Improving the yield of diagnostic medical thoracoscopy for undiagnosed exudative pleural effusions using a rigid diagnostic algorithm

Jolsana Augustine; Anand Vijay; Divya Ramachandran; Melcy Cleetus; Aparna Soman Nirmal; Susan John; Sunitha Thomas; Rajesh Venkitakrishnan

doi:10.4103/ijmy.ijmy_214_21

ORIGINAL ARTICLE

Year : 2021 | Volume : 10 | Issue : 4 | Page : 405-410

Improving the yield of diagnostic medical thoracoscopy for undiagnosed exudative pleural effusions using a rigid diagnostic algorithm

Jolsana Augustine¹, Anand Vijay¹, Divya Ramachandran¹, Melcy Cleetus¹, Aparna Soman Nirmal¹, Susan John², Sunitha Thomas³, Rajesh Venkitakrishnan¹
¹ Department of Pulmonary Medicine, Rajagiri Hospital, Kochi, Kerala, India
² Department of Clinical Epidemiologist, Rajagiri Hospital, Kochi, Kerala, India
³ Department of Pathology, Rajagiri Hospital, Kochi, Kerala, India

Date of Submission	10-Sep-2021
Date of Decision	10-Oct-2021
Date of Acceptance	19-Oct-2021
Date of Web Publication	13-Dec-2021

Correspondence Address:
Rajesh Venkitakrishnan
Rajagiri Hospital, Kochi, Kerala
India

Source of Support: None, Conflict of Interest: None

Check

DOI: 10.4103/ijmy.ijmy_214_21

Abstract

Background: Establishing the etiology of exudative pleural effusions in the setting of an unrewarding pleural fluid analysis often requires biopsies from the parietal pleura. However, it may be noted that diagnosis such as pulmonary embolism and connective tissue diseases can result in an exudative pleural effusion where a pleural biopsy can yield nonspecific results. Medical thoracoscopy (MT) is a minimally invasive procedure performed under local anesthesia or moderate sedation with excellent yield and favorable safety profile. We analyzed the diagnostic yield of MT for exudative pleural effusions after employing a rigid diagnostic algorithm. The study was undertaken to ascertain the yield of MT in establishing the diagnosis in diagnosis of exudative pleural effusions, to find out the relative contribution of pleural tuberculosis (TB) as a cause of undiagnosed exudative pleural effusion, to describe the etiology of undiagnosed exudative pleural effusion in patients undergoing MT and to determine the correlation between pleural fluid adenosine deaminase (ADA) levels and TB pleuritis in patients undergoing MT. Methods: This was a retrospective study. Patients with undiagnosed exudative pleural effusion were included in the study. MT was performed with semirigid thoracoscope (Olympus LTF 160) under local anesthesia and conscious sedation. Gross appearance and ADA level of pleural fluid were noted. Pleural biopsy material was subjected to histopathology examination and culture for mycobacteria along with cartridge-based nucleic acid amplification test for TB. The yield of MT for establishing the etiology of pleural effusion and the relative contribution of tuberculous pleuritis as a cause of undiagnosed pleural effusion was ascertained. Correlation of pleural fluid ADA levels was done with a final diagnosis of TB pleuritis in patients undergoing MT. Results: Twenty-five patients with undiagnosed exudative pleural effusion underwent thoracoscopy of which 16 were male and 9 were female. MT was able to establish the diagnosis in all cases, providing a diagnostic yield of 100%. Histopathological examination of biopsy specimens yielded a diagnosis of malignant involvement of pleura in 10 patients and granulomatous pleuritis consistent with TB in 14 patients. Pleural TB contributed to 60% of undiagnosed pleural effusions in the present study. The mean ADA value among those who turned positive was 56.338 and 35.300 among those who turned negative using genexpert, which was found to be statistically significant. A value of 31 IU/L showed a sensitivity of 93.3% and specificity of 99.8% and hence can be taken as a cut off value for the diagnosis of pleural TB based on receiver-operating characteristic analysis. Conclusion: TB contributed to 60% of undiagnosed exudative pleural effusions in the present study. MT had 100% yield in the diagnosis of undiagnosed exudative effusions. Pleural fluid ADA levels may help in differentiating TB versus malignant effusion.

