Combined spinal and epidural anaesthesia for open radical cystectomy: A controlled study

To evaluate the feasibility of loco‐regional anaesthesia and to compare perioperative outcomes between loco‐regional and standard general anaesthesia in patients with bladder cancer undergoing open radical cystectomy (ORC).


| INTRODUCTION
Radical cystectomy (RC) with pelvic lymph node dissection and urinary diversion is the gold standard surgical treatment for patients with muscle-invasive and high-risk recurrent non-muscle-invasive bladder cancer. 1 RC is a highly complex surgical procedure associated with a significant risk for intraoperative and postoperative complications, prolonged length of stay (LOS) and perioperative mortality.Recent series have reported a rate of perioperative complications and perioperative mortality ranging from 11% to 68% and 3% to 7%, respectively. 2tempts at reducing morbidity and mortality have included refinements in patient selection, improvement in anaesthetic and surgical techniques, introduction of minimally invasive approaches, centralisation to highvolume centres and optimisation of perioperative care pathways. 3,4though robot-assisted radical cystectomy (RARC) has become more and more popular in the last decade, open radical cystectomy (ORC) is still considered the standard approach. 16][7] Moreover, RARC requires a prolonged Trendelenburg position that may pose anaesthesiology challenges especially in patients with a higher anaesthesiology risk.
Whereas RARC obligatorily requires general anaesthesia (GA), pure loco-regional anaesthesia has been proposed by some authors for patients undergoing ORC. 8 Spinal and/or epidural anaesthesia is, in fact, a valid alternative to standard GA, especially in elderly patients and/or in those with multiple comorbidities in whom reducing the risk for postoperative respiratory and/or cardiovascular complications is a relevant goal. 9In addition, loco-regional anaesthesia has been shown to provide an excellent muscle relaxation, facilitate the return of normal bowel function, reduce intraoperative blood loss and improve postoperative pain control by reducing or avoiding opioids. 10,11r all these reasons, loco-regional anaesthesia seems particularly indicated for ORC candidates.However, only very few studies have explored the feasibility of spinal and/or epidural anaesthesia in this setting, 8 and none has included a control group undergoing surgery with standard GA.
The objectives of this exploratory, controlled study were (1) to further evaluate the feasibility of combined spinal and epidural anaesthesia (SpEA) and (2) to compare perioperative outcomes between SpEA and standard GA in patients undergoing ORC.

| Patients
A prospectively maintained database for RC patients collecting clinical, oncological and functional data has been active at the University of Messina Urology Section since 1 May 2017.For the purpose of the present comparative, non-randomised study, 60 consecutive patients with bladder cancer scheduled for ORC from sMay 2020 were selected, and one every three patients was assigned to the study group.The study group, thus, included 15 patients undergoing surgery with SpEA, and the control group was composed by 45 patients being operated on under GA.Exclusion criteria were contraindications for SpEA, such as skin infection at the site of spinal and/or epidural puncture, severe coagulopathies and spinal disorders.
After discussion with an anaesthesiologist and a treating surgeon, patients agreed to participate in the study by signing an informed consent where they accepted the possibility to receive SpEA instead of GA and authorised data collection for scientific purposes.All patients had computed tomography of thorax, abdomen and pelvis for staging purpose.

| Anaesthesia techniques
ORC including extended pelvic lymph node dissection and urinary diversion (orthotopic ileal neobladder, retrosigmoid ileal conduit or cutaneous ureterostomy) was performed by a single expert surgeon via an infraumbilical incision according to previously described surgical techniques. 12,13th SpEA and GA were performed by two dedicated anaesthesiologists with extensive experience in both loco-regional anaesthesia and GA techniques.
Patients in the study group received pure SpEA.With patients in a sitting position, a deep thoracic epidural catheter was placed at Th10-11 or Th11-12 level through a 17-G Tuohy needle.Spinal anaesthesia was administered by puncturing the space between L2-3 and L3-4 with a 25-G Whitacre needle with introducer and infusing 3 ml levobupivacaine 0.5% (15 mg) and 10 μg fentanyl.During surgery, 5 ml ropivacaine 7.5% (37.5 mg) with 4 ml saline 0.9% and 50 μg fentanyl every 2-3 h were administered.Although patients breathed spontaneously during the entire surgical procedure, they received an escalating i.v.sedation with dexmedetomidine under continuous measurement of bispectral index to monitor its depth, with a target value of 65-55.Dexmedetomidine was infused at 0.7 μg/kg/h velocity with on-demand adjustments of +/À 0.1-0.2μg/kg/h.This enabled patients to sleep, to be responsive to verbal stimuli and to maintain a slow and shallow breathing.

