- Email: [email protected]
Determining the optimal number of specimens to obtain with needle biopsy of the pleura
Vol. 96 (2002) 14 ^17
Determining the optimal number of specimens to obtain with needle biopsy of the pleura D. JIMEŁNEZ*, E. PEŁREZ-RODRIGUEZ*,G. DIAZ*, L. FOGUEw AND R.W. LIGHTz *Department of Pneumology and wPathological Anatomy, Hospital RamoŁny Cajal, Madrid, Spain and zPulmonary Division, Saint Thomas Hospital and Vanderbilt University, Nashville, Tennessee, U.S.A. Abstract The aim of this study was to de¢ne the number of pleural biopsy samples necessary for optimum diagnostic performance and determineto whatextentthey are complementary.Eighty-fourclosed pleuralbiopsies were performed in our department between June 1996 and January 1998 on 55 males and 29 females with an average age of 64?4716?7 years.The study of the pleural £uid included: pH, biochemical testing of pleura/serum (proteins, lactate dehydrogenase, glucose, cholesterol, triglycerides, albumin and adenosine deaminase), haemogram, cytology and microbiological testing (Gram-staining, aerobes, anaerobes and mycobacteriae cultures). The biopsies were performed using a Cope needle, with a total of ¢ve biopsies for each patient: four for pathological examination (taken numerically in the order in which they were performed: D1, D2, D3 and D4) and one for microbiological testing. In those cases in which the diagnosis was uncertain or e¡usion persisted, a thoracoscopy or thoracotomy was performed.There were no signi¢cantdi¡erencesin the diagnostic yield of each individual sample (D1, D2, D3 and D4), but there were di¡erences in the sum of the samples, depending onthe number of biopsies performed.This was true for totalgroup and the group with carcinomas, but notfor the group with tuberculosis. The increase in diagnostic yield with the number of biopsies was more remarkable in the carcinoma cases, where it increased by 35% when four biopsies were performed (54% with one biopsy versus 89% with four biopsies, Po0?002).In conclusion, the diagnostic yieldincreased withthe number of biopsy samplesinthe totalgroup and the group with malignancy, but not in the group with tuberculous e¡usions. The best diagnostic performance for malignant pathology was obtained with four samples. In pleural tuberculosis, the diagnostic yield did not increase with more biopsy samples.One high quality sample should be enough to obtain a diagnosis.r 2001Harcourt Publishers Ltd doi:10.1053/rmed.2001.1200, available online at http://www.idealibrary.com on
Keywords pleural biopsy; carcinoma; tuberculosis; diagnostic performance; pleural e¡usion.
INTRODUCTION The diagnostic yields of pleural £uid analysis and pleural biopsy are complementary. In general, the biopsy is more useful than the £uid analysis with pleural tuberculosis (1), while the opposite is true with malignant pleural disease (2). With needle biopsy of the pleura, some authors have also stated that the increase in diagnostic yield is proportional to the number of pleural samples obtained (3). However, no systematic analysis has been performed to assess the diagnostic yield from the individual samples, to assess whether or not the yield from di¡erent samples is complementary, and to assess whether the optimum number of biopsies differs in tuberculous pleuritis and malignant pleural e¡usions. The purpose of the present study was to systematically analyse the yield when obtaining one, Received19 April 2001, accepted in revised form16 August 2001and published online 14 November 2001. Correspondence should be addressed to: Dr D. JimeŁnez Castro, Servicio de Neumolog|¤ a, Hospital RamoŁn y Cajal, Ctra.Colmenar Km 9100, 28034 Madrid, Spain.
two, three or four specimens with needle biopsy of the pleura. We hypothesized that the additional specimens would be more useful with pleural malignancy than with pleural tuberculosis because pleural involvement is more patchy with pleural malignancy.
MATERIAL AND METHODS In this prospective study between June 1996 and January 1998, closed needle biopsy of the pleura was performed in 84 cases in the Pleural Unit of our department. During the same time period, 274 thoracentesis were performed. In all cases, a minimum of 10 ml of pleural £uid were recovered for analysis. Of the patients that underwent closed pleural biopsy, 55 (65%) were male and the average age of the patients was 64?4716?7 years. The reasons for performing closed pleural biopsy were: suspected malignant pleural e¡usions, suspected granulomatous diseases (tuberculosis, connective disorders and others) and unexplained exudate. At our
NEEDLE BIOPSYOF THE PLEURA
facility, closed pleural biopsy is performed immediately, at the same time as the ¢rst diagnostic thoracentesis, when there is clinical suspicion of malignancy or tuberculous pleural e¡usion, given its low rate of complications and the speed with which a diagnosis can be obtained. When the diagnosis was uncertain after thoracentesis or closed pleural biopsy and the e¡usion persisted, we referred the patient for thoracoscopy and/or thoracotomy. The routine study of the pleural £uid includes: pH, biochemical testing of pleura/serum (proteins, LDH, glucose, cholesterol, triglycerides, albumin and ADA), haemogram, cytology and microbiological testing (Gram, Ziehl, aerobic and anaerobic cultures, and a mycobacterial culture). The biopsies were performed using a Cope needle (4), with ¢ve pleural samples per patient: four for histological analyses numbered in the order in which they were taken (D1, D2, D3, D4) and one for microbiological analyses. The same needle was used for each patient and di¡erent areas of the pleura (12, 3, 6, 9 and 12 o’ clock) were sampled.The diameter of the histological samples ranged from1to 3 mm. During the analyses, morphology and immunohistochemistry were evaluated to determine the type of malignancy and the Ziehl ^Neelsen method was used for mycobacterial analyses. In all cases the pathologist read each slide independently without knowing the results of the other slides in the same patient. Statistical comparisons of continuous data were done with the Mann ^Whitney test and comparisons of categorical data were done with Fischer’s exact test. We accepted a Po0?05 as statistically signi¢cant.
