- Email: [email protected]
Long-term Visual Outcomes Comparing Descemet Stripping Automated Endothelial Keratoplasty and Penetrating Keratoplasty
Accepted Manuscript Long-term Visual Outcomes comparing Descemet Stripping Automated Endothelial Keratoplasty and Penetrating Keratoplasty Matthias Fuest, Marcus Ang, Hla Myint Htoon, Donald Tan, Jodhbir S. Mehta PII:
S0002-9394(17)30309-4
DOI:
10.1016/j.ajo.2017.07.014
Reference:
AJOPHT 10212
To appear in:
American Journal of Ophthalmology
Received Date: 2 April 2017 Revised Date:
12 July 2017
Accepted Date: 14 July 2017
Please cite this article as: Fuest M, Ang M, Htoon HM, Tan D, Mehta JS, Long-term Visual Outcomes comparing Descemet Stripping Automated Endothelial Keratoplasty and Penetrating Keratoplasty, American Journal of Ophthalmology (2017), doi: 10.1016/j.ajo.2017.07.014. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Abstract Purpose: To compare 5-year visual acuity and refraction outcome in Descemet stripping automated endothelial keratoplasty (DSAEK) and penetrating keratoplasty (PK) for Fuchs endothelial dystrophy (FED) or bullous keratopathy (BK) in Asian eyes. Design: Prospective interventional case series. Methods: We analyzed 828 consecutive cases of DSAEK (423) or PK (405) for FED and BK from the Singapore Cornea Transplant Registry performed from 1991 to 2011. Our main outcome measures were best spectaclecorrected visual acuity (BSCVA) with astigmatism (cylinder) and spherical equivalent (SE) over 5 years follow-up. Results: Mean age was 67.5 ±11.5 years (50.1% male, 49.9% female) and majority Chinese (76.6%, n=634) in our multi-racial Asian population. DSAEK eyes had significantly better BSCVA (P<.001-.037) with lower SE (P<.001.017) and cylinder (P<.001), independent of surgical indication compared to PK over 5 years. DSAEK was superior to PK over 5 years (P<.001-.026) in FED, but only over 3 years in BK (P<.001-.031). DSAEK in FED eyes had significantly better BSCVA compared to BK eyes (P=.006 at 4 year follow-up). DSAEKs with preoperative BSCVA <1.3 logMAR had significantly better visual outcomes than cases with ≥1.3 (P<.001-.042). PKs had significantly higher postoperative refractive correction than DSAEKs with no significant influence of the surgery indication. Conclusions: In our study cohort, DSAEK provided significantly better longterm BSCVA and lower astigmatism than PK over 5 years follow-up. Visual outcomes of DSAEK for FED were better than BK. In some analyses for years 1 to 3, ANCOVA adjustment indicated that this DSAEK-associated better long-term BSCVA was independent of better pre-operative vision in DSAEK eyes.
ACCEPTED MANUSCRIPT Long-term Visual Outcomes comparing Descemet Stripping Automated Endothelial Keratoplasty and Penetrating Keratoplasty
RI PT
Matthias Fuest1,2, Marcus Ang1,3,4, Hla Myint Htoon1,4, Donald Tan1,3,4, Jodhbir S Mehta1,3,4,5 1 Singapore Eye Research Institute, Singapore 2 Department of Ophthalmology, RWTH Aachen University, Aachen, Germany 3 Singapore National Eye Centre, Singapore 4 Eye-ACP, Duke-NUS Graduate Medical School, Singapore 5 School of Material Science and Engineering, Nanyang Technological University, Singapore
M AN U
SC
Short title: Visual Acuity and Refraction in DSAEK vs. PK Key words: penetrating keratoplasty, lamellar keratoplasty, visual acuity, refraction Number of Figures: 5 Number of Tables: 4 Ethics: Approved by SingHealth Centralized Institutional Review Board, R847/42/2011
EP
TE D
Abbreviations/Acronyms: • best spectacle-corrected visual acuity (BSCVA) • bullous keratopathy (BK) • Descemet membrane endothelial keratoplasty (DMEK) • Descemet stripping automated endothelial keratoplasty (DSAEK) • Descemet stripping endothelial keratoplasty (DSEK) • Fuchs endothelial dystrophy (FED) • logarithm of the minimum angle of resolution (LogMAR) • penetrating keratoplasty (PK) • spherical equivalent (SE)
Assoc. Prof. Jodhbir S. Mehta, Singapore National Eye Centre, 11 Third Hospital Avenue, Singapore 168751 E-mail: [email protected]
AC C
Correspondence:
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Introduction Endothelial keratoplasty (EK) has become the gold standard treatment for corneal endothelial dysfunction, as it combines several major advantages over penetrating keratoplasty (PK), including more rapid visual recovery, superior postoperative refractive outcomes, decreased rates of rejection and increased postoperative wound strength.1, 2 Compared with Descemet stripping automated endothelial keratoplasty (DSAEK), Descemet membrane endothelial keratoplasty (DMEK) is reported to produce even better visual acuity results.3-7 However, DMEK remains a challenging technique, limited to few centers and the vast majority of EK procedures in 2015 in the U.S. were DSAEK (22,514) over DMEK (4,694).8 Whether the final best-corrected visual acuity (BCVA) outcome of DSAEK is indeed superior to PK currently still remains controversial.9-11 This is partially attributed to the heterogeneous reporting of visual outcomes, highlighted by the technology assessment reported by the American Academy of Ophthalmology.1 In this review, eleven studies reported an average Snellen visual acuity (VA), 8 reported the percentage of patients seeing at a specific Snellen VA level, 4 reported on both and 7 reviewed studies did not report VA results at all. In studies reporting average Snellen best spectacle-corrected visual acuity (BSCVA) after DSAEK with follow-ups from 3 to 21 months, vision ranged from 20/34 to 20/66.1 In terms of visual recovery, 38-100% of eyes after DSAEK achieved 20/40 or better by 20 months after surgery;1 while up to 47% could achieve BCVA 20/20 or better at 36 months.12 However, this implies that more than 50% of DSAEK eyes with no other comorbidities may not achieve 20/20 vision. Few studies directly compared visual outcomes of DSAEK to PK in the same study population. Moreover, the majority of studies reporting VA outcome following DSAEK are in patients with Fuchs endothelial dystrophy (FED), the most common indication for EK in the US,9 while worldwide bullous keratopathy (BK) is still a leading indication for corneal transplantation in many countries.13 Due to the difference in geographical distribution of endothelial cell damage between FED and BK, pre-operative visual acuity is often better in patients with FED than in patients with progressive BK – which may affect the perception of visual outcomes after DSAEK versus PK.14 The Singapore Cornea Transplant Registry provides a long follow-up in a large cohort of patients with a significant proportion of eyes with BK as well as FED.11 There are currently few studies that provide results of DSAEK beyond five years follow-up,15 and to our knowledge, no direct comparative studies that compare long-term VA outcomes of DSAEK to PK for different endothelial diseases. Therefore, in this study we compared FED and BK cases undergoing PK or DSAEK for optical reasons, to detect possible differences in visual and refractive outcomes. Methods We analyzed BSCVA and refraction data from the on-going prospective cohort of the Singapore Corneal Transplant Study (SCTS), an audited, prospective study, which tracks all clinical data and outcomes of patients who undergo corneal transplants in Singapore.11 Our inclusion and exclusion criteria have been described previously.16 In brief, inclusion of 828 consecutive patients with either FED or BK, who underwent either a primary
ACCEPTED MANUSCRIPT
TE D
M AN U
SC
RI PT
DSAEK or PK for purely optical reasons at Singapore National Eye Centre (SNEC), excluding eyes with other ocular comorbidities, that would limit visual outcome assessment, including (but not limited to) age-related macular degeneration, pre-existing advanced glaucoma (on optic disc assessment or as defined by the glaucoma severity staging system),17 amblyopia, iris abnormalities and retinal abnormalities. Regrafts, rejections and graft failures were excluded from the time of occurrence. Graft failure was defined as irreversible loss of optical clarity sufficient to compromise vision for a minimum of 3 consecutive months. Also patients with high risk of failure requiring systemic immunosuppression were not included in the analysis.18 The DSAEK cohort also included cases, that were previously analyzed for their BSCVA and contrast sensitivity outcomes over a 2-year follow-up period.19 Apart from the following exceptions all patients were pseudophakic at the time of surgery or underwent triple-procedures. Four patients in the FED PK group remained phakic and were excluded from BSCVA analysis. Nine patients (6 DSAEK and 3 PK) were aphakic during and after surgery and were also excluded. Eight corneal surgeons at SNEC performed all surgeries over the same period (1991-2011), which included cases performed or partially performed by local or international corneal fellows in training under direct supervision. Our main outcome measure from this audit was BSCVA and refraction outcome (cylindrical correction and spherical equivalent (SE)) acquired exclusively with spectacle correction by trained nurses in our department. Visual acuity with or spectacle correction over contact lenses was not considered. BSCVA was measured using the Snellen visual acuity chart and we analyzed results using logarithm of the minimum angle of resolution (LogMAR) equivalent units. Our study followed the principles of the Declaration of Helsinki, with ethics approval obtained from our local institutional review board (SingHealth Centralized Institutional Review Board, R847/42/2011).
