Pleural fluid lysozyme as a diagnostic biomarker of pleural tuberculosis: A systematic review and meta-analysis : Lung India

Secondary Logo

Journal Logo

Original Article

Pleural fluid lysozyme as a diagnostic biomarker of pleural tuberculosis

A systematic review and meta-analysis

Aggarwal, Ashutosh Nath; Agarwal, Ritesh; Dhooria, Sahajal; Prasad, Kuruswamy Thurai; Sehgal, Inderpaul Singh; Muthu, Valliappan

Author Information
doi: 10.4103/lungindia.lungindia_738_21
  • Open

Abstract

INTRODUCTION

Tuberculosis (TB) is a common cause of exudative pleural effusion, particularly in regions with a high TB burden. A definitive diagnosis of tuberculous pleural effusion (TPE) requires demonstration of mycobacteria in pleural fluid (by nucleic acid amplification methods, microscopy, or culture), or documentation of granulomatous inflammation on pleural biopsy. The yield from microbiological testing is quite suboptimal, whereas the latter is invasive and not widely performed.[1] Therefore, physicians use surrogate laboratory biomarkers for initiating empiric anti-tuberculous therapy (ATT) among those with suspected TPE. Pleural fluid adenosine deaminase (ADA) is one such commonly used investigation having good sensitivity and specificity.[2] More recently, pleural fluid interferon-gamma levels too have shown good accuracy for identifying TPE.[3] However, none of these tests is a perfect discriminator, and there is an unmet need to identify other biomarkers for pleural TB.

Lysozyme is a low molecular weight bacteriolytic protein distributed in several body fluids and passively enters the pleural space through blood. Activated macrophages in tuberculous granulomas actively secrete lysozyme into the pleural fluid in patients with TPE, and both pleural fluid lysozyme levels (LP) and pleural fluid to serum lysozyme ratio (LP/LS) are thus greater in TPE than in other effusions.[4] However, lysozyme assays are poorly automated and time-consuming, and different studies report significant variability in diagnostic accuracy. Therefore, although the test is considered useful in differentiating tuberculous from non-tuberculous pleural effusions, it has not still been widely adopted.[4] Recent proteomics studies on pleural fluid and pleural biopsy samples, however, suggest significantly greater expression of lysozyme precursor in TPE compared to other pleural effusions.[56] Higher LP levels in patients with TPE also correlate with residual pleural thickening.[7] We, therefore, conducted a systematic review and meta-analysis to evaluate the utility of LP and LP/LS estimation in the diagnosis of TPE. We also specifically explored if LP or LP/LS could differentiate TPE from parapneumonic or malignant pleural effusions. Both these disorders are frequent diagnostic considerations during the assessment of pleural effusions suspected to be due to TB.

METHODS

We registered our systematic review and meta-analysis protocol with the PROSPERO database (registration number CRD42021287632) and followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines for this review.[89] An approval was not required from our Ethics Committee as we used only summary data from studies already published.

Search strategy

We explored the PubMed and Embase electronic databases on October 31, 2021, using the following free-text search terms: (lysozyme, or muramidase); (tuberculosis, tubercular, tuberculous, TB, Mycobacterium, or mycobacterial); and (pleura, pleural, pleuritis, or pleurisy).

Study selection

After excluding duplicate records, two reviewers (ANA and RA) screened all titles and abstracts retrieved through the search process. We excluded publications in non-English languages and studies not focused on pleural TB. We also excluded review articles, conference abstracts, case reports, letters to editors not describing original data, and editorials. Full texts of all articles judged potentially eligible were then retrieved for independent evaluation by both reviewers.

We included a study for data synthesis if it (a) included patients with TPE and at least one other cause for exudative pleural effusion, (b) employed a microbiologic (presence of acid-fast bacilli, or positivity for Mycobacterium tuberculosis on nucleic acid amplification tests or culture, in pleural fluid, pleural biopsy, or another clinical specimen), histopathologic (pleural biopsy demonstrating granulomatous inflammation), and/or a clinical (compatible clinical profile with adequate resolution of effusion after empiric anti-tubercular treatment) reference standard for diagnosing TPE, and (c) provided numerical data for calculating both sensitivity and specificity of LP or LP/LS for diagnosis of TPE, or provided measures of central tendency (mean or median) and dispersion (standard deviation [SD], or interquartile range [IQR], or range) of LP levels or LP/LS in patients with TPE and other pleural effusions. If the same patient population was evaluated in two or more studies, only the one assessing the largest dataset was selected. In case of any disagreement, study inclusion was decided through consensus between the two reviewers.

