Combined values of serum albumin, C-reactive protein and body mass index at dialysis initiation accurately predicts long-term mortality

Ryo Takahashi, Yasuhiko Ito, Hiroshi Takahashi, Hideki Ishii, Hirotake Kasuga, Masashi Mizuno, Yasuhiro Suzuki, Yukio Yuzawa, Shoichi Maruyama, Toyoaki Murohara, Enyu Imai, Seiichi Matsuo

Research output: Contribution to journalArticlepeer-review

41 Citations (Scopus)

Abstract

Background: Protein-energy wasting and chronic inflammation are prevalent in patients with end-stage renal disease (ESRD). We investigated the combination of serum albumin, C-reactive protein (CRP) and body mass index (BMI) at initiation of hemodialysis therapy as a predictor of all-cause and cardiovascular disease (CVD) mortality in Japanese ESRD patients. Methods: A total of 1,228 consecutive Japanese ESRD patients on hemodialysis therapy were enrolled and followed for up to 10 years. Patients were divided into quartiles according to levels of albumin, CRP and BMI. Furthermore, to clarify the joint role of these factors, albumin <3.5 g/dl, CRP >4.0 mg/l and BMI <19.6 were defined as risk factors using receiver operating characteristic analysis; thereafter, patients were divided into groups according to the positive number of these factors. Results: Adjusted hazard ratios (HRs) for lower serum albumin, elevated CRP and lower BMI for 10-year all-cause mortality were 1.97, 3.13 and 2.61, respectively. Regarding the combination of these variables, adjusted HRs for mortality were 2.31, 4.28 and 8.07, respectively, in patients having any one factor, any two factors and all three factors. The C-index for an established risk model with these three positive markers was the most accurate for predicting mortality (0.768), as compared to other models with one or two markers. Similar results were seen for CVD mortality. Conclusions: Serum albumin, CRP and BMI at the start of hemodialysis therapy were able to individually stratify the risk of long-term mortality in ESRD patients. Furthermore, a combination of these variables could more accurately predict mortality.

Original languageEnglish
Pages (from-to)136-143
Number of pages8
JournalAmerican Journal of Nephrology
Volume36
Issue number2
DOIs
Publication statusPublished - 08-2012

All Science Journal Classification (ASJC) codes

  • Nephrology

Fingerprint

Dive into the research topics of 'Combined values of serum albumin, C-reactive protein and body mass index at dialysis initiation accurately predicts long-term mortality'. Together they form a unique fingerprint.

Cite this