ORIGINAL CONTRIBUTION

Improving Conventional Recovery With Enhanced Recovery in Minimally Invasive Surgery for Rectal Cancer Wael Khreiss, M.D.1 • Marianne Huebner, Ph.D.2,3 • Robert R. Cima, M.S., M.D.1 Eric R. Dozois, M.D.1 • Heidi K. Chua, M.D.1 • John H. Pemberton, M.D.1 William S. Harmsen, M.S.2 • David W. Larson, M.B.A., M.D.1 1 Division of Colon and Rectal Surgery, Mayo Clinic College of Medicine, Mayo Clinic, Rochester, Minnesota 2 Department of Health Sciences Research, Mayo Clinic College of Medicine, Mayo Clinic, Rochester, Minnesota 3 Department of Statistics and Probability, Michigan State University, East Lansing, Michigan

BACKGROUND:  Enhanced recovery pathways have been shown to decrease the length of hospital stay in patients undergoing colorectal surgery. Few reports have studied patients undergoing minimally invasive surgery for rectal cancer. OBJECTIVE:  Our aim was to review our experience in minimally invasive rectal cancer surgery. We report ­shortterm outcomes and evaluate the potential advantages of the enhanced recovery protocol compared with our less intensive conventional pathway. DESIGN:  This is a consecutive retrospective study of all minimally invasive rectal cancers treated from February 2005 to December 2011. Multivariable logistic regression models were constructed to identify factors contributing to a short length of stay. SETTINGS:  This study was performed at Mayo Clinic, Rochester, Minnesota, between 2005 and 2011. PATIENTS:  A total of 346 patients were retrospectively reviewed. Seventy-eight patients were managed under the enhanced recovery pathway. Patients underwent either laparoscopic-, robotic-, or hand-assisted laparoscopic surgery for rectal cancer.

Funding/Support: The Department of General Surgery Research Committee funded this series. Financial Disclosures: None reported. Correspondence: David W. Larson, M.D., M.B.A, Department of Surgery, Division of Colorectal Surgery, Mayo Clinic, 200 First St SW, Rochester, MN 55905. E-mail: [email protected]. Dis Colon Rectum 2014; 57: 557–563 DOI: 10.1097/DCR.0000000000000101 © The ASCRS 2014 Diseases of the Colon & Rectum Volume 57: 5 (2014)

INTERVENTIONS:  All patients followed either a standardized conventional pathway or an enhanced recovery pathway for perioperative care. MAIN OUTCOME MEASURES:  The primary outcome was the length of stay. Secondary outcomes were postoperative complications and 30-day readmissions. RESULTS:  Hospital stay was significantly decreased for patients who underwent minimally invasive surgery for rectal cancer and were managed with an enhanced recovery protocol, 4.1 days, vs 6.1 days for the conventional pathway (95% CI, −2.9 to −1.2 days; p < 0.0001). Rates of complications were similar between the 2 groups. Factors associated with shorter length of stay included the enhanced recovery protocol and laparoscopic or robotic surgery compared with h ­ andassisted laparoscopic surgery. LIMITATIONS:  This was a retrospective study at a single institution. Additional limitations include the comparison with historical controls and the potential for selection bias. CONCLUSION:  The enhanced recovery pathway is associated with a significantly decreased length of hospital stay after minimally invasive surgery for rectal cancer in this series. Decreased hospital stay was achieved without affecting short-term outcomes. KEY WORDS:  Rectal cancer; Enhanced recovery protocol; Minimally invasive surgery.

M

inimally invasive surgery (MIS) has demonstrated superiority in terms of recovery and complications rates, without affecting oncological resection parameters or overall and disease-free survival.1 The CLASSIC trial from the United Kingdom has 557

