Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Open Access

Peer-reviewed

Research Article

Meta-analysis of diagnostic accuracy of nucleic acid amplification tests for abdominal tuberculosis

  • Yanqin Shen,

    Roles Data curation, Formal analysis, Methodology, Resources, Software, Writing – original draft, Writing – review & editing

    Affiliation Zhejiang Tuberculosis Diagnosis and Treatment Center, Affiliated Hangzhou Chest Hospital, School of Medicine, Zhejiang Chinese Medicine and Western Medicine Integrated Hospital, Zhejiang University, Hangzhou, Zhejiang, China

    ⨯
  • Likui Fang,

    Roles Data curation, Formal analysis, Methodology, Resources, Validation

    Affiliation Zhejiang Tuberculosis Diagnosis and Treatment Center, Affiliated Hangzhou Chest Hospital, School of Medicine, Zhejiang Chinese Medicine and Western Medicine Integrated Hospital, Zhejiang University, Hangzhou, Zhejiang, China

    ⨯
  • Bo Ye,

    Roles Investigation, Resources, Supervision, Validation

    Affiliation Zhejiang Tuberculosis Diagnosis and Treatment Center, Affiliated Hangzhou Chest Hospital, School of Medicine, Zhejiang Chinese Medicine and Western Medicine Integrated Hospital, Zhejiang University, Hangzhou, Zhejiang, China

    ⨯
  • Guocan Yu

    Roles Conceptualization, Data curation, Methodology, Project administration, Software, Writing – review & editing

    dabaitwo@163.com

    Affiliation Zhejiang Tuberculosis Diagnosis and Treatment Center, Affiliated Hangzhou Chest Hospital, School of Medicine, Zhejiang Chinese Medicine and Western Medicine Integrated Hospital, Zhejiang University, Hangzhou, Zhejiang, China

    ⨯

Registered Report Protocol

14 Dec 2020: Shen Y, Fang L, Ye B, Yu G (2020) Meta-analysis of diagnostic accuracy of nucleic acid amplification tests for abdominal tuberculosis: A protocol. PLOS ONE 15(12): e0243765. https://doi.org/10.1371/journal.pone.0243765 View registered report protocol

Abstract

Background

Abdominal tuberculosis (TB) is a severe extrapulmonary TB, which can lead to serious complications. Early diagnosis and treatment are very important for the prognosis and the diagnosis of abdominal TB is still difficult.

Methods

We searched PubMed, the Cochrane Library, Embase, China National Knowledge Infrastructure, and the Wanfang database for studies evaluating the diagnostic accuracy of NAATs for abdominal TB until August 2020. Any types of study design with full text were sought and included. The risk of bias was assessed using the Quality Assessment of Diagnostic Accuracy Studies tool. Subgroup analysis, meta-regression analysis and sensitivity analysis were used to explore the sources of heterogeneity. Stata version 15.0 with the midas command packages was used to carry out meta-analyses.

Results

We included a total of 78 independent studies from 53 articles; 64 with CRS as the reference standard, and 14 with culture as the reference standard. The pooled sensitivity, specificity, and the areas under summary receiver operating characteristic (SROC) curves (AUC) were 58% (51%–64%; I2 = 87%), 99% (97%–99%; I2 = 81%), and 0.92 (0.89–0.94) compared with CRS, respectively. The pooled sensitivity, specificity, and the AUC values of the SROC were 80% (66%–90%; I2 = 56%), 96% (92%–98%; I2 = 84%), and 0.97 (0.95–0.98) compared with culture, respectively. The heterogeneity of sensitivity and specificity was significant.

Conclusions

NAATs had excellent efficacy in the diagnosis of abdominal TB regardless of the reference standard and regardless of the subtype of abdominal TB. Multiplex PCR with multiple target genes may improve diagnostic sensitivity, and stool specimens may also be used for the diagnosis of abdominal TB in addition to tissue and ascites.

Citation: Shen Y, Fang L, Ye B, Yu G (2023) Meta-analysis of diagnostic accuracy of nucleic acid amplification tests for abdominal tuberculosis. PLoS ONE 18(11): e0289336. https://doi.org/10.1371/journal.pone.0289336

Editor: Kamal Kishore Chopra, New Delhi Tuberculosis Centre, INDIA

Received: January 22, 2021; Accepted: July 14, 2023; Published: November 27, 2023

Copyright: © 2023 Shen et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: All relevant data are within the paper and its Supporting Information files.

Funding: The author(s) received no specific funding for this work.

Competing interests: The authors have declared that no competing interests exist.

Introduction

Tuberculosis (TB) is a serious threat to global health [1]. Severe types of extrapulmonary tuberculosis (EPTB) increase tuberculosis-related mortality, especially in immunodeficient populations. Abdominal TB is a common form of EPTB caused by Mycobacterium TB (MTB) infection of the abdominal organs, mainly including intestinal and peritoneal TB [2]. Abdominal TB can cause many complications, such as intestinal obstruction, intestinal perforation, which seriously affect the quality of life and prognosis of patients [3]. Therefore, early diagnosis and treatment of abdominal TB is very important to reduce the incidence of serious abdominal complications. Crohn’s disease (CD), inflammatory bowel disease (IBD) and abdominal TB have similar clinical presentations and pathologies [4]. It is easy to misdiagnose abdominal TB as CD and IBD, thus delaying the treatment. The diagnosis of abdominal TB is still challenging.

Nucleic acid amplification tests (NAATs) play a huge role in the diagnosis of microbiological infections, making it faster and more accurate [5]. NAATs are widely used in the diagnosis of TB, which make the early diagnosis of TB possible [6, 7]. In the diagnosis of EPTB, NAATs are also fast, accurate and efficient, and they improve the detection rate of TB, especially in specimens with low bacterial content, such as tuberculous lymphadenitis and tuberculous meningitis [8, 9]. Abdominal TB is a type of paucibacillary EPTB and NAATs also have these advantages in its diagnosis. However, the diagnostic efficacy of NAATs for abdominal TB remains controversial. The aim of this systematic review and meta-analysis is to assess the diagnostic validity of NAATs for the diagnosis of abdominal TB.

Methods

Design and registration

We conducted a systematic review and meta-analysis of diagnostic test accuracy to assess the diagnostic efficacy of NAATs for abdominal TB. We have registered the protocol on the International Platform of Registered Systematic Review and Meta-analysis Protocols (INPLASY), and the registration number is INPLASY202060030 [10]. The Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) 2020 statement was followed for reporting our systematic review [11].

