Malapposition of Graft-host Interface after Keratoplasty: an Optical Coherence Tomography Study

Background: Previous studies of internal graft-host malappositions have not dealt with the precise ways in which each wound malapposition affected post-keratoplasty visual outcomes. In this study, reviewing our postsurgical keratoconic patients using anterior segment optical coherence tomography (AS-OCT) to evaluate the correlation between characteristics of the graft-host interface (GHI) and visual outcomes was aimed. Methods: Correlations between characteristics of GHI and the postsurgical visual outcomes, including 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, were evaluated of 45 eyes (patients). Results: The graft-host touch (GHT) varied with different alignment patterns. LogMAR BCVA correlated positively with GHT (r=0.32, P =0.030) and junctional graft thickness (Tg) (r=0.49, P =0.001). SE had positive correlation with frequency of step [F (step)] (r=0.46, P =0.001), graft step [F (graft step)] (r=0.40, P =0.028), total prevalence of malapposition proportion (Pm) (r=0.35, P =0.018), size of malapposition (Sm) (r=0.31, P =0.037), Tg (r=0.03, P =0.022), disparity between junctional graft and host thickness (|Tg-Th|) (r=0.40, P =0.007), and correlated negatively with GHT (r=-0.34, P =0.021). Similar results were acquired in DS. Keratometric astigmatism was found to correlate positively with Sm (r=0.30, P =0.047). Conclusion: In keratoconic eyes, corneal keratometric astigmatism increased by 0.017 with Sm. BCVA increased by 0.001 with both GHT and Tg. Investigation of the visions for keratoconus patients. Our study might have potential reference value for future technological promotion. We verified the wide existence of internal graft-host malappositions using AS-OCT and specifically quantified the relationship between GHI characteristics and visual outcomes in post-surgery keratoconic patients for the first time. GHT increased in hill and tag patterns and decreased in step and gape patterns. F (step), F (graft step) and Pm influenced SE and DS positively. Central corneal keratometric astigmatism increased by 0.017 with Sm increase. LogMAR BCVA increased by 0.001 with both GHT and Tg. In conclusion, investigations of the characteristics of GHI would be valuable for explanation of different post-surgery visions for keratoconus patients. Our study has potential reference value for future technological promotion. Further studies are warranted to determine ways of achieving optimal graft-host apposition during surgeries.


Introduction
Patients with advanced keratoconus usually have quite low vision due to progressive high myopia and irregular astigmatism. Therefore, the primary objective of keratoplasties for these patients should be rebuilding normal corneal curvature and acceptable vision.1.2 Penetrating keratoplasty (PK) and deep anterior lamellar keratoplasty (DALK) are two main surgical treatment options for keratoconus.
However the persistence of postoperative ametropia remains the mainly unsolved cause of suboptimal vision outcomes even with progressing ophthalmological operation and examination techniques.3 Approximately 40% of cases experienced astigmatism after corneal transplantations, among which 19% to 38% had high astigmatism (> 5 diopters) and could not be corrected satisfactorily by spectacles or contact lenses.4-7 Since keratoconus usually attacks young and middleaged people, lifelong satisfying vision should be the ultimate goal of surgical treatment. 2,3,5,6,8 Anterior-segment optical coherence tomography (AS-OCT) can penetrate through deeper tissues with a stretched wavelength (1.3 μm) and be safely applied in perioperative stages to obtain crosssectional images of anterior segments, which is critical for preoperative evaluation and postoperative follow-ups. 17 Hence, it had been used to observe alignment patterns of the posterior graft-host junctions after PK. However, reports of similar observations in post-DALK cases were rarely seen.9,10 Until now, the influential factors in postsurgical refraction errors have remained inconclusive due to insufficiency of relevant reports. 11 Kaiserman and Bahar11 reported that internal graft-host malappositions were associated with increased postoperative ametropia, astigmatism. However, the precise ways in which each malapposition affected postsurgical visual outcomes were rarely seen.12 The aim of this study was to analyze and quantify the correlations between characteristics of the posterior graft-host interfaces (GHI) and postoperative visual outcomes in keratoconus patients.