AC C
EP
Surgical Technique Surgical techniques for all PK and DSAEK surgeries have been described previously.16 Essentially, PK surgeries were performed using a standard technique with a Hanna vacuum trephine system (Moria SA, Antony, France). A 0.25- to 0.50-mm oversized donor cornea was punched out from the endothelial side and sutured to the recipient with 10-0 nylon (Ethicon, Johnson & Johnson Medical Corporation, New Brunswick, USA), using either an 8-bite, 10-0 nylon double continuous running suture, a combination of a single 8-bite 10-0 nylon continuous and 8 interrupted sutures or 16 interrupted sutures. All DSAEK surgeries were performed using pull-through techniques as described previously.20 Donors (approximately 100 - 150- µm thickness) were prepared by the surgeon or eye bank technician using an automated lamellar therapeutic keratoplasty system (ALTK; Moria SA). After recipient Descemet membrane stripping and insertion of anterior chamber maintainer and preplaced venting incisions, a DSAEK forceps (ASICO, Westmont, USA) was used to pull the donor cornea through the scleral incision using a sheets glide (Beaver-Visitec International, Waltham, USA)20 or a donor inserter device (Endoglide, Network Medical Products, North Yorkshire, United Kingdom). An inferior peripheral iridectomy was performed through a limbal stab incision. Wounds were secured with 10-0 nylon interrupted sutures, and
ACCEPTED MANUSCRIPT
RI PT
a full air tamponade under slight compression was achieved with a large bubble in the anterior chamber for 6 to 8 minutes, while removing interface fluid from the venting incisions. For both PK and DSAEK surgeries, a bandage contact lens was placed at the end, and dexamethasone 0.1% (Merck & Co., Rahway, USA), 14 mg/ml gentamicin (Schering AG, Berlin-Wedding, Germany) and 50 mg/ml cefazolin (GlaxoSmithKline, Durham, USA) was injected subconjunctivally. All PK and DSAEK patients received a standard postoperative steroid regimen during the study period as part of our SCTS protocol: topical prednisolone acetate 1% every 3 hours for 1 month, 4 times daily for 2 months, which was tapered by 1 drop every 3 months to 1 drop daily by 1 year, and thereafter continued indefinitely.
M AN U
SC
Statistical Analysis All statistical analysis was performed by a biostatistician (H.M.H.) using the Statistical Package for the Social Sciences (IBM Corp. Released 2013. IBM SPSS Statistics for Windows, Version 22.0. Armonk, NY: IBM Corp.). Comparisons between categorical variables were conducted using the Fisher's Exact or Chi-Square test as appropriate. For continuous measures we used Student´s t test or non-parametric Mann-Whitney U test. Preoperative BSCVA differed significantly between groups. We performed analysis of covariance (ANCOVA), a general linear model, which blends ANOVA and regression. ANCOVA is used to adjust for pre-existing differences in non-equivalent groups, that cannot be made equal through random assignment. A P-value of <.05 was considered statistically significant.
AC C
EP
TE D
Results All DSAEK vs. PK Patient characteristics for all cases of PK and DSAEK are displayed in Table 1. PK and DSAEK patients did not differ significantly in age or gender. However, the PK group had a higher percentage of Chinese patients (81.2 vs. 72.1%) and more BK (75.6 vs. 61%) and less FED (24.4 vs. 39%) cases than the DSAEK group. 31.9% of all PKs and 30.3% of all DSAEK cases underwent triple-procedure (P=.62). Following surgery DSAEK patients had significantly better BSCVA than PK for five years of follow-up (Figure 1 Top Left and Right). As the preoperative BSCVA was significantly better in the DSAEK (1.2±0.6) than in the PK (1.68±0.5) group (P<.001, Figure 1 Top Left and Right), we corrected for preoperative BSCVA with ANCOVA analysis, and found that DSAEK patients still had significantly better vision until 3 years of follow-up (Figure 1 Top Left). In addition PK led to significantly higher negative cylinders and SE correction than DSAEK during the entire follow-up period (Figure 1 Bottom Left, Middle and Bottom Right). Subanalysis DSAEK vs. PK in FED Patient characteristics are displayed in Table 2 with no differences between DSAEK and PK in terms of age, gender or race. 37.5% of PK and 62.5% of DSAEK cases for FED underwent triple-procedure (P=.21). The preoperative BSCVA was significantly better in the DSAEK (0.83±0.5) than in the PK (1.4±0.6) group (P<.001, Figure 2 Top Left and Right). Following surgery DSAEK patients had significantly better BSCVA than PK for five years of
ACCEPTED MANUSCRIPT follow-up. Correcting for preoperative BSCVA with ANCOVA, DSAEK patients had significantly better vision at 1 and 3 years of follow-up (Figure 2 Top Left). The fraction of DSAEK cases reaching BSCVA ≤0.3 and ≤0.2 LogMAR was significantly greater than for PK cases (Figure 2 Middle Left, Right and Bottom). PK led to significantly higher negative cylinders and SE correction during the entire follow-up period (Suppl. Figure 1 Top Right).
M AN U
SC
RI PT
Subanalysis DSAEK vs. PK in BK Patient characteristics are displayed in Table 3 with no differences between DSAEK and PK in terms of age and gender. 19.0% of PK and 8.5% of DSAEK cases for BK underwent triple-procedure (P=.01). The preoperative BSCVA was significantly better in the DSAEK (1.5±0.5) than in the PK (1.8±0.4) group (P<.001, Figure 3 Top Left and Right). Following surgery DSAEK patients had significantly better BSCVA than PK for the first 3 years of follow-up. Correcting for preoperative BSCVA DSAEK patients had significantly better vision in the first year of follow-up (Figure 3 Top Left). The fraction of DSAEK cases reaching BSCVA ≤0.3 and ≤0.2 LogMAR was significantly greater than for PK cases in years one to three (Figure 3 Middle Left, Right and Bottom). Penetrating keratoplasty led to significantly higher negative cylinder and SE correction during the entire follow-up period (Suppl. Figure 1 Top Left), only in the fifth year differences in SE did not reach significance (PK -2.2 ±3.1 diopters (dpt.) (n=26), DSAEK -0.89 ±1.9 dpt. (n=9), P=.1).
AC C
EP
TE D
Subanalysis FED vs. BK in DSAEK Patient characteristics are displayed in Table 4. Bullous keratopathy DSAEK patients were significantly older (FED 66.2 ±9.9 years, BK 69.1 ±11.8, P=.008) and less frequently of Chinese origin (P=.005). Females represented the majority in the FED DSAEKs (57%) and males in the BK cases (53.9%). 8.5% of DSAEK cases for BK and 62.5% of DSAEK cases for FED had a triple-procedure (P=.002). The preoperative BSCVA was significantly better in the DSAEK FED (0.83 ±0.5) than in the BK (1.5±0.5) group (P<.001, Figure 4 Top Left and Right). Following DSAEK FED patients had significantly better BSCVA than BK until year 5. Following ANCOVA correction for preoperative BSCVA, FED patients had significantly better vision in the first 2 years of follow-up (Figure 4 Top Left). Splitting all DSAEK cases by preoperative median BSCVA 1.3 LogMAR (interquartile range 0.60-1.82), the group with better preoperative VA maintained significantly better BSCVA outcome for 5 years of follow-up (Figure 4 Bottom Left and Right). In terms of refractive outcome BK and FED DSAEK groups only differed significantly in cylinder in the first year (FED -1.8 ±1.2 dpt., BK -2.1 ±1.5 dpt., P=.007, Suppl. Figure 1 Bottom Left).
Discussion In this study we analyzed the BSCVA and refractive outcome over 5 years of follow-up for DSAEK and PK in patients with FED and BK. We found that in our cohort overall DSAEK led to better BSCVA over 5 years of follow-
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
up, as well as lower degrees of cylindrical and SE correction. However, we found surgery indication to play a critical role, as FED DSAEK patients had significantly better BSCVA over PK in comparison to BK. Directly comparing FED and BK DSAEK cases, FED patients were superior in BSCVA for the first two years even after ANCOVA correction for the significantly worse preoperative visual acuity of BK patients. We also observed that eyes with better preoperative BSCVA (LogMAR VA <1.3) that underwent DSAEK had significantly better BSCVA outcomes for 5 years of follow-up. In our analysis overall patients after DSAEK had better VAs than those after PK. This can be explained by significantly lower degrees of cylinder and SE being induced by the sutureless EK technique, the smaller incisions and minor corneal nerve damage with associated effects on tear film quality and less allogenic material being transplanted.1, 2 However, such superiority in VA outcome has been disputed by previous publications.1, 9 This might be partly due to variations in keratoplasty techniques and indications, preoperative VAs and the outcome reporting style.9, 10 A 2014 Cochrane analysis by Nanavaty et al. found no strong evidence from RCTs of any difference in the final visual outcome between EK and PK.9 This result is particularly interesting since they only evaluated FED patients, the group that most profited from DSAEK in our study. However, they only looked at a two-year follow-up, a time when PKs, also in our cohort, performed best. In addition, this and most studies looking at EKs also include DMEK and Descemet stripping endothelial keratoplasty (DSEK) cases, in which lamellae are prepared mechanically with more allogenic stromal tissue producing interface irregularities and worse VA outcomes than DSAEK.21 A 2009 AAO Technology Assessment by Lee et al. concluded similarity in VA outcomes between EK and PK.1 However, they once again also included DSEKs in their evaluation and additionally stated a crucial point, namely that overall comparison was limited by heterogenous reporting of VA outcomes.1 Specifically comparing DSAEK and PK, Ishiyama et al. similarly found the average BSCVA to be significantly better in DSAEK rather than PK cases both preoperatively and postoperatively.22 However, they did not perform subgroup analyses for different indications. Even though preoperative BSCVA was comparable between their and our group, their postoperative outcomes were better with DSAEKs reaching best VA results at year 3 (0.13 LogMAR vs 0.36 LogMAR (our results)). PKs had their best results in year 4 and 5 (0.32 LogMAR), while in our group it was year 3 (0.58 LogMAR). However, Ishiyama´s analysis was retrospective and few FED cases were included, all limitations of a comparison to our data. Heinzelmann et al. calculated a 2.6 times higher chance for DSAEK of reaching BCVA 0.5 Snellen or more in comparison to PK for FED and BK cases. However, they did not report absolute VA outcomes.23 No prior study has specifically sub-analyzed absolute VA and refraction outcomes in FED and BK for DSAEK and PK. In case of FED, our data shows a clear superiority of DSAEK over PK for 5 years with better BSCVA outcome and higher percentages of patients reaching ≤0.3 and ≤0.2 LogMAR. Heinzelmann et al. reported that in patients with FED, the criteria of a BSCVA of Snellen 6/7.5 (0.1 LogMAR) or better at 24 months postoperatively was reached in 15 % after DSAEK, and 10 % after PK. 23 Interestingly, in our data
ACCEPTED MANUSCRIPT
SC
RI PT
at year 2 PK reached some of it´s best results, possibly due to targeted suture removal. In all other years DSAEK similarly showed higher percentages of patients with good VA (≤0.2 LogMAR). Nevertheless, in our patients with BK, the superiority of DSAEK over PK was not so clear, with better BSCVA and higher percentages of patients with good vision only reaching significance in the first 3 years. Chronic progressive corneal edema in BK causes pathological changes in the corneal stroma such as subepithelial fibrosis, accumulation of extracellular matrix, transdifferentiation of keratocytes into fibroblasts and myofibroblasts and disorganized collagen lamellae at the posterior stroma,13 leading to limited long-term BCVA and contrast sensitivity outcomes.19, 24 Consequently in BK cases with long-term severe corneal edema and stromal scarring, PK, completely removing the diseased tissue, could still be a valuable option from an optical point of view.
