- Email: [email protected]
Intensified Topical Steroids as Prophylaxis for Macular Edema After Posterior Lamellar Keratoplasty Combined With Cataract Surgery
Intensified Topical Steroids as Prophylaxis for Macular Edema After Posterior Lamellar Keratoplasty Combined With Cataract Surgery ROBERT HOERSTER, TISHA PRABRIPUTALOONG STANZEL, BJOERN OLOV BACHMANN, SEBASTIAN SIEBELMANN, MORITZ FELSCH, AND CLAUS CURSIEFEN PURPOSE:
To analyze the effect of intensified topical steroid therapy after Descemet membrane endothelial keratoplasty combined with cataract surgery (tripleDMEK) on the incidence of postoperative cystoid macular edema (CME). DESIGN: Single-center comparative clinical study with historical controls. METHODS: SETTING: Department of Ophthalmology, University of Cologne, Germany, tertiary hospital, performing 500 corneal transplant surgeries per year. PATIENTS: Total of 131 patients (150 eyes) undergoing triple-DMEK surgery. INCLUSION CRITERION: TripleDMEK surgery. EXCLUSION CRITERIA: Prior retinal surgery, history of prior CME. INTERVENTIONS: Prednisolone acetate eye drops 1% 5 times daily for the first week after surgery. After an internal change of therapy regimen: Prednisolone acetate eye drops 1% hourly for the first postoperative week. We compared 75 consecutive eyes before with 75 consecutive eyes after the change of therapy regimen. Patients received macular spectraldomain optical coherence tomography (SD OCT) preoperatively, as well as 6 weeks and 3 and 6 months post surgery. MAIN OUTCOME MEASURE: Development of CME detected by macular SD OCT during 6 months postoperatively. RESULTS: Both groups were comparable regarding baseline age, sex, central corneal thickness, rebubbling rate, and visual acuity. With topical steroid therapy 5 times per day during the first postoperative week, we observed 9 cases of subsequent CME (12%). With hourly topical steroid therapy none of the patients developed CME subsequently (P [ .003). Apart from the topical steroids during the first week, medical treatment was identical in both groups. CONCLUSIONS: Early intensified postoperative topical steroid therapy constitutes an effective prophylactic treatment to reduce incidence of CME after
Accepted for publication Dec 1, 2015. From the Department of Ophthalmology (R.H., T.P.S., B.O.B., S.S., C.C.) and the Institute of Medical Statistics, Informatics and Epidemiology (M.F.), University of Cologne, Cologne, Germany. Inquiries to Robert Hoerster, Department of Ophthalmology, University of Cologne, Kerpener Strasse 62, Cologne 50924, Germany; e-mail: [email protected]; [email protected] 0002-9394/$36.00 http://dx.doi.org/10.1016/j.ajo.2015.12.008
Ó
2016 BY
triple-DMEK surgery. (Am J Ophthalmol 2016;-: -–-. Ó 2016 by Elsevier Inc. All rights reserved.)
C
YSTOID MACULAR EDEMA (CME) IS A FREQUENT
postoperative complication of posterior lamellar keratoplasty with or without cataract surgery (triple-Descemet stripping automated endothelial keratoplasty [DSAEK]/Descemet membrane endothelial keratoplasty [DMEK]).1–4 A recent prospective study with regular spectral-domain optical coherence tomography (SD OCT) controls reported an incidence of CME after DMEK and triple-DMEK of 12%–13%.5 CME here was defined as newly developed accumulation of intra- or subretinal fluid within the central retinal subfield of the SD OCT. Most frequently CME occurs within the first 6 months after surgery.5 Low-grade sterile inflammation is discussed to be the cause of CME.1,2 Elevated protein levels in the anterior chamber of CME patients after cataract surgery also indicate an inflammatory cause.6 Although CME usually resolves after initiation of intensive topical anti-inflammatory therapy and long-term prognosis for best-corrected visual acuity (BCVA) seems to be favorable,5 the initially decreased visual acuity burdens the patient and in some cases even intravitreal injection of steroids becomes necessary.5 Intravitreal injections constitute a risk for long-term elevated eye-pressure, as well as infection. Given this information, we changed our internal therapy regimen during the first week from topical steroid therapy applied 5 times a day to an hourly regimen. We hypothesized that the increased release of proinflammatory cytokines during triple-DMEK surgery may explain the high rate of postoperative CME and speculated that by increased early postoperative anti-inflammatory therapy the rate of CME could be reduced. We therefore compared the rate of CME, defined as described above, in 75 consecutive eyes prior to the change of therapy regimen with 75 consecutive eyes after the change of therapy regimen.
METHODS THIS SINGLE-CENTER COMPARATIVE CLINICAL STUDY WITH
historical controls was conducted at the Department of
ELSEVIER INC. ALL
RIGHTS RESERVED.
