Abstract 20776: Glomerular Hyperfiltration and Microalbuminuria as Predictors of Cardiovascular Events in the Abp-International Study

    loading  Checking for direct PDF access through Ovid

Abstract

Objective: The association between glomerular hyperfiltration (GHF), microalbuminuria and cardiovascular events (CVE) is poorly known.

Hypothesis: We tested the hypothesis that GHF is associated with risk of adverse cardiovascular outcome independently of microalbuminuria in a multiethnic population.

Methods: The analysis was performed in 8794 participants aged 52±16 years (55.0% men) enrolled in 8 prospective studies in Australia, Italy, Japan, and U.S.A.. Hypertension was present in 88.7% of the participants. All were untreated at baseline assessment. Using the 5th and 95th percentiles of the age and sex specific quintiles of CKD-EPI-calculated estimated glomerular filtration rate (eGFR), we identified 3 groups: low (LF), high (GHF) and normal (NF) eGFR. Urinary albumin data were obtained in 6401 participants. The independent prognostic role of eGFR categories and microalbuminuria was tested using study-stratified multivariable Cox models.

Results: The median (IQR) eGFR were: 48.2 (38.1, 58.7), 81.4 (69.6, 95.8) and 111.2 (99.1, 126.7) ml/min/1.73m2 for LF, NF, and GHF participants, respectively (p<0.001). During a median follow-up of 5.7 years there were 722 CVE. Crude event rates were higher for both GHF (1.8 per 100-p-years) and LF (2.1 per 100-p-years) as compared with NF (1.2 per 100-p-years) (p<0.001). In multivariable Cox models including age, sex, average 24-hour BP, smoking, diabetes, and cholesterol, both GHF (HR 1.5, 95% CI,1.2-2.1, p<0.001) and LF (HR 2.0, 95% CI,1.5-2.6, p<0.001) participants had a higher risk of CVE as compared to NF. In the subgroup with urinary albumin data, 565 participants had microalbuminuria and 468 had a CVE. In unadjusted Cox models, GHF (p<0.001) and MA (p<0.001) considered separately were associated with risk of CVE. However, in a multivariable regression model including both eGFR category and MA only GHF remained an independent predictor of CVE (HR 1.5, 95%CI, 1.00-2.22) whereas MA was not accepted by the model (HR 1.0, 95%CI, 0.8-1.3).

Conclusions: These data show that GHF is an independent predictor of CVE in a large multiethnic population of predominantly hypertensive individuals. This association is independent of the effect of GHF on urinary albumin.

Related Topics

    loading  Loading Related Articles