---
title: "Temporal patterns of tuberculosis-associated hyperglycaemia and effects on treatment outcomes in n"
id: "plos-one-0-temporal-patterns-of-tuberculosis-associated-hyperglycaemia-and-its-effect-on"
canonical_url: "https://medichelpline.com/clinical-feed/plos-one-0-temporal-patterns-of-tuberculosis-associated-hyperglycaemia-and-its-effect-on"
content_type: "clinical_feed_article"
specialty: "Infectious Disease"
source_name: "PLOS ONE (Medicine)"
source_url: "https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0358593"
published_at: "2026-09-15T14:00:00.000Z"
evidence_level: "Journal Feed"
license: "CC-BY-NC-4.0 / Informational Use"
---
# Temporal patterns of tuberculosis-associated hyperglycaemia and effects on treatment outcomes in n
## Provenance & Clinical Metadata
- **Canonical URL:** https://medichelpline.com/clinical-feed/plos-one-0-temporal-patterns-of-tuberculosis-associated-hyperglycaemia-and-its-effect-on
- **Specialty:** [Infectious Disease](https://medichelpline.com/clinical-feed/infectious-disease.md)
- **Primary Source:** PLOS ONE (Medicine)
- **Source URL:** [Original Journal Publication](https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0358593)
- **Published At:** 2026-09-15T14:00:00.000Z
- **Evidence Rating:** Journal Feed
## Executive GIST (TL;DR)
- This prospective five-month cohort study in Nairobi, Kenya, followed 87 adults with bacteriologically confirmed pulmonary **tuberculosis** (PTB) who were not previously diabetic to describe the prevalence, temporal patterns, and clinical impact of TB-associated **hyperglycaemia** using serial **HbA1c** measurements. - Participants were tested at baseline (prior to TB treatment), month two, and month five using HbA1c and classified per ADA thresholds: normoglycaemia (4.0–5.6%), prediabetes (5.7–6.4%), and **diabetes mellitus** (≥6.5%). - At enrolment the cohort was mostly male (76%), median age 37 years (IQR 29–44), and predominantly HIV-negative (84%). - Baseline prevalence of hyperglycaemia was high: 75% (95% CI 65–83), comprising 45% prediabetes and 30% diabetes by HbA1c criteria. - Longitudinal measurements showed significant variation in **HbA1c** across time points (p < 0.001). By month two and month five, prediabetes prevalence rose from 24% to 30%, while diabetes prevalence declined from 21% to 9%. - Glycaemic trajectories were categorized as persistent normoglycaemia, persistent hyperglycaemia, transient hyperglycaemia, and incident hyperglycaemia; overall patterns indicated both persistence and variability during anti-TB therapy. - In GEE models adjusted for age, BMI, HIV status, alcohol use, and smoking, hyperglycaemia was associated with classic diabetic symptoms (frequent micturition, increased thirst, unintentional weight loss, blurred vision) and independently associated with increased risk of **sputum positivity** (adjusted RR 4.22, 95% CI 2.09–8.52). - Glycaemic variability (fluctuating between normoglycaemic and hyperglycaemic ranges) was significantly associated with unfavourable TB treatment outcomes (p < 0.001). - The authors conclude that new-onset hyperglycaemia is common in newly diagnosed TB and frequently persists through five months of treatment, supporting integration of glycaemic monitoring and management into routine TB follow-up. - Data are not publicly available due to privacy and ethical restrictions; access can be requested via the University of Nairobi–Kenyatta National Hospital Ethics and Research Committee. Funding was provided by NIH/Fogarty International Center; no competing interests declared.
