Enhanced Recovery Program in Laparoscopic Colorectal Surgery: an Observational Controlled Trial

1. Introduction One of the most important achievements in modern surgery is the implementation in perioperative care of Enhanced Recovery Programs (ERPs), pioneered by Henrik Kehlet in the mid-1990s [1]. ERP is an evidence-based perioperative care approach, involving a multidisciplinary team work including surgeons, anaesthesiologists, dieticians, and nurses, which aim to reduce the surgical stress and metabolic responses as well as organ dysfunction, thus leading to a faster recovery after surgery [2]. Such a team works with a multimodal approach to perioperative care, sharing all the endpoints of management throughout the perioperative phases, and acting in close relationship with the patient along his/her "journey in hospital". In other words, each perioperative phase comprises several core items based on solid scientific evidence that are delivered by different professionals in different parts of the hospital [2]. A few randomized clinical trials and meta-analysis provide evidence in the literature for the benefits of ERPs in colorectal surgery [3-6]. Most non-randomized controlled studies have compared patients undergoing colorectal enhanced recovery surgery with controls being either historical groups or patients operated on at different hospitals [7-9]. In a previous investigation, the investigators showed the benefits of implementing an ERP in elective colorectal surgery by comparing a prospective series of 100 patients operated on receiving enhanced recovery care (ERP group), with a retrospective cohort of 100 patients operated on by the same professionals at the same institution before the introduction of the ERP (Pre-ERP group) [10]. The present study aims to investigate the differences between a prospective series of patients undergoing elective colorectal surgery following an ERP and a retrospective cohort of patients operated on with traditional perioperative care in the same period by a different team of surgeons and hospitalized in a different ward of the same hospital.
2. Materials and Methods

The study was designed as a prospective cohort with retrospective coeval control. A prospective series of 100 consecutive patients undergoing elective colorectal surgery completing a standardized ERP protocol at the S. Anna University Hospital in Ferrara (Italy) in 2013-2015 (ERP group), was compared to a retrospective series of 100 consecutive patients operated at the same hospital, in the same period of time, but with a traditional perioperative care protocol (Non-ERP group). Despite different surgeons and ward nurses, the two groups of patients shared the same anaesthesiologists and were located in two separate wards.

In 2012 the enhanced recovery team was assembled, the ERP protocol was built up, according to the fast-track protocol proposed by Kehlet and Wilmore [11], and periodical audits were then performed. Eligible for this study were all patients between 18 and 80 years old scheduled for elective colorectal resection. Exclusion criteria in both study groups were: age > 80 years old, American Society of Anaesthesia (ASA) score IV, TNM stage IV, and diagnosis of inflammatory bowel disease. Discharging criteria were also the same for both groups: 1) free oral diet, 2) complete intestinal function recovery (passage of flatus and stool), 3) dynamic pain control (Numerical Rate Scale - NRS ≤ 3) by oral analgesia, 4) independent walking, 5) no signs of infection [10].

The study protocol was approved by the local Ethics Committee (Ethical Committee for Human Subject Research study number 52-2011), it complies with the Declaration of Helsinki, and all patients provided written informed consent The primary outcome of this study was the length of stay (LoS) in the hospital, adopted as a proxy of functional recovery. Secondary outcomes were: postoperative complications, readmission rate, mortality, and adherence to the protocol.

All complications were recorded until 30 days after surgery, as well as mortality and hospital readmission. To evaluate adherence to the protocol, all ERP items were prospectively collected, recorded and checked by an independent observer using a detailed care program for each postoperative day. The grading of complications was defined according to a validated classification [12].

Data were collected prospectively for patients in the ERP group and retrieved from a prospectively maintained database for those in the Non-ERP group.

Surgery was preferably performed by laparoscopy, but alternative approaches were allowed and discussed with patients. A converted laparoscopy was defined as an unplanned extension of the surgical incision (either a median or a transverse suprapubic incision). All operations were performed by surgeons experienced in colorectal surgery and advanced laparoscopy (CVF, GA).

2.1 ERP Group Patients received extensive preoperative counselling about protocol items, all phases of perioperative care, and discharge criteria, along with an illustrated booklet conceived by the multidisciplinary team. The patients were motivated to adhere to protocol items by members of the team throughout the preoperative and postoperative phases and were informed about the daily objectives to be achieved. Incentive spirometers were provided preoperatively and patients were instructed for perioperative breathing exercises.

A normal meal was allowed on the day before surgery; a carbohydrate loaded drink was prescribed in the evening (100 grams of maltodextrin in 800 ml liquid drink) and at least two hours (50 grams of maltodextrin in 400 ml liquid drink) before surgery. No patient was prescribed mechanical bowel preparation.

The nasogastric tube (NGT) and bladder catheter were removed at the end of the procedure and within the second postoperative day [13], respectively, whereas abdominal drains were selectively used.

Postoperative pain control was achieved with thoracic epidural analgesia within the first three postoperative days and with Nonsteroidal Anti-Inflammatory Drugs (NSAIDs) or paracetamol thereafter. Resumption of liquid and solid diets and discontinuation of intravenous fluids were all achieved within the first and the second postoperative day. Patients were out of bed on postoperative day one and encouraged by nurses to deambulate according to a predefined schedule.

The anaesthetic protocol was standardized as previously described elsewhere [10]. In brief, a blended anaesthesia was performed with short-acting anaesthetic agents and mid-thoracic epidural intraoperative analgesia. Attention was put on maintaining normothermia with forced air warming devices during surgical procedures and a multimodal approach to prevent PostOperative Nausea and Vomiting (PONV) was pursued. Intravenous fluid overload was avoided.

2.2 Non-ERP Group Patients were treated according to traditional perioperative care principles in use at the institution. Specifically, on the day before surgery patients fasted from midnight, received no carbohydrate loaded drink, and performed a mechanical bowel preparation.

An opioid-based anaesthesia with or without thoracic epidural was used, while postoperative pain control was achieved by intravenous opioids, thoracic epidural or NSAIDs.

The NGT was removed at the restoration of intestinal activity when oral feeding was resumed, while the abdominal drain was removed at bowel movements.

No predefined schedule was followed for patients' mobilization and the bladder catheter was removed at full mobilization and after discontinuation of intravenous fluids.

Patients received no daily goal assignments or specific preoperative counselling.

In order to determine in between groups perioperative care differences, an independent observer reviewed the medical charts of all patients to evaluate the adherence to the 21 protocol items.

2.3 Statistical Analysis Clinical parameters were expressed as median (interquartile range - IQR 25-75) and mean ± standard deviation according to distribution assessed by Shapiro-Wilk test. Categorical data were presented as numbers. Clinical and pathological variables were analysed with Chi-square, t-student, and Mann-Whitney tests as appropriate. The Kaplan-Meier test method and Log-Rank test were used to compare duration of surgical operation, time to functional recovery and hospital LoS between groups. A P value of less than 0.05 was considered statistically significant. Statistical analysis was performed with IBM SPSS Statistics for Windows, Version 24.0 (IBM Corp. Armonk, NY: IBM Corp.).