Controlled Attenuation Parameter and Gallstone Disease Risk: A Retrospective Cohort Study with Mediation Analysis and Longitudinal Trajectory Modeling, Emphasizing Gender-Specific Effects

Background: The association between the controlled attenuation parameter (CAP)-a noninvasive marker for hepatic steatosis-and gallstone disease (GSD) remains poorly understood. This study aimed to investigate the impact of CAP levels and their longitudinal trajectories on GSD risk and explore potential mediating pathways. Methods: A dynamic cohort of 6,308 adults without baseline GSD underwent serial health examinations in Ya’an, China, from January 2020 to April 2025. The statistical methods included multivariate Cox regression analysis, restricted cubic spline (RCS) analysis, and mediation analysis. For sensitivity analyses, we performed gender-stratified analysis, trial sequential analysis (TSA), cumulative risk curve plotting, and re-analysis after excluding individuals with baseline diseases. Among 412 participants with ≥ 3 CAP measurements, group-based trajectory modelling (GBTM) identified CAP patterns and related them to GSD incidence. Results: Each 1-SD increase in CAP was associated with a 40% higher GSD risk, with a stronger association in women (HR = 1.96, P = 0.02) than in men (HR = 1.36, P = 0.04). RCS revealed a strictly linear dose-response relationship between CAP and GSD risk ( P non-linear = 0.43). Compared with the reference group (CAP ≤ 238 dB/m), moderate (259– 292 dB/m) and severe (> 292 dB/m) steatosis were associated with significantly higher GSD risk (HR = 2.56, P < 0.01; HR = 2.47, P = 0.01, respectively), whereas mild steatosis (238– 259 dB/m) showed no significant association ( P = 0.11). GBTM identified “persistently low” (61.41%) and “persistently high” (38.59%) CAP trajectories, with the latter group having a 3.85-fold higher GSD risk ( P = 0.04). Mediation analysis identified a significant pathway from CAP to GSD via high-density lipoprotein cholesterol (HDL-C, P < 0.01, proportion of mediation = 27.49%). Conclusion: CAP is positively and linearly associated with GSD risk, particularly in women. Persistently high CAP levels confer a significantly elevated GSD risk. HDL-C is a significant mediator. Keywords: longitudinal association analysis, cox regression model, group-based trajectory model