Fast-track rehabilitation after robot-assisted laparoscopic cystectomy accelerates postoperative recovery

Authors


Correspondence: Matthias Saar, Department of Urology and Paediatric Urology, University of Saarland, D-66421 Homburg/Saar, Germany.

e-mail: matthias.saar@uks.eu

Abstract

What's known on the subject? and What does the study add?

  • There is evidence from large abdominal surgeries and some open cystectomy series that multifactorial fast-track regimens shorten postoperative convalescence without any effect on morbidity and mortality. Such a regimen is of particular interest in combination with minimally invasive techniques, as early patient recovery demands for more rapid nutrition and mobilisation schemes.
  • The present study, in a single institution, reports on the design, application and results of a fast-track protocol in patients undergoing robot-assisted laparoscopic cystectomy. There was no evidence of a higher incidence of complications with the fast-track regimen and postoperative recovery was faster.

Objectives

  • To evaluate the feasibility and effectiveness of a multifactorial fast-track (FT) regimen on perioperative outcomes in patients undergoing robot-assisted laparoscopic cystectomy (RALC) with extracorporeal urinary diversion.
  • To point out that morbidity and mortality of radical cystectomy have improved markedly over the last decades and RALC is an emerging technique showing further advances in postoperative recovery, thus demanding for more rapid nutrition and mobilisation schemes.

Patients and Methods

  • A non-randomised cohort study of 63 patients who underwent RALC at one institution between January 2007 and March 2010.
  • In all, 31 patients underwent RALC without FT and 31 RALC with FT. One patient required conversion to open surgery and was therefore excluded from the study.
  • The FT regimen included early nutrition and the quickest possible mobilisation, while mechanical bowel preparation before surgery, as well as preoperative fasting and nasogastric or abdominal drains after surgery, were omitted.
  • Demographics, perioperative and complication data (according to modified Clavien system), as well as required opioid pain medication were documented prospectively and compared between RALC patients with and without FT.

Results

  • Groups were comparable for demographics, risk factors and clinical stage as well as operative parameters, e.g. mean operating room time, estimated blood loss, lymph nodes removed and postoperative haemoglobin level.
  • In the FT group, abdominal drains were mostly omitted and nasogastric tubes were removed immediately after surgery.
  • There were significant differences in the mobilisation within the room (17.5 vs 31.2 h), the time to a regular diet (4.0 vs 6.6 days) and a remarkably lower use of postoperative morphine equivalents (57.3 vs 92.4 mg) for patients receiving FT.
  • There were no significant differences in the overall complication rates or major complications based on Clavien classification.
  • The informative value of the study is limited by its single-centre, non-randomised design, a relatively small sample size and a possible learning curve bias.

Conclusions

  • Combining RALC with FT is feasible in the perioperative treatment of these patients.
  • Multifactorial postoperative regimens seem to quicken postoperative recovery of RALC patients without increasing their risk of postoperative complications.

Introduction

Radical cystectomy with urinary diversion is the standard procedure for muscle-invasive bladder cancer. Since the early 1900s with operative death rates of >50% [1], perioperative morbidity and mortality have significantly reduced [2] due to improved surgical techniques, better urinary diversion and modern anaesthetic management. However, it still remains an important endeavour to minimise surgical trauma and optimise perioperative care [3]. From its first descriptions in 2003 [4, 5], robot-assisted laparoscopic cystectomy (RALC) developed with evidence of acceptable morbidity and good short-term oncological results [6]. Compared with the open approach, RALC seems to be favourable in several perioperative parameters [7, 8] and is an independent predictor of fewer overall and major complications [9]. The advantages of laparoscopy are well documented including decreased surgical blood loss, less postoperative pain, earlier return of bowel function and accelerated postoperative convalescence [10-13], demanding for more rapid nutrition and mobilisation schemes.

