Gas Versus Air Tamponade in Surgery for Full-Thickness Macular Hole: A Systematic Review and Meta-analysis

Gas versus room air tamponade after pars plana vitrectomy for full-thickness macular hole.

Room air is increasingly used instead of long-acting fluorinated gas after macular hole surgery, but the comparative evidence has not previously been synthesized. Clarifying whether gas improves anatomical closure or visual outcomes is important because room air may reduce tamponade duration and postoperative burden.

MEDLINE, Embase, CENTRAL, and Web of Science were searched through August 7, 2026, without language restriction (PROSPERO: CRD420261474491). Randomized and nonrandomized studies comparing intraocular gas with room air, with at least 10 eyes per arm, were included. Studies were screened and data extracted in duplicate. Random-effects meta-analysis with Hartung-Knapp adjustment was performed, with randomized and nonrandomized evidence pooled separately. Primary outcomes were single-operation anatomical closure and best-corrected visual acuity at the longest available follow-up. Certainty was assessed using GRADE.

Fourteen studies involving 1585 eyes were included, comprising only 3 randomized trials, 2 of which were at high risk of bias, and 11 nonrandomized studies. Among 10 studies reporting primary closure in primary idiopathic holes, gas was not associated with significantly greater closure than room air (RR, 1.07; 95% CI, 0.95-1.19; P=.23; I2=76.7%). Results were similar in randomized trials (RR, 1.16; 95% CI, 0.85-1.58) and nonrandomized studies (RR, 1.02; 95% CI, 0.88-1.18; P for interaction=.18). Neither gas species, mean hole diameter, surgical era, internal limiting membrane technique, combined phacoemulsification, centre or surgeon number, nor timing of closure assessment explained the heterogeneity. Visual acuity did not differ (MD, -0.01 logMAR; 95% CI, -0.29 to 0.27). Certainty was very low for every outcome.

Current evidence, of very low certainty throughout and resting on only 3 randomized trials, does not demonstrate superior anatomical or visual outcomes with fluorinated gas over room air after macular hole surgery. Confidence intervals remain compatible with a clinically important benefit of gas, the one trial powered for noninferiority did not establish it for air in holes of 400 micrometers or smaller, and the only stratified data above 520 micrometers favored gas. Randomized trials in larger holes, with posturing standardized between arms, are needed.