Information sources

PubMed, the Cochrane Library, Embase, China National Knowledge Infrastructure (CNKI), and the Wanfang database were searched for studies that evaluate NAAT’s diagnostic accuracy for abdominal TB until August 2020.

Search strategy

The search strategies were conducted by Yanqin Shen and Likui Fang.

There was no language restriction on our search. Search strategy of PubMed was as follows:

#1 "Tuberculosis, Gastrointestinal"[Mesh] OR "Gastrointestinal Tuberculosis" OR “Intestinal tuberculosis" OR "Peritonitis, Tuberculous"[Mesh] OR "Tuberculosis, Peritoneal" OR "peritoneal tuberculosis" OR "Tuberculous ascites" OR "Tuberculous Peritonitis" OR "Abdominal TB" OR "intra-Abdominal TB"

#2 ("Nucleic Acid Amplification Techniques"[Mesh] OR "Polymerase Chain Reaction"[Mesh] OR "Real-Time Polymerase Chain Reaction"[Mesh] OR "Reverse Transcriptase Polymerase Chain Reaction"[Mesh] OR "Multiplex Polymerase Chain Reaction"[Mesh] OR "genexpert"[tw] OR Xpert OR "genotype"[tw])

#3 #1 AND #2

Similar search formulae were used for Embase, the Cochrane Library, CNKI, and Wanfang databases.

Eligibility criteria

Type of study.

Any types of studies can be included, such as retrospective studies, prospective studies, case-control studies. We included original researches with full text that assessed the diagnostic accuracy of NAATs for abdominal TB. The reference standard should be appropriate and precisely defined in the study. True positive (TP), false positive (FP), false negative (FN), and true negative (TN) values were provided directly in the articles or contain necessary data to calculate these values. We excluded articles reported in languages other than Chinese and English, case reports, studies with a specimen size of less than 10, conference coverages, and studies with abstracts but no full text.

Patients.

We included studies, which contain patients diagnosed with abdominal TB through NAATs. We had no restrictions on age, gender, and nations.

Index tests.

NAATs were considered as index test.

Reference standards.

Bacteriological confirmation of MTB (positive culture of MTB and/or microscopic identification of acid‐fast bacilli on stained specimen smear) was reference gold standard.

Composite reference standard (CRS): Radiological characteristics (such as tree-in-bud pattern and cavity) and histopathological features of the suspected tissue specimen (features of chronic granulomatous inflammation with caseous necrosis/ caseating granuloma). Positive of reference standard test and/or positive of all CRS mentioned were considered abdominal TB. If all factors were negative, it was considered as non- abdominal TB.

Literature screening and selection.

Primary search results matching the search strategy were imported into the ENDNOTE X9.2 literature management software. Two investigators (Yanqin Shen and Likui Fang) screened candidate studies independently by reviewing the titles and abstracts followed by the full text. Disagreements between two the researchers were resolved by discussion with a third researcher (Guocan Yu).

Data extraction

Name of first author; year of publication; country of study; reference standard; TP, FP, FN, and TN values of the test; method of patient selection; test method; NAAT target genes; subtypes of abdominal TB (such as intestinal and peritoneal TB); type of specimen; specimen processing procedures (e.g., homogenization) and specimen condition along with other parameters were extracted. One article that simultaneously reported the accuracy of different specimen types, different reference standards, or different NAAT target genes for abdominal TB diagnosis was considered as to include separate studies based on different specimen types, or different reference standards, or different NAAT target genes [12]. The same two researchers independently extracted relevant data from each included study and cross-check their respective information. Disagreements between two the researchers were resolved by discussion with a third researcher, similar to that used during the literature selection phase.

Quality evaluation

The two researchers assessed the quality of the relevant literature using a revised tool for Quality Assessment of Diagnostic Accuracy Studies (QUADAS-2) [13] independently for the different reference standards and the disagreements between researchers were solved by discussion with a third researcher (Guocan Yu). According to the PRISMA-DTA statement, systematic review and meta-analysis of diagnostic test accuracy studies was not required to assess publication bias. The strength of the body of evidence was assessed using The Grading of Recommendations Assessment, Development and Evaluation (GRADE) guideline.

Data synthesis and statistical analysis

TP, FP, FN, and TN values were obtained from each included study, and the estimated pooled sensitivity and specificity of NAAT for abdominal TB associated with the 95% confidence interval (CI) were calculated against culture or CRS, using bivariate random-effects models. We generated forest plots for sensitivity and specificity for each study and calculate the areas under summary receiver operating characteristic (SROC) curves (AUC). We assessed heterogeneity between studies using I2 statistics. An I2 value of 0% was indicative of no heterogeneity, while a value greater than 50% indicated significant heterogeneity [14]. Subgroup analyses were done to evaluate the diagnostic accuracy of NAATs for abdominal TB, such as different test methods, subtypes of abdominal TB, types of specimens, methods of patient selection (consecutive or convenience), methods of decontamination (with or without N-acetyl-L-cysteine/sodium hydroxide [NALC-NaOH]), conditions of sample (frozen or fresh), homogenization methods (mechanical or otherwise). If the heterogeneity was obvious, meta-regression analyses and sensitivity analysis were used to explore the source of heterogeneity. The meta-analysis for predefined variable types were performed using at least four published studies. We analyzed the data from studies against CRS and culture separately. We used Stata version 15.0 (Stata Corp., College Station, TX, USA) with the midas command packages to generate forest plots of sensitivity and specificity with 95% CI and carry out meta-analyses and meta-regression analyses.