The concept of ‘fast-track surgery’ emphasises techniques that facilitate early recovery after surgery, involving modern anaesthesia, minimally invasive procedures, optimal pain control and active postoperative rehabilitation with early mobilisation and oral nutrition in a self-contained care programme [14].

Whereas this concept has been intensively studied in various surgical procedures [15], no data for a fast-track (FT) rehabilitation programme in RALC have been reported yet.

Some publications have indicated the advantages of multimodal perioperative regimen in radical cystectomy patients for high-risk minorities [16, 17]. Especially the combination of such multimodal perioperative approaches with laparoscopy assisted techniques is desirable to achieve all previously mentioned and maybe even some new synergistic benefits in terms of complications, duration of hospital stay and overall convenience. The present study describes our first experience of RALC combined with a FT approach, developed over 3 years.

Patients and Methods

From January 2007 to March 2010, 63 consecutive patients underwent RALC by three different surgeons at one institution (robotic centre Homburg/Saar). One patient within the first 32 RALCs required conversion to open cystectomy and was therefore excluded from the study. Thus, 62 patients could be evaluated. In all, 31 patients (group 1) were treated perioperatively with a conservative regimen as used for open cystectomy including preoperative mechanical bowel preparation, i.v. antibiotics for 5 days, nasogastric tubes and abdominal drains after surgery, postoperative fasting for up to 5 days and patient mobilisation regularly starting 2 days after surgery. Introducing RALC, we saw a faster patient recovery compared with those undergoing conventional open procedure, demanding for new regimens in postoperative care. Therefore, different elements of FT surgery were introduced and subsequently summarised in a perioperative care plan (Table 1). Whereas the first 31 patients after RALC where mainly treated like patients for open cystectomy, treatment of the following 31 patients was based on our designed multimodal FT protocol (group 2).

Table 1. Perioperative care plan.
DayNo FTFT
Preoperative:  
−2

Normal breakfast

Admission to hospital

Liquid diet (soup)

Tap water enema

Fleets enema in the evening

Patient at home
−1

Liquid diet (soup)

Unrestricted clear fluids

Tap water enema

Fleets enema in the evening

Normal breakfast

Admission to hospital

Liquid diet (soup)

Unrestricted clear fluids

2 high carbohydrate drinks 1800 hours

Fleets enema in the evening

Surgical

Intraoperative antibiotic treatment and postoperative maintenance for 5 days

Abdominal drainage

Nasogastric tube

2 high carbohydrate drinks 0500 hours

Intraoperative antibiotic single injection

No abdominal drainage

Removal of nasogastric tube at the end of procedure

Postoperative:

No fluids

Mobilisation as possible

Pain medication using i.v. non-opioids and opioids as required

Free fluids as tolerated 2 h after surgery

Magnesium citrate 300 mg

Mobilise first time 5 h after surgery

Non-narcotic analgesics (diclofenac 75 mg every 12 h)

Pain medication using i.v. non-opioids and opioids as required

+1

Fluids as tolerated

Pain medication using i.v. non-opioids and opioids as required

i.v. antibiotic treatment

Mobilise and refer to physiotherapist

Protein drinks and light diet as tolerated

Free fluids in >1500 mL

Magnesium citrate 3 × 300 mg

Non-narcotic analgesics (diclofenac 75 mg every 12 h)

Pain medication using i.v. non-opioids and opioids as required

Mobilise and refer to physiotherapist

8 h out of bed; twice walking in the corridor

+2

Fluids as tolerated

Pain medication using i.v. non-opioids and opioids as required

i.v. antibiotic treatment

Nasogastric tube removal missing fluid quantity

Mobilise and refer to physiotherapist

Light diet, regular diet possible

Rectal laxative (bisacodyl suppository)

Non-narcotic analgesics (diclofenac 75 mg every 12 h)

Pain medication using i.v. non-opioids and opioids as required

Mobilise and refer to physiotherapist

8 h out of bed; twice walking in the corridor

+3

Rectal laxative (bisacodyl suppository); i.v. antibiotics

Remove abdominal drainage and start oral diet after first bowel movement; full mobilisation