M AN U
Large registry studies from the Netherlands Organ Transplantation Registry (NOTR)25 and the United Kingdom National Transplant Registry 26 similarly reported better visual outcomes for EK (including DMEK and DSEK) over PK and FED over BK. Surprisingly, A recent analysis of 13920 PKs and 2287 EKs performed between January 1996 and February 2013 in Australia by Coster et al. found visual outcomes to be significantly better for PKs than for EKs in cases of FED, while in BK patients EK achieved better visual outcomes than PK.10 These observed differences may be due to inclusion of data from surgeons with varying experience and multiple centers, as opposed to a single tertiary center that has a more homogenous surgical training regime and clinical protocols.
AC C
EP
TE D
Comparing only the DSAEK cases for FED and BK we found FED cases to reach significantly better BSCVA until 4 years of follow-up (2 years ANCOVA), which agrees with the literature.19, 23 These differences are once again attributed to more extensive corneal edema in BK cases with all aforementioned anatomic changes.13 These alterations most likely also explain the different VA outcomes of our DSAEK cohort depending on the preoperative BSCVA, with cases with <1.3 LogMAR performing significantly better over 5 years. Worse preoperative VA can therefore be interpreted with more extensive corneal edema and/or scarring. Consequently, what was recently shown for DSEK, namely that performing earlier DSEK for pseudophakic corneal edema appears to be associated with improved vision,27 also seems to be the case in DSAEK. It would therefore be advisable to inform patients about limited postoperative VA outcomes with prolonged waiting periods. Further studies have to specify preoperative VA thresholds in correlation with postoperative DSAEK outcomes. In our study PKs had significantly higher postoperative refractive correction than DSAEKs with no significant influence of the surgery indication. The postoperative cylindrical correction of approximately -3.3 dpt. and SE of -2 dpt. in our PKs has been reported before and did not improve after suture removal at 2 years of follow-up. 25, 28 DSAEK led to cylinders of -2 dpt. and SE of -0.5 dpt. In particular the difference in SE could have caused even greater differences in uncorrected than corrected-visual acuity, as reported
ACCEPTED MANUSCRIPT
SC
RI PT
previously.22 The lack of induced astigmatism with DSAEK likely was the reason why patients previously preferred vision with DSAEK to PK.1, 29 The 2009 AAO Technology Assessment by Lee et al. described the induction of hyperopia of 0.7 to 1.5 dpt. and astigmatism of 0.4 to 0.6 dpt. following DSAEK.1 As SCTS does not include preoperative refractive data, we were unable to quantify induced changes in refraction by indication or intervention. The hyperopic shift is believed to originate from the graft thickness profile. Donor lenticules prepared with microkeratomes are usually thinner centrally and thicker in the graft periphery, resulting in a reduced radius of curvature of the posterior corneal surface and reduced effective corneal power, creating a clinical shift toward hyperopia.1, 30 Astigmatism in DSAEK is believed to be induced by the limbal incision similar to standard phacoemulsification cataract surgical incision, except that it is typically 1 to 2 mm wider.1 A 5-mm scleral incision has been found to induce only 0.10 dpt. of astigmatism.31 Interestingly, BK had a significantly higher cylinder correction in the first year after DSAEK in comparison to FED, which might partly be due to a slower remodeling of the more extensive corneal scarring in BK.19
AC C
EP
TE D
M AN U
As a limitation to our study, we had substantial loss of case numbers to follow-up, which most likely explains the loss of significance between groups after years, although the graphical data gives the impression of a continuous trend. For example in Figure 4 Top Left and Right, in the fifth year of follow-up BSCVA for FED and BK DSAEK is mathematically not significantly different anymore (Fuchs n=14, BK n=11), even though the trend of superiority of FED over BK continues. In addition, preoperative VA differed largely between groups, so that we additionally performed ANCOVA correction. However, in PK the diseased stroma is entirely removed so that theoretically preoperative VA and potential stromal scarring should not be as crucial as in EK. In addition triple-procedure rates partially differed significantly between groups. Nevertheless, all patients being evaluated for VA outcomes after PK and DSAEK surgery were pseudophakic, which improves comparability between groups. We also recognize the limitations of our study, as a single-center observational study involving a predominantly Asian cohort. As we included data from a 20 year period (1991-2011) bias could also have been caused by a shifting preference from PK towards DSAEK over the years, potentially lowering the threshold for surgery with evolving DSAEK techniques. In addition the data also contains PK and DSAEK cases performed by fellows under supervision. The nature of our clinical study did not allow for randomization or stratification among the FED/BK and PK/DSAEK groups; and ideally, we would have performed complete preoperative assessment of corneal stroma damage and duration of corneal edema. To summarize, DSAEK led to better VA outcomes than PK, particularly in FED. In BK the superiority of DSAEK over PK was not that clear, which most likely is due to more extensive stromal scarring. In DSAEK preoperative VA significantly influenced postoperative outcomes. Earlier EK surgery consequently might be advisable. PKs had significantly higher postoperative refractive correction than DSAEKs with no significant influence of the surgery
ACCEPTED MANUSCRIPT indication.
M AN U
References
SC
RI PT
Acknowledgements/ Disclosure A. Funding/Support: None B. Financial Disclosures: Donald Tan and Jodhbir S Mehta, inventors of the EndoGlide, have financial interests in the device (AngioTech, Reading, Pennsylvania, USA/Network Medical Products, North Yorkshire, UK). The following authors have no financial disclosures: Matthias Fuest, Marcus Ang, Hla Myint Htoon C. Other Acknowledgments: Contributions of Authors: Design and conduct of the study (MF, MA, HH, DT, JSM); collection (MF, MA, HH, DT, JSM); management (DT, JSM); analysis (MF, MA, HH, DT, JSM) and interpretation of the data (MF, MA, HH, DT, JSM); and preparation of the manuscript (MF, MA, HH, DT, JSM).
1. Lee WB, Jacobs DS, Musch DC, Kaufman SC, Reinhart WJ, Shtein RM. Descemet's stripping endothelial keratoplasty: safety and outcomes: a report by the American Academy of Ophthalmology. Ophthalmology 2009;116(9):1818-1830.
TE D
2. Anshu A, Price MO, Tan DT, Price FW, Jr. Endothelial keratoplasty: a revolution in evolution. Surv Ophthalmol 2012;57(3):236-252.
EP
3. Hamzaoglu EC, Straiko MD, Mayko ZM, Sales CS, Terry MA. The First 100 Eyes of Standardized Descemet Stripping Automated Endothelial Keratoplasty versus Standardized Descemet Membrane Endothelial Keratoplasty. Ophthalmology 2015;122(11):2193-2199.
AC C
4. Goldich Y, Showail M, Avni-Zauberman N, et al. Contralateral eye comparison of descemet membrane endothelial keratoplasty and descemet stripping automated endothelial keratoplasty. Am J Ophthalmol 2015;159(1):155-159. 5. Ang M, Wilkins MR, Mehta JS, Tan D. Descemet membrane endothelial keratoplasty. Br J Ophthalmol 2016;100(1):15-21. 6. Bose S, Ang M, Mehta JS, Tan DT, Finkelstein E. Cost-effectiveness of Descemet's stripping endothelial keratoplasty versus penetrating keratoplasty. Ophthalmology 2013;120(3):464-470. 7. Ang M, Mehta JS, Newman SD, Han SB, Chai J, Tan D. Descemet Membrane Endothelial Keratoplasty: Preliminary Results of a Donor Insertion Pull-through Technique Using a Donor Mat Device. Am J Ophthalmol 2016;171:27-34.
ACCEPTED MANUSCRIPT 8. America EBAo. 2015 Eye Banking Statistical Report www-restoresightorg 2016;Accessed November 12, 2016. 9. Nanavaty MA, Wang X, Shortt AJ. Endothelial keratoplasty versus penetrating keratoplasty for Fuchs endothelial dystrophy. Cochrane Database Syst Rev 2014(2):CD008420.
RI PT
10. Coster DJ, Lowe MT, Keane MC, Williams KA. A comparison of lamellar and penetrating keratoplasty outcomes: a registry study. Ophthalmology 2014;121(5):979-987.
SC
11. Tan D, Ang M, Arundhati A, Khor WB. Development of Selective Lamellar Keratoplasty within an Asian Corneal Transplant Program: The Singapore Corneal Transplant Study (An American Ophthalmological Society Thesis). Trans Am Ophthalmol Soc 2015;113:T10.
M AN U
12. Li JY, Terry MA, Goshe J, Davis-Boozer D, Shamie N. Three-Year Visual Acuity Outcomes after Descemet's Stripping Automated Endothelial Keratoplasty. Ophthalmology 2012;119(6):1126-1129. 13. Morishige N, Sonoda KH. Bullous keratopathy as a progressive disease: evidence from clinical and laboratory imaging studies. Cornea 2013;32(Suppl 1):77-83.