558

reported these data in rectal cancer by demonstrating no difference in overall survival, disease-free survival, and local recurrence at 3 years,2 and up to 10 years of follow-up.3 Enhanced recovery pathways (ERPs) are clinical programs designed to reduce the physiological stress response to surgery, and to facilitate recovery by focusing on a preoperative, intraoperative, and postoperative approaches to pain control, early feeding, avoidance of fluid overload, and early ambulation.4–6 Enhanced recovery pathways have been shown to clearly benefit patients in open colorectal surgery.7 However, the use of ERPs has been limited in MIS,8,9 and the comparison of ERPs with other forms of standardized care is even more rare.10,11 Although there are reports of decreased hospital stay with MIS rectal surgery managed with accelerated recovery pathways, they remain in a smaller population size of between 50 and 100 patients.6 Hence, the applicability of ERPs in these complex rectal MIS procedures warrants further investigation. Recently, our own institutional version of ERP, which includes preoperative, intraoperative, and postoperative intervention demonstrated benefit over our previous conventional pathway for patients undergoing minimally invasive colon surgery.10 Our aim was to review the Mayo Clinic experience of MIS rectal cancer surgery and determine if ERP would improve the length of stay without associated negative outcomes.12,13

MATERIALS AND METHODS An ERP protocol was developed by a multidisciplinary team, which combined recommendations from the literature and a consensus agreement of colorectal surgeons at our institution.10 The institutional review board approved this consecutive retrospective study. The records of all patients over the age of 17 who underwent minimally invasive resection for rectal cancer from February 2005 until December 2011 were obtained from a surgical registry and included in this analysis. A total of 349 patients were identified; 2 were excluded from the final database because their surgery was palliative, and 1 was excluded because of missing research authorization. Therefore, a total of 346 patients were analyzed. All patients underwent either hand-assisted, laparoscopic-assisted, or robotic-assisted resections. In brief, ERP is composed of a multifaceted preoperative, intraoperative, and postoperative protocol. This includes preoperative and postoperative analgesia regimens and perioperative interventions, such as intrathecal analgesia, which are not a part of the conventional pathway. In general, intrathecal analgesia was at the discretion of the anesthesiologist; however, the majority of patients received this intervention. In addition, goal-directed intravenous fluids were provided to patients at a rate of 5 mL/ kg per hour of operation. Intraoperative administration consisted of lactated Ringer's solution and 500 mL of

Khreiss et al: Enhanced Recovery Pathway in Rectal Cancer

5% albumin, if the anesthesiologist believed that it was necessary upon evaluation of intravascular status with noninvasive means. Postoperatively, patients are rapidly introduced to diet and early ambulation (Table 1).10 Patients in this study were treated perioperatively with either ERP or the conventional pathway as a defined order set. These orders are instituted as a bundle before the end of the surgical procedure for the conventional pathway, and before surgery for ERP. Compliance with either order set was measured and deemed compliant if the order for the intervention was executed and the patient received the intervention. These orders were sent and enacted through an order set within Computerized Physician Order Entry in an intention-to-treat manner. Execution of this order in ERP patients was evaluated and recorded as the ultimate measure of compliance. The ERP protocol was rolled out to all surgeons within our division over the course of 18 months beginning in late 2009. High-volume surgeons were the first to enroll patients in enhanced recovery. Once enrollment in enhanced recovery began, all patients were enrolled from that point forward in this postoperative pathway. By the middle of 2011, all MIS rectal cancer patients followed ERP. All MIS patients currently are treated under the protocols defined by ERP. Surgeons with patients on ERP also had treated patients on the conventional pathway before this change in practice standard. The conventional pathway was the standard postoperative order set before surgeon initiation of ERP. Therefore, all MIS rectal operations during the study period followed 1 of these 2 standardized postoperative pathways. Demographics (including age, sex, height, weight, type of disease), ASA physical status classification, postoperative complications, outcomes, and readmissions were abstracted from the electronic medical record and the institutional colorectal prospective surgery database. All notes in the records, including outside records and reports, were included in the review. The primary end point was length of stay. This was defined as the time between the date of surgery and date of discharge. Patients were dismissed when they were tolerating general diet, pain was well controlled by oral medications, bowel function had returned, and there were no complications. These dismissal criteria were consistently used in both pathways. Secondary outcomes were readmission within 30 days of surgery, complications, and mortality. Thirty-day results are obtained through direct patient contact at the Mayo Clinic, or phone conversation, and occurred in all patients. All postoperative complications were included, defined a priori as follows: abscess, leak, deep space infection, superficial infection (antibiotics/opening), nonsurgical infection, cardiac (myocardial infarction, atrial fibrillation, or other cardiac event), pulmonary (pneumonia, respiratory depression, atelectasis), acute renal failure as defined by