Regular diet after gradual return to oral feeding

Mobilise and refer to physiotherapist

8 h out of bed; full mobilisation

+4
+5

Surgical Procedure

The surgical technique of RALC has been described previously [6, 18]. The operation included the removal of the bladder, prostate and seminal vesicles in males and the removal of the bladder, uterus and adnexae as well as the anterior vaginal wall in female patients. For both genders a bilateral pelvic lymph node dissection was performed in cases of malignant bladder tumour (60 patients). Furthermore, the left ureter was transposed to the right side laparoscopically and in case of an orthotopic neobladder, five to six double-armed sutures were preplaced at the urethra for anastomosis with the robotic system. After the robot was undocked and all ports were removed, urinary diversion was performed by open surgery. A 5–6 cm lower midline laparotomy served for the retrieval of the cystectomy and lymph node specimens as well as for the creation of an ileal conduit or an orthotopic ileal neobladder. The neobladder–urethra anastomosis was made through this access using the robotically preplaced double-armed sutures that were tied after corresponding sutures through the neobladder.

Perioperative Care Plan

After a systematic literature research and in agreement with surgical, anaesthetic and nursing staff, a FT protocol was designed that differed in main principles from previously existing standards (Table 1). The key features of this perioperative care plan include an early oral nutrition within the first 24 h and the first mobilisation of the patient within a few hours after surgery. There was no more mechanical bowel preparation before surgery. A liquid diet <24 h and high carbohydrate drinks before surgery were introduced to shorten preoperative fasting. Avoidance of nasogastric tubes and abdominal drains after surgery was recommended. For perioperative pain management placement of epidural catheters (6 mg morphine/24 h) was done in both groups whenever possible and accepted by the patient. Additionally, a standardised non-opioid medication to reduce opioid consumption and an omission of regular postoperative antibiotics was carried out in the FT protocol. With the introduction of the FT regimen, patients received preoperative information about these new standards and their importance for postoperative recovery as part of the informed consent discussion for surgery.

Outcome Measures

The influence of an advanced perioperative recovery programme on postoperative results was evaluated with special regards to postoperative recovery. Therefore, time to first flatus, time to first bowel movement, time to first mobilisation and time to regular diet was measured for each patient. Opioid pain medication was administered in both groups upon patient request. Parenteral opioids were used when i.v. access was still present. Otherwise oral opioids were added. An epidural catheter was offered to all patients according to their eligibility and consent. Overall opioid consumption was documented and morphine equivalents were calculated in due consideration of different preparations during the first 5 days. Finally, the 30-day complication rate was recorded during hospital stay and after discharge according to the modified Clavien system including all readmissions.

Statistical Analysis

Comparisons were calculated for significance using the Mann–Whitney statistical test for quantitative measurements and the Fisher's exact test for results with two possible outcomes. All P-values were based on two-sided tests, and the threshold to accept statistical significance was set at the alpha level 0.05.

Results

Of the consecutive 63 patients treated by RALC, one patient within the first 32 RALCs required conversion to open cystectomy, presenting an advanced tumour with macroscopic infiltration of the intestine, not feasible for laparoscopic resection. Thus, 62 patients could be clinically evaluated for perioperative measurements and outcome. Table 2 summarises the demographic and clinical characteristics of the two cohorts. The first 31 patients were not treated according to our developed FT protocol whereas the following 31 patients did receive FT care. Nevertheless, some patients of the first group still had a few modifications in the perioperative standard of care compared with the open procedure (e.g. no postoperative nasogastric tube, no placement of a drain). Comparing both cohorts, no statistically significant differences were found for sex, mean age, body mass index (BMI), American Society of Anesthesiologists (ASA) score and number of previous abdominal surgeries.

