Shared quality data are associated with increased protamine use and reduced bleeding complications after carotid endarterectomy in the Vascular Study Group of New England.

OBJECTIVE: The goal of this study was to evaluate whether protamine usage after carotid endarterectomy (CEA) increased within the Vascular Study Group of New England (VSGNE) in response to studies indicating that protamine reduces bleeding complications associated with CEA without increasing the risk of stroke. METHODS: We reviewed 10,059 CEAs, excluding concomitant coronary bypass, performed within the VSGNE from January 2003 to July 2012. Protamine use and reoperation for bleeding were evaluated monthly using statistical process control. Twelve centers and 77 surgeons entering the VSGNE between 2003 and 2008 were classified as original participants, and 14 centers and 60 surgeons joining after May 2009 were considered new. Protamine use for surgeons was categorized as rare (<10%), selective (10%-80%), or routine (>80%). Outcome measures were in-hospital reoperation for bleeding, postoperative myocardial infarction (POMI), and stroke or death. RESULTS: Two significant increases occurred in protamine use for all VSGNE centers over time. From 2003 to 2007, the protamine rate remained stable at 43%. In 2008, protamine usage increased to 52% (P < .01), coincident with new centers joining the VSGNE. Protamine usage then increased to 62% in 2010 (P < .01), shortly after the presentations of the data showing a benefit of protamine. This effect was due to 10 surgeons in the original VSGNE centers who increased their usage of protamine: six surgeons from rare use to selective use and four surgeons to routine use. Reoperation for bleeding was reduced by 0.84% (relative risk reduction, 57.2%) in patients who received protamine (0.6% vs 1.44%; P < .001). There were no differences in POMI (1.1% vs 1.09%) or stroke or death (1.1% vs 1.03%) between protamine treated and untreated patients, respectively. Reoperation for bleeding was decreased for surgeons who used protamine routinely (0.5%; P < .001) compared with selective (1.4%) and rare users (1.5%) of protamine. There were no differences in POMI (0.9%, 1.2%, 1.1%; P = .720) and stroke or death rates (1.0%, 1.2%, 1.0%; P = .656) for rare, selective, and routine users of protamine. CONCLUSIONS: Protamine use increased over time by VSGNE surgeons, most significantly after the presentations of VSGNE-derived data showing the benefit of protamine, and was associated with a decrease in reoperation for bleeding. Improvements in processes of care and outcomes can be achieved in regional quality groups by sharing safety and efficacy data.

Use of a postoperative insulin protocol decreases wound infection in diabetics undergoing lower extremity bypass.

OBJECTIVE: Strict glucose control in patients undergoing coronary bypass grafting has been shown to decrease infectious complications, arrhythmias, and mortality. Our objective was to determine if strict glucose control reduced morbidity after lower extremity bypass (LEB). METHODS: A prospective pilot study at a single institution within the Vascular Study Group of New England was conducted from January 2009 to December 2010. Patients with diabetes and without undergoing LEB were placed on an intravenous (IV) insulin infusion for 3 days after surgery with titration of blood glucose from 80 to 150 mg/dL. The IV insulin study group (n = 104) was compared to a historic control group (n = 189) that received standard insulin treatment from the preceding 3 years. The Fisher exact test, t-tests, Wilcoxon rank-sum tests, χ(2), and logistic regression analyses were used to compare in-hospital morbidity. Stratified analyses were conducted to determine if findings differed based on the presence or absence of diabetes. RESULTS: There was no difference in postoperative complications between the two groups with regard to graft infection, myocardial infarction, dysrhythmia, primary patency at discharge, or mortality. Patients in the IV insulin group had significantly fewer in-hospital wound infections (4% vs 11%; odds ratio [OR], 0.32; 95% confidence interval [CI], 0.11-0.96; P = .047). This association strengthened after adjusting for potentially confounding baseline differences in gender, body mass index, and smoking status (adjusted OR, 0.22; 95% CI, 0.05-0.84; P = .03). When stratified by presence of diabetes, wound infections were decreased in the IV insulin group (0/44 [0%] vs 9/90 [10%]; P = .03). In patients without diabetes treated with IV insulin, there was no significant difference in wound infections (7% vs 12%; P = .42). CONCLUSIONS: Strict glucose control with a postoperative insulin infusion protocol significantly decreased the incidence of postoperative in-hospital wound infection in the diabetic population. These previously unreported findings from this single-institution prospective study warrant further investigation.