Symposium : Pulmonology Part II

Indian J Pediatr 1996; 63 : 163-169

Bronchoalveolar Lavage in Pediatrics Soumya Swaminathan, D. Vijayasekaran and N. Somu

Tuberculosis Research Centre, ICMR, Spurtank Road, Chetput, Madras and Department of Pediatric Pulmonology, Institute for Child Health and Hospitalfor Children; Egmore, Madras Abstract. Bronchoalveolar lavage is a relatively new technique that is used to study the local cellular, biochemical and immunological changes occuring in the lower respiratory tract. The procedure involves instilling a fixed volume of saline into a lung segment after the flexible fibreoptic bronchoscope is wedged into a distal bronchus. The saline is aspirated back and can be used for microbiological and other studies. Recently, attempts have been made to standardise the procedure in children and obtain data on BAL cellular profile in healthy children. The main indications for BAL are diagnostic, particularly to diagnose unusual infections in immunocompromised chlidren. It is also helpful in the diagnosis of a number on non infectious conditions, based on the cellular profile and other constituents. With the availability of new techniques like flow cytometry, analysis of lymphoctye and other cell subsets has become possible leading to a better understanding of the immunopathgenesis of various respiratory diseases. (Indian J Pediatr 1996; 63 : 163-169)

Key words : Bronchoalveolar lavage; Pediatric. B ronchoalveolar lavage (BAL) is a tech- bronchoscopes in recent years, BAL has nique that samples the alveolar lining fluid been increasingly performed in the pediati.e. the cellular and biochemical compo- ric population. However, the technique so nents of the lower respiratory tract. It is far, has not been standardised and it is benow accepted as an important diagnostic ing done mainly to retrieve infectious orand research tool in various diseases of the ganisms, particularly in immunocomprolung in adults. It has been shown that the mised children with pulmonary infections. cellular profile of the BAL fluid closely re- Some recent studies have attempted to adsembles the changes in the surrounding dress the issue of standardisation of the tissues of the lung and it therefore serves 9 Table 1. Diseases in Which BAL is Diagnostic as an easier and relatively noninvasive alternative to a lung biopsy. This procedure Infectious diseases Non-Infectious dis. is increasingly used both to diagnose and Pneumocystis carinii Alveolar proteinosis assess treatment response as well as to un- Toxoplasma gondii Eosinophilic granuloma derstand the pathogenesis, particularly lo- Strongyloides Tropicalpulrn, eosinophilia cal immunological mechanisms, in various Legionella Cancer pulmonary diseases 1'2. Histoplasrna With the availability of pediatric flexible Mycobacterium tuberculosis Mycoplasrna pneumoniae Reprint requests: Dr. Soumya Swaminathan, Influenza virus Assistant Director, Tuberculosis Research Respiratory syncytial virus Centre, Chetput, Madras 600 031.

164

THE INDIAN JOURNAL OF PEDIATRICS

Table 2. BAL Helpful in Diagnosis

Infectious Dis. Herpes simplex Cytomegalovirus Bacteria Aspergillus Candida Cryptococus Atypical Mycobact.

Non Infectious Dis. Pulmonary hemosiderosis Eosinophilic pneumonia Beryliosis Hypersensitivity pneum. Asbestosis Silicosis Sarcoidosis

1996; Vol. 63. No. 2

brosis and alveolar microlithiasis b u t its value in these conditions is yet to be proved. BAL IN INFECTIOUS DISEASES

