Long-Term Outcomes of Boston Type 1 Keratoprosthesis in Patients With Limbal Stem Cell Deficiency From Aniridia Versus Other Causes
Purpose
To describe long-term outcomes of Boston Type 1 Keratoprosthesis (KPro) implantation in eyes with limbal stem cell deficiency (LSCD) from aniridia versus other causes.
Methods
Records were reviewed of 130 KPro implantations at a university-based tertiary care institution from February 2007 to July 2014. Eighteen eyes from 17 patients with aniridia underwent KPro implantation. Fourteen eyes of 13 patients with limbal stem-cell deficiency from other causes (10 chemical burn, 2 keratoconjunctivitis, 2 of unknown etiology) underwent KPro implantation. Outcome measures included preoperative and last recorded post-operative best-corrected visual acuity (BCVA), intraoperative, and post-operative complications, including retroprosthetic membrane development (RPM) and implant extrusion.
Results
Mean age at KPro implantation for aniridics was 44±18 years. BCVA preoperatively measured 20/400 or worse in all eyes. After a mean follow-up of 33±23 months, post-operative BCVA measured 20/250 or better in 10 eyes. Postoperative complications included RPM(12) in 67% and corneal melt(3) and extrusion(3) in 16.7%. Mean age at KPro implantation for non-aniridic LCSD patients was 59±15.5. Pre-operatively, BCVA measured 20/200 or worse in all eyes. After a mean follow-up of 46±16.3 months, post-operative BCVA measured 20/60 or better in 6 eyes and 20/400 or better in 9 eyes. Postoperative complications included RPM(7) in 50% and corneal melt(5) and extrusion(5) in 36%. The overall retention rate was 83% in aniridic and 64% in non-aniridic LSCD eyes, with a mean implant survival of 1635±1132 and 1022±485 days, respectively.
Conclusion
Long-term follow-up of keratoprosthesis implantation in patients with LCSD from aniridia or other causes supports Kpro as a useful therapeutic option in these challenging patients. The majority of eyes in each group demonstrated improved post-operative visual acuity. Prior studies have questioned LCSD etiology as a prognostic factor, but no statistically significant differences (p>0.05) were seen in rates of retroprosthetic membrane, sterile keratolysis, and implant extrusion.