Journal Mobile Options
Table of Contents
Vol. 67, No. 1, 2000
Issue release date: January–February 2000
Respiration 2000;67:18–23
(DOI:10.1159/000029457)

Light’s Criteria Revisited: Consistency and Comparison with New Proposed Alternative Criteria for Separating Pleural Transudates from Exudates

Romero S. · Martinez A. · Hernandez L. · Fernandez C. · Espasa A. · Candela A. · Martin C.
Servicios de Neumologia y Laboratorio, Hospital General Universitario, Alicante, España
email Corresponding Author

Abstract

Objectives: The first objective was to assess the diagnostic value of new biochemical criteria proposed to discriminate pleural transudates from exudates and to compare their efficiency with those of Light’s criteria. The second objective of the study was to assess the interstudy variability of the parameters repeatedly determinated in two different groups of patients with pleural effusion. Patients and Methods: We recorded clinical characteristics and final diagnoses and measured pleural fluid (PF) and serum levels of protein, LDH, cholesterol and cholinesterase of 243 patients with pleural effusion. Results: Sixty-one (25%) pleural effusions were transudates and 182 were exudates. The sensitivity (99%) and accuracy (96%) of Light’s criteria were higher than those of the other criteria tested, although the differences with those of the PF LDH-cholesterol combination (96 and 93%) did not show statistical significance. Pleural LDH concentration was the criterion with the highest specificity (95%), being significantly higher (p < 0.05) than that of Light’s criteria. The sensitivity, specificity and accuracy of most criteria tested did not vary when compared with those obtained in a study performed 5 years previously. Conclusions: Light’s criteria remain the criteria of choice for segregating exudates from transudates. Based on cost-efficiency reasons, the PF LDH-cholesterol combination appears as an alternative. Because both sets of criteria misdiagnose a substantial percentage of transudates, exceptions based on good clinical judgment and the complementary use of a more specific criterion, as the PF concentration of LDH, must be considered.

Copyright © 2000 S. Karger AG, Basel


 Outline


 goto top of outline Key Words

  • Pleural diseases
  • Pleural effusion
  • Transudates
  • Exudates
  • Cholinesterase
  • Cholesterol
  • Lactate dehydrogenase

 goto top of outline Abstract

Objectives: The first objective was to assess the diagnostic value of new biochemical criteria proposed to discriminate pleural transudates from exudates and to compare their efficiency with those of Light’s criteria. The second objective of the study was to assess the interstudy variability of the parameters repeatedly determinated in two different groups of patients with pleural effusion. Patients and Methods: We recorded clinical characteristics and final diagnoses and measured pleural fluid (PF) and serum levels of protein, LDH, cholesterol and cholinesterase of 243 patients with pleural effusion. Results: Sixty-one (25%) pleural effusions were transudates and 182 were exudates. The sensitivity (99%) and accuracy (96%) of Light’s criteria were higher than those of the other criteria tested, although the differences with those of the PF LDH-cholesterol combination (96 and 93%) did not show statistical significance. Pleural LDH concentration was the criterion with the highest specificity (95%), being significantly higher (p < 0.05) than that of Light’s criteria. The sensitivity, specificity and accuracy of most criteria tested did not vary when compared with those obtained in a study performed 5 years previously. Conclusions: Light’s criteria remain the criteria of choice for segregating exudates from transudates. Based on cost-efficiency reasons, the PF LDH-cholesterol combination appears as an alternative. Because both sets of criteria misdiagnose a substantial percentage of transudates, exceptions based on good clinical judgment and the complementary use of a more specific criterion, as the PF concentration of LDH, must be considered.

Copyright © 2000 S. Karger AG, Basel


goto top of outline introduction

Light’s criteria [1], the universally accepted gold standard to discriminate transudates from exudates, correctly identify almost 100% of pleural exudates but misdiagnose 10–30% of transudates [2, 3, 4, 5, 6]. Recently proposed alternative criteria, as pleural fluid (PF) cholesterol [2, 3] and serum-PF albumin gradient [4], although being more sensitive for transudates than Light’s criteria, can confound a substantial proportion of exudates [5, 6]. Because the number of exudates that require a diagnostic study exceeds that of transudates [6], using these new criteria the number of properly managed pleural effusions can decrease.