BAL is useful in the diagnosis of infections occurring-in the i m m u n o c o m p r o m i s e d child, as these are atypical and maybe due to unusual organisms. Recovery of Pneumocystis, Toxoplasma or M. tuberculosis procedure in children as well as to define from BAL fluid in an i m m u n o comprothe normal cytology of the lower respirato- mised child is diagnostic of disease due to ry tract. This review will discuss some of these organisms. In one study, six out of 7 the indications for BAL, the suggested pro- i m m u n o s u p p r e s s e d children had a posicedure, possible complications and the ref- tive microbial result on BAL (Pneumocystis, Candida or CMV) 5. However, the diagerence data available. nosis of bacterial infections is more complicated as there is a possibility of contaminaINDICATIONS tion. It has been suggested that a semi(a) Diagnostic : BAL is useful as a diagnos- quantitative method be used and a bacteritic procedure in diagnosing infections both al colony ,qount of more than 10s cfu/ml in in immunocompromised and nonimmuno- BAL fluid be used as a cut-off to differenticompromised children 3 (Table 1). It is also ate contamination from disease. 6 In addiuseful in assessing the nature and severity tion, a combination of gram stain of BAL of alveolar inflammation and in following fluid and semiquantitative aerobic culture the course of diseases. In the infectious dis- could differentiate active pneumonia from eases listed here, retrieval of organisms resolving pneumonia, bronchitis or upper from BAL implies the presence of disease. airway colonisation 7. It has now been established that gastric In the non infectious conditions, the cellular profile of BAL fluid is fairly diagnostic lavage is better than BAL for the diagnosis of these diseases. In table 2 are listed con- of pulmonary tuberculosis in chiktren 8,9. In ditions in which BAL findings are not di- a study where 50 children with suspected agnostic per se but are suggestive of partic- pulmonary tuberculosis had both BAL and GL performed, Somu et al found that M. ular diseases and therefore maybe helpful. tuberculosis could be isolated from BAL in (b) Therapeutic : BAL is useful in the treat- only 12% of suspected cases whereas a ment of a rare condition called pulmonary simple gastric lavage (GL) was positive in alveolar proteinosis where regular lavages 32% ~. Abadco & Steiner isolated M. tuberhelp to clear the alveoli of proteinacious culosis from BAL in 10% of their cases material and improve lung function and while the yield from 3 consecutive early oxygenation. BAL has been suggested as a morning gastric lavages was 50%. The reatherapeutic procedure in asthma, cystic fi- sons for the better yield from GL could be

1996; Vol. 63. No. 2

THE INDIAN JOURNAL OF PEDIATRICS

165

that it is a concentrated specimen of swal- wal! suction pressure of 100-150 mm Hg or lowed respiratory secretions whereas BAL by band suction. There is no special time is an one-time and diluted specimen. These -allotted for lavage fluid to dwell in the studies underline the fact that the diagnos- lung but if the patient takes a breath or two tic utility of BAL needs to be scientifically before the fluid is withdrawn, more mixing evaluated for different diseases before it is occurs and a better specimen is obtained. accepted as being superior to conventional Usually, lavage fluid is frothy in the first few aliquots recovered, denoting that surtechniques. factant is present. From the first bolus inPROCEDURE fused, only a minimal amount of fluid is recovered, as most of it will remain in the Till recently, there had been no uniform lung segment. Thereafter, the return of recommendation for the performance of fluid from subsequent aliquots is about 60BAL in children, as was made in adults. 70% of that instilled. Overall, when The procedure has now been studied and lavaging normal airways about 50-60% restandardised by 2 groups 1~ It is recom- turn is expected, with airway obstruction mended that a 3.5 mm flexible fibreoptic the return drops significantly to the 10-40% bronchoscope be used for children less range. than 8 years and the larger 4.8 mm one for older children. The procedure is usually Monitoring performed under sedation with diazepam or meperidine along with atropine 0.01 It is advisable to monitor the heart rate, mg/kg. Atropine helps to prevent vagal respirat3ry rate and oxygen saturation by indticed bradycardia and reduces airway pulse oximetry throughout the procedure. secretions. The older child also needs to be An intravenous line should be secured bereassured and explained the procedure to fore starting and resuscitation equipment avoid anxiety. The bronchoscope is kept handy. Supplemental oxygen should inserted transnasally after local be given via nasal prongs or intermittently anaesthesia of the upper and lower airway via the suction channel of the with ldgnocaine (2-4%, maximum dose bronchoscope to maintain oxygen satura4mg/kg). The scope is then wedged into a tion over 90%. Desaturation during the segmental or subsegmental bronchus and lavage is common particularly in patients BAL performed. In localised disease, BAL with severe airway disease. is performed in the involved lung segment whereas in diffuse lung diseases, the right Complications middle lobe or lingula is chosen. When the Scope is wedged into a 4th or 5th order BAL is a relatively safe procedure and the bronchus, it is estimated that 1.5 to 3% of complications reported have been few. The commonest is fever occurring a few hours the lung or 106 alveoli are sampled. Three aliquots of 1 m l / k g each of after the procedure and lasting upto 24 nonbacteriostatic normal saline warmed to hours-this is probably a pyrogenic reacbody temperature are instilled and aspi- tion. The fever usually responds to rated back into a specimen trap, using a antipyretics. Transient tachypnea or tran-