TE D
14. Nakatani S, Murakami A. Three-year outcome of Descemet stripping automated endothelial keratoplasty for bullous keratopathy after argon laser iridotomy. Cornea 2014;33(8):780-784.
EP
15. Price MO, Calhoun P, Kollman C, Price FW, Jr., Lass JH. Descemet Stripping Endothelial Keratoplasty: Ten-Year Endothelial Cell Loss Compared with Penetrating Keratoplasty. Ophthalmology 2016;123(7):1421-1427.
AC C
16. Ang M, Mehta JS, Lim F, Bose S, Htoon HM, Tan D. Endothelial cell loss and graft survival after Descemet's stripping automated endothelial keratoplasty and penetrating keratoplasty. Ophthalmology 2012;119(11):2239-2244. 17. Mills RP, Budenz DL, Lee PP, et al. Categorizing the stage of glaucoma from pre-diagnosis to end-stage disease. Am J Ophthalmol 2006;141(1):24-30. 18. Ang M, Mehta JS, Sng CC, Htoon HM, Tan DT. Indications, outcomes, and risk factors for failure in tectonic keratoplasty. Ophthalmology 2012;119(7):1311-1319. 19. Ang M, Lim F, Htoon HM, Tan D, Mehta JS. Visual acuity and contrast sensitivity following Descemet stripping automated endothelial keratoplasty. Br J Ophthalmol 2016;100(3):307-311.
ACCEPTED MANUSCRIPT 20. Ang M, Saroj L, Htoon HM, Kiew S, Mehta JS, Tan D. Comparison of a donor insertion device to sheets glide in Descemet stripping endothelial keratoplasty: 3-year outcomes. Am J Ophthalmol 2014;157(6):1163-1169 e1163.
RI PT
21. Chen M, Gong L, Xu J, Zhu W, Devine EE. Ultrastructural and in vivo confocal microscopic evaluation of interface after Descemet's Stripping Endothelial Keratoplasty in rabbits. Acta Ophthalmol 2012;90(1):43-47. 22. Ishiyama S, Mori Y, Nejima R, et al. Comparison of Long-Term Outcomes of Visual Function and Endothelial Cell Survival After Descemet Stripping Automated Endothelial Keratoplasty and Penetrating Keratoplasty Using Mixed-Effects Models. Cornea 2016;35(12):1526-1532.
M AN U
SC
23. Heinzelmann S, Bohringer D, Eberwein P, Reinhard T, Maier P. Outcomes of Descemet membrane endothelial keratoplasty, Descemet stripping automated endothelial keratoplasty and penetrating keratoplasty from a single centre study. Graefes Arch Clin Exp Ophthalmol 2016;254(3):515-522.
24. Ang M, Li L, Chua D, et al. Descemet's stripping automated endothelial keratoplasty with anterior chamber intraocular lenses: complications and 3year outcomes. Br J Ophthalmol 2014;98(8):1028-1032.
TE D
25. Dickman MM, Peeters JM, van den Biggelaar FJ, et al. Changing Practice Patterns and Long-term Outcomes of Endothelial Versus Penetrating Keratoplasty: A Prospective Dutch Registry Study. Am J Ophthalmol 2016;170:133-142.
EP
26. Greenrod EB, Jones MN, Kaye S, et al. Center and surgeon effect on outcomes of endothelial keratoplasty versus penetrating keratoplasty in the United Kingdom. Am J Ophthalmol 2014;158(5):957-966.
AC C
27. Weissbart SB, Hammersmith KM, Ayres BD, et al. Influence of Early Descemet Stripping Endothelial Keratoplasty on Visual Outcomes in Pseudophakic Corneal Edema. Am J Ophthalmol 2016;172:58-63. 28. Han DC, Mehta JS, Por YM, Htoon HM, Tan DT. Comparison of outcomes of lamellar keratoplasty and penetrating keratoplasty in keratoconus. Am J Ophthalmol 2009;148(5):744-751. 29. Bahar I, Sansanayudh W, Levinger E, Kaiserman I, Srinivasan S, Rootman D. Posterior lamellar keratoplasty--comparison of deep lamellar endothelial keratoplasty and Descemet stripping automated endothelial keratoplasty in the same patients: a patient's perspective. Br J Ophthalmol 2009;93(2):186-190. 30. Dupps WJ, Jr., Qian Y, Meisler DM. Multivariate model of refractive shift in Descemet-stripping automated endothelial keratoplasty. J Cataract Refract Surg 2008;34(4):578-584.
ACCEPTED MANUSCRIPT
Figure legends
RI PT
31. Chen ES, Terry MA, Shamie N, Hoar KL, Friend DJ. Descemetstripping automated endothelial keratoplasty: six-month results in a prospective study of 100 eyes. Cornea 2008;27(5):514-520.
SC
Figure 1: Comparison of best spectacle-corrected visual acuity (BSCVA, LogMAR; Top Left and Right), cylinder (Bottom Left and Middle Right) and spherical equivalent (SE; Bottom Left and Bottom Right) in diopters (dpt.) of patients after Descemet stripping automated endothelial keratoplasty (DSAEK) or penetrating keratoplasty (PK) for Fuchs endothelial dystrophy (FED) or bullous keratopathy (BK).
M AN U
Figure 2: Comparison of best spectacle-corrected visual acuity (BSCVA, LogMAR) of patients after Descemet stripping automated endothelial keratoplasty (DSAEK) or penetrating keratoplasty (PK) for Fuchs endothelial dystrophy (FED; Top Left and Right). Percentage of FED patients reaching BSCVA ≤0.3 LogMAR (Middle Left and Bottom) and ≤0.2 (Middle Right and Bottom).
TE D
Figure 3: Comparison of best spectacle-corrected visual acuity (BSCVA, LogMAR) of patients after Descemet stripping automated endothelial keratoplasty (DSAEK) or penetrating keratoplasty (PK) for bullous keratopathy (BK; Top Left and Right). Percentage of BK patients reaching BSCVA ≤0.3 LogMAR (Middle Left and Bottom) and ≤0.2 (Middle Right and Bottom).
AC C
EP
Figure 4: Comparison of best spectacle-corrected visual acuity (BSCVA, LogMAR) of patients after Descemet stripping automated endothelial keratoplasty (DSAEK) for bullous keratopathy (BK) or Fuchs endothelial dystrophy (FED; Top Left and Right). Visual acuity outcome of all DSAEK cases divided by preoperative median BSCVA 1.3 LogMAR (Bottom Left and Right). Supplementary Figure 1: Cylinder and spherical equivalent (SE) in diopters (dpt.) of patients after Descemet stripping automated endothelial keratoplasty (DSAEK) or penetrating keratoplasty (PK) for bullous keratopathy (BK; Top Left), Fuchs endothelial dystrophy (FED; Top Right) or DSAEK for FED and BK (Bottom Left).
ACCEPTED MANUSCRIPT Tables
M AN U
SC
RI PT
Table 1. Characteristics of patients who underwent Descemet stripping automated endothelial keratoplasty or penetrating keratoplasty for Fuchs endothelial dystrophy or bullous keratopathy Characteristics Total PK DSAEK P value (n=828) (n=405) (n=423) Age (SD) 67.5 11.5) 67.1 12.0) 67.9 11.2) 0.26 Gender (%) 0.78 Male 415(50.1) 205(50.6) 210(49.6) Female 413(49.9) 200(49.4) 213(50.4) Race (%) <0.001 Chinese 634(76.6) 329(81.2) 305(72.1) Malay 43(5.2) 20(4.9) 23(5.4) Indian 41(5.0) 24(5.9) 17(4.0) others 110(13.3) 32(7.9) 78(18.4) Surgical <0.001 Indication (%) FED 264(31.9) 99(24.4) 165(39) BK 564(68.1) 306(75.6) 258(61) BK = bullous keratopathy; DSAEK = Descemet stripping automated endothelial keratoplasty; FED = Fuchs endothelial dystrophy; PK = penetrating keratoplasty
AC C
EP
TE D
Table 2. Characteristics of patients who underwent Descemet stripping automated endothelial keratoplasty or penetrating keratoplasty for Fuchs endothelial dystrophy Characteristics PK DSAEK P value (n=99) (n=165) Age (SD) 66.5(11.1) 66.2(9.9) 0.068 Gender (%) 0.119 Male 33(33.3) 71(43.0) Female 66(66.7) 94(57.0) Race (%) 0.234 Chinese 82(82.8) 124(75.2) Malay 7(7.1) 15(9.1) Indian 5(5.1) 6(3.6) Others 5(5.1) 20(12.1) DSAEK = Descemet stripping automated endothelial keratoplasty; PK = penetrating keratoplasty
M AN U
SC
Table 3. Characteristics of patients who underwent Descemet stripping automated endothelial keratoplasty or penetrating keratoplasty for bullous keratopathy Characteristics PK DSAEK P value (n=306) (n=258) Age (SD) 67.3(12.1) 69.1(11.8) 0.068 Gender (%) 0.579 Male 172(56.2) 139(53.9) Female 134(43.8) 119(46.1) Race (%) <0.001 Chinese 247(80.7) 181(70.2) Malay 13(4.2) 8(3.1) Indian 19(6.2) 11(4.3) Others 27(8.8) 58(22.5) DSAEK = Descemet stripping automated endothelial keratoplasty; PK = penetrating keratoplasty
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
Table 4. Characteristics of patients who underwent Descemet stripping automated endothelial keratoplasty for Fuchs endothelial dystrophy or bullous keratopathy Characteristics FED BK P value (n=165) (n=258) 66.2 9.9 69.1 11.8 0.008 Age (SD) ) ) Gender (%) 0.03 Male 71(43) 139(53.9) Female 119(57) 94(46.1) Race (%) 0.005 Chinese 124(75.2) 181(70.2) Malay 15(9.1) 8(3.1) Indian 6(3.6) 11(4.3) Others 20(12.1) 58(22.5) BK = bullous keratopathy; FED = Fuchs endothelial dystrophy
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
S0002-9394(17)30309-4
DOI:
10.1016/j.ajo.2017.07.014
Reference:
AJOPHT 10212
To appear in:
American Journal of Ophthalmology
Received Date: 2 April 2017 Revised Date:
12 July 2017
Accepted Date: 14 July 2017
Please cite this article as: Fuest M, Ang M, Htoon HM, Tan D, Mehta JS, Long-term Visual Outcomes comparing Descemet Stripping Automated Endothelial Keratoplasty and Penetrating Keratoplasty, American Journal of Ophthalmology (2017), doi: 10.1016/j.ajo.2017.07.014. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Abstract Purpose: To compare 5-year visual acuity and refraction outcome in Descemet stripping automated endothelial keratoplasty (DSAEK) and penetrating keratoplasty (PK) for Fuchs endothelial dystrophy (FED) or bullous keratopathy (BK) in Asian eyes. Design: Prospective interventional case series. Methods: We analyzed 828 consecutive cases of DSAEK (423) or PK (405) for FED and BK from the Singapore Cornea Transplant Registry performed from 1991 to 2011. Our main outcome measures were best spectaclecorrected visual acuity (BSCVA) with astigmatism (cylinder) and spherical equivalent (SE) over 5 years follow-up. Results: Mean age was 67.5 ±11.5 years (50.1% male, 49.9% female) and majority Chinese (76.6%, n=634) in our multi-racial Asian population. DSAEK eyes had significantly better BSCVA (P<.001-.037) with lower SE (P<.001.017) and cylinder (P<.001), independent of surgical indication compared to PK over 5 years. DSAEK was superior to PK over 5 years (P<.001-.026) in FED, but only over 3 years in BK (P<.001-.031). DSAEK in FED eyes had significantly better BSCVA compared to BK eyes (P=.006 at 4 year follow-up). DSAEKs with preoperative BSCVA <1.3 logMAR had significantly better visual outcomes than cases with ≥1.3 (P<.001-.042). PKs had significantly higher postoperative refractive correction than DSAEKs with no significant influence of the surgery indication. Conclusions: In our study cohort, DSAEK provided significantly better longterm BSCVA and lower astigmatism than PK over 5 years follow-up. Visual outcomes of DSAEK for FED were better than BK. In some analyses for years 1 to 3, ANCOVA adjustment indicated that this DSAEK-associated better long-term BSCVA was independent of better pre-operative vision in DSAEK eyes.
