0
Original Research: PLEURAL TUBERCULOSIS |

Diagnostic Value of Interferon-γ in Tuberculous Pleurisy*: A Metaanalysis FREE TO VIEW

Jing Jiang, MD; Huan-Zhong Shi, MD, PhD; Qiu-Li Liang, MD; Shou-Ming Qin, MD; Xue-Jun Qin, MD
Author and Funding Information

*From the Institute of Respiratory Diseases, First Affiliated Hospital, Guangxi Medical University, Nanning, Guangxi, People’s Republic of China.

Correspondence to: Huan-Zhong Shi, MD, Institute of Respiratory Diseases, First Affiliated Hospital, Guangxi Medical University, Nanning 530021, Guangxi, People’s Republic of China; e-mail: hzshi@vip.tom.com



Chest. 2007;131(4):1133-1141. doi:10.1378/chest.06-2273
Text Size: A A A
Published online

Background: Conventional tests are not always helpful in making a diagnosis of tuberculous pleurisy. Many studies have investigated the usefulness of interferon (IFN)-γ measurements in pleural fluid for the early diagnosis of tuberculous pleurisy. We conducted a metaanalysis to determine the accuracy of IFN-γ measurements in the diagnosis of tuberculous pleurisy.

Methods: After a systematic review of English-language studies, sensitivity, specificity, and other measures of accuracy of IFN-γ concentrations in the diagnosis of pleural effusion were pooled using random-effects models. Summary receiver operating characteristic curves were used to summarize overall test performance.

Results: Twenty-two studies met our inclusion criteria. The summary estimates for IFN-γ in the diagnosis of tuberculous pleurisy in the studies included were as follows: sensitivity, 0.89 (95% confidence interval [CI], 0.87 to 0.91); specificity, 0.97 (95% CI, 0.96 to 0.98); positive likelihood ratio, 23.45 (95% CI, 17.31 to 31.78); negative likelihood ratio, 0.11 (95% CI, 0.07 to 0.16); and diagnostic odds ratio, 272.7 (95% CI, 147.5 to 504.2).

Conclusions: IFN-γ determination is a sensitive and specific test for the diagnosis of tuberculous pleurisy. The measurement of IFN-γ levels in pleural effusions is thus likely to be a useful tool for diagnosing tuberculous pleurisy. The results of IFN-γ assays should be interpreted in parallel with clinical findings and the results of conventional tests.

Figures in this Article

Tuberculosis is the leading cause of death from a curable infectious disease. In China, the prevalence of active pulmonary tuberculosis in 2000 was 367 per 100,000 population, the prevalence of smear positive pulmonary tuberculosis was 122 per 100,000 population, and the prevalence of bacteriological positive pulmonary tuberculosis was 160 per 100,000 population.1On the basis of results of surveys of the prevalence of infection and disease, assessments of the effectiveness of surveillance systems, and death registrations, there were an estimated 8.9 million new cases of tuberculosis in 2004, fewer than half of which were reported to public-health authorities and World Health Organization.2Tuberculous pleural effusion (TPE) is caused by a severe delayed-type hypersensitivity reaction in response to the rupture of a subpleural focus of Mycobacterium tuberculosis infection. Although TPE occurs in about 10% of untreated individuals who test positive by the tuberculin test, it may also develop as a complication of primary pulmonary tuberculosis.3 Actually, tuberculosis is the major cause of pleural effusions in areas of high tuberculosis prevalence, and TPE usually manifests as lymphocytic exudative effusion.4Many studies have investigated the usefulness of interferon (IFN)-γ in pleural fluid for the early diagnosis of tuberculous pleural effusion (TPE), and an early metaanalysis has shown that the value of pleural IFN-γ measurements for the diagnosis of TPE was reasonably good5; however, likelihood ratios including both positive likelihood ratio (PLR) and negative likelihood ratio (NLR), and diagnostic odds ratio (DOR) have not been evaluated in the metaanalysis. In that metaanalysis, 13 early related studies were included. Since that time, additional clinical studies determining the concentrations of IFN-γ have been reported. It is well defined that accuracy is the degree of conformity of a measured or calculated value to its actual or specified value. Although the accuracy of IFN-γ detection for the diagnosis of TPE has been extensively studied, the exact role of these detections remains controversial. We performed the present metaanalysis to establish the overall accuracy of IFN-γ measurement for the diagnosis of TPE.

Search Strategy and Study Selection

Since the present study was a metaanalysis that was based on published articles, we did not include the consents of patients and the approval of internal review boards. We searched the following electronic databases: Medline (1980 to 2006); Embase (1980 to 2006); Web of Science (1990 to 2006); BIOSIS (1993 to 2006); and LILACS (1980 to 2006). We also reviewed the Cochrane Library to find relevant articles. All searches were up to date as of October 2006. The search terms used were “tuberculosis,” “Mycobacterium tuberculosis,” “pleurisy/pleuritis,” “pleural effusion/pleural fluid,” “interferon/IFN,” “sensitivity and specificity,” and “accuracy.” We contacted experts in the specialty, and searched the reference lists from primary and review articles. Although no language restrictions were imposed initially, for the full-text review and final analysis our resources only permitted the review of articles published in the English language. Conference abstracts and letters to the journal editors were excluded because of the limited data presented in them.

A study was included in the metaanalysis when it provided both the sensitivity (true-positive rate) and the specificity (false-positive rate) of IFN-γ for the diagnosis of TPE, or when it provided IFN-γ values in a dot-plot form, allowing test results to be extracted for individual study subjects. The studies including at least 10 TPE specimens were selected for inclusion in the study, since very small studies may be vulnerable to selection bias. Two reviewers (J.J. and H.Z.S.) independently judged study eligibility while screening the citations. Disagreements were resolved by consensus.

Data Extraction and Quality Assessment

The final set of English language articles was assessed independently by two reviewers (J.J. and H.Z.S.). The reviewers were blinded to publication details, and disagreements were resolved by consensus. Data retrieved from the reports included participant characteristics, test methods, sensitivity and specificity data, cutoff values, publication year, and methodological quality. Where pleural IFN-γ values were provided in dot plots, scalar grids were placed over the plots, and the values were measured and used to produce a receiver operating characteristic (ROC) curve for each study (SPSS; Chicago, IL). The numbers of true-positive, false-positive, false-negative, and true-negative results are displayed for each study in Table 1 .

