia.

Osteochondromas also have a male predominance, beginning to grow just before puberty and occurring most commonly at the chostochondral junctions. These lesions can cause significant pain, can fracture and have the potential for malignant transformation, especially if the lesions are multiple.

Primary malignant tumours that affect the chest wall include rhabdomyosarcoma and osteosarcoma.

Rhabdomyosarcomas are high-grade tumours that arise from skeletal muscle. Chest wall involvement can occur. However, it more commonly occurs in the head and neck region or in the abdominal wall. They are highly aggressive and have the potential to invade the surrounding bone, indicating advanced disease. Treatment of these lesions includes both surgical resection and chemotherapy.

Osteosarcomas are also highly aggressive tumours. They can originate anywhere from the bone, soft tissue or pleura. The peak incidence is during adolescence to young adulthood. The chief presenting feature is pain. Most commonly these lesions affect the long bones and 1 - 2% of osteosarcomas affect the chest wall. As with rhabdomyosarcomas, treatment involves surgery and chemotherapy.[25]

Conclusion

Tumours of the chest in children are very rare as a whole. The clinical features range from asymptomatic lesions to nonspecific symptoms of cough. The physician needs to always have a high index of suspicion.

Compartmentalising the different possible causes into the most likely

source of origin, whether it be airway or parenchymal, definitely aids in a reasonable differential diagnosis.

References 1. Amini B, Huang SY, Tsai J, Benveniste MF, Robledo HH, Lee EY. Primary lung and large airway neoplasms in children: Current imaging evaluation with multidetector computed tomography. Radiol Clin North Am 2013;51(4):637-657. [http://dx.doi. org/10.1016/j.rcl.2013.04.005] 2. Dichop MK, Kuruvilla S. Primary and metastatic lung tumours in the paediatric population. Arch Pathol Lab Med 2008;132(7):1079-1103.

3. Eyssartier E, Ang P, Bonnemaison E, et al. Characteristics of endobronchial primitive tumours in children. Paediatr Pulmonol 2014;49(6):E121-125. [http://dx.doi. org/10.1002/ppul.22987] 4. Hardison SA, Dodson KM, Rhodes JL. Subglottic hemangioma treated with propranolol. Eplasty 2014;14:ic2.

5. Tabatabaii SA, Khanbabaii G, Khatami AR, Sharifnia SA. Characteristic and follow-up of subglottic haemangiomas in Iranian children. J Res Med Sci 2010;15(4):235-239.

6. Masekela R, Smit A, Tshifularo MI, Green R. Juvenile recurrent respiratory papillomatosis a cause of airway obstruction: A case study. S Afr Ped Rev 2008;5(2):30-35.

7. Plerhoples TA, Krummel TM. Tumours of the chest. In: Wilmott RW, Boat TF, Bush A. Kendig and Chernick’s Disorders of the Respiratory Tract in Children. 8th ed.

Philadelphia: Elsevier Saunders, 2012:605-630.

8. Saini ML, Krishnamurthy S, Kumar RV. Intrapulmonary mature teratoma. Diag Path 2006;1(38):1-3. [http://dx.doi.org/10.1186/1746-1596-1-38] 9. Dar RA, Mushtaque M, Wani SH, Malik RA. Giant intrapulmonary teratoma: A rare case. Case Rep Pulmonol 2011;201: 298653. [http://dx.doi.org/10.1155/2011/298653] 10. Sawant AC, Kandra A, Narra SR. Intrapulmonary cystic teratoma mimicking malignant pulmonary neoplasm. BMJ Case Rep 2012;2012:bcr0220125770. [http:// doi.org/10.1136/bcr.02.2012.5770] 11. Terenziani M, D’Angelo P, Inserra A, et al. Mature and immature teratoma: A report from the second Italian pediatric study. Paediatr Blood Cancer 2015;62(7):1202-1208. [http://dx.doi.org/10.1002/pbc.25423] 12. Khan AA, El-Borai AK, Alnoajii M. Pleuropulmonary blastoma: A case report and review of the literature. Case Rep Path 2014;2014:509086. [http://dx.doi.org/10.1155/2014/509086] 13. Libretti L, Ciriaco P, Casiraghi M, Arrigoni G, Zannini P. Pleuropulmonary blastoma in the area of a diagnosed congenital lung cyst. Ann Thorac Surg 2008;85(2):658-660. [http://dx.doi.org/10.1016/j.athoracsur.2007.08.012] 14. Schultz KAP, Williams GM, Stewart DR, et al. Recurrent or progressive Type II and Type III pleuropulmonary blastoma (PPB) are associated with poor outcome: A report from the International PPB Registry. J Clin Oncol 2015;33(15S):10014.

15. Fauroux B. Congenital cystic adenomatous malformation (CCAM): Remove or not, how can we progress? Paediatr Respir Rev 2013;14(3):168. [http://dx.doi. org/10.1016/j.prrv.2013.06.001] 16. Delacourt C, Hadchouel A, Khen Dunlop N. Shall all congenital cystic malformations be removed? The case in favour. Paediatr Respir Rev 2013;14(3):169-170. [http:// dx.doi.org/10.1016/j.prrv.2013.06.003] 17. Kotecha S. Should asymptomatic congenital cystic malformations be removed? The case against. Paediatr Respir Rev 2013;14(3):171-172. [http://dx.doi.org/10.1016/j. prrv.2013.06.002] 18. Papagiannopoulos KA, Sheppard M, Bush AP, Goldstraw P. Pleuropulmonary blastoma: Is prophylactic resection of congenital lung cysts effective? Ann Thorac Surg 2001;72(2):604-605.

19. Graham SM. Non-tuberculosis opportunistic infections and other lung diseases in HIV-infected infants and children. Int J Tuberc Lung Dis 2005;9(6):592-602.

20. De Bruin GP, Stefan DC. Children with Kaposi sarcoma in two South African hospitals: Clinical presentation, management and outcome. J Trop Med 2013;2013:213490. [http://dx.doi.org/10.1155/2013/213490] 21. Theron S, Andronikou S, Du Plessis J, et al. Pulmonary Kaposi sarcoma in six children.

Paediatr Radiol 2007;37(12):1224-1229. [http://dx.doi.org/10.1007/s00247-007-0632-9] 22. Stefan DC, Stones DK, Wainwright L, Newton R. Kaposi sarcoma in South African children.

Paediatr Blood Cancer 2011;56(3):392-396. [http://dx.doi.org/10.1002/pbc.22903] 23. Williams HJ, Alton HM. Imaging of paediatric mediastinal abnormalities. Paediatr Respir Rev 2003;4(1):55-66. [http://dx.doi.org/10.1016/S1526-0542(02)00310-X] 24. Newman B. Thoracic neoplasms in children. Radiol Clin North Am 2011;49(4):633664. [http://dx.doi.org/10.1016/j.rcl.2011.05.010] 25. Baez JC, Lee EY, Restrepo R, Eisenber RL. Chest wall lesions in children. AJR Am J Roentgenol 2013;200(5):W402-W419. [http://dx.doi.org/10.2214/AJR.12.8883]

SARJ VOL. 22 NO. 1 2016

11

REVIEW

Eosinophilic lung diseases: A review K Dudgeon, MB BCh, FCP (SA) Division of Pulmonology, Charlotte Maxeke Johannesburg Academic Hospital, University of the Witwatersrand, Johannesburg, South Africa Corresponding author: K Dudgeon (kadet130@hotmail.com)

The terms ‘eosinophilic pneumonia’ and ‘eosinophilic lung disease’ loosely describe a heterogeneous group of pulmonary diseases of varying aetiologies and severity. The diseases are characterised by infiltration of lung parenchyma by eosinophils; peripheral eosinophilia is not required for diagnosis. In this article, major clinical entities are appraised with respect to clinical, pathological and radiological features. Diseases without pulmonary infiltration or radiographic abnormalities, such as allergic asthma, are not included in this review.

S Afr Respir J 2016;22(1):12-18. DOI: 10.7196/SARJ.2016.v22i1.39

There are several recognised clinical and radiographic presentations of eosinophilic lung disease. These include simple pulmonary eosinophilia (SPE), chronic eosinophilic pneumonia (CEP), acute eosinophilic pneumonia (AEP), allergic bronchopulmonary aspergillosis (ABPA) and pulmonary eosinophilia associated with a systemic disease.[1] Systemic diseases implicated include eosinophilic granulomatosis with polyangiitis (EGPA), formerly known as Churg-Strauss syndrome (CSS) and the hypereosinophilic syndrome (HES). Eosinophilic pneumonias may be idiopathic or secondary to a known cause. Causes may include drugs, irradiation, toxins and infections. The infections may be fungal, parasitic or mycobacterial in nature. AEP, CEP and ABPA have radiographic features that may be suggestive, if not pathognomonic, in several instances. In addition, varying degrees of pulmonary eosinophilia may be associated with diffuse lung diseases, neoplasia and connective tissue diseases. Finally, a hallmark of eosinophilic lung diseases (with the exception of HES) is their exquisite sensitivity to corticosteroids.

The eosinophil leukocyte is of obvious importance. It is a granulocyte named for its abundance of eosinophilic granules in the cytoplasm. Mature eosinophils circulate for approximately 24 hours before being recruited into target tissues where they undergo rapid apoptosis if no survival factors are present.[2] Recent studies have suggested a role for eosinophils apart from that of end-stage anti-parasitic cells. These include roles in both innate and adaptive immunity, including antigen presentation to Th2 cells and other interactions with mast and T cells.[3] Eosinophils release several toxic substances from the small and large granules in their cytoplasm that are thought to contribute to the pathophysiology of these diseases. The smaller granules contain the characteristic cationic proteins: major basic protein (MBP), eosinophilic cationic protein (ECP), eosinophil-derived neurotoxin (EDN) and eosinophil peroxidase (EPO).[4]

General concepts

Disorders may be classified as eosinophilic lung disease in one of three ways:[5] 1. Peripheral blood eosinophilia and radiographic infiltrates (pulmonary infiltrates with eosinophilia or PIE syndrome).

Is it important to note that blood eosinophilia does not prove

12 SARJ VOL. 22 NO. 1 2016

lung eosinophilia and that lung involvement is not invariably accompanied by blood eosinophilia. The absolute eosinophil count is preferred over the percentage. A normal blood eosinophil count ranges from 50 to 250 cells/µL.[5] 2. Lung biopsy. This can be accomplished by the transbronchial route or open lung biopsy. Open lung biopsy is considered the ‘gold standard’ because it yields adequate amounts of alveolar and vascular tissue. Histological findings are remarkably consistent across all forms of these diseases and include an intra-alveolar and interstitial exudate of histiocytes and eosinophils, eosinophilic microabscesses, and findings of an organising pneumonia.

There may be small areas of interstitial necrosis and fibrosis. A small degree of vasculitis is allowed, as long as granulomata are absent. Frank eosinophilic vasculitis is indicative of EGPA and granulomata are found in parasitic infections, EGPA and ABPA.[1] 3. Bronchoalveolar lavage (BAL). This technique has several advantages in that it is minimally invasive, safe and can be used to monitor response to therapy. Eosinophilia is defined as >5%, severe eosinophilia as >25%.[6]

Classification

There is no widely accepted classification of the eosinophilic lung diseases. Table 1 suggests a classification in terms of clinical and radiological presentation and aetiology.[1,5] Table 2 describes the infectious causes of pulmonary eosinophilia and the primary mechanism whereby they exert their effects. A detailed review of the infectious causes is not possible due to space constraints. Drugs are another major cause of pulmonary eosinophila. Non-steroidal anti-inflammatory drugs, immunosuppressants and antibiotics are most commonly implicated. A comprehensive list is available at www.pneumotox.com.