| Postoperative management
Postoperative pain control was standardised.In the first 24 h, in the study group, 5-6 ml/h ropivacaine 2 mg/ml was infused through the epidural catheter, which was then removed.In the control group, continuous i.v.infusion of chlorinated tramadol 200 mg and ketorolac 60 mg 2 ml/h was administered.The i.m. morphine 1 mg/10 kg was given as a rescue medication if VAS score was >3.After 24 h, patients in both groups received i.v.acetaminophen 1 g three times a day and i.m. diclofenac 75 mg twice a day for further 3 days.The i.v.chlorinated tramadol 100 mg was given as a rescue medication if VAS score was >3.No intravenous patient-controlled analgesia was used.
An Enhanced Recovery after Surgery (ERAS) pathway was followed as previously described. 14Briefly, in addition to opioid-sparing anaesthesia and analgesia, the protocol included no preoperative dietary restrictions, no bowel preparation, no nasogastric tube, no postoperative stay in the intensive or intermediate care unit, no total parenteral nutrition, early nutritionist-guided oral diet and early mobilisation.Details are reported in Figure 1.Postoperative complications observed within 90 days from surgery were recorded and graded according to the Clavien-Dindo classification. 15Grade 1 or 2 complications were considered as minor, and Grade 3 to 5 complications were classified as major.Specific SpEA-related complications were also recorded.

| Data collection and study outcomes
F I G U R E 1 Details of the enhanced recovery pathway after open radical cystectomy adopted at our institution for patients receiving general anaesthesia

| Statistical analyses
The premises for the above study design were the scarcity of literature data on the role of loco-regional anaesthesia in patients undergoing RC with only very few feasibility studies, the lack of comparative controlled studies and the still exploratory nature of our analysis.The 1:3 assignment ratio was conceived to resemble a semi-randomised study, while limiting the number of patients in the study group receiving a still experimental anaesthesia technique with potential safety issues.For all these reasons, an RCT with a formal sample size calculation was deemed not justified at the current stage of the research.Furthermore, no formal distinction between primary and secondary study outcomes was made, under the assumption that all above reported variables might be of potential interest to correctly appraise the tested anaesthesia technique.
Parametric continuous variables were reported as means ± standard deviation, whereas median and interquartile range (IQR) were used for non-parametric continuous variables.Student's t-test, Mann-Whitney U test and Pearson's chi-square test were used to compare continuous parametric, non-parametric and categorical variables, respectively, as appropriate.All clinical records were inserted in a dedicated database, and data were analysed using SPSS v. 21.0 software (IBM Corp., Armonk, NY).All reported p values were two-sided, and statistical significance was set at p < 0.05.

| RESULTS
The study period terminated on 31 December 2021.No patients were excluded from the study as per exclusion criteria.No patients assigned to the study group refused the proposed anaesthesia technique after signing the informed consent.Moreover, there was no technical failure in spinal puncture or epidural catheter placement in the study group.Therefore, no patients assigned to the study group crossed over to the control one.The two groups were comparable for all demographic, clinical and pathological characteristics.Moreover, no differences were detected in terms of urinary diversion received (Table 1 patients in the study and in 39/45 (86.7%) patients in the control group (p = 0.13).Reasons for protocol non-compliance were failure of mobilisation on POD 1 and delayed start of solid oral diet beyond POD 2 in four and two patients, respectively.However, no significant differences between the two groups were detected with regard to return to oral diet, bowel function recovery and time to ambulation (Table 3).
Rate of 90-day postoperative complications was comparable in the two groups (p = 0.76).In detail, minor complications were detected in five (33.3%) patients in the study and in 14 (31.1%)patients in the control group.Major complications were observed in two (13.3%)patients in the study and in five (11.1%) patients in the control group, including one fatal event because of acute myocardial infarction on POD 65.Specifically, no SpEA-related complications (such as epidural hematoma or other bleeding complication) were recorded.

