Diagnostic value of CD4, CD8 and CD68 immunohistochemical profiling in EBUS-TBNA samples for differentiating sarcoidosis and tuberculosis

Abstract

Background

Differentiating sarcoidosis from tuberculosis in patients with mediastinal and/or hilar granulomatous lymphadenopathy remains challenging due to overlapping clinical, radiological, and histopathological findings. Endobronchial ultrasound-guided transbronchial needle aspiration (EBUS-TBNA) allows minimally invasive sampling of lymph nodes; however, histopathological evaluation alone may be insufficient for accurate differentiation. This study aimed to investigate the diagnostic value of CD4 and CD8 T lymphocyte profiles and CD68 expression assessed by immunohistochemistry in EBUS-TBNA samples.

Methods

This retrospective single-center study included patients who underwent EBUS-TBNA for mediastinal and/or hilar lymphadenopathy between January and December 2019 and were diagnosed with granulomatous lymphadenitis. CD4 + and CD8 + T lymphocyte counts, percentages, CD4/CD8 ratios, and CD68 staining intensity were evaluated immunohistochemically in cell block specimens. Patients were classified as sarcoidosis or tuberculosis based on clinical, radiological, microbiological, and follow-up data. Diagnostic performance was assessed using receiver operating characteristic (ROC) analysis.

Results

Eighty-five patients were included (sarcoidosis: n = 55; tuberculosis: n = 30). CD4 + T lymphocyte count and percentage were significantly higher in the sarcoidosis group (p < 0.001 and p = 0.005, respectively). The CD4/CD8 ratio was also significantly elevated in sarcoidosis (median 1.44 vs. 0.35, p = 0.004). ROC analysis demonstrated moderate diagnostic performance for CD4 count (AUC 0.770), CD4% (AUC 0.686), and CD4/CD8 ratio (AUC 0.717). CD68 staining intensity did not differ significantly between groups. Granuloma necrosis was more frequent in tuberculosis (23.3% vs. 7.3%, p = 0.046).

Conclusions

Immunohistochemical evaluation of CD4 + T lymphocyte profiles and CD4/CD8 ratio in EBUS-TBNA samples provides meaningful adjunctive information in differentiating sarcoidosis from tuberculosis. Integration of immunophenotypic analysis into routine pathological assessment may improve diagnostic accuracy in patients with granulomatous mediastinal lymphadenopathy.

Introduction

Mediastinal and hilar lymphadenopathy are common clinical findings associated with a wide spectrum of diseases, including infectious, inflammatory, and malignant conditions [1–3].

Among these, granulomatous lymphadenitis represents a heterogeneous group of disorders with diverse etiologies. Infectious causes, particularly Mycobacterium tuberculosis, and non-infectious causes such as sarcoidosis are the most frequent underlying conditions [4–6].

Differentiating sarcoidosis from tuberculosis remains a major clinical challenge, especially in regions with intermediate to high tuberculosis prevalence. Both diseases may present with similar clinical symptoms, radiological findings, and histopathological features, including granuloma formation [2, 3, 6, 7]. While microbiological confirmation remains the gold standard for tuberculosis diagnosis [8, 9], no definitive diagnostic test exists for sarcoidosis. Therefore, the diagnosis of sarcoidosis is typically based on compatible clinical and radiological findings, histopathological evidence of non-caseating granulomas, and exclusion of alternative causes [4, 10].

In clinical practice, this diagnostic overlap may lead to misdiagnosis and inappropriate treatment. Particularly in tuberculosis-endemic regions, patients with sarcoidosis may be unnecessarily exposed to anti-tuberculosis therapy, whereas delayed diagnosis of tuberculosis may result in disease progression and transmission [5, 11].

Endobronchial ultrasound-guided transbronchial needle aspiration (EBUS-TBNA) has emerged as a minimally invasive and highly effective method for sampling mediastinal and hilar lymph nodes. It has largely replaced more invasive procedures such as mediastinoscopy due to its high diagnostic yield and favorable safety profile [12–16]. EBUS-TBNA allows not only histopathological evaluation but also additional analyses, including microbiological, molecular, and immunohistochemical assessment [17, 18].

