Surgical practice: evidence or anecdote.

OBJECTIVES: Our objective is to highlight a few surgical practices that are not based on evidence but are still taught in surgical education, and to assess our experience with these practices. DESIGN: We identified 3 practices (clamping of nasogastric tubes before removal, bowel preparation before elective colon resection, and elective sigmoid colectomy following 2 bouts of diverticulitis), identified the data supporting each practice, and administered a survey to faculty and residents at our institution. SETTING: Wright State University Department of Surgery, Boonshoft School of Medicine, Dayton, Ohio. PARTICIPANTS: Twenty-one faculty and 35 residents responded to the survey. RESULTS: No studies were found relating to clamping nasogastric tubes before removal. Seven faculty (33%) and 11 residents (31%) used this practice. Two faculty (10%) and 0 residents felt this was an evidence-based practice. Faculty were more likely to have reviewed the evidence (85% vs 40%, p < 0.001). Level 2 evidence has shown bowel preparation did not improve outcomes relating to anastomotic leak, wound infection, or septic complications in elective colon resection. Twenty faculty (95%) and 34 residents (97%) used this practice. Faculty were more likely to believe this to be evidence-based (85% vs 49%, p = 0.01). There has been no level 1 or 2 evidence showing that sigmoid colectomy following 2 bouts of diverticulitis improves morbidity or mortality. Fourteen faculty (70%) and 26 residents (76%) reported using this practice. Twelve faculty (60%) and 21 residents (60%) felt this was evidence-based. CONCLUSIONS: Frequent use of surgical practices without evidence support can create a misperception that such practices are evidence-based. Faculty are more likely to change a practice after obtaining continuing medical education, suggesting that residents may need validation by faculty practice of evidence-based procedures before incorporation into their clinical care.

Predictors of successful implementation of preoperative briefings and postoperative debriefings after medical team training.

BACKGROUND: The purpose of this study was to examine which factors at a medical team training learning session predict future success in the implementation of preoperative briefings and postoperative debriefings at health care facilities. METHODS: A Likert score rating for physician involvement, leadership support, and composition of the implementation team was recorded for 64 VHA facilities at the time of a learning session by 3 medical team training educators. At a mean follow-up period of 8.2 months (standard error, .4 mo), a briefing score was established from quarterly semistructured interviews with the facility's implementation team. RESULTS: In a multivariable regression, leadership involvement at the time of the learning session was the best predictor of future briefing/debriefing success (R = .34, P = .03). CONCLUSIONS: Full implementation of the patient safety tool preoperative briefings and postoperative debriefings is dependent on facility leadership support.

Impact of splenic injury guidelines on hospital stay and charges in patients with isolated splenic injury.

BACKGROUND: The purpose of this study was to assess the impact of care guidelines for patients with isolated blunt splenic trauma on length of stay (LOS) and patient charges. METHODS: We conducted a review of the hospital trauma registry and identified patients admitted with blunt splenic injury from 2000 to 2007. Splenic injury guidelines were initiated in November 2004. Patients with other major injuries were excluded. Patients were grouped according to their American Association for the Surgery of Trauma (AAST) splenic injury grade, I-V. Hospital LOS, intensive care unit (ICU) LOS, and patient charges before and after the guidelines were compared. RESULTS: We identified 137 patients with isolated splenic injuries. Sixty-three patients were admitted before and 70 patients after implementation of the guidelines. ICU and hospital LOS were significantly decreased after the guidelines (ICU LOS, 1.35 days before, 0.80 after [P < .01]; and hospital LOS, 4.17 before, 3.27 after [P < .01]). When grouped by AAST grade, grade II injuries had a decrease in hospital LOS (4.5 before vs 2.29 after; P < .01) and ICU LOS (1.43 before vs 0.29 after; P < .01). Adjusted hospital charges showed no significant increase overall after the guideline implementation (mean hospital charges before $23,047 vs after, $24,116; P = .62). CONCLUSION: Implementing guidelines for the observation of blunt splenic injury decreased the overall hospital LOS and ICU LOS at our institution, but hospital charges remained the same. Trauma programs should institute splenic injury guidelines to reduce resources needed for the care of isolated splenic injuries.

Effect of the 30-hour work limit on resident experience and education.

PURPOSE: To evaluate the effect of the 30-hour restriction on resident operative participation and assess whether the 30-hour restriction can be extended in certain cases to enhance educational experience and continuity of care without being detrimental to the 80-hour limit. METHODS: In September 2006, we administered a 10-item Likert scale survey to 41 general surgery residents to assess their experience with the 30-hour work restriction. We also reviewed the operative reports from the busiest general surgery service in April 2003 and April 2005 to assess surgical participation before and after the 30-hour restriction. RESULTS: Twenty-three (56%) residents reported missed operations each month because of the 30-hour restriction. Thirty-four (83%) reported occasions where participating in an operation would require only an additional 1-4 hours. Thirty-six (88%) residents reported a better educational experience when operating on patients whom they had evaluated and a preference to operate on patients whom they had evaluated. The operative log review revealed that in April 2003, the resident assigned to the service participated in 47 out of 134 (35%) total operations and 11 out of 30 (37%) operations beginning after noon. In April 2005, the resident assigned to the service participated in 49 out of 109 (45%) total operations and 20 out of 45 (44%) of the operations beginning after noon. CONCLUSION: The difference in the amount of operations involving resident participation before and after the 30-hour restriction, including afternoon cases that would be most affected by the work restriction, was minimal. However, we identified occasions when the 30-hour work restriction could be extended to provide continuity of care and a better educational operative experience while maintaining weekly duty hours within the approved limit. Extensions beyond the 30 hours should be limited to providing unique and comprehensive experiences for residents where the additional time or episodes would not cause resident fatigue.