Hyperglycemia should be avoided during neurosurgery in order to decrease the risk of neurological injury. Dexamethasone has been associated with increased blood glucose during surgery. In this prospective, nonrandomized study, we documented the blood glucose concentration changes for 12 h in 34 nondiabetic patients undergoing craniotomy and compared patients who received intraoperative dexamethasone (10 mg IV on induction and 4 mg IV 6 h later), with or without preoperative dexamethasone, with patients who did not receive dexamethasone. Blood glucose concentrations increased from the preinduction value in all groups. Patients not taking dexamethasone before surgery, but who were given it intra- and postoperatively, had the largest peak blood glucose concentrations (11.0 +/- 2.0 mmol/L, mean +/- sd; P< 0.01) compared with patients who received no dexamethasone (7.8 +/- 2.1 mmol/L) or those who had been taking dexamethasone before surgery and continued it during surgery (8.5 +/- 1.2 mmol/L). The peak blood glucose concentrations in this group occurred 9 +/- 2 h after the induction of anesthesia. We recommend that the blood glucose concentration be monitored for at least 12 h in nondiabetic patients having neurosurgery who are newly administered dexamethasone.

No abstract

INTRODUCTION: Cardiovascular surgery is still associated with significant morbidity and mortality due to bleeding. Several case reports have suggested that a new agent, recombinant Factor VIIa (rVIIa), may reduce bleeding in patients failing conventional treatment (1; 2). The purpose of this study was to determine if adding rVIIa to the already thrombogenic environment of new vascular anastomoses could result in higher incidence of graft occlusion. METHODS: With Animal Care Committee approval, 19 rabbits were anesthesized with ketamine (10mg/kg), xylazine (2mg/kg), and 1-2% isoflurane in 100% oxygen. Through a midline neck incision, the right jugular and both carotid arteries were exposed. The animals were anticoagulated with heparin, and a 2-3 cm section of right jugular vein was then excised and grafted to the right carotid artery with two end to side anastomoses. The left carotid artery was ligated and re-anastomosed in an end to end fashion. Following protamine administration the grafts were inspected before skin closure to ensure adequate flow. Animals then received either placebo or 300ug/kg of rVIIa intravenously. An ultrasound was performed at 3 hours and 24 hours to assess graft flow, and the presence of occlusive clot. On sacrifice, the grafts were visually inspected for thrombus. The primary outcome was ultrasound evidence of no flow or presence of occlusive thrombus in the graft. Data was analyzed using ANOVA, chi-square or fisher's exact test where appropriate, with p<0.05 considered significant. RESULTS: Three animals were excluded for technical reasons. rVIIa treated animals had a significantly higher incidence of graft occlusion (vein 7/8 vs 1/8, p=0.01; artery 7/8 vs 2/8, p<0.05) and lower average vein graft flow (26.7 +/-15.34 vs 5.5 +/-13.47 ml/min, p<0.05). There was no significant difference between the two groups in graft diameter, physiological variables, hemodynamics or anticoagulation. DISCUSSION: This study suggests that high dose rVIIa (300ug/kg) leads to an increased incidence of fresh vascular graft thrombosis. It is still unknown if these results would be obtained with lower doses. Our findings may guide further research and clinical use of rVIIa