Recently, the cholinesterase ratio between PF and serum has been proposed as a new alternative criterion, after showing a correct identification of 99% of the transudates and 98% of the exudates [7]. On the other hand, almost simultaneously, a different group of investigators has found an accuracy of 99% combining the determination of pleural LDH and cholesterol concentrations [8].

The first objective of the present study was to assess the diagnostic value of these new criteria and to compare their efficiency with those of Light’s criteria. Because recently a similar study [6], albeit without including cholinesterase, has been carried out at our hospital in a different group of patients, the second objective of this study was to assess the interstudy variability of the parameters repeatedly determinated.

 

goto top of outline patients and methods

From November 1991 to September 1995, 514 patients were prospectively studied following norms previously communicated that did not include cholinesterase [6]. Those patients in whom a simultaneous determination of pleural and serum cholinesterase concentrations had been randomly performed were included in this study.

The following studies were performed on all PF samples: glucose, protein, lactate dehydrogenase (LDH), cholesterol, cholinesterase, cell count, differential cell count, bacterial and fungal culture, acid-fast bacillus smear and culture and cytology. Only the results of the first thoracentesis were considered. Simultaneously to the thoracentesis, a sample of serum was obtained to measure glucose, protein, LDH, cholesterol and cholinesterase levels. Further studies were done at the discretion of the primary physician.

Biochemical parameters were determined using a selective discrete multichannel analyzer (Hitachi 747). Total protein concentrations (g/dl) were measured using the biuret method. The LDH level (IU/l) was measured using a kinetic UV optimized standard method conforming to the recommendations of the ‘Deutsche Gesellschaft für Klinische Chemie’; the upper normal limit (serum) is defined at 460 IU/l. Cholesterol levels (mg/dl) were measured using the enzymatic colorimetric method (cholesterol oxidase-phenol-4 aminophenazone, high performance). Cholinesterase activity was measured by a colorimetric method using as substrate butyrylthiocholine (commercial kit from Boehringer Mannheim, Meylan, France), normal activity (serum) ranged from 5,300 to 12,900 U/l.

Effusions were considered malignant if one of the following criteria was met: demonstration of malignant cells at cytologic examination or in a biopsy specimen; or histologically proven primary malignancy with exclusion of any other cause known to be associated with pleural effusion. A pleural effusion was considered to be parapneumonic when there was an acute febrile illness with purulent sputum and pulmonary infiltrate in the absence of malignancy or diseases causing transudates. Tuberculous pleurisy was diagnosed with positive culture for Mycobacterium tuberculosis or a pleural biopsy specimen showing typical epithelioid cell granuloma. A diagnosis of pulmonary embolus or infarction was made when there was a strong clinical suspicion and a high-probability perfusion scan or abnormal angiogram.

Congestive heart failure (CHF) was determined by an enlarged heart, pulmonary venous congestion on radiograph, peripheral edema, response to CHF treatment, and the absence of malignancy or pulmonary infiltrates associated with an inflammatory process. The nephrotic syndrome was diagnosed when the patient had proteinuria, edema, and hypoalbuminemia. Effusions were considered secondary to liver cirrhosis when clinical or laboratory evidence of hepatic damage with portal hypertension or hypoalbuminemia occurred in the absence of heart failure or malignancy.

The effusions secondary to diseases that directly involve pleural surfaces were considered exudates and the rest were considered transudates.

The following criteria for exudates were initially evaluated: Light’s criteria [1], PF to serum protein ratio >0.5, PF to serum LDH ratio >0.6, PF LDH >307 IU/l, the modified Light criteria (PF to serum protein ratio >0.6, PF to serum LDH ratio >0.9 or PF LDH >280 IU/l) [6], PF cholesterol >60 mg/dl, and PF to serum cholesterol ratio >0.3. A PF cholinesterase ratio >1,390 U/l, a PF to serum cholinesterase ratio >0.27, and a combination of cholesterol >60 mg/dl plus LDH >307 U/l were also applied, after being established as the best discriminating levels between transudates and exudates in this series, using receiver operating characteristic curves.

We calculated sensitivity, specificity and accuracy of each biochemical parameter for identifying exudates. Specificity for exudates equals sensitivity for transudates and vice versa. Therefore, we report sensitivities and specificities only for exudates which the reader can use to determine corresponding values for transudates. To compare the usefulness of the different criteria we used the McNemar test and to assess the consistency of interstudies the =χ2 analysis. A p value less than 0.05 was considered significant.