Data extraction

We extracted the following data from the studies finally eligible for inclusion: study design, year of publication, countries where the studies were performed, the etiology of non-tuberculous pleural effusions, human immunodeficiency virus (HIV) status, lysozyme assay method and its threshold, blinding, the proportion of TPE patients diagnosed using microbiologic or pathologic criteria (referred to hereafter as having “definite TB”), number of subjects in each group, the number of positive and negative assay results for each category of subjects, and the mean and SD of pleural fluid lysozyme for tuberculous, malignant, and parapneumonic effusions. If data dispersion was expressed as a range of values, or as a standard error of the mean, we approximated the SD assuming a normal distribution of data.[10]

Assessment of study quality

We graded the methodological quality of studies reporting on diagnostic accuracy through the QUADAS-2 (Quality Assessment of Diagnostic Accuracy Studies, version 2) tool.[11] We employed the Newcastle–Ottawa Scale to assess the methodological quality of studies describing differences in LP levels or LP/LS between TPE and other effusions. Any study with a score of at least 7 (out of a maximum possible score of 9) was judged as having good quality.[12]

Statistical analysis

We computed sensitivity, specificity, and diagnostic odds ratio (DOR) for all studies reporting on diagnostic accuracy data. We calculated the 95% confidence interval (95% CI) for each study using the Clopper–Pearson approach.[13] We applied a continuity correction of 0.5 before any logarithmic or logit transformation in studies describing zero cell frequencies. We used the Rutter and Gatsonis hierarchical model to pool diagnostic accuracy data across these studies.[14]

For studies comparing LP levels or LP/LS between different categories of pleural effusion, we estimated the standardized mean difference (SMD) and their 95% CI as bias-adjusted Hedges’g.[15] We made formal pairwise comparisons between TPE and malignant and parapneumonic pleural effusions. We calculated summary effect sizes for SMDs using DerSimonian and Laird random effects model.[16]

We expressed between-study heterogeneity using Higgins’ inconsistency index (I2) and judged it high if the I2 value exceeded 0.75.[17] We explored heterogeneity through subgroup analysis if 10 or more studies were retrieved for any analysis. For this, we stratified data based on prespecified covariates that included study design, the national burden of TB (high or not), the prevalence of TB in the entire study population (below 50% or more), the robustness of reference standard (definite TB or composite clinical criteria), the inclusion of transudative effusions, a technique of lysozyme analysis, and blinding in study. The World Health Organization (WHO) guidelines were used to designate countries as high-burden.[18] We conducted sensitivity analysis by excluding one study at a time from the analysis to evaluate if the summary results were unduly influenced by any single publication. We assessed publication bias using Deek’s plot for diagnostic accuracy studies, funnel plots, and the non-parametric rank correlation (Begg’s) test for descriptive studies. We utilized the GRADE (Grading of Recommendations, Assessment, Development and Evaluation) approach to report the quality of evidence on diagnostic accuracy.[1920]

We analyzed our data using the statistical package Stata (Intercooled edition 12.0, Stata Corp, Texas, USA). The MetaDAS macro was additionally applied to fit the hierarchical summary receiver operating characteristic (HSROC) model in the SAS environment (SAS OnDemand for Academics, SAS Institute Inc., North Carolina, USA).[21]

RESULTS

Our literature search yielded 89 citations [Figure 1], of which we ultimately included 11 studies reporting information on diagnostic accuracy (N = 7) or comparing LP data between tuberculous and other effusions (N = 10).[22232425262728293132] The major attributes of these 11 publications are summarized in Table 1. The maximum number of studies (5, 45.5%) were conducted in Spain.[2224303132] Only two studies (18.2%), both from India, were conducted in a country with a high burden of TB.[2729] One Indian study reported data exclusively from pediatric subjects.[27] Blinding was reported in only one (9.1%) study.[32] Only one (9.1%) Spanish study reported the inclusion of HIV seropositive patients.[32] Two (18.2%) studies employed definite (microbiologic and pathologic) reference criteria for diagnosing TPE.[2430] Most studies (7, 63.6%) assayed lysozyme through a turbidimetric method [Table 1].[22262728293132]