1
Ophthalmology, University of Cologne, Germany, a tertiary hospital specialized for corneal surgery, performing more than 500 corneal transplant surgeries per year. The protocol followed the tenets of the Helsinki protocol. All patients were prospectively included in the Cologne DMEK Databank: Long Term Safety and Outcome, and were followed up in highly standardized fashion. The ethics committee of the University of Cologne, Germany approved data collection and analysis within the Cologne DMEK Databank (file number 14-373). All patients gave informed consent for storage and analysis of their data before surgery. Inclusion criterion was the need for tripleDMEK surgery. Exclusion criteria were prior retinal surgery and history of prior CME because of uncontrollable risk for CME. Since a single surgeon (C.C.) conducted all surgeries according to a standardized technique,7 surgeon effects are excluded. Out of more than 1000 DMEK surgeries performed so far, prior to first inclusions within this study, the surgeon had already performed over 200 triple-DMEK surgeries. Surgeries were performed between December 3, 2012 and January 9, 2015. We performed surgery as previously described.7 In short, we obtained a standardized 8 mm graft under balanced salt solution (Alcon, Freiburg, Germany) in a stepwise manner with the donor cornea fixed to a Hanna punch block (Moria, Doylestown, Pennsylvania, USA). The spontaneously formed roll of endothelium–Descemet membrane was transferred into culture medium for immediate transplantation. All preparation steps were performed directly before transplantation surgery. Before grafting Descemet membrane into the recipient, we performed the cataract surgery according to the following specifications. After clear cornea tunnel incision and capsulorrhexis, the lens was extracted using the phaco-chop technique. After polishing of the posterior lens capsule, a preloaded 1-piece acrylic lens was implanted into the capsular bag under viscoelastic substance cover. All lenses could be implanted into the capsular bag. The corneal tunnels were hydrated after irrigation and aspiration and complete removal of viscoelastics and did not require suturing in any case. In the host cornea the central 9 mm of the Descemet membrane was peeled under air filling of the complete anterior chamber. The donor graft was stained with trypan blue solution (VisionBlue; Dutch Ophthalmic Research Corp, Rotterdam, Netherlands) and injected into the fluid-filled anterior chamber with a conventional lens injector cartridge endothelial side outward. The graft was rotated 90 degrees by blunt strikes to the corneal surface, unfolded by a sterile air-bubble injection on top of the graft, and then pressed against the corneal stroma by complete filling of the anterior chamber with sterile air. If necessary, Descemet folds were removed by a LASIK roller (BD Visitec, Abingdon, UK). The anterior chamber was filled with air for up to 90% of anterior chamber volume. Patients had to maintain a postoperative supine position for 24 hours. The air dissolved spontaneously during approximately 2
5 days postoperatively. Only in case of angle closure mechanism with pronounced intraocular pressure (IOP) elevation (>40 mm Hg) due to insufficient iridectomy was the air partly removed via a paracentesis. This was necessary in 5 cases. On the day before surgery (approximately 16 hours preoperatively) all 150 patients received a yttriumaluminum-garnet laser iridotomy at 6 o’clock in miosis (1 drop pilocarpine 2%), which was enlarged intraoperatively by a 20 gauge cutter. In our clinical experience, this procedure results in less pronounced iris bleeding than iridectomy with the 20 gauge cutter alone. Preoperatively, as well as at every follow-up visit, Snellen BCVA (given as logMAR in the following), macular SD OCT (Spectralis HRAþOCT; Heidelberg Engineering GmbH, Dossenheim, Germany), slit-lamp OCT (Heidelberg Engineering GmbH), and Scheimpflug corneal topography as well as thickness analysis (PentaCam; Oculus, Wetzlar, Germany) were performed. Macular SD OCT was performed with the following specifications: scan area 20 3 15 degrees (5.8 3 4.4 mm), 37 B-scans, distance between B-scans 121 mm, 29 images averaged per scan. Follow-up visits were performed at 6 weeks, 3 months, and 6 months postoperatively. We performed additional visits in any case of questionable postoperative result according to the patients’ ophthalmologist. In 2 cases we contacted the patients’ ophthalmologist, who performed the follow-up visits, including SD OCT. Prior to April 1, 2014 all patients were treated with topical steroids (prednisolone acetate 1%) 5 times daily for the first postoperative month. Steroids were then tapered 1 drop per month (the last drop to be carried on for at least 1 year). Based on published results5,6 and on our own clinical experience, we changed our routine therapy regimen to hourly topical steroids for the first postoperative week, starting from April 1, 2014. These were applied until 11 o’clock in the evening. One week later topical steroids were reduced to 5 times daily for the rest of the first month and thereafter again tapered 1 drop per month (the last drop to be carried on for at least 1 year, ie, the old scheme used before). All other surgical and medical treatments were unchanged. We analyzed the last 75 eyes of patients before and the first consecutive 75 eyes of patients after the change of therapy regimen for the development of CME within the first 6 months post surgery. Macular edema was defined as newly developed accumulation of intra- or subretinal fluid within the central retinal subfield of the SD OCT.5,6 Statistical analysis was performed using commercially available software (GraphPad Prism; GraphPad, La Jolla, California, USA). Significance levels were calculated with Fisher exact test to compare frequencies of CME and possible confounding factors, as well as MannWhitney test to compare age, BCVA, and central corneal thickness. Significance levels of P < .05 were deemed significant. The confidence intervals were calculated with a
AMERICAN JOURNAL OF OPHTHALMOLOGY
--- 2016
computer program written in the free statistical software ‘‘R’’ (http://CRAN.R-project.org), using the formulas given in Altman and associates to provide the calculations. The program was validated using the calculation examples given by Altman and associates.8
RESULTS BASELINE CHARACTERISTICS:
We analyzed 150 eyes of 131 patients. The last 75 consecutive eyes of patients before the change of therapy regimen (Group 1) were compared to the first 75 consecutive eyes of patients after the change of therapy regimen (Group 2). Both groups were comparable regarding possible confounding factors influencing macular integrity (Table). Corneal disorders leading to tripleDMEK procedure were Fuchs endothelial dystrophy in 147 cases and keratopathy due to pseudoexfoliation in 3 cases (none of these 3 cases developed CME). We saw inactive choroidal neovascularization without CME development during the observation period in 1 patient. Rebubbling was necessary in 50 cases in Group 1 (67%) and in 39 cases in Group 2 (52%). No repositioning of the membrane was necessary. The rate of rebubbling did not differ significantly between the 2 groups but showed a trend toward fewer rebubblings in Group 2 (P ¼ .096). However, patients with CME did not receive rebubbling more often than patients without CME (55% vs 68%; P ¼ .470). Six patients needed rebubbling twice. None of them developed CME. In Group 1, 2 pairs of corneas were from 1 donor each. None of the respective patients developed CME. In