Keywords: Adenosine deaminase levels, medical thoracoscopy, pleural tuberculosis, undiagnosed exudative effusion

How to cite this article:
Augustine J, Vijay A, Ramachandran D, Cleetus M, Nirmal AS, John S, Thomas S, Venkitakrishnan R. Improving the yield of diagnostic medical thoracoscopy for undiagnosed exudative pleural effusions using a rigid diagnostic algorithm. Int J Mycobacteriol 2021;10:405-10

How to cite this URL:
Augustine J, Vijay A, Ramachandran D, Cleetus M, Nirmal AS, John S, Thomas S, Venkitakrishnan R. Improving the yield of diagnostic medical thoracoscopy for undiagnosed exudative pleural effusions using a rigid diagnostic algorithm. Int J Mycobacteriol [serial online] 2021 [cited 2022 Jan 21];10:405-10. Available from: https://www.ijmyco.org/text.asp?2021/10/4/405/332362

Introduction

Pleural effusion is a frequently encountered entity in pulmonary practice and can be due to varying diseases with different pathogenic mechanisms. Low oncotic pressure (e.g., in hypoalbuminemia), elevated pulmonary capillary pressure, increased permeability, lymphatic obstruction, and diminished negative intrapleural pressure are main pathophysiological processes that can cause pleural effusion.^[1]

Establishing the etiology of pleural effusion is of paramount importance, as the treatment and prognosis of pleural effusion largely depend on its cause. Medical thoracoscopy (MT) is a minimally invasive and relatively safe procedure which helps the interventional pulmonologist to directly enter the pleural cavity and take pleural samples under direct visualization. MT is usually performed when multiple attempts at thoracentesis (typically two) have failed to achieve a diagnosis in patients with a persistent exudative pleural effusion. This approach has been endorsed by the 2010 British Thoracic Society Pleural Disease Guideline^[2],[3] and is based upon the high diagnostic accuracy of thoracoscopic biopsy in this setting compared with closed needle- or image-guided biopsy.^[4] It may be noted that diseases such as rheumatoid arthritis and systemic lupus erythematosus (SLE) can result in exudative pleural effusions and a thorough extrapulmonary clinical evaluation coupled with serological testing holds the key for successful diagnosis as opposed to MT and pleural biopsy. In similar lines, pulmonary embolism can result in undiagnosed exudative effusion and MT can have normal pleural findings. Hence, these possibilities need to be carefully excluded in indicated cases by appropriate investigations like computed tomography (CT) pulmonary angiogram before contemplating on an invasive investigation like MT. Viral infections can also be an etiology to persistent, subacute self-limiting exudative effusions with nonspecific findings on pleural biopsy.

Pleural malignancy (either primary or metastatic) and pleural tuberculosis (TB) are the two commonest causes of undiagnosed exudative pleural effusion^{[5],[6],[7],[8]} most Indian studies reveal that pleural TB contributes to 12%–30% of undiagnosed pleural effusions needing MT.^{[9],[10],[11],[12]} The yield of MT in establishing the etiology of undiagnosed effusion ranges between 66% and 97% in these studies. The diagnosis of pleural TB can be achieved in 99% of patients with thoracoscopy, which is higher than the 51% yield for closed pleural biopsy.^[13] Similarly, yield of thoracoscopic pleural biopsy is higher in patients with suspected pleural malignancy. A diagnosis could be achieved in 90% of patients as against 61% patients using closed pleural biopsy.^[14] We postulated that the need for MT may be brought down and yield increased by excluding conditions such as connective tissue diseases and pulmonary embolism before proceeding with MT.

Methodology

The present study was a retrospective study based on review of hospital records. The study was conducted in the Department of Pulmonary Medicine, Rajagiri Hospital, Kochi, Kerala, a tertiary care institution.

Objectives of the study

The present study was conducted with the following objectives:

To know the overall diagnostic yield of MT in exudative pleural effusions of obscure cause
To determine relative incidence of pleural TB as a cause of undiagnosed exudative pleural effusion.
To unravel the etiology of pleural effusion in patients undergoing MT
To determine whether any association exists between pleural fluid adenosine deaminase (ADA) levels and histopathologic presence of granulomas in cases of pleural TB diagnosed by MT
To describe whether any association exists between pleural fluid color and pleural appearance on the final histopathological diagnosis of pleural TB versus malignancy.