F1-7
Figure 1:
Study selection process
T1-7
Table 1:
Characteristics of studies selected for analysis

Seven studies reported information regarding diagnostic test accuracy of LP/LS on 224 TPE patients and 630 patients having pleural effusions due to other disorders [Table S1 of online supplement].[23242728303132] Four of these studies also provided data for diagnostic test accuracy of LP on 88 TPE patients and 267 patients having pleural effusions due to other disorders [Table S1 of online supplement].[23272830]Only one (14.3%) study was published from a high TB burden country.[27] Five (71.4%) studies included patients only with an exudative pleural effusion.[2324303132] Only one (14.3%) study performed assays in a blinded fashion.[32] Four (57.1%) studies employed a composite reference standard to diagnose TPE.[24273132] Most studies (4, 57.1%) assayed lysozyme through a turbidimetric method.[27283132]

T2-7
Table:
No title available.

The diagnostic thresholds varied widely between 10.0 mg/L and 15.0 g/mL for LP, and between 1.0 and 2.0 for LP/LS [Table S1 of online supplement]. A high risk of bias was observed in all studies, except one, when assessed through the QUADAS-2 tool [Figure S1 of online supplement].[32] The bias was primarily related to the absence of blinding and failure to specify diagnostic thresholds before the start of the study. All these studies additionally showed applicability concerns in the patient selection domain as well. There was no publication bias [Figure S2 of online supplement].

F2-7
Figure
F3-7
Figure

Table S2 of the online supplement provides the diagnostic accuracy estimates calculated from individual studies. There was substantial heterogeneity between the studies providing information on LP (I2 92.24%), as well as LP/LS (I2 86.42%). The sensitivity of LP for diagnosis of TPE varied from 0.63 to 1.00, and specificity from 0.62 to 1.00 [Figure 2]. The summary sensitivity, specificity, and DOR were 0.94 (95% CI 0.53–1.00), 0.89 (95% CI 0.63–0.97), and 129.88 (95% CI 6.26–2695), respectively. The summary positive and negative likelihood ratios were 8.30 (95% CI 2.14–32.15) and 0.06 (95% CI 0.01–0.80), respectively. A low positive likelihood ratio (below 10) and a low negative likelihood ratio (below 0.1) for the summary estimate indicate that LP might prove useful for excluding, but not confirming, TPE. The sensitivity of LP/LS for diagnosis of TPE varied from 0.72 to 1.00, and specificity from 0.70 to 1.00 [Figure 2]. The summary sensitivity, specificity, and DOR were 0.98 (95% CI 0.58–1.00), 0.91 (95% CI 0.84–0.96), and 708.47 (95% CI 11.42–43946), respectively. The summary positive and negative likelihood ratios were 11.31 (95% CI 5.76–22.22) and 0.02 (95% CI 0.00–0.75), respectively. A high positive likelihood ratio (above 10) and a low negative likelihood ratio (below 0.1) for the summary estimate suggest that LP/LS could be useful both for confirming and excluding TPE. The HSROC plots [Figure 3] appeared symmetrical implying that test accuracy was not dependent on the test threshold for either LP or LP/LS. However, the HSROC plot for LP/LS exhibited a much narrower 95% confidence zone and was positioned closer to the desired upper left corner of the graph [Figure 3], implying that LP/LS had better accuracy for diagnosing TPE than LP. Our summary estimates for LP/LS were robust in the sensitivity analysis and did not change much after excluding any single study from the meta-analysis [Table S3 of online supplement]. Sensitivity analysis could not be performed on data for LP levels due to the insufficient number of studies. Because of the small number of studies, a subgroup analysis was also not conducted for any of the prespecified covariates.

