Aims/hypothesis The purpose of this study was to evaluate the optimal renoprotective effect of ultrahigh doses of lisinopril, as reflected by short-term changes in urinary albumin excretion rate (UAER), in type 1 diabetic patients with diabetic nephropathy. Methods At the Steno Diabetes Center, 49 type 1 diabetic patients with diabetic nephropathy completed this double-masked randomised crossover trial consisting of an initial washout period followed by three treatment periods each lasting 2 months, where all patients received lisinopril 20, 40 and 60 mg once daily in randomised order in addition to slow-release furosemide. Allocation was concealed by sequentially numbered opaque sealed envelopes. UAER, 24 h ambulatory blood pressure (ABP) and estimated GFR were determined at baseline and after each treatment period. Results All 49 patients completed all three treatment periods. Baseline values were: UAER (geometric mean [95% CI]) 362 (240-545) mg/24 h, 24 h ABP (mean [SD]) 142 (14)/74 (8) mmHg and estimated GFR 75 (29) ml min-1 1.73 m -2. Reductions in UAER from baseline were 63%, 71% and 70%, respectively, with the increasing doses of lisinopril (p∈<∈0.001). Compared with lisinopril 20 mg there was a further reduction in UAER of 23% with lisinopril 40 mg and 19% with 60 mg, p∈<∈0.05. ABP was reduced from baseline by 10/5, 13/7 and 12/7 mmHg (p∈<∈0.001 vs baseline, p∈<∈0.05 for diastolic ABP 20 vs 40 mg, otherwise NS between doses). The difference in UAER between 20 and 40 mg lisinopril was significant after adjustment for changes in ABP (p∈<∈0.01). Two patients were excluded from the study because of an increase in plasma creatinine and one because of high BP; otherwise the study medication was well tolerated with few, mild, dose-independent adverse effects. Conclusions/interpretation Lisinopril 40 mg once daily is generally safe and offers additional reductions in BP and UAER in comparison with the currently recommended dose of 20 mg. Lisinopril 60 mg offers no further beneficial effect. Trial registration: ClinicalTrials.gov NCT00118976 Funding: This study was financed out of local funds and was not supported by the medical industry.