Objective:
To evaluate the reproducibility and the prognostic utility of the Retina Society and Silicone Study Classification Systems in eyes after surgery for severe proliferative vitreoretinopathy (PVR).
Design:
Subgroup analysis of the Silicone Study—a randomized, multicentered, surgical trial.
Setting:
Community and university-based ophthalmology clinics.
Materials:
Three hundred forty eyes with preoperative and intraoperative evaluations using both systems of grading PVR (reproducibility study), and 287 eyes with preoperative and intraoperative evaluations using both systems of grading PVR and with a 24-month follow-up examination (prognosis study).
Interventions:
Vitrectomy for PVR with long-acting perfluoropropane gas or silicone oil as the intraocular tamponade.
Outcome Measures:
Retinal reattachment, visual acuity (≥5/200), intraocular pressure, corneal clarity, and the need for reoperation.
Results:
The reproducibility of the Silicone Study Classification System was 64% (type of contraction), 77% (number of clock hours), 67% (posterior PVR), 88% (anterior and posterior PVR), and 94% (anterior, posterior, and subretinal PVR). The reproducibility of the Retina Society Classification System was 99%. Using the Silicone Study Classification System, location of PVR predicted visual acuity (P=.004, χ2 test for trend) and hypotony (P=.03, χ2 test for trend). Using the Retina Society Classification System, the grade of PVR predicted only visual acuity (P=.01, χ2 test for trend). For eyes with anterior and posterior PVR, there was a decreasing trend in successful visual acuity outcome with increasing severity of PVR (from C-3 to D-3, P=.02, χ2 test for trend).
Conclusions:
Although the classification of PVR using the Silicone Study Classification System was not reproducible for the type of contraction or for posterior PVR, identification of the anteroposterior extent of the PVR was prognostic of visual acuity and hypotony at 24 months. The joint knowledge of the location of PVR (using the Silicone Study Classification System) and the tightness of the funnel for retinas with 9 to 12 clock hours involved by fixed folds (using the Retina Society Classification System) has prognostic utility for eyes with anterior and posterior PVR.