Eosinophilic lung diseases

Simple pulmonary eosinophilia In 1932, Wilhelm Löffler described a syndrome of migratory pulmonary infiltrates with peripheral eosinophilia and minimal pulmonary symptoms.[7] In his original series, most cases were due to Ascaris infection. The term is now used more broadly to describe

REVIEW SPE resulting from any fungal, parasitic or drug-induced cause. The transient nature of eosinophilic pulmonary infiltrates and symptoms mirror the transpulmonary passage of larvae in the lifecycle of parasites including Ascaris, [7] hookworms such as Ancylystoma duodenale or Nector americanus and Schistosoma spp.[7] Though most patients remain minimally symptomatic, 8 - 15% display respiratory symptoms such as wheeze, cough and haemoptysis approximately 9 - 12 days post ingestion of eggs. Symptoms may last 5 - 10 days; severity correlates with worm burden. If required, a definitive diagnosis may be made by recovery of larvae via respiratory secretions and gastric lavage fluid. Eggs will be detectable in stools 14 days after ingestion.

The radiographic pattern often consists of patchy peripheral infiltrates with a pleural base. Coalescence may occur in severe cases.

Spontaneous resolution of the syndrome within 30 days is the norm and therapy is rarely required. Corticosteroids have been used successfully in severe cases.[9] Importantly, transient radiographic infiltrates occur in other forms of eosinophilic lung disease, including ABPA, EGPA and HES. Chronic eosinophilic pneumonia The clinical disease entity CEP may be due to drugs, parasitic infections, irradiation or severe stressors such as childbirth. [10] Typically, however, it is idiopathic.

Carrington and colleagues first described a series of patients with this disease entity in 1969.[11] Idiopathic CEP (ICEP) is a rare disorder of unknown aetiology and there are no clear diagnostic criteria available. Table 3 suggests a schema.[12] The exact prevalence of CEP remains unknown, but the disease is reported to contribute to <2.5% cases included in interstitial lung disease registries.[13] The peak incidence occurs in the fifth decade of life and females are twice as likely to be affected. [5] The overwhelming majority of patients are non-smokers, leading to the hypothesis that smoking may be protective.

Presenting complaints commonly include cough, fever, dyspneoa and weight loss.[14] Wheezing, night sweats, malaise and a productive cough are less common, and haemoptysis is rare. Asthma is present in 50% of cases, and has usually been present for <5 years. Respiratory failure is less common than in AEP but has been reported in cases where

Table 1. Classification of pulmonary eosinophilia[1] Principal forms of pulmonary eosinophilia (based on clinical and radiological presentation) 1. Simple pulmonary eosinophilia 2. Chronic eosinophilic pneumonia 3. Acute eosinophilic pneumonia 4. Allergic bronchopulmonary aspergillosis 5. Pulmonary eosinophillia associated with systemic disease •

EGPA

•

HES

Aetiology 1. Primary (idiopathic) 2. Secondary a) Known cause •

Drugs

•

Toxins or irradiation

•

Infections: parasitic, fungal and mycobacterial

b) Diseases associated with a degree of pulmonary eosinophilia •

Diffuse lung disease: idiopathic pulmonary fibrosis, hypersensitivity pneumonitis, cryptogenic organising pneumonia, sarcoidosis and pulmonary Langerhans cell histiocytosis

•

Malignancies: leukaemia, lymphoma, lung cancer, metastatic adenocarcinoma and metastatic squamous cell carcinoma

•

Auto-immune diseases: rheumatoid arthritis, Sjögren’s syndrome

the diagnosis was delayed.[15] Extrathoracic manifestations are ordinarily absent in CEP; when present, a diagnosis of EGPA or HES should be considered. Notably, a few patients with a diagnosis of CEP may develop minor extrathoracic manifestations without fulfilling diagnostic criteria for EGPA or HES. In addition, several authors have suggested that CEP may be a presenting feature of EGPA, suggesting a disease continuum.[16] The most distinctive radiographic feature is the so-called ‘photographic negative’ of acute pulmonary oedema (Fig. 1). This is characterised by peripheral pulmonary infiltrates (consolidation or ground-glass opacification) which are usually bilateral but may occasionally be unilateral and migratory. Unfortunately this pattern is present in only 25% of cases, and in addition, has been described in cryptogenic organising pneumonia (COP), sarcoidosis or drug-induced pneumonia. Less frequent abnormalities include nodules, atelectasis and cavities.[14] Typically, interstitial fibrosis is minimal but there are reports of cases with progression to honeycombing. [17] Pleural effusions are rare, but a case of CEP presenting

Fig. 1. Computed tomography (CT) scan showing CEP. Note the peripheral distribution of ground-glass opacities and the interlobular septal thickening. (Taken from www.chestatlas. com, with permission from Dr Harry Shulman.) as bilateral massive pleural effusions has been reported.[18] Peripheral eosinophilia, usually in excess of 1 000 cells/mm3 is found in the majority of cases. There are anecdotal reports of cases of CEP without peripheral eosinophilia. In these cases the diagnosis is made on BAL, by the demonstration of >40% eosinophils in the fluid. In CEP, BAL always reveals an

SARJ VOL. 22 NO. 1 2016

13

REVIEW abnormally high level of eosinophils. In contrast, transbronchial biopsy, when performed, may not reveal a significant eosinophilic inflitrate.[5] For this reason, in the rare instances where lung biopsy is required, the open modality is preferred. In keeping with the high proportion of atopic patients who develop CEP, immunoglobulin E (IgE) levels are elevated in 50% of cases.[10,12] Pulmonary function tests may be normal in mild cases but usually show restrictive abnormalities with a reduced diffusing capacity. In patients with pre-existing asthma, obstructive defects may be noted.

This is not due to CEP per se. Obstructive disease of the small airways may reflect a degree of bronchiolitis.[14] All patients with CEP will be hypoxaemic or demonstrate an increased A-a gradient.[5] Clinical response to corticosteroids provides support for the diagnosis. The response is typically rapid: blood eosinophilia regresses within hours and radiographic abnormalities within days.

Symptoms improve within weeks. There is no consensus with respect to dose or optimal duration of therapy but most authors recommend prednisone as the drug of choice. A dose of between 0.5 mg and 1 mg/kg/day is used initially and gradually tapered over a period of 6 months. Relapses are common and are said to occur in between 30 and 50% of cases.[10] Relapses may occur in the same or different parts of the lung[2] and respond as well to corticosteroids as the initial episode. There is a suggestion that 3 months of therapy may be as effective as 6 months, with no difference in relapse rate.[19] The use of inhaled corticosteroids has been suggested as a modality to reduce relapse rates and oral corticosteroid use. Marchand et al.[20] reported a reduced relapse rate in patients with CEP and asthma.More than half of patients with CEP require long-term systemic corticosteroid therapy due to frequent relapses or severe asthma. The side-effects of such therapy are well documented and, therefore, steroid-sparing strategies require consideration. Kaya et al.[21] reported the successful treatment of a single case of a patient with CEP and high IgE levels with omalizumab (a monoclonal antibody directed at IgE). This modality requires further study.

Acute eosinophilic pneumonia AEP, first described in 1989,[22] is differentiated from CEP by the duration and severity of symptoms and the absence of relapse after recovery. The diagnostic criteria proposed by Allen et al.[5] in 1994 are the most widely accepted criteria (Table 4). Some authors, however, have challenged the disease duration criterion[23] and have included patients with symptoms of up to 1 month’s duration in case series.

Typically, patients are previously well and present with an acute febrile illness and hypoxaemic respiratory failure, sometimes meeting criteria for acute respiratory distress syndrome. Blood eosinophilia is typically absent; frank alveolar eosinophilia (usually >25% of cells) at BAL is the norm and can obviate the need for lung biopsy.[2] The mean age of patients at diagnosis is 30 years. There is a male predominance and typically no prior history of atopy. Idiopathic AEP (IAEP) is a diagnosis of exclusion and, to that end, very close attention should be paid to respiratory exposures within the days prior to presentation. Several exposures are purported to lead to the intense pulmonary eosinophilic infiltrate noted in AEP. Examples reported in the literature include cave exploration, plant repotting, indoor renovations, tank cleaning and exposure to various dusts including dusts at the World Trade Centre.[23,24] There have been several case reports of the development

14 SARJ VOL. 22 NO. 1 2016

Table 2. Infectious causes of pulmonary eosinophilia* Disease manifestation

Cause

Löffler’s

Acaris Hookworm Schistosomiasis

Large parasite burden

Strongyloidiasis

Direct pulmonary penetration

Paragonimiasis Visceral larval migrans

Immunologic response to organisms Filariasis (tropical filarial pulmonary eosinophilia) Dirofilariasis Cystic disease (rare)

Echinococcus Cysticercosis

Fungal aetiologies

Coccidiomycosis Cryptococcosis Paracoccidiomycosis Basidiobolomycosis

*Adapted from Akuthota P, Weller PF.[7]

Table 3. Suggested diagnostic schema for ICEP

• Subacute or chronic respiratory and general symptoms (average of 7.7 months before diagnosis) • Alveolar eosinophilia (>40% eosinophils at BAL) or peripheral eosinophilia (blood eosinophilia count >1 000 cells/mm3 • Pulmonary infiltrates on chest imaging (usually peripheral predominance) • Exclusion of known causes of eosinophilic lung diseases

of AEP after the initiation of cigarette smoking,[2,25] but given that cigarette smoking is common and AEP is rare, smoking is unlikely the sole cause of AEP. Drugs, parasites and fungi are also known causes of this syndrome.

The precise mechanism of disease in IAEP has yet to be elucidated.

An acute hypersensitivity reaction to an unidentified inhaled antigen has been put forward as a cause. The degree of respiratory failure in AEP is related to both the intensity of eosinophilic infiltration of pulmonary parenchyma and the mediators released by the eosinophils.

In addition to cationic granule proteins and inflammatory lipid mediators, vascular endothelial growth factor is elevated in the lungs of patients with AEP, where it causes increased vascular permeability and alveolar filling.[26] The proteolytic potential of eosinophils is lower than that of neutrophils and this allows complete resolution.

The initial radiographic finding is a subtle interstitial infiltrate which progresses to a diffuse mixed interstitial and alveolar infiltrate within hours to days. Small to moderate-sized pleural effusions are common. In contrast to CEP, peripherally-based infiltrates are uncommon. The combination of diffuse areas of ground-glass attenuation, defined nodules, smooth interlobular septal thickening and pleural effusions may correctly identify the diagnosis in up to 81% of cases.[27]

REVIEW Pulmonary function tests show a restrictive defect with a low diffusing capacity in the acute phase; these return to normal following treatment.

The key diagnostic differential is the exclusion of an infectious cause.

Fungal pneumonia should always be excluded by fungal culture, as it can mimic the presentation of AEP. A rapid response to corticosteroids is a clinical hallmark, but several cases of spontaneous resolution have been reported. High doses are required, but the minimum effective dose is not known. A regimen of intravenous methylprednisolone until respiratory failure has resolved, followed by oral prednisone for another 2 weeks has been used successfully.[5,23] Steroids are then tapered for the next 2 4 weeks. Importantly, several cases of spontaneous resolution have been reported.[28] Failure to respond to corticosteroid therapy should prompt a thorough search for an alternate diagnosis, particularly fungal infection.

AEP may be rapidly progressive, with patients occasionally requiring mechanical ventilation within hours. Fatalities have been reported in severe cases. AEP is easily diagnosed and treated and should be considered in all cases of unexplained respiratory failure and pulmonary infiltrates.

Allergic bronchopulmonary aspergillosis The entity of ABPA refers to a complex hypersensitivity reaction to colonisation of the airways with Aspergillus spp. Exact prevalence is unknown. A vicious cycle with repeated episodes of bronchial obstruction, inflammation and mucoid impaction is set up. This can lead to bronchiectasis, fibrosis and eventual respiratory compromise.[29] Although most commonly seen with Aspergillus fumigatus, allergic bronchopulmonary disease has been described in association with Candida albicans, Aspergillus terreus and other fungal diseases.

