Bradley R. Davis

Complications of Colorectal Anastomoses: Leaks, Strictures, and Bleeding

Intestinal anastomosis is an essential part of surgical practice, and with it comes the inherent risk of complications including leaks, strictures, and bleeding, which result in significant morbidity and occasional mortality. Understanding the myriad of risk factors and the strength of the data helps guide a surgeon as to the safety of undertaking an operation in which a primary anastomosis is to be considered. This article reviews the risk factors, management, and outcomes associated with anastomotic complications.

Risk factors and consequences of anastomotic leak after colectomy: a national analysis.

BACKGROUND: Previous research has identified a number of patient and operative factors associated with anastomotic leak after colectomy; however, a study that examines these factors on a national level with direct coding for anastomotic leak is lacking. OBJECTIVE: The purpose of this work was to identify risk factors associated with anastomotic leak on a national level and quantify the additional morbidity and mortality experienced by these patients. DESIGN: We performed a retrospective analysis of patients who underwent segmental colectomy with anastomosis from the 2012 American College of Surgeons National Surgical Quality Improvement Program colectomy procedure-targeted database. Anastomotic leak was defined as minor leak requiring percutaneous intervention or major leak requiring laparotomy. Multivariate logistic regression was used to determine predictors of anastomotic leak and its impact on postoperative outcomes. SETTINGS: This study was conducted at a tertiary university department. PATIENTS: This study includes 13,684 patients who underwent segmental colectomy with anastomosis at American College of Surgeons National Surgical Quality Improvement Program–affiliated hospitals in 2012. MAIN OUTCOME MEASURES: The primary outcome studied was anastomotic leak. RESULTS: The overall leak rate was 3.8%. Male sex, steroid use, smoking, open approach, operative time, and preoperative chemotherapy were associated with increased anastomotic leaks and diverting ileostomy with decreased incidence of leaks on multivariate analysis. Increased length of stay (13 vs 5 days; p < 0.001) and increased 30-day mortality (6.8% vs 1.6%; p < 0.001) were also seen in patients who experienced leaks. These patients also experienced increased readmission rates (43.5% vs 8.3%; p < 0.001) and were 37 times more likely to require reoperation as a complication of their primary procedure (p < 0.001). LIMITATIONS: The main limitations of this study include its retrospective nature and the limited 30-day outcomes recorded in the American College of Surgeons National Surgical Quality Improvement Program database. CONCLUSIONS: This study identified patient and operative risk factors for anastomotic leak on a national scale. It also demonstrates that these patients have increased morbidity and 30-day mortality rates, experience multiple readmissions to the hospital, and have a higher likelihood of requiring further operative intervention.

Readmission for dehydration or renal failure after ileostomy creation.

BACKGROUND: Ileostomy creation is a commonly performed operation in colorectal surgery; however, many patients develop complications such as dehydration postoperatively. Dehydration is often severe enough to warrant hospital readmission and may result in renal failure. The true incidence of this complication has not been well described. OBJECTIVE: The aim of this study was to identify the rate of hospital readmission secondary to dehydration or renal failure within 30 days of ileostomy creation. DESIGN: Retrospective review of all patients undergoing ileostomy creation from 2007 to 2011 in a single colorectal practice of 4 surgeons was performed. Charts were reviewed to identify patients readmitted for dehydration or renal failure within 30 days of operation. Data were then analyzed to identify predictors of readmission, dehydration, and renal failure. Subset analysis compared patients readmitted with simple dehydration versus patients with renal failure. PATIENTS: Two hundred one patients undergoing colorectal operations that included ileostomy creation within a 4-year period at a single institution for a variety of indications were included. MAIN OUTCOME MEASURES: The primary outcome measured was readmission for dehydration or renal failure. RESULTS: We observed a 17% 30-day readmission rate for dehydration or renal failure following ileostomy creation. Age greater than 50 was identified as an independent predictor of readmission with renal failure, whereas IPAA was predictive of readmission for simple dehydration, but not renal failure. Patients admitted with renal failure had significantly longer hospital stays and median hospital charges after readmission in comparison with patients admitted with simple dehydration. LIMITATIONS: This study was limited by its retrospective nature and its limited sample size. CONCLUSION: Hospital readmission due to dehydration or renal failure following ileostomy creation is common, with age >50 being the strongest predictor for renal failure. Appropriate strategies to decrease dehydration and renal failure following ileostomy creation need to be investigated.

