LAUSR

BACKGROUND Although urinary alpha-1-microglobulin has been used as a marker of tubular dysfunction, its clinical and prognostic values in patients with acute heart failure have not been validated. METHODS We analyzed 623 patients (74 ± 13 years old, 60.0% male) with acute heart failure in whom urinary alpha-1-microglobulin (A1MG) levels were measured as tubular markers at the time of admission. The primary endpoint was all-cause mortality. RESULTS The median levels of urinary alpha-1-microglobulin with and without correction for urinary creatinine concentration were 8.80 (interquartile range: 4.20-17.7) mg/dL and 12.9 (5.92-30.7) mg/gCr, respectively. Urinary A1MG levels were significantly correlated with all of beta-2-microglobulin (r = 0.77), N-acetyl-β-D-glucosaminidase (r = 0.51), and estimated glomerular filtration rate (r = -0.42); however, alpha-1-microglobulin was most often predicted using beta-2-microglobulin or N-acetyl-β-D-glucosaminidase. During the 488-day (interquartile range: 185-938 days) follow-up, 141 deaths occurred. Higher A1MG levels were associated with higher mortality even after adjustment for other covariates. Only A1MG, but not beta-2-microglobulin or N-acetyl-β-D-glucosaminidase, yielded incremental prognostic information in addition to the preexisting prognostic factors (net-reclassification improvement: 0.254, P = 0.023; integrated discrimination improvement: 0.015, P = 0.028). CONCLUSIONS In patients hospitalized due to acute heart failure, urinary alpha-1-microglobulin was a marker of tubular dysfunction. High alpha-1-microglobulin was associated with all-cause mortality independent of glomerular function and was a better predictor of mortality than urinary beta-2-microglobulin.