and of the nonselective (3-adrenergic antagonist propranolol (80 mg) both ... not safer than oral propranolol with respect to recovery from hypoglycemia in ...
o
ral Propranolol and Metoprolol Both Impair Glucose Recovery from Insuliri'induced Hypoglycemia in Insulin*dependent Diabetes Mellitus DENNIS A. POPP, M.D., THOMAS F. TSE, M.D., SURESH D. SHAH, M.S., WILLIAM E. CLUTTER, M.D., AND PHILIP E. CRYER, M.D.
To the extent that they have deficient glucagon secretory responses to plasma glucose decrements, as they commonly do, patients with insulin-dependent diabetes mellitus (IDDM) are dependent on epinephrine-mediated p-adrenergic mechanisms to promote recovery from hypoglycemia. Thus, they are at increased risk for prolonged hypoglycemia if treated with a nonselective p-adrenergic antagonist such as propranolol. If the hyperglycemic actions of epinephrine are mediated through (32-adrenergic mechanisms, therapeutic efficacy (e.g., for hypertension or ischemic heart disease) could be accomplished without increased risk of hypoglycemia by selective P^adrenergic blockade in such patients. However, oral administration of the relatively selective pi-adrenergic antagonist metoprolol (100 mg) and of the nonselective (3-adrenergic antagonist propranolol (80 mg) both impaired recovery from insulin-induced hypoglycemia in patients with IDDM. Thus, at a dose of 100 mg, oral metoprolol is not safer than oral propranolol with respect to recovery from hypoglycemia in patients with IDDM. DIABETES CARE 7: 243-247, MAY-JUNE 1984.
M
aintenance of the plasma glucose concentration is critical to survival. In addition to dissipation of insulin, increased glucagon secretion plays a primary role in promoting glucose recovery from insulin-induced hypoglycemia; epinephrine secretion is not critical normally, but its enhanced secretion compensates largely, and it becomes critical, when glucagon secretion is deficient. '~3 Other hormones, neural mechanisms, or hepatic autoregulation may be involved but need not be invoked and are not sufficiently potent to promote glucose recovery when both of the key counterregulatory hormones—glucagon and epinephrine—are deficient. Hypoglycemia is a fact of life for persons with insulindependent (type 1) diabetes mellitus (IDDM) who must use exogenous insulin to survive. To the extent that they have deficient glucagon secretory responses to plasma glucose decrements, as they commonly do, patients with IDDM are dependent on epinephrine to promote recovery from hypoglycemia.4'5 The importance of epinephrine to such patients is underscored by the fact that those with deficient epinephrine (as well as glucagon) secretion have a 25-fold increased risk of severe hypoglycemia during intensive treatment of their diabetes.6 The hyperglycemic effects of epinephrine are complex.
They include both direct and indirect (hormone-mediated) actions, involve both stimulation of glucose production and limitation of glucose utilization, and are mediated through both a- and (3-adrenergic receptors in humans.7"9 Since the a-adrenergic component is largely,78 but not entirely,9 the result of limitation of insulin secretion, it follows that the hyperglycemic actions of epinephrine are mediated primarily through p-adrenergic receptors in patients with IDDM.10 It is not known, in either nondiabetic persons or patients with IDDM, whether the fi-adrenergic component is mediated through P r or f$2-adrenergic receptors or both. Metoprolol, a relatively selective fVadrenergic antagonist,11'12 did not prevent an increase in plasma glucose following intravenous terbutaline, a 32-adrenergic agonist, whereas propranolol, a nonselective fi-adrenergic antagonist, did, suggesting (32-adrenergic mediation;13 however, metoprolol prevented a significant increase in plasma glucose during intravenous isoproterenol.13 Hepatic (3-adrenergic receptors have been reported to be of the (i2 subtype14 and adrenergic induction of hepatic tyrosine aminotransferase appears to be mediated through (32adrenergic receptors15 in rats. Given their dependence on epinephrine-mediated (3-adrenergic mechanisms to prevent and correct hypoglycemia, patients with IDDM are at increased risk for prolonged hy-
D1ABETES CARE, VOL. 7 NO. 3, MAY-JUNE 1984
243
HYPOGLYCEMIA IN IDDM/DENNIS A. POPP AND ASSOCIATES
i i
c 2 E 100 - i ^^^ ^
X
.