ABPA occurs primarily in asthmatics (2 - 32%) and patients with cystic fibrosis (1 - 15%).[30] Both genders and any age group may be affected. The disease cycle mentioned above is the dominant presenting clinical feature. In addition to this, peripheral eosinophilia is present and haemoptysis may occur. Wheezing is often absent and an incidental finding of pulmonary consolidation may be the presenting feature. Chest X-rays (CXR) and CT scans may classically show bronchiectasis, patchy infiltrates and evidence of mucous impaction. Central bronchiectasis is associated with ABPA and is present in 85% of patients at diagnosis.[31] Those without central bronchiectasis should not be excluded, since it may be absent early in the disease course. Pulmonary function testing usually shows an obstructive defect.

The pathophysiology is not understood completely. Healthy individuals display low levels of IgG and IgA against fungal antigens, suggesting that they are able to eliminate fungal spores, even when inhaled in sufficient quantities to behave as an allergen. This is in contrast to atopic individuals who respond to inhalation of fungal spores by forming IgG and IgE. In addition, a Th2 response is elicited in affected individuals, leading to an increase in IL-4, IL-5 and IL-13, explaining the eosinophilia and raised IgE levels. The characteristic central bronchiectasis of ABPA is likely multifactorial in nature, with proteolytic enzymes, Aspergillus mycotoxins and neutrophilic and eosinophilic inflammation contributing.

The diagnosis of ABPA (at least in the USA) is based on the Patterson criteria.[31] Diagnosis is made using a combination of clinical, radiographic, serologic and immunologic findings. Four of

the major clinical features listed in Table 5 are required for diagnosis.

Importantly, these criteria are not universally applied, making overall prevalence difficult to study.

Pathology is not required for diagnosis but findings may include eosinophilic inflammation, mucoid impaction and bronchocentric granulomatosis. In addition, non-invasive, septated hyphae may be visible.

ABPA is said to progress through five clinical stages, as described by Patterson: acute, remission, exacerbation, corticosteroid-dependent asthma and fibrosis.[32] Treatment varies according to stage. Acute or recurrent flares are treated with systemic glucocorticoids; these are tapered over 3 - 6 months.[33] Antifungal agents effective against Aspergillus spp are used as adjunctive therapy to reduce the antigenic stimulus. Itraconazole is considered first-line therapy but voriconazole has also been used. Omalizumab has been reported as being effective in ABPA for reduction of exacerbations.[34]

Eosinophilic lung disease associated with systemic conditions

Eosinophilic granulomatosis with polyangiitis This eponymous syndrome was first described in autopsied cases in 1951 [35] and is a small vessel vasculitis. [36] Multiple organs may potentially be affected, including the sinuses, heart, lungs, gastrointestinal tract, skin and kidneys. The initial description was revised in 1990, when the American College of Rheumatology (ACR) established criteria for diagnosis (Table 6).[1] At least four are necessary to confirm the diagnosis. Notably, this means that the histopathological criterion is not necessary for the diagnosis and clinical diagnosis is possible.[1] Lung biopsy is still considered the gold standard. However, nerve, muscle and skin biopsies may reveal perivascular eosinophilic infiltration and confirm the diagnosis. Renal biopsy tends to be nonspecific and therefore not useful. Despite the name EGPA, granulomata are not required for diagnosis. Antineutrophil cytoplasmic antibody (ANCA), particularly perinuclear ANCA, is positive in 50 - 70% of cases.[37]

Table 4. Diagnostic criteria for AEP • • • • • • •

Acute febrile illness <5 days’ duration Hypoxaemic respiratory failure Diffuse alveolar or mixed alveolar-interstitial infiltrates on CXR >25% eosinophils on BAL Absence of parasitic, fungal or other infection Prompt and complete response to corticosteroids Failure to relapse after discontinuation of steroids

Table 5. Diagnostic criteria for ABPA • • • • • •

Asthma Peripheral blood eosinophilia Immediate skin prick test for Aspergillus antigens Serum precipitating antibodies against Aspergillus antigens Increased serum IgE levels CXR infiltrates

SARJ VOL. 22 NO. 1 2016

15

REVIEW

Table 6. ACR diagnostic criteria for EGPA • • • • • •

Asthma Eosinophil in peripheral blood >1 500 cells/mm3 of blood Paranasal involvement Transient pulmonary infiltrates Mononeuropathy or polyneuropathy Biopsy findings of vasculitis

The clinical features of EGPA are well defined. The disease occurs most commonly in the fourth and fifth decades, both genders being affected. The syndrome is characterised by three phases:[38] • Allergic phase: Asthma is always present and usually severe; rhinitis occurs in 75% of cases and is often accompanied by recurrent sinusitis and polyps.

• Eosinophilic phase: Severe persistent eosinophilia (more than 1 500 cells/mm3) for at least 6 months.

• Vasculitic phase: Systemic manifestations and small-vessel vasculitis involving two or more extrapulmonary organs.

These phases may be dissociated.[1] Asthma precedes vasculitis by an average of 3 - 9 years,[38] but the interval may be longer or the two entities may coincide. The advent of vasculitis may be associated with a reduction is asthma severity.[1] Blood eosinophilia commonly parallels vasculitic activity and BAL fluid may contain in excess of 60% eosinophils. IgE levels are markedly increased and correlate with disease activity.[38] Ill-defined migratory pulmonary infiltrates are present in 37 - 72% of cases; the CXR may remain normal. High-resolution chest CT demonstrates nonspecific features that allow a correct diagnosis in fewer than half of cases.[27] Features include ground-glass attenuation, airspace consolidation, centrilobular nodules, bronchial wall thickening or bronchial dilatation, interlobular septal thickening, hilar or mediastinal lymphadenopathy, and pleural and pericardial effusions. Pulmonary cavitatory lesions are rare.

Multi-organ involvement is possible. Upper airway manifestations have been alluded to. Skin manifestations are present in 70% of cases and can include nodules, palpable purpura or urticaria. Nervous system involvement includes mononeuritis multiplex in 66% of cases.

Gastrointestinal symptoms may include abdominal pain, diarrhoea and bleeding. Cardiac findings include cardiac failure, pericarditis, endomyocardial fibrosis, valvulitis, coronary vasculitis and systemic hypertension. Cardiac manifestations are a poor prognostic feature and contribute to 50% of deaths. Many patients have fever, myalgias or arthralgias and lymphadenopathy.[38] Uncommon manifestations include hearing loss, reversible exophthalmos and pulmonary capillaritis.[39] Corticosteroids alter the natural history of this disease. Fifty per cent of untreated patients die within 3 months of the onset of vasculitis.

This increases dramatically to a mean of 9 years in those undergoing corticosteroid treatment.[40] Several weeks of prednisone in high doses are required to halt the vasculitis and mononeuritis may require more prolonged treatment. Daily or alternate-day doses of prednisone are typically continued for a year and then weaned. Relapses after this are uncommon.[39] Treatment of asthma with inhaled corticosteroids

16 SARJ VOL. 22 NO. 1 2016

may allow reduction in the dose of systemic steroids. Alternative treatment options for non-responders include high-dose pulses of methylprednisolone, azathioprine or cyclophosphamide.[41,42] In some patients, the treatment of the severe asthma associated with EGPA with systemic steroids may mask the vasculitis and discontinuation of steroids or reduction in the dose may unmask it. Cases of ‘limited’ EGPA, the so-called ‘formes frustres’ have been reported. These refer to forms with single organ involvement and may resemble other eosinophilic syndromes such as ICEP.[43] There have been several reports of EGPA associated with leukotriene inhibitors, but a causative role has yet to be conclusively established. [44] Some authors suggest that these agents should be avoided in asthmatics with marked eosinophilia or features compatible with EGPA.[2] The hypereosinophilic syndrome The hypereosinophilic syndromes are a group of diseases defined by sustained eosinophil overproduction in association with tissue infiltration or damage. The term HES is reserved for those cases fulfilling the above definition, in which all known potential causes have been excluded (parasites, drugs or non-haematological neoplasia).

Blood hypereosinophilia is defined as an absolute eosinophil count of greater than 1.5 × 109/L on two examinations, at least one month apart. Tissue infiltration is defined as: • >20% eosinophils on bone marrow biopsy and/or • extensive tissue infiltration of eosinophils (in the pathologist’s opinion) and/or • marked deposition of eosinophil granule proteins in the absence of marked eosinophilic infiltration.[45] HES may be defined as primary, secondary or idiopathic. Primary (neoplastic) HES is the result of eosinophilic expansion that is clonal, such as in underlying stem cell, myeloid or eosinophilic neoplasia.

Secondary (reactive) HES is the result of polyclonal stimulation of eosinophil cytokines. Long-term follow-up of patients with ‘idiopathic’ HES often reveals a clonal process. The HES has been further sub-categorized into several variants (Table 7). These variants are distinct entities with clinically important differences in diagnosis, therapeutics and prognosis.

Table 7. HES variants Myeloproliferative variants T-cell lymphocytic variants (L-HES) Familial HES Idiopathic HES Organ-restricted HES Specific or defined syndromes associated with hypereosinophilia episodic angioedema with eosinophilia In the lymphocytic variant, it is postulated that an abnormal clonal proliferation of Th2 helper cells is responsible for the profound eosinophilia.[46] The myeloproliferative type is so termed because it shares features common to other myeloproliferative diseases,

REVIEW including hepatosplenomegaly, cytopenias, elevated serum vitamin B12 and presence of immature forms in peripheral blood.

Males are seven to nine times more likely to be affected than females. Usual age of onset is in the third of fourth decade.

Constitutional symptoms such as anorexia, night sweats and fever dominate the presentation. Cardiac involvement portends a poor prognosis; endomyocardial fibrosis is the main cardiac manifestation and is more commonly seen in the myeloproliferative variant.[47] In a retrospective review of 50 patients with HES,[48] 40% of patients suffered pulmonary involvement, 62% of patients in the study suffered neurological involvement including thrombotic cerebrovascular accidents, cognitive decline, movement disorders and peripheral neuropathy and 56% of patients had skin manifestations (angioedema, dermatographism and urticaria).

Organ cytotoxicity is largely caused by eosinophilic cationic granules such as MBP. In addition, cationic proteins induce a hypercoagulable state resulting in endothelial dysfunction with microangiopathies and cardiac mural thrombi. Long-standing HES may result in pulmonary fibrosis.

HES should be considered in patients with persistent blood hypereosinophilia on two occasions at least a month apart, regardless of whether symptoms are present or not. Once all known causes have been ruled out, a search for end-organ damage should be undertaken.

Functional and anatomical assessments of the cardiovascular, pulmonary and gastrointestinal system should be undertaken and tissue samples obtained where appropriate. Bone marrow aspiration and trephine (BMAT) will always show increased mature eosinophils and precursor forms. It may help to identify a clinically important subtype or previously unknown cause. Routinely performed tests on BMAT include karyotyping, in situ hybridisation techniques for known mutations such as the Fip1-like1-platelet-derived growth factor receptor alpha (FIP1L1-PDGFRA)-associated mutation, CD34 expression and molecular testing for the JAK 2 mutation.[49] Corticosteroids are effective first-line therapy in less than half of patients,[5] necessitating the use of other treatment modalities including chemotherapeutic agents, cyclosporine and interferon-α.[47,50,51] Imatinib, a tyrosine kinase inhibitor, is effective in patients with the myeloproliferative form of HES who are refractory to steroids, hydroxyurea or interferon-α. Mepolizumab, an anti-IL-5 monoclonal antibody, may be effective as well.[52] Due to better diagnostics and therapeutic options, 10-year survival rates may be as high as 70%.[50]

Conclusions

Clearly, eosinophilic lung diseases are a heterogeneous group of disorders that are not easily classifiable. Peripheral eosinophilia is a diagnostic clue. In its absence, the diagnosis is often not considered until eosinophilia is noted on BAL fluid or a lung biopsy specimen or radiological appearance is thought to be suggestive. History and examination provide vital diagnostic information and a thorough inquiry into prescription, non-prescription and illicit drug use and other possible exposures is vital. Information regarding the severity and duration of symptoms can help to narrow the differential. A history of asthma may be found in EGPA, CEP and ABPA. Travel to tropical areas (recent or remote) raises the possibility of parasitic infections. Ancillary investigations can be of use, especially in ABPA and pulmonary eosinophilia associated with systemic disease.

