Implementation of ERAS protocol in robot-assisted radical cystectomy with intracorporeal ileal conduit urinary diversion: An outcome analysis beyond the learning curve.

INTRODUCTION: The objective of this study was to evaluate the perioperative outcomes of patients undergoing robot-assisted radical cystectomy (RARC) with intracorporeal ileal conduit (IIC) urinary diversion treated in line with the enhanced recovery after surgery (ERAS) protocol.
METHODS: After approval from the institutional ethics committee, we conducted an analysis of a prospectively maintained database of patients undergoing RARC + IIC using ERAS protocol by a single surgical team with the da Vinci Xi® system from March 2016 till December 2018. To minimize the effect of the learning curve of this complex procedure, we excluded the first thirty patients from analysis.
RESULTS: Thirty-five consecutive patients (33 males and 2 females) with a median age of 69 years (range: 50-82) were evaluated. The median total console time and console time for diversion were 253 min (range: 191-370) and 80 min (range: 65-90), respectively. The median estimated blood loss was 300 cc (range: 50-500). The median length of stay was 8 days (range: 4-30). Per-urethral pelvic drain was removed at a median of 2 days (range: 1-17). Overall, complications occurred in 16/35 (45.7%) patients, of which major complications (≥Grade 3) were seen in 5/35 (14.3%) patients, without any 90-day mortality. The median follow-up for the cohort was 14 months (1-34).
CONCLUSIONS: While the initial outcomes of this combined treatment strategy appear promising in terms of complication rates and perioperative parameters, greater insight is required from multi-institutional data sets and prospective comparative studies to establish the true value of RARC + IIC and ERAS protocol for bladder cancer. Copyright:

INTRODUCTION

The basic goal for a minimally invasive (MI) approach is to reduce surgical stress and enable early recovery while ensuring minimal complications in the postoperative period. The concept of surgery-induced stress dates back to 1997 when Dr. Henrik Kehlet described the multimodal approach for controlling postoperative pathophysiology and promoting rehabilitation.[1] He emphasized that the neurohormonal storm has implications on the postoperative recovery and morbidity.[2] Radical cystectomy (RC), being a major procedure, has high rates of overall morbidity, including problems related to the recovery of bowel functions.[3] The use of a robotic approach for RC was popularized by Menon et al. in 2003.[4] In the same year, Karolinska Institute reported the reconstruction of ileal conduit in an intracorporeal manner after robot-assisted RC (RARC).[5] Since then, a few major centers across the world have reported their outcomes of intracorporeal urinary diversion (IUD) following RARC.[678] The potential advantages of an IUD technique include a faster bowel recovery due to reduced bowel handling, reduced fluid loss by evaporation, decreased incidence of ureteric strictures since the ureters do not need to be mobilized extensively, and lesser postoperative pain along with faster surgical recovery. However, none of these advantages have been categorically established, as published trials comparing open versus robotic RC till date have all utilized an extracorporeal diversion after RARC.[91011]

Enhanced recovery after surgery (ERAS) protocol represents another attempt to reduce surgical stress and complications and hasten the recovery after major surgery. Although there is Level 1 evidence in favor of ERAS protocols in colorectal surgery, its role in RC is not fully defined.[12] After the promulgation of RC guidelines by the ERAS society in 2013, a number of institutions have implemented these for open and MIRC.[131415161718] However, the literature on the utilization of ERAS principles combined with IUD after RARC remains sparse.[1618] This is quite surprising in view of the fact that a combination of RARC/IUD and ERAS protocol likely represents the most promising strategy to decrease the morbidity of this challenging procedure – a worrisome statistic which has largely remained unchanged in the last few decades. To provide greater insight into the true outcomes of this combination, we analyzed our prospectively maintained database of RARC with intracorporeal ileal conduit (IIC) urinary diversion. To exclude the impact of the learning curve of IIC on the outcomes, we only included those procedures that were performed after overcoming the initial learning curve of 30 patients, as defined by the Pasadena consensus.[19]