Patients And Methods
A retrospective observational cross-sectional study was conducted from March to August 2016 at the Eye & ENT Hospital of Fudan University, Shanghai, China. We reviewed the clinical records of patients who had surgical treatments (PK or DALK). All subjects received comprehensive ophthalmologic examinations including visual acuity, intraocular pressure (IOP), anterior segment photographs, slitlamp biomicroscopy, and refraction. We strictly selected patients with transparent grafts at the time of enrollment who had no postsurgical complications such as secondary glaucoma, cataract, or iris synechia. According to the widely known Amsler-Krumeich classification, preoperative diagnosis of each subject was grade 4 keratoconus without any associated history of ophthalmic diseases, surgeries, or trauma.13 Written informed consent was obtained from all patients for the participation 4 in the study. This investigation adhered to the tenets of the Declaration of Helsinki and was approved by the Ethics Committee of Shanghai Eye & ENT Hospital of Fudan University. SURGICAL TECHNIQUE: All corneal transplantations were performed by skilled anterior segment surgeons (X.J.J., Z.Z.R. et al.) from our hospital who have more than 20 years' corneal surgery experience. Fresh full-thickness cornea materials preserved and provided by the local eye bank were used. All surgeries were performed under both retrobulbar and peribulbar anaesthesia, with complete akinesia of the eyeballs together with eyelids. PKs were performed using standard techniques. The diameter of each recipient bed (7.50 mm to 8.25 mm) was decided according to preoperative AS-OCT examinations. The donor was excised to the same diameter as the recipient using a manual trephine system. Intraocular viscoelastic injections (VISCOAT®; Alcon Laboratories Ltd, Ft. Worth, TX) were applied to protect corneal endothelium of grafts. The graft was secured to the host bed using 12 or 16 interrupted 10-0 nylon non-absorbable surgical suture (USIOL, Inc.; Lexington, KY), according to the surgeon's preference. At the end of each operation, anterior chamber was restored by saline injection and the watertightness of corneal wound was carefully checked. To perform DALK, the diameter range (7.50 mm to 8.25 mm) of host bed was similar to that in PK procedures. After a partial trephination of approximately 50% of the host corneal stroma and manual stripping of the superficial stromal layer, pneumatic pressure was used to detach the Descemet's membrane (DM) from the deep stroma by injecting sterile air into the latent space between these two adjacent corneal microstructures with a 30-gauge needle. Air injection would produce a dome-shaped bubble that could be seen under the surgical microscope. Corneal stromal tissue above the "bubble" was manually dissected with scissors and spatula until a complete exposure of DM underneath was achieved. The same-size donor cornea without DM and endothelium was then sutured to the recipient using a 12-or 16-bite interrupted suturing technique.

INSTRUMENTS AND METHODS
OCULUS ANALYSIS: Corneal topography images were automatically taken by the system software 5 (OCULUS Keratograph® 5M; OCULUS, Wetzlar, Germany) when the best manual adjustments of eyeball position were acquired by one single ophthalmologist (Z.Y.J). Data from images in which cornea coverage was greater than 70% was used in the following statistical procedures. Each parameter included within the device was an average of three consecutive measurements and was collected by one single examiner (W.D). STATISTICAL ANALYSIS: Data were analyzed using SPSS® version 19 (IBM® Corp.; Armonk, NY). All measurements were expressed as mean ± standard deviation (SD). After estimating the normality and homoscedasticity of all the data using the Shapiro-Wilk normality test and Levene test, One-way analysis of variance was used for continuous variables, while ranked data were analyzed by Mann-Whitney U test or Kruskal-Wallis test. The Pearson and the Spearman correlations were employed to assess the relationships between characteristics of GHI and post-surgery visual outcomes. Linear regression analysis was performed to establish equations for visual outcome parameters which showed significant correlation with AS-OCT parameters. Probabilities of less than 5% were considered statistically significant. shorter than that in PK group. DALK patients had relatively higher LogMAR BCVA and lower SE and DS values than PK patients. Graft thickness was thinner in post-PK cohort comparing with post-DALK cohort. Correspondingly, |Tg-Th| value in PK group was smaller. Details are listed in Table 1.

AS-OCT
Although GHIs had quite smooth epithelial surfaces in all scans, 58.3% of them had malappositions on the internal surface. Of all the internal graft-host interfaces, there were 43.9% steps (122 graft steps and 36 host steps), 11.7% protrusions (33 hills and 9 tags), and 2.8% gapes (10 scans). GHT size varied with alignment types. In summary, comparing with the well-apposition, internal step and gape significantly reduced GHT, while opposite effects were observed in hill and tag patterns (see Table 2).
LogMAR BCVA was found to have a positive correlation with GHT (r=0.030) and Tg (r=0.001), but correlated negatively with frequency of protrusion [F (protrusion)] (r=0.01). SE was found to have a positive correlation with F (step) (r=0.001), F (graft step) (r=0.028), Pm (r=0.018), Sm (r=0.037), Tg (r=0.022), |Tg-Th| (r=0.007), and correlated negatively with GHT (r=0.021). As for DS, very similar results were acquired. We failed to report any relationships between DS and OCT parameters. Astig value, representing keratometric astigmatism, was found to have a very slight positive correlation with Sm (r=0.047). Details are listed in Table 3.
Based on the graft-host characteristics with significant correlations, linear regression equations of most visual outcomes were established except for the one between LogMAR BCVA and F(protrution) (F=3.937, P=0.054). Details are listed in Table 4.