3 cases in Group 1 and 2 cases in Group 2, partial removal of the anterior chamber air fill became necessary owing to insufficient iridectomy and angle closure mechanism with pronounced IOP elevation (>40 mm Hg). We observed 2 cases of elevated intraocular pressure (defined as IOP elevation >25 mm Hg) 6 weeks and 5 months post surgery, which were putatively related to the use of topical steroid eye drops. In 1 case the patient had a known history of open-angle glaucoma. Both cases were in Treatment Group 2. In both cases topical steroids were changed to rimexolon 1% (Vexol; Alcon, Freiburg, Germany). IOP elevation was treated temporarily with topical antiglaucomatous agents. IOP normalized within 2 weeks after adaption of therapy in both cases. We did not observe any other complications ascribable to the intensified use of topical steroids. EFFECT OF HOURLY TOPICAL STEROIDS:
In Group 1 we observed 9 cases of CME (12%, 95% confidence interval 6.4%–21.3%) (Figure). In all cases CME was observed within the first 3 months post surgery (Figure). All cases of CME appeared homogenously scattered over the observational period. In 8 cases CME resolved under intensified topical anti-inflammatory therapy (Figure). CME therapy consisted of topical steroids in all cases (prednisolone acetate 1% hourly for 1 week, then tapered to 83/d for 1 week, then to 53/d) and additional oral acetazolamide in 1 case (250 mg 23/d orally for 2 weeks). In 1 case (11%), despite additional therapy with intravitreal steroids (dexamethasone 700 mg [Ozurdex; Allergan, Frankfurt, Germany]) CME became persistent and did not resolve completely over an observation period of 12 months (#3, Figure). In Group 2 none of the patients developed
TABLE. Possible Confounding Factors for the Development of Cystoid Macular Edema After Posterior Lamellar Keratoplasty Combined With Cataract Surgery At Baseline
Male, n (%) Female, n (%) Age, mean 6 SD (range) BCVA, mean 6 SD [range] Central corneal thickness, mean 6 SD (range) Diabetes mellitus, n Open-angle glaucoma, n
Dry AMD, n Macular pucker,n
Group 1 (N ¼ 75) (Topical Steroids 53/d)
Group 2 (N ¼ 75) (Topical Steroids Hourly)
38 (51%) 37 (49%) 70 6 8 years (51–89) 0.51 6 0.22 logMAR (20/60 Snellen) [range 1.30–0.22] 631 6 82 mm (513–1038 mm) Total: 7 With CME: 1 Total: 2 With CME: 0 With prostaglandin therapy: 2 Total: 1 With CME: 1 Total: 3 With CME: 0
48 (64%) 27 (36%) 68 6 10 years (24–85) 0.51 6 0.38 logMAR (20/60 Snellen) (range 2.00–0.22) 642 6 87 mm (515–987 mm) Total: 4 With CME: 0 Total: 6 With CME: 0 With prostaglandin therapy: 0 Total: 3 With CME: 0 Total: 5 With CME
P Value
.137 .376 .157 .666 .533 .276
.620 .719
AMD ¼ age-related macular degeneration; BCVA ¼ best-corrected visual acuity; CME ¼ cystoid macular edema. All confounding factors at baseline were distributed equally among the groups. No relation to CME development could be seen.
VOL. -, NO. -
PROPHYLAXIS OF MACULAR EDEMA AFTER LAMELLAR KERATOPLASTY
3
FIGURE. Spectral-domain optical coherence tomography (SD OCT) of patients with cystoid macular edema (CME) after posterior lamellar keratoplasty combined with cataract surgery (triple-Descemet membrane endothelial keratoplasty). Rows represent different cases; columns represent the SD OCT slice before surgery, at the first detection of CME and after treatment for CME, respectively. Lower right corner of each slice: Postoperative days (d). In 8 cases CME resolved under topical steroids. In 1 case (#3) CME persisted despite intravitreal steroids (dexamethasone 700 mg). In Cases #4 and #6 the patients’ ophthalmologist performed the SD OCT.
4
AMERICAN JOURNAL OF OPHTHALMOLOGY
--- 2016
CME (95% confidence interval 0%–4.9%); difference between groups 12% (95% confidence interval 4.6%– 21.3%, P ¼ .003). BCVA after 6 months was 0.22 6 0.22 logMAR (approximately 20/30 Snellen) (range 1.30– 0.00) in Group 1 and did not differ from Group 2 (0.23 6 0.16 logMAR (approximately 20/30 Snellen) (range 1.00–0.00), P ¼ .208). EFFECT OF MACULAR EDEMA ON CENTRAL RETINAL THICKNESS AND VISUAL ACUITY: Mean central retinal
thickness (CRT) of patients with macular edema (occurring only in Group 1) was increased significantly compared to preoperative baseline (262 6 31 vs 507 6 170, P ¼ .004). After initiation of therapy mean CRT decreased again (349 6 199, P ¼ .039). Mean BCVA decreased by 0.15 6 0.3 logMAR. After initiation of therapy BCVA increased again by 0.16 6 0.2 logMAR. Because of high variability and the small number of cases, the difference did not reach significance level (P ¼ .262, P ¼ .055, respectively).
DISCUSSION CME AS A COMPLICATION AFTER TRIPLE-DMEK SURGERY IS
discussed to be at least partly of inflammatory origin.1,2,5,6 With respect to these publications, we have changed our routine early postoperative therapy regimen from topical steroids 5 times per day to an hourly application for the first postoperative week. We aimed to provide a more sufficient postoperative suppression of inflammation. We found that of all patients treated with topical steroids 5 times per day during the first postoperative week, 12% developed CME within the 6 months follow-up. By contrast none of the patients treated with topical steroids hourly for the first postoperative week suffered from CME during follow up. Our older rate of CME is in accordance with recently published results.5 Patients with CME suffered from a marked increase of CRT and a simultaneous decrease in visual acuity of 0.15 logMAR. This corresponds with approximately 1.5 lines Snellen BCVA. The prognosis of CME is generally good.5 There is no uniform therapy regimen recommended; however, most therapy regimens consist of intense topical steroidal or nonsteroidal anti-inflammatory agents. These are often complemented by oral carbonic anhydrase inhibitors, which can reduce the macular edema. Intravitreal or periocular injections of steroids or anti–vascular endothelial growth
factor (VEGF) agents are to be considered if noninvasive therapy fails.9,10 The fact that in our study, out of 9 patients with CME 1 patient developed a persistent CME, despite intravitreal steroid therapy, emphasizes the relevance of reducing the CME rate, if possible. However, several factors are suspected to play a role in the development of postoperative CME. After cataract surgery incomplete vitreous detachment from the macula seems to increase the risk for CME.11 This may be attributed to a mechanical irritation of the macula upon cataract surgery. Heinzelmann and associates suspected (triple)DMEK procedures to pose an increased risk for CME development, as compared to cataract surgery, since they include pronounced manipulation of the iris by iridectomy and intracameral air tamponade.5 Manipulation of the iris is known to be a risk factor for postoperative inflammation.9 Interestingly, a longer axial length seems to have a protective effect on CME development.5 Heinzelmann and associates explain this by a gradient of inflammatory factors from the iris along the axial length toward the macula.5 In addition, iris manipulation may be reduced in (myopic) eyes with deeper anterior chamber. The finding that patients with CME have increased aqueous flare values further supports the hypothesis of increased inflammatory factors originating from the iris.6,12 Like Heinzelmann and associates, we also found no direct influence of possible ocular or nonocular confounding factors (Table). The rate of rebubbling was slightly higher in the group with steroids 5 times per day. However, since in patients with CME rebubbling was not performed at higher rates compared to patients without CME, it does not seem to be causative for the increased rate of CME seen in Group 1. This is in accordance with the results of the aforementioned group.5 In 2 cases from Treatment Group 2 we observed elevated IOP. However, IOP was not elevated earlier than 6 weeks post surgery, while intensified topical steroids were administered only for 1 week and then tapered in a similar fashion in both groups. Although it seems likely that the elevated IOP is a response to the steroid therapy,13 the sample size does not allow ascription to the intensified topical steroids, administered only for a short period post surgery. We conclude that the simple measure of hourly topical steroids during the first postoperative week after tripleDMEK surgery can reduce the risk for later developing CME, as well as further complications like intravitreal injections or long-term persistent CME.
