Effects of various routes of caffeine administration on oral and intravenous glucose tolerance tests in dogs

Abstract

Oral glucose tolerance tests were performed four times on 15 dogs with an interval of at least three days between each test: glucose solution alone formed the test meal; caffeine sodium benzoate was mixed with the glucose solution and fed to the animal; glucose solution was administered orally and caffeine was given intravenously; the glucose solution was given orally and the caffeine was administered intramuscularly. The order of administration of the respective tests was randomized. Blood samples were taken at serial intervals and analyzed for blood glucose and serum immunoreactive insulin levels. Intravenous glucose tolerance tests were also run four times on 11 dogs where the regimen consisted of glucose solution alone, glucose intravenously and caffeine intravenously, glucose I.V. and caffeine orally, and glucose I.V. and caffeine I.M. Serial blood samples were drawn and likewise analyzed for glucose and immunoreactive insulin values. The oral glucose tolerance tests were markedly altered by administration of caffeine. Thirty, 60 and 90 minutes after the ingestion of caffeine blood glucose was significantly lower than the untreated group (p=<0.01). When caffeine was administered intramuscularly, similar lowering of the blood glucose levels were found for these time intervals. When the caffeine was administered intravenously, however, at 60 minutes the lowering of blood sugar just about reached statistical significance at the 5 per cent level while at 90 minutes the lowering was truly significant (p=<0.05). In the intravenous glucose tolerance tests, the differences were less dramatic. When caffeine was administered intravenously, no statistically significant differences between the blood sugar of the untreated and the caffeine-treated group were found. When caffeine was administered orally, the blood sugars were lower than those of the untreated animals only in the samples drawn 45 minutes after the ingestion of caffeine and the differences were only statistically significant at the 5 per cent level. When caffeine was administered intramuscularly, again the 45-minute sample was the only one to show values lower than those of the untreated group. Here, however, the differences were statistically significant at the 1 per cent level. No significant differences between the untreated and the caffeine-treated group were found at any time period for the serum immunoreactive insulin values during the oral or the intravenous glucose tolerance tests.