Martin J. Wiseman

Glycaemia, arterial pressure and micro-albuminuria in Type 1 (insulin-dependent) diabetes mellitus

SummaryPlasma glucose control and arterial pressure were assessed in 28 Type 1 (insulin-dependent) diabetic patients with different degrees of micro-albuminuria. They were divided into two groups according to their urinary albumin excretion rate: a low micro-albuminuria group (n= 16) with albumin excretion ranging between 12.1 and 28.9 μg/min and a high micro-albuminuria group (n= 12) with albumin excretion between 32.4 and 91.3 μg/min. The groups were matched for age, sex and duration of diabetes with the same number of normo-albuminuric (2.0–10.4 μg/min) diabetic control subjects. Both the low and high micro-albuminuria groups had significantly higher glycosylated haemoglobin levels and mean plasma glucose concentrations during a 24-h profile than their respective normo-albuminuric control subjects. A correlation between glycosylated haemoglobin level and urinary albumin excretion rate was found in the whole study group (r= 0.48; p< 0.001). Arterial pressure (both systolic and diastolic) was significantly higher in the high micro-albuminuria group than in either the control group or the low microalbuminuria group. A significant correlation was found between arterial pressure and albumin excretion rate in the whole study population (r= 0.49; p< 0.001) as well as in the pooled micro-albuminuria groups (r= 0.43; p< 0.05). Multiple regression analysis showed that glycosylated haemoglobin and arterial pressure levels were independently correlated with albumin excretion rates. Diabetic patients with micro-albuminuria of any degree have worse glycaemic control than normo-albuminuric patients. Higher levels of arterial pressure, though often sub-hypertensive, are associated with levels of micro-albuminuria predictive of later development of clinical proteinuria. Thus high plasma glucose and high arterial pressure, or both, characterise those diabetic patients at increased risk of nephropathy. These indices of risk are potentially reversible.

Effect of blood glucose control on increased glomerular filtration rate and kidney size in insulin-dependent diabetes.

To investigate the relation between blood glucose control on the one hand and an increased glomerular filtration rate and enlarged kidneys on the other, we studied 12 patients with insulin-dependent diabetes and an increased glomerular filtration rate for a year after they were randomly assigned either to continuous subcutaneous insulin infusion or to unchanged conventional therapy. Glycemic control, measured by mean plasma concentrations of glucose and glycosylated hemoglobin, was rapidly and significantly improved (P less than 0.001) in the pump group but did not change in the conventional-treatment group. In the pump group, the glomerular filtration rate fell significantly in the study period (P less than 0.001) and became normal in four of the six patients. It did not change in the conventional-treatment group. There was no change in kidney volume in either group. At the end of a year, a return to conventional insulin treatment in the pump group resulted in both metabolic deterioration and a significant rise in the mean glomerular filtration rate toward base-line values. We conclude that in patients with established insulin-dependent diabetes, strict glycemic control normalizes the glomerular filtration rate, although the kidneys may remain enlarged.

Glomerular Hyperfiltration in the Prediction of Nephropathy in IDDM: A 10-Year Follow-Up Study

Glomerular hyperfiltration has been proposed as an independent risk factor for the development of diabetic nephropathy in patients with IDDM. In a case-controlled prospective study of IDDM patients without albuminuria, serial glomerular filtration rate (GFR) measurements were performed over an observation period of 10 years. A group of 25 IDDM patients (20 men, 5 women; initial age, 29 [17–49] years) with glomerular hyperfiltration (GFR >135 ml · min−1 · 1.73 m−2) were matched for age, sex, and duration of diabetes with 25 IDDM patients (20 men, 5 women; initial age, 30 [17–48] years) with glomerular normofiltration (GFR 83–135 ml · min−1 · 1.73 m−2). GFR, urinary albumin excretion rate (AER), blood pressure, and glycated hemoglobin were measured at baseline and at 5, 8, and 10 years. The two groups had similar entry levels of blood pressure, AER, and glycated hemoglobin. Metabolic control was similar in the two groups during follow-up. The final GFR remained higher in the group with hyperfiltration (122 [109–135] vs. 103 [95–111] ml · min−1 · 1.73 m−2; P = 0.02) despite a nonsignificantly faster rate of fall of GFR compared with that of the control group (2.54 [1.20–3.88] vs. 1.50 [1.01–1.99] ml · min−1 · year−1; P = 0.14). A similar number of patients in each group progressed to either microalbuminuria or macroalbuminuria (n = 4 vs. n = 3) or developed hypertension (blood pressure, >160/95 mmHg; n = 3 vs. n = 4). End-of-study AER was, however, higher in the group with hyperfiltration (geometric mean [95% CI]: 18.9 [11.3–31.6] vs. 11.0 [8.1–15.0]; P = 0.05), and baseline glomerular hyperfiltration was an independent determinant of end-of-study blood pressure (P = 0.04). The strongest predictors of end-of-study AER and blood pressure were their baseline values (P < 0.04 and P < 0.01, respectively). In conclusion, levels of AER and blood pressure are the main risk factors for renal outcome, while glomerular hyperfiltration appears to play a lesser role.