559

Diseases of the Colon & Rectum Volume 57: 5 (2014)

TABLE 1.   Comparison between conventional pathway and enhanced recovery pathway10 Pathway component

Enhanced recovery pathway

Conventional pathway

Preoperative analgesia Postoperative nausea prophylaxis Fluid balance

Celecoxib 400 mg Gabapentin 600 mg Standardized based on risk factors Dexamethasone 4-8 mg IV + 5HT-3 antagonist Can add droperidol 0.625 mg Fluids at 500 mL/h unless hypovolemic Postoperative rate 40 mL/h, then saline lock at 0800 hours on day after surgery No epidural Intrathecal analgesia injection at discretion of anesthetist IV opioids IV ketorolac (end of case) No epidural Oral oxycodone as needed for pain score ≥ 4 Scheduled acetaminophen every 6 h Scheduled NSAIDS (ketorolac IV for 4 doses, then ibuprofen scheduled) NGT removed at end of procedure Goal 800 mL oral intake within 4 h of return to patient care area Regular diet started 4 h after procedure Encourage 2000-2500 mL/day fluids/supplements Evening of surgery: out of bed for >2 h (1 walk + up in chair) Postoperative day 1: out of bed for >8 h (4 or more walks + up in chair) Day 2: out of bed for >8 h (4 or more walks + up in chair)

None

Intraoperative analgesia

Postoperative analgesia

Diet

Ambulation

Recommended based on risk factors Dexamethasone 4-8 mg IV or 5HT-3 antagonist No target rates in the operating room Postoperative rate 80-125 mL/h until oral intake > 600 mL No epidural IV opioids ± ketorolac

No epidural IV PCA transitioned to as-needed oxycodone Scheduled acetaminophen every 6 h Scheduled NSAIDs for 3 days, then as needed NGT removed by 0600 hours on postoperative day 1 Evening of surgery: sips of clear fluid Postoperative day 1: full liquid diet (800 mL oral intake) Day 2: soft diet Day 3: normal diet Evening of surgery: mobilize 1 or more times Postoperative day 1: mobilize at least 4- 6 times Day 2: mobilize >6 times Day 3: normal activity

IV = intravenous; 5HT-3 antagonist = ondansetron or granisetron; PCA = patient-controlled analgesia; NSAID = nonsteroidal anti-inflammatory drug; NGT = nasogastric tube.

the Risk, Injury, Failure, Loss, and End-stage kidney disease criteria,14 urinary tract infection, urinary retention, venous thromboembolism, bleeding requiring transfusion, bowel obstruction, postoperative ileus, reoperation or other (dehydration requiring intravenous fluids, unplanned intensive care unit admission, hypotension requiring treatment). Ileus was defined as the inability to tolerate oral intake for >5 days with nausea and vomiting and/or requiring the insertion of a nasogastric tube (Table 2). All laparoscopic-assisted and robotic cases were completely performed in a MIS approach. They were

c­ onsidered either incisionless because the specimen was removed via the anus, or, for higher anterior resections, a Pfannenstiel incision of 4 to 5 cm was created to remove the specimen. Incision length for hand-assisted laparoscopic surgery (HALS) varied between 7 and 10 cm depending on the patient’s body habitus and the surgeon’s technique. Within this technique, much of the pelvic dissection was performed through the exteriorization port. Over time, the number of HALS and laparoscopic-assisted/robotic surgeries did not change, although robotic surgery did replace the laparoscopic-assisted technique.