T3-7
Table:
No title available.
F4-7
Figure 2:
Coupled forest plot from studies reporting on the diagnostic accuracy of pleural fluid lysozyme levels (top panel) and pleural fluid to serum lysozyme ratio (bottom panel). Individual study estimates are depicted by hollow squares, and the horizontal lines correspond to their 95% confidence intervals (95% CI). Solid diamonds represent the summary sensitivity and specificity estimates
F5-7
Figure 3:
Hierarchical summary receiver operating characteristic (HSROC) plots to summarize diagnostic accuracy for pleural fluid lysozyme (left panel) and pleural fluid to serum lysozyme ratio (right panel) in diagnosing tuberculous pleural effusion. Each open circle represents an individual study, with a circle size proportionate to the inverse standard error of sensitivity and specificity. Summary estimates of diagnostic accuracy are indicated by black squares, and the surrounding dashed regions outline the zone of 95% confidence around this estimate
T4-7
Table:
No title available.

In addition, 10 studies provided comparative data on biomarker estimation in pleural effusions due to TB or other disorders.[22232526272829303132] Eight and five studies each compared LP levels between TPE and malignant or parapneumonic pleural effusions, respectively [Table S4 of online supplement]. Seven and four studies each compared LP/LS between TPE and malignant or parapneumonic pleural effusions, respectively [Table S4 of online supplement]. Only five (50.0%) of these studies had a Newcastle–Ottawa Scale score of 7/9 or higher and were thus considered of high quality [Table S5 of online supplement].[2226282932] There was no significant publication bias [Figure S3 of online supplement].

T5-7
Table:
No title available.
T6-7
Table:
No title available.
F6-7
Figure

Mean LP levels were higher among TPE patients for all pairwise comparisons [Table S6 of online supplement]. Mean LP/LS values were similarly higher among TPE patients for all pairwise comparisons, except in a single study involving Indian children.[27] The SMDs exhibited a substantial variability for LP levels, as well as LP/LS values, between TPE and other pleural effusions [Figure 4]. There was considerable heterogeneity for LP, as well as LP/LS, for comparisons involving malignant pleural effusions (I2 79.18% and 75.21%, respectively). There was lesser heterogeneity for LP, as well as LP/LS, for comparisons involving parapneumonic pleural effusions (I2 32.91% and 53.42%, respectively). After pooling observations from different studies, LP levels were significantly greater in TPE than in malignant pleural effusions (summary SMD 1.51, 95% CI 1.04–1.98) or parapneumonic pleural effusions (summary SMD 0.86, 95% CI 0.51–1.22)[Figure 4]. Similarly, LP/LS was significantly higher in TPE than in malignant pleural effusions (summary SMD 1.77, 95% CI 1.31–2.22) or parapneumonic pleural effusions (summary SMD 1.15, 95% CI 0.64–1.66). A single outlier result (SMD − 0.11) was noted among the studies comparing this ratio between TPE and malignant pleural effusions [Figure 4].[27] Removal of this study improved the summary SMD from 1.77 to 1.93 (95% CI 1.63–2.23) with a considerable reduction in heterogeneity (I2 41.82%). Apart from this, our sensitivity analysis did not suggest any appreciable alteration in summary SMD if any one study was eliminated from that meta-analysis [Table S6 of online supplement]. However, the removal of a single Spanish study markedly improved the homogeneity in comparisons between TPE and parapneumonic pleural effusions [Table S6 of online supplement].[31] A formal subgroup analysis was not undertaken for any comparison due to the small number of studies.

T7-7
Table:
No title available.
F7-7
Figure 4:
Forest plots from studies comparing pleural fluid lysozyme levels (top panel) and pleural fluid to serum lysozyme ratio (bottom panel) in tuberculous pleural effusions (TPE) and malignant effusions or parapneumonic effusions. Individual standardized mean difference estimates from each study are depicted by hollow squares, and the horizontal lines correspond to their 95% confidence intervals (95% CI). Solid diamonds represent the summary estimates

Overall, we found low-grade evidence regarding the diagnostic performance of LP and LP/LS for the diagnosis of TPE [Table 2]. Based on our pooled data for LP, the false-positive rate was quite high for scenarios with low pre-test probabilities of TPE. The false-positive rate was somewhat lower, but still substantial, for LP/LS in such situations [Table 2]. The diagnostic performance of both tests appeared much better in settings with a higher pre-test probability of TPE.