Results

Identification of studies and study characteristics

By searching the relevant databases using our predefined search strategies, we found 966 candidate articles. By eliminating duplicates followed by the screening of titles, abstracts, and full texts, 53 articles met the inclusion criteria and were included in this study for meta-analyses [2, 4 12, 15–64]. The PRISMA flow chart of literature retrieval was shown in Fig 1. Fifteen articles were published in Chinese [16–19, 21, 24, 27, 29, 30, 38, 42, 43, 51, 59, 64], and the remaining 38 articles were published in English. The kappa index of agreement value between the two researchers for selection and data extraction was 0.735 (95% CI, 0.621–0.937). Eleven articles published in languages other than Chinese and English were excluded. We excluded 21 articles that reported sensitivity only (no specificity values were listed) and 17 articles did not report separate abdominal TB data. Three articles used the same data from the included articles so we also excluded them [65–67]. Three articles used reference standards other than those defined in this study, one article used genome sequencing as the reference standard [68], and the other two used histopathology data as the reference standard [69, 70]; as these did not meet the inclusion criteria of this study, they were excluded. The types of specimens used in the included articles were ascites fluid, stool, and tissue, abdominal TB types were mainly peritoneal and intestinal TB, and NAAT target genes were mainly IS6110, ropB, and MPB64. According to the principles defined in the methodology section, we included a total of 78 independent studies; 64 with CRS as the reference standard, and 14 with culture as the reference standard. Table 1 demonstrates all the included study characteristics. When CRS was used as the reference standard, a total of 4383 specimens were included in 64 studies, with specimen sizes ranging from 13 to 191 with a mean specimen size of 68.5. When culture was used as the reference standard, a total of 740 specimens were included in the 14 studies. Specimen sizes ranged from 10 to 139 with a mean specimen size of 52.9.

thumbnail
Download:
Fig 1. PRISMA flow chart of literature retrieval.

In total, 104, 31, 429, 171, and 231 articles were found in PubMed, the Cochrane Library, Embase, China National Knowledge Infrastructure, and the Wanfang database, respectively. NAAT: nucleic acid amplification test.

https://doi.org/10.1371/journal.pone.0289336.g001

thumbnail
Download:
Table 1. Characteristics of the included studies.

https://doi.org/10.1371/journal.pone.0289336.t001

Study quality

Fig 2 shows the results of the methodological evaluation on the quality of the included studies comparing CRS and culture. Most of the included studies used the nonconsecutive patient selection method. In the CRS reference standard, some studies did not include the anti-TB treatment response, and some studies included the results of the index test. These were the major sources of bias. The risk of bias that originated from the two aspects of the index test and the flow and timing were relatively low. According to the GRADE guidelines, the quality of evidence of this meta-analysis was high, and the recommendation level was moderate when CRS was used as the gold standard. The quality of evidence of this meta-analysis was high, and the recommendation level was strong when culture was used as the gold standard.

thumbnail
Download:
Fig 2. Methodological quality graphs (risk of bias and applicability concerns) across the included studies.

a) composite reference standard as gold standard. b) culture as gold standard.

https://doi.org/10.1371/journal.pone.0289336.g002

Diagnostic accuracy of NAATs for abdominal TB

The efficacy of NAATs for the diagnosis of abdominal TB was assessed in 64 studies when CRS was the reference standard. The pooled sensitivity, specificity, and the area under the curve (AUC) of the SROC were 58% (51%–64%; I2 = 87%), 99% (97%–99%; I2 = 81%), and 0.92 (0.89–0.94), respectively (Fig 3). The heterogeneity of sensitivity and specificity was significant. Fourteen studies assessed the efficacy of NAATs for abdominal TB compared with culture. The pooled sensitivity, specificity, and the AUC values of the SROC were 80% (66%–90%; I2 = 56%), 96% (92%–98%; I2 = 84%), and 0.97 (0.95–0.98), respectively (Fig 4). The heterogeneity of sensitivity and specificity was also significant.

thumbnail
Download:
Fig 3. Forest plot for the sensitivity and specificity of NAATs for the diagnosis of abdominal TB compared with a composite reference standard.

https://doi.org/10.1371/journal.pone.0289336.g003

thumbnail
Download:
Fig 4. Forest plot for the sensitivity and specificity of NAATs for the diagnosis of abdominal TB compared with culture.

https://doi.org/10.1371/journal.pone.0289336.g004

Subgroup analyses

Subgroup analysis of parameters for which sufficient data were available. The analyzed results of each subgroup are listed in Table 2 (compared with CRS). The results show that the heterogeneity in sensitivity and specificity remained high significant in the majority of subgroups. The heterogeneity of sensitivity and specificity in the diagnosis of abdominal TB based on the use of multiplex polymerase chain reaction (PCR) alone was insignificant. The heterogeneity in sensitivity for the remaining subgroups was significant. The heterogeneity of the specificity of the retrospective study, paraffin-embedded samples, and mechanical method for homogenization was nonsignificant, while the heterogeneity of the specificity of the remaining subgroups was highly significant. Table 3 demonstrates the analyzed results of each subgroup (sufficient data) compared with culture. For subgroups with limited data, subgroup analysis was dropped. The heterogeneity of sensitivity was insignificant except in the convenience patient selection method group. Additionally, the heterogeneity of specificity in most subgroups was insignificant.

thumbnail
Download:
Table 2. Subgroup analysis for different parameters with sufficient data compared with a composite reference standard.

https://doi.org/10.1371/journal.pone.0289336.t002

thumbnail
Download:
Table 3. Subgroup analysis for different parameters with sufficient data compared with culture.

https://doi.org/10.1371/journal.pone.0289336.t003

Meta-regression and sensitivity analysis

Meta-regression analysis could not be performed with Stata (with the midas command packages) if there were more than two subgroups that used the same parameter. Studies that did not report relevant data were removed from the meta-regression analysis of the relevant parameters. When compared with CRS, methods of patient selection (consecutive or convenience) and methods of decontamination (with or without NALC-NaOH) had no effect on the sensitivity and specificity of NAATs for abdominal TB (meta-regression P > 0.05), and homogenization methods (mechanical or otherwise) may affect the sensitivity and specificity of NAATs (meta-regression P < 0.05). Compared with culture, subtypes of abdominal TB (peritoneal or intestinal TB), types of specimens (ascitic fluid or tissue), and homogenization methods, may have effects on sensitivity and specificity of NAATs (meta-regression P < 0.05). By contrast, methods of patient selection had no effect on sensitivity and specificity of NAATs (meta-regression P > 0.05).

