Nonlinear association between cockcroft-gault creatinine clearance and all-cause mortality in critically ill cerebrovascular disease patients: a retrospective cohort study

BACKGROUND: The prognostic value of Cockcroft-Gault estimated creatinine clearance (CG-CrCl) in critically ill cerebrovascular disease (CVD) patients is unclear. METHODS: This retrospective cohort study included 8,732 adults from MIMIC-IV. CG-CrCl was examined continuously and by quartiles. Primary outcome was 30-day mortality; secondary outcomes were 90-day and 365-day mortality. Multivariable Cox models, restricted cubic splines, and two-piecewise linear regression assessed associations. RESULTS: Overall 30-day, 90-day, and 365-day mortality rates were 17.6%, 22.2%, and 28.8%, respectively. In fully adjusted models, each 1-standard deviation (SD; 45.46 mL/min) decrease in CG-CrCl was independently associated with increased mortality risk (30-day: HR 1.35, 95% CI 1.25-1.45; 90-day: HR 1.32, 95% CI 1.24-1.41; 365-day: HR 1.31, 95% CI 1.24-1.39). A significant nonlinear threshold effect was identified: the association became pronounced only below inflection points of 119.66, 114.84, and 114.89 mL/min for 30-, 90-, and 365-day mortality, respectively. Adding CG-CrCl did not significantly improve discrimination at any time point (all DeLong P > 0.05). CONCLUSION: CG-CrCl is independently associated with short- and long-term mortality in critically ill CVD patients with a nonlinear threshold effect, suggesting it may serve as a readily available risk marker worthy of further validation.