The increased proximal tubular reabsorption of sodium and water is maintained in long-term insulin-dependent diabetics with early nephropathy

Proximal tubule cell hypothesis for cardiorenal syndrome in diabetes

Incidence of cardiovascular disease (CVD) is remarkably high among patients with chronic kidney disease (CKD), even in the early microalbuminuric stages with normal glomerular filtration rates. Proximal tubule cells (PTCs) mediate metabolism and urinary excretion of vasculotoxic substances via apical and basolateral receptors and transporters. These cells also retrieve vasculoprotective substances from circulation or synthesize them for release into the circulation. PTCs are also involved in the uptake of sodium and phosphate, which are critical for hemodynamic regulation and maintaining the mineral balance, respectively. Dysregulation of PTC functions in CKD is likely to be associated with the development of CVD and is linked to the progression to end-stage renal disease. In particular, PTC dysfunction occurs early in diabetic nephropathy, a leading cause of CKD. It is therefore important to elucidate the mechanisms of PTC dysfunction to develop therapeutic strategies for treating cardiorenal syndrome in diabetes.

Effect of acute hyperglycaemia on sodium handling and excretion of nitric oxide metabolites, bradykinin, and cGMP in Type 1 diabetes mellitus

AIMS

The aim of this study was to evaluate the effect of acutely induced hyperglycaemia on renal sodium handling and to explore the role of the bradykinin-nitric oxide-cGMP signalling pathway.

PATIENTS AND METHODS

We compared 20 Type 1 diabetic (DM1) patients without microalbuminuria with 15 weight-, age-, and sex-matched healthy controls (C). Clearances of para-aminohippuric acid (CPAH), inulin (Cin), lithium, sodium, and urinary nitrite/nitrate (NOx), cGMP and bradykinin excretion rates were measured in two 90-min periods: a glycaemic clamp-induced euglycaemia (5 mmol/l-period I) and hyperglycaemia (12 mmol/l-period II) (Study 1) and during time-controlled euglycaemia (5 mmol/l-period I and 5 mmol/l-period II) to avoid the effects of time and volume load (Study 2).

RESULTS

Cin and CPAH were not significantly different during euglycaemia (period I of Study 1) in DM1 and controls, whereas fractional excretion of sodium was decreased in DM1 (1.84 +/- 0.75 vs. 2.36 +/- 0.67%; P < 0.05) due to an increase in fractional distal tubular reabsorption of sodium (94.01 +/- 1.94 vs. 92.24 +/- 2.47%; P < 0.05). A comparison of changes during Study 1 and Study 2 revealed acute hyperglycaemia did not change renal haemodynamics significantly, while fractional distal tubular reabsorption of sodium increased (DM1: P < 0.05; C: P < 0.01) and fractional excretion of sodium decreased (P < 0.01) in both groups. The urinary excretion rates of NOx were comparable during euglycaemia in DM1 and C. While in C, they significantly increased during Study 1 (period I: 382 +/- 217 vs. period II: 515 +/- 254 nmol/min; P < 0.01) and Study 2 (period I: 202.9 +/- 176.8 vs. period II: 297.2 +/- 267.5 nmol/min; P < 0.05) as a consequence of the water load, no changes were found in DM1. The urinary excretion of bradykinin was lower in DM1 compared with C (0.84 +/- 0.68 vs. 1.20 +/- 0.85 micro g/min; P < 0.01) during euglycaemia; it was not affected by hyperglycaemia. There were no significant differences between DM1 and C and in cGMP urinary excretion rates following hyperglycaemia.

CONCLUSION

This study demonstrates that DM1 without renal haemodynamic alterations is associated with impaired renal sodium handling. Moreover, we did not find a relationship between the renal excretion rates of vasoactive mediators and sodium handling due to hyperglycaemia.

