High incidence of potentially preventable emergency department visits after major elective colorectal surgery.

INTRODUCTION: Emergency department (ED) visits after surgery represent a significant cost burden on the healthcare system. Furthermore, many ED visits are related to issues of healthcare delivery services and may be avoidable. Few studies have assessed the reasons for ED visits after colorectal surgery. The main objectives of this study were to: (1) identify the reasons why patients presented to the ED within 30 postoperative days and (2) determine if these visits were potentially preventable. METHODS: A retrospective chart review was conducted on elective major colorectal surgery cases performed in a single center between 01/2017 and 07/2019. Data collected included demographics, medical history, intraoperative details, postoperative complications, ED visits within 30 postoperative days, and readmissions. Each ED visit was assessed by two reviewers and graded on a scale adapted from the New York University ED algorithm. The gradings were: (1) non-emergent, (2) emergent but treatable in an ambulatory setting, (3) emergent/ED-care required but preventable if timely outpatient care was available, and (4) emergent/ED-care required and non-preventable. Grades 1-3 were deemed potentially preventable. Logistic regression identified independent predictors of potentially preventable visits. RESULTS: Six hundred and twenty five patients were included in the final analysis. 110 (17.6%) patients presented to the ED within 30 days. The most common cause of ED visit were ileus/small bowel obstruction (SBO) (16.4%), superficial wound infection (15.5%), genitourinary issues (10.9%), and non-infectious gastrointestinal issues (nausea, malnutrition, diarrhea, high output stomas) (10.9%). After review, 51.8% of visits were considered potentially preventable (Grade 1-3). The most common causes of preventable ED visits were superficial wound infection (24.6%), non-infectious gastrointestinal issues (19.3%), and minor bleeding (14.0%). Creation of a new stoma was the only independent risk factor for potentially preventable ED visits (OR 2.14, 95%CI 1.03-4.47). CONCLUSION: Approximately half of ED visits within 30 days of discharge were potentially preventable. These findings indicate a need to improve access to outpatient care to reduce preventable ED visits after elective colorectal surgery.

Patients' preferences for sphincter preservation versus abdominoperineal resection for low rectal cancer.

BACKGROUND: Surgery for low rectal cancer can be associated with severe bowel dysfunction and impaired quality of life. It is important to determine how patients value the trade-off between anorectal dysfunction versus abdominoperineal resection. Therefore, the objective was to determine patients' preferences for treatment for low rectal cancer. METHODS: Ambulatory patients without colorectal cancer at a single high-volume academic colorectal referral center from September 2019 to March 2020 were included. Patients with prior stoma or malignancy were excluded. Participants were presented with a hypothetic scenario describing a low rectal cancer. A threshold task identified preferences for functional and oncologic outcomes for sphincter preservation versus abdominoperineal resection. RESULTS: A total of 123 patients were recruited. Patients preferred abdominoperineal resection over sphincter preservation if there were more than a mean of 6.7 (standard deviation 4.0) daily bowel movements, 1.9 (standard deviation 2.6) daily episodes of stool incontinence, and 6.5 (standard deviation 3.2) gas incontinence. Abdominoperineal resection was preferred over sphincter preservation in 38% if daily activities were altered owing to fecal urgency. Patients were willing to accept a 10% (interquartile range, 5-25) absolute increase in risk of margin involvement with sphincter preservation to avoid abdominoperineal resection. Abdominoperineal resection was the preferred option overall for 18% of patients. CONCLUSION: An important proportion of patients would prefer abdominoperineal resection over sphincter preservation owing to the impairments in anorectal function associated with sphincter preservation. The decision to perform sphincter preservation or abdominoperineal resection should consider how the patients' value functional outcomes with a low anastomosis.

The Effect of Preoperative Anemia and Perioperative Transfusion on Surgical Outcomes After Gastrectomy for Gastric Cancer.