Sensitivity analysis did not identify specific articles as sources of heterogeneity in sensitivity and specificity compared with CRS. When compared with culture, the heterogeneity of both sensitivity and specificity decreased significantly after the article published by Salman et al. was eliminated. The pooled sensitivity, specificity, and the AUC value of the SROC after the article was eliminated were 77% (64%–86%, I2 = 49%), 97% (94%–98%, I2 = 27%), and 0.97 (0.95–0.98), respectively (Fig 5). The heterogeneity of sensitivity and specificity was insignificant.

thumbnail
Download:
Fig 5. Forest plot for the sensitivity and specificity of NAATs for the diagnosis of abdominal TB after eliminating the article published by Salman et al., when compared with culture [71].

https://doi.org/10.1371/journal.pone.0289336.g005

Discussion

Despite NAATs important role in the diagnosis of TB, its diagnostic ability for abdominal TB remains inconsistent. This study systematically evaluated the diagnostic efficacy of NAATs for abdominal TB using a meta-analysis method. The study’s findings suggested that NAATs can be used as means for early and rapid diagnosis of abdominal TB. However, the heterogeneity of sensitivity and specificity was significant regardless of the reference standard which made the results less credible. Satisfactory effects of NAATs have been observed in other types of TB as well [71, 72]. The overall diagnostic accuracy of NAATs in pulmonary TB is better than in abdominal TB, which correlates with the higher bacterial content of respiratory specimens compared to abdominal TB specimens (such ascites) [71]. In other types of EPTB, the diagnostic accuracy of NAATs in different types of specimens was highly variable, but overall, the results were similar to those observed in abdominal TB [72–74].

NAATs contain multiple assays, each of which may have its own characteristics, and multiple manifestations of abdominal TB which may be a source of heterogeneity among studies [75]. Different types of PCR, Xpert MTB/RIF (Xpert), and LAMP are the common types of NAATs. In this study, the numbers of studies using real-time PCR, fluorescent quantitative PCR, and LAMP assays, were limited, and subgroup analysis could not be performed compared with CRS. When compared with culture, only the studies which used the Xpert assay had sufficient numbers for subgroup analysis. Subgroup analysis showed that the sensitivity was the highest when multiplex PCR assays were used, followed by those which used PCR, and the lowest by those using Xpert assay, but the AUC was indeed the highest for Xpert compared with CRS. This may be attributed to the fact that multiplex PCR can detect multiple target genes simultaneously, thus increasing the MTB detection rate. However, the method did not improve specificity compared to Xpert, and the total diagnostic efficacy was thus still lower than that of Xpert. Heterogeneity in sensitivity and specificity was insignificant when using multiplex PCR, and the result was reliable. This result suggested that the increase of the number of target genes in the same test may improve the sensitivity of the diagnosis. When the CRS was compared in different target genes, only the studies which used the IS6110 and ropB as target genes had sufficient numbers for subgroup analysis. IS6110 is an extensively used target gene in NAATs and ropB is a specific target gene for Xpert [76, 77]. This meta-analysis revealed that the sensitivity of IS6110 was superior to that of ropB, but the specificity and AUC were slightly lower than that of ropB. Both of these two target genes were efficient for the diagnosis of abdominal TB. In comparison with culture, only ropB was available for subgroup analysis, thus suggesting that this target gene had good diagnostic efficacy for abdominal TB.

Abdominal TB has various subtypes [78]. The common ones are peritoneal and intestinal TB, and intestinal TB can occur in any part of the gastrointestinal tract [79]. The original studies cited herein did not distinguish the specific site of intestinal TB infection, so we also did not distinguish the site of intestinal TB infection, but categorized it as intestinal TB for analysis. The specimen type which corresponded to different types of abdominal TB was also different. The common specimen for peritoneal TB is ascites, but peritoneal tissue specimens can also be obtained laparoscopically. The common specimens for intestinal TB are endoscopically obtained biopsy tissue specimens, but the stool can also be used for testing. This meta-analysis showed that for different types of specimens, the greatest diagnostic efficacy of NAATs was obtained in tissue specimens followed by stool specimens. The diagnostic efficacy in ascites was the lowest. The higher bacterial content of MTB in tissue specimens compared with ascites specimens, and the higher efficacy of NAATs with tissue specimens compared with ascites, were consistent with previous studies. However, both tissue and ascites specimens need to be obtained with invasive procedures, such as laparotomy, laparoscopy, or colonoscopy. However, these procedures are associated with some risks. Furthermore, the stool specimens can be obtained very easily and noninvasively. This study demonstrated that the use of stool specimens to detect abdominal TB also had very good diagnostic efficacy, and is an excellent alternative diagnostic route in cases in which tissue or ascites cannot be obtained. However, the diagnostic efficacy of the studies which used stools were all related with intestinal TB cases, and had limited specimen sizes. The diagnostic efficacy of peritoneal TB is still unknown, and multicenter studies with large samples are needed to further evaluate it in different types of abdominal TB. NAATs were more effective in the diagnosis of intestinal TB compared with peritoneal TB. This may be related to the different specimens tested. Intestinal TB was commonly detected by biopsy tissue specimens, whereas peritoneal TB was commonly detected by ascites. This result was different from previous studies which evaluated the diagnostic efficacy of Xpert in abdominal TB [80]. Our study showed that other NAATs, such as PCR and multiplex PCR, had superior sensitivity in abdominal TB compared with Xpert, and our study included a richer set of studies. These factors contributed to the superior sensitivity of our study.

For different conditions of specimens, our subgroup analysis showed that fresh and frozen specimens had better diagnostic efficacy compared with paraffin-embedded specimens. The reason may be attributed to the fact that the paraffin-embedded specimens may have been stored for a long time and the handling of deparaffinized specimens may have had an effect on the specimens, the exact cause of which needs to be further clarified. The efficacy of fresh and frozen specimens was similar and both were excellent. Therefore, fresh or frozen specimens should be selected for testing as far as possible, and paraffin-embedded specimens should be selected only as a last resort. However, the heterogeneity among most subgroup studies was significant especially when CRS was compared. Accordingly, the results needed to be treated with caution.