Table 2. Patient demographics and clinical characteristics.
VariableNo FTFTP
Number of patients3131
Male, n27271
Female, n441
Mean (SD) (median):   
Age, years61.6 (12.6) (62)67.2 (10.2) (69)0.09
BMI, kg/m226.2 (4.9) (25.4)26.9 (4.6) (26.8)0.36
ASA score2.16 (0.58) (2)2.32 (0.60) (2)0.35
N (%):   
Patients with ASA score 1–223 (74.2)19 (61.3)0.42
Patients with previous abdominal surgery25 (80.6)23 (74.2)0.76
Diversion type:0.42
Ileal conduit19 (61.3)23 (74.2)
Neobladder12 (38.7)8 (25.8)

Perioperative and pathological data are shown in Table 3. The pT-stages were comparable in both cohorts; six patients with urothelial carcinoma in each group had a positive lymph node status.

Table 3. Perioperative and pathological data.
VariableNo FTFTP
  1. Hb, haemoglobin.
Number of patients3131 
Mean (SD):   
Operative duration, h6.92 (1.3)6.76 (1.3)0.49
Hb before surgery, g/dL13.0 (2)13.6 (2)0.29
EBL, mL424 (182)383 (203)0.45
Hb after surgery, g/dL11.2 (1.6)11.6 (1.7)0.37
LOS, days18.1 (6.3)18.0 (5.1)0.98
Median (range) LOS, days17 (7–36)16 (12–32)
N (%):   
Urothelial carcinoma26 (83.8)30 (96.7)0.2
≤pT19 (34.6)9 (30)
pT28 (30.8)10 (33.3)
pT36 (23.1)10 (33.3)
pT43 (11.5)1 (3.4)
N020 (76.9)24 (80)
≥N16 (23.1)6 (20)
Positive surgical margins01 (3.3)
Mean (SD) nodes removed14.8 (6.3)13.5 (6.6)0.39

In the FT group only two patients had an abdominal drain placed after continent urinary diversion (formation of an orthotopic neobladder) at the discretion of the surgeon, for the remaining 29 patients a drain was omitted. All nasogastric tubes were removed immediately at the end of surgery in the FT group. Both cohorts were comparable for operative parameters, e.g. mean operating room time, estimated blood loss (EBL), number of lymph nodes removed and postoperative haemoglobin level. Implementation of the FT protocol resulted in earlier patient mobilisation (17.5 vs 31.2 h; P < 0.001) and a faster return to a regular diet (4.0 vs 6.6 days; P < 0.002). An earlier return of bowel function as evidenced by a mean time to first flatus of 1.9 days compared with 2.4 days, and a mean time to first bowel movement of 2.6 days compared with 3.1 days, was seen in the FT cohort, although these differences were not statistically significance. Postoperative care according to the FT protocol was not associated with an increase of clinically symptomatic ileus. In both cohorts, two patients had clinical signs of an ileus that were treated with reinsertion of a nasogastric tube in the FT patients. Remarkably, patients in the FT cohort required fewer analgesics, calculated in milligrams of morphine equivalents (57.3 vs 92.4 mg; P = 0.01; Table 4). Half of the patients in both groups could receive epidural analgesia (13 vs 15 patients). As shown in Table 5, the two cohorts did neither differ significantly in the absolute number of complications (FT group 12; non-FT group 15; P = 0.6) nor in the number of major complications, as defined as Clavien ≥ 3 (9.7% vs 9.7%; P = 1). Major complications in the FT group were urinary leakage (two patients) and one young but multimorbid female patient died 11 days after surgery due to an uncontrolled epileptic attack, complicated by cardiac failure (after full mobilisation, 1 day before planned discharge). In the non-FT group leakage of the neobladder anastomosis (one patient), wound dehiscence (one), and hydronephrosis with the necessity of a conduit revision (one) were seen as major complications. The re-admission rate was not influenced by the introduction of the FT protocol in group 2 (two vs two patients).

