A combined eGFR-CD4 index and mortality risk among people with HIV: a retrospective cohort study in Taizhou city, Zhejiang province, China

期刊论文
出版模式：

开放获取
发表日期：

2026年6月7日

查看全文

Congcong Guo,
Xinxin Xing,
Qiguo Meng,
Shanling Wang,
Yali Xie,
Tailin Chen,
Jiyuan Ren,
Xiaoxiao Chen,
Haijiang Lin

摘要

Background

Mortality among people with HIV (PWH) increasingly reflects both HIV-related and non-HIV-related causes. Whether a simple index integrating routinely measured kidney and immune function can improve risk stratification remains uncertain.

Methods

We conducted a retrospective cohort study in Taizhou, Zhejiang, China using the China HIV/AIDS Comprehensive Response Information Management System linked to a cause-of-death surveillance system. Adults aged ≥ 18 years diagnosed with HIV between 1998 and 2024 who had at least one paired serum creatinine and CD4 measurement were followed through 31 December 2024. Time-updated estimated glomerular filtration rate (eGFR), CD4, and a combined eGFR-CD4 index (CBI) were analysed using time-dependent Cox models. Restricted cubic splines assessed non-linearity, and 12-month landmark analyses with time-dependent receiver operating characteristic curves evaluated discrimination. Bootstrap internal validation was performed.

Results

Among 3683 PWH (median follow-up 5.61 years), 434 deaths occurred, including 67 HIV-related and 367 non-HIV-related deaths. Each 1-SD increase in CBI was associated with higher risks of all-cause, HIV-related, and non-HIV-related mortality, with adjusted hazard ratios of 4.12 (95% CI: 2.76–6.15), 37.86 (95% CI: 9.03–158.80), and 3.30 (95% CI: 2.20–4.93), respectively (all P < 0.001). Landmark analyses showed graded risk separation across CBI quartiles. After bootstrap internal validation, the combined model showed modestly higher discrimination for all-cause and non-HIV-related mortality, whereas CD4 alone performed best for HIV-related mortality.

Conclusions

A time-varying combined eGFR-CD4 index was associated with mortality outcomes in this cohort of PWH in Taizhou, Zhejiang, China. Its discriminatory advantage after internal validation was modest and was evident for all-cause and non-HIV-related, but not HIV-related, mortality. External validation is warranted.