Establishing a diagnosis of intrathoracic tuberculous lymphadenitis remains challenging because microbiological confirmation rates are often suboptimal [11, 17]. Mediastinal tuberculous lymphadenitis represents a paucibacillary form of extrapulmonary tuberculosis, in which uneven bacillary distribution and low mycobacterial burden contribute to the limited sensitivity of conventional microbiological methods [5, 11, 17, 19]. Acid-fast bacilli (AFB) smear microscopy from BAL or EBUS-TBNA specimens has high specificity but relatively low sensitivity [17, 20, 21]. Although mycobacterial culture remains the diagnostic gold standard, its prolonged turnaround time may delay clinical decision-making [11, 17].

Molecular assays such as TB-PCR and Xpert MTB/RIF have improved diagnostic performance in EBUS-TBNA specimens; however, negative molecular results do not reliably exclude tuberculosis, particularly in paucibacillary intrathoracic lymph node disease [17, 21]. Therefore, histopathological assessment together with clinical and radiological correlation remains essential for establishing an accurate diagnosis [2, 3, 7, 22, 23]. This diagnostic uncertainty further highlights the potential value of adjunctive tissue-based biomarkers such as the immunohistochemical parameters evaluated in the present study [24, 25].

The immunopathogenesis of granulomatous diseases is largely driven by cellular immune responses. T lymphocytes, particularly CD4 + and CD8 + subsets, and macrophages play a central role in granuloma formation and maintenance [26, 27]. Sarcoidosis is characterized by a Th1-dominant immune response with accumulation of activated CD4 + T lymphocytes, whereas tuberculosis involves a complex host–pathogen interaction with immune modulation by Mycobacterium tuberculosis [5, 19, 27].

Previous studies have investigated CD4/CD8 T lymphocyte ratios, particularly in bronchoalveolar lavage (BAL) fluid, demonstrating higher ratios in sarcoidosis compared to tuberculosis [28–30]. However, BAL findings may not accurately reflect the immunological microenvironment within lymph nodes [31, 32]. Data on immunohistochemical evaluation of lymphocyte subsets in EBUS-TBNA samples remain limited.

Therefore, this study aimed to evaluate the diagnostic value of CD4 and CD8 T lymphocyte profiles and CD68 expression in EBUS-TBNA samples for differentiating sarcoidosis from tuberculosis. By integrating immunophenotypic analysis with conventional histopathological assessment, we sought to improve diagnostic accuracy in patients with granulomatous mediastinal lymphadenopathy.

Methods

Study design and patient selection

This retrospective, single-center study included patients who underwent EBUS-TBNA between January 1 and December 31, 2019 due to mediastinal and/or hilar lymphadenopathy. Among these, patients with histopathologically confirmed granulomatous lymphadenitis and at least 6 months of clinical follow-up were included (Fig. 1).

Fig. 1.

Flow diagram of patient selection, including inclusion and exclusion criteria, and final study cohort

Diagnostic classification

Clinical diagnosis was established by multidisciplinary evaluation integrating clinical, radiological, pathological, microbiological, and follow-up findings. Sarcoidosis was diagnosed after exclusion of alternative diagnoses, whereas tuberculosis was diagnosed by microbiological confirmation or compatible clinical and radiological findings together with response to anti-tuberculosis therapy.

Of the 30 patients with tuberculosis, 3 (10.0%) had microbiologically confirmed disease based on bacteriological positivity, whereas the remaining 27 patients (90.0%) were diagnosed clinically based on compatible clinical and radiological findings. These patients completed 6 months of anti-tuberculosis therapy, demonstrated clinical and radiological improvement, and had a mean follow-up duration of 12 months without an alternative diagnosis emerging during follow-up.