We assessed the methodological quality of the studies using guidelines published by the standards for reporting diagnostic accuracy (STARD) initiative6(maximum score, 25) [ie, guidelines that aim to improve the quality of reporting in diagnostic studies] and the quality assessment for studies of diagnostic accuracy (QUADAS) tool7 (maximum score, 14) [ie, appraisal by use of empirical evidence, expert opinion, and formal consensus to assess the quality of primary studies of diagnostic accuracy]. In addition, for each study the following characteristics of study design were also retrieved: (1) cross-sectional design (vs case-control design); (2) consecutive or random sampling of patients; (3) blinded (single or double) interpretation of determination and reference standard results; and (4) prospective data collection. If no data on the above criteria were reported in the primary studies, we requested the information from the authors. If the authors did not respond to our letters, the “unknown” items were treated as “no.”

Statistical Analysis

We used standard methods recommended for metaanalyses of diagnostic test evaluations.8 Analyses were performed using several statistical software programs (Stata, version 8.2; Stata Corporation; College Station, TX; Meta-Test, version 0.6; New England Medical Center; Boston, MA; and Meta-DiSc for Windows; XI Cochrane Colloquium; Barcelona, Spain). We computed the following measures of test accuracy for each study: sensitivity; specificity; PLR; NLR; and DOR.

The analysis was based on a summary ROC (SROC) curve.89 The sensitivity and specificity for the single test threshold identified for each study were used to plot an SROC curve.910 A random-effects model was used to calculate the average sensitivity, specificity, and the other measures across studies.1112

The term heterogeneity when used in relation to metaanalyses refers to the degree of variability in results across studies. We used the χ2 and Fisher exact tests to detect statistically significant heterogeneity. To assess the effects of STARD and QUADAS scores on the diagnostic ability of IFN-γ, we included them as covariates in univariate metaregression analysis (inverse variance weighted). We also analyzed the effects of other covariates on DOR (ie, cross-sectional design, consecutive or random sampling of patients, single or double interpretation of determination and reference standard results, and prospective data collection). The relative DOR (RDOR) was calculated according to standard methods to analyze the change in diagnostic precision in the study per unit increase in the covariate.1314 Since publication bias is of concern for metaanalyses of diagnostic studies, we tested for the potential presence of this bias using funnel plots and the Egger test.15

After independent review, 34 publications dealing with pleural IFN-γ concentrations for the diagnosis of TPE were considered to be eligible for inclusion in the analysis.1649 Of these publications, two studies3839 were excluded because IFN-γ concentration was determined only in TPE patients, two studies4041 were excluded because they recruited < 10 patients with confirmed TPE, four studies4245 were excluded because they did not allow the calculation of sensitivity or specificity, and four studies4649 were excluded because they included groups of patients that had been described elsewhere. The data in the study by Villena et al46 were reported in another study by Villena et al29 (V. Villena, MD; personal communication; July 2006). The data in the study by Sharma et al47 were reported in the study by Sharma and Banga32, and those in the studies by Hiraki and colleagues4849 were reported in the study by Aoe et al.28 Subsequently, 22 studies1637 including 782 patients with TPE and 1,319 non-TPE patients were available for analysis, and the clinical characteristics of these studies, along with QUADAS scores, are outlined in Table 1.

Quality of Reporting and Study Characteristics

The average interrater agreement between the two reviewers for items in the quality checklist was 0.85. The average sample size of the included studies was 96 (range, 21 to 595). In one study,34 patients with TPE received diagnoses based on clinical presentation, pleural fluid analysis, radiology findings, and the responsiveness of the patient to antituberculous chemotherapy. In two studies,22,26 a small proportion of the patients studied received diagnoses according to the clinical presentation, pleural fluid analysis, radiology findings, and the responsiveness of the patient to antituberculous chemotherapy, but the diagnosis of pleural tuberculosis was confirmed in most of the TPE patients based on the conventional “gold standard,” which is a smear or culture that is positive for M tuberculosis taken from pleural fluid and/or histology showing a caseating granuloma. In the remaining 19 studies,,1621,2325,2733,3537 the diagnoses of all patients with TPE were made based on a smear or culture that was positive for M tuberculosis that had been taken from pleural fluid and/or histology showing a caseating granuloma. Our initial data were affected by the poor quality of reporting in the primary studies. To overcome this problem, we contacted all authors of the 22 studies included by air mail as well as e-mail when e-mail addresses were available. Seventeen authors responded who could provide additional data for 18 studies.,1618,2027,2932,34,3637 As shown in Table 2 , in 10 of 22 studies (45.5%), the study was cross-sectional design. In 13 studies (59.1%), the samples were collected from consecutive patients. Eleven studies (50.0%) reported blinded interpretation of the IFN-γ assay independent of the reference standard. Fourteen studies (63.6%) that reported that the study design was prospective can be identified from Table 2.

Diagnostic Accuracy

Figure 1 shows the forest plot of sensitivity and specificity for 22 IFN-γ assays in the diagnosis of TPE. The sensitivity ranged from 0.64 to 1.00 (mean, 0.89; 95% confidence interval [CI], 0.87 to 0.91), while specificity ranged from 0.86 to 1.00 (mean, 0.97; 95% CI, 0.96 to 0.98). We also noted that PLR was 23.45 (95% CI, 17.31 to 31.78), NLR was 0.11 (95% CI, 0.07 to 0.16), and DOR was 272.7 (95% CI, 147.5 to 504.2). χ2 values of sensitivity, specificity, PLR, NLR, and DOR were 67.54 (p < 0.001), 32.66 (p = 0.050), 21.21 (p = 0.446), 62.36 (p < 0.001), and 30.04 (p = 0.091), respectively, indicating a significant heterogeneity for sensitivity and NLR between studies.

Unlike a traditional ROC plot that explores the effect of varying thresholds (ie, cut points for determining test positives) on sensitivity and specificity in a single study, each data point in the SROC plot represents a separate study. The SROC curve presents a global summary of test performance, and shows the tradeoff between sensitivity and specificity. A graph of the SROC curve for the IFN-γ determination showing true-positive rates vs false-positive rates from individual studies is shown in Figure 2 . As a global measure of test efficacy we used the Q-value, the intersection point of the SROC curve with a diagonal line from the left upper corner to the right lower corner of the ROC space, which corresponds to the highest common value of sensitivity and specificity for the test. This point does not indicate the only or even the best combination of sensitivity and specificity for a particular clinical setting but represents an overall measure of the discriminatory power of a test. Our data showed that the SROC curve is positioned near the desirable upper left corner of the SROC curve, and that the maximum joint sensitivity and specificity (ie, the Q value) was 0.95; while the area under the curve (AUC) was 0.99 (weighted AUC, 0.98), indicating a high level of overall accuracy.

