Faculty Bibliography

OBJECTIVE:To investigate the current rate of attrition in general surgery residency, assess the risk factors, and identify prevention strategies. DESIGN:A literature review of the PubMed and MEDLINE databases, from January 1, 1980 to February 1, 2016, for relevant articles. The calculated attrition rate and the statistically significant influencing factors were the main measures and outcomes. SELECTION:All English language articles that described attrition from a general surgery residency were included. Articles that performed an assessment of attrition rates, academic performance, reasons for resident loss, and demographics were identified and data from these studies were collected. Random-effect meta-analysis and meta-regression based on a generalized mixed-effects model was performed. RESULTS:A total of 26 studies were included. Reported attrition rates ranged from 2% to 30% over the course of residency training. Random-effect meta-analysis is indicative of a yearly attrition rate of 2.4% (95% CI: 1.3%-3.5%) and a cumulative 5-year attrition rate of 12.9% (95% CI: 7.9%-17.8%). Most of them leave residency during their first 2 years, and the rate significantly decreases with increasing postgraduate year (p < 0.0001). The Accreditation Council for Graduate Medical Education mandated 80-hour week is associated with a higher rate, though not significantly (3.2% [95% CI: 1.3%-5.1%] vs. 2.2% [0.9%-3.5%], p = 0.37). Pooled analysis demonstrates no statistically significant difference in the rate of attrition between males and females (2.1% [95% CI: 1.1%-3%] vs. 2.9% [95% CI: 1.6%-4.1%], p = 0.73). Most remain in graduate medical education and pursue residency training in other specialties. CONCLUSION:Attrition in general surgery most commonly occurs within the first 2 years of training and, in contrast to previous findings, is not related to female sex. Restrictions on work hours seem to have increased the rate, whereas remediation practices can prevent it. Training programs should direct efforts towards attrition-prevention strategies.

Importance/UNASSIGNED:Previous studies of resident attrition have variably included preliminary residents and likely overestimated categorical resident attrition. Whether program director attitudes affect attrition has been unclear. Objectives/UNASSIGNED:To determine whether program director attitudes are associated with resident attrition and to measure the categorical resident attrition rate. Design, Setting, and Participants/UNASSIGNED:This multicenter study surveyed 21 US program directors in general surgery about their opinions regarding resident education and attrition. Data on total resident complement, demographic information, and annual attrition were collected from the program directors for the study period of July 1, 2010, to June 30, 2015. The general surgery programs were chosen on the basis of their geographic location, previous collaboration with some coauthors, prior work in surgical education and research, or a program director willing to participate. Only categorical surgical residents were included in the study; thus, program directors were specifically instructed to exclude any preliminary residents in their responses. Main Outcomes and Measures/UNASSIGNED:Five-year attrition rates (2010-2011 to 2014-2015 academic years) as well as first-time pass rates on the General Surgery Qualifying Examination and General Surgery Certifying Examination of the American Board of Surgery (ABS) were collected. High- and low-attrition programs were compared. Results/UNASSIGNED:The 21 programs represented different geographic locations and 12 university-based, 3 university-affiliated, and 6 independent program types. Programs had a median (interquartile range [IQR]) number of 30 (20-48) categorical residents, and few of those residents were women (median [IQR], 12 [5-17]). Overall, 85 of 966 residents (8.8%) left training during the study period: 15 (17.6%) left after postgraduate year 1, 34 (40.0%) after postgraduate year 2, and 36 (42.4%) after postgraduate year 3 or later. Forty-four residents (51.8%) left general surgery for another surgical discipline, 21 (24.7%) transferred to a different surgery program, and 18 (21.2%) exited graduate medical education altogether. Each program had an annual attrition rate ranging from 0.73% to 6.0% (median [IQR], 2.5% [1.5%-3.4%]). Low-attrition programs were more likely than high-attrition programs to use resident remediation (21.0% vs 6.8%; P < .001). Median (IQR) Qualifying Examination pass rates (93% [90%-98%] vs 92% [86%-100%]; P = .92) and Certifying Examination pass rates (83% [68%-84%] vs 81% [71%-86%]; P = .47) were similar. Program directors at high-attrition programs were more likely than their counterparts at low-attrition programs to agree with this statement: "I feel that it is my responsibility as a program director to redirect residents who should not be surgeons." Conclusions and Relevance/UNASSIGNED:The overall 5-year attrition rate of 8.8% was significantly lower than previously reported. Program directors at low-attrition programs were more likely to use resident remediation. Variations in attrition may be explained by program director attitudes, although larger studies are needed to further define program factors affecting attrition.

Necrotizing soft tissue infections (NSTIs) are aggressive infections requiring prompt diagnosis and extensive surgical debridement. Traditionally, patients undergo mandatory re-exploration to ensure adequacy of source control. The purpose of this study is to determine if re-exploration in the operating room is mandatory for all patients with NSTIs. An eight-year retrospective analysis of adult patients with NSTIs was performed comparing two groups: mandatory operative re-exploration versus operative re-exploration based on clinical examination findings. Outcomes measured included mortality, number of debridements, and length of stay (LOS). Twenty-two per cent of patients underwent a mandatory re-exploration. These patients were older, had a higher incidence of diabetes, and a longer duration of symptoms. There were no significant differences between groups with regard to the physical examination, severity of sepsis, time to repeat debridements, or in-hospital mortality, whereas LOS and the total number of debridements were increased in mandatory re-exploration. Bacteremia and septic shock were predictive of the need for further debridement in patients in the operative re-exploration based on clinical examination findings group. Mandatory re-exploration after initial debridement may not be necessary in all patients with NSTIs. Instead, bedside wound checks may be a safe strategy to determine the need for further operative debridement.

