Related Experiment Videos
Dynamic serum chloride trajectories predict short- and long-term mortality in patients with malignancies: a
1Department of Oncology, Zibo Municipal Hospital, Zibo, Shandong, China.
Background:
Hypochloremia is associated with poor prognosis in various malignant tumors. However, longitudinal trajectories of serum chloride in cancer patients and their relationship with mortality remain unexplored. This study aimed to delineate dynamic serum chloride trajectory patterns and analyze their association with mortality in patients with malignancies.
Methods:
This retrospective cohort study included 4,115 patients with malignant tumors from the MIMIC-IV database who had serial serum chloride measurements within the first 5 days of ICU admission, plus an additional 518 patients from Zibo Municipal Hospital for external validation. We applied group-based trajectory modeling (GBTM) to stratify patients into three distinct serum chloride trajectory classes. The primary outcomes were 28-day and 365-day all-cause mortality, which were assessed using Kaplan-Meier survival analysis and multivariable Cox proportional hazards regression. Subgroup and sensitivity analyses were further conducted to verify the robustness of the study findings.
Results:
Three distinct serum chloride trajectory patterns were identified in both the MIMIC-IV primary cohort and the external validation cohort: Class 1 (initial hyperchloremia resolving to normal levels), Class 2 (persistent normochloremia), and Class 3 (progressive hyperchloremia). In the fully adjusted Cox regression model (Model 3), Class 3 was an independent predictor of 28-day (HR = 1.55, 95% CI: 1.26-1.92) and 365-day all-cause mortality (HR = 1.38, 95%CI: 1.19-1.60) in the MIMIC-IV cohort. This prognostic value was validated in the external validation cohort, where Class 3 was also associated with a significantly increased 28-day mortality risk (HR = 1.90, 95%CI: 1.08-3.33). Kaplan-Meier survival analysis revealed that Class 3 patients had significantly lower survival rates at both 28-day and 365-day follow-ups. Subgroup analyses and sensitivity analyses further verified the robustness of the association between progressive hyperchloremia and adverse mortality outcomes.
Conclusions:
Progressive hyperchloremia trajectory independently predicts early and late mortality in cancer patients, and may offer a dynamic, low-cost biomarker for risk stratification.