T8-7
Table 2:
Summary of findings from studies evaluating the diagnostic accuracy of lysozyme in tuberculous pleural effusion

DISCUSSION

To our knowledge, the diagnostic utility of lysozyme for identifying TPE has never been systematically reviewed. Our meta-analysis suggests that LP exhibits good sensitivity (0.94) and moderate specificity (0.89) for diagnosing TPE [Table 2]. LP/LS shows better diagnostic discrimination (sensitivity 0.98, specificity 0.91). These results suggest a marginally better sensitivity and similar specificity, as compared to pleural fluid ADA estimation.[2] Further, both LP concentration and LP/LS were significantly higher in TPE than in malignant or parapneumonic pleural effusions.

Pleural fluid analysis is always the initial investigation while evaluating any patient suspected to have pleural TB. Because microbiological testing on pleural fluid has a low yield, clinicians useclinical details and findings from other pleural fluid investigations while judging the need for ATT. We, therefore, combined microbiologic, pathologic, and clinical criteria as study inclusion parameters to represent information relevant to real-life scenarios. Unfortunately, this also led to a rather imperfect reference standard for TPE diagnosis in several studies, and therefore we cannot entirely rule out misclassification bias. Some studies also included patients with transudative effusions, whereas others enrolled only malignant pleural effusions as a comparator. Because this is not the usual spectrum of clinical scenarios where pleural TB is suspected, specificity figures from such studies could be erroneous. Almost all studies enrolled a small number of subjects, and several were of poor quality. Some of these factors may compromise the validity and applicability of our findings. There was also considerable heterogeneity across the included publications. Because of the small number of eligible studies, we could not further explore potential reasons for heterogeneity. Also, various investigators employed a very wide range of diagnostic thresholds, mostly in a post-hoc fashion, and it was not feasible to identify a clinically acceptable range of values that could optimize diagnostic test performance.

How do our observations impact routine clinical practice? In a setting of low TPE prevalence (e.g., 5% pre-test probability), nearly 70% of positive LP test results and more than 60% of LP/LS results are likely to be falsely positive, implying that a large proportion of these patients may not undergo more definitive evaluation for an alternate etiology and could be unnecessarily prescribed empiric ATT [Table 2]. However, both tests are unlikely to be falsely positive for patients with nontuberculous pleural effusions. Conversely, in a high prevalence situation (e.g., 50% pre-test probability), about 6% of patients who test negative with LP will actually have a disease (but would be denied appropriate therapy); this rate is much lower at around 1.5% for LP/LS. More than 10% of positive LP test results and nearly 8% of LP/LS results are likely to be falsely positive. Overall, both LP and LP/LS appear to be reasonably good biomarkers for pleural TB, more so in high TB prevalence settings. LP/LS also seems to be a better discriminator than LP. However, our analysis was limited to evaluating the performance of lysozyme as a single isolated assay and we cannot comment on its additive utility when considered along with other test results. There are some data to suggest that the diagnostic accuracy improves further if it is combined with pleural fluid ADA estimation.[323334] Finally, there is a need for standardizing simpler automated assays for lysozyme determination, given its good diagnostic performance in TPE.

CONCLUSION

In conclusion, findings from our meta-analysis provide low-quality evidence that both LP and LP/LS exhibit good diagnostic accuracy for diagnosis of TPE, the latter being marginally superior. Good-quality studies are needed to better define clinically useful thresholds for LP and LP/LS.

Author contributions

ANA Conceptualization; methodology; investigation; formal analysis; data curation; supervision; writing-original draft; writing-review/editing.

RA Methodology; investigation; formal analysis; data curation; writing- original draft; writing- review/editing.

SD Methodology; investigation; formal analysis; writing- original draft; writing- review/editing.

KTP Methodology; investigation; formal analysis; writing- original draft; writing- review/editing.

IPS Methodology; investigation; formal analysis; writing-original draft; writing-review/editing.

VM Methodology; investigation; formal analysis; writing-original draft; writing-review/editing.