## Clinical Analysis & Structured Key Points
Temporal patterns of tuberculosis-associated hyperglycaemia and its effect on treatment outcomes in individuals not previously diabetic: A prospective cohort study | PLOS One Browse Subject Areas ? Click through the PLOS taxonomy to find articles in your field. For more information about PLOS Subject Areas, click here . Article Authors Metrics Comments Media Coverage Reader Comments Figures Figures Abstract Background Active tuberculosis (TB) infection can induce hyperglycaemia and insulin resistance, increasing the risk of type 2 diabetes mellitus and worsening TB disease severity. However, data on the dynamics of TB-associated hyperglycaemia and its effect on TB treatment outcomes among individuals not previously diabetic remain limited, particularly in high TB-burden settings. In this study, we aimed to investigate the prevalence and temporal patterns of TB-associated hyperglycaemia and its impact on disease presentation and TB treatment outcomes among a cohort of individuals not previously diabetic. Methods This was a five-month longitudinal prospective cohort study conducted at two urban specialized TB healthcare facilities in Kenya among adults aged >18 years with bacteriologically confirmed pulmonary TB (PTB). Known diabetes mellitus (DM) or prediabetic individuals were excluded during enrolment. Glycaemic status was assessed at baseline and months two and five using glycated haemoglobin (HbA1c) according to the American Diabetes Association guideline. Hyperglycaemia was defined as HbA1c ≥5.7% and categorized as normoglycaemia, persistent, transient, or incident hyperglycaemia based on longitudinal measurements. Socio-demographic and clinical data were collected using a structured questionnaire and analysed using SPSS version 26.0 and GraphPad Prism version 9.0. Longitudinal trends in HbA1c levels and the association of hyperglycaemia with DM-related symptoms, as well as TB treatment outcomes, were assessed using the Friedman test and the Generalized Estimating Equations (GEE), respectively. Results We enrolled and followed 87 individuals newly diagnosed with PTB. Of these, 66 were tested for hyperglycaemia in month two after 7 died and 14 were lost to follow-up/transferred out (LTFU/TO), and 70 in month 5, after an additional death, 7 re-engagements, and 2 LTFU/TO were reported. At enrolment, the cohort was predominantly male (76%), not married (53%), and HIV-negative (84%), with a median age of 37 years (IQR 29–44). Prevalence of hyperglycaemia at baseline was 75% (95% Confidence Interval [CI] 65–83), with 45% of the individuals presenting as prediabetic and 30% as diabetic. At the month-two and month-five time points, the proportion of participants with prediabetes (pre-DM) increased from 24% (16/66) to 30% (21/70), while the prevalence of DM declined from 21% (14/66) to 9% (6/70). Across the time points, HbA1c levels varied significantly (p < 0.001). In GEE analysis, hyperglycaemic participants were more likely to experience classic symptoms of DM such as frequent micturition (adjusted Relative Risk [aRR] 1.76, 95% CI 1.12–2.77), increased thirst (aRR 2.37, 95% CI 1.38–4.05), unintentional weight loss (aRR 1.84, 95% CI 1.24–2.75), and blurred vision (aRR 1.95, 95% CI 1.20–3.18). After adjusting for age, BMI, HIV, alcohol use, and smoking, hyperglycaemia was independently associated with an increased risk of sputum positivity (aRR 4.22, 95% CI 2.09–8.52). Glycaemic variability was significantly associated with unfavourable TB treatment outcomes (p < 0.001). Conclusions New-onset hyperglycaemia was highly prevalent in adults newly diagnosed with TB and persisted through the fifth month of TB treatment. Hyperglycaemia was associated with classic diabetic symptoms and persisting sputum positivity, highlighting the need to integrate glycaemic monitoring and management into routine TB follow-up care. Citation: Musyoki VM, Mureithi M, Heikinheimo A, Maleche-Obimbo E, Maina D, Musau S, et al. (2026) Temporal patterns of tuberculosis-associated hyperglycaemia and its effect on treatment outcomes in individuals not previously diabetic: A prospective cohort study. PLoS One 21(9): e0358593. https://doi.org/10.1371/journal.pone.0358593 Editor: Sheryar Afzal, King Faisal University College of Veterinary Medicine and Animal Resources, SAUDI ARABIA Received: February 6, 2026; Accepted: September 2, 2026; Published: September 15, 2026 Copyright: © 2026 Musyoki et al. This is an open access article distributed under the terms of the Creative Commons Attribution License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Data Availability: The datasets generated and analysed during the current study are not publicly available because they contain sensitive clinical information and are subject to ethical and privacy restrictions imposed by the University of Nairobi–Kenyatta National Hospital Ethics and Research Committee (UoN-KNH ERC). Public deposition of the data could compromise participant confidentiality and increase the risk of re-identification, which would be inconsistent with the conditions under which ethical approval and informed consent were obtained. Data are available to qualified researchers upon reasonable request and subject to review and approval by the UoN-KNH ERC. Requests for access should be directed to the UoN-KNH ERC), a non-author institutional body responsible