Effects of Perindopril on Renal Histomorphometry in Diabetic Subjects With Microalbuminuria: A 3-Year Placebo-Controlled Biopsy Study A. Nankervis, K. Nicholls, G. Kilmartin, P. Allen, S. Ratnaike, and F. I. R. Martin We conducted a 3-year randomized placebo-controlled double-blind study to determine the effects of the angiotensin- converting enzyme (ACE) inhibitor perindopril (PE) on the progress of renal function and histology in subjects with diabetes and microalbuminuria. Forty non-insulin-dependent (NIDDM) and insulin-dependent (IDDM) diabetic subjects, either normo- tensive or hypertensive, were randomly assigned to receive PE (n = 20) or placebo (n = 20). A pereutaneous renal biopsy was performed initially in all patients and repeated in 29 patients after 3 years. The mean glomerular volume, glomerular basement membrane (GBM) thickness, interstitial fibrosis, sclerosed glomeruli, and volume fraction of capillary lumina were measured histomorphometrically. Before treatment, both groups had similar clinical characteristics, blood pressure, glycosylated hemoglobin (Hb), albumin excretion rate, glomerular filtration rate (GFR), serum creatinine, and renal structural damage. Blood pressure was well controlled in both groups. After 3 years' therapy, there was no significant change in renal function and albuminuria in the PE or placebo groups. The increase in GBM thickness in nine paired biopsies was significantly less in PE-treated subjects (P = .0275). Interstitial fibrosis tended to increase less in the PE group, although this did not reach statistical significance. This study indicates that long-term therapy with PE may decrease or delay the progression of structural glomerular damage in microalbuminuric diabetic subjects. Copyright © 1998 by W.B. Saunders Company I N DIABETES MELLITUS, kidney involvement is character- ized by the development of both histological and functional abnormalities. The effects of angiotensin-converting enzyme (ACE) inhibitors on the progression of renal function have been extensively investigated. Studies in both insulin-dependent (IDDM) and non-insulin-dependent diabetics with nephropathy have shown that ACE inhibitors delay or prevent the progres- sion of proteinuria and reduce the risk of end-stage renal failure in both normotensive and hypertensive patients. 1-4 Moreover, experimental studies have established that treatment with ACE inbJbitors arrests or retards the progression of glomerular damage in various models of diabetic nephropathy (DN). 5-6 However, the effects of ACE inhibitors on the progression of structural renal damage in diabetic subjects with microalbumin- uria have not been previously investigated. We therefore conducted a 3-year randomized placebo- controlled double-blind study to determine the effects of perindopril (PE), a long-acting inhibitor of angiotensin 1-con- verting enzyme, on renal function and histology in microalbu- minuric diabetic subjects. SUBJECTS AND METHODS Patients Forty subjects with NIDDM or IDDM entered the study. The criteria for inclusion were as follows: age 18 to 65 years, microalbuminuria defined as urinary albumin excretion of 20 to 200 mg/L on two of three consecutive occasions, serum creafinine less than 120 gmol/L, and stable glycemic control. Patients were excluded if they had nondiabefic renal disease or other major disease. Patients were either normotensive or hypertensive at entry, but were excluded if they previously received From the Departments of Diabetes and Endocrinology, Nephrology, Anatomical Pathology, and Biochemistry, Royal Melbourne Hospital, Melbourne, Victoria, Australia. Supported by Servier Laboratories Australia. Address reprint requests to A. Nankervis, MD, Department of Diabetes and Endocrinology, Royal Melbourne Hospital Post Office, Victoria 3050. Copyright © 1998 by W.B. Saunders Company 0026-0495/98/4712-I00453.00/0 treatment with ACE inhibitor. Hypertension was defined as a resting blood pressure greater than 140 mm Hg systolic (SBP) or 90 mm Hg diastolic (DBP). Subjects were evaluated prospectively for 3 years. Table 1 lists the clinical characteristics at entry. Study Design The study was approved by the Board of Medical Research and Ethics Committee of The Royal Melbourne Hospital, and patients provided written consent before starting the study. The subjects were randomly allocated to receive either 4 mg PE (n = 20) or placebo (n = 20) orally once per day in a double-blind manner. Unmarked formulations were provided by Servier Laboratory (Cedex, France). The supine and erect SBP and DBP were measured using a Dinamap (Critikon, USA) machine, initially weekly and then every 3 months. The dose of PE was not altered for blood pressure control; if blood pressure became or remained elevated, further antihypertensive medication was added. Clinical and metabolic surveillance was maintained throughout the study. Blood pressure and weight were measured at 3-month intervals. The urine protein excretion, albumin excretion, serum creatinine, glucose, glycated hemoglobin (Hb), and glomerular filtration rate (GFR) were measured at baseline and at 12-month intervals over 3 years. A percntaneous renal biopsy was obtained from all subjects at entry. Thirty-one subjects continued the study for 3 years, and 29 underwent a second renal biopsy at the end of the study period. Renal biopsy. Biopsies were performed percutaneously under ultra- sound control. Tissue was divided for light microscopy, immunoperoxi- dase, and electron microscopy. Tissue for light microscopy was immediately fixed in mercuric-Formalin solution. Paraffin-embedded blocks were cut at 1 grn and stained with periodic acid-Schiff, silver methionine, and silver masson trichrome. Histomorphometry. The mean glomerular volume and capillary volume were calculated by the method of Hirose et al7 using the arithmetic mean profile surface area described by Bilous et al. 8 Point counting was performed at 400× magnification using a 25-1am eyepiece graticule. The distance between points corresponded to 25 gm. The glomerulus for point counting was defined as PAS-positive material and capillary lumen, but excluded Bowman's capsule and the space between glomemlar segments and capillary loops. Sclerosed glomeruli were those in which all glomerular tufts were involuted or replaced by scar, and are expressed as a percentage of total glomeruli. The relative cortical interstitial area was calculated and expressed as a percentage of the total area. Interstitial fibrosis was estimated as points on the tissue between tubules but excluding glomeruli, arteries, and arterioles. 12 Metabolism, Vo147, No 12, Suppl 1 (December),1998:pp 12-15