List of abbreviations

95% CI 95% confidence interval

ADA Adenosine deaminase

ATT Anti-tuberculous therapy

DOR Diagnostic odds ratio

HIV Human immunodeficiency virus

HSROC Hierarchical summary receiver operating characteristic

I2 Higgins’ inconsistency index

LP Pleural fluid lysozyme

LP/LS Pleural fluid to serum lysozyme ratio

PRISMA Preferred Reporting Items for Systematic Reviews and Meta-Analyses

QUADAS-2 Quality Assessment of Diagnostic Accuracy Studies, version 2

SD Standard deviation

SMD Standardized mean difference

TB Tuberculosis

TPE Tuberculous pleural effusion.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.

REFERENCES

1. Sehgal IS, Dhooria S, Aggarwal AN, Behera D, Agarwal R Diagnostic performance of Xpert MTB/RIF in tuberculous pleural effusion:Systematic review and meta-analysis J Clin Microbiol 2016 54 1133–6
2. 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
3. Aggarwal AN, Agarwal R, Dhooria S, Prasad KT, Sehgal IS, Muthu V Unstimulated pleural fluid interferon gamma for diagnosis of tuberculous pleural effusion:Asystematic review and meta-analysis J Clin Microbiol 2021 59 e02112–20
4. Light RW Light RW Clinical manifestations and useful tests Pleural Diseases (Sixth edition) 2013 Philadelphia, USA Lippincott Williams &Wilkins 86–127
5. Lee CY, Hong JY, Lee MG, Suh IB Identification of 10 candidate biomarkers distinguishing tuberculous and malignant pleural fluid by proteomic methods Yonsei Med J 2017 58 1144–51
6. Xuan WX, Li JJ, Zhang QC, Sun GN, Xu ZW, Sun ZF, et al Protein expression shift and potential diagnostic markers through proteomics profiling of tuberculous pleurisy biopsy tissues Int J Infect Dis 2020 99 245–52
7. de Pablo A, Villena V, Echave-Sustaeta J, Encuentra AL Are pleural fluid parameters related to the development of residual pleural thickening in tuberculosis? Chest 1997 112 1293–7
8. Salameh JP, Bossuyt PM, McGrath TA, Thombs BD, Hyde CJ, Macaskill P, et al Preferred reporting items for systematic review and meta-analysis of diagnostic test accuracy studies (PRISMA-DTA):Explanation, elaboration, and checklist BMJ 2020 370 m2632
9. Liberati A, Altman DG, Tetzlaff J, Mulrow C, Gøtzsche PC, Ioannidis JP, et al The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate health care interventions:Explanation and elaboration J Clin Epidemiol 2009 62 e1–34
10. Walter SD, Yao X Effect sizes can be calculated for studies reporting ranges for outcome variables in systematic reviews J Clin Epidemiol 2007 60 849–52
11. Whiting PF, Rutjes AW, Westwood ME, Mallett S, Deeks JJ, Reitsma JB, et al QUADAS-2:Arevised tool for the quality assessment of diagnostic accuracy studies Ann Intern Med 2011 155 529–36
12. Wells G, Shea B, O'Connell D, Peterson J, Welch V, Losos M, et al The Newcastle-Ottawa Scale (NOS) for assessing the quality of nonrandomised studies in meta-analyses 2013 Available from: http://www.ohri.ca/programs/clinical_epidemiology/oxford.asp Last accessed on 2021 Nov 15
13. Clopper CJ, Pearson ES The use of confidence or fiducial limits illustrated in the case of the binomial Biometrika 1934 26 404–13
14. Rutter CM, Gatsonis CA A hierarchical regression approach to meta-analysis of diagnostic test accuracy evaluations Stat Med 2001 20 2865–84
15. Hedges LV Distribution theory for Glass's estimator of effect size and related estimators J Educ Stat 1981 6 107–28
16. DerSimonian R, Laird N Meta-analysis in clinical trials Control Clin Trials 1986 7 177–88
17. Higgins JP, Thompson SG, Deeks JJ, Altman DG Measuring inconsistency in meta-analyses BMJ 2003 327 557–60
18. World Health Organization Global Tuberculosis Report 2021 2021 Geneva World Health Organization
