Microalbuminuria in diabetic patients is associated with ischemic heart disease and insulin resistance. We previously found a 9% prevalence of microalbuminuria in a nondiabetic population that we have reassessed, investigating associations of microalbuminuria with hypertension, dyslipidemia, hyperinsulinemia, and sodium-lithium countertransport. Of 125 subjects reexamined, 42 had been microalbuminuric 3 years previously. Twelve of these (29%) were microalbuminuric on at least one sample at follow-up, and 30 (76%) were normoalbuminuric on two. Of the 79 previously normoalbuminuric subjects, 12 (15%) became microalbuminuric on one sample, while 67 (85%) remained normoalbuminuric. Subjects who were microalbuminuric at both screening and recall were older (mean +/- SD, 65.9 +/- 11 versus 59.1 +/- 10.2 years, P = .04), with a higher waist-to-hip ratio (0.93 +/- 0.09 versus 0.86 +/- 0.08, P = .008) and at recall, on univariate analysis, had higher specific insulin (44.2 [range, 16.9 to 157.0] versus 28.4 [7.4 to 129.0] pmol/L, P = .005), intact proinsulin (5.1 [1.5 to 11.0] versus 3.0 [0.8 to 14.6] pmol/L, P = .003), and des-31,32-proinsulin (5.0 [0.5 to 9.8] versus 1.0 [0.5 to 12.2] pmol/L, P = .004) concentrations. There was also a significant difference in des-31,32-proinsulin concentration, after adjustment for covariates (P = .013), between subjects classified either as microalbuminuric or as normoalbuminuric at screening. There was no difference in body mass index; fasting blood glucose; systolic or diastolic blood pressure; total, HDL, or LDL cholesterol; triglycerides; plasminogen activator inhibitor-1; or sodium-lithium countertransport activity between consistently normoalbuminuric and persistently microalbuminuric subjects.(ABSTRACT TRUNCATED AT 250 WORDS)