Diagnosis of extrapulmonary tuberculosis by ultrasound-guided biopsy: A retrospective comparison study

Objective To compare the diagnostic performance of laboratory assays on the ultrasound-guided core needle biopsy samples for diagnosis of extra-pulmonary tuberculosis (EPTB) in HIV-positive and HIV-negative patients. Methods A total of 217 patients suspected to have EPTB underwent lesion biopsy from 2017 to 2020. Results of laboratory tests on the biopsy and non-biopsy samples were collected with clinical data for retrospective analysis of test utility. The calculated diagnostic accuracy of the tests was stratified according to the specimen types and HIV status. Results The cohort contained 118 patients with a final positive diagnosis of extrapulmonary tuberculosis (EPTB group, 54.4%) and 99 finally diagnosed as without TB (non-EPTB group, 45.6%). The risk factor for EPTB was HIV co-infection (OR 2.22, 95% CI 1.17-4.28, p = 0.014). In biopsy samples, GeneXpert (Xpert) showed higher sensitivity (96.6% [91.6-98.7], p < 0.0001) than culture (56.1% [47.0-64.9]). Regardless of HIV status, Xpert had the highest sensitivity (>95%) and specificity (nearly 100%) of any methods. In non-biopsy samples, only T-SPOT.TB (T-SPOT) showed higher sensitivity than culture (90.9% [62.3-99.5] vs 35.3% [17.3-58.7], p = 0.0037). Furthermore, the sensitivities of Xpert were lower in non-biopsy samples (60.0% [23.1-92.9], p = 0.022) than in biopsy samples (100% [86.7-100]). Even in smear-negative biopsy samples, Xpert still had higher sensitivity than culture and retained high specificity (100% [95.7-100]). Conclusion Superior performance of Xpert in diagnosing EPTB was observed regardless of HIV status and specimen types. Nevertheless, the biopsy samples still substantially facilitated the accurate diagnosis of extrapulmonary tuberculosis.


Introduction
Tuberculosis (TB) is a communicable disease caused by the mycobacterium tuberculosis complex (MTBC). Globally, an estimated 10.6 million people (range, 9.9-11.0 million) fell ill with TB in 2021, an increase of 4.5% from 10.1 million (95% UI: 9.5-10.7 million) in 2020; and about 1.6 million died from TB in the same year, up from a best estimate of 1.5 million in 2020 (WHO, 2022).
Extrapulmonary tuberculosis (EPTB) refers to TB occurring in parts of the body other than the lungs (e.g., lymph nodes, meninges, abdomen, pleura, genitourinary tract, skin, joints, and bones) (Golden and Vikram, 2005). As per the Global TB Report 2020, EPTB constituted 16% of the 7.5 million notified TB cases in 2019, ranging from 8% in the Western Pacific Region to 24% in the Eastern Mediterranean Region (WHO, 2020). In China, EPTB accounted for approximately 24% of TB cases, with a maximum of 33% in the western region (Li et al., 2022). In the context of WHO's End TB Strategy, timely diagnosis and treatment of EPTB is a challenge we have to face.
The main risk factors associated with EPTB vary widely and include human immunodeficiency virus (HIV) co-infection, female sex, age (young children or over 65 years of age), and diabetes (Shivakoti et al., 2017;Ohene et al., 2019;Pang et al., 2019;Banta et al., 2020). Due to the absence of typical TB symptoms, EPTB is often misdiagnosed as other diseases, such as cancers (Xiang et al., 2021) and inflammatory diseases (Aisenberg et al., 2005;Jain, 2011). Laboratory diagnosis plays a decisive role in the diagnosis of EPTB. However, studies comparing various laboratory assays based on biopsy samples are limited, probably because biopsy samples are not readily available (Norbis et al., 2014;Park and Kon, 2021).
This study analyzed the records from laboratory investigations of specimens from suspected extrapulmonary tuberculosis patients in an infectious disease hospital from 2017 to 2020 to compare the accuracy of different methods of laboratory diagnosis.