METHODS

Of the total 65 RARC + IIC procedures performed by a single surgical team, the last 35 cases (after completion of the learning curve in the initial 30 cases) from our prospectively maintained, ethics committee approved, RARC database were analyzed. These procedures were performed at our tertiary care dedicated oncology institute, from March 2016 till December 2018, on patients who were candidates for RC due to the presence of muscle-invasive disease, or high-grade pT1 disease treated with early cystectomy, or for Bacillus Calmette–Guérin failure. A dedicated fellowship-trained console surgeon with an experience of over 300 robotic prostatectomies and thirty robotic cystectomies before March 2016 was the primary operating surgeon for all patients in this study. All patients were operated using the da Vinci Xi® (Intuitive Surgical, Sunnyvale, California, USA) system. We tried to implement the standard ERAS principles in all our patients. Of the 22 defined principles, we had minor variations in certain parameters, as discussed subsequently [Table 1]. We involved a dedicated anesthetist for intraoperative and postoperative management to achieve maximum compliance with the protocol. Salient steps of surgery were as follows – standard port placement, positioning, docking of robot, completion of RC and extended template pelvic lymphadenectomy, reduction of Trendelenburg position before conduit formation, intracorporeal conduit formation with use of Endo-GI staplers, Bricker's type ureteroenteric anastomosis with PDS 4-0 (polydioxanone) or Stratafix 4-0 (Ethicon) suture over the internalized double-J stents, and specimen extraction from a Pfannenstiel incision (or vaginal vault in females) followed by stoma maturation. A multidisciplinary team held regular discussions on a daily basis for formulating plans as per patient progress and clinical parameters which included the surgeon and anesthetist, intensive care unit (ICU) team, physiotherapy, nutritionist, and stoma care nurse, as needed. Ryle's tube (RT) was removed on extubation after completion of surgery, and all patients were started on clear liquid diet on postoperative day 1 (POD 1). Indications for reinsertion of RT included significant bilious vomiting or abdominal distension after the resumption of oral diet. These patients were also considered for computed tomogram abdomen with oral contrast if the clinical situation did not improve within 48 h after RT insertion. Unless indicated otherwise, patients were shifted out of the ICU on POD 2 and planned for discharge on POD 5. Discharge from hospital was initiated when the patient was self-mobilized with full medical optimization with regular passage of flatus and/or stools while tolerating soft diet without any nausea or vomiting.

Table 1

Pathway for robot-assisted radical cystectomy with intracorporeal urinary diversion with application of enhanced recovery after surgery protocol (Table 1)

Timeline	ERAS principle followed
Preoperative	Preoperative counseling and education
	Preoperative medical optimization
	Smoking cessation
	Stoma nurse counseling
One day prior to surgery	No mechanical bowel preparation
	Clear liquids till 2 h prior to surgery
	Light solids till 6 h prior to surgery
	Carbohydrate loading 2 h prior to surgery in nondiabetic patients
Intraoperative	No epidural catheter
	Antibiotic prophylaxis (third-generation cephalosporin) at the time of induction
	Induction of anesthesia-propofol 2 mg/kg + midazolam 1 mg + fentanyl 2 mcg/kg + atracurium 0.5 mg/kg
	Maintenance of anesthesia-oxygen-air-desflurane mixture titrated to a minimum alveolar concentration of 1-1.3, dexmedetomidine 0.2-0.5 mcg/kg/h titrated to blood pressure and heart rate, and atracurium 20 mg/h
	Goal-directed fluid intravenous fluids 1 L every 6 h of surgery to avoid bowel edema
	Euthermia maintenance
	Minimally invasive approach with total intracorporeal urinary diversion
	Consistent surgical team
	Nasogastric tube removal immediately after the procedure
	Per-urethral catheter as surgical drain
	Avoidance of a separate abdominal drain
Postoperative	Early mobilization
	Pain management
	IV paracetamol 1 g 6 hourly
	IV diclofenac 75 mg 12 hourly (if normal renal profile), dexmedetomidine infusion in case of severe pain not controlled by the above medications (titration of dexmedetomidine dose to a Ramsay sedation scale of 2)
	Day 1 - oral clear liquids at 30 ml/h
	Chewing gum
	PONV- assessment
	Optimization of serum electrolytes
	Magnesium supplementation if needed - intravenous 2 g twice a day
	Thromboprophylaxis - LMWH
	No continuation of antibiotics after day 1
	Criteria for discharge - tolerating on full soft diet, medically optimized, mobilized without assistance, possible self-care of stoma, passing flatus and or stools regularly
Outpatient visit	Evaluation
Audit

