Background and Objectives: There are few data in the literature concerning the learning curve of tractional retinal detachment (TRD) surgery. We have analyzed the experience gained by a vitreoretinal surgeon over 10 years. Materials and Methods: A retrospective, comparative study of 34 TRD cases operated using 20G instruments between 2008 and 2011 (group A) and 94 cases operated using 23G instruments between 2015 and 2019 (group B). The preoperative characteristics, the type of endotamponade, and the anatomical and functional success were reviewed. Results: The group A patients had a significantly higher rate of concomitant vitreous hemorrhage (VH) at presentation (64.7% vs. 37.2%) and of non-macular retinal detachments (52.9% vs. 39.3%). The rate of silicone oil endotamponade was high in both groups (76.4% vs. 68.1%), but in group B 25.5% were left without a tamponade (vs. none in group A). A postoperative anatomical success was obtained in 76.5% of eyes in group A and 84.04% of eyes in group B (where it was improved to 89.3% by reinterventions). The presenting visual acuity (VA) was very low in both groups (0.01 and 0.05, respectively). The proportion of eyes with improved or stabilized VA was 85.3% in group A and 79.8% in group B (statistically non-significant difference). Conclusions: The anatomical success rate improves quite slowly with increasing surgeon experience and can be further improved by reinterventions. Visual improvement does not match the rate of anatomical improvement. With increasing experience and self-confidence, the surgeon will approach more difficult cases, a fact that may slow down the increase in surgical success rates.
Keywords: learning curve; pars plana vitrectomy; tractional retinal detachment.