ACCEPTED MANUSCRIPT Long-term Visual Outcomes comparing Descemet Stripping Automated Endothelial Keratoplasty and Penetrating Keratoplasty
RI PT
Matthias Fuest1,2, Marcus Ang1,3,4, Hla Myint Htoon1,4, Donald Tan1,3,4, Jodhbir S Mehta1,3,4,5 1 Singapore Eye Research Institute, Singapore 2 Department of Ophthalmology, RWTH Aachen University, Aachen, Germany 3 Singapore National Eye Centre, Singapore 4 Eye-ACP, Duke-NUS Graduate Medical School, Singapore 5 School of Material Science and Engineering, Nanyang Technological University, Singapore
M AN U
SC
Short title: Visual Acuity and Refraction in DSAEK vs. PK Key words: penetrating keratoplasty, lamellar keratoplasty, visual acuity, refraction Number of Figures: 5 Number of Tables: 4 Ethics: Approved by SingHealth Centralized Institutional Review Board, R847/42/2011
EP
TE D
Abbreviations/Acronyms: • best spectacle-corrected visual acuity (BSCVA) • bullous keratopathy (BK) • Descemet membrane endothelial keratoplasty (DMEK) • Descemet stripping automated endothelial keratoplasty (DSAEK) • Descemet stripping endothelial keratoplasty (DSEK) • Fuchs endothelial dystrophy (FED) • logarithm of the minimum angle of resolution (LogMAR) • penetrating keratoplasty (PK) • spherical equivalent (SE)
Assoc. Prof. Jodhbir S. Mehta, Singapore National Eye Centre, 11 Third Hospital Avenue, Singapore 168751 E-mail: [email protected]
AC C
Correspondence:
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Introduction Endothelial keratoplasty (EK) has become the gold standard treatment for corneal endothelial dysfunction, as it combines several major advantages over penetrating keratoplasty (PK), including more rapid visual recovery, superior postoperative refractive outcomes, decreased rates of rejection and increased postoperative wound strength.1, 2 Compared with Descemet stripping automated endothelial keratoplasty (DSAEK), Descemet membrane endothelial keratoplasty (DMEK) is reported to produce even better visual acuity results.3-7 However, DMEK remains a challenging technique, limited to few centers and the vast majority of EK procedures in 2015 in the U.S. were DSAEK (22,514) over DMEK (4,694).8 Whether the final best-corrected visual acuity (BCVA) outcome of DSAEK is indeed superior to PK currently still remains controversial.9-11 This is partially attributed to the heterogeneous reporting of visual outcomes, highlighted by the technology assessment reported by the American Academy of Ophthalmology.1 In this review, eleven studies reported an average Snellen visual acuity (VA), 8 reported the percentage of patients seeing at a specific Snellen VA level, 4 reported on both and 7 reviewed studies did not report VA results at all. In studies reporting average Snellen best spectacle-corrected visual acuity (BSCVA) after DSAEK with follow-ups from 3 to 21 months, vision ranged from 20/34 to 20/66.1 In terms of visual recovery, 38-100% of eyes after DSAEK achieved 20/40 or better by 20 months after surgery;1 while up to 47% could achieve BCVA 20/20 or better at 36 months.12 However, this implies that more than 50% of DSAEK eyes with no other comorbidities may not achieve 20/20 vision. Few studies directly compared visual outcomes of DSAEK to PK in the same study population. Moreover, the majority of studies reporting VA outcome following DSAEK are in patients with Fuchs endothelial dystrophy (FED), the most common indication for EK in the US,9 while worldwide bullous keratopathy (BK) is still a leading indication for corneal transplantation in many countries.13 Due to the difference in geographical distribution of endothelial cell damage between FED and BK, pre-operative visual acuity is often better in patients with FED than in patients with progressive BK – which may affect the perception of visual outcomes after DSAEK versus PK.14 The Singapore Cornea Transplant Registry provides a long follow-up in a large cohort of patients with a significant proportion of eyes with BK as well as FED.11 There are currently few studies that provide results of DSAEK beyond five years follow-up,15 and to our knowledge, no direct comparative studies that compare long-term VA outcomes of DSAEK to PK for different endothelial diseases. Therefore, in this study we compared FED and BK cases undergoing PK or DSAEK for optical reasons, to detect possible differences in visual and refractive outcomes. Methods We analyzed BSCVA and refraction data from the on-going prospective cohort of the Singapore Corneal Transplant Study (SCTS), an audited, prospective study, which tracks all clinical data and outcomes of patients who undergo corneal transplants in Singapore.11 Our inclusion and exclusion criteria have been described previously.16 In brief, inclusion of 828 consecutive patients with either FED or BK, who underwent either a primary
ACCEPTED MANUSCRIPT
TE D
M AN U
SC
RI PT
DSAEK or PK for purely optical reasons at Singapore National Eye Centre (SNEC), excluding eyes with other ocular comorbidities, that would limit visual outcome assessment, including (but not limited to) age-related macular degeneration, pre-existing advanced glaucoma (on optic disc assessment or as defined by the glaucoma severity staging system),17 amblyopia, iris abnormalities and retinal abnormalities. Regrafts, rejections and graft failures were excluded from the time of occurrence. Graft failure was defined as irreversible loss of optical clarity sufficient to compromise vision for a minimum of 3 consecutive months. Also patients with high risk of failure requiring systemic immunosuppression were not included in the analysis.18 The DSAEK cohort also included cases, that were previously analyzed for their BSCVA and contrast sensitivity outcomes over a 2-year follow-up period.19 Apart from the following exceptions all patients were pseudophakic at the time of surgery or underwent triple-procedures. Four patients in the FED PK group remained phakic and were excluded from BSCVA analysis. Nine patients (6 DSAEK and 3 PK) were aphakic during and after surgery and were also excluded. Eight corneal surgeons at SNEC performed all surgeries over the same period (1991-2011), which included cases performed or partially performed by local or international corneal fellows in training under direct supervision. Our main outcome measure from this audit was BSCVA and refraction outcome (cylindrical correction and spherical equivalent (SE)) acquired exclusively with spectacle correction by trained nurses in our department. Visual acuity with or spectacle correction over contact lenses was not considered. BSCVA was measured using the Snellen visual acuity chart and we analyzed results using logarithm of the minimum angle of resolution (LogMAR) equivalent units. Our study followed the principles of the Declaration of Helsinki, with ethics approval obtained from our local institutional review board (SingHealth Centralized Institutional Review Board, R847/42/2011).