We used meta-regression and sensitivity analyses to explore the sources of heterogeneity. When compared with CRS, homogenization methods (mechanical or otherwise) may have affected sensitivity and specificity of NAATs. Subgroup analyses showed decreased heterogeneity between studies with mechanical homogenization methods, especially in terms of specificity. This may be related to the fact that the mechanical homogenization method resulted in a more homogeneous MTB within the specimens. However, the effects of this method on the results and its causes needed to be further explored. When compared with culture, subtypes of abdominal TB, types of specimen and homogenization methods might have effects on sensitivity and specificity of NAATs. However, the analysis of subgroups with a sufficient number of studies revealed that heterogeneity between studies within subgroups was still highly significant, and subgroup analyses of other variables under the same parameter were not performed owing to the limited number of studies. Therefore, the effect of these factors on heterogeneity still needs to be clarified in large sample studies. Sensitivity analysis found that the article published by Salman et al. was the main source of heterogeneity when compared with culture. The pooled sensitivity, specificity, and the AUC values of the SROC after the elimination of the article were 77%, 97%, and 0.97, respectively. The heterogeneity of sensitivity and specificity was insignificant. We read the article again carefully and found that the study population was long-term peritoneal dialysis patients, which was different from other studies, and may be the source of the heterogeneity of the article. For CRS, it may constitute a source of heterogeneity. The definition of CRS in the included original studies may also be inconsistent. For example, some studies did not include clinical manifestations, some did not include the effect of antiTB treatment, and some included index tests in the CRS, all of which can generate heterogeneity.

The present study is associated with some limitations. Research omissions were unavoidable despite our best efforts to identify relevant studies. Data from some studies could not be extracted because they did not report the results we needed. Data for some subgroups were limited, especially when culture were used as the reference standard.

Conclusions

To the best of our knowledge, this was the first diagnostic meta-analysis for the diagnostic efficacy of NAATs for abdominal TB. We found that NAATs had excellent efficacy in the diagnosis of abdominal TB regardless of the reference standard and regardless of the subtype of abdominal TB. The test efficacy was good when different types of NAATs, different target genes, and different specimen types were used. Multiplex PCR with multiple target genes may improve diagnostic sensitivity, and stool specimens may also be used for the diagnosis of abdominal TB in addition to tissue and ascites. We hope that the results of the study will help clinicians and patients to understand in depth the role of NAATs in the diagnosis of abdominal TB.

Supporting information

S1 File. Search strategies.

https://doi.org/10.1371/journal.pone.0289336.s001

(DOCX)

S2 File. Excluded articles.

https://doi.org/10.1371/journal.pone.0289336.s002

(DOCX)

S3 File. PRISMA DTA checklist.

https://doi.org/10.1371/journal.pone.0289336.s003

(DOC)

S1 Fig. Methodological quality summary.

a) composite reference standard as gold standard. b) culture as gold standard.

https://doi.org/10.1371/journal.pone.0289336.s004

(TIF)

Acknowledgments

We thank the authors of the studies included in this meta-analysis.

