EyeWorld is the official news magazine of the American Society of Cataract & Refractive Surgery.
Issue link: https://digital.eyeworld.org/i/137624
54 EW RESIDENTS June 2013 Review continued from page 52 ence was not significant at three months postoperatively (p=0.56). The mean flare intensity was not significantly different between the two groups (p>0.33), nor were changes in mean foveal thickness (p>0.44). Subgroup analysis comparing the patients with PXF and glaucoma to those with only PXF did not reveal significant differences in the outcome measures. The authors conclude that PXF is associated with a significantly greater percent endothelial cell loss after phacoemulsification irrespective of nuclear firmness (per EmeryLittle classification), operative factors, or whether the patient has PXF-associated glaucoma. Based on the lack of difference in flare intensity or postoperative foveal thickness between the two groups, they conclude that PXF does not increase the risk of postoperative inflammation. Though the authors find a difference in endothelial cell integrity and density between the two groups, they suggest that these differences are not visually significant given the comparable visual outcomes at three months. Before attributing the above difference to PXF endotheliopathy, a careful look at the preoperative and intraoperative metrics of the PFX and control groups is warranted. With the exception of two parame- ters, shallower ACD and poorer dilation, the other factors portray a consistent trend predicting greater corneal endothelial damage from phacoemulsification energy as the PXF patients were older (74.6 years vs. 72.0 years, p=0.07), had greater nuclear firmness (2.2 vs. 1.9, p=0.10), longer surgery time (7.9 min vs. 7.6 min, p=0.12), longer mean ultrasound time (43.1 vs. 42.6 s, p=0.39), greater mean ultrasound power (50.6% vs. 46.7%, p<0.08), and greater mean irrigation volume (64.4 vs. 59.4 mL, p<0.08). While the p-value for each of these parameters individually does not reach statistical significance, their relatively small amount (often ≤0.1) suggests the differences between the groups have often at most a 10% probability of having occurred through chance alone. A larger sample size may have demonstrated a statistically significant difference among these preoperative and intraoperative factors. Because the authors did not present, in this study, a multiple linear regression analysis to account for each of these cumulative factors, it is unclear if the greater ECC loss among PXF patients was primarily due to the PXF disease process itself, and not intraoperative issues. Curiously, the lead author did present a multiple linear regression analysis in a precursor to this study from 1996, where in a group of 859 eyes, such analysis suggested that nuclear firmness and irrigation volume were the only metrics that were statistically significant with ECC loss.10 Interestingly, total amount of ultrasound energy was not found to be a significant factor. Applying those results to the current study, increased nuclear firmness and increased irrigation volume among the PXF patient group suggest that the relative ECC loss may be less attributable to the PXF disease itself than the authors suggest. It would also be interesting to see if there was any correlation between the endothelial cell loss and the increase in CCT. Additionally, the Emery-Little classification is not as commonly used as the more widely accepted LOCS III scale. Finally, the authors did not report on the power of the study, and it is not immediately clear whether with only 36 eyes in each group, the size of the effect that the authors report is meaningful. Such reassessment of the analysis would be more aligned with prior studies reporting that PXF patients do not experience greater ECC loss than controls. In such a study by Wirbelauer et al11 with 25 patients in