FUNDING/SUPPORT: TISHA PRABRIPUTALOONG STANZEL, BJOERN OLOV BACHMANN, AND CLAUS CURSIEFEN RECEIVED grants from the European Union (EU): COST BM 1302; FP7 STRONG (www.biocornea.eu) and from the German Research Foundation (DFG): FOR 2240 (www.for2240.de). Financial disclosures: Claus Cursiefen has the following financial disclosures: Consultant to Gene Signal, Lausanne Switzerland; Novaliq, Heidelberg, Germany. Lectures for Allergan, Frankfurt, Germany; Santen, Osaka Japan; Ursapharm, Saarbruecken, Germany; Alcon Pharma, Freiburg, Germany; Bayer, Leverkusen, Germany; Novartis, Berlin, Germany. The following authors have no financial disclosures: Robert Hoerster, Tisha Prabriputaloong Stanzel, Bjoern Olov Bachmann, Sebastian Siebelmann, and Moritz Felsch. All authors attest that they meet the current ICMJE criteria for authorship.
VOL. -, NO. -
PROPHYLAXIS OF MACULAR EDEMA AFTER LAMELLAR KERATOPLASTY
5
REFERENCES 1. Framme C, Wolf S. Retinal complications after damaging the vitreolenticular barrier. Ophthalmologica 2012;227(1): 20–33. 2. Packer M, Lowe J, Fine H. Incidence of acute postoperative cystoid macular edema in clinical practice. J Cataract Refract Surg 2012;38(12):2108–2111. 3. Dapena I, Ham L, Droutsas K, van Dijk K, Moutsouris K, Melles GR. Learning curve in Descemet’s membrane endothelial keratoplasty: first series of 135 consecutive cases. Ophthalmology 2011;118(11):2147–2154. 4. Suh LH, Yoo SH, Deobhakta A, et al. Complications of Descemet’s stripping with automated endothelial keratoplasty: survey of 118 eyes at One Institute. Ophthalmology 2008;115(9):1517–1524. 5. Heinzelmann S, Maier P, Bohringer D, Huther S, Eberwein P, Reinhard T. Cystoid macular oedema following Descemet membrane endothelial keratoplasty. Br J Ophthalmol 2015; 99(1):98–102. 6. Ersoy L, Caramoy A, Ristau T, Kirchhof B, Fauser S. Aqueous flare is increased in patients with clinically significant cystoid
6
7.
8.
9.
10. 11. 12.
13.
macular oedema after cataract surgery. Br J Ophthalmol 2013; 97(7):862–865. Heindl LM, Riss S, Laaser K, Bachmann BO, Kruse FE, Cursiefen C. Split cornea transplantation for 2 recipients review of the first 100 consecutive patients. Am J Ophthalmol 2011;152(4):523–532.e2. Altman D, Machin D, Bryant T, Gardner M, eds. Statistics With Confidence: Confidence Intervals and Statistical Guidelines. 2nd ed. Oxford, UK: Wiley; 2000:46–50. Zur D, Fischer N, Tufail A, Mones J, Loewenstein A. Postsurgical cystoid macular edema. Eur J Ophthalmol 2011;21(Suppl 6):62–68. Lobo C. Pseudophakic cystoid macular edema. Ophthalmologica 2012;227(2):61–67. Roldan M, Serrano JM. Macular edema and vitreous detachment. Ann Ophthalmol 1989;21(4):141–148. Ursell PG, Spalton DJ, Whitcup SM, Nussenblatt RB. Cystoid macular edema after phacoemulsification: relationship to blood-aqueous barrier damage and visual acuity. J Cataract Refract Surg 1999;25(11):1492–1497. Razeghinejad MR, Katz LJ. Steroid-induced iatrogenic glaucoma. Ophthalmic Res 2012;47(2):66–80.
AMERICAN JOURNAL OF OPHTHALMOLOGY
--- 2016
Biosketch Robert Hoerster, MD is specialist for Ophthalmology at the Center of Ophthalmology, University Hospital of Cologne since 2009. His main areas of research are mediation and prophylaxis of fibrotic diseases of the eye. He is a member of the Association for Research in Vision and Ophthalmology (ARVO), the Deutsche Ophthalmologische Gesellschaft (DOG) and the Deutsche Retinologische Gesellschaft (RG).
VOL. -, NO. -
PROPHYLAXIS OF MACULAR EDEMA AFTER LAMELLAR KERATOPLASTY
6.e1
Biosketch Claus Cursiefen, MD is Chairman and Professor of the Department of Ophthalmology, University of Cologne, Germany and Speaker of the DFG research Group FOR 2240 ‘‘Lymphangiogenesis and Cellular Immunity in Inflammatory Diseases of the Eye’’. Professor Cursiefen is a cornea specialist and Action Chair of EU COST Action BM1302 ‘‘Joining Forces in Corneal Regeneration’’. He did his Residency at the Department of Ophthalmology in Erlangen/Germany and his Postdoctoral Fellowship at Schepens Eye Research Institute, Boston.
