Malapposition of graft-host interface after penetrating keratoplasty (PK) and deep anterior lamellar keratoplasty (DALK): an optical coherence tomography study

Abstract

Backgroud Previous studies of internal graft-host malappositions have not dealt with the precise ways in which each malapposition affected post-penetrating keratoplasty (post-PK) visual outcomes. In this study, we reviewed our post-PK and post-deep anterior lamellar keratoplasty (post-DALK) keratoconic patients and used anterior segment optical coherence tomography (AS-OCT) to evaluate the associations between graft-host interface (GHI) characteristics and visual outcomes. Methods Novel GHI metrics included: mean graft-host touch (GHT), total prevalence of malapposition proportion (Pm), frequency of apposition (F), size of malapposition (Sm), junctional graft thickness (Tg), junctional host thickness (Th) and the absolute value of difference between Tg and Th (|Tg-Th|). We connected the external and internal junction points of GHI (GHT) and drew a straight line through the central point, perpendicular to both sides of the cornea. Tg and Th were the thicknesses at cross-points 1 mm away from the meeting point on the external side of the graft and host, respectively. Linear regression analysis was used to describe associations between GHI metrics and postsurgical visual outcomes [logarithm of minimum angle of resolution best-corrected visual acuity (logMAR BCVA), spherical equivalent diopter (SE), diopter of spherical power (DS), diopter of cylindrical power (DC) and keratometric astigmatism (Astig value)]. Results We enrolled 22 post-PK and 23 post-DALK keratoconic patients. Compared with the regular-apposition results, GHT was decreased in step and gape patterns, and increased in hill and tag patterns. SE increased averagely by 6.851, 5.428 and 5.164 diopter per 1% increase in: F (step) [β = 6.851; 95% Confidence interval (CI) = 2.975–10.727; P = 0.001]; F (graft step) [β = 5.428; 95% CI = 1.685–9.171; P = 0.005]; and Pm [β = 5.164; 95%CI = 0.913–9.146; P = 0.018], respectively. SE increased averagely by 0.31 diopter per 10-μm increment in |Tg-Th| [β = 0.031; 95% CI = 0.009–0.054; P = 0.007]. LogMAR BCVA increased (on average) by 0.01 per 10-μm increment in both GHT [β = 0.001; 95% CI = 0–0.002; P = 0.030]. and Tg [β = 0.001; 95% CI = 0.001–0.002; P = 0.001]. Astig value increased on average by 0.17 diopter per 10-μm increment in Sm [β = 0.017; 95% CI = 0–0.033; P = 0.047]. Conclusion This investigation of GHI characteristics suggests explanations for varied ametropia in keratoconic eyes and has potential significance as a reference for promoting pre-surgical planning and technology for corneal transplantation.

A New Pre-descemetic Corneal Ring (Neoring) in Deep Anterior Lamellar Keratoplasty for Moderate-Advanced Keratoconus: A Pilot 2-Year Long-Term Follow-Up Study

“…DALK in keratoconus aims to improve visual quality through the restoration of corneal morphology. The graft-host matching is a key factor to achieve optimal visual and refractive outcomes ( 16 ). The thickness of the corneal bed is also a factor that limits the visual acuity restoration after DALK, a recipient thickness of more than 80 μm reduces the post-operative visual acuity ( 17 ).…”

“…Finally, it is essential to note that the implant added volume to the retained peripheral recipient cornea ( Figure 5 ), decreasing the difference between the Tg and Th (Tg-Th). Zhao et al ( 16 ) found that the larger the (Tg-Th) value, the worse the visual and refractive outcomes. The (Tg-Th) value was lower in our study than that reported by Zhao et al ( 16 ) (45.1 and 85.74 μm, respectively).…”

“…Zhao et al ( 16 ) found that the larger the (Tg-Th) value, the worse the visual and refractive outcomes. The (Tg-Th) value was lower in our study than that reported by Zhao et al ( 16 ) (45.1 and 85.74 μm, respectively). Furthermore, the thickness of the complex (host + Neoring) was thicker than Tg in all cases.…”

“…Furthermore, at the last follow-up visit (24 months), the root mean square (RMS) for coma-like aberrations [computed for the Zernike terms Z (3, 1) and Z (3, −1)] and spherical aberration [Zernike coefficient Z (4, 0)] of the total cornea and the posterior corneal surface were evaluated for a pupil size of 4.5 mm. Finally, at the last visit, we used Casia II–OCT to measure the junctional graft thickness (Tg) and junctional host thickness (Th) and calculate the absolute value of the difference between Tg and Th (Tg-Th) according to the study of Zhao et al ( 16 ). The corneal thickness of the complex host + Neoring was also measured.…”

“…The most common intraoperative complication is the DM rupture ( 14 ), being mainly associated with the D-DALK technique ( 2 , 7 , 8 , 15 ). Complications related to the graft-host interface and the irregularity of the recipient corneal bed may make it difficult to achieve an optimal visual restoration in patients with severe keratoconus ( 16 , 17 ).…”