RESULTS The ¢nal diagnoses in the 84 patients who underwent needle biopsy of the pleura were as follows: 35 pleural carcinoma (21 lung cancer, seven breast cancer, ¢ve unknown, one colon carcinoma, one melanoma) 16 tuberculosis and 33 other diagnosis [12 parapneumonic, nine paramalignant, four idiopathic, two congestive heart failure, one lymphoma, one chronic lymphoid leukaemia (LLC), one mesothelioma, one chylothorax, one traumatic and one reumatoid arthritis] (Table 1). There were two pneumothoraces related to the procedure, but none of them required chest drainage. Five thoracoscopies were performed before a ¢nal diagnosis could be made. For the 35 patients with carcinoma involving the pleura, the cytology was positive in 28 cases (80%) and the ¢rst biopsy sample (D1) was positive in 19 (54%). The yield of the cytology combined with the pleural biopsy was 91?4%. The histologic type more frequently founded was adenocarcinoma 22/35 (62?8%). For the 16 patients with tuberculous pleural e¡usions, the pleural £uid acid-fast bacilli (AFB) stain was positive
15
TABLE 1. Aetiology of pleural e¡usions Diagnosis Carcinoma Paramalignancy Tuberculosis Parapneumonia Empyema Congestive heart failure Lymphoma LLC Mesothelioma Chylothorax Traumatic Rheumatoid arthritis Idiopathic
n=84 35 9 16 6 6 2 1 1 1 1 1 1 4
LLC: chronic lymphoid leukaemia.
TABLE 2. Yield of individual biopsy samples
n D1 D2 D3 D4 D1+2 D1+2+3 D1+2+3+4
General cases
Carcinoma
Tuberculosis
84 35 (42%) 27 (32%) 36 (43%) 34 (40%) 41 (49%) 47 (56%) 49 (58%)
35 19 (54%) 17 (49%) 21 (60%) 21 (60%) 24 (69%) 27 (77%) 31 (89%)
16 13 (81%) 7 (43%) 11 (68%) 9 (56%) 13 (81%) 14 (87?5%) 14 (87?5%)
in one (6?25%), while the pleural £uid culture was positive in ¢ve (31?2%). In the pleural tissues the AFB-stain was positive in six (37?5%) while culture was positive in nine (56?2%). Granulomas were present in D1 in 13 patients (81%). Fifteen patients (93?7%) had either positive smears of cultures or granulomas on D1. When the ADA levels in pleural £uid were analysed and a cuto¡ level of 40 IU was used to diagnose tuberculosis,14 of the 16 patients (87%) with tuberculous pleuritis had a positive test. There were, however, two false-positives (3%); one patient had an empyema while the other had a lymphoma. When the individual pleural biopsy samples were analysed, in the cases overall, each biopsy provided a de¢nite diagnosis in about 40% (Table 2). The cases diagnosed with pleural biopsy included a lymphoma, a LLC, a mesothelioma and a rheumatoid arthritis. However, with each successive biopsy the yield increased when the results of the biopsies where combined and with all four biopsies considered together, the diagnostic yield was 58% (Tables 2 and 3). With malignancy, each of the biopsies was positive in about 55% of cases, and when all four biopsies were considered together, the diagnosis
16
RESPIRATORY MEDICINE
TABLE 3. Increase in diagnostic yield depending on the number of biopsy samples General cases (%)
Carcinoma (%)
Tuberculosis (%)
42 7 14 16
54 15 23 35
81 0 6?5 6?5
D1 D1+2 D1+2+3 D1+2+3+4
TABLE 4. Signi¢cance of the increase in diagnostic yield depending on the number of biopsy samples P
D1/D1+2
D1/D1+2+3
D1/D1+2+3+4
NS NS NS
NS Po0?03 NS
Po0?01 Po0?02 NS
General cases Carcinoma Tuberculosis
was established in 89%. Somewhat di¡erent results where obtained when the 16 patients with tuberculosis were considered. The diagnostic yield with each of the biopsies was more varied, ranging from 43 to 81%. Biopsies D2, D3 and D4 added very little to biopsy D1 (only one additional patient was diagnosed), but this was due in large part to the high diagnostic yield with D1. Table 4 shows the statistical signi¢cance of the increase in diagnostic yield with the number of pleural biopsy samples. Statistical signi¢cance is obtained in the diagnostic yield with increasing number of samples in the group overall and in the group with carcinoma, but not in the group with tuberculous pleuresy.