6.e2
AMERICAN JOURNAL OF OPHTHALMOLOGY
--- 2016
To analyze the effect of intensified topical steroid therapy after Descemet membrane endothelial keratoplasty combined with cataract surgery (tripleDMEK) on the incidence of postoperative cystoid macular edema (CME). DESIGN: Single-center comparative clinical study with historical controls. METHODS: SETTING: Department of Ophthalmology, University of Cologne, Germany, tertiary hospital, performing 500 corneal transplant surgeries per year. PATIENTS: Total of 131 patients (150 eyes) undergoing triple-DMEK surgery. INCLUSION CRITERION: TripleDMEK surgery. EXCLUSION CRITERIA: Prior retinal surgery, history of prior CME. INTERVENTIONS: Prednisolone acetate eye drops 1% 5 times daily for the first week after surgery. After an internal change of therapy regimen: Prednisolone acetate eye drops 1% hourly for the first postoperative week. We compared 75 consecutive eyes before with 75 consecutive eyes after the change of therapy regimen. Patients received macular spectraldomain optical coherence tomography (SD OCT) preoperatively, as well as 6 weeks and 3 and 6 months post surgery. MAIN OUTCOME MEASURE: Development of CME detected by macular SD OCT during 6 months postoperatively. RESULTS: Both groups were comparable regarding baseline age, sex, central corneal thickness, rebubbling rate, and visual acuity. With topical steroid therapy 5 times per day during the first postoperative week, we observed 9 cases of subsequent CME (12%). With hourly topical steroid therapy none of the patients developed CME subsequently (P [ .003). Apart from the topical steroids during the first week, medical treatment was identical in both groups. CONCLUSIONS: Early intensified postoperative topical steroid therapy constitutes an effective prophylactic treatment to reduce incidence of CME after
Accepted for publication Dec 1, 2015. From the Department of Ophthalmology (R.H., T.P.S., B.O.B., S.S., C.C.) and the Institute of Medical Statistics, Informatics and Epidemiology (M.F.), University of Cologne, Cologne, Germany. Inquiries to Robert Hoerster, Department of Ophthalmology, University of Cologne, Kerpener Strasse 62, Cologne 50924, Germany; e-mail: [email protected]; [email protected] 0002-9394/$36.00 http://dx.doi.org/10.1016/j.ajo.2015.12.008
Ó
2016 BY
triple-DMEK surgery. (Am J Ophthalmol 2016;-: -–-. Ó 2016 by Elsevier Inc. All rights reserved.)
C
YSTOID MACULAR EDEMA (CME) IS A FREQUENT
postoperative complication of posterior lamellar keratoplasty with or without cataract surgery (triple-Descemet stripping automated endothelial keratoplasty [DSAEK]/Descemet membrane endothelial keratoplasty [DMEK]).1–4 A recent prospective study with regular spectral-domain optical coherence tomography (SD OCT) controls reported an incidence of CME after DMEK and triple-DMEK of 12%–13%.5 CME here was defined as newly developed accumulation of intra- or subretinal fluid within the central retinal subfield of the SD OCT. Most frequently CME occurs within the first 6 months after surgery.5 Low-grade sterile inflammation is discussed to be the cause of CME.1,2 Elevated protein levels in the anterior chamber of CME patients after cataract surgery also indicate an inflammatory cause.6 Although CME usually resolves after initiation of intensive topical anti-inflammatory therapy and long-term prognosis for best-corrected visual acuity (BCVA) seems to be favorable,5 the initially decreased visual acuity burdens the patient and in some cases even intravitreal injection of steroids becomes necessary.5 Intravitreal injections constitute a risk for long-term elevated eye-pressure, as well as infection. Given this information, we changed our internal therapy regimen during the first week from topical steroid therapy applied 5 times a day to an hourly regimen. We hypothesized that the increased release of proinflammatory cytokines during triple-DMEK surgery may explain the high rate of postoperative CME and speculated that by increased early postoperative anti-inflammatory therapy the rate of CME could be reduced. We therefore compared the rate of CME, defined as described above, in 75 consecutive eyes prior to the change of therapy regimen with 75 consecutive eyes after the change of therapy regimen.
METHODS THIS SINGLE-CENTER COMPARATIVE CLINICAL STUDY WITH
historical controls was conducted at the Department of
ELSEVIER INC. ALL
RIGHTS RESERVED.