TABLE 2.   Postoperative outcomes Complication classification

ERP (N = 78)

FTP (N = 268)

p

Any complication, n (%) Any major complication, n (%) Ileus (partial SBO), n (%) SBO predismissal, n (%) ARF, n (%) ARF, predismissal, n (%) Self-catheterization after 1 mo, n (%) Readmission within 30 days, n (%) Death within 30 days, n (%) Anastomotic leak, n (%) Reoperation for bleeding, n (%) Reoperation for bleed predismissal, n (%) Reoperation for SBO, n (%) Wound infection (requiring opening), n (%)

36 (46.2) 14 (18.0) 10 (12.8) 0 (0.0) 3 (3.8) 3 (3.8) 3/77 (3.9) 6 (7.7) 0 (0.0) 2(2.6) 1 (1.3) 1 (1.3) 0 (0.0) 1 (1.3)

150 (56.0) 63 (23.5) 59 (22.0) 1 (0.4) 7 (2.6) 4 (1.7) 16/266 (6.0) 37 (13.8) 2 (0.7) 7 (2.6) 1 (0.4) 1 (0.4) 1 (0.4) 4 (1.5)

0.126 0.2990 0.078 1.00 0.700 0.19 0.583 0.150 1.00 0.323 0.401 0.400 1.000 0.400

ERP = enhanced recovery pathway; FTP = fast-track pathway; ARF = acute renal failure; SBO = small-bowel obstruction.

560

STATISTICAL ANALYSIS Continuous variables were summarized as mean (SDs) or median (range) as appropriate, whereas categorical variables were summarized as frequencies and percentage. Two-sample t tests or Wilcoxon rank sum tests were used to compare continuous characteristics of patients following the ERP and the conventional pathway. For the comparison of categorical variables, a χ2 test or Fisher exact test were used. The subset of patients who did not experience major complications was used to assess patient and surgical characteristics for their association with hospital length of stay in ERP and the conventional pathway. Major complications were defined as ileus, small-bowel obstruction, leak or abscess, wound infection, renal failure, or reoperation for bleeding. To account for variability among the 9 surgeons in multivariable models, surgeon volume was defined as high if the number of cases exceeded 100 per year (3-year case load average, 2009–2011). Age, BMI, ERP or the conventional pathway, ASA, neoadjuvant therapy, surgeon volume, and sex were factors examined for association with longer length of stay (LOS). Least absolute shrinkage and selection operator (lasso) with 10-fold cross-validation was used for variable selection.15 This method uses a penalty function on the regression coefficients to produce sparse models, and the coefficients are smaller (“shrunk”). Such shrinkage adjustment provides more conservative estimates for odds ratios (“lasso OR”). The selected variables were entered in multivariable logistic regression models to test short LOS (1–3 days) for patients without complications. As a measure of the predictive ability of the model the concordance index (c-index) and the adjusted c-index after internal model validation with 200 bootstrap samples were reported. The concordance index ranges from 0.5 to 1.0, with values of 0.5 indicating no predictive ability and a value of 1.0 indicating perfect predictive ability. A model is considered reliable when the concordance index is more than 0.8. In addition to the ORs from the logistic regression model, we also report a shrinkage OR from the penalized regression model. All statistical tests were 2-sided, and a level of 0.05 was considered statistically significant. Data analyses were performed using R v.2.14.1 and SAS 9.2 (SAS Institute, Inc, Cary, NC).

RESULTS Sixty-eight percent of the patients were male, with a mean age of 59 years for the entire cohort. The majority of patients, 68%, had an ASA of 1. There was no significant difference with regard to age, ASA, sex, and BMI between groups. A total of 78 patients were cared for with the ERP, and 268 patients were cared for with the conventional pathway. Patients underwent 1 of 3 procedures: low anterior resection (n = 250), abdominoperineal resection (n = 68), and proctocolectomy (n = 28). There was no significant

Khreiss et al: Enhanced Recovery Pathway in Rectal Cancer

difference in the procedure used between the 2 groups, p = 0.29. There was no difference in previous laparotomy between groups. Twenty-six percent of the conventional pathway patients had a previous laparotomy compared with 22% in the ERP group, with a p value of 0.06. Significantly more patients in the ERP group underwent robotic resection than in the conventional pathway group, 33% and 7%; p 

Improving conventional recovery with enhanced recovery in minimally invasive surgery for rectal cancer.

Enhanced recovery pathways have been shown to decrease the length of hospital stay in patients undergoing colorectal surgery. Few reports have studied...
232KB Sizes 0 Downloads 3 Views