THE INDIAN JOURNAL OF PEDIATRICS

1996; Vol. 63. No. 2

sient alveolar infiltrates on the Chest Xray may also occur and patients should be carefully observed and given oxygen if necessary for a few hours after the procedure. In patients who have hyperreactive airways, bronchoscopy and BAL may aggravate their condition and produce acute bronchospasm. It is better to start such patients on an Aminophylline drip before the b r o n c h o s c o p y and continue it for a few hours afterwards. Transient desaturation during BAL occurs and m a y be particularly severe in patients with underlying diffuse lung disease. Oxygen saturation should be monitored and the procedure a b a n d o n e d if saturation falls consistently b e l o w 85% inspite of oxygen. Bleeding from the nose occurs sometimes because of the t r a u m a associated with passing the scope and usually responds to simple measures. Other rarer complications of BAL are pneumonia, pneumothorax, bleeding and dissemination of infection to other parts of the lung. Patients should be closely monitored after the procedure and appropriate action taken. In our o w n experience, tran-

sient fever and tachypnea were the only complications encountered and both responded to symptomatic treatment.

166

PROCESSING AND ANALYSING THE FLUID

The aliquots of BAL fluid can either all be pooled or the first aliquot can be analysed separately as it contains proportionately more bronchial airway material. The total amount of recovered fluid is measured and recovery rate calculated. Samples can then be sent for microbiological studies. Total cell count is p e r f o r m e d on' uncentrifuged BAL fluid using a Neubauer h a e m o c y t o m e t e r counting chamber and expressed as cells x 106 per 100 ml BAL fluid. The fluid is then filtered thr6ugh sterile gauze to remove mucus. Cell viability can be assessed using trypan blue dye solution-viable cells will not take up the dye. There are 3 methods which are used to perform the differential cell count : Millipore filtration, cytocentrifugation and flow cytometry. a. Millipore filter preparation : Most laboratories use 2 x 105 cells per filter and use

Table 3. Reference Values for Pediatric BAL in Healthy Children Ronchetti 14 (mean _+SD) Total cells (106/dl) Macrophages (%) Lymphocytes (%) Neutrophils (%) Eosinophils (%) Basophils CD4 (% total lyrnpho) CD8 (% total lympho) CD4/CD8 ratio

80 -+84 84_+8 10+_6 6-+6

Ratgen l~ (mean ~ SD)

Riedlern Median (25-75th percentile)

12 _+9.8 79 _+11 19 +- 11 2_+2.3

15.5 (7.5-25.8) 91 (84-94) 7.5 (4.7-12.8) 1.7 (0.6-3.5) o.2 (o-0.3)

33 _+12.8 57 +_13 0.68 + 0.44

27 (22-32) 45 (34-57) 0.6 (0.4-1.0)

I996; Vol. 63. No. 2

THE INDIANJOURNALOF PEDIATRICS

gravity drainage 12. Millipore filter preparations can be stained with a modified haematoxylin-eosin or with a PAP technique. The advantages of this technique are that cell retention by the filter is high so it provides an accurate representation of cell types. Also, the slides can be sto~ed indefinitely without significant loss of cell detail or staining. The disadvantages are that it is costlier and more time consuming and may under estimate the number of neutrophilC. b. CytocentrifuOe preparatiWnS DhUr} are different methgdc foj uaki~g catgcendrifege preparatiovs. I hey can be made bA Cpin/ninG 400-1000 ul (dependiVg on total cell c unt) of unconceftradet l~owlud BqL fdu%id.0Other ivvestiwators pre&fer tg perform the cAtGspiNs after centrifugation (500 g, i0 rains.) f BaL ftuad6 th} cellc are recucpended qn bufvezed mediem avd a volume of fluid containing 5-20 x 104 cells is placed in the cytocentrifuge. Cytocentrifugation speed (23-165 g) and time (4-10 min) vary considerably in different labs and may affect the results of the differential cell counts. Generally, a.speed of 1000-1500 rpm for 10 minutes is used. The cytospin smears are then stained with Wright-Giemsa stain for differential