Multiple Regression Analysis and Publication Bias

By use of the STARD guidelines,6a quality score for every study was compiled on the basis of title and introduction, methods, results, and discussion (Table 1). Quality scoring was also done by use of QUADAS,7 in which a score of 1 was given when a criterion was fulfilled, 0 if a criterion was unclear, and –1 if the criterion was not achieved (Table 1). These scores were used in the metaregression analysis to assess the effect of study quality on the RDOR of IFN-γ in the diagnosis of TPE. As shown in Table 3 , studies with higher quality (STARD score, ≥ 13; QUADAS score, ≥ 10) produced RDOR values that were not significantly higher than those studies with lower quality. We also noted that differences for studies with or without blinded, cross-sectional, consecutive/random, and prospective designs did not reach statistical significance, indicating that the study design did not substantially affect the diagnostic accuracy.

The evaluation of publication bias showed that the Egger test was significant (p = 0.023). The funnel plots for publication bias (Fig 3 ) also show some asymmetry. These results indicate a potential for publication bias.

Making a differential diagnosis between TPE and non-TPE is a critical clinical problem, and the conventional methods, such as the direct examination of pleural fluid by Ziehl-Neelsen staining, culture of the pleural fluid, and pleural biopsy, are not always helpful in making the diagnosis since they have limitations. Findings of microscopy of the pleural fluid is rarely positive (< 5%).5052 Culture of pleural fluid has low sensitivity (24 to 58%), and several weeks are required to grow M tuberculosis.,51,53 Biopsy of pleural tissue and culture of biopsy material are widely held to be the best methods of confirming the diagnosis.50,52 Although not perfect, culture and/or biopsy, therefore, are widely considered to be the standard of diagnosis.50 However, pleural biopsy is invasive, operator-dependent, and technically difficult (particularly in children).54 Sometimes, a differential diagnosis of TPE mandates the use of more invasive procedures like thoracoscopy or thoracotomy. These procedures, which require expertise, may cause complications and may even increase the morbidity of the patients.

Pleural levels of a number of biomarkers have been proposed as aids in the diagnosis of TPE, including those of INF-γ, adenosine deaminase, interleukin-12p40, interleukin-18, immunosuppressive acidic protein, and soluble interleukin-2 receptor, the levels of which are all significantly higher in TPE patients than in non-TPE patients.45,55 Hiraki and colleagues49 have compared directly the sensitivities of these markers in a study. ROC analysis demonstrated that INF-γ is the most sensitive and specific indicator of TPE among the above six biological markers (AUC, 1.00). The next most sensitive indicator was soluble interleukin-2 receptor (AUC, 0.99), followed by adenosine deaminase (AUC, 0.96), interleukin-18 (AUC, 0.95), immunosuppressive acidic protein (AUC, 0.93), and interleukin-12p40 (AUC, 0.87). The SROC curve and its AUC present an overall summary of test performance, and display the tradeoff between sensitivity and specificity. The present metaanalysis has shown that the mean sensitivity of the IFN-γ assay was 0.89 while the mean specificity was 0.97, and that the maximum joint sensitivity and specificity (Q value) was 0.95 while the AUC was 0.99, indicating a high level of overall accuracy. We also noted that only one study19 showed relatively low sensitivity (< 0.70) and low specificity (< 0.90) for the detection of IFN-γ in diagnosing TPE, which had low START and QUADAS scores with the smallest study size (n = 21) among all studies included in the present metaanalysis. This situation might account for the low sensitivity and low specificity of this study.

The DOR is a single indicator of test accuracy56 that combines the data from sensitivity and specificity into a single number. The DOR of a test is the ratio of the odds of positive test results in the patient with disease relative to the odds of positive test results in the patient without disease. The value of a DOR ranges from 0 to infinity, with higher values indicating better discriminatory test performance (ie, higher accuracy). A DOR of 1.0 indicates that a test does not discriminate between patients with the disorder and those without it. In the present metaanalysis, we have found that the mean DOR was 272.7, also indicating a high level of overall accuracy.

Since the SROC curve and the DOR are not easy to interpret and use in clinical practice,57 and since likelihood ratios are considered to be more clinically meaningful,5758 we also presented both PLR and NLR as our measures of diagnostic accuracy. Likelihood ratios of > 10 or < 0.1 generate large and often conclusive shifts from pretest to posttest probability (indicating high accuracy).58 A PLR value of 23.45 suggests that patients with TPE have an approximately 23-fold higher chance of being IFN-γ assay-positive compared with patients without TPE. This high probability would be considered high enough to begin or to continue antituberculosis treatment of TPE patients, especially in the case of the absence of any malignant evidences. On the other hand, NLR was found to be 0.11 in the present metaanalysis. If the IFN-γ assay result was negative, the probability that this patient has TPE is approximately 10%, which is not low enough to rule out TPE. These data suggest that a negative IFN-γ assay result should not be used alone as a justification to deny or to discontinue antituberculosis therapy. The choice of therapeutic strategy should be based on the results of microscopic examination of a smear or culture for M tuberculosis and/or histologic observation of pleural tissue, as well as the other clinical data, such as the response to antituberculosis therapy.

An exploration of the reasons for heterogeneity rather than the computation of a single summary measure is an important goal of metaanalysis.59 In our metaanalysis, both STARD and QUADAS scores were used in the metaregression analysis to assess the effect of study quality on RDOR. We did not observe that the studies with higher quality (ie, STARD score of ≥ 13 or QUADAS score of ≥ 10) had better test performances than those with lower quality. Although we found a significant heterogeneity for sensitivity and NLR among the studies analyzed, we noted that differences for studies with or without blinded, cross-sectional, consecutive/random, and prospective design did not reach statistical significance, indicating that the study design did not substantially affect diagnostic accuracy.

