Relationship between serum phosphate trajectories and mortality in patients with acute myocardial infarction: a retrospective cohort study based on the MIMIC-IV database

BMC Cardiovasc Disord. 2026 Jul 2. doi: 10.1186/s12872-026-05759-z. Online ahead of print.

ABSTRACT

BACKGROUND: Serum phosphate concentrations have been shown to significantly predict clinical outcomes among critically ill populations; however, their association with mortality following acute myocardial infarction (AMI) is not well established. This research seeks to examine how early trajectories of serum phosphate relate to prognosis in individuals diagnosed with AMI.

METHODS: This retrospective cohort study utilized data from the Medical Information Mart for Intensive Care IV (MIMIC-IV) database. A latent growth mixture modeling (LGMM) approach was utilized to examine changes in mean serum phosphate concentrations, which were assessed at 24-hour intervals during the initial 96-hour period following intensive care unit (ICU) admission among AMI patients, thereby classifying their longitudinal trajectory profiles. Using a landmark analysis design, we applied Cox proportional hazards regression models to evaluate the association between trajectory patterns and mortality after the 96-hour landmark. To evaluate the incremental prognostic value of phosphate trajectories beyond conventional clinical variables, Harrell's C-statistic, net reclassification improvement (NRI), and integrated discrimination improvement (IDI) were calculated.

RESULTS: This study included 1,522 patients with AMI. The LGMM model classified serum phosphate trajectories into three distinct patterns. Patients with different trajectories exhibited significant differences in prognosis. Compared with Trajectory 1 (normal stable), patients in Trajectory 3 (persistently elevated) had the poorest prognosis, with significantly increased 28-day mortality risk after the landmark (HR = 1.60, 95% CI: 1.21-2.11; P = 0.001) and 1-year mortality risk after the landmark (HR = 1.71, 95% CI: 1.33-2.20; P < 0.001). For Trajectory 2 (highly fluctuating), the association with 28-day mortality after the landmark was not statistically significant in the fully adjusted model, while the risk for 1-year mortality after the landmark remained significantly elevated (adjusted HR = 1.42, 95% CI: 1.04-1.94; P = 0.026). The addition of phosphate trajectories provided modest but significant improvement in risk discrimination(ΔC = 0.007, IDI and NRI both P < 0.05). Similar improvements were observed for 1-year mortality after the landmark.

CONCLUSION: Persistently elevated serum phosphate trajectories during the early ICU period were associated with increased short-term and long-term mortality in patients with AMI. These findings suggest that dynamic monitoring of serum phosphate may provide additional prognostic information beyond traditional clinical indicators. However, given the observational design and the requirement for survival during the initial 96-hour trajectory window, the findings should be interpreted as associations rather than causal relationships.

PMID:42393537 | DOI:10.1186/s12872-026-05759-z