Reply




We thank Carifi and associates for their interest in our paper, wherein we had compared the visual outcomes of 794 laser cataract cases with 420 controls by 2 surgeons.


To make cases and controls comparable, outliers with respect to preoperative cylinder >1.5 diopters and axial length <22.0 mm and >30.5 mm were sequentially removed. We had stated that the distribution of intraocular lens (IOL) types was similar. Repeated data analysis confirmed no statistical difference in distribution between laser cataract cases and controls for the 4 classes of IOL types (cases: monofocal = 262, monofocal toric = 165, multifocal = 108, multifocal toric = 102; controls: monofocal = 141, monofocal toric = 109, multifocal = 62, multifocal toric = 46; Pearson χ 2 , P = .327). Hence, we deemed it reasonable to present visual outcomes based on all IOLs. The unaided visual acuity (UAVA) of 20/25 or better was higher in laser cataract cases (cases = 68.9%, controls = 56.4%; P < .0001). Analysis of visual outcomes based only on the same standard monofocal IOL would have made the sample size too small for meaningful statistical analysis.


A re-review of visual results by a clinically relevant regrouping of IOLs—monofocals (plain = 427 and toric = 250) and multifocals (plain = 210 and toric = 108)—also showed laser cataract patients fared better (monofocal [UAVA 20/32 or better] Pearson χ 2 , P = .043; multifocal [UAVA 20/25 or better] Pearson χ 2 , P = .029).


Our paper on all IOLs stated there was no difference if UAVA comparison was at the lesser benchmark of 20/32. This suggested that laser cataract cases received an incremental visual benefit, which could be advantageous when implanting multifocal and toric implants. For the average patient with less visually demanding needs, this benefit may not be subjectively appreciated. The marginally superior UAVA in laser cataract cases could be attributed to smaller manifest residual spherical equivalent (MRSE), although mean absolute error (MAE) was not different. Further studies are required to provide a better understanding of this finding. Miháltz and associates had reported a reduction on postoperative visual aberrations in laser cataract eyes. IOL decentration and tilt may have contributed to poorer vision in controls, but our study was not designed to address this issue.


In response to the suggestion that ultrasound biometry eyes be excluded owing to other optical media abnormalities affecting postoperative vision, we deemed this exclusion unnecessary, since all eyes entered for visual outcome analysis were normal except for cataract (no cornea haze/scars, no asteroid hyalosis).


We had made all attempts to compare cases to a similar group of controls with respect to surgeons, axial length, and astigmatism. Cases and controls were considered as unpaired samples for the following reasons: (1) they did not come from the same patient; and (2) the cases and controls came from nonidentical time points. Controls included cases from a retrospective database, with a longer time interval for random sampling to reduce risk of selection bias.


The Mann-Whitney U test was employed for MAE and MRSE as our samples did not follow a normal distribution based on skewness and kurtosis (>±1.0).

Only gold members can continue reading. Log In or Register to continue

Jan 7, 2017 | Posted by in OPHTHALMOLOGY | Comments Off on Reply
Premium Wordpress Themes by UFO Themes