Symptoms of an underlying auto-immune disease or malignancy should be specifically sought.

References

1. Campos LEM, Pereira LFF. Pulmonary eosinophilia. J Bras Pneumol 2009;35(6):561573.

2. Cottin V, Cordier J-F. Eosinophilic pneumonias. Allergy 2005;60(7):841-857. [http:// dx.doi.org/10.1111/j.1398-9995.2005.00812.x] 3. Shi HZ, Xiao CQ, Li CQ, et al. Endobronchial eosinophils preferentially stimulate T helper cell type 2 responses. Allergy 2004;59(4):428-435.

4. Rothenberg ME, Hogan SP. The eosinophil. Annu Rev Immunol 2006;24:147-174. [http://dx.doi.org/10.1146/annurev.immunol.24.021605.090720] 5. Allen JN, Davis WB. Eosinophilic lung diseases. Am J Respir Crit Care Med 1994;150(5 Pt 1):1423-1438. [http://dx.doi.org/10.1164/ajrccm.150.5.7952571] 6. Allen JN, Davis WB, Pacht ER. Diagnostic significance of increased bronchoalveolar lavage fluid eosinophils. Am Rev Respir Dis 1990;142(3):642-647. [http://dx.doi. org/10.1164/ajrccm/142.3.642] 7. Löffler W. Zur differential-diagnose der lungenininfiltrierungen. Ill Űber flüchtige succedan-infiltrate (mit eosinophilie). Beitr Klin Tuberk 1932;79:368-392.

8. Lenczner M, Spaulding WB, Sanders DE. Pulmonary manifestations of parasitic infestations. Can Med Assoc J 1964;91:421-434.

9. Phills JA, Harrold AJ, Whiteman GV. Pulmonary infiltrates, asthma and eosinophilia due to Ascaris suum infestation in man. N Engl J Med 1972;286(18):965-970.

10. Marchand E, Raynaud-Gaubert M, Lauque D. Idiopathic chronic eosinophilic pneumonia: A clinical follow-up study of 62 cases. Medicine (Baltimore) 1998;77(5):299-312.

11. Carrington CB, Addington WW, Goff AM. Chronic eosinophilic pneumonia. N Engl J Med 1969;280(15):787-798.

12. Marchand E, Cordier J-F. Idiopathic chronic eosinophilic pneumonia: A review.

Orphanet Journal of Rare Diseases 2006;1(11):1-4.

13. Thomeer MJ, Constabel U, Rizzato G. Comparison of registries of interstitial lung diseases in three European countries. Eur Res J Suppl 2001;32:114s-118s.

14. Jederlinic PJ, Sicilian L, Gaensler EA. Chronic eosinophilic pneumonia: A report of 19 cases and review of the literature. Medicine (Baltimore) 1988;67(3):154-162.

15. Libby DM, Murphy TF, Edwards A, Gray G, King TK. Chronic eosinophilic pneumonia: An unusual cause of acute respiratory failure. Am Rev Respir. Dis 1980;122(3):497-500.[http://dx.doi.org/10.1164/arrd.1980.122.3.497] 16. Steinfeld S, Golstein M, De Vuyst P. Chronic eosinophilic pneumonia (CEP) as a presenting feature of Churg-Strauss syndrome (CSS). Eur Resp J 1994;7(11):2098.

17. Yoshida K, Shijubo N, Koba H, et al. Chronic eosinophilic pneumonia progressing to lung fibrosis. Eur Respir J 1994;7(8):1541-1544.

18. Yaseen S, Samman MD, Siraj O, et al. Chronic eosinophilic pneumonia presenting with recurrent massive bilateral pleural effusion: Case report. Chest 2001;119(3):968-970.

19. Oyama Y, Fujisawa T, Hashimoto D, et al. Efficacy of short-term prednisolone treatment in patients with chronic eosinophilic pneumonia. Eur Respir J 2015;45(6):1624-1631. [http://dx.doi.org/10.1183/09031936.00199614] 20. Marchand E, Etienne-Mastroïanni B, Chanez P, et al. Idiopathic chronic eosinophilic pneumonia and asthma: How do they influence each other? Eur Respir J 2003;22(1):8-13.

21. Kaya H, Gümüȿ S, Uçar E, et al. Omalizumab as a steroid-sparing agent in chronic eosinophilic pneumonia. Chest 2012;142(2):513-516.[http://dx.doi.org/10.1378/chest.11-1881] 22. Allen JN, Pacht ER, Gadek JE, Davis WB. Acute eosinophilic pneumonia as a reversible cause of non-infectious respiratory failure. N Engl J Med 1989;321(9):569574. [http://dx.doi.org/10.1056/NEJM198908313210903] 23. Philit F, Ettiene-Mastroïanni B, Parrot A, Guérin C, Robert D, Cordier JF. Idiopathic eosinophilic pneumonia: A study of 22 patients. Am J Respir Crit Care Med 2002;166(9):1235-1239.[http://dx.doi.org/10.1164/rccm.2112056] 24. Rom WN, Weiden M, Garcia R, et al. Acute eosinophilic pneumonia in a New York City firefighter exposed to World Trade Centre dust. Am J Respir Crit Care Med 2001;166(6):797-800.[http://dx.doi.org/10.1164/rccm.200206-576OC] 25. Brackel CLH, Ropers FG, Vermaas-Fricot SFN, et al. Acute eosinophilic pneumonia after recent start of smoking. Lancet 2015;385(9973):1150. [http://dx.doi.org/10.1016/ S0140-6736(15)60128-3] 26. Akuthota P, Weller PF. Eosinophilic pneumonias. Clin Microbiol Rev 2012;25(4):649660. [http://dx.doi.org/10.1128/CMR.00025-12] 27. Johokoh T, Müller NL, Akira M, et al. Eosinophilic lung disease: Diagnostic accuracy of thin-section CT in 11 patients. Radiology 2000;216(3):773-780. [http://dx.doi. org/10.1148/radiology.216.3.r00se01773] 28. Hayakawa H, Sato A, Toyoshima M, Imokawa S, Taniquchi M. A clinical study of idiopathic eosinophilic pneumonia. Chest 1994;105(5):1462-1466.

29. Stevens DA, Moss RB, Kurup VB, et al. Allergic bronchopulmonary aspergillosis in cystic fibrosis - state of the art: Cystic Fibrosis Foundation Consensus Conference.

Clin Infect Dis 2003;37 Suppl 3:S225-S264.

SARJ VOL. 22 NO. 1 2016

17

REVIEW 30. Agarwal R, Aggarwal AN, Gupta D, Jindal SK. Aspergillus hypersensitivity and allergic bronchopulmonary aspergillosis in patients with bronchial asthma: Systematic review and meta-analysis. Int J Tuberc Lung Dis 2009;13(8):936-944.

31. Patterson R, Greenberger PA, Halwig JM, Liotta JL, Roberts M. Allergic bronchopulmonary aspergillosis: Natural history and classification of early disease by serologic and roentgenographic studies. Arch Intern Med 1986;146(5):916-918.

32. Patterson R, Greenberger PA, Radin RC, Roberts M. Allergic bronchopulmonary aspergillosis: Staging as an aid to management. Ann Intern Med 1982;96(3):286-291.

33. Akuthota P, Weller PF [Internet]. Allergic bronchopulmonary aspergillosis. [accessed Oct 2015]. Available from: http://www.utpodate.com 34. Voskamp AL, Gillman A, Symons K, et al. Clinical efficacy and immunological effects of omalizumab in allergic bronchopulmonary aspergillosis. J Allerg Clin Immunol Pract 2015;3(2):192-199. [http://dx.doi.org/10.1016/j.jaip.2014.12.008] 35. Churg J, Strauss L. Allergic granulomatosis, allergic angiitis, and periarteritis nodosa.

Am J Pathol 1951;27(2):277-301.

36. Jeanette JC, Falk RJ, Andrassy K, et al. Nomenclature of systemic vasculitides. Proposal of an international consensus conference. Arthritis Rheum 1994;37(2):187-192.

37. Radu AS, Levi M. Antineutrophil cytoplasmic antibodies. J Bras Pneumol 2005; 31(suppl1):S16-S20.

38. Lanham JG, Elkon KB, Pusey CD, Hughes GR. Systemic vascultitis with asthma and eosinophilia: A clinical approach to the Churg-Strauss syndrome. Medicine (Baltimore) 1984;63(2):65-81.

39. Sale S, Patterson R. Recurrent Churg-Strauss vasculitis. With exophthalmos, hearing loss, nasal obstruction, amyloid deposits, hyperimmunoglobulin E, and circulating immune complexes. Arch Intern Med 1982;141(10):1363-1365.

40. Chumbley LC, Harrison EG, DeRemee RA. Allergic granulomatosis and angiitis (Churg-Strauss syndrome). Report and analysis of 30 cases. Mayo Clin Proc 1977;52(8):477-484.

18 SARJ VOL. 22 NO. 1 2016

41. Cooper BJ, Bacal E, Patterson R. Allergic angiitis and granulomatosis: Prolonged remission induced by combined prednisone-azathioprine therapy. Arch Intern Med 1978;138(3):367-371.

42. MacFayden R, Tron V, Keshmiri M, Road JD. Allergic angiitis of Churg and Strauss syndrome. Response to pulse methylprednisolone. Chest 1987;91(4):629-631.

43. Lie JT. Limited forms of Churg-Strauss syndrome. Pathol Annu 1993;28 Pt2:199-220.

44. Keogh KA, Specks U. Churg-Strauss syndrome: Clinical presentation, antineutrophil cytoplasmic antibodies and leukotriene receptor antagonists. Am J Med 2003;115(4):284-290.

45. Valent P, Klion AD, et al. Contemporary consensus proposal on criteria and classification of eosinophilic disorders and related syndromes. J Allergy Clin Immunol 2012; 130(3):607.

46. Raghavacher A, Fleischer S, Frickhofen N, Heimpel H, Fleischer B. T lymphocyte control of human eosinophil granulopoeisis. Clonal analysis in an idiopathic hypereosinophilic syndrome. J Immunol 1987;139(11)9:3753-3758.

47. Zielinski RM, Lawrence WD. Interferon-alpha for the hypereosinophilic syndrome.

Ann Intern Med 1990;113(9):716-718.

48. Fauci AS, Harley JB, Roberts WC, et al. NIH conference. The idiopathic hypereosinophilic syndrome. Clinical, pathophysiologic and therapeutic considerations. Ann Intern Med 1982;97(1):78-92.

49. Roufosse F, Goldman M, Cogan E. Hypereosinophilic syndrome: Lymphoproliferative and myeloproliferative variants. Sem Respir Crit Care Med 2006;27(2):158-170. [http://dx.doi.org/10.1055/s-2006-939519] 50. Zabel P, Schlaak M. Cyclosporin for hypereosinophilic syndrome. Ann Hematol 1991;62(6):230-231.

51. Smit AJ, van Essen LH, de Vries EG. Successful long-term control of idiopathic hypereosinophilic syndrome with etoposide. Cancer 1991;67(11):2826-2827.

52. Khoury MJ, Newman JH, Murray JJ. Reversal of hypereosinophilic syndrome and lymphomatoid papulosis with mepolizumab and imatinib. Am J Med 2003;115(7):587-589.

