Diabetic Control and the Renin-Angiotensin System, Catecholamines, and Blood Pressure

Transcription

1 Diabetic Control and the Renin-Angiotensin System, Catecholamines, and Blood Pressure J. BARRY FERRISS, JAMES A. O'HARE, CECILY C. M. KELLEHER, PATRICK A. SULLIVAN, MARGARET M. COLE, HAZEL F. ROSS, AND DENIS J. O'SULLIVAN SUMMARY Diabetic ketoacldosis is usually associated with marked secondary hyperaldosteronism. Plasma levels of renin, angiotensin II, and aldosterone are markedly raised before treatment in most patients, with values falling rapidly toward normal as metabolic control is restored. In a few patients, mostly those with long-term complications of diabetes, plasma levels of renin, angiotensin II, and aldosterone before treatment remain within the normal range. In moderately hyperglycemic patients who have glycosuria but not ketonuria, plasma levels of all three substances are significantly higher than when control is improved. Occasionally, moderately hyperglycemic patients have mild secondary hyperaldosteronism. Improved metabolic control in such patients causes a rise in plasma volume and a rise in total exchangeable sodium, the latter to levels significantly above normal. Plasma catecholamine levels are markedly elevated in diabetic ketoacidosis, probably as a consequence of the ketoacidotic state. In nonketotic patients with moderate hyperglycemia, basal plasma norepinephrine levels are normal; catecholamine responses to exercise may be exaggerated, however. Epidemiological and animal studies suggest a relationship between blood pressure and blood glucose levels. There are few clinical studies of the effects of altering metabolic control of diabetes on blood pressure, and this is an important area for further study. (Hypertension 7 [Suppl II]: II , 1985) KEY WORDS angiotensin II aldosterone plasma renin activity plasma volume exchangeable sodium * norepinephrine * diabetes mellitus hypertension THE abnormalities that occur in diabetic ketoacidosis have been investigated in detail. Less attention has been given to disturbances associated with poorly controlled nonketotic diabetes, although this metabolic state is all too common among our diabetic patients. In particular, the long-term effect of poor metabolic control on blood pressure is poorly understood. In this review we briefly discuss abnormalities of the renin-angiotensin system and catecholamines in ketoacidosis. Disturbances that occur in nonketotic, poorly controlled diabetes mellitus are also examined; these include changes in the renin-angiotensin system, body sodium, and plasma volume with altering metabolic control. The findings emphasize the importance of assessing metabolic control when examining these variables in diabetic patients. Finally, the possible effect of poor metabolic control of diabetes on blood pressure is discussed. From the Department of Medicine and Department of Statistics (M.M. Cole), Regional Hospital and University College, Cork, Ireland. Address for reprints: Prof. J. Barry Ferriss, Department of Medicine, Regional Hospital, Cork, Ireland The Renin-Angiotensin System in Ketoacidosis Ten years ago Christlieb et al. 1 described marked secondary hyperaldosteronism in diabetic ketoacidosis. Plasma renin activity (PRA) was grossly elevated in 13 patients in ketoacidosis, levels falling rapidly toward normal as metabolic control was restored. Plasma aldosterone concentrations, measured in four patients, were also raised. Several groups confirmed these observations. Scott et al. 2 reported raised levels of PRA and aldosterone in seven of eight patients in ketoacidosis. Waldhausl et al. 3 also described marked elevations of renin and aldosterone in a small group of severely hyperglycemic and mostly ketoacidotic patients. We found high levels of plasma angiotensin II 4 and aldosterone 5 in most patients in ketoacidosis (Figure 1). Plasma angiotensin II concentrations were within the normal range in 5 of 14 patients, however, all of whom had long-term complications of diabetes such as retinopathy, neuropathy, and nephropathy. Even in these patients there was some decrease in plasma angiotensin II with improved metabolic control. High renin levels in ketoacidosis have been attributed to fluid volume depletion,' although other factors

