Abstract
Purpose:
The purpose of our study is to review cases of corneal perforations and to compare surgical outcomes of primary penetrating keratoplasty (PK) versus repair with a small diameter lamellar or full-thickness patch graft. Traditionally, larger grafts have been discouraged in an inflamed eye due to the increased risk of immune rejection. To our knowledge, no study has directly compared these two groups in management of corneal perforations.
Methods:
A retrospective review of all corneal perforations from 2009 through 2014 was performed. Corneal perforations managed non-surgically with glue, and cases of primary closure were excluded. Patients with fewer than 30 days of total follow up were excluded. Success was determined by graft clarity at longest follow up or at last follow up prior to further corneal surgery. Statistical analysis was performed using SAS.
Results:
35 eyes with perforated corneas were included. The mechanisms of perforation included infection (n=17), trauma (n=1), autoimmune (n=6), ocular surface (n=5), surgical (n=4), and corneal ectasia (n=2). Mean age was 61.5 years (range 29-90 years). Mean follow up time was 2.2 years. 15 of 35 eyes (42%) were treated with patch grafts, 14 of which were lamellar. 11 of 15 (73%) of the patch grafts were clear at final follow up as opposed to 8 of 20 (40%) of the PKs. Multivariate regression was performed relating graft clarity to age, gender, etiology, time from perforation to surgery, use of glue, and surgical management (PK versus patch graft). Patients who underwent primary patch grafts were significantly more likely to maintain graft clarity at final follow up compared to those who underwent primary PKs (p=.02). Of the 15 patch graft patients, one patient underwent a subsequent PK with best corrected vision of 20/60 20 months post-operatively.
Conclusions:
Our study suggests that patients with corneal perforations who received patch grafts were more likely to maintain graft clarity at final follow up compared to those who underwent PKs. One possible explanation is that the majority of these patch grafts were lamellar which reduces the risk of immune rejection. Limitations of the study include inherent selection bias of a retrospective review and lack of clinical documentation of perforation measurement. Larger perforations may require PKs based on size of involvement which may lead to poorer outcomes. Future prospective studies are needed.