Discussion
The persistence of high regular or irregular astigmatism remains major postsurgical problems for keratoconus patients who resort to corneal transplantation therapies. Until now, there were still conflicting opinions about the causes of post-surgery astigmatism, such as small intraoperative trephinations, uneven suture tension, or graft-recipient misalignments.8,10 Limberg et al14 proposed 8 that imprecise graft-host matching might result in astigmatism of about 4-6 diopters.1 It has been reported that preoperative corneal pathology could influence the graft-host apposition patterns.13,15 Kaiserman et al11 reported slight misalignment-associated astigmatism after PK. Jhanji et al16 categorized the alignment patterns as step and ledge, but failed to find any relationships between malappositions and post-surgery visual outcomes, yet they raised a theory that oversize graft would affect the GHI alignment due to curling of the larger graft at the posterior surface. Despite all considerations, the potential relationships between internal graft-recipient misalignment and postoperative refractive error in keratoconic cohorts remain unclear. In the current study, graft and host beds were prepared isometrically in all cases so graft-host disparity was not an issue. Suture tension within one case was proved to be nearly even according to the OCT cross-sectional images, in which post-keratoplasty corneal morphology were quite symmetric.
Referring to lamellar keratoplasty, the GHI manifest as a moderate to high reflective interface in AS-OCT image.18 However, the cutting depth of corneal stroma in our DALK procedure virtually approached the DM layer. Hence, the very little residual stroma make the cross-sectional images using AS-OCT from PK and DALK groups very much alike.18 In our study, PK group had better LogMAR the relative merits of these two surgeries for years, there was strong evidence through register of controlled trials suggesting better LogMAR BCVA at ≧ 6 months with PK in a recent systematic review.
Our results concurred with this conclusion. 9 We found that post-surgery SE and DS were more severe, which might be contributed to the relatively worse keratoconus status and higher degree myopia before surgery.2 Due to longer post-surgery period, thinner Tg was observed in PK group.
In the current study, we found that misalignments of internal GHI existed in both postsurgical groups. only to keratoconic eyes in the current study, the most common malapposition was graft step (122 cases, 33.9%). We hold the opinion that pre-operation asymmetrical thinning of cornea in different disease staging caused more graft step, because donors' normal corneal grafts were mostly thicker than the recipient beds that had already thinned.
GHT represented the contact area between graft and recipient bed. Generally, step and gape significantly reduced GHT, while hill increased GHT. In the current study, tag pattern also increased GHT slightly. Since tag is a small piece of DM layer protruding from the corneal wound, while hill is a protrusion of both DM layer and deep corneal stroma.11,12 Hence, we suppose that tag would increase GHT, but only to a very limited extent. F (graft step), Pm, Sm, Tg, and |Tg-Th|. Given the above situation, F (step), F (graft step) and Pm seemed to be the factors that most responsible for post-keratoplasty ametropia. Moreover, the relationship between |Tg-Th| and visual outcomes indicates that greater graft-host disparity would lead to more serious ametropia, too. Hence, we assumed that the thinner the recipient bed was, the worse visual outcomes might be. The increase of graft thickness mainly occurs early in the postoperative period because of tissue edema; hence, this could explain the increase of LogMAR BCVA, SE, and DS with Tg, because most cases were observed shortly after operations.
Our study has several limitations. Because it only represented a single-center experience with relatively small-scale subjects, there may be some insignificant statistical conclusions. Next, the study population was restricted to keratoconic patients, so our results might not be applicable to postkeratoplasty cases with other corneal diseases. Moreover, we unavoidably missed some information 11 about the internal GHI, since only eight images were obtained for one eye at a time in a raster scan.
We verified the wide existence of internal graft-host malappositions using AS-OCT and specifically quantified the relationship between GHI characteristics and visual outcomes in post-surgery keratoconic patients for the first time. GHT increased in hill and tag patterns and decreased in step and gape patterns. F (step), F (graft step) and Pm influenced SE and DS positively. Central corneal keratometric astigmatism increased by 0.017 with Sm increase. LogMAR BCVA increased by 0.001 with both GHT and Tg. In conclusion, investigations of the characteristics of GHI would be valuable for explanation of different post-surgery visions for keratoconus patients. Our study has potential reference value for future technological promotion. Further studies are warranted to determine ways of achieving optimal graft-host apposition during surgeries.

Declarations
Ethics approval and consent to participate Written informed consent was obtained from all patients for the participation in the study. Approval from the Ethics Committee of Shanghai Eye & ENT Hospital of Fudan University was received.

Consent for publication
Not applicable.

Availability of data and materials
All data generated or analyzed during this study are included in this published article.

Competing interest
The authors declare that they have no competing interests.

Funding
The authors were supported by grants from the National Natural Science Foundation of China        Measurements of graft and host thickness at the corneal wound interface.