for overseeing data access, at uonknh_erc@uonbi.ac.ke . Additional information on the committee is available through the University of Nairobi website ( https://erc.uonbi.ac.ke/) . The study data are securely stored by the University of Nairobi in accordance with institutional policies and ethical requirements and will be maintained to ensure long-term preservation and availability for researchers who meet the criteria for access. All relevant summary data supporting the findings of this study are provided within the manuscript. Funding: This study was funded by the National Institute of Health/Fogarty International Centre (Grant number 5D43TW011817) through the Tuberculosis and HIV Co-infection Training Program. The funders had no role in study design, data collection and analysis, the decision to publish or the preparation of the manuscript. Competing interests: The authors have declared that no competing interests exist. Introduction Tuberculosis (TB), a highly infectious disease caused by Mycobacterium tuberculosis (MTB) bacteria, remains the world’s leading cause of death from a single infectious agent [ 1 , 2 ]. According to the World Health Organization (WHO) Global TB Report 2025, an estimated 10.7 million people developed TB worldwide in 2024, of whom approximately 1.23 million died [ 2 ]. Countries such as China, the Philippines, India, Pakistan, Indonesia, Bangladesh, Nigeria, and other sub-Saharan African countries continue to report the highest number of TB cases [ 2 , 3 ]. Despite high TB cases reported in most of the developing countries, treatment using conventional antibiotics has effectively increased the cure rates, although recurrent TB and mortality cases during and after TB treatment have been reported [ 2 , 4 , 5 ]. Recent literature shows that these cases are partly attributed to the co-existence of TB with other conditions and, lately, with hyperglycaemia as a pathological characteristic of TB disease in non-diabetic individuals [ 6 ]. The co-existence of TB with chronic conditions such as diabetes mellitus (DM) and the non-diabetic hyperglycaemic state is attracting global attention due to the probable association with increasing cases of TB morbidity, poor TB treatment outcomes, and T2DM [ 1 , 5 , 6 ]. Pathophysiologically, infection with MTB disrupts hormonal and cytokine homeostasis, increasing the production of pro-inflammatory cytokines such as tumor necrosis factor-alpha (TNF-α), interleukin-1 (IL-1), and interleukin-6 (IL-6) that impair pancreatic β-cell function and insulin signalling, altering insulin production and activity [ 7 , 8 ]. This initiates a glucose intolerance state, stimulating glucose production by the liver; hence, an increase in blood glucose levels in these patients [ 7 – 11 ]. Similarly, available literature suggests that anti-TB drugs such as rifampicin and isoniazid have a hyperglycaemia-inducing effect, especially during the intensive phase of TB treatment [ 12 , 13 ]. Hyperglycaemia weakens the immune system, creating an imbalance between pro-inflammatory and anti-inflammatory cytokines [ 10 , 14 ]. This condition halts the effective response and control of MTB before and during TB treatment [ 14 ]. Studies on TB treatment outcomes have shown that hyperglycaemic patients may experience unfavourable outcomes [ 1 , 15 , 16 ]. Hyperglycaemia reduces the absorption and metabolism of drugs; hence, the prediction that patients presenting with TB and hyperglycaemia have low concentrations of anti-TB drugs in the blood and tissues and are at high risk of transmitting TB [ 1 , 13 , 15 , 16 ]. Information on hyperglycaemia during TB disease and treatment, especially among individuals not previously diabetic, remains limited worldwide, including in low- and middle-income countries where TB is endemic and prevalent [ 1 , 7 ]. Studies from Peru, India, China, South Africa, and Nigeria have reported over 25% prevalence of hyperglycaemia among patients with pulmonary TB [ 1 , 7 , 12 , 17 ]. However, the majority of these and similar studies have largely been of cross-sectional design, providing little insight into the dynamics of hyperglycaemia during TB treatment. In the present study, we employed a prospective cohort design to assess the glycaemic changes of TB-associated hyperglycaemia in newly diagnosed TB patients previously not diabetic and its impact on TB treatment outcome. The study aimed to generate new evidence to inform future clinical management strategies, address a key gap in the current literature, and improve understanding of the effects of metabolic response to TB infection and treatment. Methods Study design and population This was a five-month longitudinal prospective cohort study conducted between February 6, 2024, and November 20, 2024. We recruited newly diagnosed individuals with active PTB who presented for care at Mbagathi County Hospital and Rhodes Chest Clinic, TB-specialized public healthcare facilities in Nairobi, Kenya. We included individuals who were above the age of 18 years with clinical symptoms of PTB, mainly persistent cough lasting more than 2 weeks, fever, chest pain, night sweats, and fatigue, and who were medically confirmed using sputum specimens tested with the Xpert ® MTB/RIF Ultra (Cepheid, Sunnyvale, CA, USA) molecular assay according to the manufacturer’s instructions. We excluded individuals who were pregnant, lactating, known pre-DM or