Sodium retention and insulin treatment in insulin-dependent diabetes mellitus

The hypothesis that total body exchangeable sodium (ENa) is elevated in type 1 (insulin-dependent) diabetic patients with short-duration diabetes and no signs of microangiopathy was tested. Also tested was whether peripheral hyperinsulinaemia, in terms of the amounts of insulin injected subcutaneously, contributes to the increased ENa. Three studies were performed. Study 1 was a cross-sectional study comprising 28 type 1 diabetic men (aged 18-35 years) with short-duration diabetes (< 5 years) and no signs of diabetic complications, and 22 control subjects. Study 2 was a prospective study of 17 newly diagnosed diabetic patients (aged 20-35 years, median 27 years) who were studied on two occasions on different insulin doses. Study 3 was a 12-month prospective intervention study of 21 type 1 diabetic patients with incipient nephropathy, who had been randomized either to receive continuous subcutaneous insulin infusion for improvement of glycaemic control or to remain on conventional insulin treatment. In study 1, ENa was higher in short-duration type 1 diabetic men than in controls (3003 +/- 325 vs 2849 +/- 207 mEq/1.73 m2, P < 0.05) and was correlated significantly with the insulin dose (r = 0.38, P < 0.05). In study 2, of the newly diagnosed diabetic patients, 11 received a reduced insulin dose and 6 an increased dose as compared with the initial study. ENa was reduced in all patients receiving less insulin (P < 0.001) and remained unchanged in patients receiving more insulin.(ABSTRACT TRUNCATED AT 250 WORDS)

Segmental tubular sodium reabsorption in type 1 diabetes

Segmental tubular sodium reabsorption in Type 1 (insulin-dependent) diabetes was measured in 36 patients in a cross-sectional study including one group (n = 13) without significant albuminuria (UalbV < 30 mg 24 h-1), one group (n = 16) with albuminuria in the range from 30 to 300 mg 24 h-1, and a group (n = 7) with nephropathy (UalbV > 300 mg 24 h-1). Lithium clearance was used to measure end proximal delivery. From end proximal delivery, 51Cr-EDTA clearance (GFR) and sodium clearance, segmental tubular reabsorption was calculated. For all patients, GFR was directly correlated with end proximal delivery (r = 0.62, p < 0.0005), while end proximal delivery was inversely correlated to fractional proximal reabsorption (r = -0.71, p < 0.0005). In the subgroup with UalbV less than 30 mg 24 h-1, the direct correlation between GFR and end proximal delivery was also significant (r = 0.77, p < 0.05). In the group with nephropathy (UalbV > 300 mg 24 h-1), mean GFR and end proximal delivery were decreased and fractional proximal reabsorption was increased, but there was still a positive correlation between GFR and end proximal delivery (r = 0.75, p < 0.05) and an inverse correlation between end proximal delivery and fractional proximal reabsorption (r = -0.85, p < 0.05). It is concluded that in these groups of diabetic patients the end proximal delivery is increased while GFR is increased.(ABSTRACT TRUNCATED AT 250 WORDS)

Magnetic resonance imaging of the kidney in type 1 (insulin-dependent) diabetes mellitus

Reductions in the physiological cortical to medullary signal intensity ratio are found in magnetic resonance scans of the kidney in non-diabetic glomerular disease. Whether this abnormality can also characterise patients with Type 1 (insulin-dependent) diabetes mellitus and nephropathy is not known. We measured the cortical to medullary signal intensity ratio in magnetic resonance images of the kidney in 34 patients with Type 1 diabetes (ten with either clinical proteinuria or raised serum creatinine or both, nine with microalbuminuria, seven with normal urinary albumin excretion and long duration of diabetes and eight with Type 1 diabetes of short duration). The cortical to medullary signal intensity ratio showed a trend to cluster at lower values in the normoalbuminuric patients with normal serum creatinine rather than in the nine healthy individuals, independent of Type 1 diabetes duration (1.47 +/- 0.06 and 1.41 +/- 0.13 vs 1.63 +/- 0.16; five groups Scheffé F-test p = 0.05-0.1). Among the Type 1 diabetic patients, significant reductions in the cortical to medullary signal intensity ratio characterised overt nephropathy (1.19 +/- 0.15, p less than 0.05 vs all groups), but not microalbuminuria (1.47 +/- 0.13, p = NS), concomitantly with low glomerular filtration rate and elevated fractional excretion of uric acid, but independent of glycaemic control. The determinants of the renal cortical to medullary signal intensity ratio in Type 1 diabetes are uncertain. Reductions in the cortical to medullary signal intensity ratio may be a late finding in diabetic nephropathy, and parallel the accompanying impairment in kidney haemodynamics.(ABSTRACT TRUNCATED AT 250 WORDS)