AC C
EP
Surgical Technique Surgical techniques for all PK and DSAEK surgeries have been described previously.16 Essentially, PK surgeries were performed using a standard technique with a Hanna vacuum trephine system (Moria SA, Antony, France). A 0.25- to 0.50-mm oversized donor cornea was punched out from the endothelial side and sutured to the recipient with 10-0 nylon (Ethicon, Johnson & Johnson Medical Corporation, New Brunswick, USA), using either an 8-bite, 10-0 nylon double continuous running suture, a combination of a single 8-bite 10-0 nylon continuous and 8 interrupted sutures or 16 interrupted sutures. All DSAEK surgeries were performed using pull-through techniques as described previously.20 Donors (approximately 100 - 150- µm thickness) were prepared by the surgeon or eye bank technician using an automated lamellar therapeutic keratoplasty system (ALTK; Moria SA). After recipient Descemet membrane stripping and insertion of anterior chamber maintainer and preplaced venting incisions, a DSAEK forceps (ASICO, Westmont, USA) was used to pull the donor cornea through the scleral incision using a sheets glide (Beaver-Visitec International, Waltham, USA)20 or a donor inserter device (Endoglide, Network Medical Products, North Yorkshire, United Kingdom). An inferior peripheral iridectomy was performed through a limbal stab incision. Wounds were secured with 10-0 nylon interrupted sutures, and
ACCEPTED MANUSCRIPT
RI PT
a full air tamponade under slight compression was achieved with a large bubble in the anterior chamber for 6 to 8 minutes, while removing interface fluid from the venting incisions. For both PK and DSAEK surgeries, a bandage contact lens was placed at the end, and dexamethasone 0.1% (Merck & Co., Rahway, USA), 14 mg/ml gentamicin (Schering AG, Berlin-Wedding, Germany) and 50 mg/ml cefazolin (GlaxoSmithKline, Durham, USA) was injected subconjunctivally. All PK and DSAEK patients received a standard postoperative steroid regimen during the study period as part of our SCTS protocol: topical prednisolone acetate 1% every 3 hours for 1 month, 4 times daily for 2 months, which was tapered by 1 drop every 3 months to 1 drop daily by 1 year, and thereafter continued indefinitely.
M AN U
SC
Statistical Analysis All statistical analysis was performed by a biostatistician (H.M.H.) using the Statistical Package for the Social Sciences (IBM Corp. Released 2013. IBM SPSS Statistics for Windows, Version 22.0. Armonk, NY: IBM Corp.). Comparisons between categorical variables were conducted using the Fisher's Exact or Chi-Square test as appropriate. For continuous measures we used Student´s t test or non-parametric Mann-Whitney U test. Preoperative BSCVA differed significantly between groups. We performed analysis of covariance (ANCOVA), a general linear model, which blends ANOVA and regression. ANCOVA is used to adjust for pre-existing differences in non-equivalent groups, that cannot be made equal through random assignment. A P-value of <.05 was considered statistically significant.
AC C
EP
TE D
Results All DSAEK vs. PK Patient characteristics for all cases of PK and DSAEK are displayed in Table 1. PK and DSAEK patients did not differ significantly in age or gender. However, the PK group had a higher percentage of Chinese patients (81.2 vs. 72.1%) and more BK (75.6 vs. 61%) and less FED (24.4 vs. 39%) cases than the DSAEK group. 31.9% of all PKs and 30.3% of all DSAEK cases underwent triple-procedure (P=.62). Following surgery DSAEK patients had significantly better BSCVA than PK for five years of follow-up (Figure 1 Top Left and Right). As the preoperative BSCVA was significantly better in the DSAEK (1.2±0.6) than in the PK (1.68±0.5) group (P<.001, Figure 1 Top Left and Right), we corrected for preoperative BSCVA with ANCOVA analysis, and found that DSAEK patients still had significantly better vision until 3 years of follow-up (Figure 1 Top Left). In addition PK led to significantly higher negative cylinders and SE correction than DSAEK during the entire follow-up period (Figure 1 Bottom Left, Middle and Bottom Right). Subanalysis DSAEK vs. PK in FED Patient characteristics are displayed in Table 2 with no differences between DSAEK and PK in terms of age, gender or race. 37.5% of PK and 62.5% of DSAEK cases for FED underwent triple-procedure (P=.21). The preoperative BSCVA was significantly better in the DSAEK (0.83±0.5) than in the PK (1.4±0.6) group (P<.001, Figure 2 Top Left and Right). Following surgery DSAEK patients had significantly better BSCVA than PK for five years of
ACCEPTED MANUSCRIPT follow-up. Correcting for preoperative BSCVA with ANCOVA, DSAEK patients had significantly better vision at 1 and 3 years of follow-up (Figure 2 Top Left). The fraction of DSAEK cases reaching BSCVA ≤0.3 and ≤0.2 LogMAR was significantly greater than for PK cases (Figure 2 Middle Left, Right and Bottom). PK led to significantly higher negative cylinders and SE correction during the entire follow-up period (Suppl. Figure 1 Top Right).
M AN U
SC
RI PT
Subanalysis DSAEK vs. PK in BK Patient characteristics are displayed in Table 3 with no differences between DSAEK and PK in terms of age and gender. 19.0% of PK and 8.5% of DSAEK cases for BK underwent triple-procedure (P=.01). The preoperative BSCVA was significantly better in the DSAEK (1.5±0.5) than in the PK (1.8±0.4) group (P<.001, Figure 3 Top Left and Right). Following surgery DSAEK patients had significantly better BSCVA than PK for the first 3 years of follow-up. Correcting for preoperative BSCVA DSAEK patients had significantly better vision in the first year of follow-up (Figure 3 Top Left). The fraction of DSAEK cases reaching BSCVA ≤0.3 and ≤0.2 LogMAR was significantly greater than for PK cases in years one to three (Figure 3 Middle Left, Right and Bottom). Penetrating keratoplasty led to significantly higher negative cylinder and SE correction during the entire follow-up period (Suppl. Figure 1 Top Left), only in the fifth year differences in SE did not reach significance (PK -2.2 ±3.1 diopters (dpt.) (n=26), DSAEK -0.89 ±1.9 dpt. (n=9), P=.1).
AC C
EP
TE D
Subanalysis FED vs. BK in DSAEK Patient characteristics are displayed in Table 4. Bullous keratopathy DSAEK patients were significantly older (FED 66.2 ±9.9 years, BK 69.1 ±11.8, P=.008) and less frequently of Chinese origin (P=.005). Females represented the majority in the FED DSAEKs (57%) and males in the BK cases (53.9%). 8.5% of DSAEK cases for BK and 62.5% of DSAEK cases for FED had a triple-procedure (P=.002). The preoperative BSCVA was significantly better in the DSAEK FED (0.83 ±0.5) than in the BK (1.5±0.5) group (P<.001, Figure 4 Top Left and Right). Following DSAEK FED patients had significantly better BSCVA than BK until year 5. Following ANCOVA correction for preoperative BSCVA, FED patients had significantly better vision in the first 2 years of follow-up (Figure 4 Top Left). Splitting all DSAEK cases by preoperative median BSCVA 1.3 LogMAR (interquartile range 0.60-1.82), the group with better preoperative VA maintained significantly better BSCVA outcome for 5 years of follow-up (Figure 4 Bottom Left and Right). In terms of refractive outcome BK and FED DSAEK groups only differed significantly in cylinder in the first year (FED -1.8 ±1.2 dpt., BK -2.1 ±1.5 dpt., P=.007, Suppl. Figure 1 Bottom Left).
Discussion In this study we analyzed the BSCVA and refractive outcome over 5 years of follow-up for DSAEK and PK in patients with FED and BK. We found that in our cohort overall DSAEK led to better BSCVA over 5 years of follow-
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
up, as well as lower degrees of cylindrical and SE correction. However, we found surgery indication to play a critical role, as FED DSAEK patients had significantly better BSCVA over PK in comparison to BK. Directly comparing FED and BK DSAEK cases, FED patients were superior in BSCVA for the first two years even after ANCOVA correction for the significantly worse preoperative visual acuity of BK patients. We also observed that eyes with better preoperative BSCVA (LogMAR VA <1.3) that underwent DSAEK had significantly better BSCVA outcomes for 5 years of follow-up. In our analysis overall patients after DSAEK had better VAs than those after PK. This can be explained by significantly lower degrees of cylinder and SE being induced by the sutureless EK technique, the smaller incisions and minor corneal nerve damage with associated effects on tear film quality and less allogenic material being transplanted.1, 2 However, such superiority in VA outcome has been disputed by previous publications.1, 9 This might be partly due to variations in keratoplasty techniques and indications, preoperative VAs and the outcome reporting style.9, 10 A 2014 Cochrane analysis by Nanavaty et al. found no strong evidence from RCTs of any difference in the final visual outcome between EK and PK.9 This result is particularly interesting since they only evaluated FED patients, the group that most profited from DSAEK in our study. However, they only looked at a two-year follow-up, a time when PKs, also in our cohort, performed best. In addition, this and most studies looking at EKs also include DMEK and Descemet stripping endothelial keratoplasty (DSEK) cases, in which lamellae are prepared mechanically with more allogenic stromal tissue producing interface irregularities and worse VA outcomes than DSAEK.21 A 2009 AAO Technology Assessment by Lee et al. concluded similarity in VA outcomes between EK and PK.1 However, they once again also included DSEKs in their evaluation and additionally stated a crucial point, namely that overall comparison was limited by heterogenous reporting of VA outcomes.1 Specifically comparing DSAEK and PK, Ishiyama et al. similarly found the average BSCVA to be significantly better in DSAEK rather than PK cases both preoperatively and postoperatively.22 However, they did not perform subgroup analyses for different indications. Even though preoperative BSCVA was comparable between their and our group, their postoperative outcomes were better with DSAEKs reaching best VA results at year 3 (0.13 LogMAR vs 0.36 LogMAR (our results)). PKs had their best results in year 4 and 5 (0.32 LogMAR), while in our group it was year 3 (0.58 LogMAR). However, Ishiyama´s analysis was retrospective and few FED cases were included, all limitations of a comparison to our data. Heinzelmann et al. calculated a 2.6 times higher chance for DSAEK of reaching BCVA 0.5 Snellen or more in comparison to PK for FED and BK cases. However, they did not report absolute VA outcomes.23 No prior study has specifically sub-analyzed absolute VA and refraction outcomes in FED and BK for DSAEK and PK. In case of FED, our data shows a clear superiority of DSAEK over PK for 5 years with better BSCVA outcome and higher percentages of patients reaching ≤0.3 and ≤0.2 LogMAR. Heinzelmann et al. reported that in patients with FED, the criteria of a BSCVA of Snellen 6/7.5 (0.1 LogMAR) or better at 24 months postoperatively was reached in 15 % after DSAEK, and 10 % after PK. 23 Interestingly, in our data
ACCEPTED MANUSCRIPT
SC
RI PT
at year 2 PK reached some of it´s best results, possibly due to targeted suture removal. In all other years DSAEK similarly showed higher percentages of patients with good VA (≤0.2 LogMAR). Nevertheless, in our patients with BK, the superiority of DSAEK over PK was not so clear, with better BSCVA and higher percentages of patients with good vision only reaching significance in the first 3 years. Chronic progressive corneal edema in BK causes pathological changes in the corneal stroma such as subepithelial fibrosis, accumulation of extracellular matrix, transdifferentiation of keratocytes into fibroblasts and myofibroblasts and disorganized collagen lamellae at the posterior stroma,13 leading to limited long-term BCVA and contrast sensitivity outcomes.19, 24 Consequently in BK cases with long-term severe corneal edema and stromal scarring, PK, completely removing the diseased tissue, could still be a valuable option from an optical point of view.
