In patients with insulin dependent diabetes, angiotensin converting enzyme inhibition delays the progression from microalbuminuria to diabetic nephropathy, but previous studies have been too short to show a preservation of kidney function.[1-3] We assessed the effectiveness of angiotensin converting enzyme inhibition on preservation of kidney function in an 8 year prospective, randomised controlled trial.
Patients, methods, and results
Forty four normotensive patients with insulin dependent (type I) diabetes and persistent microalbuminuria (30-300 rog/24 h) were enrolled as previously described in detail. The treatment group (n = 21) was given captopril (100 mg/24 h) and bendrofluazide (2.5 mg/24 h). The 23 remaining patients were left untreated. Diabetic nephropathy was defined as albuminuria persistently [is greater than] 300 mg/24 h. Glomerular filtration rate was measured annually with Crom EDTA plasma clearance over 4 hours.
After 4 years two patients in each group were excluded because they did not attend follow up sessions. Four of the patients in the control group started antihypertensive treatment with diuretics, [Beta] blockers, or a calcium channel blocker. Three patients in the treatment group were changed from bendrofluazide to frusemide because of oedema or diastolic blood pressure [is greater than] 95 mm Hg. After 8 years 16 of the 21 patients in the treatment group and two patients from the control group were subsequently investigated after a treatment pause of 2 months.
The proportion of patients who progressed to diabetic nephropathy was 40% (9/23) in the control group and 10% (2/21) in the captopril group (survival analysis P = 0.019). In the captopril group there was a significant increase in urinary albumin excretion (P [is less than] 0.001) during the treatment pause. In six (38%) of the 16 patients albuminuria exceeded 300 mg/24 h.
Glomerular filtration rate in the captopril group declined from 126 (24) at baseline to 114 (23) ml/min after 8 years but rose again to 126 (21) during the pause in treatment (table). Follow up values of glomerular filtration rate measured during the treatment pause were therefore used whenever available. The decline in mean glomerular filtration rate (ml/min) was 11.8 (95% confidence interval 1.2 to 22.0; t test P value 0.03) and 1.4 (-4.9 to 7.7; P = 0.65) in the control and captopril group, respectively (P = 0.09 between the groups). The fall in glomerular filtration rate during the 8 year study period in the eight control patients who developed nephropathy was 27.3 (3.7 to 51.0; P = 0.03) while glomerular filtration rate increased by 3.8 ( - 3.5 to 11.0) in the six patients treated with captopril with urinary albumin excretion [is greater than] 300 mg/24 h during the treatment pause (P = 0.02 between the groups). Haemoglobin [A.sub.1c] and blood pressure did not differ between the two groups at any time during the study.
Mean glomerular filtration rates (ml/min)in normotensive patients with insulin dependent diabetes at baseline, after 8 years' follow up, and during pause in treatment in captopril and control groups
(*) Patient developed diabetic nephropathy during 8 years of follow up or during treatment pause.
([dagger]) Difference from baseline significant at P = 0.03.
Our study has shown that the beneficial effect of angiotensin converting enzyme inhibition in the prevention of diabetic nephropathy is long lasting and associated with preservation of normal glomerular filtration rate. To obtain a valid determination of the rate of decline in glomerular filtration rate the applied glomerular filtration rate method should have a good accuracy and precision and the observation period should exceed 2 years. These requirements have been fulfilled in our study in contrast with previous studies.[2 3] The second part of the study showed a return in glomerular filtration rate to the values before treatment after 2 months of withdrawal of antihypertensive treatment. The temporary fall in glomerular filtration rate in the intervention group was therefore regarded as a reversible haemodynamic phenomenon. Patients with persistent microalbuminuria at follow up had a stable normal glomerular filtration rate. The clinically significant effect of angiotensin converting enzyme inhibition on preservation of normal glomerular filtration rate was related to prevention of progression from micro-albuminuria to diabetic nephropathy in patients with insulin dependent diabetes.
Contributors: HHP had the original idea for the study. ERM and HHP were responsible for conducting the study and interpreting the results and are guarantors. ERM and EH conducted the clinical evaluation during the 8 years of study. HPH and ERM conducted the clinical evaluation during the treatment pause. UMS performed the assessments of glomerular filtration rate. All authors participated in the interpretation of the results and reporting.
Funding: ERM was funded by a senior research fellowship from the University of Copenhagen. Steno Diabetes Center supplied us with equipment for glomerular filtration analysis and laboratory tests. Squibb donated the tablets and a 1 month research fellowship for ERM.
Competing interests: None declared.
 Mathiesen ER, Hommel E, Giese J, Parving H-H. Efficacy of captopril in postponing nephropathy in normotensive insulin dependent diabetic patients with microalbuminuria. BMJ 1991;303:210-6.
 Viberti G, Mogensen CE, Groop LC, Pauls JF. Effect of captopril on progression to clinical proteinuria in patients with insulin dependent diabetes mellitus and microalbuminuria. JAMA 1994;271:275-9.
 Laffel LMB, McGill JB, Cans DJ. The beneficial effect of angiotensin-converting enzyme inhibition with captopril on diabetic nephropathy in normotensive IDDM patients with microalbuminuria. Am J Med 1995;99:497-504.
 Levey AS, GassmanJ, Hall PM, Walker WG. Assessing the progression of renal disease in clinical studies: effects of duration of follow-up and regression to the mean. J Am Soc Nephrol 1991;1:1087-94.
 Mathiesen ER, Feldt-Rasmussen B, Hommel E, Deckert T, Parving H-H. Stable glomerular filtration rate in normotensive IDDM patients with stable microalbuminuria: a 5 year prospective study. Diabetes Care 1997;20:286-9.
Steno Diabetes Center, DK 2860 Gentofte, Copenhagen, Denmark
Elisabeth R Mathiesen, consultant
Eva Hommel, chief physician
Henrik P Hansen, research fellow
Ulla M Smidt, laboratory technician
Hans-Henrik Parving, professor
Correspondence to: Dr E R Mathiesen, Medical Endocrine Department, University Hospital of Copenhagen, Rigshospitalet, 2100 Copenhagen O, Denmark email@example.com
COPYRIGHT 1999 British Medical Association
COPYRIGHT 2000 Gale Group