Microalbuminuria in critically ill medical patients: prevalence, predictors, and prognostic significance.

OBJECTIVE: To ascertain the prevalence, predictors, and prognostic significance of microalbuminuria in critically ill patients.

DESIGN: Prospective cohort study.

SETTING: Medical intensive care unit of a community teaching hospital.

PATIENTS: Admitted critically ill patients.

MEASUREMENTS AND MAIN RESULTS: We measured serial spot urine albumin-creatinine ratios in 104 critically ill patients, with a median age of 64.5 yrs and median Acute Physiology and Chronic Health Evaluation (APACHE) II and Sequential Organ Failure Assessment (SOFA) scores of 20.5 and 5.0, respectively. Sixty-nine percent of the patients had microalbuminuria or clinical proteinuria and 43.3% had an albumin-creatinine ratio >/=100 mg/g at admission. The acuity of illness, being non-White, and having diabetes mellitus were independent predictors of albumin-creatinine ratio >/=100 mg/g. The overall mortality rate was 26.9% (28/104). Patients with an albumin-creatinine ratio >/=100 mg/g were 2.7 times as likely to die compared with those with an albumin-creatinine ratio <100 mg/g, even after simultaneous adjustments for age, and APACHE II and SOFA scores (odds ratio, 2.7; 95% confidence interval, 1.1-7.2, p =.04). The association of albumin-creatinine ratio >/=100 mg/g with death was consistent across age, ethnicity, renal function, acuity of illness, and comorbid conditions. Among survivors, patients with an albumin-creatinine ratio >/=100 mg/g stayed approximately 5 days longer in the hospital (p =.0007). Overall, the albumin-creatinine ratio shared similar predictive characteristics with APACHE II and SOFA scores.

CONCLUSIONS: This study confirms a high prevalence of microalbuminuria in critically ill patients and suggests that an albumin-creatinine ratio >/=100 mg/g is an independent predictor of mortality and hospital stay.