Table 4. Postoperative recovery.
VariableNo FTFTP
Number of patients3131 
Intra-abdominal drainage, n (%)212<0.001
Mean (SD) time of intra-abdominal drainage, days4.12 (3.44)0.39 (1.53)<0.001
Nasogastric tubes, n (%)9 (29)00.002
Mean (SD) time of nasogastric tube, days0.71 (1.3)00.05
Reinsertion of nasogastric tubes, n (%)02 (6.4)0.49
Peridural catheter, n (%)15 (48)13 (42)0.80
Mean (SD):   
In-house analgesia (morphine equivalents), mg92.4 (62)57.3 (26)0.01
Postoperative hospital stay, days18.1 (6.3)18.0 (5.1)0.98
Time to mobilisation within the room, h31.2 (15.5)17.5 (10.2)<0.001
Time to mobilisation on the floor, days2.8 (1.3)2.3 (1.1)0.18
Time of i.v. fluids, days5.4 (3.3)4.7 (1.6)0.60
Time to regular diet, days6.6 (3.5)4.0 (2.2)<0.002
Time to flatus, days2.4 (1.3)1.9 (0.8)0.32
Time to bowel movement, days3.1 (1.4)2.6 (0.9)0.3
Readmission ≤ 30 days, n (%)6 (19.4)2 (6.4)0.26
Table 5. Patients with complications ≤ 30 days of RALC.
VariableNo FTFTP
Number of patients3131
N (%):   
Patients with complications15 (48.4)12 (38.7)0.61
Patients with major complications3 (9.7)3 (9.7)1
Grade of complication (Clavien):   
Grade 016 (51.6)19 (61.3)
Grade 14 (12.9)3 (9.7)
Grade 28 (25.8)6 (19.4)
Grade 33 (9.7)2 (6.4)
Grade 400
Grade 501 (3.2)

Discussion

Perioperative clinical care pathways with FT programmes have been successfully used in a wide variety of surgical procedures over the last decade [15, 19], leading to a faster patient recovery without increasing their risks of postoperative complications. For open radical cystectomy an enhanced recovery programme has already been tested and resulted in quicker postoperative convalescence with no effect on morbidity and mortality [20, 21]. To our knowledge, the present study analyses for the first time the feasibility and effects of a standardised perioperative FT regimen in patients receiving RALC.

The use of minimally invasive, laparoscopic techniques for cystectomy already results in less pain and postoperative pain medication [8, 11], in a shortened time of postoperative ileus and in a shorter hospital stay [18, 21] compared with open surgery. When we started performing RALC at our centre, we quickly realised that an adaptation of our perioperative care plans (originally designed for open cystectomy) was necessary to account for the benefits of minimally invasive surgery. As the minimally invasive approach itself led to more favourable patient outcomes in the early postoperative period, patients did not accept the old scheme any more. These considerations led to a standardised FT protocol as presented in the present study. Key features of this protocol include:

  1. minimisation of preoperative fasting by using high carbohydrate drinks to reduce postoperative insulin resistance [22],
  2. omission of mechanical bowel preparation as recent studies give no evidence supporting bowel preparation before ileal surgery [23-25],
  3. removal of nasogastric tubes at the end of surgery and re-starting oral intake on the first day after surgery and
  4. early patient mobilisation beginning on the day of surgery.

A reduction of systemic opioids was achieved by standard oral medication with diclofenac, which has been shown to reduce use of opioid analgesia after laparoscopy [26], and using perioperative epidural analgesia, to minimise opioid induced bowel atony. These features of the care plan may explain the low rate of gastrointestinal complications (6.4%) in this cohort compared with other series investigating FT regimens in open cystectomy with postoperative ileus in 16% [20] and 22.5% [21] or 12% gastrointestinal complications after RALC [9]. Pruthi and Wallen [18] already discussed that RALC itself is able to reduce time to flatus and time to bowel movement. In the present cohort, the addition of a FT regimen tended to further shorten the time for a regain of bowel function. As a consequence, earlier regular diet was possible (from 6.6 to 4 days), which was also seen in other studies applying FT regimen to patients after open cystectomy [20, 21].

