Importance of Insulin in Subjective, Cognitive, and Hormonal Responses to Hypoglycemia in Patients With IDDM DAVID KERR, MANSUR REZA, NICOLA SMITH, AND BRIAN A. LEATHERDALE
Not all episodes of hypoglycemia are recognized as such by diabetic patients, suggesting that it is possible for them to adapt to a low blood glucose level, although the mechanism involved is not known. The aim of this study was to examine whether insulin has an effect, independent of blood glucose, on the subjective, cognitive, and hormonal responses to hypoglycemia. Nine patients with insulindependent diabetes mellitus (IDDM) participated in three hyperinsulinemic glucose-clamp studies. After 60 min at 4.5 mM, blood glucose was randomized to be 1) maintained at 4.5 mM for 240 min, 2) lowered to 2.8 mM for 180 min followed by 60 min at 2 mM with an insulin infusion rate of 40 mil • rrr 2 • m \ and 3) fitted to the same protocol as 2 but with an infusion rate of 120 mil • rrr 2 • m i n 1 . Symptoms and awareness of hypoglycemia (100-mm visual analogue scales), cognitive function, and counterregulatory hormone levels were assessed every 30 min. There were no subjective or cognitive changes during the euglycemic study. Awareness and hypoglycemic symptoms (hunger, facial flushing, trembling, and sweating) were attenuated by the higher insulin infusion rate (P < 0.05 and P < 0.01, respectively). Cognition was significantly impaired after 60 min at 2.8 mM (P < 0.001) and deteriorated further when the blood glucose level was lowered to 2 mM (P < 0.01). Levels of cortisol (P < 0.01) and growth hormone (P < 0.05) but not epinephrine were suppressed by the higher insulin infusion rate. We conclude that the prevailing insulin level does influence the perception of symptomatic and hormonal responses to low blood glucose levels. Cognitive impairment is dependent on level and duration of hypoglycemia. Diabetes 40:1057-62,1991
From the Department of Diabetes and Endocrinology, Royal South Hants Hospital, Southampton, United Kingdom. Address correspondence to Dr. David Kerr, Section of Endocrinology, Department of Internal Medicine, School of Medicine, Fitkin I, PO Box 3333, New Haven, CT 06510-8056. Received for publication 21 November 1990 and accepted in revised form 8 March 1991.
DIABETES, VOL. 40, AUGUST 1991
P
atients with insulin-dependent diabetes mellitus (IDDM) learn to recognize specific physical sensations that tell them that their blood glucose level is low (1). Some patients do not recognize hypoglycemia, thus running the risk of sudden and severe neuroglycopenia (2), whereas others do not come to any harm (3). This suggests an ability of some individuals to adapt (i.e., have decreased perception without significant cognitive impairment) to hypoglycemia (4). The mechanism for adaptation is uncertain, but we wondered whether insulin concentrations may, independent of blood glucose levels, determine the physiological responses to a low blood glucose level. A hyperinsulinemic glucose-clamp technique (5) was used to maintain mild (2.8 mM) and moderate (2 mM) hypoglycemia, and the hormonal, cognitive, and symptomatic responses were compared at two different insulin infusion rates (with a 3-fold difference in circulating insulin levels). Patients with stable glycosylated hemoglobin levels were chosen because blood glucose thresholds for production of warning symptoms and release of counterregulatory hormones may be modified by changes in glycemic control (6).
RESEARCH DESIGN AND METHODS Nine patients with IDDM who had no history of hypoglycemic unawareness gave written consent for the study, which was approved by the hospital ethics committee. Patients were selected based on their willingness to participate and on the basis of stable glycemic control (HbA, 8-9% [normal range c
15 -
Blood glucose 2.8 mM
30
60
90
60
90
120 150 < 80 2 1 0 2 4 0 2 7 0
400 T
s S.
300 200 •
1 0 •
Not all episodes of hypoglycemia are recognized as such by diabetic patients, suggesting that it is possible for them to adapt to a low blood glucose level, although the mechanism involved is not known. The aim of this study was to examine whether insulin has an effect, independent of blood glucose, on the subjective, cognitive, and hormonal responses to hypoglycemia. Nine patients with insulindependent diabetes mellitus (IDDM) participated in three hyperinsulinemic glucose-clamp studies. After 60 min at 4.5 mM, blood glucose was randomized to be 1) maintained at 4.5 mM for 240 min, 2) lowered to 2.8 mM for 180 min followed by 60 min at 2 mM with an insulin infusion rate of 40 mil • rrr 2 • m \ and 3) fitted to the same protocol as 2 but with an infusion rate of 120 mil • rrr 2 • m i n 1 . Symptoms and awareness of hypoglycemia (100-mm visual analogue scales), cognitive function, and counterregulatory hormone levels were assessed every 30 min. There were no subjective or cognitive changes during the euglycemic study. Awareness and hypoglycemic symptoms (hunger, facial flushing, trembling, and sweating) were attenuated by the higher insulin infusion rate (P < 0.05 and P < 0.01, respectively). Cognition was significantly impaired after 60 min at 2.8 mM (P < 0.001) and deteriorated further when the blood glucose level was lowered to 2 mM (P < 0.01). Levels of cortisol (P < 0.01) and growth hormone (P < 0.05) but not epinephrine were suppressed by the higher insulin infusion rate. We conclude that the prevailing insulin level does influence the perception of symptomatic and hormonal responses to low blood glucose levels. Cognitive impairment is dependent on level and duration of hypoglycemia. Diabetes 40:1057-62,1991
From the Department of Diabetes and Endocrinology, Royal South Hants Hospital, Southampton, United Kingdom. Address correspondence to Dr. David Kerr, Section of Endocrinology, Department of Internal Medicine, School of Medicine, Fitkin I, PO Box 3333, New Haven, CT 06510-8056. Received for publication 21 November 1990 and accepted in revised form 8 March 1991.
DIABETES, VOL. 40, AUGUST 1991
P
atients with insulin-dependent diabetes mellitus (IDDM) learn to recognize specific physical sensations that tell them that their blood glucose level is low (1). Some patients do not recognize hypoglycemia, thus running the risk of sudden and severe neuroglycopenia (2), whereas others do not come to any harm (3). This suggests an ability of some individuals to adapt (i.e., have decreased perception without significant cognitive impairment) to hypoglycemia (4). The mechanism for adaptation is uncertain, but we wondered whether insulin concentrations may, independent of blood glucose levels, determine the physiological responses to a low blood glucose level. A hyperinsulinemic glucose-clamp technique (5) was used to maintain mild (2.8 mM) and moderate (2 mM) hypoglycemia, and the hormonal, cognitive, and symptomatic responses were compared at two different insulin infusion rates (with a 3-fold difference in circulating insulin levels). Patients with stable glycosylated hemoglobin levels were chosen because blood glucose thresholds for production of warning symptoms and release of counterregulatory hormones may be modified by changes in glycemic control (6).
RESEARCH DESIGN AND METHODS Nine patients with IDDM who had no history of hypoglycemic unawareness gave written consent for the study, which was approved by the hospital ethics committee. Patients were selected based on their willingness to participate and on the basis of stable glycemic control (HbA, 8-9% [normal range c
15 -
Blood glucose 2.8 mM
30
60
90
60
90
120 150 < 80 2 1 0 2 4 0 2 7 0
400 T
s S.
300 200 •
1 0 •
