Glomerulopathy in normoalbuminuric adolescents with type 1 diabetes : relations between structure, function, metabolic control and ambulatory blood pressure

Abstract

An increase in the incidence of type 1 diabetes in childhood has been reported in several countries, especially among the youngest children. Although clinically apparent diabetic vascular disease is rare in children and adolescents, 30-40% are at risk of developing nephropathy during their lifetime which may lead to renal failure. The determination of risk factors and screening for diabetic microvascular complications should be done even in childhood and adolescence to find those in the early stages who may respond to aggressive intervention strategies. The main aims of this thesis were to improve the understanding of the very early events that initiate the process leading to albuminuria and renal injury and to find patients at risk of developing diabetic renal damage.

Children and adolescents with type 1 diabetes in the Department of Paediatrics at Karolinska University Hospital Huddinge have been followed longitudinally with renal function tests since 1978 and with HbA1/HbA1c since 1980. In 1992-1994 cross-sectional investigations were done on 46 patients over 12 years of age and who had had diabetes for more than five years. These included: 1. renal biopsies to evaluate the renal morphology on electron microscopy; 2. 24-h ambulatory blood pressure measurements to determine the day- and nighttime blood pressures (BP) and heart rate (HR); 3. renal clearances of inulin (glomerular filtration rate, GFR) and PAH (effective renal plasma flow, ERPF) to calculate the filtration fraction (FF=GFR/ERPF); and 4. the urinary albumin excretion rate (UAE). The mean age of the patients was 18 years and they had had a mean diabetes duration of 10 years. None of them had persistent microalbuminuria.

Basement membrane thickness (BMT), mesangial matrix volume fraction per glomerulus (Vv(matrix/glom)) and foot process width were larger in the patients than healthy controls. Spontaneous remission (period from the onset to the last visit when the insulin requirement was <0.5 U/kg/day and HbA1c <7.0%) occurred in 58% of the patients with a median duration of 6.6 (range 2.6-16.2) months. Remitters had lower day- and nighttime diastolic BPs, thinner BMT and less Vv(matrix/glom) than those without a remission. At least one third of the patients had systolic and diastolic nighttime BPs above the 90 'h percentile while their HR was within normal range. The nighttime mean arterial BP (MAP) correlated directly with the BMT, Vv(matrix/glom) and foot process width. In multiple regression analyses, 44% of the nighttime MAP and 57% of the BMT were explained by the long-term HbA1c, nighttime HR and body height and 43 % of the Vv(matrix/glom) by long-term HbA1c, nighttime HR and duration of diabetes. The effect of BP on BMT and Vv(matrix/glom) disappeared when the HR was included.

GFR and FF at the time of biopsy and the mean previous GFR and FF were higher in the patients with diabetes than in healthy controls. The mean previous FF correlated directly with the longterm HbA1c and both variables correlated with the BMT and Vv(matrix/glom). UAE correlated directly with the mean previous FF and foot process width. In conclusion, after about 10 years' duration of type 1 diabetes, we found typical diabetic glomerular changes, higher nighttime BP, increases in GFR and FF before and at the time of the biopsy in normoalbuminuric adolescents, as compared to controls. Poor metabolic control from the onset of diabetes was related to glomerulopathy and hypertension, and may in itself affect later metabolic control. Nighttime BP and especially nighttime HR seemed to be related to glomerulopathy changes which suggest that a disturbance in sympathovagal balance may have a pathogenic effect. An increase in FF may reflect intraglomerular hypertension and lead to glomerular changes and albuminuria.