Selection of study participants and inclusion criteria

Patients with undiagnosed exudative pleural effusions were considered for the study. The study period spanned from March 2018 to December 2020. Data were collected retrospectively from in-patient records. All subjects who underwent MT for undiagnosed exudative pleural effusions were recruited. A working definition of undiagnosed exudative pleural effusion was made based on (all criteria need to be present).

Exudative pleural fluid as per Light's criteria
No evidence of lobar consolidation, mass lesion, or cavitory lesion in chest radiograph or CT chest
No evidence of pulmonary thromboembolism in a CT pulmonary angiogram
Pleural fluid for bacterial culture and TB (acid-fast bacilli [AFB] smear and cartridge-based nucleic acid amplification test [CB-NAAT]) negative
Pleural fluid cytology and cell block for malignant cells negative on two separate pleural fluid samples.

Exclusion criteria

The following patients were excluded from the study:

Patients <40 years of age with exudative pleural effusions, pleural fluid ADA levels >60 and mantoux test positivity (>10 mm induration in transverse dimensions). These patients were initiated on empirical anti-TB chemotherapy and were serially followed up for response
Patients with exudative effusion but with microbial confirmation of TB (smear or CB-NAAT) on sputum or other lower respiratory tract specimen
Previous diagnosis or current clinical evidence of an active rheumatological disease potentially accountable for effusion. A rheumatologist's evaluation, and if deemed appropriate serological tests for rheumatic diseases were carried out on a case-to-case basis.

Sample size

Since the inclusion and exclusion criteria were more stringent than previous studies conducted with similar objectives, only small number of patients fulfilled the study entry requisites. Hence, all patients who fulfilled the inclusion criteria were recruited into the study.

Study methods and data entry

Patients who satisfied the inclusion criteria within the study period were recruited into the study. Demographic details, comorbid conditions, baseline blood reports (Complete blood count, glycemic status, renal function, hepatic function, prothrombin time international normalized ratio, erythrocyte sedimentation rate), and Mantoux reactivity of study patients were noted down from electronic medical records and case files of patients. Chest radiographs and CT chest images were reviewed by one member of the radiology team and at least two members of the treating pulmonology team. The interpretations were noted down. Gross appearance of pleural fluid and test reports were entered into study pro forma. The pleural fluid tests included biochemical parameters (Pleural fluid protein, albumin, sugar, lactate dehydrogenase), total and differential cell count, microbial investigations (bacterial culture, Xpert MTB/RIF), and ADA levels. Two separate samples for cytology and cell block were performed serially to look for any evidence of malignant cells. MT was performed with semirigid thoracoscope (Olympus LTF 160) under local anesthesia topped up with conscious sedation with fentanyl and midazolam. At least 5 biopsy samples were taken from abnormal appearing areas. The descriptive appearance of pleural surfaces at the time of MT and results of tests done on pleural biopsy material were noted down.

Data presentation and statistical analysis

Data analysis was done using IBM SPSS v. 25 (IBM Corporation, Armonk, N.Y.). For the description of the continuous variables, methods of descriptive statistics with the calculation of average and standard deviation (SD) were used. Qualitative variables were expressed as proportion or percentage. Quantitative variables were expressed as mean ± SD Association between two groups were assessed using Chi-square test for qualitative variables and Mann–Whitney test (nonparametric) for quantitative variables due to small sample size (<30). The level of statistical significance was considered to be P < 0.05.

Ethical committee approval

Approval to conduct the study was obtained from the hospital ethical committee. The anonymity of study participants was ensured. Considering the retrospective nature of the study, the need for informed patient consent was waived off by the ethical committee.

Results

A total of 62 patients with undiagnosed exudative pleural effusions were encountered in the study period after pleural fluid analysis. However, after a CT thorax with CT pulmonary angiogram and rheumatology evaluation, a diagnosis could be arrived at in 32 of these patients. Fourteen patients had pulmonary thromboembolism, 10 patients had consolidation patches that were not evident in chest radiograph and eight patients had features of connective tissue disease (rheumatoid arthritis in 6 and SLE in 2). Thirty patients had persistent undiagnosed pleural effusion. Thus, our rigid diagnostic algorithm was able to decrease the need for MT in over 50% of undiagnosed exudative effusions.