CASE REPORT

Pneumocystis jiroveci and cytomegalovirus co-infection in an immunocompromised patient D Simon,1 MB ChB, FCP, MMed; K E Greyling,2 MB ChB, FCP, MMed, Cert Infectious Diseases; E M Irusen,1 MB ChB, FCP, PhD; J Rigby,3 MB ChB, FC Path (SA) Anat; J J Taljaard,2 MB ChB, MMed, DTM&H; C F N Koegelenberg,1 MB ChB, FCP, FRCP, PhD Division of Pulmonology, Department of Medicine, Stellenbosch University and Tygerberg Academic Hospital, Cape Town, South Africa Division of Infectious Diseases, Department of Medicine, Stellenbosch University and Tygerberg Academic Hospital, Cape Town, South Africa 3 Division of Anatomical Pathology, Department of Pathology and National Health Laboratory Service, Stellenbosch University and Tygerberg Academic Hospital, Cape Town, South Africa 1 2

Corresponding author: D Simon (siya_bonga@hotmail.com)

Pneumocystis jiroveci pneumonia (PCP) and cytomegalovirus (CMV) are opportunistic infection seen in patients with advanced immunocompromised states, such as HIV infection. We present a case of PCP-CMV co-infection in a patient with newly diagnosed HIV disease. The presence of CMV in the context of another opportunistic respiratory tract infection is often presumed non-invasive and not treated. Our report highlights that this is not always the case. Invasive CMV disease can be easily misdiagnosed and remains a potentially fatal affliction. We postulate that the use of high-dose corticosteroids used as an adjunct in the treatment of serious PCP can lead to the reactivation of CMV infection and clinical disease. Moreover, we suggest that there may be a role for serial viral load measurement, analogous to protocols often utilised in solid-organ transplant patients receiving immunosuppression.

S Afr Respir J 2016;22(1):20-21. DOI: 10.7196/SARJ.2016.v22i1.40

Pneumocystis jiroveci pneu­ monia (PCP) and cyto­ megalovirus (CMV) are opportunistic infections seen in patients with advanced immunocompromised states, including human immunodeficency virus (HIV). PCP and CMV pneumonitis present in a similar manner, but the diagnosis of CMV infection is often far more challenging.[1] The poor sensitivity and specificity of CMV viral load often delays treatment of this potentially lethal co-infection. In addition, the presence of CMV in bronchoalveolar lavage (BAL) fluid or PCP polymerase chain reaction (PCR) in patients with pneumocystis pneumonia denotes a poor prognosis.[1]

Case report

A 48-year-old female presented with a 2-week history of a productive cough, dyspnoea and fever. She reported weight loss of >10 kg over the preceding year. She was not known with any significant past medical history and had a 10 pack-year smoking history.

Clinical examination revealed an acutely ill patient who was tachypnoeic with oxygen saturation on room air of 84%. Bilateral

20 SARJ VOL. 22 NO. 1 2016

Fig. 1. Chest radiograph on admission. Note the diffuse reticulo-nodular infiltrates. inspiratory crackles were present. The rest of her clinical examination was noncontributory.

A chest radiograph at presentation revealed bilateral interstitial infiltrates (Fig. 1). An HIV enzyme-linked immunosorbent assay test (ELISA) confirmed that she was HIV infected.

Her serum lactate dehydrogenase was raised at 646 IU/L and her CD4 count was 87 cell/mm3 Sputum analysis confirmed P. jiroveci cysts. No other pathogens (including Mycobacterium tuberculosis) were identified in her sputum.

The patient was initiated on high-dose oral trimethoprim-sulphomethoxazole and

Fig. 2. A high-resolution computed tomography scan showing ground-glass infiltrates. oral prednisone in addition to supplemental face-mask oxygen. She remained hypoxic and oxygen-dependent at the end of her 3-week treatment course. Her chest radiograph showed a worsening ground-glass pattern.

A high-resolution computed tomography chest confirmed an interstitial process with a ground-glass appearance as well as honeycombing (Fig. 2).

The differential diagnosis at this point included cotrimoxazole-resistant P. jiroveci infection, other opportunistic infections and non-infective causes such as lymphocytic interstitial pneumonitis or nonspecific interstitial pneumonia. Her CMV viral load

CASE REPORT

A

B

Fig. 3. The histology was diagnostic of CMV infection with (A) a diagnostic basophilic nuclear inclusion surrounded by a clear halo and indistinct basophilic cytoplasmic inclusions (H&E). Immunohistochemistry confirmed diffuse nuclear and cytoplasmic positivity. (B) Black intra-alveolar cup-shaped cysts of P. jiroveci highlighted by Grocott’s methenamine silver stain were also seen. was 957 768 copies/mL. She was deemed unfit for bronchoscopy due to persistent respiratory failure. An open lung biopsy demonstrated a frothy alveolar exudate with numerous cysts of P. jiroveci as well as CMV inclusion bodies (Fig. 3). Intravenous ganciclovir was initiated, but the patient died 4 days later despite full medical support, including mechanical ventilation.

Discussion

The advent of antiretroviral treatment has brought about a dramatic reduction in the number of HIV-infected patients presenting with opportunistic infections with P. jiroveci and CMV. Historically, the presence of CMV in the context of another opportunistic respiratory tract infection was presumed non-invasive and not treated. [2] Evidence suggests no difference in outcome between CMV-treated and CMV-untreated groups.[2]

The majority of studies were, however, conducted prior to the use of corticosteroids as standard ancillary treatment for hypoxia in PCP infection.[3] It is plausible that the use of high-dose corticosteroids can lead to the reactivation of CMV infection and clinical disease.

The diagnostic challenge is to distinguish between CMV infection and invasive CMV disease. CMV infection implies the presence of CMV evidenced by direct isolation, culture, seroconversion, a fourfold increase in titre, antigen detection or identification of CMV DNA by molecular techniques.

In contrast, invasive CMV disease is characterised by an infectious syndrome and specific organ dysfunction together with pathological changes.[4] Radiological findings may include ground-glass attenuation, interstitial infiltrates (both reticular and/or nodular), dense consolidation, bronchial wall thickening and even apparent mass lesions.[1] Toyoda et al.[3] investigated the use of the total CMV viral burden in the serum as a marker for invasive CMV disease. Using quantitative PCR in a cohort of cardiac and renal transplant recipients, they found that there was a direct correlation between the CMV viral load and invasive disease.

Diagnostic techniques have improved considerably since 1997 with real-time PCR currently the gold standard in determining viral load. Due to the absence of an internationally standardised reference reagent for quantitative PCR prior to 2010, it is impossible to compare studies using different assays from different laboratories in that era. Viral loads of 2 000 - 5 000 IU/mL in the context of a clinical syndrome consistent with CMV infection are generally believed to be consistent with invasive CMV disease in transplant patients. The Transplantation Society International CMV Consensus Group and the American Society of Transplantation guidelines on CMV infection in solid-organ transplantation recommend plasma blood viral load testing and state that laboratories must establish their own cut-offs to validate the trigger points used to initiate therapy.[5] We believe that a similar approach should be followed in other immunocompromised

groups and that in-house cut-offs should be developed in conjunction with the local virology laboratory.

CMV viral load must be monitored using real-time quantitative PCR at baseline when steroids are initiated and then weekly thereafter until there is clinical improvement. An increase in viral load is often predictive of early disease and should be considered as an alternative to weekly testing in patients who are clinically deteriorating. [6] All viral load testing should be followed by thorough clinical investigation e.g., fundoscopy, repeat chest radiography and if pneumonitis is suspected, a BAL for CMV quantitative PCR and appropriate cytological examination.

The index case not only demonstrates that true PCP and CMV pneumonitis can occur in a newly diagnosed HIV-infected patient, but also that CMV pneumonitis carries a high mortality despite appropriate treatment.

Consent: Informed consent to report this case was obtained from one of the patient’s living relatives as the patient was on invasive ventilatory support and sedated at the time.

References 1. Salomon N, Gomez T, Perlman D, Laya L, Eber C, Mildvan D. Clinical features and outcome of HIV-related cytomegalovirus pneumonia. AIDS 1997;11(3):319-324.

2. Jacobson MA, Mills J, Rush J, et al. Morbidity and mortality of patients with AIDS and first-episode Pneumocystis carinii pneumonia unaffected by concomitant pulmonary cytomegalovirus infection.

Am Rev Respir Dis 1991;144(1):6-9.[http://dx.doi. org/10.1164/ajrccm/144.1.6] 3. Toyoda M, Carlos JB, Galera OA, et al. Correlation of cytomegalovirus DNA levels with response to antiviral therapy in cardiac and renal allograft recipients.

Transplantation 1997;63(7):957-963.

4. Kraft CS, Armstrong WS, Caliendo AM. Interpreting quantitative cytomegalovirus DNA testing: Understanding the laboratory perspective. Clin Infect Dis 2012;54(12):17931797. [http://dx.doi.org/10.1093/cid/cis212] 5. Kotton CN, Kumar D, Caliendo AM, et al. Updated international consensus guidelines on the management of cytomegalovirus in solid-organ transplantation.

Transplantation 2013;96(4):333-360. [http://dx.doi. org/10.1097/TP.0b013e31829df29d] 6. Emery VC, Sabin CA, Cope AV, et al. Application of viral-load kinetics to identify patients who develop cytomegalovirus disease after transplantation.

Lancet 2000;355(9220):2032-2036. [http://dx.doi. org/10.1016/S0140-6736(00)02350-3]

SARJ VOL. 22 NO. 1 2016

21

CASE REPORT

A paradoxical cause of hypoxia and orthodeoxia in a stroke patient P Soma,1 MB ChB, MSc (Clin Epi); D Joseph,2 MB ChB; S Ahmad,3 MBBS FCRAD (SA) Diploma in Neurointervention; S Ellemdin,2 MB ChB MMed (Int Med) Department of Physiology, School of Medicine, Faculty of Health Sciences, University of Pretoria, South Africa epartment of Internal Medicine, School of Medicine, Faculty of Health Sciences, University of Pretoria and Steve Biko Academic Hospital, Pretoria, South Africa D 3 Department of Radiology, School of Medicine, Faculty of Health Sciences, University of Pretoria and Steve Biko Academic Hospital, Pretoria, South Africa 1 2

Corresponding author: P Soma (prashilla.soma@up.ac.za)

The classic triad of dyspnoea on exertion, cyanosis and clubbing should alert the clinician to the possibility of pulmonary arteriovenous malformation (PAVM). Despite it being a rare condition, it is associated with significant clinical complications. A common feature is paradoxical emboli, with consequent complications such as brain abscess and stroke.

S Afr Respir J 2016;22(1):22-23. DOI: 10.7196/SARJ.2016.v22i1.42

Pulmonary arteriovenous mal­ formations (PAVMs) are best defined as low-resistance, high-flow abnormal vascular structures that often connect a pulmonary artery to a pulmonary vein, bypassing the normal pulmonary capillary bed and resulting in an intrapulmonary right-to-left shunt.[1] The development of the right-left shunt results in hypoxia with shunting of micro emboli with central nervous system compli­cations.[2] PAVMs are most commonly congenital in nature with about 70% of cases associated with hereditary haemorrhagic telangiectasia (HHT). [2] Physiologic consequences depend on the degree of right-to-left shunt and include hypoxaemia, dyspnoea, and cyanosis.[1] Treatment options include surgery, embolisation therapy as well as hormonal and pharmacological interventions to prevent bleeding from arteriovenous malformations (AVMs).[2] The triad of dyspnoea, cyanosis and clubbing is described as classic for PAVMs but may not always be present.[2] We describe a case of a young female presenting with a brain abscess and progressive neurological fallout. The presence of hypoxia and clubbing were clues to her underlying lung condition.

Case report

A 21-year-old woman presented to her local hospital with weakness of her right arm and leg as well as aphasia. Her symptoms started in September 2014 with impairment of speech followed by weakness over a period of 1 week.

This was accompanied by a severe headache, nausea, vomiting, photophobia and blurred vision. Her past medical history was noncontributory. The patient assumed her unusual

22 SARJ VOL. 22 NO. 1 2016

nails and blue discolouration of her lips and tongue were normal for her. She was burdened by dyspnoea on exertion while growing up. A computed tomography (CT) brain scan was done and she was discharged as having had a stroke after 2 weeks’ inpatient care. With no improvement in her condition, she presented to our academic institution, Steve Biko Academic Hospital, Pretoria, in October 2014 with a complaint of chest pain.