References

  1. 1. World Health Organization. Global tuberculosis report 2020. Ghana, Geneva.
  2. 2. Fei BY, Lv HX, Zheng WH. Fluorescent quantitative PCR of Mycobacterium tuberculosis for differentiating intestinal tuberculosis from Crohn’s disease. Brazilian Journal of Medical and Biological Research. 2014;47(2):166–70. pmid:24519133
  3. 3. Ahmad QA, Sarwar MZ, Fatimah N, Ahmed AS, Changaizi SH, Ayyaz M. Acute Presentation and Management of Abdominal Tuberculosis. J Coll Physicians Surg Pak. 2020;30(2):129–33. pmid:32036817
  4. 4. Jin XJ, Kim JM, Kim HK, Kim L, Choi SJ, Park IS, et al. Histopathology and TB-PCR kit analysis in differentiating the diagnosis of intestinal tuberculosis and Crohn’s disease. World J Gastroenterol. 2010;16(20):2496–503. pmid:20503449
  5. 5. Rossignol L, Feuillepain L, Ndeikoundam Ngangro N, Souty C, Fournet N, Le Strat Y, et al. Estimate of male urethritis incidences in France between 2007 and 2017 with a specific focus on Neisseria gonorrhoeae, Chlamydia trachomatis, and Trichomonas vaginalis infections. BMC Infect Dis. 2019;19(1):561. pmid:31248368
  6. 6. Lv C, Wu J, Pierre-Audigier C, Lu L, Alame-Emane AK, Takiff H, et al. Combination of Xpert MTB/RIF and MTBDRplus for Diagnosing Tuberculosis in a Chinese District. Med Sci Monit. 2020;26:e923508. pmid:32504464
  7. 7. Zhou YC, He SM, Wen ZL, Zhao JW, Song YZ, Zhang Y, et al. A Rapid and Accurate Detection Approach for Multidrug-Resistant Tuberculosis Based on PCR-ELISA Microplate Hybridization Assay. Lab Med. 2020;51(6):606–13. pmid:32447387
  8. 8. Sivaratnam L, Nawi AM, Abdul Manaf MR. An evidence-based clinical pathway for the diagnosis of tuberculous lymphadenitis: A systematic review. Int J Mycobacteriol. 2020;9(2):107–15. pmid:32474531
  9. 9. Foppiano Palacios C, Saleeb PG. Challenges in the diagnosis of tuberculous meningitis. J Clin Tuberc Other Mycobact Dis. 2020;20:100164. pmid:32462082
  10. 10. Shen Y, Fang L, Ye B, Yu G. Diagnostic accuracy of nucleic acid amplification tests for abdominal tuberculosis: a protocol of systematic review and meta-analysis. 2020.
  11. 11. Page MJ, McKenzie JE, Bossuyt PM, Boutron I, Hoffmann TC, Mulrow CD, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ 2021;372:n71. pmid:33782057
  12. 12. Hallur V, Sharma M, Sethi S, Sharma K, Mewara A, Dhatwalia S, et al. Development and evaluation of multiplex PCR in rapid diagnosis of abdominal tuberculosis. Diagnostic Microbiology & Infectious Disease. 2013;76(1). pmid:23608350
  13. 13. Whiting PF, Rutjes AW, Westwood ME, Mallett S, Deeks JJ, Reitsma JB, et al. QUADAS-2: a revised tool for the quality assessment of diagnostic accuracy studies. Ann Intern Med. 2011;155(8):529–36. pmid:22007046
  14. 14. Puhan M, Scharplatz M, Troosters T, Walters EH, Steurer J. Pulmonary rehabilitation following exacerbations of chronic obstructive pulmonary disease. Cochrane Database Syst Rev. 2009(1):Cd005305. pmid:19160250
  15. 15. H G, Q O, H B, S L, D C, G L, et al. Value of polymerase chain reaction assay in diagnosis of intestinal tuberculosis and differentiation from Crohn’s disease. Chinese medical journal. 1995;108(3).
  16. 16. Jiang H, Zhao R, Zhang Y, Li J, Wang M, Yao X. Diagnostic value of Mycobacterium tuberculosis deoxyribonucleic acid for tuberculous ascites. Chinese Journal of Internal Medicine. 1997(07):55. Chinese.
  17. 17. Wu Z, Yang J, Yin B, Wang Q. Clinical utility of polymerase chain reaction for the diagnosis of tuberculous peritonitis. Journal of Anhui Medical University. 1997(03):54–5. Chinese.
  18. 18. Hao Z, Chang X, Lei J. Diagnostic value of ascites TB-PCR for tuberculous peri-tonitis. Journal of Xi’an Medical University (Chinese Edition). 1998(03):3–5. Chi-nese.
  19. 19. Liu W, Pan A,Li H. Evaluation of polymerase chain reaction for the diagnosis of tuberculous peritonitis. Chinese Journal of Antituberculosis. 1998(02):3–5. Chinese.
  20. 20. Gan HT, Chen YQ, Ouyang Q, Bu H, Yang XY. Differentiation between intestinal tuberculosis and Crohn’s disease in endoscopic biopsy specimens by polymerase chain reaction. Am J Gastroenterol. 2002;97(6):1446–51. pmid:12094863
  21. 21. Huang W, Wang W, Gong F. Clinical diagnosis of intestinal tuberculosis by endoscopic biopsy. Chinese Journal of Abdominal Diseases. 2003;3(8):563–5. Chinese.
  22. 22. Amarapurkar DN, Patel ND, Amarapurkar AD, Agal S, Baigal R, Gupte P. Tissue polymerase chain reaction in diagnosis of intestinal tuberculosis and Crohn’s disease. J Assoc Physicians India. 2004;52:863–867. pmid:15906835
  23. 23. Balamurugan R, Venkataraman S, John KR, Ramakrishna BS. PCR amplification of the IS6110 insertion element of Mycobacterium tuberculosis in fecal samples from patients with intestinal tuberculosis. Journal of Clinical Microbiology. 2006;44(5):1884–6. pmid:16672431
  24. 24. Yang G, Li L,Guan H, Xue C, Bai, H. Differential diagnosis of tuberculous and malignant ascites in the elderly. Chinese Journal of Gerontology. 2007;27(15):1497–8. Chinese.
  25. 25. Bandyopadhyay D, Gupta S, Banerjee S, Gupta S, Ray D, Bhattacharya S, et al. Adenosine deaminase estimation and multiplex polymerase chain reaction in diagnosis of extra-pulmonary tuberculosis. Int J Tuberc Lung Dis. 2008;12(10):1203–8. pmid:18812052
  26. 26. Pulimood AB, Peter S, Rook GWA, Donoghue HD. In situ PCR for Mycobacterium tuberculosis in endoscopic mucosal biopsy specimens of intestinal tuberculosis and Crohn disease. American Journal of Clinical Pathology. 2008;129(6):846–51. pmid:18479999
  27. 27. Ma Z, Yang J, Yang Z, Wang Y, Feng L, Lv C. Differential diagnostic value of tuberculosis antibody and TB-DNA tests for benign and malignant ascites. Shandong Medicine. 2008(32):83–4. Chinese.
  28. 28. Kim SH, Cho OH, Park SJ, Ye BD, Sung H, Kim MN, et al. Diagnosis of abdominal tuberculosis by T-cell-based assays on peripheral blood and peritoneal fluid mononuclear cells. Journal of Infection. 2009;59(6):409–15. pmid:19778553
  29. 29. Gu Q, Ou Y, Xia Q. FQ-PCRDiagnostic value of detecting intestinal mucosal Mycobacterium tuberculosis DNA for intestinal tuberculosis. China Experimental Diagnostics. 2009;13(06):741–4. Chinese.