 

goto top of outline results

Two hundred eighty-one patients with pleural effusion were evaluated. In 21 patients, despite extensive evaluation, the cause of the pleural effusion, always an exudate, was either indeterminate (9 patients), or was due to multiple superimposed diseases (12 patients). These and 17 additional patients, in whom not all the biochemical parameters were simultaneously determined, were excluded from the study.

The average age of the remaining 243 patients with an effusion of single and known cause was 59 ± 19 years (range 12–91 years); there were 137 men and 106 women. Sixty-one patients (25%) were defined as having transudates and 182 were identified as having exudates. Among the 61 patients with transudates, 34 (56%) were men and 27 (44%) women, with an average age of 67 ± 13 years. Among the 182 patients with exudates, 103 (57%) were men and 79 (43%) women, with an average age of 55 ± 19 years. The causes of the 243 effusions are shown in table 1. Forty-four of the 61 transudates (72%) were secondary to CHF. Eighty-six of the 182 exudates (47%) were considered malignant effusions. Receiver operating characteristic curves and pairwise comparison of individual tests are shown in figure 1. PF to serum protein ratio showed the highest area under the curve. The difference was statistically significant with respect to most parameters (p < 0.05). Table 2 shows the percentage of correctly classified transudates and exudates, and the accuracy of each criterion used. The sensitivity and accuracy of Light’s criteria were higher than those of the other criteria used, although the differences with the PF LDH-cholesterol combination did not show statistical significance. The specificity of the PF LDH concentration (95%), the highest of all criteria tested, was significantly higher (p < 0.05) than that of Light’s criteria (85%). The etiology of misclassified pleural effusions and the criteria used are shown in table 3. Of the 9 transudates misclassified by Light’s criteria, 6 (67%) were classified correctly by the LDH concentration in PF, and 3 of these were receiving diuretics. When comparing the results of the present series with those obtained in our previous series, no statistically significant difference in sensitivity, specificity or accuracy was found between most criteria evaluated. Only the accuracy (p < 0.05) and sensitivity (p < 0.05) of PF LDH were significantly higher in the present series (table 4).

TAB01

Table 1. Cause of pleural effusions

TAB02

Table 2. Transudates and exudatescorrectly classified for every criterionstudied: comparison with Light’s criteria [1]

TAB03

Table 3. Number of misclassified pleural effusions in each diagnostic group for every parameter studied

TAB04

Table 4. Sensitivity for exudates and accuracy for every criterion tested: comparison with those of a previous study

FIG01

Fig. 1. Receiver operating characteristic curves, area under the curve (AUC) and pairwise comparison of individual tests. Cholest = Cholesterol; cholin = cholinesterase.

 

goto top of outline discussion

In a previous study, we examined the sensitivity and specificity of Light’s original criteria for the separation of transudates and exudates, and evaluated the contribution that cholesterol can provide to improve the diagnostic accuracy [6]. Now, we have extended that study showing that, in both groups, there is a significant superiority of Light’s criteria over the other alternative criteria, including the PF/serum cholinesterase ratio. However, the recently proposed combination of LDH and cholesterol in PF has a discriminative capacity that may be considered equivalent to that of Light’s criteria. Because this equivalent accuracy has been repeatedly found [8, 9, 10], the decision of changing to this last option may be considered justified based on the cost saving of the determination of PF protein concentrations as well as the simultaneous collection and processing of a serum sample. Moreover, the routine determination of cholesterol in PF may be useful in the identification of some lipid effusions. However, to abandon the use of criteria traditionally used with satisfactory results based on convenient more than sound advantages will be a decision not everyone will take.

Because the sensitivity for exudative effusions and the overall accuracy of Light’s criteria are in the upper 90s, looking for alternative criteria has been discouraged [11]. However, using those criteria, 10–30% of all transudates will be considered exudates, a shortcoming which is not even solved with the alternative use of the criteria proposed by Costa et al. [8]. Patients in whom a false diagnosis of exudate is done may be subjected to a blind pleural biopsy, followed by pleuroscopy and thoracotomy. All these techniques, now useless, will probably have an increased morbidity in patients with an underlying heart or liver failure. Moreover, this misdiagnosis may contribute to defer the treatment of a potentially reversible condition.