1
Ophthalmology, University of Cologne, Germany, a tertiary hospital specialized for corneal surgery, performing more than 500 corneal transplant surgeries per year. The protocol followed the tenets of the Helsinki protocol. All patients were prospectively included in the Cologne DMEK Databank: Long Term Safety and Outcome, and were followed up in highly standardized fashion. The ethics committee of the University of Cologne, Germany approved data collection and analysis within the Cologne DMEK Databank (file number 14-373). All patients gave informed consent for storage and analysis of their data before surgery. Inclusion criterion was the need for tripleDMEK surgery. Exclusion criteria were prior retinal surgery and history of prior CME because of uncontrollable risk for CME. Since a single surgeon (C.C.) conducted all surgeries according to a standardized technique,7 surgeon effects are excluded. Out of more than 1000 DMEK surgeries performed so far, prior to first inclusions within this study, the surgeon had already performed over 200 triple-DMEK surgeries. Surgeries were performed between December 3, 2012 and January 9, 2015. We performed surgery as previously described.7 In short, we obtained a standardized 8 mm graft under balanced salt solution (Alcon, Freiburg, Germany) in a stepwise manner with the donor cornea fixed to a Hanna punch block (Moria, Doylestown, Pennsylvania, USA). The spontaneously formed roll of endothelium–Descemet membrane was transferred into culture medium for immediate transplantation. All preparation steps were performed directly before transplantation surgery. Before grafting Descemet membrane into the recipient, we performed the cataract surgery according to the following specifications. After clear cornea tunnel incision and capsulorrhexis, the lens was extracted using the phaco-chop technique. After polishing of the posterior lens capsule, a preloaded 1-piece acrylic lens was implanted into the capsular bag under viscoelastic substance cover. All lenses could be implanted into the capsular bag. The corneal tunnels were hydrated after irrigation and aspiration and complete removal of viscoelastics and did not require suturing in any case. In the host cornea the central 9 mm of the Descemet membrane was peeled under air filling of the complete anterior chamber. The donor graft was stained with trypan blue solution (VisionBlue; Dutch Ophthalmic Research Corp, Rotterdam, Netherlands) and injected into the fluid-filled anterior chamber with a conventional lens injector cartridge endothelial side outward. The graft was rotated 90 degrees by blunt strikes to the corneal surface, unfolded by a sterile air-bubble injection on top of the graft, and then pressed against the corneal stroma by complete filling of the anterior chamber with sterile air. If necessary, Descemet folds were removed by a LASIK roller (BD Visitec, Abingdon, UK). The anterior chamber was filled with air for up to 90% of anterior chamber volume. Patients had to maintain a postoperative supine position for 24 hours. The air dissolved spontaneously during approximately 2
5 days postoperatively. Only in case of angle closure mechanism with pronounced intraocular pressure (IOP) elevation (>40 mm Hg) due to insufficient iridectomy was the air partly removed via a paracentesis. This was necessary in 5 cases. On the day before surgery (approximately 16 hours preoperatively) all 150 patients received a yttriumaluminum-garnet laser iridotomy at 6 o’clock in miosis (1 drop pilocarpine 2%), which was enlarged intraoperatively by a 20 gauge cutter. In our clinical experience, this procedure results in less pronounced iris bleeding than iridectomy with the 20 gauge cutter alone. Preoperatively, as well as at every follow-up visit, Snellen BCVA (given as logMAR in the following), macular SD OCT (Spectralis HRAþOCT; Heidelberg Engineering GmbH, Dossenheim, Germany), slit-lamp OCT (Heidelberg Engineering GmbH), and Scheimpflug corneal topography as well as thickness analysis (PentaCam; Oculus, Wetzlar, Germany) were performed. Macular SD OCT was performed with the following specifications: scan area 20 3 15 degrees (5.8 3 4.4 mm), 37 B-scans, distance between B-scans 121 mm, 29 images averaged per scan. Follow-up visits were performed at 6 weeks, 3 months, and 6 months postoperatively. We performed additional visits in any case of questionable postoperative result according to the patients’ ophthalmologist. In 2 cases we contacted the patients’ ophthalmologist, who performed the follow-up visits, including SD OCT. Prior to April 1, 2014 all patients were treated with topical steroids (prednisolone acetate 1%) 5 times daily for the first postoperative month. Steroids were then tapered 1 drop per month (the last drop to be carried on for at least 1 year). Based on published results5,6 and on our own clinical experience, we changed our routine therapy regimen to hourly topical steroids for the first postoperative week, starting from April 1, 2014. These were applied until 11 o’clock in the evening. One week later topical steroids were reduced to 5 times daily for the rest of the first month and thereafter again tapered 1 drop per month (the last drop to be carried on for at least 1 year, ie, the old scheme used before). All other surgical and medical treatments were unchanged. We analyzed the last 75 eyes of patients before and the first consecutive 75 eyes of patients after the change of therapy regimen for the development of CME within the first 6 months post surgery. Macular edema was defined as newly developed accumulation of intra- or subretinal fluid within the central retinal subfield of the SD OCT.5,6 Statistical analysis was performed using commercially available software (GraphPad Prism; GraphPad, La Jolla, California, USA). Significance levels were calculated with Fisher exact test to compare frequencies of CME and possible confounding factors, as well as MannWhitney test to compare age, BCVA, and central corneal thickness. Significance levels of P < .05 were deemed significant. The confidence intervals were calculated with a
AMERICAN JOURNAL OF OPHTHALMOLOGY
--- 2016
computer program written in the free statistical software ‘‘R’’ (http://CRAN.R-project.org), using the formulas given in Altman and associates to provide the calculations. The program was validated using the calculation examples given by Altman and associates.8
RESULTS BASELINE CHARACTERISTICS:
We analyzed 150 eyes of 131 patients. The last 75 consecutive eyes of patients before the change of therapy regimen (Group 1) were compared to the first 75 consecutive eyes of patients after the change of therapy regimen (Group 2). Both groups were comparable regarding possible confounding factors influencing macular integrity (Table). Corneal disorders leading to tripleDMEK procedure were Fuchs endothelial dystrophy in 147 cases and keratopathy due to pseudoexfoliation in 3 cases (none of these 3 cases developed CME). We saw inactive choroidal neovascularization without CME development during the observation period in 1 patient. Rebubbling was necessary in 50 cases in Group 1 (67%) and in 39 cases in Group 2 (52%). No repositioning of the membrane was necessary. The rate of rebubbling did not differ significantly between the 2 groups but showed a trend toward fewer rebubblings in Group 2 (P ¼ .096). However, patients with CME did not receive rebubbling more often than patients without CME (55% vs 68%; P ¼ .470). Six patients needed rebubbling twice. None of them developed CME. In Group 1, 2 pairs of corneas were from 1 donor each. None of the respective patients developed CME. In