On examination, she was afebrile, blood pressure was 97/77 mmHg and heart rate was 90 bpm. She had both peripheral and central cyanosis with clubbing but insufficient criteria to meet the diagnosis of HHT (no history of epistaxis, absent telangiectasia and no family history). Her Glasgow coma scale (GCS) was 10/15, with abnormalities on the right side of the body including an upper motor neuron facial nerve palsy, 0/5 power, reduced tone and positive clonus. In addition, there was hemisensory loss of all modalities.

Her CT brain scan confirmed a left frontal lobe, thick-walled mass with surrounding vasogenic oedema suggestive of a brain abscess, and drainage thereof was performed.

Laboratory findings revealed a PaO 2 7.67 kPa (57.5 mmHg), saturation of 91.4% on room air on admission. Supine and sitting blood gases varied with sitting PaO2 5.87 kPa (44.0 mmHg), saturation of 77.8% compared with supine PaO2 6.91 kPa (51.8 mmHg), saturation of 85.9%. Despite the vascular abnormality located in the upper lobe, the patient presented with orthodeoxia. To exclude other causes of her stroke, antinuclear antibodies screen was performed and HIV serology, both of which were negative.

Her electrocardiogram and echocardiogram were normal. Her chest X-ray showed a large lobulated opacity in the right upper lung zone almost adjacent to the right hilum with some air bronchograms noted through the opacity.

A CT scan of the chest confirmed the right upper lobe PAVM. This is likely causing a shunt that precipitated the septic embolism to the brain. The patient was referred to interventional radiology for AVM coiling where multiple coils were inserted as shown A

B

Fig. 1. (A) Pulmonary angiogram of right lung.

(B) Embolisation of PAVM with few minor branches in AVM still visible.

CASE REPORT in Fig 1. Post intervention her blood gases improved to a PaO2 12.22 kPa (91.7 mmHg) and saturation of 96.8%. However, upon followup her hypoxia had recurred, so on the 14 October 2015, the patient underwent a right and middle lobe lobectomy as the embolization resulted in partial improvement, the coils had opened up.

Discussion

PAVMs are abnormal communications between pulmonary veins and arteries.[2] Despite their rarity, mortality attributable to PAVMs is caused by serious neurological complications such as stroke, brain abscess, transient ischaemic attacks, migraine and seizures.[1,3] The natural history of PAVMs is that they are inclined to increase in size and rarely regress spontaneously. Complications are more likely in the presence of hereditary haemorrhagic telangiectasia.[4] Clinical presentation is a result of the direct communications between pulmonary and systemic circulations, bypassing the capillary bed, which leads to physiological abnormalities, in particular right-to-left shunts which cause hypoxaemia with subsequent paradoxical embolism.[5] PAVMs may be single or multiple, unilateral or bilateral, and simple or complex. Most PAVMs are congenital with acquired causes including conditions such as post-thoracic surgery, trauma, tuberculosis, actinomycosis and schistosomiasis.[2,3] 53 - 70% of PAVMs are found in lower lobes with a preponderance of unilateral disease.[3] Clinically, patients present most commonly with epistaxis, dyspnoea, sometimes platypnoea and haemoptysis, but may be asymptomatic. Less common complaints include cough, chest pain, migraine headaches, tinnitus, dizziness, dysarthria, syncope, vertigo and diplopia. Signs commonly found include bruit, clubbing, cyanosis and telangiectasia.[2]

Conclusion

The clinical features of PAVMs resemble many respiratory conditions, a chest radiograph is thus mandatory to diagnosis PAVMs with CT or pulmonary angiography being diagnostic. Chest radiography is an important diagnostic tool and reveals abnormalities in 98% of patients.[3] Classic finding on chest radiograph is that of a round or oval mass of uniform density, often lobulated but sharply defined, more commonly

in lower lobes and ranging from 1 to 5 cm in diameter.[2] However, chest CT is more accurate in identifying connecting vessels and PAVM. The diagnosis of PAVM should be suspected with any of the following presentations: (a) classic findings on chest radiograph as described, (b) mucocutaneous telangiectasis and (c) unexplained findings such as dyspnoea, haemoptysis, hypoxaemia, polycythaemia, clubbing, cyanosis, cerebral embolism or brain abscess.[2] Advances in endovascular techniques have made embolotherapy the preferred treatment option which reduces the need for surgical intervention. The practice of careful and precise angiography techniques utilising modern coaxial catheters, coils and other embolic devices, means almost all PAVMs can be successfully treated.[2] Key learning points • PAVMs are rare pulmonary vascular anomalies, with more than 80% being congenital.

• PAVMs can cause dyspnoea with right-to-left shunts, commonly with cyanosis, clubbing and pulmonary vascular bruit.

• Because of paradoxical emboli, patients can present with central nervous system complications.

• Chest radiography and contrast CT are essential diagnostic tools, but pulmonary angiography is the gold standard.

• Treatment options include angiography techniques utilising modern coaxial catheters, coils and other embolic devices.

• Advanced treatment modalities means almost all PAVMs can be successfully treated.

References 1. Meek ME, Meek JC, Beheshti MV. Management of pulmonary arteriovenous malformations.

Semin Intervent Radiol 2011;28(1):24-31.[http://dx.doi.org/10.1055/s-0031-1273937] 2. Gossage JR, Kanj G. Pulmonary arteriovenous malformations: A state of the art review. Am J Respir Crit Care Med 1998;158(2):643-661.[http://dx.doi.org/10.1164/ajrccm.158.2.9711041] 3. Bandyopadhyay SK, Nandy A, Sarkar S, Ghosal AG. Massive haemothorax: A presentation of pulmonary arteriovenous malformation. Indian J Chest Dis Allied Sci 2008;50(3):285-287.

4. Todo K, Moriwaki H, Higashi M, Kimura K, Naritomi H. A small pulmonary arteriovenous malformation as a cause of recurrent brain embolism. AJNR Am J Neuroradiol 2004;25(3):428-430.

5. Shovlin CS, Letarte M. Hereditary haemorrhagic telangiectasia and pulmonary arteriovenous malformations: Issues in clinical management and review of pathogenic mechanisms. Thorax 1999;54(8):714-729.

Superior mediastinal masses in children – two cases of lymphoma A C Jeevarathnum, MB BCh, FCPaed (SA), Dip Allergy (SA), MMed, Cert Paed Pulm (SA), European Respiratory Diploma; A van Niekerk, MB BCh, MMed; D Parris, BSc, MB BCh, FCPaed (SA), Dip Allergy (SA); K De Campos, MB ChB, MMed, Dip Allergy (SA); W Wijnant, MD Paed, Dip Allergy (SA), Cert Paed Pulm(SA); X Deadren, MB ChB, FCPaed (SA), MMed; A Büchner, MB ChB, DCH (SA), FCPaed(SA), MMed, Dip Pall Med, Cert Med Oncol (Paed)(SA); F Omar, MB ChB, FCPaed (SA), Cert Paed Med, Onc Paed (SA); D Reynders, MB ChB, FCPaed (SA), MRCPCH, Cert Paed Med, Onc Paed (SA); R J Green, PhD, DSc Department of Paediatrics and Child Health, School of Medicine, Faculty of Health Sciences, University of Pretoria and Steve Biko Academic Hospital, Pretoria, South Africa Corresponding author: A C Jeevarathnum (acjeevarathnum@gmail.com)

The exact incidence of superior mediastinal masses in children is largely unknown. They present as a spectrum of disease ranging from an incidental finding on a chest X-ray to being markedly symptomatic with superior vena caval syndrome or obstruction of the upper airways.

Lymphomas are the most common causes of superior mediastinal masses in children. We present two cases of confirmed T-cell lymphoma

SARJ VOL. 22 NO. 1 2016

23

CASE REPORT in children with superior mediastinal masses. In doing so, we explore a diagnostic approach and visit the complications the physician needs to be aware of when confronted with a child with a superior mediastinal mass.

S Afr Respir J 2016;22(1):23-25. DOI: 10.7196/SARJ.2016.v22i1.58

Case 1

The first patient is a 4-yearold HIV-negative male who presented to Steve Biko Academic Hospital with a 1-week history of cough and shortness of breath. This was the first time this child had been ill. There was no other significant history and there were no tuberculosis (TB) contacts.

Clinically this child was in moderate respiratory distress and was oxygen dependent. There was shift of the mediastinum to the right and stony dullness to percussion on the left, indicative of a left-sided pleural effusion. There were no significant peripheral nodes that could be biopsied and there was no hepatosplenomegaly.

A frontal chest X-ray (CXR) of this child confirmed a left-sided pleural effusion and a widened superior mediastinum as evidenced

by Fig. 1. The widened superior mediastinum became more apparent on drainage of the effusion. The computed tomography (CT) confirmed the presence of a large superior mediastinal mass as in Fig. 2. The effusion was exudative in nature with a very high adenosine deaminase level of 184 U/L. Cytology of the effusion revealed atypical lymphocytes suggestive of a malignancy.

The childâ&#x20AC;&#x2122;s white cell count was normal with no atypical lymphocytes on smear.

Tumour markers including serum lactate dehydrogenase (LDH) were not elevated.

The superior medistinal mass was biopsied by the cardiothoracics surgery team and a bone marrow aspirate and trephine (BMAT) was performed concurrently. The anaesthetic for the entire process was conducted in

Fig 1. Frontal chest X-ray of patient 1 revealing a widened superior mediastinum (white arrows) and an intercostal drain on the left (red arrows).

24 SARJ VOL. 22 NO. 1 2016

an extremely cautious fashion with a gas induction and with spontaneous respiration via endotracheal tube during the procedure.

The BMAT was not suggestive of malignancy.

Histology of the mass itself revealed a T-cell lymphoblastic lymphoma as depicted in Fig. 3.

Discussion

The second case was a 2-year-old HIV-negative male who presented with an acute history of cough and shortness of breath following a choking episode. Considering the history, a foreign body was the initial concern. This was a clinically well child with no respiratory symptomatology. The frontal CXR revealed an incidental finding of a widened superior mediastinum as depicted in Fig. 4. A contrasted CT scan of the chest confirmed a homogenous superior mediastinal mass in the anterior compartment, as shown in Figs. 5. There were no calcifications or cystic changes that would suggest a teratoma. The CT did reveal evidence of compression of the trachea although the child was clinically asymptomatic. In this case as well, there were no peripheral lymph nodes to biopsy and haematological workup including tumour markers was non-contributory. On awaiting theatre for a histological specimen of the mass, the patient had an unexpected cardiorespiratory arrest and unfortunately died.

A postmortem examination revealed no evidence of a foreign body aspiration as suggested by the history. Histology of the mass revealed a T-cell lymphoma. The cause of death was most likely upper airway obstruction from a very large tumour.

Neither patient received corticosteroids while awaiting theatre.

Depending on the compartment of the mediastinum involved, there are a number of causes of a widened mediastinum, as shown in Table 1.[1-3] This is imperative in trying to define the aetiology. The majority of mediastinal masses in children are malignant.[2] Lymphomas are the most common cause of mediastinal masses in the paediatric population. [1,4] Between 50 and 70% of patients with lymphoblastic lymphomas present with an anterior mediastinal or intrathoracic mass. [1] In the paediatric population, two-thirds of lymphomas occurring in the mediastinum are nonHodgkinâ&#x20AC;&#x2122;s lymphoma, and the remainder are Hodgkinâ&#x20AC;&#x2122;s lymphoma. The second most common cause of mediastinal masses in the anterior mediastinum are germ cell tumours including benign teratomas in addition to malignant seminomas and yolk sac tumours.

Germ cell tumours peak in incidence at 3 years of age and at adolescence. The mediastinum is the fourth most common site for teratomas. Neurogenic tumours including neuroblastoma are the most common causes of posterior mediastinal masses.