Study population and specimens
This study was conducted in Shanghai Public Health Clinical Center, one of the designated National Tuberculosis Hospitals in China. Patients with suspected EPTB (WHO, 2021a) who had undergone biopsy between July 01 2017 and September 30 2020 were enrolled. The inclusion criteria were patients with lymph node enlargement and typical symptoms of TB (fever, wasting, night sweats, etc.), or a positive PPD/TSPOT.TB test, or suspicion of TB on imaging, and willing to receive puncture procedures. The exclusion criteria were the patient refusing the biopsy or patients with contraindications to puncture, such as coagulation dysfunction. The biopsy samples were collected by an ultrasoundguided core needle biopsy. For the non-biopsy samples, we collected data from the hospital's Laboratory Examination Control System by matching the patient's ID and the exact test date. Demographic information (sex, age, HIV status, and diagnosis) and anatomical locations of EPTB were recorded upon enrollment. The results of pathological and microbiological tests were included.

Clinical definition and classification
The culture (combined with the MPB64 test) and Xpert results were used as a microbiological reference standard. Patients were eventually classified into the EPTB group (culture (4 cases), Xpert (54 cases), or culture-Xpert (60 cases) positive) and non-EPTB group [culture and Xpert negative (99 cases)].

Laboratory methods
Biopsy samples were collected by ultrasound-guided biopsy in the Ultrasound Intervention Department and sent to the Laboratory and Pathology Departments for diagnostic tests and histological examinations. Non-biopsy samples were collected and tested routinely in the Laboratory Department. An optimized sample pre-treatment process was used to concentrate mycobacteria in the specimens and thus improve the accuracy of the assays (Rickman and Moyer, 1980;Peterson et al., 1999). Briefly, Largevolume liquid specimens were first centrifuged at 3000-3800g for 15 min, the supernatant was discarded and digested with 2-4% NaOH for 15-20 min. Solid samples were digested directly with 2-4% NaOH. After digestion, the samples were neutralized with sterile PBS, then centrifuged at 3000-3800g for 15 min and the supernatant was discarded. The digested samples were mixed with 0.1-1mL of PBS and used for subsequent assays. Routine tests included culture (BACTEC MGIT 960 rapid culture method), smear (Auramine O staining kit, Zhuhai Baso Biotechnology Co.), and Xpert (Gene X-Pert MTB/RIF, Cepheid, USA). T-SPOT.TB (Oxford Immunotec Ltd, UK), was carried out using kits based upon the hospital's programmatic laboratory procedures. Species identification was carried out with an MPB64 monoclonal antibody assay (Hangzhou Genesis Biodetection & Biocontrol Co., Ltd, Hangzhou, China) based on positive cultures. Next-generation sequencing is done by Shanghai Simple Gene Medical Laboratory (Kindstar Globalgene Technology, Inc. Shanghai, China) when required.
The pathological tissues were fixed with 4.0% formaldehyde, routinely dehydrated and paraffin-embedded, and serially sectioned at a thickness of 4 mm. HE stain and acid-fast stain (Zhuhai Baso Biotechnology Co. Zhuhai, China) were performed in sections for routine microscopic diagnosis. EPTB positive was identified when there was typical epithelioid granuloma formation, caseation, and positive acid-fast staining.

Statistical analysis
We used R studio version 4.0.0 to process the data and GraphPad Prism version 8.0 for all analyses. The baseline table was performed using the R-based tableone package (version 0.13.2). The c² test (including McNemar's test) was used to calculate differences in diagnostic accuracy metrics; the Mann-Whitney U test was used to calculate differences in non-parametric data; the two-sample proportion test (Chi-square test) was used to compare, for example, sensitivity across two groups.
Males were more likely to have EPTB than females (OR 1.89, 95%CI 1.10-3.29, p = 0.021). Using patients < 25 years of age as a control group, we found that patients exhibited an increased risk of extrapulmonary TB with increasing age (OR 1.16, 95% CI 0.60-2.25 for patients 25-44 years of age; OR 1.30, 95% CI 0.58-2.93 for patients 45-64 years of age; OR 2.05, 95% CI 0.76-5.76 for patients > 65 years of age). As expected, HIV-positive patients were more frequently affected by EPTB than HIV-negative ones (OR 2.22, 95% CI 1.17-4.28, p = 0.014). However, the lump diameter, pus volume, and length of patients' biopsy samples were not related to the likelihood of a positive diagnosis of EPTB (Table 1).
Firstly, we compared the diagnostic accuracy of conventional assays with 217 biopsy samples (  (Bennani et al., 2019). Unexpectedly, the sensitivity and NPV of the smear were slightly higher than that of the culture, but the difference was not significant (   Table S1).
In 53 non-biopsy samples ( , p = 0.0026). The sensitivity and specificity of Xpert and culture did not differ significantly, most likely due to the small sample size (Table 3). We also compared the performance of different assays between biopsy samples and non-biopsy samples. The