A good indicator for patients' convenience is the need of analgesics. For RALC, Nix et al. [8] described an up to 40% reduction of in-house analgesia compared with the open surgical approach. Whereas the need for pain medication in morphine equivalents in the first group without FT was comparable with Nix et al. [8], the introduction of the FT regimen further drecreased the intake of analgesics by almost another 40%. Epidural analgesia is considered to be one important element of FT surgery, but could only be realised in half of our patients. Moreover, epidural analgesia was already part of our care programme before starting the FT regimen and therefore may not be responsible for the demonstrated benefits in the FT group. The remarkably low needs of pain medication in both groups certainly also contributed to the early regain of bowel function and may have reduced the impact of epidural analgesics in general.

Recent reports of improved outcomes with RALC have been criticised for selecting younger patients with fewer co-morbidities and no prior abdominal surgery [27, 28]. However, in the present study up to 80% of the patients had had prior abdominal surgery. Moreover, ASA scores and the mean age were higher in the FT group. A selection bias for less risk factors in the FT group seems therefore very unlikely. In fact the present results are similar to previously published data [29] showing that RALC appears to be well tolerated, independent of risk factors such as age, BMI, prior radiation, or ASA score.

Successful application of a FT programme was hypothesised to improve postoperative recovery via early mobilisation and quick return to oral diet, and may lead to a shorter length of stay (LOS). In previous studies patient discharge after 5–8 days [20, 30] and even as early as 3 days [31] has been reported. In the present cohort, the LOS remained at 18 days with or without the FT programme. The differences in national health care systems of the different countries most likely account for these imbalances and not patient or surgical factors. In Germany, early discharge requires consideration of short-term healthcare arrangements to bridge patients until rehabilitation. As the German reimbursement system does cover a longer hospital stay, most patients prefer to transfer directly from the hospital to rehabilitation without returning home first. For these reasons, the effectiveness of the FT programme was not reflected in the LOS in the present study so far. Nevertheless, an early discharge programme is under development in our clinic.

There are certainly important and noteworthy limitations in the present study. A major shortcoming is the non-randomised and single-centre design. Additionally, the study is also very prone to a selection bias, as we analysed a cohort of consecutive patients, where no matching was performed. As mentioned before, the development of a FT protocol was an evolutionary process. Therefore, single aspects of the FT regimen were already tested in the first 31 patients, leading to a less clear distinction between the two groups, as the second group of 31 patients received a standardised FT regimen according to the presented protocol. Patients treated in the present study represent the early experience with RALC at our institution and the learning curve of the surgeons may represent the biggest confounder in this study, when comparing both groups. Three surgeons performed RALC in our centre starting their first surgery at different time points during the study, with an average of 8–14 RALCs per year per surgeon. All three surgeons were already experienced with the Da Vinci® system (prostatectomy, partial nephrectomy) and although unexpected there were no differences in operating room time and EBL between the first and the second group, suggesting that case number is not necessarily associated with an increased incidence of intraoperative complications and blood loss as already investigated by Hayn et al. [32]. Nevertheless, the equally low complication rate in the FT group may be attributed to the gain of surgical experience. Moreover, not only surgeons improve over time, but also the whole medical team adapts to the special requirements of a new perioperative care plan, biasing the study results. Finally, the significant differences in time to mobilisation and time to full diet in the FT group just reflect the successful implementation for the FT protocol.

In conclusion advances in care were realised in the preoperative, surgical and postoperative period. Together these improvements have led to a more rapid return to regular diet, earlier mobilisation, lower opioid consumption and overall improved patient recovery, especially when compared with such outcomes as described few years ago not using such a regimen and an open surgical approach.

Conflict of Interest

None declared.

Abbreviations
ASA

American Society of Anesthesiologists

BMI

body mass index

EBL

estimated blood loss

FT

fast track

LOS

length of stay

RALC

robot-assisted laparoscopic cystectomy

Ancillary