PONV=Prevention of postoperative nausea and vomiting, LMWH=Low-molecular-weight heparin, ERAS=Enhanced recovery after surgery

All relevant data of these 35 patients were entered in our prospectively maintained database on a real-time basis. Preoperative details included age, gender, body mass index (BMI), and comorbidities. Operative details included console time, diversion time, estimated blood loss (EBL) and blood transfusions, and intraoperative complications, if any. Postoperative parameters included day of RT and drain removal, day of starting liquid diet and soft diet, day of passage of flatus and stools, complications (Clavien–Dindo [CD] grade), and length of stay (LOS).[20] On follow-up, histopathology report, requirement of adjuvant treatment, 90-day complication rates (CD grade), and 30-day readmission rates were also recorded in the database.

RESULTS

A total of 35 consecutive patients (33 males and 2 females) with a median age of 69 years (range: 50–82) underwent RARC + IIC from March 2016 to December 2018. Totally, five patients required adjuvant procedures along with cystectomy [Table 2]. The predominant American Society of Anesthesiologists (ASA) class of patients was ASA 3 (n = 28) with five patients being on antiplatelet therapy preoperatively. Aspirin 75 mg, once a day, was continued through the surgery for these patients. The median BMI was 24.2 kg/m2 (range: 14.3–42). The indications of cystectomy are given in Table 2. Five patients opted for neoadjuvant cisplatin-based chemotherapy before proceeding with RC among all those who were offered the same as per the indication.

Table 2

Demographic profile for patients of robot-assisted radical nephroureterectomy + intracorporeal ileal conduit urinary diversion

Parameter	n
Total	35
Male:female ratio	33:2
Adjunct procedures	5
RARC + IIC + left RANU	3
RARC + IIC + low anterior resection	1
Total robotic pelvic exenteration	1
Age (range), years	69 (50-82)
ASA3:ASA2	28:7
BMI (range), kg/m2	24.2 (14.3-42)
Median console time (range), min	253 (191-370)#
Median console time for diversion (range), min	80 (65-90)
Median EBL (range), ml	300 (50-500)
Blood transfusion (%)	2 (5.7)
Median LOS (range), days	8 (4-30)
Median day of RT removal (range), days	0 (0-1)
Median day of per-urethral drain removal (range), days	2 (1-17)
Median serum magnesium levels on POD 1 (before optimization), mEq/L	1.75
Median day of passing flatus (range), days	2.5 (1-5)
Median day of passing stools (range), days	4 (3-11)
Final histopathological distribution
ypT0	6
ypTis	1
pT1	6
pT2	11
pT3	7
pT4	4
Margin positivity (%)	1 (2.8)
Median lymph node yield (range)	24 (12-54)
Lymph node positivity (%)	14 (40)
Follow up (range), months	14 (1-34)
30-day readmissions (%)	2 (5.7)##
Open conversion	0
90-day complications, overall (%)	16 (45.7)
90-day complications, minor (CD grade 1/2) (%)	11 (31.4)
90-day complications, major (CD grade 3/4/5) (%)	5 (14.3)
90-day mortality rate (%)	0 (0)