^illilftlili
0*A 90 0) _D
il
«•»—•»
1 1 1 LXJ
i i
80
. I I In /^iU*PLB "
f
-
)L
I—
70 1—
60
j/*PRP
TftRTiiri*iVAVtTrrpj-*• MTL Iili fi UAi i / II
: *
LU
X
50
i
i
i
i
0
80
160
240
i
E E
LU Q^
140
SYSTOLIC
120
oO 100
oO LU
or a. Q
O
60
CO
40
g
^PLB
an ou
.
PRP DIASTOLIC
0
80
160 240
TIME (minutes) poglycemia if treated with nonselective f$-adrenergic antagonists such as propranolol. If the hyperglycemic actions of epinephrine are mediated through f$2'adrenergic receptors, therapeutic efficacy (e.g., for hypertension or ischemic heart disease) could be accomplished without increased risk of hypoglycemia by selective fVadrenergic blockade in such patients. Accordingly, we compared the effects of the relatively selective 3i-adrenergic antagonist metoprolol and of the nonselective 3-adrenergic antagonist propranolol with those of placebo on recovery from insulin-induced hypoglycemia in patients with IDDM. METHODS
FIG. I. Mean (± SE) heart rates and systolic and diastolic blood pressures before and after rapid intravenous injection of regular insulin (0.075 VI kg) after ingestion of placebo (PLB, solid circles); propranolol (PRP, open circles), 80 mg; or metoprolol (MTL, open squares), 100 mg, at —60 min in six patients with IDDM. The shaded areas are I SE around the mean of data from seven nondiabetic controls.*
pranolol and metoprolol (Figure 1), indicating similar degrees of (3i-adrenergic blockade. A small diastolic pressor response appeared to be associated with hypoglycemia following both propranolol and metoprolol (Figure 1); this could be interpreted to indicate some degree of 32-adrenergic antagonism following metoprolol as well as propranolol. As shown in Figure 2, glucose recovery from hypoglycemia was equally impaired following ingestion of metoprolol and of propranolol compared with that after placebo. Mean (±SE) plasma glucose concentrations before insulin injection were 75 ± 5 mg/dl after placebo, 83 ± 6 mg/dl after propranolol, and 92 ± 7 mg/dl after metoprolol. Mean post-insulin-injection nadir values did not differ (40 ± 8 mg/dl after placebo, 36 ± 4 mg/dl after propranolol, and 41 ± 9 mg/dl after metoprolol). Values following propranolol and metoprolol were 118 ± 22 and 125 ± 19 mg/dl, respectively, at the final sampling point, 240 min after insulin injection, significantly (P < 0.02 by t test) lower than the corresponding value of 181 ± 30 mg/dl following placebo. Plasma free insulin and glucagon concentrations were comparable under all three study conditions (Figure 2). The glucagon responses to hypoglycemia were markedly subnormal as expected. Cortisol and growth hormone responses tended to be greater following propranolol and metoprolol compared with those following placebo (Figure 2), undoubtedly the result of lower glucose concentrations.
Six patients (five women, one man) with IDDM, ranging in age from 25 to 45 yr, consented to participate in this study, which was approved by the Washington University Human Studies Committee and performed on the Washington University General Clinical Research Center. After an overnight fast, with maintenance of euglycemia with a variable intravenous infusion of regular insulin,4 patients ingested propranolol (Inderal, Ayerst Laboratories, New York, New York), 80 mg; metoprolol (Lopressor, CIBAGEIGY, Ardsley, New York), 100 mg; or placebo at 0700 h on each of three separate occasions. At 0800 h regular insulin (0.075 U/kg) was administered by rapid intravenous injection. Observations were made at 10-min intervals from — 30 through 240 min in relation to insulin injection. This DISCUSSION was a single-blind study, with the sequence of drug adminhese data confirm and extend our previous finding istration varied and a washout period of at least 1 wk between of impaired glucose recovery from insulin-induced drug administrations. hypoglycemia during high-grade p-adrenergic Plasma glucose, free insulin, glucagon, cortisol, and growth blockade produced by intravenous infusion of prohormone were measured with methods referenced previpranolol in patients with IDDM and deficient glucagon seously.4 Heart rates and blood pressures were recorded. cretory responses to hypoglycemia.4 In the present study, a single, oral dose of 80 mg of propranolol also impaired reRESULTS covery from hypoglycemia in similar patients. Thus, to the Heart rates were suppressed, and tachycardic responses to extent that they have deficient glucagon secretory responses, hypoglycemia were blunted, comparably following both pro- patients with IDDM are dependent on epinephrine to pro-
