Permeability of blood-brain and blood-nerve barriers in experimental diabetes mellitus in the anaesthetized rat.

Diabetes was induced with streptozotocin in rats weighing about 160 g. These were maintained with age-matched controls for up to 14 months, blood glucose being periodically monitored. Half the diabetic and control rats received the aldose reductase inhibitor, Ponalrestat, in their diet. At 3 weeks, 6-7 months and 13-14 months, the vascular permeability in regions of brain, and in optic and sciatic nerves, were measured by maintaining radiotracers in the bloodstream--125I-albumin (100 min), [14C]sucrose (60 min) and 131I-albumin (5 min)--followed by tissue sampling and counting at termination. 131I-albumin estimated residual intravascular plasma. Diabetes of up to 13-14 weeks caused no measurable increase in the sucrose permeability of microvessels in eight different brain regions, in optic or in sciatic nerve. At 3 weeks of diabetes, sucrose permeability in all brain regions and in optic nerve was reduced relative to that in controls. Extravascular albumin entry into different regions of brain and optic nerve was insignificant and insensitive to diabetes, except in the hypothalamus and optic nerves where it was raised with increasing duration of diabetes. In sciatic nerve, extravascular albumin distribution was markedly increased by diabetes, but sucrose permeability was not demonstrably affected. At the level used in the diet, Ponalrestat reduced the sorbitol content of diabetic sciatic nerve but did not protect again the increased permeability to albumin.