Robotic-Assisted Versus Laparoscopic Colectomy: Cost and Clinical Outcomes

Background and Objectives: Laparoscopic colectomies, with and without robotic assistance, are performed to treat both benign and malignant colonic disease. This study compared clinical and economic outcomes for laparoscopic colectomy procedures with and without robotic assistance. Methods: Patients aged ≥18 years having primary inpatient laparoscopic colectomy procedures (cecectomy, right hemicolectomy, left hemicolectomy, and sigmoidectomy) identified by International Classification of Diseases, Ninth Edition procedure codes performed between 2009 and the second quarter of 2011 from the Premier Hospital Database were studied. Patients were matched to a control cohort using propensity scores for disease, comorbidities, and hospital characteristics and were matched 1:1 for specific colectomy procedure. The outcomes of interest were hospital cost of laparoscopic robotic-assisted colectomy compared with traditional laparoscopic colectomy, surgery time, adverse events, and length of stay. Results: Of 25 758 laparoscopic colectomies identified, 98% were performed without robotic assistance and 2% were performed with robotic assistance. After matching, 1066 patients remained, 533 in each group. Lengths of stay were not significantly different between the matched cohorts, nor were rates of major, minor, and/or surgical complications. Inpatient procedures with robotic assistance were significantly more costly than those without robotic assistance (

Factors associated with 30-day readmission after restorative proctocolectomy with IPAA: a national study.

17 445 vs

Gearing up for milestones in surgery: Will simulation play a role?

15 448, P = .001). Operative times were significantly longer for robotic-assisted procedures (4.37 hours vs 3.34 hours, P < .001). Conclusion: Segmental colectomies can be performed safely by either laparoscopic or robotic-assisted methods. Increased per-case hospital costs for robotic-assisted procedures and prolonged operative times suggest that further investigation is warranted when considering robotic technology for routine laparoscopic colectomies.

The effect of surgical approach on short-term oncologic outcomes in rectal cancer surgery

BACKGROUND:Hospital readmission has been identified by many payers as a surrogate for surgical quality. The 30-day readmission rate and factors associated with hospital readmission after restorative proctocolectomy with IPAA have not been well studied. OBJECTIVE:The purpose of this work was to identify the rate of and factors associated with hospital readmission within 30 days of restorative proctocolectomy with IPAA. DESIGN:A retrospective review of patients undergoing IPAA from 2009 to 2012 in the University HealthSystem Consortium database was performed. Hospitals were stratified into quartiles according to the number of cases performed annually. SETTING:This study was conducted using a national database of university hospitals. PATIENTS:A total of 4952 patients within the 4-year study period were included in the analysis. MAIN OUTCOME MEASURES:The primary outcome measured was readmission within 30 days of discharge. RESULTS:The 30-day readmission rate was 22.8% overall, although high-volume centers performed significantly better than low-volume centers (high vs low volume: 19.7% vs 28.2%; p < 0.001). When controlling for confounding variables, multivariate analysis identified female sex (OR, 1.191; p = 0.02), government-based (vs private) insurance (OR, 1.364; p < 0.001), and higher preoperative severity of illness (OR, 1.491; p = 0.001) to be associated with readmission. In addition, a significant volume-dependent relationship on 30-day readmission was identified, wherein undergoing operation at the higher-volume hospitals was protective for predicting readmission. Hierarchical regression modeling indicated that 31% of the variation in readmission rates among individual hospitals was accounted for by hospital volume. LIMITATIONS:This study was limited by its retrospective nature and limited postoperative complication data. CONCLUSIONS:The national 30-day readmission after IPAA creation was 22.8%, at least double that of other colorectal procedures. This high rate of readmission was mitigated by centers performing the highest volume of cases. Avoidance of referral to centers performing very few of these procedures annually may improve perioperative outcomes and reduce associated morbidity.