A total of 30 patients presented with undiagnosed exudative pleural effusion during the study period of which 25 underwent MT. Five patients who opted out of procedure were managed with intercostal drainage and conservative measures. Out of 25 patients, 16 (64%) were male and 9 were female. The mean age of patients who underwent procedure was 54.4 with a standard deviation of 16.8 (minimum age 21 years, maximum 75 years). Demographic details and personal characteristics of the study population are shown in [Table 1].

Table 1: Demographic details of study population

Click here to view

Seventeen out of the 25 patients (68%) who underwent MT had right-sided effusion. Pleural fluid was straw-colored in 15 (60%) patients and hemorrhagic in the rest. Visual inspection of the pleural surfaces during MT showed nodules, exudative lesions, and septations in both TB and malignancies and no significant association was noted.

A final diagnosis was obtained in all the patients who underwent MT, thus providing a diagnostic yield of 100%. Pleural TB contributed to 60% (15 out of 25) of undiagnosed exudative effusions. Rest 40% of effusions (10 out of 25) were due to malignant disease involving the pleura. [Figure 1] shows the relative contribution of different diseases to the etiology of pleural effusion.

Figure 1: Pie chart showing diagnosis obtained form thoracoscopic evalution

Click here to view

Histopathological examination (HPE) of thoracoscopically performed biopsy samples demonstrated features suggestive of malignancy in 10 patients (40%). Granulomas were demonstrated in 14 patients (56%) suggestive of tuberculous involvement of pleura. One patient had features of xanthomatous pleuritis but had CB-NAAT positivity in pleural tissue. Microbial confirmation of TB by CB-NAAT testing (of pleural tissue and pleural fluid combined) could be made in only 8 of the 14 patients (57%) who had granulomas. Six patients had granulomas on HPE but pleural fluid and tissue for CB-NAAT were negative. In the 8 CB-NAAT positive cases, histopathology features of granulomatous inflammation were seen in all patients except one (87.5%), who had features of xanthomatous pleuritis.

Mantoux reactivity was seen in 17 (68%) of the study population. Among the 17 patients with Mantoux positivity, 15 (88.2%) had granulomas on HPE and none of the negative patients showed the presence of granulomas (P < 0.01). It may be noted that two patients who had features of malignant involvement of pleura also exhibited mantoux positivity. [Table 2] summarizes the clinical aspects and investigation results of study patients.

Table 2: Clinical aspects and investigation results of study patients

Click here to view

No significant association between gross pleural appearance and histopathology were noted. In patients whose histopathology showed granuloma, 80% had a straw-colored pleural effusion and in those with malignancy 70% had hemorrhagic effusion (P = 0.018.). The mean length of hospital stay was 8.53 days (SD 3.662) and 9.60 (SD 2.86) in the granuloma group and malignancy groups, respectively. A mean difference of 1.067 days was noted between these groups but it was not statistically significant. Seventy percent smokers had evidence of malignancy in HPE. Among nonsmokers 76.5% had granuloma in biopsy. There was a significant statistical correlation between histopathology and smoking history (P value 0.022). [Table 3] summarizes the statistical association between various parameters and etiology of pleural effusion.

Table 3: Relationship between etiology of pleural effusion and clinical/investigation parameters

Click here to view

Among the patients with a final diagnosis of TB after MT, mean ADA level was 53.140 ± (15.199) IU/L whereas in the malignancy group mean ADA level was 25.370 (±11.066) IU/L. The difference in the mean of ADA between patients with TB versus malignant effusions was 27.77. This difference in mean was statistically significant (P < 0.05). A receiver-operating characteristic (ROC) curve [Figure 2] was constructed to find out the optimum cut-off of pleural fluid ADA values to diagnose pleural TB.

On plotting the ROC curve, area under the curve was obtained as 93.7% (95% confidence interval – 84.2%–100%) which was statistically significant. A value of 31 IU/L showed a sensitivity of 93.3% and specificity of 99.8% and can be taken as a cut off value.