Effects of hyperglycaemia on kidney function, atrial natriuretic factor and plasma renin in patients with insulin-dependent diabetes mellitus

In normoalbuminuric patients with insulin-dependent diabetes mellitus, plasma atrial natriuretic factor (ANF), cyclic GMP and active renin and the renal clearances of [99Tcm]-diethylenetriaminepentaacetic acid (DTPA) lithium and sodium were studied on a hyperglycaemia day and a euglycaemia day. Baseline euglycaemia was achieved by an overnight variable insulin infusion, which during study days was fixed at the rate necessary to maintain euglycaemia in the morning. After a baseline euglycaemic clearance period of 90 min, measurements were repeated in a new 90-min period beginning 150 min later. On the hyperglycaemia day i.v. infusion of 20% glucose was started at the end of the euglycaemic baseline period, increasing blood glucose (5.3 +/- 1.3 vs 12.1 +/- 1.2 mmol l-1, p less than 0.01). On the euglycaemia day blood glucose declined (5.1 +/- 1.0 vs 4.2 +/- 1.0 mmol l-1, p less than 0.02). Glomerular filtration rate (GFR) was unchanged by acute hyperglycaemia (127 +/- 16 vs 129 +/- 24 ml min-1, NS), but nearly normalized during maintained euglycaemia on the euglycaemia day (124 +/- 17 vs 105 +/- 16 ml min-1, p less than 0.01). When comparing the hyperglycaemic study period with the similarly timed period on the euglycaemia day, GFR was elevated by hyperglycaemia (129 +/- 24 vs 105 +/- 16 ml min-1, p less than 0.01), while the renal clearances of lithium and sodium were similar. Consequently, the calculated absolute proximal reabsorption rate of sodium and water was elevated during hyperglycaemia. Hyperglycaemia reduced the slight decline in plasma concentrations of ANF and cyclic GMP observed on the euglycaemia day. Active renin, glucagon and plasma osmolality were unchanged. In conclusion, marked changes in glomerular filtration rate are induced by changes in blood glucose concentration, but the effect is delayed and thus not directly related to renal tubular transport of glucose. Hyperglycaemia does not affect renal clearances of lithium and sodium, while proximal tubular reabsorption is markedly stimulated. These changes are not related to changes in ANF, renin, glucagon or plasma osmolality.

Effect of insulin on renal sodium handling in hyperinsulinaemic type 2 (non-insulin-dependent) diabetic patients with peripheral insulin resistance

The sodium retaining effect of insulin was studied in ten Type 2 (non-insulin-dependent) diabetic patients (mean age 56 (43-73) years, mean body mass index 29.5 (24.2-33.7) kg/m2) and eight age-matched control subjects (mean age 57 (43-68) years, mean body mass index 23.4 (20.8-26.6) kg/m2). The renal clearances of 99mTc-DTPA, lithium, sodium and potassium were measured over a basal period of 90 min. Then insulin was infused at a rate of 40 mU.min-1.m-2. After an equilibration period of 90 min, the clearance measurements were repeated during a new 90 min period. Blood glucose was clamped at the basal level (diabetic patients: 9.9 +/- 3.5, control subjects: 5.3 +/- 0.5 mmol/l) by a variable glucose infusion. Basal plasma insulin concentration was elevated in the diabetic patients (0.12 +/- 0.05 vs 0.05 +/- 0.02 pmol/ml, p less than 0.01). Insulin infusion resulted in comparable absolute increments in plasma insulin concentrations in the diabetic group and in the control group (0.44 +/- 0.13 vs 0.36 +/- 0.07 pmol/ml, NS). The metabolic clearance rate of glucose during the last 30 min of insulin infusion was lower in the diabetic patients (155 +/- 62 vs 320 +/- 69 ml.min-1.m-2, p less than 0.01), reflecting peripheral insulin resistance. The decline in sodium clearance during insulin infusion was similar in diabetic subjects (1.8 +/- 1.1 vs 0.7 +/- 0.4 ml.min-1.1.73 m-2, p less than 0.01) and in control subjects (1.7 +/- 0.3 vs 0.8 +/- 0.3 ml.min-1.1.73 m-2, p less than 0.01).(ABSTRACT TRUNCATED AT 250 WORDS)