DISCUSSION It is often di⁄cult to establish the cause of a pleural e¡usion.Closed pleural biopsy using a needle is a simple procedure with a low morbidity and mortality rate. In most cases, it is possible to obtain an adequate pleural sample for testing.The pleural biopsy is complementary to study of the pleural £uid in establishing a diagnosis (5), particularly when e¡usion is caused by tuberculosis (6). In addition, the chances of obtaining a diagnosis are increased when the biopsy is repeated (7,8). In addition to standard pathology the sample of parietal pleura may be tested by other methods such as microbiology and immunohistochemistry which increase its diagnostic yield (7,9). Closed pleural biopsy has proven useful in pleural tuberculosis, with a diagnostic yield of 60 ^95%. In cases involving malignant pleural e¡usion, it ranges from 5 to 7% for mesotheliomas and 40 to 87% for other neoplasias (10 ^12). When a patient has a malignant pleural e¡usion,
cytology in general is more useful than needle biopsy of the pleura. This is illustrated in the study of Prakash and Reiman (2), who studied 281 patients with malignant disease and reported that the cytology was positive in 58% while the pleural biopsy was positive in 43%.The pleural biopsy was positive in only 20 of the119 patients with malignant disease and non-diagnostic cytology and thereby increased the overall yield by 7%. It should be noted that the yield with cytology depends on the histologic type. The yield is higher for adenocarcinomas (60%), than it is with small cell carcinoma (19%), or squamous cell carcinoma (6%) (13).The diagnostic yield with the combination of cytology and pleural biopsy in the present study (91%) is comparable to that reported by Bueno et al. (9) (79%), Prakash and Remain (2) (65%) and Salyer et al. (14) (90%). The false-negative results obtained in patients with malignant pleural e¡usions are usually due to the fact that the samples were taken from areas is not involved by the tumour (13,15) or that there are other explanations for the pleural e¡usion such as lymphatic obstructions, obstructive pneumonia, hypoproteinemia, respiratory infections or heart failure. Computer tomography (CT) scan-guided biopsies can increase the diagnostic yield in pleural malignancy if nodules are present (16). After a thoracentesis and a pleural biopsy no diagnosis is found in 10 ^25% of patients with malignant pleural e¡usions. A second thoracentesis and pleural biopsy can be performed when no diagnosis is obtained with the ¢rst one and there are reasons to suspect neoplasia and tuberculosis (9). If the e¡usion persists and no de¢nite diagnosis can be achieved, diagnostic thoracoscopy should be performed (17). On the other hand, as far as pleural tuberculosis is concerned, many authors stress the high sensitivity of the pleural biopsy for diagnosis (3, 15, 18, 19) and the added e¡ect of the culture (20). With both procedures, a global sensitivity of up to 87% is obtained (3) [60% for the culture, 80% for the biopsy and 87% for both; 94% in our experience (21)]. Recently, much interest has been focused on adenosine deaminase (ADA) for the diagnosis of pleural tuberculosis (22).Tuberculous pleural £uid contains higher levels of ADA than do most other exudates (6,23), but elevated pleural £uid ADA levels do occur with most empyemas (23), rheumatoid pleuresy (24), some lymphomas (23) and rarely with non-lymphoma malignancies (23). ADA has been recently separated into ADA1and ADA2. Patients with tuberculous pleuresy had predominantly ADA2 and the determination of ADA isoenzymes could overcome the problem of false-positive cases (21). Even after extensive diagnostic work-up of the pleural £uid the aetiology of a number of pleural e¡usions remains undetermined (25). In these cases, some authors stress the yield of diagnostic thoracoscopy, particularly when pleural malignancy is suspected (26). In view of our results, thoracoscopy is not necessary for the
NEEDLE BIOPSYOF THE PLEURA
diagnosis of most malignant pleural e¡usions (91% of diagnostic yield with cytology and pleural biopsy). In any case, in centres with a lower prevalence of adenocarcinomas, it can be justi¢ed in cases of suspected malignant pleural e¡usions. Our studies also suggest that there is a very limited role for thoracoscopy in patients with a suspected tuberculous pleural e¡usion. The diagnostic yield of pleural £uid studies and pleural biopsy was 96?2% in a recent paper (21) and the use of ADA, and specially ADA1/ADAp has shown an accuracy of 91?2% and 99?02% respectively (21). The goal of this paper was to determine the number of biopsy samples for optimum yield according to the type of pleural pathology. In pleural tuberculosis, 81% of the diagnoses could be obtained with a single good sample due to the disseminated nature of the disease, which a¡ects di¡erent areas more or less homogeneously. We could not ¢nd any signi¢cant improvement in the diagnostic yield by taking four specimens instead of one. On the other hand, in malignant pleural pathology, a single biopsy sample only provides a diagnosis 54% of the time. When four samples were taken, diagnoses were obtained 89% of the time because the involvement of the pleura is not homogeneous, and the results depend on where the biopsy sample is taken (13). One could argue that malignant pleural e¡usions are often mistaken for pleural tuberculosis and that one should always take at least four pleural biopsies. In a recent paper from our own group we found that only ¢ve patients out of 278 with pleural malignancies were suspected as having pleural tuberculosis (1?79%) before thoracentesis was performed (27). In conclusion, a single good biopsy sample is generally su⁄cient when the suspected cause of the pleural e¡usion is tuberculosis, but four samples should be obtained when the suspected cause is malignant or unknown.