167

cell counting. Special stains which can be used if indicated are Perl's stain for iron positive alveolar macrophages, GomoriGrocott stain for Pneumocystis and Aspergillus, Z / N stain for AFB, indirect immunofluorecence for Legionella and Pap stain for malignant cells. The advantages of this technique are that it is easy, smears can be rapidly stained, special stains can be used where required and cell structure is better observed. The selective loss of lymphocytes with this technique can be minimised by not adding serum to the medium. c. Flow cytometry : Analysis of lymphocyte subsets in BAL fluid can now be done using flow cytometry. This method relies on the use of monoclonal antibodies directed against antigens present on the cell surface of lymphocytes, monocytes and macrophages. Many different types of flow cytometer are available but the m i n i m u m requirement for most current applications is that the machine must be fitted with both forward angle and 90 degree scatter detectors and a minimum of 2 or 3 fluorescence detectors. For application with BAL samples, it is also important that the machine should have the capacity for logarithmic, as well as linear acquisition of

Table 4. Predominant Cellular Profile in different Diseases

Lymphocytosis

Neutrophilia

Eosinophilia

Tuberculosis Hypersens. pneum. Sarcoidosis Berryliosis Drugs : Methotrexate, gold salts, Dilantin, Propranolol, Amiodarone

Pulm. fibrosis Pyogenic infections Asbestosis Chr. bronchitis Smokers ILD ass. with collagen vasc. dis. Drugs : Bleomycin, Busulphan

Tropical pulm. eos. Eos. granuloma Tuberculosis (rare) Churg-Strauss syndr. Parasitic diseases Asthma Collagen vasc. dis. Drugs : Bleomycin, Cotrimoxazo|e, Penicilin, Salazopyrin, Nitrofurantion

THE INDIAN JOURNAL OF PEDIATRICS

1996; Vol. 63. No. 2

data and for nonrectangular 'gating'. BAL fluid is centrifuged, the cell pellet washed, resuspended at a suitable cell con-centration (e.g. 1 x 107 cells/ml) and then stained with monoclonal antibodies. The stained sample should be filtered prior to r u n n i n g it through the instrument to remove all cell clumps and mucus. It is also essential to selectively "gate" different Cell populations for separate analysis because different types of cells in BAL have different amounts of background auto-fluorescence. Using flow cytometry, it is possible to rapidly identify different cell populations and subsets and also to detect cells sim u l t a n e o u s l y expressing more than one surface marker. There is also a lot of research going on about the biochemical markers of various diseases and their usefulness in diagnosis and prognosis. However, quantitative expression of the non cellular components is complex and controversial because of variability in lavage return and the unavailability of satisfactory reference standards to control for the dilution of soluble components by the lavage liquid. They can be expressed either as u n i t s / m g albumin or u n i t s / m l lavage fluid. The use of urea or methylene blue as a marker for estimating the v o l u m e of ECF recovered by BAL remains controversial.

The values for total and differential cell count in BAL fluid obtained by the different groups in healthy children are shown in table 3. The cellular profile of BAL fluid helps in the differential diagnosis of various pulm o n a r y diseases (Table 4). By analysing the predominant cell type, one can narrow down the differential diagnosis of various respiratory diseases. Analysis of the different lymphocyte subsets also gives valuable information about the local immunological changes taking place. These changes are not always reflected in peripheral blood and that is w h y BAL is such an important tool. BAL fluid analysis has also been used to monitor the progress of diseases like sarcoidosis and interstitial lung diseases associated with collagen vascular disease and to predict prognosis. However, most of these studies have been conducted in adults 15. There have been only a few studies of BAL cellular profile in children with lung diseases, particularly AIDS 16 and tuberculosis 17. More research is required using this useful technique and the newer molecular biologic approaches in order to understand better the pathogenesis of various respiratory diseases of children.