Discussion
In this retrospective analysis, we used biopsy samples and nonbiopsy samples from patients with presumptive EPTB to determine

FIGURE 3
Head-to-head comparison of test accuracy in biopsy samples, by HIV status. Data are shown stratified according to: (A) sensitivity, (B) specificity, (C) positive predictive value, (D) negative predictive value. Xu et al. 10.3389/fcimb.2023.1154939 Frontiers in Cellular and Infection Microbiology frontiersin.org the diagnostic accuracy, sensitivity, and specificity of the assays. Our key finding was that Xpert performed better than other laboratory assays regardless of the HIV status of the patients or the types of specimens. Overall, the biopsy samples provided more realistic pictures of the patient's conditions and a more accurate diagnosis of EPTB than non-biopsy samples. From the demographic aspects, several studies have reported similar findings that HIV co-infection and age (> 65 years old) contribute to EPTB infection (Lakoh et al., 2020;Winter et al., 2020;Barreto-Duarte et al., 2021), consistent with our results. However, we found that males were more likely to have EPTB than females (OR 1.89, 95%CI 1.10-3.29, p = 0.021), which was not consistent with some previous studies (Peto et al., 2009;Pang et al., 2019). This may be attributed to a higher proportion of HIV-positive men than women (34.8% vs 10.8%, p < 0.0001; Supplementary Table S3), although the relationship between gender and EPTB is controversial in current studies (Liu et al., 2020;Barreto-Duarte et al., 2021).
We found little statistical difference in the sensitivity, specificity, and predictive values of the non-biopsy samples-based Xpert compared to culture, mainly due to the small sample size. However, in agreement with previous studies, TSPOT showed a high sensitivity (70-100%) compared to culture and Xpert (Zhou et al., 2015;Li et al., 2020), predicting that TSPOT can be used as a powerful screening method for EPTB (Antel et al., 2020). Furthermore, the culture and Xpert performed better with biopsy samples than with non-biopsy samples, suggesting that biopsy is important for the accurate diagnosis of EPTB.
There are several limitations to this study. The smaller sample size of non-biopsy samples may affect the methodological comparison between non-biopsy samples and biopsy samples. The small number of samples assayed by various methods in non-biopsy samples was not conducive to evaluating the diagnostic accuracy of the method, for instance, the sensitivity of Xpert may be underestimated. In addition, we did not exclude patients with both pulmonary and extrapulmonary TB (42 cases), which may affect the comparison of assays between biopsy and non-biopsy samples.
In summary, our study compared the diagnostic accuracy of commonly used EPTB diagnostic methods across HIV status and sample types, highlighting the superiority of Xpert in different clinical settings and the critical contribution of biopsy samples in the diagnosis of EPTB. Further clinical studies evaluating the performance of the different laboratory assays in extrapulmonary samples and HIV populations are warranted to help clinicians choose the best diagnostic methods when faced with various dilemmas.

Data availability statement
The original contributions presented in the study are included in the article/Supplementary Material. Further inquiries can be directed to the corresponding authors.

Ethics statement
The studies involving human participants were reviewed and approved by Ethics Committee of Shanghai Public Health Clinical Center (2019-S030-02). Written informed consent to participate in this study was provided by the participants' legal guardian/next of kin. Shown are the numbers of positive results, n/N and % with 95% CI in parentheses, with p-values where appropriate. NPV, negative predictive value; PPV, positive predictive value. Bold means the P value is less than 0.05, with a statistical difference.

Author contributions
Project concept conceived (X-YF and HZ), experiments performed (J-CX and W-FG), sampling (XM and XS), data analysis (J-CX, XM, W-FG, X-YF, and HZ), and paper writing (J-CX and X-YF). All authors contributed to the article and approved the submitted version.

Funding
This study was supported by the National Key Research and Development Program of China (2022YFC2302900, 2021YFC2301503), National Natural and Science Foundation of China (82171815, 82171739), Shanghai Municipal Health Bureau (2022XD060) and Shanghai Science and Technology Commission (20Y11903400, 18411970800).