Red cell distribution width to albumin ratio (RAR) as a predictor of clinical outcomes and in-hospital mortality in acute gastrointestinal bleeding: a retrospective analysis.

BACKGROUND

Nutritional status and systemic inflammation drive outcomes in acute gastrointestinal bleeding. While albumin and red cell distribution width track these states individually, their combination, the red cell distribution width to albumin ratio (RAR), may offer superior prognostic value.

METHODS

Using the MIMIC-IV 3.1 database, 1508 adults admitted to the intensive care unit with gastrointestinal bleeding were analyzed. RAR was calculated and divided into quartiles. We assessed hospital, 28-day, and 1-year mortality using stepwise logistic regression, Cox proportional hazards models, and receiver operating characteristic analyses, adjusting for severity via Sequential Organ Failure Assessment and AIMS65 scores.

RESULTS

The highest RAR quartile was independently associated with hospital mortality (adjusted odds ratio = 1.54 vs Q1, P = 0.02), and continuous RAR showed a progressive risk increase (odds ratio = 1.18 per unit, P < 0.001). Results remained consistent in sensitivity analysis excluding patients with cirrhosis and variceal bleeding. Cox models with continuous RAR predicted both 28-day (hazard ratio = 1.17, P < 0.001) and 1-year mortality (hazard ratio = 1.18, P < 0.001). RAR alone exhibited moderate discrimination (area under the curve [AUC] = 0.62) but provided incremental value when combined with Sequential Organ Failure Assessment (AUC = 0.77) or AIMS65 (AUC = 0.67).

CONCLUSION

The study suggests that RAR is a readily available marker that correlates with illness severity and is an independent predictor of mortality in acute gastrointestinal bleeding, even after accounting for chronic liver disease.