HPLC-detected albuminuria predicts mortality

Dianna J. Magliano, Kevan R. Polkinghorne, Elizabeth L M Barr, Qing Su, Steven J. Chadban, Paul Z. Zimmet, Jonathan E. Shaw, Robert C. Atkins

Research output: Contribution to journalArticlepeer-review

28 Citations (Scopus)


Albuminuria is a risk factor for all-cause mortality. Urinary albumin is traditionally measured with an immunochemical method such as immunonephelometry (IN). High-performance liquid chromatography (HPLC) detects additional albumin that is missed by IN, but it is unknown if HPLC-detected albuminuria also predicts mortality in the general population. We measured urinary albumin at baseline with both IN and HPLC in 10,175 participants in the longitudinal Australian Diabetes, Obesity, and Lifestyle (AusDiab) study. Over 5.2 yr of follow-up, 319 participants died. In a Cox proportional hazards regression model, participants with albuminuria measured by IN or HPLC were approximately twice as likely to die as participants who were normoalbuminuric on both assays. Receiver-operator characteristic curve analyses suggest that each test has a similar ability to predict mortality. However, the HPLC assay provided information on mortality risk that the IN assay did not capture. We conclude that albuminuria, detected by either IN or HPLC, independently predicts mortality in a community-based sample. HPLC identifies some people at increased risk of mortality that IN would not detect.

Original languageEnglish
Pages (from-to)3171-3176
Number of pages6
JournalJournal of the American Society of Nephrology
Issue number12
Publication statusPublished - 1 Dec 2007

ASJC Scopus subject areas

  • Nephrology

Cite this