Figure 2: Receiver-operating characteristic curve to determine the sensitivity and specificity of pleural fluid Adenosine deaminase level to diagnose tuberculosis pleuritis

Click here to view

Discussion

The first step in the management of exudative pleural effusions is to determine the etiology, which is of paramount importance in prognostication and treatment planning. Exudative effusions have varying etiology including infections, malignancies, connective tissue diseases, and pulmonary embolism. The etiology remains obscure in some effusions after an initial clinical evaluation, pleural fluid analysis, and chest radiograph. MT is a valuable tool for establishing the etiology of exudative effusions which remain undiagnosed after the initial evaluation.

We adopted a stringent inclusion criteria to subject our patients to MT, including the performance of CT chest with CT pulmonary angiogram, rheumatology evaluation, and serological markers for connective tissue diseases as deemed necessary. Of the total of 62 patients with undiagnosed exudative pleural effusions after pleural fluid analysis, a diagnosis could be arrived at in 32 of these patients after a CT thorax with CT pulmonary angiogram and rheumatology evaluation. Fourteen patients had pulmonary thromboembolism, 10 patients had consolidation patches that were not evident in chest radiograph and eight patients had features of connective tissue disease (rheumatoid arthritis in 6 and SLE in 2). Thus, our rigid diagnostic algorithm was able to decrease the need for MT in over 50% of undiagnosed exudative effusions.

Our study had an excellent diagnostic yield of MT establishing a diagnosis in all patients subjected to the test. The yield quoted in Indian studies varies between 66% and 97%. We carefully selected our patients avoiding patients with connective tissue disease, pulmonary embolism, and slow resolving pneumonia with effusion, which might otherwise have led to histopathology showing only nonspecific pleuritis and nonrewarding diagnosis. TB contributed to 60% of cases in the present study, which is higher than in most other Indian studies. We opted against empirical anti TB treatment in exudative lymphocytic effusions with high ADA, unless the patient was <40 years of age and had a positive mantoux. This might account for the relatively high proportion of patients in our series having pleural TB. The incidence of TB reported by other studies with thoracoscopicpleural biopsy in TB endemic populations have been 52.4% (Kannan et al.^[15]) and 22.8% (Mootha et al.^[5] A Chinese study^[16] revealed 92% yield for MT with TB contributing to 39.9% cases. [Table 4] summarizes the yield of thoracoscopy and relative proportion of TB with various Indian studies.

Table 4: Yield of thoracoscopy and relative proportion of tuberculosis in studies

Click here to view

ADA, an enzyme produced from lymphocytes and involved in purine metabolism, has been extensively studied as a biochemical marker in pleural fluid for tubercular pleuritis.^[17] Despite its widespread use, there is still no clarity on an optimum discriminatory threshold value. Its diagnostic utility varies considerably across different geographical regions, clinical settings, and test result thresholds. In general, a positive test improves posttest probability of tuberculous effusion to a much lower extent in high prevalence than in low prevalence settings, and a negative test performs much better at excluding TPE in low prevalence settings. Our study detected a mean ADA level of 53.140 in patients with TB as opposed to a value of 25.370 in subjects with malignant effusions, which indicated a statistically significant association. Despite four decades of research, the optimum pleural fluid ADA cut-off for the diagnosis of TPE remains unclear. In general, ADA levels higher than 70 IU/L are highly suggestive of TBE, whereas levels less than 40 IU/L are more helpful in excluding disease. A meta-analysis published in 2019^[18] looked at 174 publications and 27009 patients. Studies with ADA threshold of 40 ± 4 IU/L showed summary sensitivity and specificity of 0.93 and 0.90, respectively. Four studies with ADA threshold >65 IU/L showed summary sensitivity and specificity of 0.86 and 0.94, respectively. ADA values need to be coupled with clinical judgment, imaging findings, and other ancillary tests to make a noninvasive diagnosis of tuberculous pleuritis. In addition, one case of xanthomatous pleuritis as a manifestation of pleural TB was encountered in the present study, which has been shared by us as a previous separate case report.^[19]

