Gas tamponade choice in low-to-moderate complexity primary rhegmatogenous retinal detachment: propensity-matched comparisons of SF<sub>6</sub>, C<sub>2</sub>F<sub>6</sub> and C<sub>3</sub>F<sub>8</sub>.

Tzoumas, Nikolaos; Li, Xiaoman S; Yorston, David; Williamson, Thomas H; Laidlaw, David Alistair; Steel, David H W; BEAVRS and EURETINA VR Retinal Detachment Outcomes Group · Br J Ophthalmol · 2026

retrospective_cohort · Level III

Where this comes from

Abstract

To compare single-operation anatomical success, postoperative visual acuity, and complications after rhegmatogenous retinal detachment (RRD) repair using sulphur hexafluoride (SF<sub>6</sub>), perfluoroethane (C<sub>2</sub>F<sub>6</sub>), or perfluoropropane (C<sub>3</sub>F<sub>8</sub>). We analysed British and Eire Association of Vitreoretinal Surgeons-European Society of Retina Specialists (BEAVRS-EURETINA) registry data from 2012 to 2024. After exclusions, 7466 eyes contributed to anatomical and 5734 to visual analyses. Intraocular gas tamponades were compared using propensity-score matching and weighted regression adjustment. Equivalence was tested using prespecified margins of OR 0.80-1.25 for anatomical success and ±0.10 logMAR for postoperative visual acuity. Matched main-outcome cohorts had effective sample sizes up to 4155. For single-operation anatomical success, estimates favoured C<sub>2</sub>F<sub>6</sub> over SF<sub>6</sub> (OR 1.19; 95% CI 0.97 to 1.48; p=0.10), and C<sub>2</sub>F<sub>6</sub> over C<sub>3</sub>F<sub>8</sub> (OR 1.28; 1.01 to 1.65; p=0.048), although the CIs included effects close to the null. C<sub>3</sub>F<sub>8</sub> showed no anatomical advantage over SF<sub>6</sub> (OR 0.83; 0.58 to 1.13; p=0.35), with CIs compatible with possible harm. For postoperative visual acuity, C<sub>2</sub>F<sub>6</sub> and SF<sub>6</sub> were similar (MD -0.01 logMAR; -0.04 to 0.01; p=0.29), while C<sub>3</sub>F<sub>8</sub> was associated with worse acuity than C<sub>2</sub>F<sub>6</sub> (MD 0.10 logMAR worse; 0.07 to 0.14; p<0.001) and SF<sub>6</sub> (MD 0.09 logMAR worse; 0.04 to 0.13; p<0.001). Cystoid macular oedema (CMO) was more common with C<sub>3</sub>F<sub>8</sub> than SF<sub>6</sub> (OR 2.72; 1.13 to 6.57; p=0.026), while raised IOP was less likely with C<sub>2</sub>F<sub>6</sub> than C<sub>3</sub>F<sub>8</sub> (OR 0.41; 0.24 to 0.68; p<0.001). In matched low-to-moderate complexity RRD cohorts, C<sub>2</sub>F<sub>6</sub> and SF<sub>6</sub> were associated with similar postoperative visual acuity. C<sub>3</sub>F<sub>8</sub> was associated with worse postoperative visual acuity compared to both other gases, as well as increased risks of CMO compared to SF<sub>6</sub> and of raised IOP compared to C<sub>2</sub>F<sub>6</sub>. Anatomical estimates were consistent with a possible benefit for C<sub>2</sub>F<sub>6</sub> over the other gases, and indicated no clear advantage for C<sub>3</sub>F<sub>8</sub>.