Differentiating transient from persistent diabetic range hyperglycemia in a cohort of people completing tuberculosis treatment in Dhaka, Bangladesh.

Yosra M A Alkabab,Scott Heysell,Shahriar Ahmed,Jyothi Nagajyothi,Kishor Paul,Samanta Biswas,Sayera Banu,Venkata Naga Srikanth Garikipati

doi:10.1371/journal.pone.0260389

Abstract

BackgroundIn recent non-pandemic periods, tuberculosis (TB) has been the leading killer worldwide from a single infectious disease. Patients with DM are three times more likely to develop active TB and poor treatment outcomes. Single glycemic measurements at TB diagnosis may inaccurately diagnose or mischaracterize DM severity. Data are limited regarding glycemic dynamics from TB diagnosis through treatment.MethodsProspective study of glycemia dynamics in response to TB treatment measured glycosylated haemoglobin (HbA1c) in patients presenting to TB screening centres in Bangladesh to determine the prevalence and risk factors of hyperglycemia before and at TB treatment completion.Results429 adults with active TB disease were enrolled and divided into groups based on history of DM and initial HbA1c range: normoglycemia, prediabetes, and DM. DM was diagnosed in 37%. At treatment completion,14(6%) patients from the normoglycemia and prediabetes groups had HbA1c>6.5%, thus increasing the prevalence of DM to 39%. The number needed to screen to diagnose one new case of DM at TB diagnosis was 5.7 and 16 at treatment completion in the groups without DM. Weight gain>5% at treatment completion significantly increased the risk of hyperglycemia in the groups without DM at TB diagnosis (95% CI 1.23–26.04, p<0.05).ConclusionHbA1c testing prior to and at TB treatment completion found a high prevalence of prediabetes and DM, including a proportion found at treatment completion and commonly in people with a higher percentage of weight gain. Further longitudinal research is needed to understand the effects of TB disease and treatment on insulin resistance and DM complications.

Highlights

In recent non-pandemic periods, tuberculosis (TB) has been the leading killer worldwide from a single infectious disease [1]
429 adults with active TB disease were enrolled and divided into groups based on history of diabetes mellitus (DM) and initial HbA1c range: normoglycemia, prediabetes, and DM
Weight gain>5% at treatment completion significantly increased the risk of hyperglycemia in the groups without DM at TB diagnosis

Summary

Introduction

In recent non-pandemic periods, tuberculosis (TB) has been the leading killer worldwide from a single infectious disease [1]. With a quarter of the world’s population infected with Mycobacterium tuberculosis (MTB), conservative models have forecasted a 20% increase in TB deaths in the five years after COVID-19 has waned compared to pre-pandemic mortality, in large part due to interruptions in TB diagnostic and therapeutic services [2]. Patients with DM-TB have a higher MTB bacillary burden than patients without DM, leading to a delayed bacillary clearance from the sputum, and subsequently, an increased risk of poor treatment outcomes: death, drug resistance, and relapse [4, 5]. Despite consistent reports that people with DM are three times more likely to develop active TB than people without DM, laboratory diagnostics for DM are rarely performed among people with TB in endemic settings [3]. Patients with DM are three times more likely to develop active TB and poor treatment outcomes. Data are limited regarding glycemic dynamics from TB diagnosis through treatment.

Objectives

Methods

Results

Conclusion