Specific challenges in the diagnosis of TB pleuritis in pediatric population deserve special mention. In childhood, a definitive diagnosis may not be possible because of low bacillus load or the patients' inability to provide adequate sputum specimens. The pleural fluid bacilli load also tends to be very low, generally resulting in negative AFB staining and mycobacterium culture. In one pediatric study,^[20] in lymphocytepredominant pleural effusion, an increased ADA level highly supported TB disease. Exudative effusion was not the rule in pediatric pleural TB and transudates were occasionally encountered. The combination of microbiological results and histopathology examination of the pleural biopsy specimens is essential for the diagnosis of tuberculous pleurisy, as microbiological examination of pleural biopsy specimens has proved to have a higher detection rate than pleural fluid examination. Furthermore, cytocentrifugation improved the sensitivity of microscopic detection for both pleural fluid and biopsy specimens.^[21]

The advances in metabolomics may assist in the differential diagnosis of diseases. This technique is based on the modern biochemioinformatics technique to identify specific metabolome for discrimination between different clinical disorders. It has also been applied for TB. Previous authors^[22] have assessed the effect of background diabetes mellitus on the metabolome in tuberculous effusion. Significant confounding and overlapping effect of citric acid in diabetes to tuberculous effusion has been revealed. The impact of this finding in day-to-day practice in pleural disease remains to be elucidated.

One aspect of our study deserves special mention. We kept stringent diagnostic criteria for making a working label of “undiagnosed exudative pleural effusion” and proceeding to MT. Patients with CT chest findings of mass, nodules, etc., were subjected to CT-guided biopsy or bronchoscopy and MT was done only if these investigations were negative. A CT pulmonary angiogram was done to rule out thromboembolic phenomena as a cause of the effusion. The presence of connective tissue disease as an etiology to effusion was reasonably ruled out with clinical evaluation by a rheumatologist and serological evaluation with RA factor and ANA screening. These measures reduced the number of undiagnosed cases of effusion from an initial number of 62–30, thereby obviating the need for an invasive diagnostic procedure like MT. The same approach would have limited the possibilities to malignant effusions and tuberculous pleuritis, the two of which accounted for all etiologies in our case series. The study also goes along to educate us that focused evaluation with less invasive investigations in exudative effusions would clinch the diagnosis in many cases and avoid the need for MT.

Limitations of the study

Our study has several limitations. This study was done in a single center that caters to a limited geographic territory in central Kerala. The number of study subjects was small, probably because we had kept very stringent criteria for labeling as undiagnosed exudative pleural effusion. The spectrum of pleural diseases that has been revealed in our study has been very limited and is restricted to pleural malignancy and pleural TB.

Conclusion and Recommendations

Pleural TB contributed to 60% of undiagnosed exudative pleural effusions in the present study. MT had a diagnostic yield of 100% in the present study. Tuberculous pleuritis and malignant pleural effusions accounted for all undiagnosed pleural effusions in the present study. On gross appearance, TB-related pleural effusion was more often straw-colored whereas malignant pleural effusion was frequently hemorrhagic. Specific patterns of pleural involvement during MT were not useful in differentiating malignant versus tuberculous pleural effusions. Pleural fluid ADA was found to be high in tubercular effusions with an optimum cut-off of 31 IU/L.

Ethical clearance

Ethical clearance was obtained from the institutional ethics committee [Study Reference Number: RAJH/A/2021/003].

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.

