Eyeworld

EW CATARACT/IOL 48 February 2011 by Matt Young EyeWorld Contributing Editor Scoring method may help predict outcomes of cataract surgery R esearchers may have de- termined better measures for predicting visual out- comes after cataract sur- gery based on a new scoring methodology. "Newly developed and validated clinical prediction scores may assist physicians and patients in decision making about the expected out- comes and benefits of cataract sur- gery," according to a study by Emilio Perea-Milla, M.D., Unidad de Investigación, Hospital de Mar- bella, CIBER Epidemiología y Salud Pública, Málaga, Spain. Further, the scores could reduce "discrepancies between expectations and actual outcomes of cataract sur- gery," Dr. Perea-Milla reported. Data from 5,512 patients await- ing phacoemulsification at 17 hospi- tals in Spain formed the basis for Dr. Perea-Milla's analysis. The statistical analysis per- formed was described thoroughly in Dr. Perea-Milla's study, published online in August 2010 in Ophthal- mology. Scores for visual acuity (VA) ranged from 0-44, while visual func- tion (VF)-14 values ranged from 0- 24, and Dr. Perea-Milla explained their usefulness in a hypothetical case. "For instance, a 68-year-old pa- tient (weight for such age group: 6) with simple cataract (weight: 6), pre- operative VA between 0.2 and 0.4 (weight: 14), and neutral surgical complexity (weight: 8), the total score obtained by adding each weight for VA would be 34," Dr. Perea-Milla explained. "For the same patient, if the preintervention VF-14 was between 45 and 70 (weight: 7) and with a surgical complexity con- sidered as neutral (weight: 2), the total score for VF-14 would be 9." The scoring results suggested this patient is a good candidate for phacoemulsification. "In this case, the scores obtained for VA and the VF-14 are greater than the estab- lished cutoff point (≥26 and ≥6, re- spectively)," Dr. Perea-Milla noted. "This means that cataract extraction would be predicted to improve this patient's VA and VF-14 above the MCID [minimal clinically important difference] for such parameters." The scoring model isn't perfect by any means, although "both scores had a positive predictive value from 74% to 85%," Dr. Perea-Milla noted. "The predictive capacity of the score is incomplete," Dr. Perea-Milla acknowledged. "Visual perception is a complex and multidimensional function involving the presence of contrast sensitivity, visual phenom- ena (halo, glare), visual aberrations, visual quality, visual fields, and in- trinsic factors related to patient per- sonality and learning. These variables usually are not included in routinely collected data." Still, Dr. Perea-Milla was able to determine predictive variables for VA and VF-14 gain with fairly good ac- curacy. "The variables identified as predictors of greater VA postopera- tive improvement were lower base- line VA, age less than 75 years, and uncomplicated cataract," Dr. Perea- Milla reported. "In the VF-14, the factors predictive of improvement were a baseline VF-14 score less than 70 and better vision in the eye that underwent surgery. Neutral and pos- itive technical complexity was asso- ciated with a greater probability of increases in both the VA and VF-14." Therefore, despite the weak- nesses in the model, Dr. Perea-Milla suggested scoring could be helpful in a number of ways. "A clinical score could help improve patient se- lection, offer patients more compre- hensive counseling, and reduce the discrepancy between expectations and actual outcomes of cataract sur- gery," Dr. Perea-Milla reported. Although the study methodol- ogy is intricate, Dr. Perea-Milla re- ported that the score itself is an "easy-to-use, practical clinical score that provides a good prediction … of which patients are more likely to benefit from the intervention and cutoff points for decision makers with estimates of the probability of benefit." This is not the first study to try to identify factors linked to im- proved VA and VF-14 outcomes. Poorer baseline VA and VF as well as ocular comorbidity were found to be predictors in this study as well as previous research. "Another important new issue introduced into our models was the anticipated technical difficulty of cataract extraction for individual cases," Dr. Perea-Milla noted. Francis S. Mah, M.D., co-med- ical director, Charles T. Campbell Ophthalmic Microbiology Labora- tory, University of Pittsburgh School of Medicine, Pittsburgh, suggested that predicting visual acuity pre-op would be helpful. "It would be nice to know what the post-op vision is going to be af- terward," Dr. Mah said. Still, Dr. Mah said he thinks it would be more useful to develop a way to predict a complicated patient's vision after cataract surgery. "More importantly, the question would be in terms of patients with macular degeneration, glaucoma, or multiple ocular conditions, how much does cataract affect vision as opposed to the other issues?" Dr. Mah said. "If I'm counseling a pa- tient who has cataract and also mac- ular degeneration, I will tell that patient, 'I can't promise what your vision will be. I can only say the quality is going to be better.' It would be better to say that if a pa- tient has cataract surgery, his or her vision will change to X or is only going to change to Y." Currently, Dr. Mah said, there are tests that attempt to predict vi- sual outcomes for these complicated patients with cataracts, but there's nothing reliable, he said. EW Editors' note: Dr. Perea-Milla has no fi- nancial interests related to this study. Dr. Mah has no financial interests re- lated to his comments. Contact information Mah: 412-647-2211, mahfs@upmc.edu Perea-Milla: eperea@hcs.es

• Cover