Changes in renal function in response to protein restricted diet in type 1 (insulin-dependent) diabetic patients

SummaryGlomerular filtration rate, renal plasma flow and urinary albumin excretion rate were measured during insulin-induced euglycaemia in 12 male Type 1 (insulin-dependent) diabetic patients after a 3-week period of low protein diet (45 g/day) or a similar period on unchanged conventional diet (103 g/day). No changes in glycaemic control, indicated by home blood glucose profiles and serum fractosamine concentration, or in arterial pressure, were noted on either diet. On low protein diet, glomerular filtration rate was lower (p<0.001) in all patients, but there was no difference in renal plasma flow between low protein diet and normal protein diet; filtration fraction fell significantly on low protein diet (p<0.001). Fractional clearance of albumin was also lower (p<0.05) on low protein diet. This study suggests that reduction of dietary protein induces, independently of changes in plasma glucose and arterial pressure, modifications in glomeralar filtration rate, filtration fraction and fractional clearance of albumin, which may be associated with a beneficial effect on the evolution of diabetic renal disease.

Threshold Effect of Plasma Glucose in the Glomerular Hyperfiltration of Diabetes

Glomerular filtration rate (GFR) is abnormally high in some, but not all, insulin-dependent diabetic patients. The potential importance of this hyperfiltration lies in its possible link with later severe diabetic kidney disease. Inadequate glycaemic control is closely related to hyperfiltration but the mechanisms of the association are obscure. GFR and prevailing plasma glucose concentration were examined in a group of insulin-dependent diabetics without clinical proteinuria and in a group of non-diabetics, and their relationships observed using linear and multiple regression analysis. A positive correlation (r = 0.30, p less than 0.05) is found between mean plasma glucose concentration and GFR up to a mean plasma glucose level of approximately 13.5 mmol/l. Glycaemia in excess of this degree tends to be associated with a lower GFR. Multiple regression analysis confirmed the independence of plasma glucose as a determinant of GFR (p less than 0.05) at concentrations below 13.5 mmol/l. GFR declined significantly with age, but independently of diabetes duration, in the diabetic group (r = -0.48, p less than 0.001). GFR in the control group showed a statistically non-significant decline with age.

Glomerular Hyperfiltration and Urinary Prostaglandins in Type 1 Diabetes Mellitus

In order to determine whether glomerular hyperfiltration in diabetes is related to renal prostaglandin production we have studied the urinary excretion of PGE2, 6‐keto‐PGF1α, and TXB2 in two sex, age and duration of diabetes matched groups of 9 and 10 Type 1 diabetic patients with either normal (mean 121, range 105–129 ml min−1 1.73 m−2) or supranormal glomerular filtration rate (154, 135–206 ml min−1 1.73 m−2). A group of 15 matched healthy volunteers served as control subjects. Urine was collected overnight for an uninterrupted period of at least 6 h. All studies in the patients were performed during insulin‐induced sustained euglycaemia to prevent the confounding effect of variable degrees of blood glucose control on urinary prostaglandin excretion. Blood pressure was normal in all subjects. Urinary excretion of 6‐keto‐PGF1α was significantly higher in the patients with glomerular hyperfiltration (median 17.1, range 4.5–33.6 ng h−1) than in those without (8.8, 1.5–13.8 ng h−1; p < 0.05) or in normal control subjects (9.6, 5.2–15.5 ng h−1; p < 0.05). No significant differences were found in the excretion rates of PGE2 and TXB2 between the three groups. Under conditions of controlled plasma glucose and insulin concentrations the urinary excretion of 6‐keto‐PGF1α, the stable breakdown product of PGI2, a compound of endothelial, possibly glomerular, origin was elevated only in the diabetic patients with glomerular hyperfiltration.

The glomerular hyperfiltration of diabetes is not associated with elevated plasma levels of glucagon and growth hormone.