REFERENCES 1. Ferrer J. Pleural tuberculosis. Eur Respir J 1997; 10: 942^947. 2. Prakash UBS, Reiman HM. Comparison of needle biopsy with cytologic analysis for the evaluation of pleural e¡usion. Analysis of 414 cases. Mayo Clin Proc 1985; 60: 158 ^164. 3. Kirsch CM, Kroe M, Azzi RL, Jensen WA, Kagawa FT, Wehner JH. The optimal number of pleural biopsy specimens for a diagnosis of tuberculous pleuresy.Chest 1997; 112: 702^706. 4. Cope C. New pleural biopsy needle.JAMA 1958; 167: 1107^1108. 5. Frist B, Kahan AV, Koss LG. Comparison of the diagnostic value of biopsies of the pleura and cytological evaluation of pleural £uids. Am J Clin Pathol 1979; 72: 48 ^51. 6. Valde¤s L, San Jose¤ E, AŁlvarez D, et al. Diagnosis of tuberculous pleuresy using the biologic parameters adenosine deaminase, lysozyme and interferon gamma.Chest 1993; 103: 458 ^ 465.
17
7. Scerbo J, Keltz H, Stone DJ. A prospective study of closed pleural biopsies.JAMA 1971; 218: 377^380. 8. Llorente JL, Zalaca|¤ n R, Galdiz JB, et al. Utilidad de la biopsia pleural con aguja de Cope: a propo¤sito de 106 biopsias consecutives. Med Clin (Barc) 1984; 82: 242^244. 9. Bueno CE, Clemente G, Castro BC, et al. Cytologic and bacteriologic analysis of £uid and pleural biopsy specimens with Cope’s needle: study of 414 patients. Arch Intern Med 1990; 150: 1190 ^1194. 10. Jarvi OH, Kunnas RJ, Laitio MT,Tirkko JES.The accuracy and signi¢cance of cytologic cancer diagnosis of pleural e¡usions. Acta Cytol 1972; 16: 152^157. 11. Grunze H.The comparative diagnostic accuracy, e⁄ciency and speci¢city of cytologic techniques used in the diagnosis of malignant neoplasm in serous e¡usions of the pleural and pericardial cavities. Acta Cytol 1964; 8: 150 ^164. 12. Deckker A, Bupp PA.Cytology of serous e¡usions. An investigation into the usefulness of cell blocks versus smears. Am J Clin Pathology 1978; 70: 855^ 860. 13. Canto¤ A, Arnau A, Galbis J, et al. El llamado derrame pleural maligno: una nueva revisio¤n sobre aspectos directos extra|¤ dos mediante la pleuroscopia diagno¤stica. Arch Bronconeumol 1996; 32: 453^ 458. 14. Salyer WR, Eggleston JC, ErozarYS.E⁄cacy of pleural needle biopsy and pleural £uid cytopathology in the diagnosis of malignant neoplasm involving the pleura.Chest 1975; 67: 536^539. 15. Light RW. Pleural Diseases. 3rd ed. Baltimore: Williams & Wilkins 1995; 95^116. 16. Beauchamp HD, Kundra NK, Arason R, Chong F, MacDonnell K. The role of closed pleural needle biopsy in the diagnosis of malignant mesothelioma of the pleura.Chest 1992; 102: 1110 ^1112. 17. Loddenkemper R.Thoracoscopy^ state of the art. Eur Respir J 1998; 11: 213^221. 18. Poe RH, Israel RH, Utell MJ, Hull WJ, Greemblatt DW, Kalley MC. Sensitivity, speci¢city and predictive values of closed pleural biopsy. Arch Intern Med 1984; 144: 325^328. 19. Sahn SA. The pleura (state of the art). Am Rev Respir Dis 1988; 138: 194 ^234. 20. Berger HW, Mejia E.Tuberculous pleuresy. Chest 1973; 63: 88 ^92. 21. Pe¤rez-Rodr|¤ guez E, Pe¤rez-Walton IJ, Sa¤nchez Herna¤ndez JJ, et al. ADA1/ADAp ratio in pleural tuberculosis. An excellent diagnostic parameter in pleural £uid. Respir Med 1999; 93: 816^ 821. 22. Light RW, Ferrer J. Diagnosis of tuberculous pleuritis. Arch Bronconeumol 1999; 35: 105^107. 23. Oca•a I, Mart|¤ nez-Va¤zquez JM, Segura RM, Ferna¤ndez-De-Sevilla T, Capdevila JA. Adenosine deaminase in pleural £uids: test for diagnosis of tuberculous pleural e¡usion.Chest 1983; 84: 51^53. 24. Petterson T, Ojala K, Weber TH. Adenosine deaminase in pleural £uids: test for diagnosis of tuberculous pleural e¡usions. Acta Med Scand 1984; 215: 299^304. 25. Storey DD, Dines DE, Coles DT. Pleural e¡usion: a diagnostic dilemma.JAMA 1976; 236: 2183^2186. 26. Emad A, Rezaian GR. Diagnostic value of closed percutaneous pleural biopsy vs pleuroscopy in suspected malignant pleural e¡usion or tuberculous pleuresy in a region with a high incidence of tuberculosis: a comparative, age-dependent study. Respir Med 1998; 92: 488 ^ 492. 27. Jime¤nez Castro D, D|¤ az Nuevo G, Pe¤rez-Rodr|¤ guez E. Clinical suspicion as the ¢rst step in the diagnosis of pleural e¡usions. Eur Respir J 1999; 514s.
