Djougarian A., Zaidman B.A., Zaidman G.
Corneal transplant surgery (CTS) is the most common surgical option for patients with keratoconus (KCN). While in adults, CTS is procedure marked by a relatively good surgical and visual outcome with low graft rejection rates [5, 6, 14, 21], in children it is a difficult procedure to perform. It is characterized by pre-, intra- and postoperative concerns .
Preoperative evaluations are difficult to perform. Intraoperatively, the procedure is complicated by low scleral rigidity, increased fibrin reaction, and positive vitreous pressure. The postoperative course is more likely to involve increased graft rejection, infection, and need for exams under anesthesia (EUA) . Complications such as the development of cataracts, secondary glaucoma, epithelial defects, band keratopathy, retinal detachment, wound leakage, retrocorneal membrane, and microbial keratitis make the postoperative course complex, and often make regrafting a necessity . Further, since graft survival gradually decreases over time , children that undergo CTS are more likely to require a regraft later on in life. It is therefore important to maintain long-term follow up for children undergoing CTS.
The aim of our study is to investigate the long-term outcome of PKP in pediatric patients with KCN.
A retrospective chart review was conducted for all children under 18 years of age that had CTS for KCN from 2000-2011.
We analyzed age, average К value and best corrected visual acuity (BCVA) at presentation; postoperative follow up time; surgical outcomes as determined by BCVA on last follow up day and finally complications such as the incidence of graft rejection, glaucoma and cataracts. We also present data on outcomes of bilateral PKP as well as the unoperated eyes of patients with bilateral KCN.
Thirty seven eyes of 21 patients with KCN were analyzed. 26 eyes had CTS. 7 were girls, 14 were boys. The average age of the patients was 15.8 years, ranging from 10 to 18.
Eighteen of the 21 patients had bilateral disease. Seven patients had bilateral penetrating keratoplasty, two of which were done at age >18, and therefore are not included in this study. Six eyes presented with hydrops. One patient had bilateral hydrop. The best corrected visual acuity (BCVA) at presentation of all eyes that did not have hydrops was recorded. Of the eyes that were operated on, 6 eyes had BCVA of 20/30-2/50, 6 eyes had BCVA of 20/70-20/100, 7 eyes had BCVA of 20/150-20/400 and 1 eye was counting fingers (CF). Of the eyes that were not operated on, 5 eyes had BCVA of 20/25 or better, 5 eyes had BCVA of 20/30-20/40, and 1 eye was 20/100. Mean К value at presentation of operated eyes was 53.3. Mean К value at presentation of unoperated eyes was 49.2.
The average time from presentation to surgery was 0.7 years, ranging from 16 days to 2.8 years. The average length of time to the last follow up exam was 2.6 years, ranging from 0.02 years to 12.5 years.
Twenty five of the 26 eyes had clear grafts. 13 eyes experienced graft rejection episodes and 12 cleared after treatment. All episodes of graft rejection occurred early after surgery, all within 1.5 years. (I don't have rejection date for Sinchi, L. OD), One graft failed due to a graft rejection. 1 eye developed medically controlled glaucoma. No eyes developed cataracts. 22 of the 25 (85%) eyes with clear grafts had a BCVA >20/40. 10 of the 25 eyes (40%) with clear grafts were 20/25 or better. One eye was 20/60 and one was CF, in two patients lost to follow-up less than 2 months after surgery. One eye was 20/100, due to a wound dehiscence after blunt trauma.
This study presents one of the longest follow up times (mean 2.6 years, ranging from 0.02 to 12.5 years) for pediatric PKP, thus allowing for a more complete picture of long-term surgical outcomes. This is particularly important in providing quality of life to a pediatric population who naturally have a longer expected life ahead of them than older patients and may have more opportunities to develop complications over time. Their age alone places them at an increased risk of graft rejection because of their more active immune systems . Furthermore, the success of the transplant varies with the indication for PKP . It is therefore important to divide patients based on diagnoses. Previous studies include data of PKP for other single diagnoses or on aggregate data on multiple diagnoses [3, 18]. Graft rejection episodes vary from 17% reported by Zaidman et al.  and 25% reported by Yang et al.  in children with Peter's anomaly to 14% reported by Al-Ghamdi  and 43.4% reported by Javadi et al.  in children with congenital hereditary endothelial dystrophy and 31% by Limaiem  and 42.3% reported by Aasuri et al.  for aggregate data on various congenital and acquired conditions.
