Elevated creatinine-to-albumin ratio (CAR) was independently and consistently associated with an increased risk of short-term mortality in critically ill patients with AF, and a nomogram incorporating CAR demonstrated strong predictive performance.
The optimal CAR cutoff value for predicting 28-day mortality was 0.5, dividing patients into high (>0.5) and low (≤0.5) CAR groups.
Kaplan-Meier analysis demonstrated significantly lower 28-day survival in patients with elevated CAR (>0.5) compared to those with low CAR (≤0.5).
CAR was independently associated with 28-day mortality as a continuous variable in multivariate Cox regression analysis.
CAR as a categorical variable (>0.5 vs ≤0.5) was independently associated with 28-day mortality in multivariate Cox regression analysis.
Restricted cubic spline analysis revealed a significant nonlinear dose-response relationship between CAR and 28-day mortality risk.
CAR showed good discriminative ability for 28-day mortality in critically ill patients with AF.
Subgroup analyses showed consistent associations between elevated CAR and 28-day mortality across most prespecified subgroups.
A nomogram incorporating CAR demonstrated good discrimination, calibration, and clinical utility for predicting 28-day mortality.
This research suggests that a simple blood test ratio — creatinine divided by albumin (called CAR) — can help predict which critically ill patients with atrial fibrillation (an irregular heart rhythm) are most likely to die within 28 days of ICU admission. Using data from over 4,500 ICU patients recorded in a large U.S. hospital database (MIMIC-IV), researchers found that patients with a CAR above 0.5 had significantly worse survival odds than those with a CAR of 0.5 or below. Patients with higher CAR values had approximately 38% greater risk of dying within 28 days compared to those with lower values, even after accounting for other health factors. The study also found that the relationship between CAR and death risk is not strictly linear — meaning the risk does not simply increase at a steady rate as CAR rises, but follows a more complex curve. The researchers used these findings to build a scoring tool (called a nomogram) that combines CAR with other clinical information to generate individualized risk predictions for patients. This tool showed strong performance in terms of both accuracy and practical usefulness. This research suggests that CAR — which can be calculated from routine blood tests already performed in the ICU — may serve as a practical and readily available marker to help clinicians identify high-risk patients early and tailor their care accordingly. Because both creatinine (a kidney function marker) and albumin (a nutrition and inflammation marker) reflect different aspects of a patient's physiological stress, their ratio captures a combination of organ dysfunction and nutritional status that appears meaningful for predicting short-term outcomes in this vulnerable population.
Li W, Guo L, Huang H. (2026). Usefulness of serum creatinine-to-albumin ratio for 28-day all-cause mortality in critically ill patients with atrial fibrillation: A retrospective study based on the MIMIC-IV database.. Medicine. https://doi.org/10.1097/MD.0000000000048651