In this canine study, glucose homeostasis after clinical pancreas transplantation is complex, with the relative effect of systemic versus portal delivery of insulin remaining unresolved. Thirty-two pancreatectomized dogs received either systemic venous drainage (SVD) with bladder exocrine drainage (n = 16), or portal venous drainage (PVD) with gastric exocrine drainage (n = 16). Cyclosporine (CsA) based immunosuppression was commenced on day -7. The effect of immunosuppression was a significant increase in fasting blood glucose (FBGL) (P = 0.002), fasting insulin (P = 0.024), AUC for insulin (P = 0.009), and K values decreased (P = 0.009). FBGL and K values remained abnormal after transplantation with no significant difference seen between SVD and PVD. However, fasting insulin became significantly lower after PVD and AUC insulin fell in both groups. CsA levels fell in both groups after transplantation, mirroring the fall in AUC insulin, and implicating CsA as a major cause of peripheral resistance to insulin. In conclusion, PVD did not demonstrate a significant advantage over SVD in handling an intravenous glucose challenge. The need for pancreatectomy in large animals may make them an unsatisfactory experimental model to evaluate the glucoregulatory effects of pancreas allotransplantation.