Mulvihill 1996.

Study characteristics
Methods	Study design: parallel‐group randomized controlled trial Number randomized: 20 participants total 10 participants in the PR group 10 participants in the SB group Exclusions after randomization: none reported Unit of analysis: individual (1 eye of each participant included) Number analyzed: 20 participants total 10 participants in the PR group 10 participants in the SB group Losses to follow‐up: none reported Power calculation: not reported
Participants	Country: Ireland Age: not reported Gender: not reported Inclusion criteria: participants diagnosed with RRD fulfilling certain criteria: single retinal break or a small group of breaks, not larger than 1 clock hour of retina in size break or breaks located within the superior 8 clock hours of retina absence of significant PVR absence of uncontrolled glaucoma person able to maintain postoperative head posturing, if required Exclusion criteria: not reported Equivalence of baseline characteristics: not reported
Interventions	Intervention 1: PR PR performed under local anesthesia in the operating theater in all cases. Following peribulbar anesthesia, the eye was softened using Honan's Balloon and the periorbital skin cleaned with 0.1% chlorhexidine and draped. The retinal breaks were identified, and gas then injected into the vitreous cavity 4 mm posterior to the limbus using a 30‐gauge needle. The gas used was either 0.6 mL of sulfur hexafluoride (SF6) or 0.3 mL of perfluoropropane (C3F8). Following rejection of gas, the patency of the central retinal artery was evaluated with the indirect ophthalmoscope. The retinal break was then sealed using either transconjunctival cryopexy or indirect (argon) laser. Cryopexy was used when a satisfactory reaction could not be obtained with laser. In some large bullous RDs the procedure was performed in 2 stages: the retina was allowed to flatten after gas injection and then the retinal break was sealed 1–2 days later. After gas injection, IOP was checked by applanation tonometry 3 and 6 hours later. Postoperatively the participants were instructed to posture for ≥ 16 hours per day and for ≥ 4 days so the retinal breaks were uppermost, tamponaded by the intraocular gas bubble. Topical antibiotic/steroid (betamethasone sodium phosphate plus neomycin sulfate) was administered for 4 weeks postoperatively.  Intervention 2: SB Conjunctiva and Tenon's capsule were opened and the subretinal fluid drained via an opening in the sclera to allow the retina to move back into position after injection of sterile air into the vitreous cavity. The retinal break was then sealed with trans‐scleral cryopexy. Finally, a hard silicone explant was sutured onto the sclera in the region of the retinal break to create a permanent scleral indentation. All SB procedures were performed under general anesthesia. Length of follow‐up: planned: not reported; actual: mean follow‐up 16.7 months (range 5–27 months) in the PR group and 16.0 months (range 8–23 months) in the SB group
Outcomes	Primary outcome: reattachment of the retina Secondary outcomes: visual acuity, recurrence of retinal detachment, and adverse events Intervals at which outcomes assessed: not reported
Identification	Not available
Notes	Type of report: published article Funding sources: not reported Disclosures of interest: not reported Study period: November 1991 to February 1994 Subgroup analyses: none reported Publication language: English
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Method of randomization not reported.
Allocation concealment (selection bias)	Unclear risk	Quote: "Participants were randomised by a closed envelope technique into one of two groups."
Masking of participants and personnel (performance bias)	Unclear risk	Masking of participants and personnel not reported, but surgeons could not have been masked.
Masking of outcome assessment (detection bias)	Unclear risk	Masking of outcome assessors not reported.
Incomplete outcome data (attrition bias) all outcomes	Low risk	All participants included and randomized were analyzed.
Selective reporting (reporting bias)	Unclear risk	Study protocol not available.
Other bias	Unclear risk	Funding source not reported.