T
244
DIABETES CARE, VOL. 7 NO. 3, MAY-JUNE 1984
i
i
i
200 160
•I
^PRP
df^MTL '
CO
I
O
1'
80
-
40 i
0
0
80
i
1 .f11il-PLB -
120
8
i
i
80
i
i
i
160 240
160 240
240 O " ° TIME (minutes)
160 240
80
160 240
FIG. 2. Mean ( ± SE) plasma glucose, free insulin, glucagon, cortisol, and growth hormone concentrations before and after rapid inti-avenous injection of regular insulin (0.075 U/kg) after ingestion of placebo (PLB, solid circles); propranolol (PRP, open circles), 80 mg; or metoprolol (MTL, open squares), 100 mg, at — 60 min in six patients with IDDM. The shaded areas are I SE around the mean of data from seven nondiabetic controls.4
HYPOGLYCEMIA IN IDDM/DENNIS A. POPP AND ASSOCIATES
mote recovery from hypoglycemia. Furthermore, this effect of epinephrine can be impaired by the nonselective (3-adrenergic antagonist propranolol at a dose commonly used to treat hypertension or ischemic heart disease in patients with IDDM. If the impaired recovery from hypoglycemia in IDDM produced by propranolol were the result of 32-adrenergic blockade, it would be preferable to use a selective (3!-adrenergic antagonist in these patients, since the therapeutic effects of propranolol are the result of P,-adrenergic blockade. Metoprolol is a relatively selective (3,-adrenergic antagonist"12 with pharmacokinetic characteristics similar, but not identical, to those of the nonselective antagonist propranolol.12 Lager et al.16 compared the effects of oral metoprolol (50 mg) and propranolol (40 mg) with those of placebo on insulin-induced hypoglycemia in patients with IDDM. Although differences in the absolute blood glucose concentrations were not apparent, the calculated rate of glucose recovery was significantly reduced following propranolol but not following metoprolol. Using the same experimental design these investigators17 reported subsequently that 40 mg of propranolol delayed glucose recovery from hypoglycemia by 43% whereas 50 mg of metoprolol delayed glucose recovery by 22%, although the latter was not statistically significant, in seven patients with IDDM. DeFeo et al.18 have recently reported that intravenous propranolol but not metoprolol impairs glucose recovery from hypoglycemia in patients with IDDM. Since the most commonly required antihypertensive dose of metoprolol is 100 mg twice daily,11 we tested a single oral dose of 100 mg against a dose of propranolol with comparable 3i-adrenergic antagonist activity and against placebo. Under these conditions both metoprolol and propranolol impaired glucose recovery from hypoglycemia to a similar degree in a group of glucagon-deficient patients with IDDM. Taken at face value, this suggests that the hyperglycemic effects of epinephrine are mediated at least in part through f^-adrenergic mechanisms, a possibility not inconsistent with some earlier data.13 On the other hand, we cannot exclude the possibility that the 100-mg dose of metoprolol produced some degree of fJ2-adrenergic blockade, although doses in excess of 100 mg are said to be required to do so.11 In either event, at a dose of 100 mg metoprolol is not safer than propranolol with respect to recovery from hypoglycemia in patients with IDDM. The authors gratefully acknowledge the technical assistance of Denise Nachowiak, Joy Brothers, and Bakula Trivedi, the secretarial assistance of Susan Hofherr, and the assistance of the nursing staff of the Washington University General Clinical Research Center. This work was supported by U.S.P.H.S. grants AM27085, AM20579, and RR00036 and by the CIBA-GEIGY Corporation.