It should be emphasized that a definite diagnosis of TPE is achieved when M tuberculosis is demonstrated in sputum or pleural specimens, or when caseating granulomas are found in pleural biopsy specimens. As abovementioned, microscopy of the pleural fluid is rarely positive (< 5%).5052 Histologic examination of pleura by needle biopsy is not conclusive in 20 to 40% of patients with TPE.5051 When the pleural biopsy finding is negative, mycobacteria can be cultured in pleural specimens in < 10% of patients,50 and this usually takes at least 3 weeks. Where diagnostic difficulty exists, measuring the levels of several biomarkers, such as adenosine deaminase and IFN-γ, in pleural fluid is useful, and clinicians can embark on empirical antituberculosis therapy while awaiting culture results, especially in young patients from areas with a high prevalence of tuberculosis. One criticism of the use biomarkers rather than cultures for the diagnosis of TPE is that culture results are not available to guide antituberculosis therapy. First and foremost among the shortcomings of this is the fact that none of the biomarkers, including IFN-γ, provide culture and sensitivity data. Culture results are particularly useful if drug-resistant tuberculosis is prevalent.60

Our data were consistent with the results of the previous metaanalysis on the accuracy of IFN-γ assays by Greco and colleagues.5 Their metaanalysis of 13 studies showed that both sensitivity and specificity estimates were heterogeneous. First of all, we updated that previous metaanalysis by including more recent related studies. An important strength of our study was its comprehensive search strategy. Screening, study selection, and quality assessment were done independently and reproducibly by two reviewers. Data extraction and quality assessment were performed in a blinded fashion to reduce bias. We reduced the problem of missing data by contacting authors. We also explored heterogeneity and potential publication bias in accordance with published guidelines.

Our metaanalysis had several limitations. First, the exclusion of conference abstracts, letters to the editors, and non–English-language studies may have led to publication bias, which was found in the present metaanalysis. However, a review of these abstracts and letters suggests that the overall results were similar to the results in the included English-language studies. Publication bias may also be introduced by the inflation of diagnostic accuracy estimates since studies that report positive results are more likely to be accepted for publication. Second, misclassification bias can occur. TPE is not always diagnosed by either histologic or microbiological examination. Actually, TPE was diagnosed in some patients with TPE based just on the clinical course. This issue regarding accuracy of diagnosis can cause nonrandom misclassification, leading to biased results.

In conclusion, current evidence suggests a potential role for IFN-γ assays in confirming a diagnosis of TPE. Since none of pleural fluid biomarkers, including IFN-γ, is specific for TPE, the results of IFN-γ assays should be interpreted in parallel with clinical findings and the results of conventional tests including microbiological examination and pleural biopsy.

Abbreviations: AUC = area under the curve; CI = confidence interval; DOR = diagnostic odds ratio; IFN = interferon; NLR = negative likelihood ratio; PLR = positive likelihood ratio; QUADAS = quality assessment for studies of diagnostic accuracy; RDOR = relative diagnostic odds ratio; ROC = receiver operating characteristic; SROC = summary receiver operating characteristic; STARD = standards for reporting diagnostic accuracy; TPE = tuberculous pleural effusion

Dr. Shi designed the study, searched the databases, extracted the data, analyzed the results, and wrote the manuscript. Dr. Jiang helped with study design, searching the databases, and writing and revising the manuscript. Dr. Liang formulated the research question, and helped with database searches and analysis. Drs. S.-M. Qin and X.-J. Qin helped to design the data abstraction form and served as second reviewers in extracting the data. All authors read and approved the final manuscript.

The authors have reported to the ACCP that no significant conflicts of interest exist with any companies/organizations whose products or services may be discussed in this article.

This study was supported in part by research grant 30660064 from National Natural Science Foundation of China, in part by Program NCET-04–0835 for New Century Excellent Talents in Chinese Universities, and in part by research grant 0639044 from the Natural Science Foundation of Guangxi Zhuang Autonomous Zone, People’s Republic of China.

Table Graphic Jump Location
Table 1. Summary of Included Studies*
* 

ELISA = enzyme-linked immunosorbent assay; RIA = radioimmunoassay; TP = true-positive; FP = false-positive; FN = false-negative; TN = true-negative.

Table Graphic Jump Location
Table 2. Characteristics of Included Studies*
* 

TB = tuberculosis; Bac = bacteriology; His = histology; Clin = clinical course.

Figure Jump LinkFigure 1. Forest plot of estimates of sensitivity and specificity for IFN-γ assays in the diagnosis of TPE. • = point estimates of sensitivity and specificity from each study; error bars = 95% CIs; numbers = reference numbers of studies cited in the reference list. Pooled estimates for IFN-γ assay were as follows: sensitivity, 0.89 (95% CI, 0.87 to 0.91); specificity, 0.97 (95% CI, 0.96 to 0.98).Grahic Jump Location
Figure Jump LinkFigure 2. SROC curves for IFN-γ assays. • = each study in the metaanalysis (the size of each study is indicated by the size of the solid circle); dark line = weighted regression; and dashed line = unweighted regression. SROC curves summarize the overall diagnostic accuracy.Grahic Jump Location
Table Graphic Jump Location
Table 3. Weighted Metaregression of the Effects of Methodological Quality and Study Design on Diagnostic Precision of Pleural IFN-γ in 22 Assays
Figure Jump LinkFigure 3. Funnel graph for the assessment of potential publication bias in IFN-γ assays. The funnel graph plots the log of the DOR against the SE of the log of the DOR (an indicator of sample size). • = each study in the metaanalysis; center line = SDOR. The result of the Egger test for publication bias was significant (p = 0.023).Grahic Jump Location

We are grateful to the following authors who sent additional information on their primary studies: A. Kaya, C. F. Wong, E. Ribera, K. Ogawa, K. Shimokata, L. Valdés, N. G. Saravia, S. K. Sharma, S. Wongtim, T. Morimoto, T. Pettersson, V. Villena, W. H. Hsu, Y. C. Kim, Y. Aoki, Y. Yamada, and Z. C. Gao.

. National Technical Steering Group of the Epidemiological Sampling Survey for Tuberculosis (2002) Report on fourth national epidemiological sampling survey of tuberculosis [in Chinese].Chin J Tuberc Respir Dis25,3-7
 
Dye, C Global epidemiology of tuberculosis.Lancet2006;367,938-940. [PubMed] [CrossRef]
 
Ellner, JJ, Barnes, PF, Wallis, RS, et al The immunology of tuberculous pleurisy.Semin Respir Infect1988;3,335-342. [PubMed]
 
Light, RW Pleural effusion.N Engl J Med2002;346,1971-1977. [PubMed]
 
Greco, S, Girardi, E, Masciangelo, R, et al Adenosine deaminase and interferon γ measurements for the diagnosis of tuberculous pleurisy: a meta-analysis.Int J Tuberc Lung Dis2003;7,777-786. [PubMed]
 
Bossuyt, PM, Reitsma, JB, Bruns, DE, et al Standards for reporting of diagnostic accuracy: towards complete and accurate reporting of studies of diagnostic accuracy; the STARD initiative.BMJ2003;326,41-44. [PubMed]
 