EBUS-TBNA procedure

EBUS-TBNA was performed under local anesthesia and conscious sedation using a convex probe bronchoscope (Olympus BFUC160F 7.5 MHz Olympus Co Ltd, Japan) and 22-gauge needles (NA-201SX-4022, 22 gauge, Olympus, Tokyo, Japan). Lymph nodes with a short-axis diameter ≥ 10 mm were sampled under real-time ultrasound guidance. At least two passes were performed for each lymph node.

Immunohistochemical analysis

Immunohistochemistry was performed using a fully automated system with CD4, CD8, and CD68 antibodies. Detection was achieved using the ultraView Universal DAB Kit.

Cell block sections were pretreated at 72 °C for 30 min and stained with monoclonal antibodies against CD4, CD8, and CD68. Antigen retrieval was performed with EDTA (CC1) at 95 °C. Primary antibody incubation was performed at 37 °C, followed by standard secondary antibody, DAB visualization, copper enhancement, and hematoxylin counterstaining.

Immunohistochemical evaluation

CD4- and CD8-positive lymphocytes and CD68-positive macrophages were identified by comparison with the corresponding hematoxylin and eosin (H&E)-stained sections. Dark brown cytoplasmic/membranous staining was considered positive. CD4- and CD8-positive lymphocytes were assessed in 15 randomly selected representative high-power fields (×400) containing granulomatous inflammation, and the total number of positively stained cells was recorded for each specimen. CD68 expression was evaluated separately at ×200 magnification using a semi-quantitative scoring system. Staining intensity was graded as follows: 0 = no staining, 1 = weak staining in a minority of cells, 2 = moderate staining in approximately 30% of cells, 3 = strong staining in more than 50% of cells, and 4 = strong staining in nearly all cells. Results were expressed as CD4 + and CD8 + T-cell counts and percentages, and CD68 staining intensity. All immunohistochemical evaluations were performed independently by a single experienced pulmonary pathologist who was blinded to the final clinical diagnosis of the patients. Since all specimens were assessed by a single observer, interobserver agreement analysis was not applicable. Representative images are shown in Figs. 2 and 3.

Fig. 2.

Immunohistochemical staining of granulomatous inflammation. A CD4-positive lymphocytes surrounding granulomatous inflammation. B CD8-positive lymphocyte aggregates. (×200)

Fig. 3.

CD68 immunostaining in granulomatous inflammation. Granulomas composed of CD68-positive epithelioid histiocytes. (×400)

The study was approved by the relevant institutional ethics committee (approval date: July 26, 2019; decision no: 1916), as detailed in the Declarations section.

Statistical analysis

Statistical analyses were performed using Stata (version 15.1; StataCorp, College Station, TX, USA) and XLSTAT (version 2021.2.2; Addinsoft, New York, USA). Continuous variables were assessed for normality using the Shapiro–Wilk test and are presented as median (interquartile range, IQR). Categorical variables are expressed as frequency and percentage.

Comparisons between groups were performed using the Mann–Whitney U test for continuous variables and the chi-square or Fisher’s exact test for categorical variables.

The diagnostic performance of CD4, CD8, and the CD4/CD8 ratio for predicting sarcoidosis was evaluated using receiver operating characteristic (ROC) curve analysis. Sensitivity, specificity, positive predictive value, and negative predictive value were calculated for significant cut-off values. Optimal cut-off values were determined using the Youden index (sensitivity + specificity − 1).

A p-value < 0.05 was considered statistically significant.

Results

A total of 1,709 patients who underwent EBUS-guided transbronchial needle aspiration for mediastinal and/or hilar lymphadenopathy between January 1, 2019 and December 31, 2019 were screened, and 85 patients meeting the study criteria were included. Of these, 55 (64.7%) had sarcoidosis and 30 (35.3%) had tuberculosis.