Determining the optimal number of specimens to obtain with needle biopsy of the pleura D. JIMEŁNEZ*, E. PEŁREZ-RODRIGUEZ*,G. DIAZ*, L. FOGUEw AND R.W. LIGHTz *Department of Pneumology and wPathological Anatomy, Hospital RamoŁny Cajal, Madrid, Spain and zPulmonary Division, Saint Thomas Hospital and Vanderbilt University, Nashville, Tennessee, U.S.A. Abstract The aim of this study was to de¢ne the number of pleural biopsy samples necessary for optimum diagnostic performance and determineto whatextentthey are complementary.Eighty-fourclosed pleuralbiopsies were performed in our department between June 1996 and January 1998 on 55 males and 29 females with an average age of 64?4716?7 years.The study of the pleural £uid included: pH, biochemical testing of pleura/serum (proteins, lactate dehydrogenase, glucose, cholesterol, triglycerides, albumin and adenosine deaminase), haemogram, cytology and microbiological testing (Gram-staining, aerobes, anaerobes and mycobacteriae cultures). The biopsies were performed using a Cope needle, with a total of ¢ve biopsies for each patient: four for pathological examination (taken numerically in the order in which they were performed: D1, D2, D3 and D4) and one for microbiological testing. In those cases in which the diagnosis was uncertain or e¡usion persisted, a thoracoscopy or thoracotomy was performed.There were no signi¢cantdi¡erencesin the diagnostic yield of each individual sample (D1, D2, D3 and D4), but there were di¡erences in the sum of the samples, depending onthe number of biopsies performed.This was true for totalgroup and the group with carcinomas, but notfor the group with tuberculosis. The increase in diagnostic yield with the number of biopsies was more remarkable in the carcinoma cases, where it increased by 35% when four biopsies were performed (54% with one biopsy versus 89% with four biopsies, Po0?002).In conclusion, the diagnostic yieldincreased withthe number of biopsy samplesinthe totalgroup and the group with malignancy, but not in the group with tuberculous e¡usions. The best diagnostic performance for malignant pathology was obtained with four samples. In pleural tuberculosis, the diagnostic yield did not increase with more biopsy samples.One high quality sample should be enough to obtain a diagnosis.r 2001Harcourt Publishers Ltd doi:10.1053/rmed.2001.1200, available online at http://www.idealibrary.com on
Keywords pleural biopsy; carcinoma; tuberculosis; diagnostic performance; pleural e¡usion.
INTRODUCTION The diagnostic yields of pleural £uid analysis and pleural biopsy are complementary. In general, the biopsy is more useful than the £uid analysis with pleural tuberculosis (1), while the opposite is true with malignant pleural disease (2). With needle biopsy of the pleura, some authors have also stated that the increase in diagnostic yield is proportional to the number of pleural samples obtained (3). However, no systematic analysis has been performed to assess the diagnostic yield from the individual samples, to assess whether or not the yield from di¡erent samples is complementary, and to assess whether the optimum number of biopsies differs in tuberculous pleuritis and malignant pleural e¡usions. The purpose of the present study was to systematically analyse the yield when obtaining one, Received19 April 2001, accepted in revised form16 August 2001and published online 14 November 2001. Correspondence should be addressed to: Dr D. JimeŁnez Castro, Servicio de Neumolog|¤ a, Hospital RamoŁn y Cajal, Ctra.Colmenar Km 9100, 28034 Madrid, Spain.
two, three or four specimens with needle biopsy of the pleura. We hypothesized that the additional specimens would be more useful with pleural malignancy than with pleural tuberculosis because pleural involvement is more patchy with pleural malignancy.
MATERIAL AND METHODS In this prospective study between June 1996 and January 1998, closed needle biopsy of the pleura was performed in 84 cases in the Pleural Unit of our department. During the same time period, 274 thoracentesis were performed. In all cases, a minimum of 10 ml of pleural £uid were recovered for analysis. Of the patients that underwent closed pleural biopsy, 55 (65%) were male and the average age of the patients was 64?4716?7 years. The reasons for performing closed pleural biopsy were: suspected malignant pleural e¡usions, suspected granulomatous diseases (tuberculosis, connective disorders and others) and unexplained exudate. At our
NEEDLE BIOPSYOF THE PLEURA
facility, closed pleural biopsy is performed immediately, at the same time as the ¢rst diagnostic thoracentesis, when there is clinical suspicion of malignancy or tuberculous pleural e¡usion, given its low rate of complications and the speed with which a diagnosis can be obtained. When the diagnosis was uncertain after thoracentesis or closed pleural biopsy and the e¡usion persisted, we referred the patient for thoracoscopy and/or thoracotomy. The routine study of the pleural £uid includes: pH, biochemical testing of pleura/serum (proteins, LDH, glucose, cholesterol, triglycerides, albumin and ADA), haemogram, cytology and microbiological testing (Gram, Ziehl, aerobic and anaerobic cultures, and a mycobacterial culture). The biopsies were performed using a Cope needle (4), with ¢ve pleural samples per patient: four for histological analyses numbered in the order in which they were taken (D1, D2, D3, D4) and one for microbiological analyses. The same needle was used for each patient and di¡erent areas of the pleura (12, 3, 6, 9 and 12 o’ clock) were sampled.The diameter of the histological samples ranged from1to 3 mm. During the analyses, morphology and immunohistochemistry were evaluated to determine the type of malignancy and the Ziehl ^Neelsen method was used for mycobacterial analyses. In all cases the pathologist read each slide independently without knowing the results of the other slides in the same patient. Statistical comparisons of continuous data were done with the Mann ^Whitney test and comparisons of categorical data were done with Fischer’s exact test. We accepted a Po0?05 as statistically signi¢cant.