Whiting, P, Rutjes, AW, Reitsma, JB, et al The development of QUADAS: a tool for the quality assessment of studies of diagnostic accuracy included in systematic reviews.BMC Med Res Methodol2003;3,25. [PubMed]
 
Deville, WL, Buntinx, F, Bouter, LM, et al Conducting systematic reviews of diagnostic studies: didactic guidelines.BMC Med Res Methodol2002;2,9. [PubMed]
 
Moses, LE, Shapiro, D, Littenberg, B Combining independent studies of a diagnostic test into a summary ROC curve: data-analytic approaches and some additional considerations.Stat Med1993;12,1293-1316. [PubMed]
 
Lau, J, Ioannidis, JP, Balk, EM, et al Diagnosing acute cardiac ischemia in the emergency department: a systematic review of the accuracy and clinical effect of current technologies.Ann Emerg Med2001;37,453-460. [PubMed]
 
Irwig, L, Tosteson, AN, Gatsonis, C, et al Guidelines for meta-analyses evaluating diagnostic tests.Ann Intern Med1994;120,667-676. [PubMed]
 
Vamvakas, EC Meta-analyses of studies of the diagnostic accuracy of laboratory tests: a review of the concepts and methods.Arch Pathol Lab Med1998;122,675-686. [PubMed]
 
Suzuki, S, Moro-oka, T, Choudhry, NK The conditional relative odds ratio provided less biased results for comparing diagnostic test accuracy in meta-analyses.J Clin Epidemiol2004;57,461-469. [PubMed]
 
Westwood, ME, Whiting, PF, Kleijnen, J How does study quality affect the results of a diagnostic meta-analysis?BMC Med Res Methodol2005;8,20
 
Egger, M, Smith, G Davey, Schneider, M, et al Bias in meta-analysis detected by a simple, graphical test.BMJ1997;315,629-634. [PubMed]
 
Ribera, E, Ocana, I, Martinez-Vazquez, JM, et al High level of interferon γ in tuberculous pleural effusion.Chest1988;93,308-311. [PubMed]
 
Hsu, WH, Chiang, CD, Chen, WT, et al Diagnostic value of adenosine deaminase and γ-interferon in tuberculous and malignant pleural effusions.Taiwan Yi Xue Hui Za Zhi1989;88,879-882. [PubMed]
 
Shimokata, K, Saka, H, Murate, T, et al Cytokine content in pleural effusion: comparison between tuberculous and carcinomatous pleurisy.Chest1991;99,1103-1107. [PubMed]
 
Maeda, J, Ueki, N, Ohkawa, T, et al Local production and localization of transforming growth factor-β in tuberculous pleurisy.Clin Exp Immunol1993;92,32-38. [PubMed]
 
Valdes, L, San Jose, E, Alvarez, D, et al Diagnosis of tuberculous pleurisy using the biologic parameters adenosine deaminase, lysozyme, and interferon γ.Chest1993;103,458-465. [PubMed]
 
Aoki, Y, Katoh, O, Nakanishi, Y, et al A comparison study of IFN-γ, ADA, and CA125 as the diagnostic parameters in tuberculous pleuritis.Respir Med1994;88,139-143. [PubMed]
 
Soderblom, T, Nyberg, P, Teppo, AM, et al Pleural fluid interferon-γ and tumour necrosis factor-α in tuberculous and rheumatoid pleurisy.Eur Respir J1996;9,1652-1655. [PubMed]
 
Kim, YC, Park, KO, Bom, HS, et al Combining ADA, protein and IFN-γ best allows discrimination between tuberculous and malignant pleural effusion.Korean J Intern Med1997;12,225-231. [PubMed]
 
Ogawa, K, Koga, H, Hirakata, Y, et al Differential diagnosis of tuberculous pleurisy by measurement of cytokine concentrations in pleural effusion.Tuber Lung Dis1997;78,29-34. [PubMed]
 
Wongtim, S, Silachamroon, U, Ruxrungtham, K, et al Interferon γ for diagnosing tuberculous pleural effusions.Thorax1999;54,921-924. [PubMed]
 
Villegas, MV, Labrada, LA, Saravia, NG Evaluation of polymerase chain reaction, adenosine deaminase, and interferon-γ in pleural fluid for the differential diagnosis of pleural tuberculosis.Chest2000;118,1355-1364. [PubMed]
 
Yamada, Y, Nakamura, A, Hosoda, M, et al Cytokines in pleural liquid for diagnosis of tuberculous pleurisy.Respir Med2001;95,577-581. [PubMed]
 
Aoe, K, Hiraki, A, Murakami, T, et al Diagnostic significance of interferon-γ in tuberculous pleural effusions.Chest2003;123,740-744. [PubMed]
 
Villena, V, Lopez-Encuentra, A, Pozo, F, et al Interferon γ levels in pleural fluid for the diagnosis of tuberculosis.Am J Med2003;115,365-370. [PubMed]
 
Wong, CF, Yew, WW, Leung, SK, et al Assay of pleural fluid interleukin-6, tumor necrosis factor-α and interferon-γ in the diagnosis and outcome correlation of tuberculous effusion.Respir Med2003;97,1289-1295. [PubMed]
 
Poyraz, B, Kaya, A, Ciledag, A, et al Diagnostic significance of γ-interferon in tuberculous pleurisy.Tuberk Toraks2004;52,211-217. [PubMed]
 
Sharma, SK, Banga, A Diagnostic utility of pleural fluid IFN-γ in tuberculosis pleural effusion.J Interferon Cytokine Res2004;24,213-217. [PubMed]
 
El-Ansary, AK, Radwan, MA Evaluation of cytokines in pleural fluid for the differential diagnosis of tuberculous pleurisy.Biochim Clin2005;29,13-20
 
Gao, ZC, Tian, RX Clinical investigation on diagnostic value of interferon-γ, interleukin-12 and adenosine deaminase isoenzyme for tuberculous pleurisy.Chin Med J (Engl)2005;118,234-237. [PubMed]
 
Okamoto, M, Kawabe, T, Iwasaki, Y, et al Evaluation of interferon-γ, interferon-γ-inducing cytokines, and interferon-γ-inducible chemokines in tuberculous pleural effusions.J Lab Clin Med2005;145,88-93. [PubMed]
 
Sharma, SK, Banga, A Pleural fluid interferon-γ and adenosine deaminase levels in tuberculosis pleural effusion: a cost-effectiveness analysis.J Clin Lab Anal2005;19,40-46. [PubMed]
 