BACKGROUND: The aim of this study is to examine the interaction between preoperative anemia and perioperative transfusions with postoperative morbidity and mortality among patients undergoing gastrectomy for cancer. MATERIALS AND METHODS: The American College of Surgeons National Surgical Quality Improvement Program database was queried from 2005 to 2016. Restricted cubic splines modeled the nonlinear relationship between preoperative hematocrit (Hct) and 30-day overall morbidity, sepsis, and mortality. Preoperative Hct was categorized based on cut points for the three models. Multiple regression modeling examined the interactive effect of preoperative anemia and postoperative transfusion on surgical outcomes. RESULTS: Among 9936 included patients, complication incidence was 38.9% (sepsis 12.7%; mortality 6.0%). Preoperative Hct cut points were identified at 29 and 42. Hct <29 was associated with higher risk of morbidity (OR 2.47, 95%CI 2.10-2.93). Postoperative transfusion was associated with lower risk of morbidity for Hct <29 (OR 0.56, 95%CI 0.43-0.73) but increased risk between 29 and 42 (OR 1.59, 95%CI 1.21-2.08). Similar relationships were found for sepsis and mortality. CONCLUSIONS: Preoperative Hct <29 is associated with an increased risk of surgical complications after gastrectomy for cancer and perioperative transfusions appear to be beneficial for Hct <29 only. There may be a role for better optimization of red cell mass among high-risk patients before gastrectomy for cancer.

Endoscopic Submucosal Dissection for Upper Gastrointestinal Neoplasia-a North American Perspective.

OBJECTIVES: Organ-sparing endoscopic submucosal dissection (ESD) is an acceptable treatment strategy for superficial neoplastic lesions of the esophagus and stomach. The adoption of this technique has lagged in North America compared with Asia, and we sought to report on our experiences with ESD for upper GI neoplasia. METHODS: A prospectively entered database of all patients undergoing endoscopic resection of esophageal and gastric neoplasia at McGill University from 2009 to 2019 was queried for those who received ESD. RESULTS: A total of 103 consecutive ESDs were identified from 2009 to 2019. Seventy-one (69%) patients were male and the median age was 72 (range: 38-90). Sixty-one (59%) cases were esophageal and 42 (41%) gastric. Forty-nine (48%) were performed in the endoscopy suite under local sedation only. Perforation occurred in 9 patients (7 esophageal and 2 stomach), of which 3 required operative repair. Histology was principally invasive carcinoma (79, 77%), with 17 (16%) dysplastic lesions (e.g., HGD), 1 (1%) neuroendocrine tumor, and 7 (7%) benign lesions. En bloc resection was achieved in 90 (87%), and the complete resection rate was 74 (72%), with 51 (50%) of procedures fulfilling the criteria for curative resection. At medium of 23-month (2-199) follow-up of these 51 curative resections, one case of recurrent carcinoma was found at follow-up and was managed with repeat endoscopic resection. Non-curative ESDs were found 45 (R1 resection = 29: risk of lymph node metastasis = 16), 21 had active surveillance, and 24 were resected. CONCLUSION: ESD is a viable, effective, and safe therapeutic and staging modality for superficial lesions of the stomach and esophagus.

Meta-analysis of the Diagnostic Accuracy of C-Reactive Protein for Infectious Complications in Laparoscopic Versus Open Colorectal Surgery.

INTRODUCTION: C-reactive protein may predict anastomotic complications after colorectal surgery, but its predictive ability may differ between laparoscopic and open resection due to differences in stress response. Therefore, the objective of this study was to perform a systematic review and meta-analysis on the diagnostic characteristics of C-reactive protein to detect anastomotic leaks and infectious complications after laparoscopic and open colorectal surgery. METHODS: A systematic review was performed according to PRISMA. Studies were included if they reported on the diagnostic characteristics of postoperative day 3-5 values of serum C-reactive protein to diagnose anastomotic leak or infectious complications specifically in patients undergoing elective laparoscopic and open colorectal surgery. The main outcome was a composite of anastomotic leak and infectious complications. A random-effects model was used to perform a meta-analysis of diagnostic accuracy. RESULTS: A total of 13 studies were included (9 for laparoscopic surgery, 8 for open surgery). The pooled incidence of the composite outcome was 14.8% (95% CI 10.2-19.3) in laparoscopic studies and 21.0% (95% CI 11.9-30.0) for open. The pooled diagnostic accuracy characteristics were similar for open and laparoscopic studies. However, the C-reactive protein threshold cutoffs were lower in laparoscopic studies for postoperative days 3 and 4, but similar on day 5. CONCLUSIONS: The diagnostic characteristics of C-reactive protein in the early postoperative period to detect infectious complications and leaks are similar after laparoscopic and open colorectal surgery. However, thresholds are lower for laparoscopic surgery, suggesting that the interpretation of serum CRP values needs to be tailored based on operative approach.