2 CHANGING DIABETIC CONTROL/Ferriss et al KETOACIDOSIS PLASMA ANGtOTENSlNI PO/ml 2SO PtASMA. AIDOSTERONE nfl/loomi 200- P«0 0O2 P<0001 Before When Wort Whan Treatment f JI_LJ- JJ led Treatment Controlled Free With of long Term Complication* Lang Term Complication! FIGURE 1. Individual plasma concentrations of angiotensin II and aldostewne in 14 patients with ketoacidosis before treatment and when metabolic control was restored (mean interval 9 days). Mean blood glucose (± SEM) before treatment was 33.3 ± 4.2 mmolll, and when control was restored, meanfasting blood glucose was 11.5 ± 1.7 mmolll fp < 0.001). The stippled areas represent the respective normal ranges in supine subjects. may also be important. Among our patients, pretreatment plasma angiotensin II concentrations were not related to indirect indexes of dehydration but were inversely related to ph (r = -0.65, p < 0.01). It is probable that hydrogen ions do not directly stimulate renin secretion, 6 but ph may reflect the metabolic disturbance that stimulates renin more accurately than indexes of dehydration. 4 Activation of the renin-angiotensin system in ketoacidosis may be clinically important. Circulating angiotensin II may be important in maintaining blood pressure in sodium-depleted subjects, 7 while secondary hyperaldosteronism, which increases renal potassium wasting, may also protect patients from even more marked hyperkalemia than actually occurs.' In some circumstances, high angiotensin II concentrations may cause multifocal myocardial necrosis and cardiac arrest 8 ; it is not known if the grossly raised concentrations sometimes seen in ketoacidosis can cause myocardial damage. Plasma Catecholamines in Ketoacidosis Circulating catecholamine levels are also elevated in established ketoacidosis. Christensen 9 described consistently raised plasma norepinephrine values in untreated patients, levels being positively related to the severity of the metabolic disturbance. Plasma epinephrine concentrations were elevated in only some patients. Once metabolic control was restored, both norepinephrine and epinephrine values were similar to those of healthy controls. Waldhausl et al. 3 reported a 10-fold increase in mean plasma norepinephrine and a 20-fold increase in mean plasma epinephrine levels in poorly controlled diabetic patients, mostly in ketoacidosis. Plasma concentrations again fell rapidly toward normal as metabolic control was restored. The lipolytic and ketogenic action of catecholamines may further aggravate established ketoacidosis. A rise in plasma norepinephrine and epinephrine does not precede the development of ketoacidosis when insulin is withdrawn, however. 10 This suggests that elevated levels are sequelae of ketoacidosis rather than an initiating cause. The Renin-Angiotensin System in Poorly Controlled Nonketotic Diabetes Abnormalities of the renin-angiotensin system in poorly controlled nonketotic diabetes are less marked than in ketoacidosis. Christlieb et al. 1 were unable to demonstrate a significant change in PRA with improved metabolic control in nonketotic patients, while in the alloxan-diabetic rat, PRA was suppressed in proportion to the severity of diabetes."