3 cases in Group 1 and 2 cases in Group 2, partial removal of the anterior chamber air fill became necessary owing to insufficient iridectomy and angle closure mechanism with pronounced IOP elevation (>40 mm Hg). We observed 2 cases of elevated intraocular pressure (defined as IOP elevation >25 mm Hg) 6 weeks and 5 months post surgery, which were putatively related to the use of topical steroid eye drops. In 1 case the patient had a known history of open-angle glaucoma. Both cases were in Treatment Group 2. In both cases topical steroids were changed to rimexolon 1% (Vexol; Alcon, Freiburg, Germany). IOP elevation was treated temporarily with topical antiglaucomatous agents. IOP normalized within 2 weeks after adaption of therapy in both cases. We did not observe any other complications ascribable to the intensified use of topical steroids. EFFECT OF HOURLY TOPICAL STEROIDS:
In Group 1 we observed 9 cases of CME (12%, 95% confidence interval 6.4%–21.3%) (Figure). In all cases CME was observed within the first 3 months post surgery (Figure). All cases of CME appeared homogenously scattered over the observational period. In 8 cases CME resolved under intensified topical anti-inflammatory therapy (Figure). CME therapy consisted of topical steroids in all cases (prednisolone acetate 1% hourly for 1 week, then tapered to 83/d for 1 week, then to 53/d) and additional oral acetazolamide in 1 case (250 mg 23/d orally for 2 weeks). In 1 case (11%), despite additional therapy with intravitreal steroids (dexamethasone 700 mg [Ozurdex; Allergan, Frankfurt, Germany]) CME became persistent and did not resolve completely over an observation period of 12 months (#3, Figure). In Group 2 none of the patients developed
TABLE. Possible Confounding Factors for the Development of Cystoid Macular Edema After Posterior Lamellar Keratoplasty Combined With Cataract Surgery At Baseline
Male, n (%) Female, n (%) Age, mean 6 SD (range) BCVA, mean 6 SD [range] Central corneal thickness, mean 6 SD (range) Diabetes mellitus, n Open-angle glaucoma, n
Dry AMD, n Macular pucker,n
Group 1 (N ¼ 75) (Topical Steroids 53/d)
Group 2 (N ¼ 75) (Topical Steroids Hourly)
38 (51%) 37 (49%) 70 6 8 years (51–89) 0.51 6 0.22 logMAR (20/60 Snellen) [range 1.30–0.22] 631 6 82 mm (513–1038 mm) Total: 7 With CME: 1 Total: 2 With CME: 0 With prostaglandin therapy: 2 Total: 1 With CME: 1 Total: 3 With CME: 0
48 (64%) 27 (36%) 68 6 10 years (24–85) 0.51 6 0.38 logMAR (20/60 Snellen) (range 2.00–0.22) 642 6 87 mm (515–987 mm) Total: 4 With CME: 0 Total: 6 With CME: 0 With prostaglandin therapy: 0 Total: 3 With CME: 0 Total: 5 With CME
P Value
.137 .376 .157 .666 .533 .276
.620 .719
AMD ¼ age-related macular degeneration; BCVA ¼ best-corrected visual acuity; CME ¼ cystoid macular edema. All confounding factors at baseline were distributed equally among the groups. No relation to CME development could be seen.
VOL. -, NO. -
PROPHYLAXIS OF MACULAR EDEMA AFTER LAMELLAR KERATOPLASTY
3
FIGURE. Spectral-domain optical coherence tomography (SD OCT) of patients with cystoid macular edema (CME) after posterior lamellar keratoplasty combined with cataract surgery (triple-Descemet membrane endothelial keratoplasty). Rows represent different cases; columns represent the SD OCT slice before surgery, at the first detection of CME and after treatment for CME, respectively. Lower right corner of each slice: Postoperative days (d). In 8 cases CME resolved under topical steroids. In 1 case (#3) CME persisted despite intravitreal steroids (dexamethasone 700 mg). In Cases #4 and #6 the patients’ ophthalmologist performed the SD OCT.
4
AMERICAN JOURNAL OF OPHTHALMOLOGY
--- 2016
CME (95% confidence interval 0%–4.9%); difference between groups 12% (95% confidence interval 4.6%– 21.3%, P ¼ .003). BCVA after 6 months was 0.22 6 0.22 logMAR (approximately 20/30 Snellen) (range 1.30– 0.00) in Group 1 and did not differ from Group 2 (0.23 6 0.16 logMAR (approximately 20/30 Snellen) (range 1.00–0.00), P ¼ .208). EFFECT OF MACULAR EDEMA ON CENTRAL RETINAL THICKNESS AND VISUAL ACUITY: Mean central retinal
thickness (CRT) of patients with macular edema (occurring only in Group 1) was increased significantly compared to preoperative baseline (262 6 31 vs 507 6 170, P ¼ .004). After initiation of therapy mean CRT decreased again (349 6 199, P ¼ .039). Mean BCVA decreased by 0.15 6 0.3 logMAR. After initiation of therapy BCVA increased again by 0.16 6 0.2 logMAR. Because of high variability and the small number of cases, the difference did not reach significance level (P ¼ .262, P ¼ .055, respectively).
DISCUSSION CME AS A COMPLICATION AFTER TRIPLE-DMEK SURGERY IS
discussed to be at least partly of inflammatory origin.1,2,5,6 With respect to these publications, we have changed our routine early postoperative therapy regimen from topical steroids 5 times per day to an hourly application for the first postoperative week. We aimed to provide a more sufficient postoperative suppression of inflammation. We found that of all patients treated with topical steroids 5 times per day during the first postoperative week, 12% developed CME within the 6 months follow-up. By contrast none of the patients treated with topical steroids hourly for the first postoperative week suffered from CME during follow up. Our older rate of CME is in accordance with recently published results.5 Patients with CME suffered from a marked increase of CRT and a simultaneous decrease in visual acuity of 0.15 logMAR. This corresponds with approximately 1.5 lines Snellen BCVA. The prognosis of CME is generally good.5 There is no uniform therapy regimen recommended; however, most therapy regimens consist of intense topical steroidal or nonsteroidal anti-inflammatory agents. These are often complemented by oral carbonic anhydrase inhibitors, which can reduce the macular edema. Intravitreal or periocular injections of steroids or anti–vascular endothelial growth
factor (VEGF) agents are to be considered if noninvasive therapy fails.9,10 The fact that in our study, out of 9 patients with CME 1 patient developed a persistent CME, despite intravitreal steroid therapy, emphasizes the relevance of reducing the CME rate, if possible. However, several factors are suspected to play a role in the development of postoperative CME. After cataract surgery incomplete vitreous detachment from the macula seems to increase the risk for CME.11 This may be attributed to a mechanical irritation of the macula upon cataract surgery. Heinzelmann and associates suspected (triple)DMEK procedures to pose an increased risk for CME development, as compared to cataract surgery, since they include pronounced manipulation of the iris by iridectomy and intracameral air tamponade.5 Manipulation of the iris is known to be a risk factor for postoperative inflammation.9 Interestingly, a longer axial length seems to have a protective effect on CME development.5 Heinzelmann and associates explain this by a gradient of inflammatory factors from the iris along the axial length toward the macula.5 In addition, iris manipulation may be reduced in (myopic) eyes with deeper anterior chamber. The finding that patients with CME have increased aqueous flare values further supports the hypothesis of increased inflammatory factors originating from the iris.6,12 Like Heinzelmann and associates, we also found no direct influence of possible ocular or nonocular confounding factors (Table). The rate of rebubbling was slightly higher in the group with steroids 5 times per day. However, since in patients with CME rebubbling was not performed at higher rates compared to patients without CME, it does not seem to be causative for the increased rate of CME seen in Group 1. This is in accordance with the results of the aforementioned group.5 In 2 cases from Treatment Group 2 we observed elevated IOP. However, IOP was not elevated earlier than 6 weeks post surgery, while intensified topical steroids were administered only for 1 week and then tapered in a similar fashion in both groups. Although it seems likely that the elevated IOP is a response to the steroid therapy,13 the sample size does not allow ascription to the intensified topical steroids, administered only for a short period post surgery. We conclude that the simple measure of hourly topical steroids during the first postoperative week after tripleDMEK surgery can reduce the risk for later developing CME, as well as further complications like intravitreal injections or long-term persistent CME.