The overall cohort included 31 males (36.5%) and 54 females (63.5%), with a mean age of. 48.0 ± 14.5 years. There were no significant differences between the sarcoidosis and tuberculosis groups in terms of age or sex (p = 0.989 and p = 0.056, respectively). The subcarinal and paratracheal stations were the most frequently sampled lymph node stations in both groups. Granuloma necrosis was significantly more frequent in the tuberculosis group compared to sarcoidosis (23.3% vs. 7.3%, p = 0.046). Demographic characteristics are summarized in Table 1.

Table 1.

Demographic characteristics and necrosis status of patients with sarcoidosis and tuberculosis

	Sarcoidosis	Tuberculosis	Total
	(n = 55)	(n = 30)	(n = 85)
	n (%)	n (%)	n (%)	p value
Age (mean ± SD)	47.4 ± 12.9	49.2 ± 17.3	48.0 ± 14.5	0.989
Gender				0.056
Male	16 (29.1%)	15 (50.0%)	31 (36.5%)
Female	39 (70.9%)	15 (50.0%)	54 (63.5%)
Comorbidity				0.178
Yes	21 (38.2%)	16 (53.3%)	37 (43.5%)
No	34 (61.8%)	14 (46.7%)	48 (56.5%)
Necrosis				0.046
Yes	4 (7.3%)	7 (23.3%)	11 (12.9%)
No	51 (92.7%)	23 (76.7%)	74 (87.1%)

Among the tuberculosis patients, 3 (10.0%) had microbiologically confirmed disease and 27 (90.0%) were diagnosed clinically based on compatible clinical and radiological findings

No significant differences were observed between groups in laboratory parameters, including hemoglobin, white blood cell count, inflammatory markers, liver and renal function tests, calcium, and ACE levels.

Lymphocyte profile analysis

CD4 + T-cell counts and CD4 + T-cell percentages were significantly higher in the sarcoidosis group compared to tuberculosis (p < 0.001, p = 0.005). No significant differences were found between groups in CD8 + T-cell counts or percentages. The CD4/CD8 ratio was significantly higher in sarcoidosis compared to tuberculosis (1.44 vs. 0.35, p = 0.004). These findings are presented in Table 2; Fig. 4.

Table 2.

Comparison of CD4 + and CD8 + T lymphocyte cell counts, percentages, and CD4/CD8 ratios

	Sarcoidosis (n = 55)	Tuberculosis (n = 30)	Total (n = 85)	p value
Variable	Median (IQR)	Median (IQR)	Median (IQR)
CD4 + T	83.0	19.0	49.0	p < 0.001
lymphocyte count	(33.0-223.0)	(8.0–66.0)	(18.0-183.0)
CD4 + T	18.5	6.1	11.5	0.005
lymphocyte percentage	(6.0-42.5)	(2.0-23.3)	(5.0-37.5)
(%)
CD8 + T	71.0	59.5	67.0	0.157
lymphocyte count	(30.0-164.0)	(22.0–98.0)	(30.0-127.0)
CD8 + T	12.0	17.5	14.0	0.244
lymphocyte percentage	(6.0–26.0)	(8.0–27.0)	(6.0–26.0)
(%)
CD4/CD8	1.44	0.35	1.0	0.004
ratio	(0.53–3.21)	(0.13–0.82)	(0.0–2.0)

Fig. 4.

Distribution of CD4 + T lymphocyte counts (A), CD4 + T lymphocyte percentages (B), and CD4/CD8 ratios (C) among the study population

Diagnostic performance (ROC analysis)

ROC analysis demonstrated that CD4 + T-cell count had good diagnostic performance for differentiating sarcoidosis from tuberculosis (AUC: 0.770, 95% CI: 0.666–0.854, p < 0.001). At a cut-off value of > 33, sensitivity was 74.6% and specificity was 70.0%. CD4 + T-cell percentage also showed diagnostic value (AUC: 0.686, 95% CI: 0.576–0.782, p = 0.003), whereas CD8 + T-cell count and percentage were not significant. The CD4/CD8 ratio demonstrated moderate diagnostic performance (AUC: 0.717, 95% CI: 0.609–0.810, p < 0.001). At a cut-off value of > 0.55, sensitivity was 74.6% and specificity was 66.7% (Fig. 5).

