To the Editor:

We read with great interest the paper by Loiudice et al. where the authors report that localized intraoperative laser retinopexy (ILR) is as effective as 360° ILR in cases of retinal detachment repair using pars plana vitrectomy (PPV) [1]. In particular, the difference in single surgery anatomic success (SSAS) rates in the localized ILR group (86.66%) and the 360° ILR group (89.58%) was not statistically significant (P = 0.46).

Recently, results from retrospective studies on the topic have been conflicting. Our retrospective study demonstrated a 75% reduction in the odds of retinal redetachment with 360° ILR [2]. Other studies like the Primary Retinal Detachment Outcomes (PRO) Study Number 4 showed no improved surgical outcomes with this technique [3]. Given the need for well-designed prospective trials to address this question, we commend the authors for their study. However, we would like to raise an important point regarding their findings.

We estimate that the study design assumption was inaccurate and resulted in a large underestimation of the sample size required to detect a difference between two independent proportions (SSAS rates for each group) to reach the defined 80% power using a 5% significance level. In this study, the SSAS rates were also better than anticipated and given the small sample size, the conclusion should have reflected that the main outcome measure could not statistically validate a difference in efficacy between the two techniques. The conclusion “Localized laser resulted to be as effective as 360° laser application” is inaccurate because failure to find a statistically significant difference is not the same as establishing clinical equivalence.

For future superiority prospective trials, we believe that the study assumption must be that SSAS rates in the 360° ILR group would have a raw treatment effect in the range of 8–10%. We base this estimate on the results of the phakic [4] and pseudophakic [5] PRO studies that reported an 8% higher SSAS rate with combined PPV/scleral buckling compared to PPV alone. We believe that such an assumption is more adequate than the one used by Loiudice et al., since it is based on real-world evidence on the additional benefits that can be obtained with combined PPV procedures. Recalculating the sample size with this assumption shows the need for 536 eyes, much more than the study’s 93 eyes.