#Excluding patients requiring adjunct procedures, ##Corrected readmission rate-11.4%. RANU=Robot-assisted radical nephroureterectomy, IIC=intracorporeal ileal conduit urinary diversion, ASA=American Society of Anesthesiology, BMI=Body mass index, EBL=Estimated blood loss, LOS=Length of stay, RT=Ryle’s tube, POD=Postoperative day, CD=Clavien-Dindo, RARC=Robot assisted radical cystectomy

The median total console time was 253 min (range: 191–370). This excludes patients who underwent simultaneous radical nephroureterectomy RNU, low anterior resection, or pelvic exenteration. The median console time for diversion was 80 min (range: 65–90). The median EBL was 300 cc (range: 50–500) without the requirement of any intraoperative packed cell transfusions. The median LOS was 8 days (range: 4–30). RT was removed on extubation in all patients, and per-urethral pelvic drain was removed at a median of 2 days (range: 1–17) after surgery. The median day for starting on clear liquids and soft diet was POD 1 and POD 3, respectively. The median day for passage of flatus and stools was 2.5 (range: 1–5) and 4 days (range: 3–11), respectively. The median serum magnesium and potassium levels were 1.75 mEq/L and 4.55 mEq/L, respectively, on the POD 1. The median serum magnesium level after optimization was 2.04 mEq/L. On final histopathology, the distribution of T-stage was 6 ypT0, 1 ypTis, 6 pT1, 11 pT2, 7 pT3, and 4 pT4, respectively. Surgical margins were negative in all except one patient who had a focally positive margin of the left ureter. He received adjuvant chemotherapy and was recurrence free at the last follow-up 11 months postsurgery. Three patients had incidental carcinoma prostate with Gleason score 3 + 3 in two and 4 + 3 in one of them. The median lymph node yield was 24 (range: 12–54). A total of 14 patients had positive lymph nodes with a median positive lymph node count of 3 and the range of positive nodes being 1–25 with a median lymph node density of 12.5% (range: 2.4%–96.2%). Seven patients underwent adjuvant platinum-based chemotherapy out of the subset which was referred as per the standard indications.

There were no intraoperative complications or open conversions. Postoperative complications were encountered in 16/35 (45.7%) patients within 90 days of surgery [Table 3]. Of these, 5 (14.3%) patients developed major (Clavien Grade 3 or 4) complications. Two of these patients developed an ileoileal anastomotic leak needing re-exploration with diverting proximal ileostomy, one at POD 7 and the other at just under 3 months after surgery after an episode of stroke and esophageal candidiasis. There was no mortality within the 90-day postoperative period. Five patients developed postoperative ileus requiring RT insertion. Two patients required readmission within 30 days (5.7%) (one of them for fever and another for hyponatremia), whereas two more patients were advised admission for fever but did not get admitted due to some logistical issues (corrected readmission rate: 11.4%). The median follow up was 14 months in this cohort with the longest duration of follow-up being 34 months (range: 1–34). On the last follow up, 6 patients had expired (four due to metastatic bladder cancer). Another two were undergoing systemic therapy for metastatic disease. The estimated cancer-specific survival at 34 months was 53.8% and the estimated overall survival at 34 months was 49.7% by Kaplan–Meier analysis [Figure 1a and b].

Table 3

90-day complications - Clavien-Dindo grades (per complication type)

Complication type (90 days)	n (%)	CD grade
Fever +/− UTI	6 (17.1)	2*
Paralytic ileus - oral contrast CT f/b RT insertion - managed conservatively	3 (8.6)	1**
Atrial fibrillation and arrhythmia - medical management	1 (2.8)	2
Hyponatremia	1 (2.8)	1
Melena - conservative management	1 (2.8)	1
Localized pelvic collection - pigtail drainage	1 (2.8)	3a
Hydronephrosis - antegrade PCN insertion in view of rising creatinine	1 (2.8)	3a
Wound infection requiring secondary suturing under LA	1 (2.8)	3a
Stroke + esophageal candidiasis + delayed anastomotic leak, exploration - prolonged ICU stay	1 (2.8)	4a
Ileoileal anastomotic leak with left ureterointestinal anastomotic leak - emergency re-exploration, bilateral PCN, hemodialysis, ICU stay	1 (2.8)	4b