3 H-60 DIABETES AND HYPERTENSION SUPPL II HYPERTENSION, VOL 7, No 6, NOV-DEC 1985 As illustrated in Figure 2, we found a significant fall in PRA, plasma angiotensin n, and plasma aldosterone with improved metabolic control of diabetes. 3 ' l2> 13 All patients were free of ketonuria, but most had glycosuria when metabolic control was poor. The changes were observed in types I and II diabetes and in patients with and without complications of diabetes. On the other hand, patients studied on two occasions without alteration of metabolic control failed to show a change in any of these variables (see Figure 2). During poor metabolic control, plasma concentrations of angiotensin II were more than two standard deviations above the mean of healthy controls in 7 (22%) of 32 patients, while plasma aldosterone concentrations were similarly elevated in 6 (20%) of 30 patients. This indicates that some patients with poorly controlled diabetes have mild secondary hyperaldosteronism, even in the absence of ketosis. 5 These findings illustrate the need to define the degree of metabolic control when studying the reninangiotensin system in diabetic patients. Plasma levels of angiotensin II close to or within the normal range may exert a pressor effect 14 and the higher plasma angiotensin II concentrations during poor metabolic control may contribute to higher blood pressure levels. Plasma Catecholamines in Poorly Controlled Nonketotic Diabetes Altered metabolic control of diabetes does not cause marked changes in basal circulating catecholamine concentrations in the absence of ketosis. Improvement of metabolic control by continuous subcutaneous insulin infusion did not alter basal concentrations of epinephrine or norepinephrine. 131 l6 Catecholamine responses to exercise were exaggerated during conventional control, however, returning to normal during insulin infusion. 15 We measured plasma norepinephrine in 11 nonketotic diabetic patients free of long-term complications, both when metabolic control of diabetes was poor and again when control improved. Plasma norepinephrine was determined using the solvent extraction system of Smedes et al. 17 and analysis by high-performance liquid chromatography (Waters Associates, Milford, MA, USA), with electrochemical detection (Bio-Analytical Systems, West Layfayette, IN, USA) (detection limit < 75 pg/ml; intraassay and interassay coefficients of variation < 10%). Samples were taken when patients were supine after overnight rest and again when upright for 10 minutes. Supine and upright plasma norepinephrine concen- Control Improved Control Unchanged FIGURE 2. Mean values ( ± SEM) of plasma renin activity (PRA), plasma angiotensin II, and aldosterone in diabetic patients free of ketonuria studied when fasting and supine overnight. Values during poor metabolic control (open circles) and when control improved (closed circles) are shown on the left; the mean of five to six blood glucose measurements over 24 hours were 15.9 ± 0.5 and 7.9 ± 0.4 mmolll respectively (p < 0.001); mean urinary volumes were 1589 ± 118 and 1436 ± 112 ml respectively (NS) on the two study occasions. The mean interval between studies was 6 weeks. Values when metabolic control was unchanged on two occasions (closed triangles) are shown on the right. Means of five to six blood glucose measurements over 24 hours were 12.9 ± 1.4andl2.6 ± 1.7mmoULrespectively (n = 10, NS). The mean interval between studies was 6 weeks. PR A (ng/ml/h) PLASMA ANGIOTENSIN H (pg/ml) PLASMA ALDOSTERONE (ng/ 100ml) n= 1 5 P < n = 6 NS 30-1O J n = 32 P < 0.02 n=10 NS 20 n = 30 P < n = 6 NS

4 CHANGING DIABETIC CONTROL!Ferriss et al trations were similar and the postural response was unchanged when metabolic control improved (Figure 3). Values were similar in patients receiving and not receiving insulin on each occasion. The findings were in keeping with observations that basal plasma catecholamines are unchanged by poorly controlled nonketotic diabetes. Plasma Volume and Exchangeable Sodium in Poorly Controlled Nonketotic Diabetes It has been suggested that osmotic volume expansion associated with hyperglycemia may contribute to the increased prevalence of hypertension in younger diabetic patients. 