M AN U
Large registry studies from the Netherlands Organ Transplantation Registry (NOTR)25 and the United Kingdom National Transplant Registry 26 similarly reported better visual outcomes for EK (including DMEK and DSEK) over PK and FED over BK. Surprisingly, A recent analysis of 13920 PKs and 2287 EKs performed between January 1996 and February 2013 in Australia by Coster et al. found visual outcomes to be significantly better for PKs than for EKs in cases of FED, while in BK patients EK achieved better visual outcomes than PK.10 These observed differences may be due to inclusion of data from surgeons with varying experience and multiple centers, as opposed to a single tertiary center that has a more homogenous surgical training regime and clinical protocols.
AC C
EP
TE D
Comparing only the DSAEK cases for FED and BK we found FED cases to reach significantly better BSCVA until 4 years of follow-up (2 years ANCOVA), which agrees with the literature.19, 23 These differences are once again attributed to more extensive corneal edema in BK cases with all aforementioned anatomic changes.13 These alterations most likely also explain the different VA outcomes of our DSAEK cohort depending on the preoperative BSCVA, with cases with <1.3 LogMAR performing significantly better over 5 years. Worse preoperative VA can therefore be interpreted with more extensive corneal edema and/or scarring. Consequently, what was recently shown for DSEK, namely that performing earlier DSEK for pseudophakic corneal edema appears to be associated with improved vision,27 also seems to be the case in DSAEK. It would therefore be advisable to inform patients about limited postoperative VA outcomes with prolonged waiting periods. Further studies have to specify preoperative VA thresholds in correlation with postoperative DSAEK outcomes. In our study PKs had significantly higher postoperative refractive correction than DSAEKs with no significant influence of the surgery indication. The postoperative cylindrical correction of approximately -3.3 dpt. and SE of -2 dpt. in our PKs has been reported before and did not improve after suture removal at 2 years of follow-up. 25, 28 DSAEK led to cylinders of -2 dpt. and SE of -0.5 dpt. In particular the difference in SE could have caused even greater differences in uncorrected than corrected-visual acuity, as reported
ACCEPTED MANUSCRIPT
SC
RI PT
previously.22 The lack of induced astigmatism with DSAEK likely was the reason why patients previously preferred vision with DSAEK to PK.1, 29 The 2009 AAO Technology Assessment by Lee et al. described the induction of hyperopia of 0.7 to 1.5 dpt. and astigmatism of 0.4 to 0.6 dpt. following DSAEK.1 As SCTS does not include preoperative refractive data, we were unable to quantify induced changes in refraction by indication or intervention. The hyperopic shift is believed to originate from the graft thickness profile. Donor lenticules prepared with microkeratomes are usually thinner centrally and thicker in the graft periphery, resulting in a reduced radius of curvature of the posterior corneal surface and reduced effective corneal power, creating a clinical shift toward hyperopia.1, 30 Astigmatism in DSAEK is believed to be induced by the limbal incision similar to standard phacoemulsification cataract surgical incision, except that it is typically 1 to 2 mm wider.1 A 5-mm scleral incision has been found to induce only 0.10 dpt. of astigmatism.31 Interestingly, BK had a significantly higher cylinder correction in the first year after DSAEK in comparison to FED, which might partly be due to a slower remodeling of the more extensive corneal scarring in BK.19
AC C
EP
TE D
M AN U
As a limitation to our study, we had substantial loss of case numbers to follow-up, which most likely explains the loss of significance between groups after years, although the graphical data gives the impression of a continuous trend. For example in Figure 4 Top Left and Right, in the fifth year of follow-up BSCVA for FED and BK DSAEK is mathematically not significantly different anymore (Fuchs n=14, BK n=11), even though the trend of superiority of FED over BK continues. In addition, preoperative VA differed largely between groups, so that we additionally performed ANCOVA correction. However, in PK the diseased stroma is entirely removed so that theoretically preoperative VA and potential stromal scarring should not be as crucial as in EK. In addition triple-procedure rates partially differed significantly between groups. Nevertheless, all patients being evaluated for VA outcomes after PK and DSAEK surgery were pseudophakic, which improves comparability between groups. We also recognize the limitations of our study, as a single-center observational study involving a predominantly Asian cohort. As we included data from a 20 year period (1991-2011) bias could also have been caused by a shifting preference from PK towards DSAEK over the years, potentially lowering the threshold for surgery with evolving DSAEK techniques. In addition the data also contains PK and DSAEK cases performed by fellows under supervision. The nature of our clinical study did not allow for randomization or stratification among the FED/BK and PK/DSAEK groups; and ideally, we would have performed complete preoperative assessment of corneal stroma damage and duration of corneal edema. To summarize, DSAEK led to better VA outcomes than PK, particularly in FED. In BK the superiority of DSAEK over PK was not that clear, which most likely is due to more extensive stromal scarring. In DSAEK preoperative VA significantly influenced postoperative outcomes. Earlier EK surgery consequently might be advisable. PKs had significantly higher postoperative refractive correction than DSAEKs with no significant influence of the surgery
ACCEPTED MANUSCRIPT indication.
M AN U
References
SC
RI PT
Acknowledgements/ Disclosure A. Funding/Support: None B. Financial Disclosures: Donald Tan and Jodhbir S Mehta, inventors of the EndoGlide, have financial interests in the device (AngioTech, Reading, Pennsylvania, USA/Network Medical Products, North Yorkshire, UK). The following authors have no financial disclosures: Matthias Fuest, Marcus Ang, Hla Myint Htoon C. Other Acknowledgments: Contributions of Authors: Design and conduct of the study (MF, MA, HH, DT, JSM); collection (MF, MA, HH, DT, JSM); management (DT, JSM); analysis (MF, MA, HH, DT, JSM) and interpretation of the data (MF, MA, HH, DT, JSM); and preparation of the manuscript (MF, MA, HH, DT, JSM).
1. Lee WB, Jacobs DS, Musch DC, Kaufman SC, Reinhart WJ, Shtein RM. Descemet's stripping endothelial keratoplasty: safety and outcomes: a report by the American Academy of Ophthalmology. Ophthalmology 2009;116(9):1818-1830.
TE D
2. Anshu A, Price MO, Tan DT, Price FW, Jr. Endothelial keratoplasty: a revolution in evolution. Surv Ophthalmol 2012;57(3):236-252.
EP
3. Hamzaoglu EC, Straiko MD, Mayko ZM, Sales CS, Terry MA. The First 100 Eyes of Standardized Descemet Stripping Automated Endothelial Keratoplasty versus Standardized Descemet Membrane Endothelial Keratoplasty. Ophthalmology 2015;122(11):2193-2199.
AC C
4. Goldich Y, Showail M, Avni-Zauberman N, et al. Contralateral eye comparison of descemet membrane endothelial keratoplasty and descemet stripping automated endothelial keratoplasty. Am J Ophthalmol 2015;159(1):155-159. 5. Ang M, Wilkins MR, Mehta JS, Tan D. Descemet membrane endothelial keratoplasty. Br J Ophthalmol 2016;100(1):15-21. 6. Bose S, Ang M, Mehta JS, Tan DT, Finkelstein E. Cost-effectiveness of Descemet's stripping endothelial keratoplasty versus penetrating keratoplasty. Ophthalmology 2013;120(3):464-470. 7. Ang M, Mehta JS, Newman SD, Han SB, Chai J, Tan D. Descemet Membrane Endothelial Keratoplasty: Preliminary Results of a Donor Insertion Pull-through Technique Using a Donor Mat Device. Am J Ophthalmol 2016;171:27-34.
ACCEPTED MANUSCRIPT 8. America EBAo. 2015 Eye Banking Statistical Report www-restoresightorg 2016;Accessed November 12, 2016. 9. Nanavaty MA, Wang X, Shortt AJ. Endothelial keratoplasty versus penetrating keratoplasty for Fuchs endothelial dystrophy. Cochrane Database Syst Rev 2014(2):CD008420.
RI PT
10. Coster DJ, Lowe MT, Keane MC, Williams KA. A comparison of lamellar and penetrating keratoplasty outcomes: a registry study. Ophthalmology 2014;121(5):979-987.
SC
11. Tan D, Ang M, Arundhati A, Khor WB. Development of Selective Lamellar Keratoplasty within an Asian Corneal Transplant Program: The Singapore Corneal Transplant Study (An American Ophthalmological Society Thesis). Trans Am Ophthalmol Soc 2015;113:T10.
M AN U
12. Li JY, Terry MA, Goshe J, Davis-Boozer D, Shamie N. Three-Year Visual Acuity Outcomes after Descemet's Stripping Automated Endothelial Keratoplasty. Ophthalmology 2012;119(6):1126-1129. 13. Morishige N, Sonoda KH. Bullous keratopathy as a progressive disease: evidence from clinical and laboratory imaging studies. Cornea 2013;32(Suppl 1):77-83.
