Predicting Corneal Graft Rejection by Confocal Microscopy

Purpose: This study aims at estimating corneal healing by activated keratocyte (AK) counting with in vivo confocal microscopy after perforating keratoplasty (PK). It assesses the value of AK counting in predicting corneal graft rejection. Methods: This prospective single-center observational study included 45 patients who benefited from PK in 2013 and were followed up over 2 years. All were monitored by confocal microscopy at day 1, day 7, and monthly thereafter. The AKs were counted in 5 optical sections in each of the anterior, middle, and posterior stroma. The ability of AKs in predicting the occurrence of corneal rejection was assessed by comparison of AK counts between patients with and without clinical signs of rejection. Results: In patients with graft rejection, the AK counts increased significantly 2 months before the clinical diagnosis of rejection, whereas it remained stable after 4 months in patients without rejection. In patient with graft rejection, the AK count reached a maximum at the rejection diagnosis and antirejection treatment initiation but decreased significantly 1 month after treatment initiation. Conclusions: This study confirmed the predictive value of AK counting in corneal graft rejection. The increase in the AK count allowed predicting graft rejection 2 months before the clinical diagnosis of rejection; it may then be the first sign of subclinical rejection.