RESULTS The ¢nal diagnoses in the 84 patients who underwent needle biopsy of the pleura were as follows: 35 pleural carcinoma (21 lung cancer, seven breast cancer, ¢ve unknown, one colon carcinoma, one melanoma) 16 tuberculosis and 33 other diagnosis [12 parapneumonic, nine paramalignant, four idiopathic, two congestive heart failure, one lymphoma, one chronic lymphoid leukaemia (LLC), one mesothelioma, one chylothorax, one traumatic and one reumatoid arthritis] (Table 1). There were two pneumothoraces related to the procedure, but none of them required chest drainage. Five thoracoscopies were performed before a ¢nal diagnosis could be made. For the 35 patients with carcinoma involving the pleura, the cytology was positive in 28 cases (80%) and the ¢rst biopsy sample (D1) was positive in 19 (54%). The yield of the cytology combined with the pleural biopsy was 91?4%. The histologic type more frequently founded was adenocarcinoma 22/35 (62?8%). For the 16 patients with tuberculous pleural e¡usions, the pleural £uid acid-fast bacilli (AFB) stain was positive
15
TABLE 1. Aetiology of pleural e¡usions Diagnosis Carcinoma Paramalignancy Tuberculosis Parapneumonia Empyema Congestive heart failure Lymphoma LLC Mesothelioma Chylothorax Traumatic Rheumatoid arthritis Idiopathic
n=84 35 9 16 6 6 2 1 1 1 1 1 1 4
LLC: chronic lymphoid leukaemia.
TABLE 2. Yield of individual biopsy samples
n D1 D2 D3 D4 D1+2 D1+2+3 D1+2+3+4
General cases
Carcinoma
Tuberculosis
84 35 (42%) 27 (32%) 36 (43%) 34 (40%) 41 (49%) 47 (56%) 49 (58%)
35 19 (54%) 17 (49%) 21 (60%) 21 (60%) 24 (69%) 27 (77%) 31 (89%)
16 13 (81%) 7 (43%) 11 (68%) 9 (56%) 13 (81%) 14 (87?5%) 14 (87?5%)
in one (6?25%), while the pleural £uid culture was positive in ¢ve (31?2%). In the pleural tissues the AFB-stain was positive in six (37?5%) while culture was positive in nine (56?2%). Granulomas were present in D1 in 13 patients (81%). Fifteen patients (93?7%) had either positive smears of cultures or granulomas on D1. When the ADA levels in pleural £uid were analysed and a cuto¡ level of 40 IU was used to diagnose tuberculosis,14 of the 16 patients (87%) with tuberculous pleuritis had a positive test. There were, however, two false-positives (3%); one patient had an empyema while the other had a lymphoma. When the individual pleural biopsy samples were analysed, in the cases overall, each biopsy provided a de¢nite diagnosis in about 40% (Table 2). The cases diagnosed with pleural biopsy included a lymphoma, a LLC, a mesothelioma and a rheumatoid arthritis. However, with each successive biopsy the yield increased when the results of the biopsies where combined and with all four biopsies considered together, the diagnostic yield was 58% (Tables 2 and 3). With malignancy, each of the biopsies was positive in about 55% of cases, and when all four biopsies were considered together, the diagnosis
16
RESPIRATORY MEDICINE
TABLE 3. Increase in diagnostic yield depending on the number of biopsy samples General cases (%)
Carcinoma (%)
Tuberculosis (%)
42 7 14 16
54 15 23 35
81 0 6?5 6?5
D1 D1+2 D1+2+3 D1+2+3+4
TABLE 4. Signi¢cance of the increase in diagnostic yield depending on the number of biopsy samples P
D1/D1+2
D1/D1+2+3
D1/D1+2+3+4
NS NS NS
NS Po0?03 NS
Po0?01 Po0?02 NS
General cases Carcinoma Tuberculosis
was established in 89%. Somewhat di¡erent results where obtained when the 16 patients with tuberculosis were considered. The diagnostic yield with each of the biopsies was more varied, ranging from 43 to 81%. Biopsies D2, D3 and D4 added very little to biopsy D1 (only one additional patient was diagnosed), but this was due in large part to the high diagnostic yield with D1. Table 4 shows the statistical signi¢cance of the increase in diagnostic yield with the number of pleural biopsy samples. Statistical signi¢cance is obtained in the diagnostic yield with increasing number of samples in the group overall and in the group with carcinoma, but not in the group with tuberculous pleuresy.
