This was a two-arm randomized controlled feasibility trial conducted at St Joseph’s Hospital, Hamilton, Ontario, Canada. The study was registered at clinicaltrials.gov with the registration number NCT03176667, and ethics approval was obtained from the Hamilton Integrated Research Ethics Board on July 18, 2017 (#3012). Due to the unavailability of rupivacaine at the start of the study, as well as input regarding epinephrine from participating anesthesiologists, patients undergoing ESP were administered 30 ml of 0.125% bupivacaine rather than 30 ml of 0.5% ropivacaine with 5 µg ml-1 of epinephrine. Patient enrollment began December 2018.
Participants
Patients were screened for trial inclusion using the preoperative booking list. Adult patients (>18 years) having elective unilateral VATS procedures with at least one overnight stay were included. Patients were excluded if they had one or more of the following: patient refusal; body mass index >40 kg m-2; contraindications to neuraxial analgesia techniques as per the American Society of Regional Anesthesia and Pain guidelines;21–23 known allergy to local anesthetics; inability to use PCA due to language or comprehension barriers; procedures with a higher chance of open surgery; and patient on any regular opioids for ≥3 months prior to surgery. During their pre-operative visit, a research assistant approached suitable patients, explained study procedures, and answered their questions. Patients had the opportunity to withdraw at any point. Baseline data, including demographics and history of comorbidities were collected at the same visit.
Randomization, allocation and blinding
Patients were randomized to intervention (US-guided continuous ESP block) and control (surgeon-performed ICNB) groups with a 1:1 allocation ratio using a computerized randomization the day before surgery. Randomization was generated by a statistician not involved with the study and the list was provided to the pharmacy. The group allocation was concealed and revealed to the anesthesiologist assigned to manage the surgical case the day before to ensure that a trained anesthesiologist was available to perform the pre-operative ESP procedure and catheter placement. Due to the nature of the study interventions, patients and research personnel were not blinded.
Intervention group
In the intervention group, patients had a pre-operatively performed US-guided ESP block with catheter placement in a designated block room equipped with monitoring and resuscitation facilities. Intravenous (i.v.) access was established and standard American Society of Anesthesiologists monitors were applied. Sedation and anxiolytics were used as considered appropriate. Patients were placed in a sitting position and the area was sterilized with disposable swabs of 2% chlorhexidine in 70% isopropyl alcohol and then draped in a sterile fashion. A high-frequency linear US transducer (GE LOGIQe, Wauwatosa, WI, USA) was used in a longitudinal parasagittal orientation 3 cm lateral to the T5 spinous process. The trapezius, rhomboid major, and erector spinae muscles were identified superficial to the tip of the T5 transverse process. The patient’s skin was anesthetized with 2% lidocaine. A B-Braun Contiplex Echo 18 gauge Tuohy needle with the Contiplex Echo 20 gauge catheter (Braun Medical Inc., Bethlehem, PA, USA) was inserted using an in-plane superior-to-inferior approach to place the tip into the fascial plane on the deep (anterior) aspect of erector spinae muscle. The location of the needle tip was confirmed by visible fluid spread lifting erector spinae muscle off the bony shadow of the transverse process. A maximum of 30 ml of 0.125% bupivacaine was injected in 5 ml aliquots through the needle (maximum of 3 mg kg-1) followed by insertion of a 19 gauge catheter under direct vision 5 cm beyond the needle tip.
Control group
In the control group, the thoracic surgeons performed ICNBs from T4 to T11 on the operated side of the chest, at the end of surgical procedure. The ICNBs were performed using 0.25% bupivacaine with epinephrine and a volume of 5 ml per block (maximum of 2.5 mg kg-1) was used.
Anesthesia management
The attending anesthesiologist provided a general anesthetic as per the routine institutional practice. For patients in the ESP block group, the attending anesthesiologist started a background infusion of bupivacaine 0.125% at 13 ml h-1 through the catheter after incision. At the end of the case a bolus of 0.125% bupivacaine 5 ml was injected through the ESP catheter and the patient was extubated. Patients in the control group had ICNBs placed at the end of surgery, then extubated. PCA was initiated postoperatively in the recovery room.
Postoperative management
All patients were monitored in the post-anesthetic care unit (PACU) for stabilization as routine practice. If the patient complained of incisional pain of intensity of >4/10 on the numerical rating scale (NRS), the nurse provided boluses of hydromorphone 0.2–0.4 mg i.v. every 7 min as needed to a maximum of 2 mg, after which PCA pumps were initiated. If the patient was allergic to hydromorphone, a morphine PCA was substituted. All patients were enrolled into the Acute Pain Service (APS) and were visited daily by a nurse practitioner and/or an anesthesiologist. In the ESP group, continuous ESP block was provided with a background infusion of 0.125% bupivacaine at 13 ml h-1.