18 Blood volume was increased in both alloxan-diabetic" and streptozotocin-diabetic 19 rats. In humans, plasma volume was higher in normotensive diabetic patients than in matched controls in one study, 20 but others found a similar plasma volume in normotensive diabetics and controls Christlieb et al. 1 did not note a change in blood volume when metabolic control improved in nonketotic patients. We studied 12 patients with nonketotic diabetes when metabolic control was poor and again when control improved. 12 On each occasion, blood pressure was measured under standardized conditions after overnight rest, using a London School of Hygiene Sphygmomanometer (Cinetronics Ltd., Mildenhall, England), and the average of five readings taken over 20 minutes was calculated. Plasma volume was measured in 11 of these patients and exchangeable sodium in 9. The mean interval between studies was 3 weeks. With improved metabolic control of diabetes, there was a significant fall in both systolic and diastolic blood pressures, while plasma volume and exchangeable sodium each rose significantly (Figure 4). Exchangeable PLASMA NOREPINEPHRINE ( pg/ml) 600 sodium was similar to that of healthy control subjects when metabolic control was poor, but rose to levels significantly above normal when metabolic control improved. 12 In this small series there was an inverse relationship between the change in exchangeable sodium and the change in both systolic (r = 0.78, p < 0.01) and diastolic (r = 0.66, p < 0.05) blood pressures. An elevated exchangeable sodium has been described in metabolically controlled diabetic patients, 21 " 24 and the increased values when metabolic control was improved are in keeping with these observations. The rise in exchangeable sodium with improved control is in keeping with short-term balance studies that showed sodium retention." It may reflect the reversal of sodium loss associated with even a mild osmotic diuresis during poor control and the direct sodium-retaining effect of insulin on the kidneys. 26 In view of the negative association between changes in blood pressure and in exchangeable sodium, sodium retention could also result from an antinatriuresis associated with a fall in blood pressure. The rise in plasma volume with improved control in humans contrasts with findings in diabetic rats. Our findings do not support the hypothesis that plasma volume is osmotically expanded in humans when metabolic control of diabetes is poor. The Effect on Blood Pressure of Altering Metabolic Control Epidemiological studies indicate an independent association between blood pressure and blood glucose. A positive relationship has been shown between both systolic and diastolic pressures and blood glucose after a glucose load in middle-aged populations. This persisted when controlled for age, body weight, and heart rate M An independent relationship between blood P < P < Poor Control Improved Control FIGURE 3. Plasma norepinephrine (mean values ± SEM) in 11 diabetic patients when supine (open circles) and after 10 minutes standing (closed circles), when metabolic control was poor (left-hand pair) and when control improved (right-hand pair). The mean of five to six blood glucose measurements over 24 hours was 15.9 ± 0.8 mmol/l when control was poor and 8.1 ± 0.4 mmol/l when control improved (p < 0.001); mean 24-hour urinary volumes were 1984 ± 128 and 1765 ± 233 ml respectively (NS). The mean interval between studies was 3 weeks. All patients were free of clinical evidence of long-term complications of diabetes and none had ketonuria. There were no significant differences in supine or upright values when metabolic control was altered.

5 11-62 DIABETES AND HYPERTENSION SUPPL II HYPERTENSION, VOL 7, No 6, NOV-DEC 1985 SYSTOLIC BP (rnmhg) 130- DIASTOLIC BP (mmhg) 80 P < 0.0S Na E (mmol) PLASMA VOLUME (ml) 3400 Improved Control Improved Control Poor Control Poor Control FIGURE 4. Mean values (± SEM) of systolic and diastolic blood pressure (n = 12), exchangeable sodiun (n = 9), and plasma volume (n = 11), when metabolic control was poor (open circles) and when control improved (closed circles). The mean of five to six blood glucose measurements over 24 hours was 16.5 ±0.9 mmoltl when control was poor and 8.0 ± 0.5 mmolll when control improved (n = 12, p < 0.001); mean 24-hour urinary volumes were 1803 ± 153 and 1507 ± 232 ml respectively (NS). The mean interval between studies was 3 weeks. All patients were free of ketonuria. pressure and postprandial glucose was also noted in schoolchildren. 