The diagnostic evaluation begins with a frontal and lateral chest X-ray in which 90% of mediastinal masses can be seen.[2] A CT scan of the chest is necessary to anatomically define the extent and nature of the mass, define the compartment of the mediastinum in which the mass occurs and to determine the degree of airway compression.[2] There are certain clues that could point to a specific diagnosis with fat, fluid and calcified components being more common in germ

Fig. 2. Axial contrasted CT)at the level of the carina demonstrating the large superior mediastinal mass (white arrows) with areas of necrosis and cystic change (red arrows).

Fig. 3. H&E section of the mass seen in figures revealing extensive lymphocyte proliferation (courtesy of Dr J Dinkel, Department of Anatomical Pathology, Tshwane Academic Division, NHLS).

Case 2

CASE REPORT

Table 1. Mediastinal masses by location Anterior

Middle

Posterior

Non-Hodgkin’s lymphoma

Vascular malformations

Neuroblastoma

Hodgkin’s lymphoma

Double aortic arch

Ganglioneuroblastoma

Germ cell tumour

Pulmonary artery sling

Ganglioneuroma

Hyperplastic/ectopic thymus

Aneurysms

Nerve sheath tumours

Fig. 4. Frontal chest X-ray of patient 2 demonstrating large superior mediastinal mass (arrows).

Thymoma

Bronchogenic/foregut cysts

cell tumours.[3,5] The involved compartment will guide the differential diagnosis.[5] Haematological markers that are useful include a full blood count with differential and (looking for atypical lymphocytes or blasts, or cytopenias indicating possible bone marrow infiltration) tumour markers (alpha fetoprotein, beta human chorionic gonadotrophin (HCG)) in the case of germ cell tumours. A BMAT is necessary in the case of a suspected haematological malignancy, as one-third of patients will have bone marrow involvement.[2] A biopsy of the mass itself is mandatory in those cases where a diagnosis cannot be made with peripheral specimens (biopsy of peripheral lymph nodes or other blood investigations). Biopsy of the mass itself can be obtained by a CT-guided procedure or via sternotomy. Certain tumour markers may assist with a diagnosis: for instance, elevated alpha foetoprotein and beta HCG would suggest a germ cell tumour.[5] A peripheral flow cytometry conducted on a patient with a very high white cell count would suggest a lymphoma/leukaemia. Elevated urinary levels of the catecholamine vanillylmandelic acid (VMA) and homovanillic acid (HVA) in a patient with a posterior mediastinal mass is suggestive of a neuroblastoma.[6] When planning a biopsy, the clinician needs to be aware that there is a significant anaesthetic risk with any form of sedation and induction of anaesthesia leading to possible acute airway obstruction, sudden cardiac arrest and death.[4,7-9] The risk is present even in the case of an asymptomatic lesion.[4] Children with superior mediastinal masses are at higher risk of an anaesthetic death than their adult counterparts.[9] Conscious sedation with spontaneous breathing during the procedure is the recommended method of anaesthesia in these patients and neuromuscular blockade is not advised.[4,8,9] The role of preoperative steroids needs to be clearly defined, the advantage of which will decrease the risk of airway obstruction

A

Mediastinal lymph nodes B

Fig. 5A and B. Axial (A) and sagittal (B) CT views of patient 2 revealing a large homogenous superior mediastinal mass (white arrows). Tracheal compression is noted in sagittal view (red arrow). by shrinking the tumour and improving the anaesthetic outcome.[10] However, this runs the risk of an inadequate biopsy specimen and could potentially interfere with a histological diagnosis.[7] In one series of 18 patients, preoperative steroids were used in patients with features of airway compromise and despite this a good histological sample was obtained in 95% of cases; prolonged use of steroids (>5 days) impaired histological diagnosis in 5% of cases.[7] Another series also concluded minimal interference with pathological diagnosis with the use of preoperative steroids in high-risk patients.[10] This is definitely an area that needs further research and exploration.

In case 2, starting preoperative steroids would have possibly resulted in tumour shrinkage and avoided an unfortunate demise.

This is definitely a learning point in the case.

Conclusion

Lymphomas are one of the most common causes of superior mediastinal masses in the paediatric population. A mass in the superior mediastinum usually requires a histological diagnosis. These patients, despite appearing clinically stable, can be challenging to manage and caution should be employed during the time of biopsy when anaesthetic is administered. The physician should consider the use of preoperative steroids in symptomatic individuals or

asymptomatic individuals with evidence of airway compression on imaging in order to improve the anaesthetic outcome.

References

1. Williams HJ, Alton HM. Imaging of paediatric mediastinal abnormalities. Paediatr Respir Rev 2003;4(1):55-66. [http://dx.doi.org/10.1016/S15260542(02)00310-X] 2. Jaggers J, Balsara K. Mediastinal masses in children.

Sem Thorac Cardiovasc Surg 2004;16(3):201-208. [http://dx.doi.org/10.1053/j.semtcvs.2004.08.005] 3. Ranganath SH, Lee EY, Restrepo R, Eisenberg RL. Mediastinal masses in children. ARJ Am J Roentgenol 2012;198(3):197216. [http://dx.doi.org/10.2214/AJR.11.7027] 4. Suominen PK, Kanerva JA, Saliba KJ, Taivainen TR. Unrecognised mediastinal tumor causing sudden tracheal obstruction and out-of-hospital cardiac arrest. J Emerg Med 2010;38(5):e63-e66. [http:// dx.doi.org/10.1016/j.jemermed.2007.10.065] 5. Kennebeck SS. Tumours of the mediastinum. Clin Ped Emerg Med 2005;6:156-164. [http://dx.doi. org/10.1016/j.cpem.2005.05.003] 6. Kushner BH. Neuroblastoma: A disease requiring a multitude of imaging studies. J Nucl Med 2004;45(7):1172-1188.

7. Hack HA, Wright NB, Wynn RF. The anaesthetic management of children with anterior mediastinal masses. Anaesthesia 2008;63(8):837-846. [http:// dx.doi.org/10.1111/j.1365-2044.2008.05515.x] 8. Gothard JW. Anaesthetic considerations for patients with anterior mediastinal masses. Anesthesiol Clin 2008;26:305-314. [http://dx.doi.org/10.1016/j. anclin.2008.01.002] 9. Slinger P, Karsli C. Management of the patient with a large anterior mediastinal mass: Recurring myths.

Curr Opin Anaesthesiol 2007;20(1):1-3.

10. Borenstein SH, Gerstle T, Malkin D, Thorner P, Filler RM. The effects of prebiopsy corticosteroid treatment on the diagnosis of mediastinal lymphoma. J Pediatr Surg 2000;35(6):973-976. [http://dx.doi.org/10.1053/jpsu.2000.6945]

SARJ VOL. 22 NO. 1 2016

25

BREATH-TAKING NEWS

The dilemma of persisting productive or wet cough in children without established chronic lung disease: How to treat and when to investigate Clinicians are frequently confronted by a chronic productive or wet cough in children. A cough of more than 4 weeks is defined as chronic. It is referred to as ‘productive’ when a child is able to expectorate. A similar cough in younger children who are unable to expectorate, will be referred to as a ‘wet’ cough. The most common conditions associated with a chronic productive/wet cough are protracted bacterial bronchitis and bronchiectasis. The early diagnosis, and effective management, of a chronic productive/wet cough is therefore important. It may signal an underlying illness where early and correct management may reduce morbidity and mortality.

Chang et al.[1] undertook two related systematic literature reviews aimed at answering two key questions (KQs) on children (≤14 years) who present for the first time with a chronic productive/wet cough in the absence of known chronic lung disease. KQ1: How effective are antibiotics in improving resolution of cough, and if so, which antibiotic(s) and for how long? KQ2: When should they be referred for further investigation? The authors followed the CHEST expert cough panel’s protocol and conducted systemic reviews of randomised controlled trials, and prospective and retrospective cohort crosssectional studies, published in English. Fifteen studies (data on 1 363 children) were included in KQ1 and 17 studies (data on 2 109 children) in KQ2. They concluded that: 1. There is high-quality evidence that the appropriate use of antibiotics improves cough resolution. The number needed to treat for benefit was 3 (95% confidence interval 2.0 - 4.3).

2. There is high-quality evidence that specific cough pointers[2] (e.g. digital clubbing, failure to thrive, cough during feeding, etc.)

should prompt further investigations (e.g. flexible bronchoscopy, chest computed tomography (CT) and immunity tests).

3. There is moderate-quality evidence that children should be referred for further investigation when a productive/wet cough does not improve after 4 weeks of antibiotic treatment.

The most common bacteria reported in the studies were Haemophilus influenzae (non-typeable), Moraxella catarrhalis and Streptococcus pneumoniae. The duration of antibiotic use varied from 7 days up to 8 weeks. A 2-week antibiotic course seemed sufficient in most patients. A minority of patients received 4 weeks of antibiotic treatment. Amoxicillin-clavulanate was the most frequent choice, followed by clarithromycin. Antibiotic stewardship must remain key.

More data are clearly needed on the specific criteria for antibiotic prescription in chronic wet cough. The clinical context of children presenting with a chronic productive/wet cough remains important when further investigations are ordered. These usually include chest CT, flexible bronchoscopy and appropriate laboratory investigations.

André van Niekerk Paediatric Pulmonologist, University of Pretoria and Steve Biko Academic Hospital

References

1. Chang AB, Oppenheimer JJ, Weinberger M, Rubin BK, Irwin RS. Children with chronic wet or productive cough – treatment and investigations: A systematic review.

Chest 2015; Epub. [http://dx.doi.org/10.1378/15-2065] 2. Chang AB, Glomb WB. Guidelines for evaluating chronic cough in paediatrics: ACCP Evidence-Based Clinical Practice Guidelines. Chest 2006;129(1_suppl):260S-283S. [http://dx.doi.org/10.1378/chest.129.1_suppl.260S]

S Afr Respir J 2016;22(1):26. DOI:10.7196/SARJ.2016.v22i1.61

Prevention of ventilator-associated pneumonia: Is there a role for colistin nebs?

Ventilator-associated pneumonia (VAP) is associated with a high mortality rate, increased duration of mechanical ventilation as well as additional costs. It is reassuring to note that the incidence of VAP has decreased over the last decade.

Colonisation of the lower respiratory tract, either by endogenous or exogenous pathogens, occurs rapidly after intubation. Endogenous pathogens come from contaminated oropharyngeal secretions and gastric contents. Exogenous contamination results from tracheal suctioning, fibre-optic bronchoscopy and/or ventilator circuit disconnection for nebulisations and patient transport. There is a

26 SARJ VOL. 22 NO. 1 2016

suggestion from previous studies that there might exist a continuum from colonisation of the lower respiratory tract to the development of VAP. Understanding the pathophysiology of VAP and improving preventive strategies has been the topic of recent research. The three broad categories of prevention include prevention of intubation itself, prevention of colonisation of the lower respiratory tract, and prevention of the progression of colonisation to development of VAP. The use of prophylactic antibiotics to prevent VAP is an area aimed at halting progression of colonisation to the development of VAP.

BREATH-TAKING NEWS A meta-analysis by Falagas et al.[2] studied the use of prophylactic inhaled gentamycin, polymixin, tobramycin and ceftazadine.

This study found VAP to be less frequent in those patients given prophylactic antibiotics. A recent study conducted by Karvouniaris et al.[3] randomised 168 patients to receive inhaled colistin (500 000 units) or normal saline three times a day for 10 days or until the patient was extubated. Unfortunately, there was no significant difference in the incidence of VAP. However, there was a significantly lower incidence of VAP associated with gram-negative bacilli.

Currently, the use of prophylactic inhaled antibiotics is not recommended in the prevention of VAP. However, further studies are required in this field.