19. Schunemann HJ, Mustafa RA, Brozek J, Steingart KR, Leeflang M, Murad MH, et al GRADE guidelines:21 part 1. Study design, risk of bias, and indirectness in rating the certainty across a body of evidence for test accuracy J Clin Epidemiol 2020 122 129–41
20. Schunemann HJ, Mustafa RA, Brozek J, Steingart KR, Leeflang M, Murad MH, et al GRADE guidelines:21 part 2. Test accuracy:Inconsistency, imprecision, publication bias, and other domains for rating the certainty of evidence and presenting it in evidence profiles and summary of findings tables J Clin Epidemiol 2020 122 142–52
21. Takwoingi Y, Deeks JJ MetaDAS:A SAS macro for meta-analysis of diagnostic accuracy studies (version 1.3) 2010 Available from: https://methods.cochrane.org/sdt/software-meta-analysis-dta-studies Last accessed on 2021 Nov 15
22. Alegre J, Jufresa J, Aleman C, Segura R, Armadans L, Marti R, et al Pleural fluid myeloperoxidase as a marker of infectious pleural effusions Eur J Intern Med 2001 12 357–62
23. Asseo PP, Tracopoulos GD, Kotsovoulou-Fouskaki V Lysozyme (muramidase) in pleural effusions and serum Am J ClinPathol 1982 78 763–7
24. Caballero M, Ruiz R, Márquez De Prado M, Seco M, Borque L, Escanero JF Development of a microparticle-enhanced nephelometric immunoassay for quantitation of human lysozyme in pleural effusion and plasma J Clin Lab Anal 1999 13 301–7
25. Klockars M, Pettersson T, Riska H, Hellstrom PE, Norhagen A Pleural fluid lysozyme in human disease Arch Intern Med 1979 139 73–7
26. Lew DP, Perrin LH, Vassalli JD, Suter S, Lambert PH, Waldvogel FA High levels of complement breakdown products in tuberculous pleural effusions Clin Exp Immunol 1983 52 569–74
27. Mishra OP, Yusuf S, Ali Z, Nath G Lysozyme levels for the diagnosis of tuberculous effusions in children J Trop Pediatr 2000 46 296–300
28. Moriwaki Y, Kohjiro N, Itoh M, Nakatsuji Y, Okada M, Ishihara H, et al Discrimination of tuberculous from carcinomatous pleural effusion by biochemical markers:Adenosine deaminase, lysozyme, fibronectin and carcinoembryonic antigen Jpn J Med 1989 28 478–84
29. Rajpal AS, Sharma GS, Gupta PK, Nanawati V Value of pleural fluid lysozyme estimation in tubercular and non-tubercular pleural effusions Indian J Tuberc 1981 28 205–8
30. Valdes L, San Jose E, Alvarez D, Sarandeses A, Pose A, Chomon B, et al Diagnosis of tuberculous pleurisy using the biologic parameters adenosine deaminase, lysozyme, and interferon gamma Chest 1993 103 458–65
31. Verea Hernando HR, Masa Jimenez JF, Dominguez Juncal L, Perez Garcia-Buela J, Martin Egana MT, Fontan Bueso J Meaning and diagnostic value of determining the lysozyme level of pleural fluid Chest 1987 91 342–5
32. Villena V, Navarro-Gonzalvez JA, Garcia-Benayas C, Manzanos JA, Echave J, Lopez-Encuentra A, et al Rapid automated determination of adenosine deaminase and lysozyme for differentiating tuberculous and nontuberculous pleural effusions Clin Chem 1996 42 218–21
33. FontanBueso J, Verea Hernando H, Garcia-Buela JP, Dominguez Juncal L, Martin Egana MT, Montero Martinez MC Diagnostic value of simultaneous determination of pleural adenosine deaminase and pleural lysozyme/serum lysozyme ratio in pleural effusions Chest 1988 93 303–7
34. San Jose E, Valdes L, Sarandeses A, Alvarez D, Chomon B Diagnostic value of adenosine deaminase and lysozyme in tuberculous pleurisy Clin Chim Acta 1992 209 73–81

ONLINE SUPPLEMENT

Pleural fluid lysozyme as a diagnostic biomarker of pleural tuberculosis: a systematic review and meta-analysis

Keywords:

Diagnostic tests; lysozyme; meta-analysis; pleural tuberculosis; sensitivity and specificity

© 2022 Lung India | Published by Wolters Kluwer – Medknow