DISCUSSION It is often di⁄cult to establish the cause of a pleural e¡usion.Closed pleural biopsy using a needle is a simple procedure with a low morbidity and mortality rate. In most cases, it is possible to obtain an adequate pleural sample for testing.The pleural biopsy is complementary to study of the pleural £uid in establishing a diagnosis (5), particularly when e¡usion is caused by tuberculosis (6). In addition, the chances of obtaining a diagnosis are increased when the biopsy is repeated (7,8). In addition to standard pathology the sample of parietal pleura may be tested by other methods such as microbiology and immunohistochemistry which increase its diagnostic yield (7,9). Closed pleural biopsy has proven useful in pleural tuberculosis, with a diagnostic yield of 60 ^95%. In cases involving malignant pleural e¡usion, it ranges from 5 to 7% for mesotheliomas and 40 to 87% for other neoplasias (10 ^12). When a patient has a malignant pleural e¡usion,
cytology in general is more useful than needle biopsy of the pleura. This is illustrated in the study of Prakash and Reiman (2), who studied 281 patients with malignant disease and reported that the cytology was positive in 58% while the pleural biopsy was positive in 43%.The pleural biopsy was positive in only 20 of the119 patients with malignant disease and non-diagnostic cytology and thereby increased the overall yield by 7%. It should be noted that the yield with cytology depends on the histologic type. The yield is higher for adenocarcinomas (60%), than it is with small cell carcinoma (19%), or squamous cell carcinoma (6%) (13).The diagnostic yield with the combination of cytology and pleural biopsy in the present study (91%) is comparable to that reported by Bueno et al. (9) (79%), Prakash and Remain (2) (65%) and Salyer et al. (14) (90%). The false-negative results obtained in patients with malignant pleural e¡usions are usually due to the fact that the samples were taken from areas is not involved by the tumour (13,15) or that there are other explanations for the pleural e¡usion such as lymphatic obstructions, obstructive pneumonia, hypoproteinemia, respiratory infections or heart failure. Computer tomography (CT) scan-guided biopsies can increase the diagnostic yield in pleural malignancy if nodules are present (16). After a thoracentesis and a pleural biopsy no diagnosis is found in 10 ^25% of patients with malignant pleural e¡usions. A second thoracentesis and pleural biopsy can be performed when no diagnosis is obtained with the ¢rst one and there are reasons to suspect neoplasia and tuberculosis (9). If the e¡usion persists and no de¢nite diagnosis can be achieved, diagnostic thoracoscopy should be performed (17). On the other hand, as far as pleural tuberculosis is concerned, many authors stress the high sensitivity of the pleural biopsy for diagnosis (3, 15, 18, 19) and the added e¡ect of the culture (20). With both procedures, a global sensitivity of up to 87% is obtained (3) [60% for the culture, 80% for the biopsy and 87% for both; 94% in our experience (21)]. Recently, much interest has been focused on adenosine deaminase (ADA) for the diagnosis of pleural tuberculosis (22).Tuberculous pleural £uid contains higher levels of ADA than do most other exudates (6,23), but elevated pleural £uid ADA levels do occur with most empyemas (23), rheumatoid pleuresy (24), some lymphomas (23) and rarely with non-lymphoma malignancies (23). ADA has been recently separated into ADA1and ADA2. Patients with tuberculous pleuresy had predominantly ADA2 and the determination of ADA isoenzymes could overcome the problem of false-positive cases (21). Even after extensive diagnostic work-up of the pleural £uid the aetiology of a number of pleural e¡usions remains undetermined (25). In these cases, some authors stress the yield of diagnostic thoracoscopy, particularly when pleural malignancy is suspected (26). In view of our results, thoracoscopy is not necessary for the
NEEDLE BIOPSYOF THE PLEURA
diagnosis of most malignant pleural e¡usions (91% of diagnostic yield with cytology and pleural biopsy). In any case, in centres with a lower prevalence of adenocarcinomas, it can be justi¢ed in cases of suspected malignant pleural e¡usions. Our studies also suggest that there is a very limited role for thoracoscopy in patients with a suspected tuberculous pleural e¡usion. The diagnostic yield of pleural £uid studies and pleural biopsy was 96?2% in a recent paper (21) and the use of ADA, and specially ADA1/ADAp has shown an accuracy of 91?2% and 99?02% respectively (21). The goal of this paper was to determine the number of biopsy samples for optimum yield according to the type of pleural pathology. In pleural tuberculosis, 81% of the diagnoses could be obtained with a single good sample due to the disseminated nature of the disease, which a¡ects di¡erent areas more or less homogeneously. We could not ¢nd any signi¢cant improvement in the diagnostic yield by taking four specimens instead of one. On the other hand, in malignant pleural pathology, a single biopsy sample only provides a diagnosis 54% of the time. When four samples were taken, diagnoses were obtained 89% of the time because the involvement of the pleura is not homogeneous, and the results depend on where the biopsy sample is taken (13). One could argue that malignant pleural e¡usions are often mistaken for pleural tuberculosis and that one should always take at least four pleural biopsies. In a recent paper from our own group we found that only ¢ve patients out of 278 with pleural malignancies were suspected as having pleural tuberculosis (1?79%) before thoracentesis was performed (27). In conclusion, a single good biopsy sample is generally su⁄cient when the suspected cause of the pleural e¡usion is tuberculosis, but four samples should be obtained when the suspected cause is malignant or unknown.