  30. 30. Han M, Lv Y, Fei F, Yang X. Clinical differentiation of intestinal tuberculosis from Crohn’s disease and significance of polymerase chain reaction test for Mycobacterium tuberculosis in the intestinal mucosa. Journal of Gastroenterology and Hepatology. 2009;18(09):820–3. Chinese.
  31. 31. Mishra PK, Bhargava A, Punde RP, Pathak N, Desikan P, Jain A, et al. Diagnosis of gastrointestinal tuberculosis: Using cytomorphological, microbiological, immunological and molecular techniques—A study from Central India. Indian Journal of Clinical Biochemistry. 2010;25(2):158–63. pmid:23105903
  32. 32. Ramadass B, Chittaranjan S, Subramanian V, Ramakrishna BS. Fecal polymerase chain reaction for Mycobacterium tuberculosis IS6110 to distinguish Crohn’s disease from intestinal tuberculosis. Indian Journal of Gastroenterology. 2010;29(4):152–6. pmid:20577845
  33. 33. Vadwai V, Boehme C, Nabeta P, Shetty A, Alland D, Rodrigues C. Xpert MTB/RIF: a new pillar in diagnosis of extrapulmonary tuberculosis? J Clin Microbiol. 2011;49(7):2540–5. pmid:21593262
  34. 34. Lei Y, Yi FM, Zhao J, Luckheeram RV, Huang S, Chen M, et al. Utility of in vitro interferon-γ release assay in differential diagnosis between intestinal tuberculosis and Crohn’s disease. Journal of Digestive Diseases. 2013;14(2):68–75. pmid:23176201
  35. 35. Sharma K, Sinha SK, Sharma A, Nada R, Prasad KK, Goyal K, et al. Multiplex PCR for rapid diagnosis of gastrointestinal tuberculosis. Journal of Global Infectious Diseases. 2013;5(2):49–53. pmid:23853431
  36. 36. Zmak L, Jankovic M, Jankovic VK. Evaluation of Xpert MTB/RIF assay for rapid molecular diagnosis of tuberculosis in a two-year period in Croatia. Int J Mycobacteriol. 2013;2(3):179–82. pmid:26785988
  37. 37. Chen Y, He Y, Chen B, Mao R, Cao K, Xu P, et al. The value of PCR method for detecting Mycobacterium tuberculosis DNA in differentiating Crohn’s disease from intestinal tuberculosis. Chinese Journal of Pathophysiology. 2013;29(01):188–92. Chinese.
  38. 38. Scott LE, Beylis N, Nicol M, Nkuna G, Molapo S, Berrie L, et al. Diagnostic accuracy of Xpert MTB/RIF for extrapulmonary tuberculosis specimens: establishing a laboratory testing algorithm for South Africa. J Clin Microbiol. 2014;52(6):1818–23. pmid:24622091
  39. 39. Bera C, Michael JS, Burad D, Shirly SB, Gibikote S, Ramakrishna B, et al. Tissue Xpert™ MTB/Rif assay is of limited use in diagnosing peritoneal tuberculosis in patients with exudative ascites. Indian Journal of Gastroenterology. 2015;34(5):395–8. pmid:26531065
  40. 40. Sekine K, Nagata N, Shindo T, Morino E, Shimbo T, Akiyama J, et al. Combined identifying granuloma and biopsy culture is useful for diagnosing intestinal tuberculosis. International Journal of Colorectal Disease. 2015;30(7). pmid:25868516
  41. 41. Yadav SK, Jha NK, Sinha DK, Singla H, Kumar S, Yadav J, et al. Evaluation of the role of ascitic fluid polymerase chain reaction targeting IS6110 of Mycobacterium tuberculosis in the diagnosis of tuberculous intestinal obstruction. Annals of Tropical Medicine and Public Health. 2015;8(5):198–201.
  42. 42. Yang B, Wang X, Jiang J, Qu F, Cao Z, Cheng X. Application of ring-mediated isothermal amplification in the diagnosis of tuberculosis. Journal of Clinical Pulmonology. 2015(3):423–5. Chinese.
  43. 43. Mei J, Lu Y, Liu H. Observations on the use of Xpert MTB/RIF test in the rapid diagnosis of intestinal tuberculosis. Journal of Practical Medicine. 2016;32(18):3073–5. Chinese.
  44. 44. Kumar S, Bopanna S, Kedia S, Mouli P, Dhingra R, Padhan R, et al. Evaluation of Xpert MTB/RIF assay performance in the diagnosis of abdominal tuberculosis. Intest Res. 2017;15(2):187–94. pmid:28522948
  45. 45. Li Y, Pang Y, Zhang T, Xian X, Wang X, Yang J, et al. Rapid diagnosis of extrapulmonary tuberculosis with Xpert Mycobacterium tuberculosis/rifampicin assay. J Med Microbiol. 2017;66(7):910–4. pmid:28708053
  46. 46. Rufai SB, Singh S, Singh A, Kumar P, Singh J, Vishal A. Performance of Xpert MTB/RIF on ascitic fluid samples for detection of abdominal tuberculosis. Journal of laboratory physicians. 2017;9(1):47‐52. pmid:28042217
  47. 47. Ullah I, Javaid A, Masud H, Ali M, Basit A, Ahmad W, et al. Rapid detection of Mycobacterium tuberculosis and rifampicin resistance in extrapulmonary tuberculosis and sputum smear-negative pulmonary suspects using Xpert MTB/RIF. J Med Microbiol. 2017;66(4):412–8. pmid:28425873
  48. 48. Bhumit P, D YV. Clinical and laboratory features of intestinal tuberculosis. Clinical and experimental gastroenterology. 2018;11.
  49. 49. Gui-Jun F, Li-Fan Z, Hui-Jun S. Values of Different Laboratory Diagnostic Approaches for Tuberculous Peritonitis. Zhongguo yi xue ke xue yuan xue bao Acta Academiae Medicinae Sinicae. 2018;40(4).
  50. 50. Rakotoarivelo R, Ambrosioni J, Rasolofo V, Raberahona M, Rakotosamimanana N, Andrianasolo R, et al. Evaluation of the Xpert MTB/RIF assay for the diagnosis of smear-negative pulmonary and extrapulmonary tuberculosis in Madagascar. Int J Infect Dis. 2018;69:20–5. pmid:29408360
  51. 51. Luo F, Huang C, Li B, Long S, Liao W, Zhu Z, et al. The value of Mycobacterium tuberculosis polymerase chain reaction assay in the differential diagnosis of intestinal tuberculosis and Crohn’s disease. Journal of Nanchang University (Medical Edition). 2018;58(02):14–8. Chinese.
  52. 52. Allahyartorkaman M, Mirsaeidi M, Hamzehloo G, Amini S, Zakiloo M, Nasiri MJ. Low diagnostic accuracy of Xpert MTB/RIF assay for extrapulmonary tuberculosis: A multicenter surveillance. Sci Rep. 2019;9(1):18515. pmid:31811239
  53. 53. Bellam BL, Mandavdhare HS, Sharma K, Shukla S, Soni H, Kumar MP, et al. Utility of tissue Xpert-Mtb/Rif for the diagnosis of intestinal tuberculosis in patients with ileocolonic ulcers. Ther Adv Infect Dis. 2019;6:2049936119863939. pmid:31431826
  54. 54. Dahale AS, Puri AS, Kumar A, Dalal A, Agarwal A, Sachdeva S. Tissue Xpert® MTB/RIF Assay in Peritoneal Tuberculosis: To be (Done) or Not to be (Done). Cureus. 2019;11(6):e5009. pmid:31497439
