Abstract
Purpose :
Retinal detachment (RD) complicated by proliferative vitreoretinopathy (PVR) is associated with an increased risk of recurrent RD. We hypothesized that features of vitreoretinal surgery may be associated with an increased risk of recurrence, such as longer operating times and more extensive use of laser retinopexy. The identification of such risk factors may help to stratify cases of RD with PVR with regard to postoperative single-surgery success and affect surveillance.
Methods :
This study analyzed 104 consecutive surgical repairs by a single surgeon for RD with PVR. The primary outcome measure was recurrence of RD. The variables analyzed were grade of proliferative vitreoretinopathy, vitrectomy probe cut time, number of laser spots, total laser energy, and lens status. A multinomial logistic regression analysis model was generated. A P value < 0.05 was considered significant.
Results :
The vast majority of cases were categorized as grade C PVR (88.5%), while the remainder were split between Grade A (2.9%) and Grade B (8.7%). In the multinomial logistic regression model, the only statistically significant covariate was lens status (P < 0.001). Of note, grade of PVR, number of laser spots, total laser energy, and vitrectomy cut time were not statistically significant variables in the regression model.
Conclusions :
Recurrent RD in the setting of pre-existing PVR may be most significantly associated with preoperative clinical parameters. While the majority of the included cases were Grade C PVR, PVR grade was not a statistically significant parameter in the regression model. Furthermore, the potential surrogate variables for the complexity of the surgical repair (vitrectomy cut time, number of laser spots, and total laser energy) were not significant covariates in the regression model. However, patient lens status may be a significant prognostic variable, with phakia being associated with a greater risk of redetachment. Lens status should be considered in the tactical planning of surgery for RD with PVR.
This is an abstract that was submitted for the 2016 ARVO Annual Meeting, held in Seattle, Wash., May 1-5, 2016.