Morimoto, T, Takanashi, S, Hasegawa, Y, et al Level of antibodies against mycobacterial glycolipid in the effusion for diagnosis of tuberculous pleural effusion.Respir Med2006;100,1775-1780. [PubMed]
 
Barnes, PF, Fong, SJ, Brennan, PJ, et al Local production of tumor necrosis factor and IFN-γ in tuberculous pleuritis.J Immunol1990;145,149-154. [PubMed]
 
Barbosa, T, Arruda, S, Chalhoub, M, et al Correlation between interleukin-10 and in situ necrosis and fibrosis suggests a role for interleukin-10 in the resolution of the granulomatous response of tuberculous pleurisy patients.Microbes Infect2006;8,889-897. [PubMed]
 
Yanagawa, H, Yano, S, Haku, T, et al Interleukin-1 receptor antagonist in pleural effusion due to inflammatory and malignant lung disease.Eur Respir J1996;9,1211-1216. [PubMed]
 
Oshikawa, K, Yanagisawa, K, Ohno, S, et al Expression of ST2 in helper T lymphocytes of malignant pleural effusions.Am J Respir Crit Care Med2002;165,1005-1009. [PubMed]
 
Chen, YM, Yang, WK, Whang-Peng, J, et al An analysis of cytokine status in the serum and effusions of patients with tuberculous and lung cancer.Lung Cancer2001;31,25-30. [PubMed]
 
Oshikawa, K, Sugiyama, Y Elevated soluble CD26 levels in patients with tuberculous pleurisy.Int J Tuberc Lung Dis2001;5,868-872. [PubMed]
 
Kim, YK, Lee, SY, Kwon, SS, et al γ-interferon and soluble interleukin 2 receptor in tuberculous pleural effusion.Lung2001;179,175-184. [PubMed]
 
Okamoto, M, Hasegawa, Y, Hara, T, et al T-helper type 1/T-helper type 2 balance in malignant pleural effusions compared to tuberculous pleural effusions.Chest2005;128,4030-4035. [PubMed]
 
Villena, V, Lopez-Encuentra, A, Echave-Sustaeta, J, et al Interferon-γ in 388 immunocompromised and immunocompetent patients for diagnosing pleural tuberculosis.Eur Respir J1996;9,2635-2639. [PubMed]
 
Sharma, SK, Mitra, DK, Balamurugan, A, et al Cytokine polarization in miliary and pleural tuberculosis.J Clin Immunol2002;22,345-352. [PubMed]
 
Hiraki, A, Aoe, K, Matsuo, K, et al Simultaneous measurement of T-helper 1 cytokines in tuberculous pleural effusion.Int J Tuberc Lung Dis2003;7,1172-1177. [PubMed]
 
Hiraki, A, Aoe, K, Eda, R, et al Comparison of six biological markers for the diagnosis of tuberculous pleuritis.Chest2004;125,987-989. [PubMed]
 
Escudero Bueno, C, Garcia Clemente, M, Cuesta Castro, B, et al Cytologic and bacteriologic analysis of fluid and pleural biopsy specimens with Cope’s needle: study of 414 patients.Arch Intern Med1990;150,1190-1194. [PubMed]
 
Epstein, DM, Kline, LR, Albelda, SM, et al Tuberculous pleural effusions.Chest1987;91,106-109. [PubMed]
 
Valdes, L, Alvarez, D, San Jose, E, et al Tuberculous pleurisy: a study of 254 patients.Arch Intern Med1998;158,2017-2021. [PubMed]
 
Seibert, AF, Haynes, J, Jr, Middleton, R, et al Tuberculous pleural effusion: twenty-year experience.Chest1991;99,883-886. [PubMed]
 
Perez-Rodriguez, E, Jimenez Castro, D The use of adenosine deaminase and adenosine deaminase isoenzymes in the diagnosis of tuberculous pleuritis.Curr Opin Pulm Med2000;6,259-266. [PubMed]
 
Wong, PC Management of tuberculous pleuritis: can we do better?Respirology2005;10,144-148. [PubMed]
 
Glas, AS, Lijmer, JG, Prins, MH, et al The diagnostic odds ratio: a single indicator of test performance.J Clin Epidemiol2003;56,1129-1135. [PubMed]
 
Deeks, JJ Systematic reviews of evaluations of diagnostic and screening tests. Egger, M Smith, GD Altman, DG eds.Systematic reviews in health care: meta-analysis in context2001,248-282 BMJ Publishing Group. London, UK:
 
Jaeschke, R, Guyatt, G, Lijmer, J Diagnostic tests. Guyatt, G Rennie, D eds.Users’ guides to the medical literature: a manual for evidence-based clinical practice2002,121-140 AMA Press. Chicago, IL:
 
Petitti, DB Approaches to heterogeneity in meta-analysis.Stat Med2001;20,3625-3633. [PubMed]
 
Light, RW Establishing the diagnosis of tuberculous pleuritis.Arch Intern Med1998;158,1967-1968. [PubMed]
 

Figures

Figure Jump LinkFigure 1. Forest plot of estimates of sensitivity and specificity for IFN-γ assays in the diagnosis of TPE. • = point estimates of sensitivity and specificity from each study; error bars = 95% CIs; numbers = reference numbers of studies cited in the reference list. Pooled estimates for IFN-γ assay were as follows: sensitivity, 0.89 (95% CI, 0.87 to 0.91); specificity, 0.97 (95% CI, 0.96 to 0.98).Grahic Jump Location
Figure Jump LinkFigure 2. SROC curves for IFN-γ assays. • = each study in the metaanalysis (the size of each study is indicated by the size of the solid circle); dark line = weighted regression; and dashed line = unweighted regression. SROC curves summarize the overall diagnostic accuracy.Grahic Jump Location
Figure Jump LinkFigure 3. Funnel graph for the assessment of potential publication bias in IFN-γ assays. The funnel graph plots the log of the DOR against the SE of the log of the DOR (an indicator of sample size). • = each study in the metaanalysis; center line = SDOR. The result of the Egger test for publication bias was significant (p = 0.023).Grahic Jump Location

Tables

Table Graphic Jump Location
Table 1. Summary of Included Studies*
* 

ELISA = enzyme-linked immunosorbent assay; RIA = radioimmunoassay; TP = true-positive; FP = false-positive; FN = false-negative; TN = true-negative.

Table Graphic Jump Location
Table 2. Characteristics of Included Studies*
* 

TB = tuberculosis; Bac = bacteriology; His = histology; Clin = clinical course.

