Gacio 2016.

Methods	Triple‐blind (participant, provider, outcome assessor), clinical RCT Sequence generation was randomized but not described Follow‐up: 6 months
Participants	Participants: 80 participants at a university hospital in Portugal Operation: lumpectomy with axillary dissection, modified radical mastectomy (MRM), and mastectomy with or without axillary dissection 2 groups, size: 40/40 Age (± SD), group 1, 2: 55.10 (9.8), 52.68 (8.9) All women Exclusion criteria: allergy to NSAIDs, LAs, propofol, opioids, paracetamol, or antiemetics, participants on chronic treatment with antibiotics, obesity (BMI > 30), bilateral or multiple surgical procedures, contraindication to PVB (including coagulation disorders/anatomical changes), severe respiratory disease, pregnancy, inability to understand the VAS
Interventions	Group 1 (ropivacaine PVB): before the induction of anaesthesia, peripheral route catheterization was performed, and participants were monitored according to ASA standards and bispectral index (BIS) anaesthetic depth. PVB was performed with single‐injection, according to the classic technique at the T4 level with Tuohy needle 18 G, with 0.5% ropivacaine + adrenaline 3 g/mL, with a volume of 0.3 mL/kg (maximum total volume of 30 mL). Subsequently, anaesthesia was induced with propofol (1.5 mg kg−1 h−1) and fentanyl (2 g kg−1) and LMA was inserted. Anaesthesia was induced with propofol (1.5 mg kg−1 h−1) and fentanyl (2g kg−1) and LMA was inserted. The maintenance of anaesthesia was performed in both groups with desflurane to maintain BIS values at 45‐60 with a mixture of O2/air. Both groups received parecoxib 40 mg IV before the start of surgery. During maintenance, fentanyl (1.5 g kg−1) was administered if there was an increase of 20% from baseline values of mean arterial pressure (MAP) and heart rate (HR). For maintenance of haemodynamic stability, ephedrine or atropine was administered, at the anaesthesiologist’s discretion, if verified a decreased in MAP > 20% or HR < 50 beats/min of baseline values. The institutional protocol for the prevention of nausea and vomiting was administered, according to the predictive model by Apfel and colleagues, with three antiemetic intervention lines. At the end of surgery, PCA with morphine was initiated, programmed with bolus of 2 mg on demand and 5 min lock‐out and a maximum dose of 6 mg h−1 during the first 24 h postoperatively. Group 2 (general anaesthesia): same anaesthetic technique as above but no PVB was administered Adjuvants: parecoxib, fentanyl, morphine, and adrenaline Immediate post‐op pain control: significantly improved
Outcomes	Dichotomous: pain vs no pain Continuous: none Other reported: anxiety was assessed using the Hospital Anxiety and Depression scale (HADS), pain at rest according to the VAS score (0‐10), as well as pain with mobilization of the ipsilateral arm interpreted as 90° arm abduction 0 h, 1 h, 6 h, and 24 h after surgery, postoperative nausea and vomiting at 24 hours after surgery
Notes	Pain defined as DN4 score > 4 We acknowledge the study author's response regarding blinding and randomization technique. Funding sources: funding for the study was not described. Conflicts of interest: "the authors declare no conflicts of interest."
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	The study author responded, quote: "a stratified randomization was performed using Excel software for that purpose."
Allocation concealment (selection bias)	Low risk	The study author responded, quote: " in this study the anesthesiologist who proceeded to the technique became aware of the randomization sequence (in groups of 4 patients) the same day of the procedure."
Blinding of participants and personnel (performance bias) All outcomes	Low risk	The study author responded, quote: " the surgical team did not know the group to which the patient belongs." However, "In the first part of the study (assessment of acute pain in the peri‐operative and up to the first 24 hours) the anesthesiologist who proceeded to the technique knew in which group the patient was."
Blinding of outcome assessment (detection bias) All outcomes	Low risk	The study author responded, quote: "the investigator who interviewed the patients and carried out the records in the peri‐operative period.did not know the group to which the patient belongs."
Incomplete outcome data (attrition bias) All outcomes	High risk	14 participants were not included in the final analysis
Selective reporting (reporting bias)	Low risk	No subgroup analysis was performed
Null bias	Low risk	"The Visual Analog Scale (VAS) values of paravertebral group at rest were lower throughout the 24 h of study"