Ashley C Jeevarathnum Paediatric Pulmonologist, University of Pretoria and Steve Biko Academic Hospital

References

1. Nseir S, Martin-Loeches I. In the name of ventilator-associated pneumonia prevention: Lung microbiota blown away by colistin! Eur Resp J 2015;46(6):15441547. [http://dx.doi.org/10.1183/13993003.01361-2015] 2. Falagas ME, Siempos II, Bliziotis IA, Michalopoulos A. Administration of antibiotics via the respiratory tract for the prevention of ICU-acquired pneumonia: A meta-analysis of comparative trials. Crit Care 2006;10(4):R123. [http://dx.doi.org/10.1186/cc5032] 3. Karvouniaris M, Makris D, Zygoulis P, et al. Nebulized colistin for ventilatorassociated pneumonia prevention. Eur Respir J 2015;46(6):1732-1739. [http://dx.doi. org/10.1183/13993003.02235-2014]

S Afr Respir J 2016;22(1):26-27. DOI:10.7196/SARJ.2016.v22i1.62

Who should receive respiratory syncitial virus prophylaxis?

Respiratory syncitial virus (RSV) is a major cause of respiratory tract infection in both infants and children during the RSV season. Conditions predisposing to disease are chronic lung disease of prematurity, significant congenital heart disease and preterm birth <35 weeks’ gestational age (wGA).

Palivizumab is a monoclonal antibody that has been approved for use in high-risk children and is administered at the start of and during the RSV season. Until now, use was limited to specifically defined population groups.

The Respiratory Events Among Preterm Infants Outcomes and Risk Tracking (REPORT) study conducted by Ambrose et al.[1] addressed the excluded subgroups, which were previously perceived to be low-risk, namely the infants born 32 - 35 wGA with no associated risk factors, but as a function of chronological and gestational age. Despite 73% of all US infants being born at 32 - 35 wGA, this is the first prospective study to be conducted in the US investigating the burden of RSV disease in this population, although four similar studies have already been conducted elsewhere in the world.

Rates of RSV hospitalisation particularly from RSV emergency department (ED) visits, ICU and mechanical ventilation were similar throughout all five studies. Previously identified risk factors such as daycare attendance, preschool siblings and tobacco smoke exposure

were observed. However, lower rates of RSV hospitalisations were noted among multiple birth infants.

REPORT is one of the few studies looking at RSV hospitalisation during the entire RSV season. It found that the risk of RSV ED visits and hospitalisation was highest at 6 months chronological age in the 32 - 34 and 35 wGA infants with risk factors. The incidence of RSV disease severity did not differ among the <3 month and 3 - 6 month age groups in 32 - 34 wGA infants. However, RSV illness in this group was more severe.

In light of the findings in this study, the question that needs to be answered is whether RSV prophylaxis should be provided to this population beyond 90 days of age. The elevated RSV hospitalisation rates among infants 35 wGA with the 2012 American Academy of Paediatrics environmental risk factors at <3 months and 3 - < 6 months also requires review.

Denise C Parris Fellow, Paediatric Pulmonology University of Pretoria and Steve Biko Academic Hospital

Reference

1. Ambrose CS, Anderson EJ, Simões AF, et al. Respiratory syncytial virus disease in preterm infants in the U.S. born at 32–35 weeks gestation not receiving immunoprophylaxis. Pediatr Infect Dis J 2014;33(6):576-582. [http://dx.doi. org/10.1097/INF.0000000000000219]

S Afr Respir J 2016;22(1):27. DOI:10.7196/SARJ.2016.v22i1.63

Early inhaled budesonide for the prevention of bronchopulmonary dysplasia Although with modern methods of prevention and management of prematurity, bronchopulmonary dysplasia (BPD) is a less common consequence of being born prematurely, the ‘new’ form of BPD is still a debilitating and difficult-to-manage problem. Attempts to prevent it, or manage it early, are quite disappointing. Systemic

steroids given to extremely premature infants with respiratory distress do reduce the incidence of BPD but have important side-effects including compromising brain development. Use of inhaled steroids has documented inconsistent findings.

Bassler et al.[1] randomly assigned 863 premature infants to early (within 24 hours of birth) inhaled budesonide or placebo and

SARJ VOL. 22 NO. 1 2016

27

BREATH-TAKING NEWS continued this therapy until they were no longer on oxygen or reached a postnatal age of 32 weeks. BPD was assessed at 36 weeks by means of standardised oxygen-saturation monitoring.

The incidence of BPD was 27.8% in the budesonide group versus 38.0% in the placebo group (relative risk (RR) 0.86, 95% confidence interval (CI) 0.75 - 1.00, p=0.05). Death occurred in 16.9% and 13.6% of the patients, respectively (RR 1.24, 95% CI 0.91 - 1.69, p=0.17).

The authors conclude that among extremely premature infants, the incidence of BPD was lower among those who received early inhaled budesonide than in those receiving placebo. However, they go on to point out that the advantage may come at the expense of increased mortality.

While seeking a solution to the devastating effects of BPD on children, it seems unlikely that this study will motivate the widespread

28 SARJ VOL. 22 NO. 1 2016

use of inhaled steroids. However, it does suggest that inhaled steroids may have a role. It would be interesting to repeat this study using newer steroids such as fluticasone, mometasone or ciclesonide, which theoretically have lower systemic effects.

Robin J Green Paediatric Pulmonologist, University of Pretoria and Steve Biko Academic Hospital

Reference

1. Bassler D, Plavka R, Shinwell ES, et al. Early inhaled budesonide for the prevention of bronchopulmonary dysplasia. N Eng J Med 2015;373:1497-1506. [http://dx.doi. org/10.1056/NEJMoa1501917]

S Afr Respir J 2016;22(1):27-28. DOI:10.7196/SARJ.2016.v22i1.64

WHO’S WHO

Who’s who in pulmonology The University of Pretoria, Division of Paediatric Pulmonology, has an active programme. There are two consultants, Prof. André van Niekerk and Dr Ashley Jeevarathnum. Prof. Robin Green leads the team. In addition we have two current Fellows, Dr Denise Parris and Dr Xandre Dearden. Ms Odette Coetzee is our Lung Function Technologist.

Dr Denise Parris

Dr Xandre Dearden

Prof. André van Niekerk

Prof. André van Niekerk

André is a Paediatrician and Paediatric Pulmonologist. He works as session consultant in the Department of Paediatrics and Child Health at the University of Pretoria and in private practice at the NetCare Clinton and NetCare Alberlito Hospitals. The NetCare Clinton practice is a Health Professions Council of South Africa (HPCSA) accredited sub-specialist training unit that offers satellite

Dr Ashley Jeevarathnum

Dr Ashley Jeevarathnum

Ashley has just qualified as a Paediatric Pulmonologist at the University of Pretoria.

In those 2 years he achieved four degrees (MMed(Paed), Diploma Allergology, Certificate in Paediatric Pulmonology and HERMES degree in Paediatric Pulmonology from the European Respiratory Society). Moreover, Ashley achieved the Eugene Weinberg Medal

Ms Odette Coetzee

training in Paediatric Pulmonology and Paediatric Critical Care medicine to fellows from the University of Pretoria. He received an Honorary Professorship from the University of Pretoria. André is the Vice Chair of the Allergy Society of South Africa (ALLSA).

He is also the ALLSA liaison for the Primary Immunodeficiency Working Group (PIDDSA) of ALLSA, the executive officer of the Allergy Foundation of South Africa (AFSA) and a member of the South African Thoracic Society (SATS) and the African Society for Immunodeficiencies (ASID).

for Excellence in the Diploma Allergology, a Travelling Fellowship to attend and present his research at the 2015 American Academy of Allergy, Asthma and Immunology Congress. He was awarded the best paper presentation at the 2014 ALLSA Congress. Ashley has now published a number of papers and given many lectures. He has been appointed as a consultant in the Department of Paediatrics and Child Health at the University of Pretoria and Steve Biko Academic Hospital.

SARJ VOL. 22 NO. 1 2016

29

PRODUCT NEWS

The South African Respiratory Journal PO Box 13725 Mowbray 7705 Any correspondence to the Editor should be sent to the same address or via email to: sarj@iafrica.com The website of The South African Thoracic Society can be found at www.pulmonology.co.za

PRODUCT NEWS Rivaroxaban reduces length of hospital stay in patients with symptomatic venous thromboembolism The phase III EINSTEIN deep vein thrombosis (DVT) and EINSTEIN pulmonary embolism (PE) trials demonstrated the potential of oral rivaroxaban (Xarelto, Bayer) – 15 mg twice daily for 21 days, followed by 20 mg once daily – for the treatment of venous thromboembolism (VTE), a term that embraces DVT and PE. A subsequent study by van Bellen et al.,[1] published in Current Medical Research and Opinion in 2014, was undertaken to assess the length of initial hospitalisation in patients presenting with either symptomatic DVT or PE using hospitalisation records from these trials.

The authors found that overall 52% of EINSTEIN DVT patients and 90% of EINSTEIN PE patients were admitted to hospital. The proportion of hospitalised DVT patients with a length of stay 5 days or fewer, receiving rivaroxaban, was 54% compared with 31% for those receiving enoxaparin/vitamin K antagonist (VKA), the current standard of care for the treatment of patients with symptomatic DVT and PE. For patients with PE, the corresponding values were 45% and 33%. Stays of 6 - 10 days were observed in 29% of rivaroxaban-treated patients compared with 45% for enoxaparin/VKA-treated patients for DVT. For patients with PE, these values were 39% and 46% in the rivaroxaban and enoxaparin/ VKA groups, respectively. Overall, length of stay was significantly shorter in the rivaroxaban group, compared with the enoxaparin/VKA group across all analyses performed (p<0.0001).

VTE is associated with significant morbidity and mortality and therefore carries a considerable healthcare burden. Rivaroxaban is as effective as enoxaparin/VKA for the treatment of acute symptomatic DVT or PE, with the additional benefit of significantly reducing the period of hospitalisation in patients being treated for an initial DVT or PE. ‘Coupled with improved patient treatment satisfaction and no requirement for routine monitoring or dose adjustment, this presents strong advantages for treating patients with VTE with rivaroxaban,’ the authors wrote.

They concluded that a single-drug regimen with rivaroxaban may reduce the burden on healthcare systems and patients by providing effective and well-tolerated treatment. ‘The convenience of a single-drug approach with oral rivaroxaban has the potential to allow discharge based on a patient’s clinical condition and to facilitate the transition from in-hospital to outpatient care. […] However, assessment of patient risk is still warranted to identify candidates who can safely receive outpatient treatment, and patient monitoring is essential to ensure adherence to the specified dosing regimen.’

Reference

1. van Bellen B, Bamber L, Correa de Carvalho F, et al. Reduction in the length of stay with rivaroxaban as a single-drug regimen for the treatment of deep vein thrombosis and pulmonary embolism. Curr Med Res Opin 2014; 30(5):829-837. [http://dx.doi.org/10.1185/03007995.2013.879439]

For full prescribing information, refer to the package insert approved by the Medicines Regulatory Authority (MCC).

PHARMACOLOGICAL CLASSIFICATION: A.8.2 Anticoagulants.

S4 XARELTO® 10. Reg. No.: 42/8.2/1046. Each film-coated tablet contains rivaroxaban 10 mg.

INDICATION: Prevention of VTE in patients undergoing major orthopaedic surgery of the lower limbs.

S4 XARELTO® 15 and XARELTO® 20. Reg. No.: 46/8.2/0111 and 46/8.2/0112. Each film-coated tablet contains rivaroxaban 15 mg or 20 mg, respectively.

INDICATIONS: Prevention of stroke and systemic embolism in patients with non-valvular atrial fibrillation; Treatment of DVT and for the prevention of recurrent DVT and PE; Treatment of PE and for the prevention of recurrent PE and DVT. Bayer (Pty) Ltd, Co. Reg. No.: 1968/011192/07, 27 Wrench Road, Isando, 1609. Tel: 011 921 5044 Fax: 011 921 5041.

L.ZA.GM.06.2014.1007

© Bayer HealthCare Pharmaceuticals June 2014

SARJ VOL. 22 NO. 1 2016

31

EVENTS

32 SARJ VOL. 22 NO. 1 2016

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 acknowledges with thanks the invaluable sponsorship of the following companies: Aspen GSK Division Bayer Healthcare