The primary end points (PEs) were fatal and nonfatal myocardial infarction and stroke and other cardiovascular deaths.
Creatinine and glomerular filtration rate (GFR), estimated using the formulas of the Modification of Diet in Renal Disease study group and Cockroft and Gault and in a subsample (n=4929) of microalbuminuria and interaction terms of microalbuminuria and renal function, were related to the risk of the PE using Cox proportional hazards model after full adjustment.
Increased creatinine (P<0.001), decreased GFR from Cockroft and Gault (P=0.001), and decreased GFR from the Modification of Diet in Renal Disease study group (P=0.001) were all independent risk factors for the PE. Stepwise exclusion of patients with the poorest renal function excluded the possibility that the relationship between decreasing renal function and the PE was driven only by patients with severely impaired renal function.
Microalbuminuria and all 3 of the indices of renal function predicted the PE independent of each other.
There was a significant interaction between microalbuminuria and GFR from Cockroft and Gault (P=0.040) in prediction of the PE. Both renal function and microalbuminuria add independent prognostic information regarding cardiovascular risk in hypertensive patients.
The cardiovascular risk associated with microalbuminuria increases with a decline in GFR, as demonstrated by a significant interaction between microalbuminuria and GFR from Cockroft and Gault. Because estimation of the total cardiovascular risk is essential for the aggressiveness of risk factor interventions, simultaneous inclusion of GFR and microalbuminuria in global cardiovascular risk assessment is essential.