We conducted the trial after obtaining approval from the Internal Review Board of the Medical University of Lublin (KE-0254/92/2018, chairman Professor M. Olejossy). We registered our study on the ClinicalTrials.gov site on 27/01/2021 with the number NCT04726878 before recruiting patients. Finally, we obtained written informed consent from each patient, and the study was conducted according to the tenets of the Declaration of Helsinki for medical research involving human subjects.
For eligibility, we assessed adults (≥ 18 to ≤ 80 years old) scheduled for single-side breast surgery due to cancer. We excluded patients unable to give informed consent, who had previously participated in the trial (the second breast or a reoperation on the same side), and who qualified for surgery on two breasts. We also disqualified patients with known coagulopathy, allergies to the studied drugs, depression, epilepsy, antidepressant drug treatment, usage of painkillers before surgery, and addiction to alcohol or recreational drugs.
One to four weeks before surgery, patients visited our preanesthetic clinic, where an attending anesthesiologist qualified them for anesthesia. An anesthesiologist identified other participants for our study. A meeting was held for the purpose of screening and affirming the patients’ willingness to participate in our trial. A day before surgery, an anesthesiologist who participated in the study discussed with each patient the potential risks and benefits of taking part in the trial. The patients then verified and signed their informed consents to participate in our study. Finally, the anesthesiologist presented and explained the QoR-40 form, the visual analog scale (VAS), and demonstrated the use of the patient-controlled analgesia (PCA) pump. All patients were informed that they could withdraw from the study at any time.
We anesthetized the patients participating in the study in a similar manner using fentanyl (Fentanyl, Polpharma S.A., Warszawa, Poland) and propofol (Propofol 1% Fresenius, Fresenius Kabi Deutschland GmbH, Bad Homburg, Germany) to induce general anesthesia. Then, an anesthesiologist inserted a laryngeal mask airway. If the risk of aspiration was high, the anesthesiologist secured the airway with an endotracheal tube. In this case, rocuronium and suxamethonium could be used. We maintained anesthesia with sevoflurane and fentanyl. We emerged patients from general anesthesia using oxygen, sugammadex, or neostigmine, as required. An anesthesiologist assessed the patient before transfer to the postoperative care unit, taking vital signs and applying the Richmond Agitation Sedation Scale.
After inducing general anesthesia, an anesthesiologist involved in the study opened a sealed envelope containing the patient’s allocation. We randomized patients into three groups—the ESP block group (ESP), the sham block group (SHAM), and the control group (CON). We continued general anesthesia in the CON group without modification. Participants were unaware of their allocations.
Patients in the ESP and SHAM groups were placed in the lateral position contralaterally for injections (Fig. 1). The anesthesiologist scanned the patient’s back to determine the injection site (Fig. 2). After preparing the injection field, the ESP block was performed at the level of T4, as shown in Figure 1. In the ESP group, we used a 0.375% solution of ropivacaine, 0.4 mL/kg, to a maximum of 40 mL administered on the unilateral side. In the SHAM group, we injected 0.4 mL/kg of normal saline into the ESP space up to 40 mL. After injections, we placed the patients supine to perform the surgery.
Analgesia and postoperative care
Approximately 30 minutes before the end of the surgery, the patient received oxycodone intravenously (IV) at a dose of 0.1 mg/kg. The analgesia regime also included IV Paracetamol, one gram every six hours. In the postoperative care unit, the anesthesiologist initiated the PCA with oxycodone, 1 mg per bolus, with a lockout period of five minutes. If pain exceeded 40 mm on the VAS, the attending nurse could administer a rescue dose of oxycodone (5 mg twice). Routine care included IV Ondansetron, 4 mg twice daily.
The primary outcome of our study was the result of the QoR-40. We also analyzed parts A and B of the survey. A higher score on the QoR-40 means better recovery following breast surgery. An anesthesiologist who was unaware of participants’ allocations assessed the QoR. We hypothesized that QoR-40 scores in the ESP group would be significantly higher than in the CON group.
Secondary outcomes included postoperative pain severity, opioid consumption, time to the first opioid demand, and treatment satisfaction. An attending nurse not directly involved in the study measured pain severity on the VAS at hours 2, 4, 8, 12, and 24 following surgery. We also assessed overall satisfaction with treatment. Satisfaction was presented on a Likert-type scale from one to five points (very poor, poor, moderate, good, and excellent). Higher scores indicate greater satisfaction. Moreover, we analyzed the impact of the surgery type—partial resection versus breast amputation—on the aforementioned goals.
We investigated the normality of the distribution for continuous variables with the Shapiro–Wilk test. We analyzed normally distributed parameters using an analysis of variance (ANOVA). These variables are presented as means with 95% confidence intervals. We used the Kruskal–Wallis test by ranks to compute parameters with non-normal distributions. If the Kruskal–Wallis test results showed statistical significance, the Bonferroni correction was applied. Then, a pairwise comparison was performed using the Mann–Whitney U test. These data are presented as medians and interquartile ranges (IQR). Qualitative parameters were compared with Fisher’s exact test. The time to the first demand for oxycodone with PCA was presented as the Kaplan-Meier curve. For this variable, we calculated statistics using the log-rank test. All measurements were performed using Statistica 13.1 software (StatSoft, Tulsa, OK, United States). Randomization was also generated with Statistica software’s random number generator by a team member who was not directly involved in recruiting, treating, and assessing patients.
A preliminary study was performed to assess the sample size. The study’s primary outcome was the quality of recovery measured with the QoR-40. We compared 14 patients, seven after ESP, and seven controls. The mean results of the QoR-40 were 185 after ESP and 172 in patients without any intervention. The calculated sample size was 11 individuals for each group, power 0.8, and alpha 0.05. Because three comparisons were necessary, we decided to randomize 75 participants into groups, with 25 individuals in each group.