How Residents Learn Predicts Success in Surgical Residency

BACKGROUND The Consortium of American College of Surgeons-Accredited Education Institutes was created to promote patient safety through the use of simulation, develop new education and technologies, identify best practices, and encourage research and collaboration. METHODS During the 7th Annual Meeting of the Consortium, leaders from a variety of specialties discussed how simulation is playing a role in the assessment of resident performance within the context of the Milestones of the Accreditation Council for Graduate Medical Education as part of the Next Accreditation System. CONCLUSION This report presents experiences from several viewpoints and supports the utility of simulation for this purpose.

The impact of pregnancy on surgical Crohn disease: an analysis of the Nationwide Inpatient Sample

BACKGROUND Although evidence to support the use of laparoscopic and robotic approaches for the treatment of rectal cancer is limited, these approaches are being adopted broadly. We sought to investigate national practice patterns and compare short-term oncologic outcomes of different approaches for rectal cancer resections. METHODS The 2010 National Cancer Database was queried for operative cases of rectal cancer. Approach was classified as open, laparoscopic, or robotic. Patient, tumor, and hospital characteristics and surgical margin status were evaluated. Propensity score matching was used to compare outcomes across approaches. RESULTS We identified 8,712 patients. Laparoscopic and robotic approaches were more common in privately insured and wealthier patients at high-volume centers (P < .001). Open approaches were used for tumors with higher histologic grade and pathologic stage (P < .001). A minimally invasive approach was associated with fewer positive margins and shorter hospital stays. After propensity score matching, the laparoscopic approach was associated with a 2.0% lesser (P = .01) and robotic surgery with a 3.8% lesser (P = .004) incidence of positive margins compared with open surgery. CONCLUSION An open approach is often used in rectal cancers with higher pathologic stages. Matched patient analysis suggests minimally invasive approaches are associated with improved R0 resections.

A systematic approach to developing a global surgery elective.

BACKGROUND Predictors of success in surgical residency have been poorly understood. Previous studies have related prior performance to future success without consideration of personal attributes that help an individual succeed. Surgical educators should consider how residents learn to gain insight into early identification of residents at risk of failing to complete their surgical training. METHODS We examined our 14-year database of surgical resident learning-style assessments, Accreditation Council for Graduate Medical Education operative log data of graduating residents from 1999 to 2012, first time pass rates on the American Board of Surgery Qualifying and Certifying examinations, and departmental records to identify those residents who did not complete their surgery training at our institution. Statistical analysis was performed using the chi-square test, Wilcoxon rank-sum, and regression analysis with significance set at p < 0.05. RESULTS We analyzed 441 learning-style assessments from 130 residents. Surgical residents are predominantly action-based learners, with converging (219, 49.7%) and accommodating (112, 25.4%) being the principal learning styles. Assimilating (66, 15%) and diverging (44, 10%) learning styles, where an individual learns by observation, were less common. Regression analysis comparing learning style with case volume revealed that residents who are action-based learners completed more cases at graduation (p < 0.05 for each). Additionally, surgical residents who transferred to a nonsurgical residency or nonphysician field were more likely to learn by observation (p = 0.0467). CONCLUSIONS Surgical residents are predominantly action-based learners. However, a subset of surgical residents learn primarily by observation. These residents are at risk for a less robust operative experience and not completing surgical training. Learning-style analysis may be utilized by surgical educators to identify the potential at-risk residents in general surgery.