Outcome of penetrating Keratoplasties performed at Kikuyu Eye Unit from 1993 to 2003
Results: Sixty percent of eyes were grafted for KC, four of whom had PKP in both eyes. The highest proportion of NKC indications included PBK (7.5%), trauma (6.6%) and dystrophies (5.3%). There was no significant difference in follow up between KC (5.25 years) and NKC (4.55 years) patients (p = 0.117). KC patients were generally younger than NKC patients (p < 0.001). The commonest immediate complications were uveitis, PEP and primary failures. There were 11 primary and 15 secondary failures accounting for 35.1% of all failures. Apart from primary and secondary failure, infections were more likely to cause graft failure compared to other complications (p<0.001). Grafted NKC eyes were more likely to fail compared with KC eyes (p < 0.001). A total of 74 eyes failed of which 36 were NKC eyes accounting for 40.4% of all NKC eyes. The highest risk for failure was seen in eyes with HSV (p = 0.025; OR = 4.44), and trachoma (p = 0.009; OR = 6.66). At 2 years, the probability of survival for KC was 81.1% (95%CI, 74.7 - 87.4%) while that for NKC was 71.3% (95%CI, 60.1 - 81.9%). Theprobability of survival for KC was 76.3% (95%CI, 70.1 - 82%) at 5 years and 72.3% (95%CI, 66.4 - 77.7%) at 10 years. For NKC cases, the probabilities of survival were evenlower at 65.3% (95%CI, 55.6 - 75.8%) at 5 years and 57% (95%CI, 48.6 - 66.1 %) at 10years. There was no association between graft survival in both KC and NKC groups and endothelial count, age of donor, and storage time (p = 0.73, 0.119 and 0.067 respectively). There was however an association between graft survival and cadaveric time (p = 0.012). In KC eyes, graft failure increased with increasing cadaveric time (p = 0.007). Postoperatively, KC eyes were more likely to have best corrected visual acuity of 6118and better (p=0.003).In total 93 (41.7%) patients were blind preoperatively. After surgery only 26 (11.7%) of all the patients remained blind. There was therefore a 72% reduction in blindness amongst all the patients who were blind preoperatively. Only 6 KC patients remained blind after PKP out of 60 patients who were blind pre-operatively, while 20 out of 33 NKC patients remained blind after PKP. There was minimal impact on blindness among the NKC patients. Conclusions: Penetrating Keratoplasty can be successful in our setting especially for KC, PBK and corneal dystrophies. There is very little role for PKP in our setting for major causes of corneal blindness like trachoma, and vitamin A deficiency. Increasing donor age and associated lower endothelial counts, and longer storage times associated with transporting donated corneas over long distances do not affect graft outcome in terms of survival. However increasing cadaveric time would lead to an increase in the number of graft failures.