SummaryPlasma concentrations of glucagon, growth hormone and glucose were measured hourly during an ordinary treatment day in 11 Type 1 (insulin-dependent) diabetic patients with high glomerular filtration rate, 11 Type 1 diabetic patients with normal glomerular filtration rate matched for age, diabetes duration and sex, and five healthy control subjects, simultaneously with the measurement of the glomerular filtration rate using 51Cr EDTA clearance. Plasma glucagon profiles were not statistically distinguishable (p = 0.49) from control values in either group, although they were somewhat lower in the hyperfiltering group. Plasma growth hormone values were higher than control (p=0.07) in both diabetic groups, but were not different between these two groups (p = 0.94). All indices of glycaemic control (glycosylated haemoglobin, urinary glucose excretion, and plasma glucose concentration) were higher in the hyperfiltering group, although no single index reached statistical significance. No correlations between concentrations of these substances and glomerular filtration rate were found. Elevated plasma concentrations of glucagon and growth hormone do not characterise those diabetic patients with high glomerular filtration rate.

Prevention of Diabetic Nephropathy: Markers of Disease and Perspectives for Intervention

Renal failure is a complication of diabetes mellitus ultimately affecting approximately 45% of insulindependent patients.’ Its onset is heralded by the development of dipstix positive proteinuria (i.e. total urinary protein excretion >0.5 g/24 h) usually after 10 years or more of diabetes in a patient who has concomitant retinopathy and rising arterial pressure, but is free from other renal disease, urinary tract infection, and cardiac failure. The incidence of diabetic nephropathy peaks at about 16 years after the onset of diabetes and once manifest the condition progresses relentlessly to end stage renal failure (ESRF). In 50% of affected patients this occurs within 7 years of onset of clinical nephropathy.’ With the appearance of persistent proteinuria glomerular filtrafion declines linearly with time at an average rate of 1.2 ml/min/month, ranging widely between different individuals from 0.5 to 2.4 ml/min/month.2 The reasons for the different rates of decline are largely unknown but varying degrees of blood pressure rise may play a part. Indeed arterial pressure starts to climb at a very early stage of renal involvement in diabete~.~ With the fall in glomerular filtration rate (GFR), proteinuria and clearances of albumin and IgG increa~e.~ There is a change from proteinuria highly selective for albumin in the early stages when the GFR is st i l l normal or only moderately reduced to a low selectivity proteinuria with proportionally more IgG being cleared when the GFR is markedly reduced. One possible interpretation of this phenomenon is that size-selectivity defects in the glomerular capillary membrane develop in late nephropathy, accounting for the proportionately increased transit of lgGI5 whereas the early selective albuminuria may be due to a change in the charge selectivity properties of the glomerular A compensatory intraglornerular pressure increase in the surviving glomeruli would almost certainly accompany the membrane injuries and perhaps perpetuate them.’

Genesis and evolution of proteinuria in diabetes mellitus

SummaryApproximately 40–45% of insulin-dependent diabetic (IDD) patients will develop, with time, clinical proteinuria, a forerunner of certain renal failure. Before this, however, up to 45% of IDD patients excrete supranormal amounts of protein in the urine, though still undetectable by dipstix test. This microproteinuria appears to be glomerular in origin, consists mainly of albumin and IgG, and is associated with poor glycemic control and marginal elevation of arterial pressure. Glomerular hemodynamic disturbances, and loss of charge selectivity of the glomerular membrane, are probably responsible for this microproteinuria, which appears reversible by correction of hyperglycemia and raised blood pressure. Once the dipstix test becomes positive (i.e. total urinary protein excretion exceeds 0.5 g/24h) and blood pressure rises into the hypertensive range, glomerular filtration rate (GFR) falls relentlessly. By the time GFR is as low as 20 ml/min/1.73 m2, more IgG relative to albumin is being filtered, giving rise to a low selectivity proteinuria, a condition consistent with changes in the size selectivity properties of the glomerular filtre. Glycemic control does not affect the decline in GFR, although blood pressure control and low protein diet can slow it, presumably by altering the self-perpetuating hemodynamic disturbances that occur in surviving glomeruli. The recent demonstration that IDD patients with microalbuminuria in excess of 30 µg/min have approximatley a 20-fold increase in risk of developing persistent detectable proteinuria has provided a link between these two phases of diabetic nephropathy. The reversibility of the early microalbuminuria heralds a real chance of preventing the later irreversible phase of end-stage renal failure.