  55. 55. Malik S, Sharma K, Vaiphei K, Dhaka N, Berry N, Gupta P, et al. Multiplex Polymerase Chain Reaction for diagnosis of gastrointestinal tuberculosis. JGH Open. 2019;3(1):32–7. pmid:30834338
  56. 56. Mbuh TP, Ane-Anyangwe I, Adeline W, Thumamo Pokam BD, Meriki HD, Mbacham WF. Bacteriologically confirmed extra pulmonary tuberculosis and treatment outcome of patients consulted and treated under program conditions in the littoral region of Cameroon. BMC Pulm Med. 2019;19(1):17. pmid:30654769
  57. 57. Tadesse M, Abebe G, Bekele A, Bezabih M, Yilma D, Apers L, et al. Xpert MTB/RIF assay for the diagnosis of extrapulmonary tuberculosis: a diagnostic evaluation study. Clin Microbiol Infect. 2019;25(8):1000–5. pmid:30583052
  58. 58. Talib A, Bhatty S, Mehmood K, Naim H, Haider I, Lal H, et al. GeneXpert in stool: Diagnostic yield in Intestinal Tuberculosis. J Clin Tuberc Other Mycobact Dis. 2019;17:100131. pmid:31788572
  59. 59. Wang Q, Zhu F, Li Q, He F, Zhu M, Xu Y, et al. A comparative study of XPert MTB/RIF and MGIT for detection and sensitivity of Mycobacterium tuberculosis. Journal of Clinical Pulmonology. 2019;24(12):2218–21. Chinese.
  60. 60. Liu R, Li J, Tan Y, Shang Y, Li Y, Su B, et al. Multicenter evaluation of the acid-fast bacillus smear, mycobacterial culture, Xpert MTB/RIF assay, and adenosine deaminase for the diagnosis of tuberculous peritonitis in China. International Journal of Infectious Diseases. 2020;90. pmid:31693941
  61. 61. Lowbridge C, Fadhil SAM, Krishnan GD, Schimann E, Karuppan RM, Sriram N, et al. How can gastro-intestinal tuberculosis diagnosis be improved? A prospective cohort study. BMC Infectious Diseases. 2020;20(1). pmid:32228479
  62. 62. Salman AE, Hassan JS, Qudus WJ. Detection of mycobacterium tuberculosis in peritoneal dialysis patients. Annals of Tropical Medicine and Public Health. 2020;23(1).
  63. 63. Sharma M, Sinha SK, Sharma M, Singh AK, Samanta J, Sharma A, et al. Challenging diagnosis of gastrointestinal tuberculosis made simpler with multi-targeted loop-mediated isothermal amplification assay. European Journal of Gastroenterology and Hepatology. 2020:971–5. pmid:32433424
  64. 64. Song J. Wang D, Ren S. Clinical application of Xpert MTB/RIF technique in the diagnosis of intestinal tuberculosis. Contemporary Medicine. 2020;26(10):89–92. Chinese.
  65. 65. Mishra PK, Gorantla VR, Bhargava A, Varshney S, Vashistha P, Maudar KK. Molecular detection of Mycobacterium tuberculosis in formalin-fixed, paraffin-embedded tissues and biopsies of gastrointestinal specimens using real-time polymerase chain reaction system. Turkish Journal of Gastroenterology. 2010;21(2):129–34. pmid:20872325
  66. 66. Gan H, Ouyang Q, Bu H. The value of polymerase chain reaction assay in the diagnosis of intestinal tuberculosis and differentiation from Crohn’s disease. Zhonghua nei ke za zhi [Chinese journal of internal medicine]. 1995;34(1):30–3.
  67. 67. Gan H, Ou Y, Bu H, Yang X. The value of Mycobacterium tuberculosis DNA testing in the differential diagnosis of intestinal tuberculosis and Crohn’s disease. Chinese Journal of Gastroenterology. 2001(04):22–5. Chinese.
  68. 68. Ye F, Chen Y, He D, Jian SY, Zheng K, Li GD, et al. [Detection of Mycobacterium tuberculosis complex in paraffin-embedded tissues by real-time fluorescent quantitative polymerase chain reaction]. Zhonghua Bing Li Xue Za Zhi. 2013;42(8):534–7. pmid:24246919
  69. 69. Ahmad R, Changeez M, Khan JS, Qureshi U, Tariq M, Malik S, et al. Diagnostic Accuracy of Peritoneal Fluid GeneXpert in the Diagnosis of Intestinal Tuberculosis, Keeping Histopathology as the Gold Standard. Cureus. 2018;10(10):e3451. pmid:30564530
  70. 70. Kulkarni S, Vyas S, Supe A, Kadival G. Use of polymerase chain reaction in the diagnosis of abdominal tuberculosis. Journal of Gastroenterology and Hepatology. 2006;21(5). pmid:16704529
  71. 71. Ling DI, Flores LL, Riley LW, Pai M. Commercial nucleic-acid amplification tests for diagnosis of pulmonary tuberculosis in respiratory specimens: meta-analysis and meta-regression. PLoS One. 2008;3(2):e1536. pmid:18253484
  72. 72. Altez-Fernandez C, Ortiz V, Mirzazadeh M, Zegarra L, Seas C, Ugarte-Gil C. Diagnostic accuracy of nucleic acid amplification tests (NAATs) in urine for genitourinary tuberculosis: a systematic review and meta-analysis. BMC Infect Dis. 2017;17(1):390. pmid:28583076
  73. 73. Pormohammad A, Nasiri MJ, McHugh TD, Riahi SM, Bahr NC. A Systematic Review and Meta-analysis of the Diagnostic Accuracy of Nucleic Acid Amplification Tests for Tuberculous Meningitis. J Clin Microbiol. 2019;57(6). pmid:30944198
  74. 74. Yu G, Zhong F, Shen Y, Zheng H. Diagnostic accuracy of the Xpert MTB/RIF assay for tuberculous pericarditis: A systematic review and meta-analysis. PLoS One. 2021;16(9):e0257220. pmid:34506587
  75. 75. MacLean E, Kohli M, Weber SF, Suresh A, Schumacher SG, Denkinger CM, et al. Advances in Molecular Diagnosis of Tuberculosis. J Clin Microbiol. 2020;58(10). pmid:32759357
  76. 76. Lin SY, Hwang SC, Yang YC, Wang CF, Chen YH, Chen TC, et al. Early detection of Mycobacterium tuberculosis complex in BACTEC MGIT cultures using nucleic acid amplification. Eur J Clin Microbiol Infect Dis. 2016;35(6):977–84. pmid:27039339
  77. 77. Nicol MP, Whitelaw A, Wendy S. Using Xpert MTB/RIF. Curr Respir Med Rev. 2013;9:187–92. pmid:24089608
  78. 78. Nayagam JS, Mullender C, Cosgrove C, Poullis A. Abdominal tuberculosis: Diagnosis and demographics, a 10-year retrospective review from a single centre. World J Clin Cases. 2016;4(8):207–12. pmid:27574607
  79. 79. Kentley J, Ooi JL, Potter J, Tiberi S, O’Shaughnessy T, Langmead L, et al. Intestinal tuberculosis: a diagnostic challenge. Trop Med Int Health. 2017;22(8):994–9. pmid:28609809
  80. 80. Sharma V, Soni H, Kumar MP, Dawra S, Mishra S, Mandavdhare HS, et al. Diagnostic accuracy of the Xpert MTB/RIF assay for abdominal tuberculosis: a systematic review and meta-analysis. Expert Rev Anti Infect Ther. 2020. pmid:32845790