The Impact of Delays to Definitive Surgical Care on Survival in Colorectal Cancer Patients.

INTRODUCTION: Treatment delay may have detrimental effects on cancer outcomes. The impact of longer delays on colorectal cancer outcomes remains poorly described. The objective of this study was to determine the effect of delays to curative-intent surgical resection on survival in colorectal cancer patients. METHODS: All adult patients undergoing elective resection of primary non-metastatic colorectal adenocarcinoma from January 2009 to December 2014 were reviewed. Treatment delays were defined as the time from tissue diagnosis to definitive surgery, categorized as < 4, 4 to < 8, and ≥ 8 weeks. Primary outcomes were 5-year disease-free (DFS) and overall survival (OS). Statistical analysis included Kaplan-Meier curves and Cox regression models. RESULTS: A total of 408 patients were included (83.2% colon;15.8% rectal) with a mean follow-up of 58.4 months (SD29.9). Fourteen percent (14.0%) of patients underwent resection < 4 weeks, 40.0% 4 to < 8 weeks, and 46.1% ≥ 8 weeks. More rectal cancer patients had treatment delay ≥ 8 weeks compared with colonic tumors (69.8% vs. 41.4%, p < 0.001). Cumulative 5-year DFS and OS were similar between groups (p = 0.558; p = 0.572). After adjusting for confounders, surgical delays were not independently associated with DFS and OS. CONCLUSIONS: Treatment delays > 4 weeks were not associated with worse oncologic outcomes. Delaying surgery to optimize patients can safely be considered without compromising survival.

Intracorporeal versus extracorporeal anastomosis for right colectomy does not affect gastrointestinal recovery within an enhanced recovery after surgery program.

INTRODUCTION: Delayed gastrointestinal (GI) recovery remains a significant morbidity after colorectal surgery. Intracorporeal anastomosis for right colectomy may hasten GI recovery. Therefore, the objective of this study was to determine the effect of intracorporeal versus extracorporeal anastomosis on GI recovery after elective laparoscopic right colectomy within an established ERAS program. METHODS: Adult patients undergoing elective laparoscopic right colectomy at a single high-volume institution from 07/2014 to 12/2018 were reviewed. Patients were divided into two groups: intracorporeal (IC) and extracorporeal (EC). The primary outcome was time to GI-3 defined as days to tolerance of solid diet and first flatus/bowel movement. Prolonged postoperative ileus (PPOI) was defined as GI-3 not met by postoperative day 4. Secondary outcomes were length of stay (LOS) and overall 30-day complications. Sensitivity analysis was performed using coarsened exact matching to account for unmeasured confounding. Multiple regression was performed using a Cox proportional hazard model to identify predictors of GI recovery. RESULTS: A total of 346 patients were reviewed, of which 226 were included (71IC, 155EC). Patient characteristics were well balanced between groups: mean age was 64.9 years (SD 15.9), BMI was 26.3 (SD 5.7), 38.1% of patients had ASA ≥ 3, and 78.3% underwent surgery for neoplasms. IC anastomosis was associated with longer operative duration (165 min (SD 40); 144 min (SD 48), p = 0.002). There was no difference in the median time to GI-3 (IC 2 days [IQR1-2]; EC 2 days [IQR2-3], p = 0.135). The incidence of PPOI (IC 8.5%; EC 10.3%, p = 0.659), superficial SSI (4.2% vs. 5.8%, p = 0.757), deep SSI (2.8% vs. 5.2%, p = 0.729), and median LOS (3 days [IQR 2-4] vs. 3 [IQR 3-5], p = 0.059) were also similar. On multivariate analysis, IC anastomosis did not independently predict faster GI recovery (HR 0.98, 95% CI 0.71-1.34). Similar results were observed in the matched cohort (185 patients (61IC, 124EC)). CONCLUSION: In this study, IC anastomosis was not associated with faster GI recovery or reduced complication rate compared to EC anastomosis. Longer term studies may be required to determine the potential benefits of IC anastomosis.

Examining the relationship between lymph node harvest and survival in patients undergoing colectomy for colon adenocarcinoma.