The inclusion of a high percentage of transudates will magnify the problem. The proportion of transudates included is overestimated by the fact that most, if not all, are diagnosed [7, 12, 13]. The absence of idiopathic transudates has been a constant in the series, which like ours refer to the exudative or transudative nature of these effusions. This fact may reflect an easier diagnosis of transudates or a deliberate inclusion of effusions already diagnosed, only justified for investigational purposes to increase the number of patients making the ‘control group’ more specific.

Whatever the reason, in clinical settings with a high prevalence of transudates, the use of alternative criteria with a better specificity may be indicated. In patients receiving diuretics, the use of the serum-effusion albumin gradient has been recommended [4, 5, 14]. Herein, where these last parameters were not determined, the pleural LDH concentration showed, consistent with most previous studies [1, 3, 6], a very high specificity and appears as the alternative election criterion. Its use will have an added advantage, to avoid the determination of parameters not included in Light’s criteria.

As long as diagnostic limitations persist, the search for better alternatives seems justified, but to accept an alternative as valid, it must be prospectively confirmed in an independent, random patient population. This may allow an endless series of articles as has been pointed out previously [11]. Searching for the reasons of these limitations seems more rational than looking for new empirical alternatives, that, as the present study shows with the use of the PF/serum cholinesterase ratio, lose part of its excellent performance when it is tried in a different setting.

Few methodological differences existed between the two studies here compared. The simultaneous collection of blood and PF specimens and the exclusion of patients in whom not all tests have been performed, only applied in the present study, may have entailed some consequences, as the observed improvement in the specificity of Light’s criteria, from 77 to 85%. However, due to the higher proportion of transudates now included, a net increase in accuracy was not evident. In fact, in this second study, the only significant variation observed was an increase in the sensitivity of pleural LDH that rose from 66 to 77%. Although the reason for this variation is not clear, the fact that it did not modify the overall discriminative capacity of Light’s criteria reinforce the value of these criteria and that of those criteria using an ‘and/or’ structure. Moreover, as the results of the present interstudy comparison demonstrate, when the methods used are fairly similar, the discriminative capacity of the different criteria tested do not vary with time.

In summary, due to its consistency and accuracy, Light’s criteria remain the criteria of choice for segregating exudates from transudates. Based on cost-efficiency reasons, the PF LDH-cholesterol combination appears to be an alternative. Because both sets of criteria misdiagnose a substantial percentage of transudates, exceptions based on clinical judgment and the complementary use of a more specific criterion, as the PF concentration of LDH, may be considered.


 goto top of outline References
  1. Light RW, Mac Gregor MI, Luchsinger PC, Ball WC: Pleural effusions: The diagnostic separation of transudates and exudates. Ann Intern Med 1972;77:507–513.
  2. Hamm H, Brohan U, Bohmer R, Missmahl HP: Cholesterol in pleural effusions: A diagnostic aid. Chest 1987;92:296–302.
  3. Valdes L, Pose A, Suarez J, et al: Cholesterol: A useful parameter for distinguishing between pleural exudates and transudates. Chest 1991;99:1097–1102.
  4. Roth BJ, O’Meara TF, Cragun WH: The serum-effusion albumin gradient in the evaluation of pleural effusions. Chest 1990;98:546–549.
  5. Burgess LJ, Maritz FJ, Taljaard JJF: Comparative analysis of the biochemical parameters used to distinguish between pleural transudates and exudates. Chest 1995;107:1604–1609.
  6. Romero S, Candela A, Martín C, Hernández L, Trigo C, Gil J: Evaluation of different criteria for the separation of pleural transudates from exudates. Chest 1993;104:399–404.
  7. García-Pachon E, Padilla-Navas I, Sánchez JF, Jiménez B, Custardoy J: PF to serum cholinesterase ratio for the separation of transudates and exudates. Chest 1996;110:97–101.
  8. Costa M, Quiroga T, Cruz E: Measurement of PF cholesterol and lactate dehydrogenase: A simple and accurate set of indicators for separating exudates from transudates. Chest 1995;108:1260–1263.
  9. Romero S, Candela A, Arriero JM: ‘Keep testing the waters’. Chest 1996;110:299.
  10. Heffner JE, Brown LK, Barbieri CA (for the Primary Study Investigators): Diagnostic value of tests that discriminate between exudative and transudative pleural effusions. Chest 1997;111:970–980.
  11. Bartter T: Communications to the editor. Chest 1996;110:300.
  12. Roth BJ: Evaluating PF. Chest 1996;110:7–8.
  13. Peterman TA, Speicher CE: Evaluating pleural effusions. A two stage laboratory approach. JAMA 1984;252:1051–1053.
  14. Light RW: Diagnostic principles in pleural disease. Eur Respir J 1997;10:476–481.

