Abstract
Purpose: :
Previous data from our institution suggests that presence of a clinically evident retroprosthetic membrane (RPM) is a risk factor for melt in patients with Type I Boston Keratoprosthesis (KPro). One proposed mechanism for this phenomenon is that RPM impairs nutrient diffusion into the carrier cornea by occluding backplate holes. However, only RPM located behind the optic portion of the KPro can be appreciated at the slit lamp. With the advent of anterior segment optical coherence tomography (AS-OCT), further characterization of retro-backplate RPM is now possible. The purpose of this study is to evaluate whether the presence and relative thickness of retro-backplate RPM on AS-OCT is correlated with risk of melt in patients with KPro.
Methods: :
Retrospective chart review of 44 eyes in 41 patients with Type I KPro and post-operative AS-OCT imaging was performed. AS-OCT images were reviewed for evidence of backplate RPM. When backplate RPM was present, thickness was measured at the thickest point adjacent to the titanium locking ring using Visante 3.0 software (Carl Zeiss Meditec, Jena, Germany). For eyes with clinical or OCT evidence of melt, RPM thickness was measured on the AS-OCT obtained at the last post-operative visit prior to melt. For eyes with no evidence of melt, the last available AS-OCT images were used to measure RPM thickness.
Results: :
Out of 44 total eyes, 5 had AS-OCT or clinical evidence of melt. AS-OCT evidence of retro-backplate RPM was observed in 4 out of 5 eyes that melted and 7 out of 39 eyes without melt (80.0% vs. 18.0%, p=0.0105 by Fisher exact test). The mean retro-backplate RPM thickness in the melt group (n=4) was 278 µm (range 220-310 µm, SD 39 µm) compared to 193 µm (range 100-280 µm, SD 71 µm) in the non-melt group (n=7; p=0.0592 by t-test).
Conclusions: :
Retro-backplate mean RPM thickness was greater in eyes with melt compared to eyes without melt at a level approaching statistical significance. Furthermore, the incidence of retro-backplate RPM was significantly higher among eyes with melt. Although this study is limited by its retrospective nature and the relatively small sample size, the evidence that retro-backplate RPM thickness may be associated with an increased risk of melt warrants further investigation. The evaluation of retro-backplate RPM following KPro implantation may be an important indication for AS-OCT in the future.
Keywords: keratoprostheses • cornea: clinical science • imaging/image analysis: clinical