Histopathological and mpt64 Immunological diagnosis of extra pulmonary tuberculosis at Muhimbili National Hospital

Abstract

Background: Extra pulmonary tuberculosis (EPTB) is reported to be on the rise worldwide particularly in developing countries. In the year 2014, the EPTB represented about twenty one per cent of the new cases of tuberculosis (TB) reported in Tanzania(1). The diagnosis of EPTB is more challenging than pulmonary tuberculosis (PTB) since it cannot be diagnosed by either sputum examination or chest imaging. The diagnosis of EPTB is made by identification of characteristic histomorphologic features on tissue biopsies from the affected organ and confirmed by Ziehl Nielsen stain (ZNS) that shows Acid-fast bacilli (AFB). However, ZNS has a low sensitivity thus it can be negative despite the presence of TB infection. New cost effective methods to identify AFB on tissues are being developed. One of them is immunohistochemistry (IHC) method using anti-MPT64 to detect mycobacterium TB complex proteins. Various studies were performed especially on cytological material and lymphoid tissues; scant information is available for other extrapulmonary sites particularly using formalin-fixed paraffin-embedded (FFPE) tissue sections. Objectives: The study aimed at determining the EPTB frequency, demography, histomorphology, clinical presentation as well as anti-MPT64 immunoreactivity on FFPE archived surgical biopsies at Muhimbili National Hospital (MNH), Histopathology Unit.

Methods: FFPE tissue blocks and haematoxylin-eosin (H&E) stained slides of all biopsies which were signed out as EPTB from July 2015 to June 2017 were retrieved from the Histopathology Unit archive, at MNH. The slides of all included cases were retrieved and reviewed to confirm the diagnosis of EPTB by histomorphology prior to retrieval and resectioning for ZN staining and IHC. Demographic, clinical and histopathologic data were collected using excel data collection sheet exported to SPSS version 20 for analysis. Proportions of AFB positivity were calculated for ZNS and anti-MPT64 immunoreactivity. Continuous variables were summarized using mean, median, mode, standard deviation and range. Frequency tables, bar charts and pie charts were made using excel. The association between demographic, clinical data and positivity for AFB was analysed. The association was considered significant for a p-value less than 0. Results: A total of 110 out of 165 tissue biopsies which were signed out as extrapulmonary TB based on histomorphology were further analysed. Biopsies were from patients aged between 2-76 years, with mean, median and mode of 36.3, 33 and 30 years respectively. Biopsies from female patients represented 60% (n=66). Necrotizing granuloma was the most (69%, n=76) frequent lesion amongst the biopsies collected. Furthermore, IHC and ZNS positivity were found in 71% and 8% of the biopsies respectively. All ZN positive biopsies were also positive to MPT64 by IHC. The lymph node presentation was more frequent (41.8%, n=46) compared to other sites.

Conclusions: EPTB is not infrequently encountered among biopsies submitted at MNH. Despite the challenges faced in establishing EPTB diagnosis, the histomorphology of necrotizing granulomas should be regarded as a strong indicator of EPTB, particularly in TB endemic areas. Furthermore, IHC by anti-MPT64 appears to be useful particularly in controversial cases where ZN staining is negative. TB lymphadenitis is the most common type of EPTB presentation at MNH. EPTB affects mostly young adults in reproductive age group and the female gender.

Recommendations: EPTB should be considered as a differential diagnosis in various clinical scenarios as it presents in a protean ways including mimicking malignancy, thus histological and/or cytological confirmation is important. All patients who present with enlarged lymphnodes should have TB infection ruled out. For granulomatous lesions it is important to do further work up including special stains for microorganisms in order exclude other causes. Furthermore, clinicopathological correlation is important in order to reach the diagnosis of EPTB and whenever there is a controversy; advanced tests including immunohistochemistry should be employed.