We read with great interest the recent article from Chee and associates, who reported on outcomes of femtosecond laser–assisted cataract surgery.
The authors claimed that their study was the first to highlight better visual outcomes (in terms of unaided visual acuity) following laser-assisted cataract surgery; however, a possible explanation for their findings was not provided. In detail, the corrected visual acuity (CDVA) was similar among groups, and the mean absolute error was also similar. Differences were noted instead in regard to manifest postoperative spherical equivalent.
If refractive accuracy in both groups was similar, and no differences were observed in CDVA, why should the unaided distance visual acuity be statistically better in the laser-assisted group? It must be highlighted that different intraocular lens (IOL) designs were adopted, and a (large?) proportion of eyes were implanted with toric IOLs, multifocal IOLs, and toric-multifocal IOLs. We believe that this alone could be a possible source of bias and we would challenge the authors to reanalyze the data, excluding from analysis all eyes implanted with “premium” IOLs (including eyes with the same standard monofocal IOL). Also, eyes with preoperative ultrasound biometry might need to be excluded, as they might have other optical media abnormalities apart from lens opacities that would affect postoperative vision.
In addition, it would appear not appropriate to employ the Mann-Whitney U test for statistical analysis in this study: the eyes included in the control group were matched for biometric and patient characteristics, and were operated on by the same surgeons. Therefore, the 2 groups should be considered as paired samples and appropriate statistics employed.