We evaluated potential preoperative and early postoperative prognostic indicators for chronic intraocular pressure control in 38 consecutive glaucoma patients undergoing single-plate Molteno implantation. Six months following surgery 22 patients had an intraocular pressure < or = 18 mm Hg and were considered successes. Twelve patients had an intraocular pressure > 18 mm Hg and were considered failures. Factors such as age, race, sex, type of glaucoma, phakic status, eye treated, intraocular pressure, bleb elevation, or number of glaucoma medicines were not significantly related to postoperative intraocular pressure control (p > 0.05). Postoperatively a significant difference in intraocular pressure was not observed between success and failure groups until three (13.8 +/- 5.4 versus 20.8 +/- 6.4 mm Hg, p = 0.010) and six (11.2 +/- 3.3 mm Hg versus 21.2 +/- 1.2 mm Hg, p < 0.001) months. Approximately 50% of patients who had an intraocular pressure > 20 mm Hg and 75% of those who measured < 20 mm Hg at any given examination in the early postoperative period were controlled six months postoperatively. Patients controlled at six months maintained control for as long as 43 months postoperatively. This study indicates that in the early postoperative period after a single-plate Molteno implant some patients have an ocular hypertensive phase but may ultimately be controlled, whereas most patients with an intraocular pressure within the normal range maintain control long-term.

We evaluated characteristics of the optic disc in ocular hypertensive patients which might portend future glaucomatous damage. We included in this study 12 patients with an intraocular pressure > 21 mmHg who on follow-up showed signs of optic nerve head damage. Each of these patients were matched to an ocular hypertensive patient who remained stable for > 5 years. This study found that neither optic disc, neural rim or peripapillary halo areas, vein or artery diameter over non-atrophied or atrophied retinal areas, differed significantly between groups (p > 0.05, Wilcoxon signed rank test). However, peripapillary atrophy was greater in patients who showed signs of glaucomatous disc damage (1.757 +/- 0.36 mm2, p = 0.02, Wilcoxon signed rank test) compared to stable patients (1.064 +/- 0.79 mm2), although this finding was not significant after the Bonferroni correction. Patients with < 0.6 mm2 area of peripapillary atrophy (n = 6), however, did not suffer damage. This study suggests that a relationship between the extent of peripapillary atrophy and the chance of developing optic disc damage in patients with ocular hypertension deserves further study.

We evaluated 19 ocular hypertensive and 32 chronic open-angle glaucoma patients to determine the correlation of pattern discrimination field loss to known markers of glaucomatous damage on the visual field and the optic nerve head. This study found no statistical association of the findings on the pattern discrimination perimeter to the areas of the optic disc, peripapillary halo, peripapillary atrophy, neural rim, or cup/pallor discrepancy (P > 0.05). In addition, no statistical relationship was observed to the diameters of the largest vein or artery in the inferior- or superior-temporal quadrant adjacent to the optic disc (P > 0.05). Between pattern discrimination and automated perimetry no agreement was observed in any visual field cluster (Glaucoma Hemifield Test) greater than that expected by chance alone (P > 0.05). When both visual function tests disagreed, the proportion of abnormal diagnoses in any cluster did not differ statistically between tests in chronic open-angle glaucoma patients (P > 0.05). However, in ocular hypertensive patients a greater proportion of abnormal diagnoses was observed with pattern discrimination perimetry (P < 0.03). This study suggests that pattern discrimination perimetry appears to measure a different physiologic property of the retina than does automated perimetry.

We evaluated the effectiveness of using the Potential Acuity Meter (PAM) and automated perimetry to predict postoperative vision in 30 advanced glaucoma patients following combined cataract extraction and trabeculectomy. Using the binomial distribution at α = 0.05, PAM and automated perimetry individually were significantly useful in predicting postoperative vision of 20/40 or better. Automated perimetry, but not PAM, was useful in predicting worse than 20/40 vision. Used together, automated perimetry and PAM testing were significantly useful in predicting vision worse or better than 20/40.