ACKNOWLEDGMENTS:
From the Metabolism Divison of the Department of Medicine, and the General Clinical Research Center and Diabetes Research
246
and Training Center, Washington University School of Medicine (T.F.T., S.D.S., W.E.C., P.E.C.), and the Section of Endocrinology, Department of Medicine, St. Louis University School of Medicine (D.A.P), St. Louis, Missouri. Address reprint requests to Philip E. Cryer, M.D., Metabolism Division, Washington University School of Medicine, 660 South Euclid, St. Louis, Missouri 63110. REFERENCES 1 Gerich, J., Davis, J., Lorenzi, M., Rizza, R., Bohannon, N., Karam, J., Lewis, S., Kaplan, S., Schultz, T., and Cryer, P.: Hormonal mechanisms of recovery from insulin-induced hypoglycemia in man. Am. J. Physiol. 1979; 236:E380-85. 2 Rizza, R. A., Cryer, P. E., and Gerich, J. E.: Role of glucagon, epinephrine and growth hormone in human glucose counterregulation: effects of somatostatin and adrenergic blockade on plasma glucose recovery and glucose flux rates following insulin induced hypoglycemia. J. Clin. Invest. 1979; 64:62-71. 3 Cryer, P. E. Glucose counterregulation in man. Diabetes 1981; 30:261-64. 4 Popp, D. A., Shah, S. D., and Cryer, P. E.: The role of epinephrine mediated (3-adrenergic mechanisms in hypoglycemic glucose counterregulation and posthypoglycemic hyperglycemia in insulin dependent diabetes mellitus. J. Clin. Invest. 1982; 69:315— 26. 5 Bolli, G., DeFeo, P., Compagnucci, P., Cartechini, M. G., Angeletti, G., Santeusanio, E, and Brunetti, P.: Important role of adrenergic mechanisms in acute glucose counterregulation following insulin-induced hypoglycemia in type I diabetes: evidence for an effect mediated by beta-adrenoreceptors. Diabetes 1982; 31:641-
47.
6 White, N. H., Skor, D., Cryer, P. E., Bier, D. M., Levandoski, L., and Santiago, J. V.: Identification of type I diabetic patients at increased risk for hypoglycemia during intensive therapy. N. Engl. J. Med. 1983; 308:485-91. 7 Rizza, R. A., Haymond, M. W., Miles, J. M., Verdonk, C. H., Cryer, P. E., and Gerich, J. E.: Effect of a-adrenergic stimulation and its blockade on glucose turnover in man. Am. J. Physiol. 1980; 238:E467-72. 8 Rizza, R. A., Cryer, P. E., Haymond, M. W., and Gerich, J. E.: Adrenergic mechanisms for the effect of epinephrine on glucose production and clearance in man. J. Clin. Invest. 1980; 65:682-89. 9 Rosen, S. G., Clutter, W. E., Shah, S. D., Miller, J. P., Bier, D. M., and Cryer, P. E.: Direct, alpha-adrenergic stimulation of hepatic glucose production in human subjects. Am. J. Physiol. 1983; 245:E616-26. 10 Serusclat, P., Rosen, S. G., Smith, E. B., Shah, S. D., Clutter, W. E., and Cryer, P. E.: Mononuclear leukocyte p2-adrenergic receptors and adenylate cyclase sensitivity in insulin-dependent diabetes mellitus. Diabetes 1983; 32:825-29. 11 Koch-Weser, J.: Metoprolol. N. Engl. J. Med. 1979; 301:698703. 12 Fishman, W. H.: (3-Adrenergic antagonists: new drugs and new indications. N. Engl. J. Med. 1981; 305:500-506. 13 William-Olsson, T , Fellenius, E., Bjorntorp, P., and Smith, U.: Differences in metabolic responses to p-adrenergic stimulation after propranolol or metoprolol administration. Acta Med. Scand. 1979; 205:201-206. 14 Minneman, K. P., Hedberg, A., and Molinoff, P. B.: Comparison of beta adrenergic receptor subtypes in mammalian tissues. J. Pharmacol. Exp. Ther. 1979; 211:502-508.
DIABETES CARE, VOL. 7 NO. 3, MAY-JUNE 1984
HYPOGLYCEV.IA IN IDDM/DENN1S A. POPP AND ASSOCIATES
15
Andersson, S. M.: (5-Adrenergic induction of tyrosine aminotransferase in organ culture of fetal rat and fetal human liver. Endocrinology 1983; 112:466-69. 16 Lager, I., Blohme, G., and Smith, U.: Effect of cardioselective and non-selective (3-blockade on the hypoglycemic: response in insulin-dependent diabetics. Lancet 1979; 1:458-62 17 Lager, I., Jagenburg, R., von Schenck, H., and Smith, U.:
Effect of beta-blockade on hormone release during hypoglycaemia in insulin-dependent diabetics. Acta Endocrinol. 1980; 95:364— 71. 18 DeFeo, P., Bolli, G., Perriello, G., DeCosmo, S., Compagnucci, P., Calafiore, R., and Brunetti, P.: The catecholamine contribution to hypoglycemia in IDDM. Diabetes 1983; 32 (Suppl. l):50A.
Y
DIABETES CARE, VOL. 7 NO. 3, MAY-JUNE 1984
247
