Factors affecting visual outcomes after small-incision phacoemulsification in diabetic patients

• Published in: Journal of cataract and refractive surgery Vol. 28; no. 8; p. 1364
• Main Authors: Somaiya, Mamta D, Burns, Jason D, Mintz, Roni, Warren, Robert E, Uchida, Tatsuo, Godley, Bernard F
• Format: Journal Article
• Language: English
• Published: United States 01-08-2002
• Subjects: Adult; African Americans; African Continental Ancestry Group; Aged; Aged, 80 and over; Cataract - complications; Cataract - ethnology; Cataract Extraction - methods; Control Groups; Diabetes Complications; Diabetes Mellitus - ethnology; European Continental Ancestry Group; Female; Hispanic Americans; Humans; Male; Middle Aged; Phacoemulsification - methods; Retrospective Studies; Treatment Outcome; Visual Acuity
• ISSN: 0886-3350
• DOI: 10.1016/S0886-3350(02)01319-6

To examine the factors affecting visual outcome after phacoemulsification and evaluate the use of preoperative visual potential in assessing the visual prognosis in diabetic patients.
Department of Ophthalmology, University of Texas Medical Branch, Galveston, Texas, USA.
In a retrospective chart review of 1345 consecutive patients who had uneventful small-incision phacoemulsification, operated eyes from 106 diabetic and 55 nondiabetic control patients were selected. Data on demographics, level of retinopathy, perioperative glycosylated hemoglobin (HbA(Ic)), surgical duration, preoperative best corrected visual acuity (BCVA), and visual potential were collected.
The age, sex, preoperative BCVA, and visual potential in the diabetic and control eyes were comparable. Throughout the postoperative period, BCVA was worse of the diabetic group. At 1 year, BCVA was 20/40 in 82.1% of the diabetic group and 94.7% of the control group (P =.01). The most important factors affecting postoperative BCVA included coexisting diabetes and preoperative level of retinopathy. No correlation was found between perioperative and postoperative BCVA. Diabetic patients were less likely than control patients to achieve a BCVA better than or equal to the preoperative visual potential at 4 years (hazard ratio 0.6; 95% confidence interval, 0.4-0.9; P =.011). Patients with nonproliferative diabetic retinopathy were nearly 5 times less likely (P =.023) and patients with proliferative diabetic retinopathy 30 times less likely (P <.0001) to achieve a postoperative BCVA of 20/40 than diabetic patients without retinopathy.
Although uneventful small-incision phacoemulsification improved visual acuity in diabetic patients, this group had an overall worse visual outcome than nondiabetic patients. The most important predictors of visual outcome were coexisting diabetes and the extent of preoperative retinopathy. Methods used to assess preoperative visual potential provided a reasonable estimate of postoperative BCVA in diabetic patients. Given the inverse association between the level of retinopathy and visual outcome, it may be better to perform cataract extraction in diabetic patients during earlier stages of retinopathy.