*One patient did not receive antibiotics - Grade 1, **No requirement of TPN. TPN=Total parenteral nutrition, UTI=Urinary tract infection, CT=Computed tomogram, RT=Ryle’s tube, ICU=Intensive care unit, CD=Clavien-Dindo, PCN= Percutaneous nephrostomy, LA=Local anaesthesia

Figure 1

(a) Kaplan–Meier analysis for cancer-specific survival. (b) Kaplan–Meier analysis for overall survival

DISCUSSION

To assess the feasibility and outcomes of implementation of ERAS principles in patients undergoing RARC with IIC, we analyzed 35 consecutive patients from our prospectively maintained RARC database. To minimize the impact of the learning curve of this complex procedure, we only considered patients who were operated after the completion of the learning curve (initial thirty cases) as by the Pasadena consensus.[19] We found that, after allowing for minor modifications, ERAS protocol could be effectively and safely applied in patients undergoing this procedure, with potential improvements in terms of LOS and complication rates.

Till date, the evidence in favor of the ERAS pathway is quite robust for colorectal surgery, whereas it still remains sparse for urology in general and RC in particular. A meta-analysis of randomized controlled trials concluded that the LOS for patients undergoing colorectal surgery could be reduced by 2.5 days by following the ERAS pathway vis-a-vis standard care.[12] The overall complication rates were also significantly lower with the ERAS protocol.[12]

Data for the use of ERAS principles for cystectomy are quite limited till date, with most publications providing only Level 3b evidence.[21] One of the largest series reported till date mentioned 362 consecutive patients undergoing open RC with a fast-track approach in 2010.[22] They reinforced the advantages of fast-track principles on early bowel recovery, with a specific note on early removal of nasogastric tube. After that, multiple single-institutional experiences have demonstrated the advantages of ERAS in patients undergoing open RC. However, comparative arms are lacking in most of these reports.[21] In fact, there has been only a single randomized controlled trial (n = 101) for the use of ERAS principles in RC which showed a significant improvement in ERAS patients with respect to morbidity, quality of life parameters, analgesic requirement, and stay in intermediate care unit in postoperative phase (level of evidence 2b).[23] Gradually, there was an evolution toward the miniaturization of surgical incisions through a laparoscopic approach, as a part of ERAS.[24] Since the addition of the surgical robot to the urological armamentarium, the combination of a robotic approach with ERAS principles for RC provided an opportunity to further improve the outcomes of this morbid procedure. Initial data from Abaza and Saar reported the encouraging outcomes of robotic cystectomy with enhanced recovery protocol, the results of which were reported as “clinical pathway for 3-day stay for robotic radical cystectomy.”[1517] Moreover, with growing experience in the last 5 years, intracorporeal diversion replaced its extracorporeal counterpart at several advanced centers in the world.[1618] While this is MIRC in its purest sense, the current data are again limited to single-center experiences without any randomization.

Our study provides further evidence on the applicability of ERAS principles to this procedure and compares favorably with the data reported so far on this topic. Our median LOS was 8 days, and our overall and major complication rates were 45.7% and 14.3%, respectively. This is likely attributable to the combination of the potential advantages of an intracorporeal technique of urinary diversion, the involvement of a dedicated multidisciplinary team (including an anesthesiologist) in the perioperative care of the patients, and consistent bedside and console surgeons who had overcome the initial learning curve of this procedure, working together. This combination of positive factors likely represents the best possible outcomes for this procedure in the current scheme of things. This is of course notwithstanding the fact that there is no definite cutoff point for a learning curve, and there is always further scope for improvement in intra- and perioperative care for this (or for that matter, any other) procedure.