 goto top of outline Author Contacts

Santiago Romero Candeira
C/ Italia, no 30, Esc. 2a, 1° Dcha
E–03003 Alicante (Spain)
Tel. +34 96 5938455, Fax +34 96 5249759


 goto top of outline Article Information

Received: Received: March 10, 1999
Accepted: May 7, 1999
Number of Print Pages : 6
Number of Figures : 1, Number of Tables : 4, Number of References : 14


 goto top of outline Publication Details

Respiration (International Review of Thoracic Diseases)
Founded 1944 as ‘Schweizerische Zeitschrift für Tuberkulose und Pneumonologie’ by E. Bachmann, M. Gilbert, F. Häberlin, W. Löffler, P. Steiner and E. Uehlinger, continued 1962–1967 as ‘Medicina Thoracalis’

Vol. 67, No. 1, Year 2000 (Cover Date: January-February 2000)

Journal Editor: C.T. Bolliger, Cape Town
ISSN: 0025–7931 (print), 1423–0356 (Online)

For additional information: http://www.karger.com/journals/res


Copyright / Drug Dosage / Disclaimer

Copyright: All rights reserved. No part of this publication may be translated into other languages, reproduced or utilized in any form or by any means, electronic or mechanical, including photocopying, recording, microcopying, or by any information storage and retrieval system, without permission in writing from the publisher or, in the case of photocopying, direct payment of a specified fee to the Copyright Clearance Center.
Drug Dosage: The authors and the publisher have exerted every effort to ensure that drug selection and dosage set forth in this text are in accord with current recommendations and practice at the time of publication. However, in view of ongoing research, changes in goverment regulations, and the constant flow of information relating to drug therapy and drug reactions, the reader is urged to check the package insert for each drug for any changes in indications and dosage and for added warnings and precautions. This is particularly important when the recommended agent is a new and/or infrequently employed drug.
Disclaimer: The statements, opinions and data contained in this publication are solely those of the individual authors and contributors and not of the publishers and the editor(s). The appearance of advertisements or/and product references in the publication is not a warranty, endorsement, or approval of the products or services advertised or of their effectiveness, quality or safety. The publisher and the editor(s) disclaim responsibility for any injury to persons or property resulting from any ideas, methods, instructions or products referred to in the content or advertisements.

Abstract

Objectives: The first objective was to assess the diagnostic value of new biochemical criteria proposed to discriminate pleural transudates from exudates and to compare their efficiency with those of Light’s criteria. The second objective of the study was to assess the interstudy variability of the parameters repeatedly determinated in two different groups of patients with pleural effusion. Patients and Methods: We recorded clinical characteristics and final diagnoses and measured pleural fluid (PF) and serum levels of protein, LDH, cholesterol and cholinesterase of 243 patients with pleural effusion. Results: Sixty-one (25%) pleural effusions were transudates and 182 were exudates. The sensitivity (99%) and accuracy (96%) of Light’s criteria were higher than those of the other criteria tested, although the differences with those of the PF LDH-cholesterol combination (96 and 93%) did not show statistical significance. Pleural LDH concentration was the criterion with the highest specificity (95%), being significantly higher (p < 0.05) than that of Light’s criteria. The sensitivity, specificity and accuracy of most criteria tested did not vary when compared with those obtained in a study performed 5 years previously. Conclusions: Light’s criteria remain the criteria of choice for segregating exudates from transudates. Based on cost-efficiency reasons, the PF LDH-cholesterol combination appears as an alternative. Because both sets of criteria misdiagnose a substantial percentage of transudates, exceptions based on good clinical judgment and the complementary use of a more specific criterion, as the PF concentration of LDH, must be considered.