The PCA pump was programmed to administer hydromorphone in boluses of 0.2 to 0.4 mg i.v., with lockout time of 7 to 10 min to a maximum 6 mg over 4 h; or morphine in boluses of 1 to 2 mg i.v., with a lockout time of 7 to 10 min to a maximum of 30 mg over 4 h. Changes to PCA opioid doses was made by the APS team, as necessitated for each patient to target minimal or tolerable pain (NRS <4). In both groups, patients had multimodal analgesia, individualized based on the allergies, comorbidities, and patient tolerance: 1) acetaminophen 975 mg orally/per rectum (PO/PR) for 48 h followed by 650 mg PO/PR every 4 h as needed until discharge; 2) naproxen 500 mg PO twice daily × eight doses (administer with food) or ketorolac 10 mg i.v. four times daily × eight doses if patient is unable to ingest PO medications. PCA was discontinued upon removal of the chest tubes or when patient was ready for discharge, as is the current standard of practice at our hospital. ESP catheter infusion was discontinued 48 h after surgery or under any one or more of the following circumstances: chest drain removal; patient discharge; any signs of local infection or systemic infection; or any sign of mechanical dysfunction including leaking or inadvertent withdrawal of the catheter.
Study outcomes
Primary outcomes (feasibility)
To assess the feasibility of conducting a large RCT comparing the ESP versus ICNB, the following outcomes were evaluated: 1) recruitment rate (number of participants recruited per week, and 2) proportion of in-hospital follow-up until discharge. The study was considered feasible if the recruitment rate was greater than two patients per week, and with complete (100%) in-hospital follow-up until discharge for the clinical outcomes.
Secondary outcomes (clinical)
Among the following clinical outcomes, differences in total opioid consumed over the first 24 h and 48 h were explored as primary outcomes for the larger RCT. We considered evaluating opioid use at both 24 and 48 h separately since a substantial percentage of patients were being discharged before 48 hours. For the main study, we plan to consider only one of the two as our primary outcome.
- Differences in total opioid consumption over the first 24 h and 48 h. This was collected using electronic medical records. Opioids will be considered as total hydromorphone equivalents for analysis.
- Difference in pain scores “at rest” and “at movement” at the following time points: 1 h after entry at PACU; average pain score during the postoperative nights as collected during the morning APS rounds; and pain scores on each morning until discharge. Pain scores at movement were collected by asking patients to cough. Both pain scores were recorded using the patient-reported NRS, an 11-point scale where 0 is no pain and 10 is the worst pain imaginable (Appendix 1). The NRS is validated and considered easy to use.24
- Adverse effects: Difference in the incidence of postoperative nausea-vomiting; respiratory depression; itching; local anesthetic toxicity; catheter leakage and catheter migration; and infection around the catheter site. The definitions and scales of each of the outcomes are provided in Appendix 1.
- Presence of sensory blockade on the operated side of the chest was collected on the first postoperative morning. Sensory assessment was performed by trained data collectors for pin prick (using a blunt needle). The dermatome distribution of the blockade on the patient’s anterior chest, mid-axillary line, mid-clavicular line and mid-scapular line was assessed using a blunt needle to test for loss of sensation to pinprick. Sensory testing was performed and recorded as shown in Appendix 1.
- Patient satisfaction with postoperative analgesia on the day of discharge. This was performed using a five-point Likert scale (Appendix 1).
Statistical analysis and Sample size
The analysis and reporting of the trial was performed in accordance with the Consolidated Standards of Reporting Trials (CONSORT) guidelines for pilot studies.25 Baseline and intraoperative data were reported using means and standard deviations or median and interquartile range as appropriate. We analyzed as intent to treat (ITT) with each patient analyzed according to the group to which they were randomized. Clinical outcomes were compared using a two-sided test with a significance level of 0.05. No subgroup tests were conducted. As suggested in literature for pilot studies, we considered a sample size of 12 patients per group to demonstrate feasibility.25,26
Data collection and confidentiality
Patients were noted using a study ID to keep their information anonymous. The patient’s study ID and corresponding hospital medical record number were recorded in an Excel file in a password-encrypted USB that was securely stored in a locked cabinet. Data collection was done using paper forms and transferred to REDCap for secure storage and analysis. Data were collected by research assistants and the APS nurse practitioner (KD). The data collection forms were also kept in the locked cabinet.