Table Graphic Jump Location
Table 3. Weighted Metaregression of the Effects of Methodological Quality and Study Design on Diagnostic Precision of Pleural IFN-γ in 22 Assays

References

. National Technical Steering Group of the Epidemiological Sampling Survey for Tuberculosis (2002) Report on fourth national epidemiological sampling survey of tuberculosis [in Chinese].Chin J Tuberc Respir Dis25,3-7
 
Dye, C Global epidemiology of tuberculosis.Lancet2006;367,938-940. [PubMed] [CrossRef]
 
Ellner, JJ, Barnes, PF, Wallis, RS, et al The immunology of tuberculous pleurisy.Semin Respir Infect1988;3,335-342. [PubMed]
 
Light, RW Pleural effusion.N Engl J Med2002;346,1971-1977. [PubMed]
 
Greco, S, Girardi, E, Masciangelo, R, et al Adenosine deaminase and interferon γ measurements for the diagnosis of tuberculous pleurisy: a meta-analysis.Int J Tuberc Lung Dis2003;7,777-786. [PubMed]
 
Bossuyt, PM, Reitsma, JB, Bruns, DE, et al Standards for reporting of diagnostic accuracy: towards complete and accurate reporting of studies of diagnostic accuracy; the STARD initiative.BMJ2003;326,41-44. [PubMed]
 
Whiting, P, Rutjes, AW, Reitsma, JB, et al The development of QUADAS: a tool for the quality assessment of studies of diagnostic accuracy included in systematic reviews.BMC Med Res Methodol2003;3,25. [PubMed]
 
Deville, WL, Buntinx, F, Bouter, LM, et al Conducting systematic reviews of diagnostic studies: didactic guidelines.BMC Med Res Methodol2002;2,9. [PubMed]
 
Moses, LE, Shapiro, D, Littenberg, B Combining independent studies of a diagnostic test into a summary ROC curve: data-analytic approaches and some additional considerations.Stat Med1993;12,1293-1316. [PubMed]
 
Lau, J, Ioannidis, JP, Balk, EM, et al Diagnosing acute cardiac ischemia in the emergency department: a systematic review of the accuracy and clinical effect of current technologies.Ann Emerg Med2001;37,453-460. [PubMed]
 
Irwig, L, Tosteson, AN, Gatsonis, C, et al Guidelines for meta-analyses evaluating diagnostic tests.Ann Intern Med1994;120,667-676. [PubMed]
 
Vamvakas, EC Meta-analyses of studies of the diagnostic accuracy of laboratory tests: a review of the concepts and methods.Arch Pathol Lab Med1998;122,675-686. [PubMed]
 
Suzuki, S, Moro-oka, T, Choudhry, NK The conditional relative odds ratio provided less biased results for comparing diagnostic test accuracy in meta-analyses.J Clin Epidemiol2004;57,461-469. [PubMed]
 
Westwood, ME, Whiting, PF, Kleijnen, J How does study quality affect the results of a diagnostic meta-analysis?BMC Med Res Methodol2005;8,20
 
Egger, M, Smith, G Davey, Schneider, M, et al Bias in meta-analysis detected by a simple, graphical test.BMJ1997;315,629-634. [PubMed]
 
Ribera, E, Ocana, I, Martinez-Vazquez, JM, et al High level of interferon γ in tuberculous pleural effusion.Chest1988;93,308-311. [PubMed]
 
Hsu, WH, Chiang, CD, Chen, WT, et al Diagnostic value of adenosine deaminase and γ-interferon in tuberculous and malignant pleural effusions.Taiwan Yi Xue Hui Za Zhi1989;88,879-882. [PubMed]
 
Shimokata, K, Saka, H, Murate, T, et al Cytokine content in pleural effusion: comparison between tuberculous and carcinomatous pleurisy.Chest1991;99,1103-1107. [PubMed]
 
Maeda, J, Ueki, N, Ohkawa, T, et al Local production and localization of transforming growth factor-β in tuberculous pleurisy.Clin Exp Immunol1993;92,32-38. [PubMed]
 
Valdes, L, San Jose, E, Alvarez, D, et al Diagnosis of tuberculous pleurisy using the biologic parameters adenosine deaminase, lysozyme, and interferon γ.Chest1993;103,458-465. [PubMed]
 
Aoki, Y, Katoh, O, Nakanishi, Y, et al A comparison study of IFN-γ, ADA, and CA125 as the diagnostic parameters in tuberculous pleuritis.Respir Med1994;88,139-143. [PubMed]
 
Soderblom, T, Nyberg, P, Teppo, AM, et al Pleural fluid interferon-γ and tumour necrosis factor-α in tuberculous and rheumatoid pleurisy.Eur Respir J1996;9,1652-1655. [PubMed]
 
Kim, YC, Park, KO, Bom, HS, et al Combining ADA, protein and IFN-γ best allows discrimination between tuberculous and malignant pleural effusion.Korean J Intern Med1997;12,225-231. [PubMed]
 
Ogawa, K, Koga, H, Hirakata, Y, et al Differential diagnosis of tuberculous pleurisy by measurement of cytokine concentrations in pleural effusion.Tuber Lung Dis1997;78,29-34. [PubMed]
 
Wongtim, S, Silachamroon, U, Ruxrungtham, K, et al Interferon γ for diagnosing tuberculous pleural effusions.Thorax1999;54,921-924. [PubMed]
 
Villegas, MV, Labrada, LA, Saravia, NG Evaluation of polymerase chain reaction, adenosine deaminase, and interferon-γ in pleural fluid for the differential diagnosis of pleural tuberculosis.Chest2000;118,1355-1364. [PubMed]
 
Yamada, Y, Nakamura, A, Hosoda, M, et al Cytokines in pleural liquid for diagnosis of tuberculous pleurisy.Respir Med2001;95,577-581. [PubMed]
 
Aoe, K, Hiraki, A, Murakami, T, et al Diagnostic significance of interferon-γ in tuberculous pleural effusions.Chest2003;123,740-744. [PubMed]
 
Villena, V, Lopez-Encuentra, A, Pozo, F, et al Interferon γ levels in pleural fluid for the diagnosis of tuberculosis.Am J Med2003;115,365-370. [PubMed]
 
Wong, CF, Yew, WW, Leung, SK, et al Assay of pleural fluid interleukin-6, tumor necrosis factor-α and interferon-γ in the diagnosis and outcome correlation of tuberculous effusion.Respir Med2003;97,1289-1295. [PubMed]
 
Poyraz, B, Kaya, A, Ciledag, A, et al Diagnostic significance of γ-interferon in tuberculous pleurisy.Tuberk Toraks2004;52,211-217. [PubMed]
 