References

1.	Jany B, Welte T. Pleural effusion in adults-etiology, diagnosis, and treatment. Dtsch Arztebl Int 2019;116:377-86.
2.	Rahman NM, Ali NJ, Brown G, Chapman SJ, Davies RJ, Downer NJ, et al. Local anaesthetic thoracoscopy: British thoracic society pleural disease guideline 2010. Thorax 2010;65 Suppl 2:i54-60.
3.	Hooper C, Lee YC, Maskell N; BTS Pleural Guideline Group. Investigation of a unilateral pleural effusion in adults: British thoracic society pleural disease guideline 2010. Thorax 2010;65 Suppl 2:i4-17.
4.	Agarwal R, Aggarwal AN, Gupta D. Diagnostic accuracy and safety of semirigid thoracoscopy in exudative pleural effusions: A meta-analysis. Chest 2013;144:1857-67.
5.	Mootha V, Agarwal R, Singh N, Agarwal A, Gupta D, Jindal S. Medical thoracoscopy for undiagnosed pleural effusions: Experience from a tertiary care hospital in North India. Indian J Chest Dis Allied Sci 2010;53:21-4.
6.	Nattusamy L, Madan K, Mohan A, Hadda V, Jain D, Madan NK, et al. Utility of semi-rigid thoracoscopy in undiagnosed exudative pleural effusion. Lung India 2015;32:119-26. [PUBMED] [Full text]
7.	Maturu VN, Dhooria S, Bal A, Singh N, Aggarwal AN, Gupta D, et al. Role of medical thoracoscopy and closed-blind pleural biopsy in undiagnosed exudative pleural effusions: A single-center experience of 348 patients. J Bronchology Interv Pulmonol 2015;22:121-9.
8.	Haridas N, Suraj KP, Rajagopal TP, James PT, Chetambath R. Medical thoracoscopy vs closed pleural biopsy in pleural effusions: A randomized controlled study. J Clin Diagn Res 2014;8:C01-4.
9.	Thangakunam B, Christopher DJ, James P, Gupta R. Semi-rigid thoracoscopy: Initial experience from a tertiary care hospital. Indian J Chest Dis Allied Sci 2010;52:25-7.
10.	Mehta A, Venkitakrishnan R, Viswam D, Patel V, Babu S, Indira KS, et al. Value of semirigid thoracoscopy in pleural effusion. Pulmon 2010;12:43-5.
11.	Prabhu VG, Narasimhan R. The role of pleuroscopy in undiagnosed exudative pleural effusion. Lung India 2012;29:128-30. [Full text]
12.	Dhooria S, Singh N, Aggarwal AN, Gupta D, Agarwal R. A randomized trial comparing the diagnostic yield of rigid and semirigid thoracoscopy in undiagnosed pleural effusions. Respir Care 2014;59:756-64.
13.	Loddenkemper R. Thoracoscopy-State of the art. Eur Respir J 1998;11:213-21.
14.	McLean AN, Bicknell SR, McAlpine LG, Peacock AJ. Investigation of pleural effusion: An evaluation of the new Olympus LTF semiflexible thoracofiberscope and comparison with Abram's needle biopsy. Chest 1998;114:150-3.
15.	Kannan SK, Lin WJ, Teck TS, Azizi AR. Pleuroscopy: Early experience in an East Malaysian state with high tuberculosis prevalence. J Bronchology Interv Pulmonol 2009;16:250-3.
16.	Wang XJ, Yang Y, Wang Z, Xu LL, Wu YB, Zhang J, et al. Efficacy and safety of diagnostic thoracoscopy in undiagnosed pleural effusions. Respiration 2015;90:251-5.
17.	Mehta AA, Gupta AS, Ahmed S, Rajesh V. Diagnostic utility of adenosine deaminase in exudative pleural effusions. Lung India 2014;31:142-4. [PUBMED] [Full text]
18.	Aggarwal AN, Agarwal R, Sehgal IS, Dhooria S. Adenosine deaminase for diagnosis of tuberculous pleural effusion: A systematic review and meta-analysis. PLoS One 2019;14:e0213728.
19.	Augustine J, Venkitakrishnan R, Ramachandran D, Abraham L. Xanthogranulomatous pleuritis-An unusual presentation of tuberculosis. Int J Mycobacteriol 2020;9:442-4. [PUBMED] [Full text]
20.	Bayhan GI, Sayir F, Tanir G, Tuncer O. Pediatric pleural tuberculosis. Int J Mycobacteriol 2018;7:261-4. [PUBMED] [Full text]
21.	Amer S, Hefnawy AE, Wahab NA, Okasha H, Baz A. Evaluation of different laboratory methods for rapid diagnosis of tuberculous pleurisy. Int J Mycobacteriol 2016;5:437-45. [Full text]
22.	Sookaromdee P, Wiwanitkit V. Tuberculous pleural effusion: Modification of metabolome by the effect of common metabolic disease, diabetes mellitus. Biomed Biotechnol Res J 2019;3:19-21. [Full text]