# Causal effects of serum testosterone on septic shock mortality: a Mendelian randomization study

## TL;DR

Higher testosterone levels are linked to better survival in septic shock patients, especially in men and those with lipid issues.

## Contribution

This study establishes a causal link between genetically predicted testosterone levels and reduced mortality in septic shock.

## Key findings

- Higher testosterone PGS was associated with a 28% lower risk of 28-day mortality.
- The protective effect was stronger in men and modified by ApoC3 levels.
- Mendelian randomization confirmed a causal relationship between testosterone and survival.

## Abstract

Sex hormones, particularly testosterone, modulate immune function during critical illness, and patients with septic shock frequently exhibit hypotestosteronemia. However, the causal relationship between testosterone and outcomes remains unclear owing to the confounding effects of illness-related changes in hormone levels during acute illness.

We investigated 469 patients with septic shock in multicenter ICUs using a testosterone polygenic score (PGS) derived from genome-wide association studies combined with two-sample Mendelian randomization to establish causal relationships independent of confounding factors. Cox proportional hazards regression was performed to assess the association with 28-day mortality. Additionally, we evaluated whether apolipoprotein C3 (ApoC3) levels modified the protective effects of testosterone using interaction models and the likelihood ratio test.

Higher genetically predicted testosterone levels were significantly associated with improved 28-day survival (adjusted hazard ratio [HR] 0.72 per 1-standard deviation increase in PGS; P = 0.024). This protective effect was more pronounced in men (HR, 0.66; P = 0.020) than in women (HR, 0.78; P = 0.37). Kaplan–Meier survival analysis revealed that the high testosterone PGS group had a 54.2% reduction in mortality hazard compared with the low testosterone PGS group (log-rank P = 0.007). Two-sample Mendelian randomization confirmed causality (inverse variance-weighted: β = − 2.79; P = 0.0042), with consistent results across complementary estimation methods. Notably, the testosterone-protective effect was significantly modified by ApoC3 levels (interaction, P = 0.041), with substantially stronger protective effects at higher ApoC3 concentrations. At high ApoC3 levels, the HR was 0.51 (95% confidence interval 0.31–0.85), suggesting that testosterone exerts a disproportionate benefit in the context of lipid dysmetabolism and inflammation.

Genetically determined higher testosterone levels are causally associated with improved survival in patients with septic shock, particularly in men and in those with lipid dysmetabolism. These findings identify testosterone as a potential therapeutic target and highlight lipid metabolism as a key modifier of the protective effects of testosterone against septic shock, warranting the investigation of testosterone-based interventions in future clinical trials.

The online version contains supplementary material available at 10.1186/s13054-026-05860-x.