Resistance to symptomatic insulin reactions after fasting.
Level 4 - case-series / case-control
Small non-randomized before-and-after physiological study
PubMed 5056667 · doi:10.1172/JCI107095
What was done
Nine obese participants underwent standard insulin tolerance tests (0.1-0.2 U/kg) before and after a 2-month fast to evaluate whether ketone utilization prevents cerebral dysfunction during acute glucopenia. Measures included symptomatic reactions, plasma glucose, insulin disappearance, urinary catecholamine excretion, and plasma beta-hydroxybutyrate. In an additional five fasting subjects, arteriovenous concentration differences for beta-hydroxybutyrate and glucose across the brain and forearm were measured during insulin-induced hypoglycemia.
What was found
Prior to fasting, insulin tolerance tests caused symptomatic hypoglycemic reactions and increased urinary catecholamines from 61 to 113 mug/24 hr (P < 0.01). Following 2 months of fasting, weight-adjusted insulin administration produced identical glucose declines (reaching as low as 0.5 mmoles/liter or 9 mg/100 ml) without triggering symptomatic hypoglycemic reactions or significant catecholamine increases. During postfast tests, mean plasma beta-hydroxybutyrate fell from 8.02 to 6.69 mmoles/liter (P < 0.01). In five tested subjects, the cerebral arteriovenous difference for beta-hydroxybutyrate increased from 0.21 to 0.70 mmoles/liter, while the cerebral arteriovenous difference for glucose decreased from 0.24 to 0.07 mmoles/liter of plasma; forearm uptake showed no consistent trend.
Why it matters
This study provides physiological evidence in humans that prolonged fasting enables the brain to extract and utilize ketoacids in place of glucose, protecting against symptomatic neuroglycopenia during profound hypoglycemia.
Limits
The study tested an extreme 2-month fast in a very small sample of obese individuals (n = 14 total), limiting applicability to shorter fasts, milder ketosis, or non-obese populations. Cerebral function was assessed via overt clinical symptoms and catecholamine response rather than formal neurocognitive testing.
Cited by
- supports George Cahill fasted human subjects for multiple days and administered insulin to drop blood glucose to approximately 20 mg/dL while they remained conscious.