BACKGROUND: Current standards for lymph node harvest in colorectal cancer surgery may be inadequate. Higher lymph node yield may improve survival, but the number of lymph nodes needed to optimize survival is unknown. The objective of this study was to examine the relationship between lymph node yield and overall survival in patients undergoing colectomy for nonmetastatic colon adenocarcinoma. METHODS: The 2010 to 2014 National Cancer Database was queried for patients undergoing colectomy for nonmetastatic colon adenocarcinoma. Adjusted restricted cubic splines were used to model the nonlinear relationship between lymph node harvest and overall survival. Cox proportional hazard determined independent predictors of overall survival. RESULTS: A total of 261,423 patients were included. Restricted cubic splines demonstrated that the adjusted improvements in overall survival stabilized after 24 nodes. Patients were divided into: <12, 12 to 23, and ≥24 nodes. On survival analysis, patients with ≥24 nodes had better survival across all N stages compared to other groups (P < .001). Lymph node harvest ≥24 nodes was independently associated with improved overall survival compared to 12 to 23 nodes (hazard ratio 0.82; 95% confidence interval, 0.80-0.85). CONCLUSION: Lymph node harvest ≥24 nodes is associated with improved survival in colorectal cancer patients. These data may provide indirect evidence for a more extensive lymphadenectomy for colon cancer.

Comparison of Dor and Nissen fundoplication after laparoscopic paraesophageal hernia repair.

BACKGROUND: Fundoplication is performed routinely during laparoscopic repairs of a paraesophageal hernia, but the degree of fundoplication remains controversial. The purpose of this study is to assess postoperative dysphagia and reflux after a Dor versus a Nissen fundoplication in patients undergoing laparoscopic repair of giant paraesophageal hernias. METHODS: We performed a retrospective cohort study of all patients undergoing laparoscopic repair of giant paraesophageal hernias with Nissen or Dor fundoplication between January 2012 and December 2017 at a high-volume center, excluding revisional and emergency cases. Primary outcomes were reflux and dysphagia at 1 and 6 months. Severe dysphagia was defined as intolerance to liquids. Balanced cohorts were created using coarsened exact matching. RESULTS: A total of 106 patients were included, and 87 were matched (Dor = 48, Nissen = 58). Baseline characteristics were well balanced between matched groups. Mean follow-up duration was 17.7 months (standard deviation 16.4). The incidence of severe dysphagia at 1 month was less in the Dor group (0 of 48 vs 8 of 58, P = .02) with similar reflux symptoms. There was no difference in severe dysphagia and reflux symptoms at 6 months and at the latest visit. CONCLUSION: Dor fundoplication is associated with less severe, early postoperative dysphagia. Future studies assessing the relative importance of dysphagia and reflux on quality of life should be conducted to tailor the operative technique and optimize patient satisfaction.

Impact of residual nodal involvement after complete tumor response in patients undergoing neoadjuvant (chemo)radiotherapy for rectal cancer.

BACKGROUND: The management of patients with a complete clinical response after neoadjuvant therapy for rectal adenocarcinoma is controversial. Those who advocate for resection point out the inaccuracy of N-staging with current imaging modalities. The objective of this study is to determine the impact of residual nodal involvement after complete tumor regression after neoadjuvant (chemo)radiotherapy. METHODS: The 2004 to 2014 National Cancer Database was queried for patients undergoing proctectomy for nonmetastatic rectal adenocarcinoma who had received neoadjuvant (chemo)radiotherapy and with ypT0 on final pathology. Patients were grouped based on pathologic nodal stage: ypT0N- and ypT0N+. The main outcome was 5-year overall survival. RESULTS: There were 5,156 patients with ypT0N- and 527 with ypT0N+. Mean lymph node harvest was similar (ypT0N- 12.2 nodes [standard deviation 9.1] vs ypT0N+ 11.6 nodes [standard deviation 10.3]; P = .086). Patients with ypT0N+ were more likely to have had clinically involved nodes (P < .001) and earlier clinical T-stage (P = .002). Overall survival at 5 years was less for patients with ypT0N+ (80% vs 86%, log-rank P = .014). ypT0N+ was independently associated with worse overall survival (hazard ratio 1.74, 95% confidence interval 1.33-2.28). CONCLUSION: Residual nodal involvement despite complete tumor regression was associated with worse 5-year overall survival compared to complete pathologic response. Additional therapy should be considered in the presence of complete clinical tumor regression after neoadjuvant (chemo)radiotherapy.

The relationship of two postoperative complication grading schemas with postoperative quality of life after elective colorectal surgery.