REFERENCES 1. Ferrer J. Pleural tuberculosis. Eur Respir J 1997; 10: 942^947. 2. Prakash UBS, Reiman HM. Comparison of needle biopsy with cytologic analysis for the evaluation of pleural e¡usion. Analysis of 414 cases. Mayo Clin Proc 1985; 60: 158 ^164. 3. Kirsch CM, Kroe M, Azzi RL, Jensen WA, Kagawa FT, Wehner JH. The optimal number of pleural biopsy specimens for a diagnosis of tuberculous pleuresy.Chest 1997; 112: 702^706. 4. Cope C. New pleural biopsy needle.JAMA 1958; 167: 1107^1108. 5. Frist B, Kahan AV, Koss LG. Comparison of the diagnostic value of biopsies of the pleura and cytological evaluation of pleural £uids. Am J Clin Pathol 1979; 72: 48 ^51. 6. Valde¤s L, San Jose¤ E, AŁlvarez D, et al. Diagnosis of tuberculous pleuresy using the biologic parameters adenosine deaminase, lysozyme and interferon gamma.Chest 1993; 103: 458 ^ 465.
17
7. Scerbo J, Keltz H, Stone DJ. A prospective study of closed pleural biopsies.JAMA 1971; 218: 377^380. 8. Llorente JL, Zalaca|¤ n R, Galdiz JB, et al. Utilidad de la biopsia pleural con aguja de Cope: a propo¤sito de 106 biopsias consecutives. Med Clin (Barc) 1984; 82: 242^244. 9. Bueno CE, Clemente G, Castro BC, et al. Cytologic and bacteriologic analysis of £uid and pleural biopsy specimens with Cope’s needle: study of 414 patients. Arch Intern Med 1990; 150: 1190 ^1194. 10. Jarvi OH, Kunnas RJ, Laitio MT,Tirkko JES.The accuracy and signi¢cance of cytologic cancer diagnosis of pleural e¡usions. Acta Cytol 1972; 16: 152^157. 11. Grunze H.The comparative diagnostic accuracy, e⁄ciency and speci¢city of cytologic techniques used in the diagnosis of malignant neoplasm in serous e¡usions of the pleural and pericardial cavities. Acta Cytol 1964; 8: 150 ^164. 12. Deckker A, Bupp PA.Cytology of serous e¡usions. An investigation into the usefulness of cell blocks versus smears. Am J Clin Pathology 1978; 70: 855^ 860. 13. Canto¤ A, Arnau A, Galbis J, et al. El llamado derrame pleural maligno: una nueva revisio¤n sobre aspectos directos extra|¤ dos mediante la pleuroscopia diagno¤stica. Arch Bronconeumol 1996; 32: 453^ 458. 14. Salyer WR, Eggleston JC, ErozarYS.E⁄cacy of pleural needle biopsy and pleural £uid cytopathology in the diagnosis of malignant neoplasm involving the pleura.Chest 1975; 67: 536^539. 15. Light RW. Pleural Diseases. 3rd ed. Baltimore: Williams & Wilkins 1995; 95^116. 16. Beauchamp HD, Kundra NK, Arason R, Chong F, MacDonnell K. The role of closed pleural needle biopsy in the diagnosis of malignant mesothelioma of the pleura.Chest 1992; 102: 1110 ^1112. 17. Loddenkemper R.Thoracoscopy^ state of the art. Eur Respir J 1998; 11: 213^221. 18. Poe RH, Israel RH, Utell MJ, Hull WJ, Greemblatt DW, Kalley MC. Sensitivity, speci¢city and predictive values of closed pleural biopsy. Arch Intern Med 1984; 144: 325^328. 19. Sahn SA. The pleura (state of the art). Am Rev Respir Dis 1988; 138: 194 ^234. 20. Berger HW, Mejia E.Tuberculous pleuresy. Chest 1973; 63: 88 ^92. 21. Pe¤rez-Rodr|¤ guez E, Pe¤rez-Walton IJ, Sa¤nchez Herna¤ndez JJ, et al. ADA1/ADAp ratio in pleural tuberculosis. An excellent diagnostic parameter in pleural £uid. Respir Med 1999; 93: 816^ 821. 22. Light RW, Ferrer J. Diagnosis of tuberculous pleuritis. Arch Bronconeumol 1999; 35: 105^107. 23. Oca•a I, Mart|¤ nez-Va¤zquez JM, Segura RM, Ferna¤ndez-De-Sevilla T, Capdevila JA. Adenosine deaminase in pleural £uids: test for diagnosis of tuberculous pleural e¡usion.Chest 1983; 84: 51^53. 24. Petterson T, Ojala K, Weber TH. Adenosine deaminase in pleural £uids: test for diagnosis of tuberculous pleural e¡usions. Acta Med Scand 1984; 215: 299^304. 25. Storey DD, Dines DE, Coles DT. Pleural e¡usion: a diagnostic dilemma.JAMA 1976; 236: 2183^2186. 26. Emad A, Rezaian GR. Diagnostic value of closed percutaneous pleural biopsy vs pleuroscopy in suspected malignant pleural e¡usion or tuberculous pleuresy in a region with a high incidence of tuberculosis: a comparative, age-dependent study. Respir Med 1998; 92: 488 ^ 492. 27. Jime¤nez Castro D, D|¤ az Nuevo G, Pe¤rez-Rodr|¤ guez E. Clinical suspicion as the ¢rst step in the diagnosis of pleural e¡usions. Eur Respir J 1999; 514s.