29 The association in these populations raises the possibility of a similar relationship in diabetic subjects. The metabolic control of diabetes may influence blood pressure levels in the rat. In streptozotocin-diabetic rats there was a dose-related blood pressure rise in normotensive animals at least, and hypertension may be reduced by insulin injections. 32 There are few clinical studies of the effects of altering metabolic control of diabetes on blood pressure. As mentioned, we found a small but significant fall in systolic and diastolic pressures with improved metabolic control (see Figure 4). Blood pressure often falls on repeated measurement, however, 33 and despite efforts to standardize measurements, familiarization may have contributed to the reductions observed. We would have liked to study patients in reverse, when control was good and subsequently when control deteriorated, but had ethical reservations about allowing deliberate worsening of metabolic control for a prolonged period. Gunderson 34 studied acute changes in young patients with insulin-dependent diabetes with altering metabolic control. Some were newly diagnosed as having diabetes, but in most, insulin was withdrawn under supervision. Mean arterial pressure was significantly higher when metabolic control was poor, compared with values when control was optimal. Blood pressure changes with altering control were not related to changes in blood glucose but were significantly related to changes in standard bicarbonate and free fatty acids. Some patients developed mild ketoacidosis, however. The influence of metabolic control of diabetes on blood pressure is an important area for further investigation. Because of blood pressure variation even under standardized conditions, it is difficult to demonstrate small changes. This may explain why alteration of metabolic control in small groups may not be associated with significant alterations in blood pressure, even if the degree of metabolic control does influence blood pressure levels. Because of blood pressure variability, we calculate that to detect an independent, unidirectional 4-mm change in systolic pressure between two study occasions, it would be necessary to investigate approximately 45 patients for a change at the 5% significance level and approximately 100 patients for a change at the 1% level. Hypertension is an important prognostic indicator for the diabetic patient. 18 Even small changes in blood

6 CHANGING DIABETIC CONTROL/Ferriss et al pressure may have long-term importance, and the effect of long-term poor metabolic control of diabetes on blood pressure needs fuller evaluation. If poor metabolic control causes even a small increase in blood pressure, this may be an additional explanation for the association between hypertension and diabetes. Hypertension is a quantitative disorder and any definition is arbitrary. 35 In a diabetic population, poor metabolic control may move some patients from just below to just above an arbitrary upper limit of normal blood pressure for the particular age and sex. This change would cause an increased prevalence of hypertension in a diabetic population at any given time unless all are in good metabolic control, an unlikely event in current diabetic practice. Acknowledgment We are grateful for the financial support of the Medical Research Council of Ireland. References 1. Christlieb AR, Assal J, Katsilambros N, Williams GH, Kozak GP, Suzuki T. Plasma renin activity and blood volume in uncontrolled diabetes: ketoacidosis, a state of secondary aldosteronism. Diabetes 1975 ;24: Scott RS, Espiner EA, Donald RA, Livesey JH. Hormonal responses during treatment of acute diabetic ketoacidosis with constant insulin infusions. Clin Endocrinol 1978;9: Waldhausl W, KleinbergerG, Korn A, Dudczak R, Bratusch- Marrain P, Nowontny P. Severe hyperglycaemia: effects of rehydration on endocrine derangements and blood glucose concentration. Diabetes 1979;28: Sullivan PA, Gonggrijp H, Crowley MJ, Ferriss JB, O'Sullivan DJ. Plasma angiotensin II concentrations in diabetic ketoacidosis and in hyperosmolar non-ketotic hyperglycaemia. Acta Diabetol Lat 1981 ;18: Quigley C, Sullivan PA, Gonggrijp H, Crowley MJ, Ferriss JB, O'Sullivan DJ. Hyperaldosteronism in ketoacidosis and in poorly controlled non-ketotic diabetes. Ir J Med Sci 1982; 151: Kisch ES, Dluhy RG, Williams GH. Regulation of renin release by calcium and ammonium ions in normal man. J Clin Endocrinol Metab I976;43: Gavras H, Ribeiro AB, Gavras I, Brunner HR. Reciprocal relation between renin dependency and sodium dependency in essential hypertension. N Engl J Med 1976;295: Gavras