Abstract
Purpose: :
To evaluate the impact of baseline factors on subsequent measures of visual field (VF) progression during long–term treatment for newly diagnosed, open–angle glaucoma.
Methods: :
607 patients with newly–diagnosed, open–angle glaucoma were randomly assigned to initial treatment with medications or with trabeculectomy, within the context of the Collaborative Initial Glaucoma Treatment Study (CIGTS). These patients underwent biannual examinations, in which a battery of standardized testing was performed, including Humphrey 24–2 full threshold VF tests. Repeated measures analyses were conducted using SAS Proc Mixed ("mixed") and Sas Proc Genmod ("GEE") to develop predictive models for VF outcomes, which were measured using the CIGTS VF score and the mean deviation (MD). Analyses included follow–up data extending through 7.5 years post–randomization.
Results: :
The interpretations of all main effects in the mixed model and most in the GEE model were conditioned by significant interactions. In the mixed model that addressed mean VF scores, patients with greater baseline VF loss (e.g., MD = –10) had better VF outcomes when treated surgically vs. medically (p=0.0003), whereas no such effect was seen for patients with minimal VF loss at baseline. Diabetics, however, showed the most VF loss over time when treated surgically (p=0.032). Blacks treated surgically fared worse than non–blacks (p=0.018), whereas race was not a predictive factor for VF loss in medically–treated patients. In the GEE model that addressed clinically substantial VF loss, black race had a significant main effect with no interactions. Blacks had a two–fold risk of VF loss vs. non–blacks (OR=2.12; 95% CI, 1.28, 3.51). Other than race, the interactions found were consistent with those reported above for the mixed model.
Conclusions: :
With increased follow–up of patients treated medically or surgically, the role played by numerous baseline factors in predicting subsequent VF loss is clarified. Initial treatment choice is important when it is considered in the context of initial VF severity. Blacks fared worse than non–blacks, especially blacks treated surgically. These results should be considered in the complicated milieu of deciding upon treatment for newly–diagnosed, open–angle glaucoma.
Keywords: clinical (human) or epidemiologic studies: risk factor assessment • clinical (human) or epidemiologic studies: treatment/prevention assessment/controlled clinical trials • visual fields