We thank the correspondents for their thoughtful comments on our randomised controlled trial examining the influence of subconjunctival anaesthesia duration on pain perception during intravitreal injections (IVI) 1, 2. Regarding performance bias and blinding, we agree that double blinding is optimal for randomised trials. However, it is difficult to implement in high-throughput IVI clinics and may deviate from routine clinical practice. To minimise performance bias, injection technique (needle gauge, injection site, anaesthetic volume and procedural steps) and injector–patient interactions were standardised. Communication during the procedure was limited to brief scripted cues required for the injection. The participants completed the questionnaire outside the procedure room with a nurse unaware of the allocated wait-time, ensuring that the responses were collected independently of the injector. Although full double blinding with third-party timing would be ideal, the standardised protocol and blinded outcome assessment make meaningful performance bias unlikely. Regarding the baseline differences in the number of prior injections, and previous IVI experiences impacting on the pain perception, it was prespecified as a covariate. ANCOVA confirmed that prior injection category did not differ significantly between wait-time groups (one-way ANOVA F = 1.71, p = 0.166). When included with wait time and laterality, prior injections were not associated with pain scores (β = −0.026, p = 0.745). Adjustment altered mean VAS estimates by only 0.03 points on average (2 min: 2.27 vs. 2.24; 3 min: 1.03 vs. 1.02; 4 min: 0.67 vs. 0.65; 5 min: 0.58 vs. 0.64). Differences between 2 min and longer wait times remained large and highly significant (2 vs. 3 min: −1.22, p 15 injections had corresponding scores of 2.36, 0.97, 0.56 and 0.67. In both groups pain decreased with increasing wait time, with a marked reduction between 2 and 3 min followed by a plateau, indicating that the analgesic benefit of waiting ≥ 3 min is consistent across prior-injection subgroups. Studying only first injections would raise ethical concerns and introduce confounding due to the lack of prior experience, although future questionnaires comparing prior and current injection experiences may provide useful insight. Laterality was examined in univariate and multivariable analyses. In the full ANCOVA model, left eyes had slightly lower pain scores than right eyes (β = −0.40, p = 0.044). Including laterality only modestly improved model fit (adjusted R2 0.163 vs. 0.152; Δ ≈ 1.1%) and did not alter the effect of wait time, which remained the dominant predictor (F(3,234) = 14.71, p < 0.001). To address imbalance in right versus left eye injections, we examined wait-time effects stratified by laterality. Among right eyes, mean pain scores were 2.82, 1.09, 0.58 and 0.83 for 2, 3, 4 and 5 min, with reductions of −1.80 (p < 0.001), −2.23 (p < 0.001) and −2.03 (p < 0.001) VAS points for 3, 4 and 5 min compared with 2 min. Among left eyes, means were 1.54, 0.96, 0.76 and 0.43, with reductions of −0.63 (p = 0.071), −0.82 (p = 0.017) and −1.08 (p = 0.001). The direction and pattern of the wait-time effect were consistent in both eyes. In summary, our trial addressed a practical question—the optimal wait time after subconjunctival anaesthesia in routine practice—using a pragmatic and standardised protocol in real life setting. Additional ANCOVA and subgroup analyses demonstrate that prior injection experience and laterality do not meaningfully influence the estimated effect of wait time. These findings still support a minimum 3-min interval between subconjunctival anaesthesia and IVI to achieve a clinically meaningful reduction in patient-reported pain without disrupting clinic workflow. The authors have nothing to report. The authors declare no conflicts of interest. The data that support the findings of this study are available from the corresponding author upon reasonable request.
Kim et al. (Tue,) studied this question.