Copyright © 2000 S. Karger AG, Basel



 goto top of outline Author Contacts

Santiago Romero Candeira
C/ Italia, no 30, Esc. 2a, 1° Dcha
E–03003 Alicante (Spain)
Tel. +34 96 5938455, Fax +34 96 5249759


 goto top of outline Article Information

Received: Received: March 10, 1999
Accepted: May 7, 1999
Number of Print Pages : 6
Number of Figures : 1, Number of Tables : 4, Number of References : 14


 goto top of outline Publication Details

Respiration (International Review of Thoracic Diseases)
Founded 1944 as ‘Schweizerische Zeitschrift für Tuberkulose und Pneumonologie’ by E. Bachmann, M. Gilbert, F. Häberlin, W. Löffler, P. Steiner and E. Uehlinger, continued 1962–1967 as ‘Medicina Thoracalis’

Vol. 67, No. 1, Year 2000 (Cover Date: January-February 2000)

Journal Editor: C.T. Bolliger, Cape Town
ISSN: 0025–7931 (print), 1423–0356 (Online)

For additional information: http://www.karger.com/journals/res


Copyright / Drug Dosage

Copyright: All rights reserved. No part of this publication may be translated into other languages, reproduced or utilized in any form or by any means, electronic or mechanical, including photocopying, recording, microcopying, or by any information storage and retrieval system, without permission in writing from the publisher or, in the case of photocopying, direct payment of a specified fee to the Copyright Clearance Center.
Drug Dosage: The authors and the publisher have exerted every effort to ensure that drug selection and dosage set forth in this text are in accord with current recommendations and practice at the time of publication. However, in view of ongoing research, changes in goverment regulations, and the constant flow of information relating to drug therapy and drug reactions, the reader is urged to check the package insert for each drug for any changes in indications and dosage and for added warnings and precautions. This is particularly important when the recommended agent is a new and/or infrequently employed drug.
Disclaimer: The statements, opinions and data contained in this publication are solely those of the individual authors and contributors and not of the publishers and the editor(s). The appearance of advertisements or/and product references in the publication is not a warranty, endorsement, or approval of the products or services advertised or of their effectiveness, quality or safety. The publisher and the editor(s) disclaim responsibility for any injury to persons or property resulting from any ideas, methods, instructions or products referred to in the content or advertisements.

References

  1. Light RW, Mac Gregor MI, Luchsinger PC, Ball WC: Pleural effusions: The diagnostic separation of transudates and exudates. Ann Intern Med 1972;77:507–513.
  2. Hamm H, Brohan U, Bohmer R, Missmahl HP: Cholesterol in pleural effusions: A diagnostic aid. Chest 1987;92:296–302.
  3. Valdes L, Pose A, Suarez J, et al: Cholesterol: A useful parameter for distinguishing between pleural exudates and transudates. Chest 1991;99:1097–1102.
  4. Roth BJ, O’Meara TF, Cragun WH: The serum-effusion albumin gradient in the evaluation of pleural effusions. Chest 1990;98:546–549.
  5. Burgess LJ, Maritz FJ, Taljaard JJF: Comparative analysis of the biochemical parameters used to distinguish between pleural transudates and exudates. Chest 1995;107:1604–1609.
  6. Romero S, Candela A, Martín C, Hernández L, Trigo C, Gil J: Evaluation of different criteria for the separation of pleural transudates from exudates. Chest 1993;104:399–404.
  7. García-Pachon E, Padilla-Navas I, Sánchez JF, Jiménez B, Custardoy J: PF to serum cholinesterase ratio for the separation of transudates and exudates. Chest 1996;110:97–101.
  8. Costa M, Quiroga T, Cruz E: Measurement of PF cholesterol and lactate dehydrogenase: A simple and accurate set of indicators for separating exudates from transudates. Chest 1995;108:1260–1263.
  9. Romero S, Candela A, Arriero JM: ‘Keep testing the waters’. Chest 1996;110:299.
  10. Heffner JE, Brown LK, Barbieri CA (for the Primary Study Investigators): Diagnostic value of tests that discriminate between exudative and transudative pleural effusions. Chest 1997;111:970–980.
  11. Bartter T: Communications to the editor. Chest 1996;110:300.
  12. Roth BJ: Evaluating PF. Chest 1996;110:7–8.
  13. Peterman TA, Speicher CE: Evaluating pleural effusions. A two stage laboratory approach. JAMA 1984;252:1051–1053.
  14. Light RW: Diagnostic principles in pleural disease. Eur Respir J 1997;10:476–481.