INTRODUCTION: Several grading schemes are available to assess surgical complications, but their relationship with patient-reported outcomes is not well understood. Therefore, our objective was to examine the effect of two complication grading schemas on health-related quality of life in colorectal surgery patients. METHODS: An analysis of adult patients undergoing elective colorectal surgery from 2005 to 2013 was performed. Health-related quality of life was measured using the SF-36 preoperatively and at 4 weeks and 8 weeks postoperatively. The 30-day morbidity was classified using Clavien-Dindo grading (I-IV) and the Comprehensive Complication Index (0-100). The main outcomes were the postoperative changes in physical summary scores and mental summary scores. Multivariate logistic and fractional polynomial regression analyses were used to determine the relationship between complication severity and health-related quality of life. RESULTS: A total of 402 patients were included in the study. Overall morbidity was 46%. Patients with complications had lower physical summary scores and mental summary scores at 4-weeks and 8-weeks postoperatively compared with patients without complications (P < .05). On multivariate regression, there was no dose-response relationship between Clavien-Dindo grade and postoperative physical summary scores and mental summary scores. Adjusted changes in the physical summary scores and mental summary scores had a more appropriate, dose-response relationship with the Comprehensive Complication Index scores. CONCLUSION: In patients undergoing colorectal surgery, there is a more consistent relationship between the Comprehensive Complication Index and postoperative health-related quality of life compared with the Clavien-Dindo classification.

Cost-Effectiveness of Extended Thromboprophylaxis in Patients Undergoing Colorectal Surgery from a Canadian Health Care System Perspective.

BACKGROUND: There is increasing evidence to support extended thromboprophylaxis after colorectal surgery to minimize the incidence of postdischarge venous thromboembolic events. However, the absolute number of events is small, and extended thromboprophylaxis requires significant resources from the health care system. OBJECTIVE: This study aimed to determine the cost-effectiveness of extended thromboprophylaxis in patients undergoing colorectal surgery for malignancy or IBD. DESIGN: An individualized patient microsimulation model (1,000,000 patients; 1-month cycle length) comparing extended thromboprophylaxis (28-day course of enoxaparin) to standard management (inpatient administration only) after colorectal surgery was constructed. SETTINGS: The sources for this study were The American College of Surgeons National Surgical Quality Improvement Project Participant User File and literature searches. OUTCOMES: Costs (Canadian dollars), quality-adjusted life-years, and venous thromboembolism-related deaths prevented over a 1-year time horizon starting with hospital discharge were determined. The results were stratified by malignancy or IBD. RESULTS: In patients with malignancy, extended prophylaxis was associated with higher costs (+113$; 95% CI, 102-123), but increased quality-adjusted life-years (+0.05; 95% CI, 0.04-0.06), resulting in an incremental cost-effectiveness ratio of 2473$/quality-adjusted life-year. For IBD, extended prophylaxis also had higher costs (+116$; 95% CI, 109-123), more quality-adjusted life-years (+0.05; 95% CI, 0.04-0.06), and an incremental cost-effectiveness ratio of 2475$/quality-adjusted life-year. Extended prophylaxis prevented 16 (95% CI, 4-27) venous thromboembolism-related deaths per 100,000 patients and 22 (95% CI, 6-38) for malignancy and IBD. There was a 99.7% probability of cost-effectiveness at a willingness-to-pay threshold of 50,000$/quality-adjusted life-year. To account for statistical uncertainty around variables, sensitivity analysis was performed and found that extended prophylaxis is associated with lower overall costs when the incidence of postdischarge venous thromboembolic events reaches 1.8%. LIMITATIONS: Significant differences in health care systems may affect the generalizability of our results. CONCLUSIONS: Despite the rarity of venous thromboembolic events, extended thromboprophylaxis is a cost-effective strategy. See Video Abstract at http://links.lww.com/DCR/A976. COSTO-EFECTIVIDAD DE LA TROMBOPROFILAXIS EXTENDIDA EN PACIENTES SOMETIDOS A CIRUGÍA COLORRECTAL DESDE UNA PERSPECTIVA DEL SISTEMA DE SALUD CANADIENSE:: Cada vez hay más pruebas que apoyen la tromboprofilaxis extendida después de la cirugía colorrectal para minimizar la incidencia de eventos tromboembólicos venosos después del alta hospitalaria. Sin embargo, el número absoluto de eventos es pequeño y la tromboprofilaxis extendida requiere recursos significativos del sistema médico.Determinar la rentabilidad (relación costo-efectividad) de la tromboprofilaxis extendida en pacientes sometidos a cirugía colorrectal por neoplasia maligna o enfermedad inflamatoria intestinal.Un modelo de microsimulación de paciente individualizado (1,000,000 de pacientes; ciclo de 1 mes) que compara la tromboprofilaxis extendida (curso de enoxaparina de 28 días) con el tratamiento estándar (solo para pacientes hospitalizados) después de la cirugía colorrectal.Archivo de usuario participante del Proyecto de Mejoramiento de la Calidad Quirúrgica del Colegio Nacional de Cirujanos Americanos (ACS-NSQIP) y búsquedas bibliográficas.Costos (en dólares Canadienses), años de vida ajustados por la calidad y muertes relacionadas con el tromboembolismo venoso prevenidas en un horizonte temporal de 1 año a partir del alta hospitalaria. Los resultados fueron estratificados por malignidad o enfermedad inflamatoria intestinal.En pacientes con neoplasias malignas, la profilaxis extendida se asoció con costos más altos (+113 $; IC del 95%, 102-123), pero con un aumento de la calidad de vida ajustada por años de vida (+0.05; IC del 95%, 0.04-0.06), lo que resultó en un incremento de relación costo-efectividad de 2473 $/año de vida ajustado por calidad. Para la enfermedad inflamatoria intestinal, la profilaxis extendida también tuvo costos más altos (+116 $; 95% IC, 109-123), más años de vida ajustados por calidad (+0.05; 95% IC, 0.04-0.06) y una relación costo-efectividad incremental de 2475 $/año de vida ajustado por calidad. La profilaxis prolongada evitó 16 (95% IC, 4-27) muertes relacionadas con tromboembolismo venoso por cada 100,000 pacientes y 22 (95% IC, 6-38) por malignidad y enfermedad inflamatoria intestinal, respectivamente. Hubo un 99.7% de probabilidad de costo-efectividad en un límite de disposición a pagar de 50,000 $/año de vida ajustado por calidad. Para tener en cuenta la incertidumbre estadística en torno a los variables, se realizó un análisis de sensibilidad y se encontró que la profilaxis extendida se asocia con menores costos generales cuando la incidencia de eventos tromboembólicos venosos después del alta hospitalaria alcanza 1.8%.Las diferencias significativas en los sistemas de salud pueden afectar la generalización de nuestros resultados.A pesar de la escasez de eventos tromboembólicos venosos, la tromboprofilaxis extendida es una estrategia rentable. Vea el video del resumen en http://links.lww.com/DCR/A976.

