Tuberculosis in advanced chronic kidney disease: An Observational Study at a Tertiary Care Center in Mexico

The management of Tuberculosis (TB) in patients with chronic kidney disease (CKD) presents unique challenges, including an immunosuppressive state, altered drug pharmacokinetics, and limited access to single-drug formulations in our setting. There is a scarcity of real-world evidence on TB outcomes in this population in Latin America. Our study aimed to compare mortality, cure, and relapse rates between TB patients with ACKD and without ACKD.

We conducted an observational-analytical study of all patients aged ≥18 years with microbiologically or histologically confirmed TB between 2013 and 2024. Patients with ACKD (GFR < 30 mL/min/1.73 m² were compared against age- and sex-matched non-ACKD (GFR ≥ 30 mL/min/1.73 m²) patients. Due to differential HIV distribution, we also performed a sensitivity analysis excluding HIV-positive patients. The primary outcome was all-cause mortality at 1 year. Outcomes were compared using the Chi-squared and Mann-Whitney U tests, as well as logistic regression.

A total of 51 patients with tuberculosis were included (17 with ACKD, 34 without ACKD). CKD was caused by lupus or diabetes in 29% of patients each. Most CKD patients (68%) received a local pragmatic alternating regimen. One-year all-cause mortality was 18% in both groups (p > 0.999), and TB-related mortality was 9% in the control group, vs 0% in the ACKD group. The cure rate was similar between groups (ACKD: 88% vs. non-ACKD: 82%; p = 0.586). No relapses occurred. In a sensitivity analysis excluding HIV-positive patients (n = 44), findings were consistent with the primary analysis, with no significant difference in mortality between groups. Due to the low event rate, we conducted a bivariate analysis in an exploratory fashion and did not perform a multivariate analysis.

In this Mexican small cohort, ACKD didn't significantly worsen TB outcomes compared with non-ACKD patients. An local pragmatic alternating regimen was used without apparent harm. These preliminary findings are limited by small sample size, limited statistical power, and lack of pharmacokinetic validation. Larger studies with drug monitoring are needed to optimize treatment for this vulnerable group.