Fig. 5.

Receiver operating characteristic (ROC) curves. A ROC curve of CD4 + T-lymphocyte count for differentiating sarcoidosis and tuberculosis. CD4, B ROC curve for CD4 + T-lymphocyte percentage, and (C) ROC curve for CD4/CD8 ratio. The diagnostic performance of each parameter is presented with corresponding AUC, sensitivity, specificity, and cut-off values

CD68 staining

CD68 staining intensity was high in both groups, with no significant difference between sarcoidosis and tuberculosis. Most cases showed strong (grade 4) staining in both groups (Table 3).

Table 3.

Comparison of CD68 immunostaining intensity in sarcoidosis and tuberculosis

CD68 immunostaining intensity	Sarcoidosis	Tuberculosis	Total
CD68 (3+) immunostaining	1	1	2
CD68 (4+) immunostaining	53	29	82

No cases demonstrated grade 0–2 staining

Discussion

In this study, we demonstrated that immunohistochemical evaluation of CD4 + and CD8 + T lymphocyte profiles in EBUS-TBNA-derived lymph node samples provides meaningful additional value in differentiating sarcoidosis from tuberculosis. Our findings showed that CD4 + T-cell count, CD4 + T-cell percentage, and CD4/CD8 ratio were significantly higher in sarcoidosis compared to tuberculosis, whereas CD8 + T-cell parameters did not differ significantly. ROC analysis further demonstrated that CD4 + T-cell count and CD4/CD8 ratio had moderate diagnostic performance, supporting their potential role as adjunctive diagnostic markers. CD68 expression did not differ between groups, indicating similar macrophage involvement in both diseases [1–3, 26].

The predominance of CD4 + T lymphocytes in sarcoidosis is consistent with its well-established Th1-driven immunopathogenesis. Activated CD4 + T cells accumulate within granulomatous lesions and produce cytokines such as interferon-γ and interleukin-2, which promote macrophage activation and granuloma maintenance [27, 33–35] Tuberculosis involves a complex host–pathogen interaction that modulates host immune responses, leading to altered T-cell function and relatively lower CD4/CD8 ratios [5, 19].

The CD4/CD8 ratio has been widely investigated in bronchoalveolar lavage (BAL) fluid; however, BAL findings may not accurately reflect the immunological microenvironment within lymph nodes [28–30, 36, 37]. Previous studies have reported variable CD4/CD8 ratios depending on the sampling method and analytical technique. Oda et al. demonstrated a correlation between BAL and lymph node CD4/CD8 ratios [38], whereas Akao et al. reported differences between these compartments [39]. In contrast, Ruiz et al. found limited concordance between BAL and lymph node findings [40]. Our study extends these observations by demonstrating that immunophenotypic differences can also be detected directly in lymph node tissue obtained via EBUS-TBNA.

Recent studies have highlighted the limitations of EBUS-TBNA in differentiating granulomatous diseases based solely on cytomorphological findings [41]. Although EBUS-TBNA provides a high diagnostic yield and has largely replaced more invasive procedures for mediastinal sampling [42–48]distinguishing sarcoidosis from tuberculosis remains challenging because of their overlapping histopathological features [22].

In daily clinical practice, non-necrotizing granulomas identified in EBUS-TBNA specimens are frequently interpreted as suggestive of sarcoidosis, particularly when microbiological investigations are negative. However, early-stage tuberculous lymphadenitis may also present with non-necrotizing granulomatous inflammation, resulting in substantial histopathological overlap between the two diseases [2, 3, 7, 23]. Granuloma morphology may vary according to.

disease stage and host immune response. While advanced tuberculosis is more commonly associated with caseating necrosis, early disease may closely mimic sarcoidosis [5, 8, 19, 49]. Consequently, conventional morphology alone may be insufficient for a confident diagnosis in patients lacking overt necrosis. In such circumstances, the significantly higher CD4 + T-cell counts and CD4/CD8 ratios observed in our study may provide useful adjunctive information and improve diagnostic confidence [25, 27, 33].