TE D
14. Nakatani S, Murakami A. Three-year outcome of Descemet stripping automated endothelial keratoplasty for bullous keratopathy after argon laser iridotomy. Cornea 2014;33(8):780-784.
EP
15. Price MO, Calhoun P, Kollman C, Price FW, Jr., Lass JH. Descemet Stripping Endothelial Keratoplasty: Ten-Year Endothelial Cell Loss Compared with Penetrating Keratoplasty. Ophthalmology 2016;123(7):1421-1427.
AC C
16. Ang M, Mehta JS, Lim F, Bose S, Htoon HM, Tan D. Endothelial cell loss and graft survival after Descemet's stripping automated endothelial keratoplasty and penetrating keratoplasty. Ophthalmology 2012;119(11):2239-2244. 17. Mills RP, Budenz DL, Lee PP, et al. Categorizing the stage of glaucoma from pre-diagnosis to end-stage disease. Am J Ophthalmol 2006;141(1):24-30. 18. Ang M, Mehta JS, Sng CC, Htoon HM, Tan DT. Indications, outcomes, and risk factors for failure in tectonic keratoplasty. Ophthalmology 2012;119(7):1311-1319. 19. Ang M, Lim F, Htoon HM, Tan D, Mehta JS. Visual acuity and contrast sensitivity following Descemet stripping automated endothelial keratoplasty. Br J Ophthalmol 2016;100(3):307-311.
ACCEPTED MANUSCRIPT 20. Ang M, Saroj L, Htoon HM, Kiew S, Mehta JS, Tan D. Comparison of a donor insertion device to sheets glide in Descemet stripping endothelial keratoplasty: 3-year outcomes. Am J Ophthalmol 2014;157(6):1163-1169 e1163.
RI PT
21. Chen M, Gong L, Xu J, Zhu W, Devine EE. Ultrastructural and in vivo confocal microscopic evaluation of interface after Descemet's Stripping Endothelial Keratoplasty in rabbits. Acta Ophthalmol 2012;90(1):43-47. 22. Ishiyama S, Mori Y, Nejima R, et al. Comparison of Long-Term Outcomes of Visual Function and Endothelial Cell Survival After Descemet Stripping Automated Endothelial Keratoplasty and Penetrating Keratoplasty Using Mixed-Effects Models. Cornea 2016;35(12):1526-1532.
M AN U
SC
23. Heinzelmann S, Bohringer D, Eberwein P, Reinhard T, Maier P. Outcomes of Descemet membrane endothelial keratoplasty, Descemet stripping automated endothelial keratoplasty and penetrating keratoplasty from a single centre study. Graefes Arch Clin Exp Ophthalmol 2016;254(3):515-522.
24. Ang M, Li L, Chua D, et al. Descemet's stripping automated endothelial keratoplasty with anterior chamber intraocular lenses: complications and 3year outcomes. Br J Ophthalmol 2014;98(8):1028-1032.
TE D
25. Dickman MM, Peeters JM, van den Biggelaar FJ, et al. Changing Practice Patterns and Long-term Outcomes of Endothelial Versus Penetrating Keratoplasty: A Prospective Dutch Registry Study. Am J Ophthalmol 2016;170:133-142.
EP
26. Greenrod EB, Jones MN, Kaye S, et al. Center and surgeon effect on outcomes of endothelial keratoplasty versus penetrating keratoplasty in the United Kingdom. Am J Ophthalmol 2014;158(5):957-966.
AC C
27. Weissbart SB, Hammersmith KM, Ayres BD, et al. Influence of Early Descemet Stripping Endothelial Keratoplasty on Visual Outcomes in Pseudophakic Corneal Edema. Am J Ophthalmol 2016;172:58-63. 28. Han DC, Mehta JS, Por YM, Htoon HM, Tan DT. Comparison of outcomes of lamellar keratoplasty and penetrating keratoplasty in keratoconus. Am J Ophthalmol 2009;148(5):744-751. 29. Bahar I, Sansanayudh W, Levinger E, Kaiserman I, Srinivasan S, Rootman D. Posterior lamellar keratoplasty--comparison of deep lamellar endothelial keratoplasty and Descemet stripping automated endothelial keratoplasty in the same patients: a patient's perspective. Br J Ophthalmol 2009;93(2):186-190. 30. Dupps WJ, Jr., Qian Y, Meisler DM. Multivariate model of refractive shift in Descemet-stripping automated endothelial keratoplasty. J Cataract Refract Surg 2008;34(4):578-584.
ACCEPTED MANUSCRIPT
Figure legends
RI PT
31. Chen ES, Terry MA, Shamie N, Hoar KL, Friend DJ. Descemetstripping automated endothelial keratoplasty: six-month results in a prospective study of 100 eyes. Cornea 2008;27(5):514-520.
SC
Figure 1: Comparison of best spectacle-corrected visual acuity (BSCVA, LogMAR; Top Left and Right), cylinder (Bottom Left and Middle Right) and spherical equivalent (SE; Bottom Left and Bottom Right) in diopters (dpt.) of patients after Descemet stripping automated endothelial keratoplasty (DSAEK) or penetrating keratoplasty (PK) for Fuchs endothelial dystrophy (FED) or bullous keratopathy (BK).
M AN U
Figure 2: Comparison of best spectacle-corrected visual acuity (BSCVA, LogMAR) of patients after Descemet stripping automated endothelial keratoplasty (DSAEK) or penetrating keratoplasty (PK) for Fuchs endothelial dystrophy (FED; Top Left and Right). Percentage of FED patients reaching BSCVA ≤0.3 LogMAR (Middle Left and Bottom) and ≤0.2 (Middle Right and Bottom).
TE D
Figure 3: Comparison of best spectacle-corrected visual acuity (BSCVA, LogMAR) of patients after Descemet stripping automated endothelial keratoplasty (DSAEK) or penetrating keratoplasty (PK) for bullous keratopathy (BK; Top Left and Right). Percentage of BK patients reaching BSCVA ≤0.3 LogMAR (Middle Left and Bottom) and ≤0.2 (Middle Right and Bottom).
AC C
EP
Figure 4: Comparison of best spectacle-corrected visual acuity (BSCVA, LogMAR) of patients after Descemet stripping automated endothelial keratoplasty (DSAEK) for bullous keratopathy (BK) or Fuchs endothelial dystrophy (FED; Top Left and Right). Visual acuity outcome of all DSAEK cases divided by preoperative median BSCVA 1.3 LogMAR (Bottom Left and Right). Supplementary Figure 1: Cylinder and spherical equivalent (SE) in diopters (dpt.) of patients after Descemet stripping automated endothelial keratoplasty (DSAEK) or penetrating keratoplasty (PK) for bullous keratopathy (BK; Top Left), Fuchs endothelial dystrophy (FED; Top Right) or DSAEK for FED and BK (Bottom Left).
ACCEPTED MANUSCRIPT Tables
M AN U
SC
RI PT
Table 1. Characteristics of patients who underwent Descemet stripping automated endothelial keratoplasty or penetrating keratoplasty for Fuchs endothelial dystrophy or bullous keratopathy Characteristics Total PK DSAEK P value (n=828) (n=405) (n=423) Age (SD) 67.5 11.5) 67.1 12.0) 67.9 11.2) 0.26 Gender (%) 0.78 Male 415(50.1) 205(50.6) 210(49.6) Female 413(49.9) 200(49.4) 213(50.4) Race (%) <0.001 Chinese 634(76.6) 329(81.2) 305(72.1) Malay 43(5.2) 20(4.9) 23(5.4) Indian 41(5.0) 24(5.9) 17(4.0) others 110(13.3) 32(7.9) 78(18.4) Surgical <0.001 Indication (%) FED 264(31.9) 99(24.4) 165(39) BK 564(68.1) 306(75.6) 258(61) BK = bullous keratopathy; DSAEK = Descemet stripping automated endothelial keratoplasty; FED = Fuchs endothelial dystrophy; PK = penetrating keratoplasty
AC C
EP
TE D
Table 2. Characteristics of patients who underwent Descemet stripping automated endothelial keratoplasty or penetrating keratoplasty for Fuchs endothelial dystrophy Characteristics PK DSAEK P value (n=99) (n=165) Age (SD) 66.5(11.1) 66.2(9.9) 0.068 Gender (%) 0.119 Male 33(33.3) 71(43.0) Female 66(66.7) 94(57.0) Race (%) 0.234 Chinese 82(82.8) 124(75.2) Malay 7(7.1) 15(9.1) Indian 5(5.1) 6(3.6) Others 5(5.1) 20(12.1) DSAEK = Descemet stripping automated endothelial keratoplasty; PK = penetrating keratoplasty
M AN U
SC
Table 3. Characteristics of patients who underwent Descemet stripping automated endothelial keratoplasty or penetrating keratoplasty for bullous keratopathy Characteristics PK DSAEK P value (n=306) (n=258) Age (SD) 67.3(12.1) 69.1(11.8) 0.068 Gender (%) 0.579 Male 172(56.2) 139(53.9) Female 134(43.8) 119(46.1) Race (%) <0.001 Chinese 247(80.7) 181(70.2) Malay 13(4.2) 8(3.1) Indian 19(6.2) 11(4.3) Others 27(8.8) 58(22.5) DSAEK = Descemet stripping automated endothelial keratoplasty; PK = penetrating keratoplasty
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
Table 4. Characteristics of patients who underwent Descemet stripping automated endothelial keratoplasty for Fuchs endothelial dystrophy or bullous keratopathy Characteristics FED BK P value (n=165) (n=258) 66.2 9.9 69.1 11.8 0.008 Age (SD) ) ) Gender (%) 0.03 Male 71(43) 139(53.9) Female 119(57) 94(46.1) Race (%) 0.005 Chinese 124(75.2) 181(70.2) Malay 15(9.1) 8(3.1) Indian 6(3.6) 11(4.3) Others 20(12.1) 58(22.5) BK = bullous keratopathy; FED = Fuchs endothelial dystrophy
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT