Sprung 2006.

Methods	Single‐blinded (outcome assessor), randomized clinical trial Sequence generation via computer‐generated list Follow‐up: 3 months
Participants	Participants: 89 women from a university hospital in Minnesota, USA Operation: elective vaginal hysterectomy (with or without repair of cystocoele and rectocoele) 2 groups, size: 45/44 Age (± SD), group 1, 2: 52.2 (± 11.9), 51.8 (± 12.8) All female participants Comorbidities: postmenopausal, group 1, 2: 21/17. Procedure, group 1, 2: hysterectomy only 27/27, hysterectomy + cystocoele 1/1, hysterectomy + rectocoele 4/4, hysterectomy + cystocoele + rectocoele 13/7
Interventions	Group 1 (regional): sedation with IV midazolam and propofol. SAB performed in lumbar region between 3rd and 5th vertebral bodies. After cerebrospinal fluid free flow, 0.75% hyperbaric bupivacaine (15 mg), preservative‐free clonidine (1 µg/kg), morphine (2 µg/kg, max 200 µg) injected to subarachnoid space. Intraoperative sedation with IV midazolam and propofol as needed. No intraoperative IV opioids. 30 mg ketorolac IV at end of surgery. On floor IV PCA 1.0 mg every 10 min with 4‐h lock out max of 15 mg in regional group (lower than general group, to decrease likelihood of delayed respiratory depression). Additional IV morphine per attending physician as needed Group 2 (general): 2 µg/kg fentanyl after pre‐oxygenation GA with sodium thiopental, succinylcholine, vecuronium bromide, isoflurane and 50% inspired nitrous oxide. A morphine sulphate 0.1 mg/kg IV in divided doses, no additional morphine was allowed. All participants received 30 mg IV ketoralac at end of surgery. On floor IV PCA 1.0 mg every 10 min, 4‐h lockout max of 30 mg Both groups: in PACU 2 mg IV morphine every 5‐10 min as needed for NRS > 3. On floor, morphine PCA, with differences in maximum noted above. Scheduled ketorolac 30 mg IM every 8 h until oral D3. After 24 h, IV PCA stopped and oral paracetamol and codeine (650 mg/30 mg) every 6 h as needed. In both groups, pruritis managed with diphenhydramine then naloxone if needed. Nausea/vomiting managed with droperidol, if later stages ondansetron, then naloxone if persisted. Adjuvants: clonidine (into subarachnoid space) Immediate post‐op pain control: significantly improved, significantly reduced analgesic consumption
Outcomes	Dichotomous: none Continuous: NRS at 3 months, SF‐36 pain subcomponent at 3 months Secondary: none Effective regional anaesthesia: reported. "Confirmation of an adequate dermatomal level of blockade" Adverse events reported on included use of intraoperative pressors, nausea/vomiting, pruritis
Notes	We acknowledge the study author's clarification on blinding methods. Funding sources: "intramural grant from the Mayo Foundation." Conflicts of interest: "none declared."
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: "computer‐generated list"
Allocation concealment (selection bias)	Low risk	Quote: "patients were randomized...using a sealed envelope"
Blinding of participants and personnel (performance bias) All outcomes	High risk	The anaesthesiologist, participants and providers were not blinded. This is acceptable for our purposes.
Blinding of outcome assessment (detection bias) All outcomes	Low risk	SF‐36 was filled out by participant and mailed in at 12 weeks. Study author contacted, stated the research co‐ordinator performing telephone follow‐up "was blinded regarding the study group"
Incomplete outcome data (attrition bias) All outcomes	Low risk	Quote. "in three patients in the SAB group, the block failed and the patient received general anesthesia. For all analyses presented in this report these patients are included in the SAB group (intention‐to‐treat)". Fairly balanced, low rate of participants lost to follow‐up at 12‐week follow‐up.
Selective reporting (reporting bias)	Low risk	All primary outcomes fully reported on.
Null bias	Low risk	Quote: "the patients in the general anesthesia group received more morphine in the PACU... compared to patients receiving SAB" and this continued into the 12 hours after PACU discharge. Numerical pain score values tended to be lower in participants receiving SAB compared to the general anesthesia group through 14:00 hr on postoperative day two (the day after surgery), with significant differences noted at the time of floor arrival and at 14:00 hr on postoperative day two"