Necrosis itself may also influence the diagnostic performance of EBUS-TBNA. Extensive nodal necrosis can reduce the amount of viable cellular material available for cytological and immunohistochemical evaluation, thereby affecting both morphological interpretation and ancillary analyses [17, 20, 50]. Although necrosis was significantly more frequent in the tuberculosis group in our cohort, adequate diagnostic material was obtained in most patients, allowing successful immunohistochemical assessment. Nevertheless, sampling limitations related to extensive necrosis should be considered when interpreting EBUS-TBNA findings in tuberculous lymphadenitis [17, 18, 24]. Consistent with previous reports, the presence of necrosis favored tuberculosis in our study; however, its absence did not exclude the diagnosis, particularly in early-stage disease [2, 3, 11].

Several studies have described characteristic EBUS sonographic differences between sarcoidosis and tuberculous lymphadenitis. Tuberculous lymph nodes more frequently demonstrate heterogeneous echotexture, coagulation necrosis signs, hypoechoic areas, and indistinct margins, whereas sarcoidosis is more commonly associated with homogeneous echotexture and well-defined borders [7, 17, 18, 20, 21]. However, substantial overlap exists between these sonographic characteristics, limiting their diagnostic specificity [22, 24, 41, 43].

Therefore, tissue acquisition remains essential, and adjunctive biomarkers such as CD4 + T-cell counts and CD4/CD8 ratios may further improve diagnostic confidence [24, 25].

In this context, our findings support the growing body of evidence suggesting that ancillary diagnostic methods are required to improve diagnostic accuracy. Dhooria et al. emphasized that integrating additional techniques, including immunological markers, may enhance the diagnostic performance of EBUS-TBNA [24]. Likewise, Mukhopadhyay et al. highlighted the need for multimodal diagnostic approaches in granulomatous diseases because of their overlapping pathological features [23]. Our results are consistent with these observations and suggest that CD4/CD8 T-lymphocyte profiling may serve as a practical adjunctive tool in this setting.

Despite the significant differences observed, the diagnostic performance of the CD4/CD8 ratio in our study was moderate, indicating that this parameter should not be used in isolation but rather in conjunction with clinical, radiological, microbiological, and histopathological findings [20, 50]. The low microbiological confirmation rate observed in our tuberculosis cohort is consistent with previous reports and may be explained by the paucibacillary nature of lymph node tuberculosis and uneven bacillary distribution [11, 17]. In addition, TB-PCR results were not systematically available for all patients because of the retrospective design and variations in routine clinical practice. Consequently, molecular findings could not be included in the statistical analyses. Future prospective studies incorporating standardized molecular testing may further clarify the complementary roles of microbiological and immunohistochemical approaches in the differential diagnosis of granulomatous lymphadenitis [17, 21, 24].

CD68 staining intensity did not differ significantly between the two groups in our study. This finding is biologically plausible because macrophages constitute the major cellular component of granulomas in both sarcoidosis and tuberculosis. CD68 primarily reflects macrophage presence rather than functional activation or polarization. Therefore, although CD68 staining supports the identification of granulomatous inflammation, it is unlikely to discriminate between these two granulomatous diseases [26, 49]. Alternatively, the absence of a statistically significant difference may also reflect the limited sensitivity of the semi-quantitative scoring system used in this study rather than true biological similarity in macrophage activity between sarcoidosis and tuberculosis.