FUNDING/SUPPORT: TISHA PRABRIPUTALOONG STANZEL, BJOERN OLOV BACHMANN, AND CLAUS CURSIEFEN RECEIVED grants from the European Union (EU): COST BM 1302; FP7 STRONG (www.biocornea.eu) and from the German Research Foundation (DFG): FOR 2240 (www.for2240.de). Financial disclosures: Claus Cursiefen has the following financial disclosures: Consultant to Gene Signal, Lausanne Switzerland; Novaliq, Heidelberg, Germany. Lectures for Allergan, Frankfurt, Germany; Santen, Osaka Japan; Ursapharm, Saarbruecken, Germany; Alcon Pharma, Freiburg, Germany; Bayer, Leverkusen, Germany; Novartis, Berlin, Germany. The following authors have no financial disclosures: Robert Hoerster, Tisha Prabriputaloong Stanzel, Bjoern Olov Bachmann, Sebastian Siebelmann, and Moritz Felsch. All authors attest that they meet the current ICMJE criteria for authorship.
VOL. -, NO. -
PROPHYLAXIS OF MACULAR EDEMA AFTER LAMELLAR KERATOPLASTY
5
REFERENCES 1. Framme C, Wolf S. Retinal complications after damaging the vitreolenticular barrier. Ophthalmologica 2012;227(1): 20–33. 2. Packer M, Lowe J, Fine H. Incidence of acute postoperative cystoid macular edema in clinical practice. J Cataract Refract Surg 2012;38(12):2108–2111. 3. Dapena I, Ham L, Droutsas K, van Dijk K, Moutsouris K, Melles GR. Learning curve in Descemet’s membrane endothelial keratoplasty: first series of 135 consecutive cases. Ophthalmology 2011;118(11):2147–2154. 4. Suh LH, Yoo SH, Deobhakta A, et al. Complications of Descemet’s stripping with automated endothelial keratoplasty: survey of 118 eyes at One Institute. Ophthalmology 2008;115(9):1517–1524. 5. Heinzelmann S, Maier P, Bohringer D, Huther S, Eberwein P, Reinhard T. Cystoid macular oedema following Descemet membrane endothelial keratoplasty. Br J Ophthalmol 2015; 99(1):98–102. 6. Ersoy L, Caramoy A, Ristau T, Kirchhof B, Fauser S. Aqueous flare is increased in patients with clinically significant cystoid
6
7.
8.
9.
10. 11. 12.
13.
macular oedema after cataract surgery. Br J Ophthalmol 2013; 97(7):862–865. Heindl LM, Riss S, Laaser K, Bachmann BO, Kruse FE, Cursiefen C. Split cornea transplantation for 2 recipients review of the first 100 consecutive patients. Am J Ophthalmol 2011;152(4):523–532.e2. Altman D, Machin D, Bryant T, Gardner M, eds. Statistics With Confidence: Confidence Intervals and Statistical Guidelines. 2nd ed. Oxford, UK: Wiley; 2000:46–50. Zur D, Fischer N, Tufail A, Mones J, Loewenstein A. Postsurgical cystoid macular edema. Eur J Ophthalmol 2011;21(Suppl 6):62–68. Lobo C. Pseudophakic cystoid macular edema. Ophthalmologica 2012;227(2):61–67. Roldan M, Serrano JM. Macular edema and vitreous detachment. Ann Ophthalmol 1989;21(4):141–148. Ursell PG, Spalton DJ, Whitcup SM, Nussenblatt RB. Cystoid macular edema after phacoemulsification: relationship to blood-aqueous barrier damage and visual acuity. J Cataract Refract Surg 1999;25(11):1492–1497. Razeghinejad MR, Katz LJ. Steroid-induced iatrogenic glaucoma. Ophthalmic Res 2012;47(2):66–80.
AMERICAN JOURNAL OF OPHTHALMOLOGY
--- 2016
Biosketch Robert Hoerster, MD is specialist for Ophthalmology at the Center of Ophthalmology, University Hospital of Cologne since 2009. His main areas of research are mediation and prophylaxis of fibrotic diseases of the eye. He is a member of the Association for Research in Vision and Ophthalmology (ARVO), the Deutsche Ophthalmologische Gesellschaft (DOG) and the Deutsche Retinologische Gesellschaft (RG).
VOL. -, NO. -
PROPHYLAXIS OF MACULAR EDEMA AFTER LAMELLAR KERATOPLASTY
6.e1
Biosketch Claus Cursiefen, MD is Chairman and Professor of the Department of Ophthalmology, University of Cologne, Germany and Speaker of the DFG research Group FOR 2240 ‘‘Lymphangiogenesis and Cellular Immunity in Inflammatory Diseases of the Eye’’. Professor Cursiefen is a cornea specialist and Action Chair of EU COST Action BM1302 ‘‘Joining Forces in Corneal Regeneration’’. He did his Residency at the Department of Ophthalmology in Erlangen/Germany and his Postdoctoral Fellowship at Schepens Eye Research Institute, Boston.
6.e2
AMERICAN JOURNAL OF OPHTHALMOLOGY
--- 2016