Accumulation of metabolic multimorbidity and its association with mortality: results from a prospective cohort study among people with HIV in China, 2010-2024.

Background: As survival among people with HIV (PWH) improves in the antiretroviral therapy (ART) era, metabolic comorbidities have become important determinants of long-term outcomes. However, longitudinal evidence quantifying the mortality impact of time-varying metabolic multimorbidity (MM) remains limited. This study aimed to characterise temporal trends in major metabolic conditions and MM patterns after ART initiation, and to quantify their associations with all-cause and non-AIDS-related mortality. Methods: We conducted a prospective two-centre cohort study of PWH who initiated ART in southern China between 2010 and 2024. Dyslipidaemia, diabetes, hypertension, metabolic dysfunction-associated steatotic liver disease, and osteoporosis were ascertained from clinical records and laboratory measurements, and MM was defined as the presence of at least two of these conditions. All-cause and non-AIDS-related deaths were identified through linkage with HIV surveillance, hospital records, and death registries. Associations between MM and mortality were evaluated using Cox regression models with MM treated as a time-varying exposure, and were reported as adjusted hazard ratios (aHRs) with 95% confidence intervals (CIs). Findings: Among 31,630 PWH who initiated ART between 2010 and 2024 (median age 35 years; 83·2% men), 3143 (9·9%) had MM at baseline, defined as present before or within 90 days after ART initiation. The incidence of all five metabolic conditions increased over time, with particularly marked rises after 2020. During follow-up, 8495 MM events occurred, most commonly as two-condition combinations of dyslipidaemia with diabetes or hypertension. All-cause mortality rates were 7·2 (95% CI 6·2-8·3), 10·3 (95% CI 9·7-10·8), and 15·9 (95% CI 14·5-17·4) per 1000 person-years among individuals with 0, 1, and ≥2 conditions, respectively. Compared with those with no metabolic condition, the aHRs for all-cause mortality were 1·7 (95% CI 1·5-2·0) for one condition and 2·1 (1·8-2·5) for ≥2 conditions; a similar dose-response relationship was observed for non-AIDS-related mortality. PWH with hypertension-diabetes-dyslipidaemia doublet or triad had substantially higher risks of all-cause mortality (aHR 2·1 and 4·1, respectively) and non-AIDS-related mortality (aHR 2·3 and 4·2, respectively) than those without metabolic comorbidity. Interpretation: As metabolic comorbidities accumulated over time, the burden of MM increased among PWH. Both overall metabolic burden and specific high-risk combinations were strongly associated with increased mortality. These findings underscore the need to move beyond virological control alone and to integrate structured metabolic comorbidity assessment and long-term metabolic risk management into routine HIV care in the ART era. Funding: National Natural Science Foundation of China, the Guangxi Key Research and Development Program, the project of the Guangdong Basic and Applied Basic Research Foundation, and the Guangdong Provincial Medical Science and Technology Research Fund Project.