Improved Disease-free Survival After Prehabilitation for Colorectal Cancer Surgery.

OBJECTIVE: The objective of this study was to investigate the effect of prehabilitation on survival after colorectal cancer surgery. SUMMARY OF BACKGROUND DATA: Preoperative multimodal exercise and nutritional programs (prehabilitation) improve functional capacity and recovery following colorectal surgery. Exercise may also affect cancer outcomes by mediating the systemic inflammatory response. The effect of prehabilitation on cancer outcomes is unknown. METHODS: Pooled data from 3 prehabilitation trials (2 randomized controlled trials, 1 cohort) in patients undergoing elective, biopsy-proven, primary non-metastatic colorectal cancer surgery from 2009 to 2014 within an enhanced recovery program were analyzed. Patients were grouped into +prehab or-prehab. The primary outcomes were 5-year disease-free (DFS) and overall survival (OS). DFS and OS were analyzed using Kaplan-Meier curves and multiple Cox regression. RESULTS: A total of 202 patients were included (+prehab 104, -prehab 98). Median prehabilitation duration was 29 days (interquartile range 20-40). Patient and tumor characteristics were well-balanced (33% stage III). Postoperative complications and time to adjuvant chemotherapy were similar. Mean duration of follow-up was 60.3 months (standard deviation 26.2). DFS was similar for the combined group of stage I-III patients (P = 0.244). For stage III patients, prehabilitation was associated with improved DFS (73.4% vs 50.9%, P = 0.044). There were no differences in OS (P = 0.226). Prehabilitation independently predicted improved DFS (hazard ratio 0.45; 95% confidence interval, 0.21-0.93), adjusting for stage and other confounders. Prehabilitation did not independently predict OS. CONCLUSION: In this report, prehabilitation is associated with improved 5-year DFS in stage III colorectal cancer. This finding should be confirmed in future trials.