Evaluation of molecular and bacteriological detection methods performed on the formalin-fixed paraffin-embedded...

Endometrial Tuberculosis is one of the most common gynecological problems known to have serious implications for the quality of life like infertility. The commonly practiced histopathology solely relies on the suggestive feature of Tuberculosis (TB) with low specificity. Regarding the alternative bacteriological and molecular detection tools, little evidence was generated on their utility in the diagnosis of endometrial tuberculosis in Ethiopia. Therefore, we aim to investigate the detection rate of molecular and bacteriological detection methods on formalin-fixed paraffin-embedded biopsy samples for the diagnosis of endometrial and lymph node TB.

A retrospective cross-sectional study was conducted on 90 formalin fixed paraffin embedded biopsy samples from patients with gynecologic and lymph problems collected between 2018 and 2022 at St. Paul's Hospital Millennium Medical College, Addis Ababa, Ethiopia. SPSS version 26 was used for statistical analysis. The diagnostic performance was calculated using the histopathology method as the reference standard. Cohen's Kappa value was used to measure the level of agreement. A test with a P-value of < 0.05 was considered statistically significant.

A total of 90 samples were analyzed in the current study. Auramine O, GeneXpert MTB/RIF assay, and Real-Time PCR tests have shown a detection rate of 32/90 (36%), 43/90 (47.8%), and 54/90 (60%) respectively (P ≤ 0.01). The sensitivity and specificity of AO were 38.1% and 95% respectively. RT PCR showed superior sensitivity followed by GeneXpert MTB/RIF assay, 70% and 58.6%. AO and molecular methods have shown a similarly low level of agreement with histopathology (Kappa value = 0.2).

In a resource-limited setting, the selection of diagnostic tools needs careful attention. Putting the patients on anti-TB treatments based solely on histopathological findings may lead to undesired and adverse complications. Therefore, applying molecular and bacteriological detection methods along with histopathology, could help minimize inappropriate antimicrobial use.