Abstract
Purpose :
To describe changing trends in surgical techniques for diabetic patients undergoing pars plana vitrectomy (PPV) and assess visual and anatomical outcomes.
Methods :
Retrospective chart review of all patients undergoing PPV at Weill Cornell Medical Center for diabetes-associated diagnosis between 2007-2015 was performed. Variables including demographics, visual acuity (VA), lens status, surgical indications, anti-vascular endothelial growth factor (VEGF) use, gauge of vitrectomy, need for additional surgery, and complications were assessed. All those with less than 1-year of follow-up data, incomplete chart data, or vitrectomy for indications unrelated to diabetes mellitus, were excluded from analysis.
Results :
A total of 140 eyes of 105 diabetic patients, with a mean follow-up of 3.3 years, met study inclusion criteria. There was a trend over time towards smaller gauge PPV for all diabetic indications (FIGURE 1), with the greatest change in preference noted for tractional retinal detachments (TRD). Those with vitreous hemorrhage (VH) or diabetic epiretinal membrane/macular edema (DME) were more likely to have smaller gauge PPV than those with TRD (p=0.007, TABLE 1). While all patients presented with similar pre-op visual acuities (p=0.818), patients undergoing 23 and 25 gauge PPV had better visual outcomes compared to 20 gauge or mix gauge PPV (p <0.05 at 1 month, 6 month, and 1 year). Re-operation rates were slightly higher for 20 gauge PPV and mix gauge PPV than single 23 or 25 gauge PPV, however this was not statistically significant (p=0.533). Major or minor complication rates did not vary by instrumentation gauge (p=0.100 and 0.300, respectively). When controlled for surgical indication, such as TRD, there was still a trend of improved visual outcomes with smaller gauge surgery, but not to a level of statistical significance (p=0.357, 0.178, 0.100 at 1 month, 6 month, and 1 year respectively). For all diabetic indications, TRDs had the worst visual outcomes. Those with post-op VA <20/200 at 1 year was 50.5% for TRD, 14.6% for VH, and 25% for DME (p=<0.001).
Conclusions :
There is a trend towards smaller gauge vitrectomy for diabetes-related complications. Those undergoing smaller gauge PPV have better visual outcomes, but this may reflect the lower complexity of these cases. TRDs had the worse visual outcomes despite the gauge of instrumentation used.
This is an abstract that was submitted for the 2017 ARVO Annual Meeting, held in Baltimore, MD, May 7-11, 2017.