INTRODUCTION OF AN ENHANCED RECOVERY PROTOCOL TO REDUCE SHORT TERM COMPLICATIONS FOLLOWING RADICAL CYSTECTOMY AND INTESTINAL URINARY DIVERSION WITH VESCICA ILEALE PADOVANA (VIP): EVALUATION OF COMPLICATION AND MORTALITY RATE

ABSTRACT Background and goals: Radical cystectomy (RC) and urinary diversion with vescica ileale Padovana (VIP) neobladder represents one option for patients needing bladder substitution. The origins of the fast track concept in the field of elective colon surgery can be traced back to the beginning of the 1990s and is based on an interdisciplinary approach by surgeons, anesthesiologists, nurses and physiotherapy staff to optimize perioperative care in order to decrease surgically induced stress. Enhanced recovery protocols (ERP) have been used successfully in patients undergoing pancreatic, colorectal and pelvic surgery, with a decrease in surgical stress and inpatient hospital stay while improving markedly postoperative recovery. The aim of this project is to develop a multimodal ERP inspired by the principles of fast track colorectal surgery, describing our results in a series of consecutive patients, with the purpose to reduce complications of RC and intestinal urinary diversion with the VIP neobladder. We decide, moreover, to compare the outcome of a small cohort of patients undergoing the ERP with a matched group of subjects who had undergone RC and VIP neobladder before implementation of the ERP in order to identify perioperative mortality risk factors. Methods: An ERP was introduced in our institution focusing on reduced bowel preparation, standardized feeding, analgesic regimens and early oral diet and mobilization. The key features of this protocol involved stopping use of mechanical bowel preparation, reduction of post-operative nausea and vomiting, implementation of postoperative pain control, early enteral feeding and mobilization soon after RC. Patient performance status was evaluated according to the Eastern Cooperative Oncology Group (ECOG) classification, while patient comorbidity was assessed by the age adjusted Charlson Comorbidity Index (CCI). American Society of Anaesthesiologists (ASA) score was used to estimate perioperative risk. Clinical and pathological staging was reported according to the 2002 and 2009 TNM system. Inclusion criteria were as follow: ASA score < 3; absence of malnutrition according to the Mini Nutritional Assessment – Short Form (MNA-SF) criteria; absence of inflammatory bowel diseases. All postoperative complications occurred within 90 days of surgery were recorded, defined and graded according to an established 5-grade modification of the original Clavien system. We analysed the impact of this ERP and reported the outcomes of all patients consecutively undergoing RC and VIP neobladder. In order to evaluate, moreover, prognostic factors of mortality in patients undergoing radical cystectomy, with the aim to ameliorate our ER protocol, we retrospectively reviewed data regarding 145 consecutive patients who underwent radical cystectomy and urinary diversion for urothelial bladder cancer at our Institute between 2002 and 2012 and 90-day mortality were collected and analysed and analyzed; moreover, Health Related Quality of Life (HRQoL) of this population were collected and analyzed. Results: Overall, 31 consecutive patients undergoing RC and VIP neobladder with curative intent were recruited to undergo our ERP. Moreover, we performed a comparative analysis between 9 patients of the ERP group and 13 patients of the control group. Twenty-nine out of 31 patients (ERP) were male with a mean age of 62.31 + 8.22 years; mean ASA score was 1.85 + 0.37. There were no specific comorbidities: median age adjusted Charlson comorbidiy was 2; 90% of patients presented an ECOG performance status 0. All patients underwent radical cystectomy with extended pelvic lymphadenectomy; in 5 cases another concomitant surgical procedure was performed and all patients underwent urinary reconstruction with a neobladder using the VIP technique. All patients were evaluated postoperatively every day, paying particular attention to nausea, vomiting, bowel activity, pain levels and adherence to the diet program. The bowel preparation with only the enema instead of the usual use of Polyethylene glycol, showed no difference in the "intestinal cleansing" of the bowel segment used for the urinary diversion. Mean operating time was 247.10 minutes; mean estimated intraoperative blood loss was 633.87 mL; 11 