Impact of recent antecedent hypoglycemia on hypoglycemic cognitive dysfunction in nondiabetic humans

To test the hypothesis that glycemic thresholds for hypoglycemic cognitive dysfunction, like those for neuroendocrine responses to and symptoms of hypoglycemia, shift to lower plasma glucose concentrations after recent antecedent hypoglycemia, 16 healthy young adult subjects (7 women and 9 men) were studied on two separate occasions in random sequence, once with hyperinsulinemic hypoglycemia (2.6 +/- 0.1 mmol/l, 47 +/- 1 mg/dl) and once with otherwise identical hyperinsulinemic euglycemia (4.8 +/- 0.1 mmol/l, 86 +/- 5 mg/dl) between 1430 and 1630. Neuroendocrine, symptomatic, and cognitive responses to hyperinsulinemic stepped hypoglycemic (4.7, 4.2, 3.6, 3.0, 2.8, 2.5, and 2.2 mmol/l; 85, 75, 65, 55, 50, 45, and 40 mg/dl) clamps were quantitated the following morning on both occasions. Cognitive function tests included measures of information processing (Serial Addition), attention (Stroop Arrow Word), pattern recognition and memory (Delayed Non-Match to Sample), and declarative memory (Paragraph Recall). As expected, plasma glucagon (P = 0.0094), epinephrine (P = 0.0063), and pancreatic polypeptide (P = 0.0046) responses to stepped hypoglycemia were reduced significantly, and symptomatic responses tended to be reduced after afternoon hypoglycemia. Performance on the cognitive function tests deteriorated (P < 0.0001) during stepped hypoglycemic clamps, but there were no significant overall effects of antecedent hypoglycemia on hypoglycemic cognitive dysfunction. Although deterioration was reduced (P < 0.05) from the 2.8 mmol/l (50 mg/dl) to the 2.5 mmol/l (45 mg/dl) steps on the Serial Addition and Delayed Non-Match to Sample tasks after afternoon hypoglycemia, comparable differences were not found on the Stroop Arrow Word or Paragraph Recall tasks. Thus, glycemic thresholds for hypoglycemic cognitive dysfunction, unlike those for neuroendocrine responses to and symptoms of hypoglycemia, do not seem to shift to substantially lower plasma glucose concentrations after recent antecedent hypoglycemia in nondiabetic humans.