1. In diabetes mellitus a selective increase in the excretion of albumin generally precedes the occurrence of demonstrable loss of glomerular size-selectivity. However, even in this (microalbuminuric) phase of diabetic nephropathy a defect in glomerular barrier function can be demonstrated during infusion of atrial natriuretic peptide. 2. The aim of this study was to investigate whether angiotensin-converting enzyme inhibition could prevent the proteinuric response to atrial natriuretic peptide in these patients. We performed infusions of atrial natriuretic peptide (0.01 μg min −1 kg −1 ) in 10 patients with insulin-dependent diabetes mellitus and microalbuminuria (urinary albumin excretion 90 ± 44 mg/day), both before and after 1 month of treatment with enalapril (20 mg once daily). 3. Despite a 40% reduction in proteinuria, angiotensin-converting enzyme inhibition did not prevent the atrial natriuretic peptide-induced increase in protein excretion. Both before and during angiotensin-converting enzyme inhibition, atrial natriuretic peptide infusion resulted in a significant increase in the fractional excretion of large dextran molecules, which is compatible with an increase in flow through large unrestrictive ‘shunt' pores. Atrial natriuretic peptide infusion also induced an increase in the transcapillary escape rate of albumin and angiotensin-converting enzyme inhibition also failed to prevent this effect of atrial natriuretic peptide on peripheral capillary permeability. 4. We conclude that angiotensin-converting enzyme inhibition during 1 month does not correct the capillary barrier function defect in patients with diabetes mellitus and microalbuminuria that is unmasked by atrial natriuretic peptide infusion.

1. The effects of frusemide on glomerular permeability were investigated in eight proteinuric patients by measuring the fractional protein and dextran clearances and correlating these observations with changes in renal haemodynamics. 2. Frusemide increased significantly the fractional albumin and IgG clearances. The fractional dextran clearances of molecules with a radius ≥ 5.4 run also increased significantly after frusemide injection. Pretreatment with indomethacin partly inhibited the increments in clearances of macromolecules. 3. The changes in the fractional protein clearances correlated significantly with those of inulin clearance There was also a high degree of correlation between the changes in fractional protein clearances and prostanoid excretion. 4. The data obtained suggest that frusemide increases glomerular permeability by influencing the effective pores of the glomerular capillary wall. The increased permeability possibly is due to changes in prostanoid synthesis.

1. The renal clearance of dextran and protein of different molecular sizes has been measured in 42 proteinuric patients with diabetes and glomerulonephritis. 2. In both patient groups the fractional low molecular (radii 3.0–3.6 nm) dextran clearances were significantly decreased whereas the fractional high molecular (radii 5.1–7.8 nm) dextran clearances were increased as compared with the controls. 3. A porosity/charge index was constructed by calculating the ratio of fractional clearances of neutral to anionic macromolecules (θ D36 /θ ALB , θ D5.1 /θ IgG ). In diabetic patients the relation between the fractional clearances of neutral and anionic macromolecules was dependent on glomerular filtration rate. 4. It seems likely that the altered glomerular permeability in diabetic nephropathy is associated with a decrease of charge as renal function deteriorates. In glomerulonephritis the process is different and there is a high degree of variation from case to case.