We present data specific to one indication for PKP, 37 eyes with KCN, and we find a rejection rate of 50%. Additionally, we also present the follow-up outcomes of these rejection episodes using medical therapy. All episodes occurred early on following surgery, within 1.5 years. This is consistent with previous reports: Aasuri et al., find that 50% of graft failures occur within the first 26 weeks. Our rejection reversal was 92.3% with medical therapy, compared to prior reports of 28% in Peter's anomaly by Yang et al. , 47% for congenital opacities by Comer et al. , and 45% by Vajpayee et al. , more than 50% by Dana et al.  in a heterogeneous group, and 83.3%> by McClellan et al. .
Early symptoms of graft rejection may be difficult to identify in younger children. They may not be able to effectively express reduction in visual acuity and ocular discomfort. Diagnosis and treatment are therefore delayed and may cause a higher degree of graft failure . However, our data shows that with consistent follow up and proper patient compliance with medical therapy, rejection episodes can be reversed in this population.
Postoperative complications include cataracts and glaucoma. This can occur, for example, by postoperative shallowing of the anterior chamber and the occurrence of anterior synechiae can lead to secondary glaucoma in these children . Further, in order to prevent graft rejection, topical corticosteroids are commonly prescribed. Prolonged use of topical corticosteroids has been associated with cataracts, glaucoma and delayed wound healing . Cataract formation has been reported at a rate between 2 to 7% [7, 10, 12]. The incidence of glaucoma has been reported at a rate of 5-9% . Wound dehiscence has been reported at a rate of 2-10% [7, 9, 10, 12]. We report no cataracts in our sample, only one case of glaucoma (3%) and one case of wound dehiscence (3%).
Ultimately the purpose of the surgery is to provide useful vision. Just as with graft survival, the functional visual results vary depending on the indication for PKP. In one study by Stulting, the percentage of patients achieving 6/12 (20/40) or better in the congenital, post-traumatic, and nontraumatic groups were 3% (1/34), 17% (5/29), and 47% (14/30), respectively. We find that BCVA is dramatically improved in children with KCN following CTS. 85% reached a BCVA of 20/40 or better. This is comparable to the results seen in CTS for adults with KCN, with 86% reaching BCVA of 20/40 over a mean of 46 month follow up of 163 PKP. (McClellan) In accordance with our findings, recently Lowe et al, found that in children do well following PKP for KCN: 75% of eyes in children aged 13-19 years achieved a best corrected Snellen acuity of at least 20/40. Furthermore, in our study 2 of the patients with less than 20/40 vision were due to trauma or loss of follow up within two months. Similarly, Limaiem et al. found that poor prognosis outcomes were especially caused by noncooperation of parents and postoperative ocular trauma . Since we see that long term monitoring of patients allows improvement over time, had these patients in our study that were lost to follow up maintained regular follow up, it is possible that they too would have seen improvement. In comparison, reports by Aasuri et al. , in a study with a shorter follow up averaging 1.3 years (ranging 1 week to 5 years), BCVA of 20/40 or better was seen only in 16.7% of all eyes regardless of pathology, and 23.3% in eyes with acquired nontraumatic (ANT) conditions, including keratoconus. Again, this data is difficult to interpret because of the difficulties with examining children; visual acuity in their study could not be assessed in a total of 32.4% of eyes, and 28.3% of eyes with ANT.
In assessing risk against benefit of PKP in pediatric patients, the risk of potential future long term complications is small when compared to the benefit of improvement in BCVA and quality of life that can be achieved. Our data suggests that with regular long term monitoring and patient compliance, acceptable results could be obtained following PKP in children.
There is also controversy over whether there is an increased risk of allograft rejection after bilateral PKP [1, 2, 11]. For example, Aasuri found that out of 14 patients that had bilateral PKP, 4 had graft failure in the second operated - on eye . In our study, we found that in five patients with bilateral PKP under 18 years of age, only 1 patient had a rejection episode in the second corneal graft.
A limitation to our study is that the minimum patient age is 10 years. Aasuri et al., found that a poorer graft survival can be associated with patient age less than 5. This may result in the favorable treatment outcomes reported in this study.
The great majority of pediatric patients did well following corneal transplant surgery. 50% of eyes experienced graft rejection and 92.3% cleared. Of the 26 eyes, 1 eye developed glaucoma, 1 eye was clear but postoperative course was complicated by corneal trauma and no eyes developed cataracts.