The integration of immunohistochemical CD4/CD8 profiling into routine evaluation of EBUS-TBNA samples may significantly enhance diagnostic accuracy. This approach may reduce the need for invasive procedures and facilitate earlier, more appropriate treatment decisions, particularly in tuberculosis-endemic regions. Furthermore, tissue-based immunophenotypic evaluation may provide a more direct reflection of the local immune microenvironment than BAL-based analyses and therefore represents a promising adjunct in the differential diagnosis of granulomatous mediastinal lymphadenopathy.

Limitations

This study has several limitations, including its retrospective single-center design and relatively small sample size. A minimum follow-up period was required to establish a reliable final diagnosis in patients with granulomatous lymphadenitis; although excluding patients without adequate follow-up may have introduced some degree of selection bias, this criterion was predefined to ensure diagnostic reliability and reduce the risk of disease misclassification. Only a minority of tuberculosis cases were microbiologically confirmed, and TB-PCR results were not systematically available for all patients; most cases were instead diagnosed based on compatible clinical and radiological findings together with a favorable response to anti-tuberculosis therapy. Although all clinically diagnosed patients showed sustained clinical and radiological improvement during a mean follow-up of 12 months, diagnostic misclassification cannot be completely excluded. In addition, extensive nodal necrosis may have affected sample adequacy in some tuberculosis cases. Finally, standardized tissue-based cut-off values for CD4 + T-cell counts and CD4/CD8 ratios have not yet been established. Larger prospective multicenter studies are needed to validate our findings and define clinically applicable diagnostic thresholds.

Conclusion

In conclusion, CD4 + T lymphocyte parameters and CD4/CD8 ratio in EBUS-TBNA samples provide valuable adjunctive information for differentiating sarcoidosis from tuberculosis. These findings support the use of tissue-based CD4/CD8 immunophenotyping as an adjunctive diagnostic tool that should be interpreted together with clinical, radiological, microbiological, and histopathological findings.

Abbreviations

AUC

Area under the curve BAL: Bronchoalveolar lavage CD: Cluster of differentiation

CD4/CD8 ratio

Ratio of CD4-positive to CD8-positive T lymphocytes CI: Confidence interval

DAB

Diaminobenzidine

EBUS-TBNA

Endobronchial ultrasound-guided transbronchial needle aspiration EDTA: Ethylenediaminetetraacetic acid

ILD

Interstitial lung disease IQR: Interquartile range

MDD

Multidisciplinary discussion

ORCID

Open Researcher and Contributor ID

ROC

Receiver operating characteristic TB: Tuberculosis

Th1

T helper type 1

Authors’ contributions

M.G.O. contributed to study conception and design, data interpretation, and critical revision of the manuscript.H.N.Ü. performed histopathological and immunohistochemical evaluation and contributed to data interpretation.M.A.U. contributed to study design, patient management, and clinical data interpretation, contributed to statistical analysisB.A.B. participated in data collection, data analysis, and manuscript drafting.E.S.A. contributed to statistical analysis and interpretation of the results.M.A.P. contributed to data acquisition and clinical evaluation of patients.U.İ. contributed to study supervision, final review, and critical revision of the manuscript.All authors read and approved the final manuscript.

Funding

The authors received no specific funding for this work.

Data availability

Data Availability StatementThe datasets generated and/or analyzed during the current study are not publicly available due to patient privacy and institutional restrictions but are available from the corresponding author upon reasonable request.

Declarations

Ethics approval and consent to participate

This study was conducted at the Department of Chest Diseases, University of Health Sciences Turkey, Yedikule Chest Diseases and Thoracic Surgery Training and Research Hospital. As this hospital did not have an independent clinical research ethics committee at the time of the study, ethical approval was obtained from the Clinical Research Ethics Committee of University of Health Sciences Turkey İstanbul Training and Research Hospital (approval date: July 26, 2019; decision no: 1916). Due to the retrospective design of the study, the requirement for informed consent was waived by the Institutional Review Board. All procedures involving human participants were performed in accordance with the ethical standards of the institutional research committee and with the 1964 Declaration of Helsinki and its later amendments or comparable ethical standards.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.