DM, and those presenting with multidrug-resistant (MDR-TB) or rifampicin-resistant TB (RR-TB). We also excluded participants who had other chronic diseases, such as cancer or cardiovascular diseases. All the participants were followed for 5 months and interviewed at baseline (prior to TB treatment initiation) and two and five months after starting TB treatment. At each visit, clinical examinations were performed to assess vital signs, the presence of TB-related symptoms, and treatment complications. Data, specimen collection, and follow-up TB diagnosis After consenting and enrolment of the study participants, data were collected using a structured questionnaire at baseline and during the 2 nd and 5 th month follow-up visits. At baseline, socio-demographic information including age, sex, marital status, education level, smoking, and alcohol use was collected. Weight and height were measured and body mass index (BMI) estimated. Based on American Diabetes Association (ADA) guidelines [ 18 ], symptoms including increased thirst, dry mouth, frequent urination, fatigue, blurred vision, and unintentional weight loss were ascertained through participant self-report and corroborated by clinical assessment where available, with each symptom recorded as a binary outcome. Additionally, we assessed HIV status through a combination of patient self-report and confirmatory diagnostic testing, in accordance with national guidelines. All participants received the same standard TB treatment regimen as per the TB guideline [ 3 ]. Approximately 4 mL of venous blood was drawn from enrolled participants in a lavender-top vacutainer, and whole blood was used for baseline hyperglycaemia testing. At follow-up, two and five months after TB treatment initiation, participants were re-examined for DM symptoms, and blood was drawn for hyperglycaemia testing. Sputum specimens were collected and tested for MTB bacilli as per the TB screening and diagnostic algorithm [ 3 ]. TB treatment outcomes were assessed using month-two and month-five sputum MTB test results and were reported as either favourable (smear negative) or unfavourable (smear positive) based on bacteriological treatment response and WHO guidelines [ 5 ]. During TB outcome evaluation, we excluded patients who were lost to follow-up. Hyperglycaemia screening criteria Enrolled participants were tested for hyperglycaemia using glycated haemoglobin (HbA1c). Approximately 4 mL of venous blood was collected in a lavender-top vacutainer before TB treatment was initiated and HbA1c levels determined by fluorescence immunoassay technology – sandwich immunodetection method (Finecare™, Guangzhou, P.R. China). At baseline, participants were divided into two: normoglycaemic and hyperglycaemic groups. Those with HbA1c levels ≥5.7% were classified under the hyperglycaemic group, while those presenting with HbA1c levels between 4.0% and 5.6% under the normoglycaemic group. The hyperglycaemic were further classified as having DM (HbA1c ≥ 6.5%) or prediabetes (pre-DM) (HbA1c = 5.7–6.4%) as per the American Diabetes Association (ADA) guideline [ 18 ]. Glycaemic status was assessed before initiation of TB treatment, two months after the intensive phase, and 3 months after the start of the continuation treatment phase. The HbA1c trajectory during the initial 2 months of intensive TB treatment and 3 months of continuation treatment phase was defined as follows: Persistent normoglycaemia if HbA1c was between 4.0% and 5.6% at all time points; persistent hyperglycaemia if HbA1c was ≥5.7% at all time points; transient hyperglycaemia if HbA1c was ≥5.7% at only one or two of the time points; and incident hyperglycaemia if the participants were normoglycaemic at baseline and then developed hyperglycaemia at either the 2 nd or 3 rd time point. After the five-month follow-up period, participants were further categorized into two groups: glycaemic stability if, at the three time points, the participant had HbA1c levels between 4.0 and 5.6%; and glycaemic variability if they had HbA1c levels fluctuating between the normoglycaemic and hyperglycaemic ranges across the three time points. Statistical analysis Data analysis was done using SPSS version 26.0 (IBM Corp., NY, USA) and GraphPad Prism version 9.0 (GraphPad Software, San Diego, CA). The Shapiro-Wilk test was used to test the normality of the sample data set. Continuous variables, such as age and HbA1c, were summarized using median and interquartile ranges (IQRs), while categorical variables, such as sex, education level, BMI, smoking, and alcohol use, were presented as frequencies and percentages. To test for differences between groups, the Mann-Whitney U test was used for numerical data, while χ2 or Fisher’s exact test, as appropriate, was used for categorical data. Overall differences in participants’ HbA1c levels across the three study time points were assessed using the Friedman test. This analysis was restricted to participants with complete paired HbA1c measurements at baseline, month 2, and month 5 (complete-case subset), as required for the Friedman test. In observations with significant differences, post hoc pairwise comparison was conducted using the Wilcoxon signed-rank test with Bonferroni correction for multiple testing. The prevalence of normoglycaemia and hyperglycaemia (pre-DM and DM) was estimated with corresponding proportions and 95% confidence