Lara D Veeken, Iceu D Kulsum, Bony W Lestari, Prayudi Santoso, Nanny N M Soetedjo, Raspati C Koesoemadinata, Adriana V Miranda, Wini Sukmawati, Argita D Salindri, Arto Y Soeroto, Reinout van Crevel
{"title":"High Rates of Mortality During Drug-Resistant Tuberculosis Treatment Among Individuals With Diabetes Mellitus and Low Body Mass Index.","authors":"Lara D Veeken, Iceu D Kulsum, Bony W Lestari, Prayudi Santoso, Nanny N M Soetedjo, Raspati C Koesoemadinata, Adriana V Miranda, Wini Sukmawati, Argita D Salindri, Arto Y Soeroto, Reinout van Crevel","doi":"10.1093/ofid/ofaf344","DOIUrl":null,"url":null,"abstract":"<p><strong>Background: </strong>Diabetes is a risk factor for mortality during rifampicin-resistant tuberculosis (RR-TB) treatment, but whether its impact differs by nutritional status is unknown. We estimated the effect of diabetes and its interaction with low body mass index (BMI) (ie, <18.5 kg/m<sup>2</sup>) on all-cause mortality during treatment of RR-TB.</p><p><strong>Methods: </strong>We used medical record data of adults treated for RR-TB in Indonesia between March 2020 and May 2022. Diabetes was defined as glycated hemoglobin ≥6.5% or prior diabetes diagnosis by healthcare providers. Cox proportional hazards regression was used to estimate the hazard rates of mortality during treatment comparing those with and without diabetes. Multiplicative and additive interactions were evaluated to determine if the effect of diabetes on mortality during treatment was moderated by BMI status.</p><p><strong>Results: </strong>Among 345 individuals (57% male, 1.7% with human immunodeficiency virus, 59% with BMI <18.5 kg/m<sup>2</sup>), 96 (28%) had diabetes and 62 (18%) died. Adjusting for confounders, the hazard rates of mortality during treatment were higher among those with diabetes (adjusted hazard rate ratio [aHR], 2.05 [95% CI, 1.17-3.58]) or those with BMI <18.5 kg/m<sup>2</sup> (aHR, 2.33 [95% CI, 1.28-4.21]). No significant multiplicative nor additive interaction was detected, but the hazard rates of mortality were highest among those with diabetes and BMI <18.5 kg/m<sup>2</sup> (aHR, 7.14 [95% CI, 2.71-18.82]) compared to those without diabetes and BMI ≥18.5 kg/m<sup>2</sup>.</p><p><strong>Conclusions: </strong>Having diabetes doubled the risk of mortality during RR-TB treatment. Highest mortality rates were observed among individuals with combined diabetes and low BMI.</p>","PeriodicalId":19517,"journal":{"name":"Open Forum Infectious Diseases","volume":"12 7","pages":"ofaf344"},"PeriodicalIF":3.8000,"publicationDate":"2025-06-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12207741/pdf/","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Open Forum Infectious Diseases","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1093/ofid/ofaf344","RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"2025/7/1 0:00:00","PubModel":"eCollection","JCR":"Q2","JCRName":"IMMUNOLOGY","Score":null,"Total":0}
引用次数: 0
Abstract
Background: Diabetes is a risk factor for mortality during rifampicin-resistant tuberculosis (RR-TB) treatment, but whether its impact differs by nutritional status is unknown. We estimated the effect of diabetes and its interaction with low body mass index (BMI) (ie, <18.5 kg/m2) on all-cause mortality during treatment of RR-TB.
Methods: We used medical record data of adults treated for RR-TB in Indonesia between March 2020 and May 2022. Diabetes was defined as glycated hemoglobin ≥6.5% or prior diabetes diagnosis by healthcare providers. Cox proportional hazards regression was used to estimate the hazard rates of mortality during treatment comparing those with and without diabetes. Multiplicative and additive interactions were evaluated to determine if the effect of diabetes on mortality during treatment was moderated by BMI status.
Results: Among 345 individuals (57% male, 1.7% with human immunodeficiency virus, 59% with BMI <18.5 kg/m2), 96 (28%) had diabetes and 62 (18%) died. Adjusting for confounders, the hazard rates of mortality during treatment were higher among those with diabetes (adjusted hazard rate ratio [aHR], 2.05 [95% CI, 1.17-3.58]) or those with BMI <18.5 kg/m2 (aHR, 2.33 [95% CI, 1.28-4.21]). No significant multiplicative nor additive interaction was detected, but the hazard rates of mortality were highest among those with diabetes and BMI <18.5 kg/m2 (aHR, 7.14 [95% CI, 2.71-18.82]) compared to those without diabetes and BMI ≥18.5 kg/m2.
Conclusions: Having diabetes doubled the risk of mortality during RR-TB treatment. Highest mortality rates were observed among individuals with combined diabetes and low BMI.
期刊介绍:
Open Forum Infectious Diseases provides a global forum for the publication of clinical, translational, and basic research findings in a fully open access, online journal environment. The journal reflects the broad diversity of the field of infectious diseases, and focuses on the intersection of biomedical science and clinical practice, with a particular emphasis on knowledge that holds the potential to improve patient care in populations around the world. Fully peer-reviewed, OFID supports the international community of infectious diseases experts by providing a venue for articles that further the understanding of all aspects of infectious diseases.