Purpose: To determine whether response of albumin excretion rate, to angiotensin-converting enzyme (ACE) inhibitors has a threshold in patients with type 1 diabetes mellitus and microalbuminuria and to examine treatment effect according to covariates. Data Sources: Studies were identified by searching MEDLINE and related bibliographies. Study Selection: selected studies included at least 10 normotensive patients with type 1 diabetes mellitus and microalbuminuria, had a placebo or nonintervention group, and included at least 1 year of follow-up. Data Extraction: Raw data were obtained for 698 patients from the 12 identified trials. Analysis of treatment effect at 2 years was restricted to trials with at least 2 years of follow-up (646 patients from 10 trials). Data Synthesis: In patients receiving ACE inhibitors, progression to macroalbuminuria was reduced (odds ratio, 0.38 [95% Cl, 0.25 to 0.57]) and the odds ratio for regression to normoalbuminuria was 3.07 (Cl, 2.15 to 4.44). At 2 years, albumin excretion rate was 50.5% (Cl, 29.2% to 65.5%) lower in treated patients than in those receiving placebo (P < 0.001). Estimated treatment effect varied by baseline albumin excretion rate (74.1% and 17.8% in patients with a rate of 200 g/min and 20 mug/min, respectively [P = 0.04]) but not by patient subgroup. Adjustment for change in blood pressure attenuated the treatment difference in albumin excretion rate at 2 years to 45.1% (Cl, 18.6% to 63.1%, P < 0.001). Conclusions: In normotensive patients with type 1 diabetes mellitus and microalbuminuria, ACE inhibitors significantly reduced progression to macroalbuminuria and increased chances of regression. Beneficial effects were weaker at the lowest levels of microalbuminuria but did not differ according to other baseline risk factors. Changes in blood pressure cannot entirely explain the antiproteinuric effect of ACE inhibitors.