We could implement most of the ERAS principles in our patients undergoing RARC + IIC with few subtle modifications [Table 1]. Although the resection site drains can be safely avoided as per ERAS recommendations, we feel that a per-urethral catheter drain can give an opportunity to test drain fluid creatinine levels in cases of higher outputs. In most of these patients, the drain is removed within 2 days, unless indicated otherwise. The expert panel too had an 82% consensus on the placement of a passive silicon drain through a port site for 1 day.[21] A subtle modification for goal-directed fluid management is to restrict the intravenous fluid supplementation intraoperatively while ensuring euvolemia with strict monitoring of central venous pressure, heart rate, and perfusion status [Table 1]. In this group of patients, we have continued antibiotic prophylaxis (third-generation cephalosporin) on POD 1 and stopped it from POD 2. The Pasadena consensus panel in 2015 recommended the use of antibiotic within 1 h before surgery in RARC with continuation in men for 24 h and women for 48 h.[19] Alvimopan is not commercially available in our country yet and hence not used by us. One consistent finding was the detection of low-to-low normal levels of serum magnesium which required replacement.

While some MI surgery-related advantages and oncological equivalence of RARC are now well established through prospective randomized trials, the role of IUD is not yet clear, since these studies have almost universally utilized an extracorporeal method for urinary diversion.[911] We routinely use intracorporeal diversions in all our patients undergoing RARC, and the diversion time (at an average of 80 min for an ileal conduit) seems to be within an acceptable range. This has further been reduced to about 60 min as we have advanced along our learning curve. There may be certain potential advantages of an intracorporeal diversion, as stated earlier, and this likely represents implementation of a robotic approach to RC in its purest sense. In an analysis of RARC from 18 international centers across the globe, the international robotic cystectomy consortium did find that intracorporeal diversion had lower 30-day and 90-day complication and readmission rates.[25] Ninety-six percent of the expert panel agreed that the recovery is different between RARC with intracorporeal diversion and RARC with extracorporeal diversion.[21]

In comparison to the Pasadena consensus panel recommendations, if we try to fit our outcomes based on console time, LOS, EBL, complication rates, and completeness of lymphadenectomy, we fall in the very experienced group (>100 cases) in terms of all parameters except for console time with conduit which falls in the experienced group (30–100 cases).[19] The apparent improvement in LOS and complication rates (comparable to very experienced group) may be attributable to the application of ERAS principles in our patients. In the light of recently published data of international robotic cystectomy consortium of 2125 RARCs (1094 with IUD) across 26 institutions, our results of LOS and overall and major complications fall in a similar range (9 days, 50%, 11% for international robotic cystectomy consortium IRCC).[26] Guru et al. have recommended gaining adequate experience in robot assisted radical prostatectomy RARP before considering RARC.[27] Furthermore, the arbitrary learning curve for RARC is considered to be 30 cases based on the time for surgery, lymph node yield, and positive surgical margin rates.[28] Hence, we excluded our first thirty patients of RARC + IIC before analyzing these results of consecutive 35 patients. The overall experience of the console surgeon (completion of more than 800 robotic procedures in uro-oncology till date) and a multidisciplinary team were the key elements behind the optimal outcomes achieved by us. In fact, the dedicated team has been recently introduced as the 23rd principle of ERAS by the EAU Robotic Urology Section Scientific Working Group consensus panel, which applies completely in our series.[21]

Our study, being a single-center, noncomparative analysis with relatively small numbers, does have its limitations. However, we do believe that it represents the real-world outcomes of RARC with simultaneous implementation of a purely intracorporeal diversion and the ERAS protocols, modified to the unique circumstances in our country. Moreover, it also eliminates the impact of the early learning curve of this procedure, thereby providing a realistic picture of this modality to aid decision-making by patients and their treating physicians.

CONCLUSIONS

Enhanced recovery protocols can be effectively implemented in patients undergoing RARC with IIC for bladder cancer. After completion of the initial learning curve, this approach provides a safe and reproducible alternative to an open surgical approach with the potential advantages of an earlier postoperative recovery and reduced complication rates. Multi-institutional data sets and prospective randomized studies are, however, required to establish these advantages conclusively.