Sharma, SK, Banga, A Diagnostic utility of pleural fluid IFN-γ in tuberculosis pleural effusion.J Interferon Cytokine Res2004;24,213-217. [PubMed]
 
El-Ansary, AK, Radwan, MA Evaluation of cytokines in pleural fluid for the differential diagnosis of tuberculous pleurisy.Biochim Clin2005;29,13-20
 
Gao, ZC, Tian, RX Clinical investigation on diagnostic value of interferon-γ, interleukin-12 and adenosine deaminase isoenzyme for tuberculous pleurisy.Chin Med J (Engl)2005;118,234-237. [PubMed]
 
Okamoto, M, Kawabe, T, Iwasaki, Y, et al Evaluation of interferon-γ, interferon-γ-inducing cytokines, and interferon-γ-inducible chemokines in tuberculous pleural effusions.J Lab Clin Med2005;145,88-93. [PubMed]
 
Sharma, SK, Banga, A Pleural fluid interferon-γ and adenosine deaminase levels in tuberculosis pleural effusion: a cost-effectiveness analysis.J Clin Lab Anal2005;19,40-46. [PubMed]
 
Morimoto, T, Takanashi, S, Hasegawa, Y, et al Level of antibodies against mycobacterial glycolipid in the effusion for diagnosis of tuberculous pleural effusion.Respir Med2006;100,1775-1780. [PubMed]
 
Barnes, PF, Fong, SJ, Brennan, PJ, et al Local production of tumor necrosis factor and IFN-γ in tuberculous pleuritis.J Immunol1990;145,149-154. [PubMed]
 
Barbosa, T, Arruda, S, Chalhoub, M, et al Correlation between interleukin-10 and in situ necrosis and fibrosis suggests a role for interleukin-10 in the resolution of the granulomatous response of tuberculous pleurisy patients.Microbes Infect2006;8,889-897. [PubMed]
 
Yanagawa, H, Yano, S, Haku, T, et al Interleukin-1 receptor antagonist in pleural effusion due to inflammatory and malignant lung disease.Eur Respir J1996;9,1211-1216. [PubMed]
 
Oshikawa, K, Yanagisawa, K, Ohno, S, et al Expression of ST2 in helper T lymphocytes of malignant pleural effusions.Am J Respir Crit Care Med2002;165,1005-1009. [PubMed]
 
Chen, YM, Yang, WK, Whang-Peng, J, et al An analysis of cytokine status in the serum and effusions of patients with tuberculous and lung cancer.Lung Cancer2001;31,25-30. [PubMed]
 
Oshikawa, K, Sugiyama, Y Elevated soluble CD26 levels in patients with tuberculous pleurisy.Int J Tuberc Lung Dis2001;5,868-872. [PubMed]
 
Kim, YK, Lee, SY, Kwon, SS, et al γ-interferon and soluble interleukin 2 receptor in tuberculous pleural effusion.Lung2001;179,175-184. [PubMed]
 
Okamoto, M, Hasegawa, Y, Hara, T, et al T-helper type 1/T-helper type 2 balance in malignant pleural effusions compared to tuberculous pleural effusions.Chest2005;128,4030-4035. [PubMed]
 
Villena, V, Lopez-Encuentra, A, Echave-Sustaeta, J, et al Interferon-γ in 388 immunocompromised and immunocompetent patients for diagnosing pleural tuberculosis.Eur Respir J1996;9,2635-2639. [PubMed]
 
Sharma, SK, Mitra, DK, Balamurugan, A, et al Cytokine polarization in miliary and pleural tuberculosis.J Clin Immunol2002;22,345-352. [PubMed]
 
Hiraki, A, Aoe, K, Matsuo, K, et al Simultaneous measurement of T-helper 1 cytokines in tuberculous pleural effusion.Int J Tuberc Lung Dis2003;7,1172-1177. [PubMed]
 
Hiraki, A, Aoe, K, Eda, R, et al Comparison of six biological markers for the diagnosis of tuberculous pleuritis.Chest2004;125,987-989. [PubMed]
 
Escudero Bueno, C, Garcia Clemente, M, Cuesta Castro, B, et al Cytologic and bacteriologic analysis of fluid and pleural biopsy specimens with Cope’s needle: study of 414 patients.Arch Intern Med1990;150,1190-1194. [PubMed]
 
Epstein, DM, Kline, LR, Albelda, SM, et al Tuberculous pleural effusions.Chest1987;91,106-109. [PubMed]
 
Valdes, L, Alvarez, D, San Jose, E, et al Tuberculous pleurisy: a study of 254 patients.Arch Intern Med1998;158,2017-2021. [PubMed]
 
Seibert, AF, Haynes, J, Jr, Middleton, R, et al Tuberculous pleural effusion: twenty-year experience.Chest1991;99,883-886. [PubMed]
 
Perez-Rodriguez, E, Jimenez Castro, D The use of adenosine deaminase and adenosine deaminase isoenzymes in the diagnosis of tuberculous pleuritis.Curr Opin Pulm Med2000;6,259-266. [PubMed]
 
Wong, PC Management of tuberculous pleuritis: can we do better?Respirology2005;10,144-148. [PubMed]
 
Glas, AS, Lijmer, JG, Prins, MH, et al The diagnostic odds ratio: a single indicator of test performance.J Clin Epidemiol2003;56,1129-1135. [PubMed]
 
Deeks, JJ Systematic reviews of evaluations of diagnostic and screening tests. Egger, M Smith, GD Altman, DG eds.Systematic reviews in health care: meta-analysis in context2001,248-282 BMJ Publishing Group. London, UK:
 
Jaeschke, R, Guyatt, G, Lijmer, J Diagnostic tests. Guyatt, G Rennie, D eds.Users’ guides to the medical literature: a manual for evidence-based clinical practice2002,121-140 AMA Press. Chicago, IL:
 
Petitti, DB Approaches to heterogeneity in meta-analysis.Stat Med2001;20,3625-3633. [PubMed]
 
Light, RW Establishing the diagnosis of tuberculous pleuritis.Arch Intern Med1998;158,1967-1968. [PubMed]
 
NOTE:
Citing articles are presented as examples only. In non-demo SCM6 implementation, integration with CrossRef’s "Cited By" API will populate this tab (http://www.crossref.org/citedby.html).

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging & repositioning the boxes below.

CHEST Journal Articles
PubMed Articles
  • CHEST Journal
    Print ISSN: 0012-3692
    Online ISSN: 1931-3543