H, Kremer D, Brown JJ, et al. Angiotensin- and norepinephrine-induced myocardial lesions: experimental and clinical studies in rabbits and man. Am Heart J 1975; 89: Christensen NJ. Plasma norepinephrine and epinephrine in untreated diabetics, during fasting and after insulin administration. Diabetes 1974;23:l Alberti KGMM, Christensen NJ, Iversen J, Orskov H. Role of glucagon and other hormones in development of diabetic ketoacidosis. Lancet 1975; 1: ] Christlieb AR. Renin, angiotensin and norepinephrine in alloxan diabetes. Diabetes 1974;23: O'Hare JA, Ferriss JB, Twomey BM, Gonggrijp H, O'Sullivan DJ. Changes in blood pressure, body fluids, circulating angiotensin II and aldosterone, with improved diabetic control. Clin Sci 1982;63:4l5s-418s 13. Sullivan PA, Gonggrijp H, Crowley MJ, Ferriss JB, O'Sullivan DJ. Plasma angiotensin II and the control of diabetes mellitus. Clin Endocrinol 1980; 13: Chinn RH, Dusterdieck G. The response of blood pressure to infusion of angiotensin II: relation to plasma concentrations of renin and angiotensin II. Clin Sci 1972;42: Tamborlane WV, Sherwin RS, Koivisto V, Hendler R, Genel M, Felig P. Normalization of the growth hormone and catecholamine response to exercise in juvenile-onset diabetic subjects treated with a portable insulin infusion pump. Diabetes 1979;28: Hershcope R, Plotnick LP, Kaya K, et al. Short-term improvement in glycemic control utilizing continuous subcutaneous insulin infusion: the effect on 24-hour integrated concentrations of counterregulatory hormones and plasma lipids in insulin-dependent diabetes mellitus. J Clin Endocrinol Metab 1982;54: Smedes F, Kraak JC, Poppe H. Simple and fast solvent extraction system for selective and quantitative isolation of adrenaline, noradrenaline and dopamine from plasma and urine. J Chromatogr 1982;231: Christlieb AR, Warram JH, Krolewski AS, et al. Hypertension: the major risk factor in juvenile-onset insulin-dependent diabetics. Diabetes 1981;3O(suppl 2): Hostetter TH, Troy JL, Brenner BM. Glomerular hemodynamics in experimental diabetes mellitus. Kidney Int 1981;19: Brochner-Mortensen J. Glomerular filtration rate and extracellular fluid volumes during normoglycemia and moderate hyperglycemia in diabetics. Scand J Clin Lab Invest 1973; 32: De Chatel R, Weidmann P, Flammer J, et al. Sodium, renin, aldosterone, catecholamines, and blood pressure in diabetes mellitus. Kidney Int 1977;12: O'Hare JA, Ferriss JB, Twomey BM, Brady D, O'Sullivan DJ. Essential hypertension and hypertension in diabetic patients without nephropathy. J Hypertension 1983;l(suppl 2): Weidmann P, Beretta-Piccoli C, Keusch G, et al. Sodiumvolume factor, cardiovascular reactivity and hypotensive mechanism of diuretic therapy in mild hypertension associated with diabetes mellitus. Am J Med 1979;67: O'Hare JA, Ferriss JB, Brady D, Twomey B, O'Sullivan DJ. Exchangeable sodium and renin in hypertensive diabetic patients with and without nephropathy. Hypertension 1985; 7(suppl II). 11^ SaudekCD, Boulter PR, Knopp RH, Arky RA. Sodium retention accompanying insulin treatment of diabetes mellitus. Diabetes 1974;23:24O De Fronzo RA. The effect of insulin on renal sodium metabolism. Diabetologia 1981;21: Stamler J, Rhomberg P, Schoenberger JA, et al. Multivariate analysis of the relationship of seven variables to blood pressure. J Chron Dis 1975;28: Jarrett RJ, Keen H, McCartney M, et al. Glucose tolerance and blood pressure in two population samples: their relation to diabetes mellitus and hypertension. Int J Epidemiol 1978 ;7: Florey CduV, Uppal S, Lowy C. Relation between blood pressure, weight, and plasma sugar and serum insulin levels in school children aged 9-12 years in Westland, Holland. BrMed J 1976;1: Kawashima H, Igarashi T, Nakajima Y, Akiyama Y, Usuki K, Ohtake S. Chronic hypertension induced by streptozotocin in rats. Naunyn Schmiedebergs Arch Pharmacol 1978; 305: Somani P, Singh HP, Saini RK, Rabinovitch A. Streptozotocin-induced diabetes in the spontaneously hypertensive rat. Metabolism 1979;28:1O75-1O Sasaki S, Buiiag RD. Insulin reverses hypertension and hypothalamic depression in streptozotocin diabetic rats. Hypertension 1983;5: Dunne JF. Variation of blood pressure in untreated hypertensive out-patients. Lancet 1969;l: Gundersen HJG. Peripheral blood flow and metabolic control in juvenile diabetes. Diabetologia 1974;10: Pickering G. Hypertension: causes, consequences and management. Edinburgh, London: Churchill Livingstone, 1974: 32-36