Correlation between urine ACR and 24-h proteinuria in a real-world cohort of systemic AL amyloidosis patients

Abstract

A 24-h urine protein collection (24hUP), the gold standard for measuring albuminuria in systemic AL amyloidosis, is cumbersome and inaccurate. We retrospectively reviewed 575 patients with systemic AL amyloidosis to assess the correlation between a urine albumin to creatinine ratio (uACR) and the 24hUP. The uACR correlated strongly with 24hUP at diagnosis (Pearson’s r = 0.87, 95% CI 0.83–0.90) and during the disease course (Pearson’s r = 0.88, 95% CI 0.86–0.90). A uACR ≥300 mg/g estimated a 24hUP ≥ 500 mg with a sensitivity of 92% and specificity of 97% (area under the receiver operating curve = 0.938, 95% CI 0.919–0.957). A uACR cutoff of 3600 mg/g best predicted a 24hUP > 5000 g (sensitivity 93%, specificity 94%), and renal stage at diagnosis was strongly concordant using either 24hUP or uACR as the proteinuria measure (k = 0.823, 95% CI 0.728–0.919). In patients with serial urine collections, a > 30% decrease in uACR predicted a > 30% decrease in 24hUP with a sensitivity of 94%. In conclusion, the uACR is a reliable and convenient method for ruling out proteinuria >500 mg per day, prognosticating renal outcomes, and assessing renal response to therapy. Further studies are needed to validate the uACR cutoffs proposed in this study.