168

CELLULAR DATA IN CHILDREN

Since BAL has only recently been applied in children as a research tool, there is very little data available on normal and diseased subjects. Total and differential cell counts have been' measured in "normal" children i.e. those without obvious respiratory disease by Ronchetti et a114 and more recently by Riedler et a111 and Ratjen et al. I~

REFERENCES

1. Reynolds HY. Bronchoalveolar lavage. Am Rev Respir Dis J987; 135 : 250-263. 2. Klech H and Pohl W. Technical recommendations and guidelines for BAL. Report of the European Society of Pneumology Task group on BAL Eur Resp J 1989; 2 : 561-585. 3 Goldstein RA, Rohatgi PK, Bergofsky EM et al. Clinical role of bronchoalveolar lavage in adults with pulmonary disease. Am Rev Respir Dis 1990; 42 : 481-486. 4. Sharma SK and Pande JN. BAL

1996; Vol. 63. No. 2

5.

6.

7.

8.

9.

10.

11.

Application in pulmonary diseases. Ind J Chest Dis & All Sci 1990; 32 : 157-176. Frankel LR, Smith DW and Lewiston NJ. Bror~choalveolar lavage for diagnosis of pneumonia in the immuno compromised child. Pediatrics 1988; 81 : 785-788. Kahn FW and Jones JM. Diagnosing bacterial respiratory infection by bronchoalveolar lavage. J Infect Dis 1987; 155 : 862-869. Thorpe JE, Baughman RP, Frame PT et al. Bronchoalveolar lavage for diagnosing acute bacterial pneumonia. J Infect Dis. 1987; 155 : 855-861. Somu N, Swaminathan S, Paramasivan CN et al. Value of bronchoalveolar lavage and gastric lavage in the diagnosis of pulmonary tuberculosis in children. Tubercle 1995; 76 : 295-299. Abadco DL and Steiner P. Gastric lavage is better than bronchoalveolar lavage for isolation of Mycobacterium tuberculosis in childhood pulmonary tuberculosis. Pediatr Infect Dis J 1992; 11 : 735-8. Ratjen F, Bredendiek M, Brendel Met al. Differential cytology of bronchoalveolar lavage fluid in normal children. Eur Resp J 1994; 7 : 1865-1870. Riedler J, Grigg J, Stone C et al. Bronchoalveolar lavage cellularity in

THE INDIAN JOURNAL OF PEDIATRICS

12.

13.

14. 15.

16.

17.

169

healthy children. A m ] Respir Crit Care Med 1995; 152; 163-168. Saltini C, Hance AJ, Ferrans VJ et al. Accurate quantification of cells recovered by bronchoalveolar lavage. Am Rev Respir Dis 1984; 130 : 650-658. Wilcox ML, Kervitsky A, Walters LC and King TE. Quantification of cells recovered by bronchoalveolar lavage. Comparison of cytocentrifuge preparations with the filter method. Am Rev Respir Dis 1988; 138 : 74-80. Ronchetti R, Villani A, Dotta A et al. Pediatric BAL : Reference values. Am Rev Respir Dis 1992; 145 (4) : A 556. Klech H and Hutter C. Clinical guidelines and indications for bronchoalveolar lavage (BAL): Report of the European Society of Pneumology Task Group on BAL. Eur Resp ] 1990; 3 : 937-974. Nowakowski M, Chan SP, Steiner P et al. Different distributions of lung and blood lymphocyte subsets in pediatric AIDS or tuberculosis. Ann Clin Lab Sci 1992; 22 : 377-384. Soumya Swaminathan, Somu N, Venkatesan Pet al. Cellular profile of bronchoalveolar lavage fluid in pulmonary tuberculosis. Arch Dis Child 1995; 73 : 182182.