Profound, prolonged hypoglycemia can cause brain death. In studies of insulin-induced hypoglycemia in monkeys, 5–6 hours of blood glucose concentrations of less than 1.1 mmol/l (20 mg/dl) were required for the regular production of neurological damage (12
); the average blood glucose level was 0.7 mmol/l (13 mg/dl). Fortunately, hypoglycemia of that magnitude and duration occurs rarely in people with diabetes.
The mechanisms of the common, hypoglycemia-induced functional brain failure and of the rare, hypoglycemia-induced brain death that occurs at very low, and at least in primates prolonged, plasma glucose concentrations (Figure ) differ. The former is the result of brain fuel deprivation per se, but the latter is not. As summarized by Suh and colleagues in their study reported in this issue of the JCI
), a variety of mechanisms are thought to be involved in the pathogenesis of hypoglycemic neuronal death. These include glutamate release and activation of neuronal glutamate receptors, production of reactive oxygen species, neuronal zinc release, activation of poly(ADP-ribose) polymerase, and mitochondrial permeability transition.
In their current report, Suh and colleagues (13
) describe additional studies of the mechanisms of hypoglycemia-induced neuronal necrosis. Based on systematic cell culture and in vivo rodent studies of glucose deprivation followed by glucose provision, they provide evidence that hypoglycemic superoxide production and neuronal death are increased by NADPH oxidase activation during glucose reperfusion. These effects were reduced by an inhibitor of NADPH oxidase, deficiency of a subunit of the enzyme, and blockade of NADPH regeneration, among other findings. Notably, superoxide formation and neuronal death increased with increasing glucose concentrations during the posthypoglycemic reperfusion period. That finding is generally consistent with earlier findings by these investigators (14
) and by others (15
In order to reproducibly cause the study endpoints, including neuronal death, these studies (13
) were generally performed at glucose concentration extremes. In the cell culture studies, glucose deprivation conditions were established by the use of a medium containing no glucose, while conditions of glucose provision were established by adding glucose to the medium at 10.0 mmol/l (180 mg/dl), several-fold greater than normal brain extracellular fluid glucose concentrations. In the in vivo studies, blood glucose concentrations averaged 0.4 mmol/l (7 mg/dl), causing an isoelectric EEG, during hypoglycemia and approximately 7.5 mmol/l (135 mg/dl) during glucose reperfusion that was documented to cause detrimental effects. Superoxide production, and presumably neuronal death, occurred as a result of hypoglycemia, but these occurred to a greater extent with glucose reperfusion, less so when posthypoglycemic blood glucose concentrations were raised to the range of 1.0–2.0 mmol/l (18–36 mg/dl) than when they were raised to the range of 5.0–10.0 mmol/l (90–180 mg/dl). Studies involving less profound hypoglycemia were not reported.
The distinction between the common hypoglycemia-induced functional brain failure and the rare hypoglycemia-induced brain death drawn here is admittedly arbitrary. Plasma glucose concentrations of less than 1.0 mmol/l (18 mg/dl) occur occasionally in people with diabetes (9
), and dying brain cells, presumably neurons, have been reported following episodes of hypoglycemia at plasma glucose levels of 1.7–1.9 mmol/l (30–35 mg/dl) — but not following episodes of hypoglycemia at plasma glucose levels of 2.5 mmol/l (45 mg/dl) — in rats (16
). Thus, it could be reasoned that these categories are not binary and that there is a continuous spectrum with increasing risk of neuronal death at progressively lower plasma glucose concentrations. Nonetheless, seemingly complete recovery follows the vast majority of episodes of clinical hypoglycemia.
The appropriate clinical extrapolation of these data is not entirely clear. As the authors point out (13
), plasma glucose concentrations must be raised in hypoglycemic patients. In the common clinical setting of hypoglycemia-induced functional brain failure, plasma glucose levels should be raised into the physiological range promptly with the expectation that recovery of brain function will follow. At this point there is no clear evidence that posttreatment hyperglycemia is detrimental to recovery, but there is no reason to think it is beneficial in that setting. On the other hand, undertreatment will delay recovery. In the rare clinical setting of profound, prolonged hypoglycemia, where the risk of neuronal death is higher, the data suggest that plasma glucose levels should be raised cautiously with avoidance of hyperglycemia (13
). Nonetheless, it would seem reasonable to raise the plasma glucose level into the physiological range (e.g., >3.9 mmol/l [70 mg/dl]) promptly. Clearly, additional studies of this important issue are needed.