intervals. The number needed to screen (NNS), to evaluate the efficiency of screening for hyperglycaemia, prediabetes, and diabetes mellitus among newly diagnosed TB individuals, was calculated as the reciprocal of the point prevalence (proportion) for each parameter. To robustly model longitudinal associations, a Generalized Estimating Equation (GEE) model with a binomial distribution and log link function to account for within-subject correlation over time was performed. The GEE approach assumed an exchangeable working structure to model the correlation between repeated measures within individuals. Robust (sandwich) standard errors were computed to obtain valid inferences even when the working correlation structure was misspecified. Estimates were reported for the association between hyperglycaemia and clinical symptoms at baseline, two months, and five months, as well as sputum smear conversion and cure results at the second and fifth months of TB treatment. In multivariable GEE analysis, a model was fitted adjusting for a priori covariates selected based on literature and their known associations with glucose metabolism, symptom reporting, and clinical relevance. In symptom overlap, clinical symptoms, including thirst and frequent urination, which are related, were modelled as distinct binary outcomes in separate model analysis, and neither outcome was included as a covariate. However, to assess the robustness of the multivariable GEE model findings given the conceptual symptom overlap, a sensitivity analysis was conducted. Analysis was repeated after excluding participants who reported both symptoms. To explore and complement the primary regression-based analysis (GEE) and further describe patterns of HbA1c at months 2 and 5, we stratified participants into glycaemic stability and glycaemic variability groups. Within these groups, differences in HbA1c distributions between participants with favourable and unfavourable TB treatment outcomes were assessed using Mann-Whitney U test. Sputum smear conversion was defined as a change from positive to negative status, and the percent conversion calculated as the number of smear-negative cases divided by the total in each group. Statistical significance was set at a p-value of <0.05, and 95% confidence intervals (CI) were reported. Missing data during the five-month study period was examined for patterns to determine the plausibility of the missing completely at random (MCAR) assumption required for the Generalized Estimating Equation (GEE) analysis. Baseline socio-demographic and clinical characteristics (age, HbA1c, BMI, sex, and marital status) were compared between participants with complete follow-up data and those with incomplete observations. Continuous variables were compared using the Mann-Whitney U test, while categorical variables were assessed using the χ2 or Fisher’s exact test, as appropriate. No significant differences in baseline characteristics were observed between the groups, supporting the plausibility of the MCAR assumption. Therefore, the standard GEE model was fitted using all available observations under an available-case approach. Ethical statement The study was reviewed and approved by the University of Nairobi – Kenyatta National Hospital Ethics and Research Committee (P553/06/2023). Permission to conduct the study in Rhodes Chest Clinic and Mbagathi County Hospital was granted by the Nairobi City County, and licensed by the National Commission for Science, Technology & Innovation (NACOSTI/P/23/31666). Participation was voluntary, and written informed consent was obtained from all participants prior to enrolment. Confidentiality was mainta
## Related Clinical Research

- [Walking pace vs. step count: Both linked to lower mortality risk](https://medichelpline.com/clinical-feed/medical-news-today-0-walk-faster-or-walk-more-study-shows-both-pace-and-step-count-lower-death-risk.md)
- [Replacing animal fats with plant fats linked to lower mortality after nonmetastatic prostate cancer](https://medichelpline.com/clinical-feed/medical-news-today-0-swapping-animal-fats-with-plant-fats-may-lower-risk-of-death-after-prostate.md)
- [Global access to diabetes medicines: TARGET-DM framework for targeted policy actions](https://medichelpline.com/clinical-feed/nature-0-global-access-to-different-classes-of-diabetes-medicines-a-framework-for.md)
- [Gestational diabetes prevalence and physiology in South African women with and without HIV — ORCHI](https://medichelpline.com/clinical-feed/pubmed-41841696.md) (DOI: 10.1210/clinem/dgag111)
- [Efgartigimod for anti-GAD65 antibody–positive epilepsy and cerebellar ataxia: case report overview](https://medichelpline.com/clinical-feed/frontiers-in-immunology-15-treatment-of-anti-gad65-antibody-positive-epilepsy-and-cerebellar-ataxia-with.md)

## Navigation
- [← Back to Infectious Disease Feed](https://medichelpline.com/clinical-feed/infectious-disease.md)
- [← All Clinical Specialties](https://medichelpline.com/clinical-feed.md)
## Medical & Regulatory Disclaimer

> [!CAUTION]
> MedicHelpline content is structured for research, educational, and professional discovery purposes. It does not constitute individual medical advice, clinical diagnosis, or treatment recommendations.
> Always verify dosing, contraindications, and regulatory alerts against official product labeling and primary regulatory sources before clinical decision-making.