patients (35%) experienced vomiting, reflecting the same trend of nausea incidence. A postoperative blood transfusion was required in 4 patients. Only 7 patients took chewing gum; bowel function was comparable in patients with and without use of chewing gum. A normal stool evacuation occurred on post-operative day (POD) 1 in 1 patient (3.2%), on POD2 in 4 patients (12.9 %), on POD3 in 15 cases (48.3%), on POD 4 in 7 (22.5%), on POD5 in 2 (6.4%) and on POD8 in the remaining 2 patients (6.4%) (mean time interval 3.5 days). Postoperative pain was well controlled in all patients, with a mean VAS score of 3.19. Concerning the nutritional supply, in 18 patients (61%) it has been possible to follow the diet program according to the protocol. For 3 patients, it was necessary to administer a total parenteral nutrition from POD 1 to POD 4, from POD 3 to POD 7, and from POD 1 to POD 3, respectively, because of their poor compliance to the proposed diet protocol. The last surgical drainage removal usually occurred on POD 12 (range 10-19). The mean interval for the first ureteral catheter removal was POD8 (range 8-11), and for the second it was POD 9 (range 9-13). In 19 patients the bladder catheter was removed on POD 14 to 16; in the remaining 12 cases it was removed two weeks after discharge when a pouch-gram showed absent significant urine leakage (median POD 15; IQR 14-28). Overall median hospitalization was 14 days (IQR 14.5 – 15.5). Hospitalization was equivalent in patients with or without complications with a median of 15 days. No patient died from surgical complications. A total of 13 complications developed in 9 of the 31 patients (29.03%), none requiring surgical intervention. We did no record major medical complications. According to Clavien grading, all complications were grade < 2. With regard to the 90 days mortality prognostic factors, at multivariate analysis, no variable was independently related to perioperative mortality risk. Evaluating the ROC curves, ASA score was found to be the single variable with the highest accuracy in predicting 90 days mortality. Discussion: The application of a fast track program has been successfully performed in several surgical procedures, ranging from colorectal and hepatobiliary surgery other specialized surgical disciplines. The ERAS (Enhanced Recovery After Surgery) care pathways reduce surgical stress, maintain postoperative physiological function, and enhance mobilisation after surgery, resulting in reduced rates of morbidity, faster recovery and shorter length of stay in hospital in case series and in randomized trials in patients undergoing elective colonic surgery, pancreaticoduodenectomy, and rectal/pelvic surgery. To date no data are available on the application of evidence-based fast track programs to the VIP neobladder. In our study we applied our ERP to RC and intestinal urinary diversion with VIP neobladder, assessing the results in terms of feasibility and effectiveness, comparing the collected data with a matched group of patients without ERP. Unfortunately, our pilot study is underpowered and unsuitable to identify predictors of complications of any grade. Conclusion: In light of our preliminary data, the use of our fast track program was proven to be feasible and effective in the management of patients undergoing RC and VIP neobladder. The quality of the postoperative course was enhanced by the absence of a NGT and the reduction in TPN with early postoperative feeding. Postoperative pain was managed with good results, making possible early mobilization. All these findings led to a more rapid recovery of bowel function without occurrence of significant complications. A drawback of our case series is the small number of patients recruited, too small to assess some perioperative factors predicting the success of a enhanced recovery protocol and early postoperative complications and the design of the study. The best study protocol could be represented by a randomized controlled trial, for which, however, we need further confirmation of the effectiveness and safety of our protocol. At the moment, the relatively low number of complications and the absence of complications of degree higher than 2 according to the Clavien Dindo classification, the good response on early nutrition and mobilization, the good pain control, encourage us to continue our project and to improve our enhanced recovery protocol, inspired by the ERAS protocols currently widely used by Colleagues in General Surgery.