The presence of ultrastructural alterations in the stromal microarchitecture of grafted corneas can provide evidence of a natural plane of separation and in these eyes, stromal peeling can be performed successfully post-penetrating keratoplasty (PK), regardless of stromal pathology, according to research results published in the American Journal of Ophthalmology.
Through a multicenter, interventional case series, researchers evaluated the outcomes of 96 attempted deep anterior lamellar keratoplasty (DALK) surgeries that utilized the stromal peeling technique for eyes that had previously undergone PK. Clinical pathologies were evaluated for various stromal pathologies, including conditions outside of keratoconus.
All procedures were performed by 3 corneal surgeons in Italy between June 2015 and February 2020. The study cohort included eyes from patients with stromal disease that required repeat keratoplasty for unsatisfactory best spectacle-corrected visual acuity (BSCVA) or poor contact lens tolerance.
In all cases, the natural plane of separation was identified intraoperatively. Stromal peeling was completed successfully in 87.5% of eyes, while 12.5% of eyes — including 10 eyes that underwent primary PK for keratoconus and 2 for herpetic keratitis — required a conversion to mushroom PK resulting from significant Descemet membrane perforation. All perforations were at the site of the PK surgical scar and occurred during attempted identification of the surgical plane (n=2) or during severing of the PK surgical wound (n=10).
Rate of successful completion of stromal peeling surgery was not significantly different among the 3 surgeons.
After complete suture removal, baseline BSCVA was significantly improved, from 0.90±0.37 to 0.20±0.14 logMAR and 0.12±0.10 logMAR at 1 and 2 year postoperative, respectively. BSCVA subsequently remained stable during the follow-up period, and sensitivity analyses adjusting from stromal pathology and primary surgeon did not lead to different results. At 3 years postoperative, 93% of eyes reached ≥20/40 and 72% reached ≥20/25.
At 1 year postoperative, endothelial cell density (ECD) averaged 1004±263 cells/mm2 with an endothelial cell loss (ECL( of 6.6%±9.5%. After year 1, average annual ECL was 2.8%±2.7%. Twenty-two eyes experienced double anterior chamber formation, 7 of which were associated with an intraoperative microperforation.
Fourteen eyes underwent sequential cataract surgery via phacoemulsification with posterior chamber intraocular lens implantation; 4 of these eyes had an immature cataract prior to stromal peeling surgery.
The 4-year cumulative risk for immunologic rejection or graft failure was 4.1% and 4.0%, respectively. All rejection episodes were successfully reversed with the use of topical steroids.
Both cases of graft failure were due to progressive ECL and seen 2 and 6 months following sequential cataract surgery, requiring Descemet stripping automated endothelial keratoplasty (DSAEK).
In patients with grafted corneas from cases with an identified plane of separation that required conversion to mushroom PK, transmission electron microscopy demonstrated that the natural plane, identified intraoperatively, occurred “along a single row of keratocytes separating a thin layer of pre-[Descemet membrane] stroma from the overlying anterior stroma.”
Study limitations include a lack of generalizability and a lack of head-to-head statistical analysis, as well as the comparative study design.
“[T]he ultrastructural alterations in the stromal microarchitecture of grafted corneas provide evidence of the natural plane of separation identified intraoperatively,” according to the research. “Stromal peeling can be successfully performed in post-PK eyes with various stromal pathology yielding excellent visual outcomes, minimal ECL, and satisfactory graft survival.”
Reference
Busin M, Bovone C, Scorcia V, et al. Ultrastructural alterations of grafted corneal buttons: The anatomic basis for stromal peeling along a natural plane of separation. Am J Ophthalmol. Published online June 9, 2021. doi:10.1016/j.ajo.2021.06.005