| DISCUSSION
Our comparative study showed that SpEA is feasible, safe and effective in patients undergoing ORC.No significant differences were observed between SpEA and GA in terms of perioperative parameters, except for shorter anaesthesia time and greater early pain control with SpEA.
Whereas several endoscopic procedures in the lower and upper urinary tract are usually performed under loco-regional anaesthesia, the routine use of spinal and/or epidural anaesthesia for oncological interventions in the pelvis or lower abdomen is exceptional.SpEA combines the advantages of spinal anaesthesia, with a rapid starting effect and limited effect on good motor blockade, with those of epidural analgesia, which enables an optimal postoperative pain control. 16,17SpEA may position itself as a viable alternative to standard GA, being particularly attractive for certain categories of patients.
For instance, in current practice, a substantial proportion of RC candidates are elderly and have multiple relevant comorbidities.In these patients, the use of SpEA instead of traditional GA might reduce the cardiovascular and respiratory stress, thus abolishing the need for postoperative intensive or semi-intensive monitoring and care, and decrease the risk of cardiopulmonary complications and neurological disorders.
Very few non-comparative studies have investigated the role of loco-regional anaesthesia in patients undergoing ORC.In 2013, Friedrich-Freksa et al. were the first to report on a case series of 28 patients receiving SpEA. 18In the same year, Karl et al. reported their experience with spinal and/or epidural anaesthesia in a smaller series of nine ORC cases. 19In 2015, Tzortzis et al. published a retrospective analysis of 18 octogenarians with ASA score ≥3 who underwent ORC under SpEA. 20Finally, in 2017, Gerullis et al.
reviewed the literature adding three further cases performed under epidural anaesthesia only. 8Overall, ORC under loco-regional anaesthesia was proved to be feasible, and no specific complications or safety issues were observed.However, it should still be considered in a development stage according to the IDEAL criteria. 8Our study is a step forward towards its clinical evaluation, being the first exploratory, controlled study comparing SpEA with standard of care GA.
Virtually all intraoperative and postoperative parameters were comparable with the two anaesthesia techniques, which is particularly reassuring considering that our population includes mostly elderly and comorbid patients.The only two significant differences were in favour of SpEA.First, median anaesthesia time was shorter by 40 min, which might be viewed as a valuable aid to the rationalisation of operating room resources, although an ad hoc cost analysis would be needed to fully assess its effect.Second, SpEA was associated with a significantly greater pain reduction in the early postoperative period compared with GA.This finding could represent a great advantage of the loco-regional technique, which could benefit even younger and fitter RC candidates.
Side effects of SpEA could be hypotension, bradycardia, cardiac arrhythmia, cerebral ischemia, cardiac arrest in rare cases, respiratory insufficiency due to high spinal anaesthesia, paraplegia because of epidural haematoma or abscess, motor blockade, nausea and vomiting, pruritus and postdural puncture headache. 21Notably, we observed none of them, thus confirming the safety of this anaesthesia technique applied to RC.
We acknowledge the following limitations to the present study.
First, the study group included only a limited number of patients.This was deliberately decided in view of the exploratory study design.Second, we have reported the experience of a single high-volume surgeon and two dedicated expert anaesthesiologists working at an academic tertiary referral institution.Our promising findings should, then, be confirmed by studies including multiple surgeons and anaesthesiologists.Third, only a single pain assessment in the early postoperative period was performed.However, we believe there would have been no differences at later time points since the epidural catheter was no longer in place.
In conclusion, our exploratory, controlled study confirmed the feasibility, safety and effectiveness of a pure loco-regional anaesthesia in patients undergoing ORC.Similar intra-and postoperative outcomes were observed when compared with current standard of care, GA.Of note, SpEA was associated with a significantly shorter anaesthesia time and greater pain reduction in the early postoperative period compared with GA.Our preliminary data warrant further exploration in larger multi-surgeon and multi-anaesthesiologist comparative studies.SpEA seems particularly suited for those RC candidates who are elderly or have multiple relevant comorbidities increasing their anaesthesiology risk for GA.
The following variables were prospectively collected: age on surgery, gender, body mass index, Charlson comorbidity index, haemoglobin and creatinine levels, clinical tumour and node stage according to the 2017 TNM staging system, American Society of Anesthesiologists (ASA) score, intraoperative colloid and crystalloid infusion, EBL, intraoperative red blood cell transfusion rate, anaesthesia time (including induction, surgery and recovery), surgery time (divided by RC and urinary diversion) and intraoperative complications.All surgical specimens were processed according to the standard protocol and reviewed by a dedicated uropathologist.The following parameters were assigned: cell type, tumour and nodal stage according to the 2017 TNM staging system, urethral and soft tissue surgical margin status and perineural and lymphovascular invasion.Pain level was assessed 24 h postoperatively by a rotating anaesthesiology resident not involved in the study team using a 10-point visual analogue scale (VAS) graduating 0 = no pain and 10 = worst pain.Number of patients requiring rescue medication and time to first rescue medication within the first postoperative 24 h and during the subsequent 3 days [postoperative day (POD) 1 to 4] were also recorded.Other postoperative parameters were oxygen saturation percentage (SpO2) in room air measured by pulse oximetry in holding area before readmission to the ward, ERAS compliance, nasogastric tube placement due to nausea/vomiting or ileus, time to return to oral diet, time to bowel function recovery (measured as resumption of bowel sounds, passage of flatus and passage of stool), time to ambulation and LOS.
). Demographic, clinical and pathological characteristics of the 60 patients included in the comparative analysis stratified by type of anaesthesia technique during open radical cystectomy Intraoperative variables for the 60 patients included in the comparative analysis stratified by type of anaesthesia technique during open radical cystectomy quate anaesthesia during surgery.Both abdominal wall and bowel were adequately relaxed during the entire procedure.Intraoperative vital signs of patients in both groups were stable.No patients in the study group showed hypotension as a potential side effect of the sympathetic blockade.Hypotension was detected and promptly managed only in one patient in the control group.No intraoperative Postoperative SpO2 was comparable in the two groups (p = 0.12).Pain VAS score 24 h after surgery was significantly lower in the study versus control group (p < 0.001) (Table3).Rescue medication within the first 24 h was required by no patients in the study versus three (6.7%)patients in the control group (time to first rescue medication: 3, 6 and 10 h, respectively).Rescue medication between T A B L E 1 (Continued) Abbreviations: ASA, American Society of Anesthesiologists; Cis, carcinoma in situ, IQR, interquartile range.T A B L E 2 T A B L E 3 Postoperative variables for the 60 patients included in the comparative analysis stratified by type of anaesthesia technique during open radical cystectomy