BACKGROUND:The benefit of intraoperative cholangiography (IOC) is controversial in patients with gallstone pancreatitis whose bilirubin levels are normalizing. IOC with subsequent endoscopic retrograde cholangiopancreatography may lengthen duration of surgery and length of stay, whereas failure to clear the common bile duct may result in recurrent pancreatitis. METHODS:We performed a 6-year retrospective cohort analysis of consecutive adult patients with mild gallstone pancreatitis undergoing same-admission cholecystectomy at 2 university-affiliated medical centers. Institution A routinely performed IOC, whereas institution B did not. The primary outcome was readmission within 30 days for recurrent pancreatitis. RESULTS:Of 520 patients evaluated, 246 (47%) were managed at institution A (routine IOC) and 274 (53%) were managed at institution B (restricted IOC). Patients at institution B had a shorter duration of surgery (1.0 vs 1.6 hours, P < .001), shorter length of stay (4 vs 5 days, P < .001), and fewer postoperative endoscopic retrograde cholangiopancreatographies performed (1.8% vs 21%, P < .001), without a difference in readmissions (1.5% vs 0%, P = .12). CONCLUSIONS:Routine IOC is not necessary in the setting of mild gallstone pancreatitis with normalizing bilirubin values.

Importance:Acute cholangitis (AC), particularly severe AC, has historically required urgent endoscopic decompression, although the timing of decompression is controversial. We previously identified 2 admission risk factors for adverse outcomes in AC: total bilirubin level greater than 10 mg/dL and white blood cell count greater than 20 000 cells/µL. Objectives:To validate previously identified prognostic factors in AC, evaluate the effect of timing of endoscopic retrograde cholangiopancreatography on clinical outcomes, and compare recent experience with AC vs an historical cohort. Design, Setting, and Participants:A retrospective analysis (2008-2015) of patients with AC (validation cohort, n = 196) was conducted at 2 academic medical centers to validate predictors of adverse outcome. Timing of endoscopic retrograde cholangiopancreatography and outcome were stratified by severity using the Tokyo Guidelines for acute cholangitis diagnosis. Outcomes for the validation cohort were compared with the derivation cohort (1995-2005; n = 114). Data analysis was conducted from July 1, 2015, to September 9, 2015. Main Outcomes and Measures:Death and a composite outcome of death or organ failure. Results:The median age of patients in the derivation cohort was 54 years (interquartile range, 40-65 years) and in the validation cohort was 59 years (45-67 years). Multivariate logistic regression analysis of the validation cohort confirmed white blood cell count of more than 20 000 cells/µL (odds ratio, 3.4; 95% CI, 1.2-9.5; P = .02) and total bilirubin level of more than 10 mg/dL (odds ratio, 5.4; 95% CI, 1.8-16.4; P = .003) as independent risk factors for poor outcomes. In the validation cohort, timing of endoscopic retrograde cholangiopancreatography was not significantly different between those with and without an adverse outcome, even when stratified by AC severity (moderate: median, 0.6 hours [interquartile range (IQR), 0.5-0.9] vs 1.7 hours [IQR, 0.7-18.0] and severe: median, 10.6 hours [IQR, 1.2-35.1] vs 25.5 hours [IQR, 15.5-58.5] for those with and without adverse events, respectively). Patients in the validation cohort had a shorter hospital length of stay (median, 7 days [IQR, 4-10 days] vs 9 days [IQR, 5-16 days]) and lower rate of intensive care unit admission (26% vs 82%), despite a higher rate of severe cholangitis (n = 131 [67%] vs n = 29 [25%]). There were no significant differences in the composite outcome between the validation and derivation cohorts (22 [18.6%] vs 44 [22.4%]; P = .47). Adjusted analysis demonstrated decreased mortality in the validation cohort (odds ratio, 0.3; 95% CI, 0.1-0.7; P = .01). Conclusions and Relevance:White blood cell count greater than 20 000 cells/µL and total bilirubin level greater than 10 mg/dL are independent prognostic factors for adverse outcomes in AC. Consideration should be given to include these criteria in the Tokyo Guidelines severity assessment. Timing of endoscopic retrograde cholangiopancreatography does not appear to affect clinical outcomes in these patients. Management of AC has improved with time, with an overall shorter hospital length of stay, lower rate of intensive care unit admission, and a decreased adjusted mortality, demonstrating improvements in care efficiency and delivery.

Consensus is lacking for ideal management of mild traumatic brain injury (mTBI) with intracranial hemorrhage (ICH). Patients are often monitored in the intensive care unit (ICU) without additional interventions. We sought to identify admission variables associated with a favorable outcome (ICU admission for 24 hours, no neurosurgical interventions, no complications or mortality) to divert these patients to a non-ICU setting in the future. We reviewed all patients with mTBI [Glasgow Coma Scale (GCS) = 13-15] and concomitant ICH between July 1, 2012, and June 30, 2015. Variables collected included demographics, vital signs, neurologic examination, imaging results, ICU course, mortality, and disposition. Of 201 patients, 78 (39%) had a favorable outcome. On univariate analysis, these patients were younger, more often had an isolated subarachnoid hemorrhage, and were more likely to have a GCS of 15 at admission. On multivariate regression analysis, after controlling for admission blood pressure, time to CT scan, and Marshall Score, age <55, GCS of 15 on arrival to the ICU, and isolated subarachnoid hemorrhage remained independent predictors of a favorable outcome. Patients meeting these criteria after mTBI with ICH likely do not require ICU-level care.