ARTICLE
Effect of the eyelid speculum on pachymetry during corneal collagen crosslinking in keratoconus patients Nienke Soeters, BOptom, Erik van Bussel, BSc, Rikkert van der Valk, MD, MSc, PhD, Allegonda Van der Lelij, MD, PhD, Nayyirih G. Tahzib, MD, PhD, FEBOphth
PURPOSE: To compare central corneal thickness (CCT) with and without using an eyelid speculum during corneal collagen crosslinking (CXL). SETTING: University Medical Center Utrecht, Utrecht, the Netherlands. DESIGN: Prospective nonrandomized comparative study. METHODS: Eyes with progressive keratoconus were treated by CXL and consecutively divided into 2 groups. In Group A, an eyelid speculum remained in place throughout the entire CXL procedure. In Group B, the eyelids remained closed during the 30-minute riboflavin instillation. Intraoperative ultrasound pachymetry measurements were performed at different timepoints. The visual acuity, refraction, keratometry, pachymetry, and endothelium were evaluated 6 months after CXL. The main outcome measures were intraoperative CCT measurements and the clinical CXL effect after 6 months. RESULTS: Fifty-two eyes (50 patients) were treated. After riboflavin instillation, a statistically significant difference in CCT reduction was found between the 2 groups (P < .001), with a mean CCT decrease of 62 mm G 53 (SD) (13% G 11%) in Group A and 11 G 35 mm (2% G 8%) in Group B. No statistically significant between-group differences were found after epithelial removal or ultraviolet-A (UVA) irradiation. Six months after CXL, no statistically significant between-group difference was found in the visual acuity, refraction, keratometry, pachymetry, or endothelium. CONCLUSIONS: Avoidance of an eyelid speculum during riboflavin instillation resulted in less CCT reduction during CXL. This finding could increase the chance of attaining the required pachymetry safety margin for applying UVA and thus decrease the chance of premature CXL treatment termination. Financial Disclosure: No author has a financial or proprietary interest in any material or method mentioned. J Cataract Refract Surg 2014; 40:575–581 Q 2014 ASCRS and ESCRS
Keratoconus is a degenerative corneal disorder that presents with corneal thinning and topographic cone formation, leading to irregular corneal shape, astigmatism, and visual loss.1 The disorder affects approximately 1 in 2000 individuals, is generally diagnosed in the late teenage years, and can be progressive in nature. The exact cause of keratoconus is unknown, although different associations and genetic factors have been described.2 Corneal collagen crosslinking (CXL), which is a molecular crosslinking of collagen fibers, has the ability to increase the mechanical corneal rigidity in
Q 2014 ASCRS and ESCRS Published by Elsevier Inc.
keratoconic eyes and to decrease corneal weakening and thinning.3 The standard CXL procedure described by Wollensak et al.4 consists of a corneal abrasion and then 30 minutes of isotonic riboflavin (vitamin B2) instillation followed by 30 minutes of ultraviolet-A (UVA) light irradiation. The riboflavin absorbs the UVA, resulting in a UVA transmission of only 7% across the cornea.5 The same group calculated that in human corneas thinner than 400 mm, a toxic UVA level is reached at the endothelium.6 Therefore, pachymetry measurements should be performed routinely before recommending CXL procedures to
0886-3350/$ - see front matter http://dx.doi.org/10.1016/j.jcrs.2013.08.060
575
576
EYELID SPECULUM DURING CXL
identify unsuitable cases.7 Since 2009, the possibility of treating corneas thinner than 400 mm using a hypoosmolar CXL technique has been studied and has shown good results in stabilizing the cornea without causing stromal scars as long as the stromal corneal thickness is at least 330 mm before administration of the hypoosmolar drops.8–11 In addition to pre-CXL pachymetry measurements, monitoring pachymetry during a CXL procedure is important because pachymetry does not necessarily remain stable during CXL.12 Even though there is now the possibility of hypoosmolar CXL procedures, unnecessary corneal thinning during a CXL procedure should be avoided. Kymionis et al.13 report a significant decrease in central corneal thickness (CCT) directly after riboflavin instillation. Our research group subsequently found stable CCT values after riboflavin instillation in 3 eyes treated using a slightly adjusted CXL technique.14 In this short report, we related our preliminary findings to the fact that no eyelid speculum was used during the 30-minute riboflavin instillation, thereby avoiding prolonged corneal dehydration by tear evaporation and endothelial pump mechanisms.15,16 The current study is an extension of our abovementioned small case study. In this study, we evaluated and compared CCT during CXL in 2 groups with continued and discontinued use of an eyelid speculum during riboflavin instillation to analyze pachymetry during CXL. PATIENTS AND METHODS Study Group and Protocol This prospective nonrandomized study enrolled patients diagnosed with documented progressive keratoconus who were scheduled to have a CXL procedure at the University Medical Center Utrecht (UMCU), the Netherlands. Written informed consent was obtained in accordance with UMCU guidelines, and the study complied with the Declaration of Helsinki. Submitted: May 29, 2013. Final revision submitted: August 20, 2013. Accepted: August 27, 2013. From the Utrecht Cornea Research Group (Soeters, van Bussel, van der Valk, Van der Lelij), Department of Ophthalmology, University Medical Center Utrecht, Utrecht, and Zonnestraal Eye Hospital (Tahzib), Amersfoort, the Netherlands. Supported by the Dr. F.P. Fischer Stichting, Amersfoort, and Stichting Nederlands Oogheelkundig Onderzoek, Rotterdam, the Netherlands (research grant to Dr. Soeters). Corresponding author: Nienke Soeters, BOptom, Utrecht Cornea Research Group, Department of Ophthalmology, University Medical Center Utrecht, HP E03.136, Heidelberglaan 100, 3508 GX Utrecht, the Netherlands. E-mail: [email protected].
Patients were consecutively divided into 2 study groups and were not matched for age, keratometry (K), or pachymetry properties. Group A included eyes that were treated according to standard international CXL protocol17 with continued use of an eyelid speculum throughout the entire procedure. Group B consisted of eyes that were treated by standard UMCU protocol in which no eyelid speculum was used during the 30-minute riboflavin instillation. The eyes remained closed during these 30 minutes and were manually opened for a few seconds every 3 minutes by the person administering the isotonic riboflavin drops. Data in both study groups were prospectively collected from consecutively planned treatments after approval of the Medical Ethics Committee UMCU. Inclusion criteria were defined as a clear central cornea, a minimum CCT of 400 mm before UVA application, the absence of corneal scars, and a documented keratometric progression over 6 to 12 months. Patients who did not meet the inclusion criteria and patients who had a hypoosmolar treatment were excluded. All patients discontinued contact lens wear at least 2 weeks before preoperative examinations and all follow-up measurements. At the pre-CXL examination and follow-up visits, uncorrected (UDVA) and corrected (CDVA) distance visual acuities, refraction, anterior segment imaging (including corneal topography and optical pachymetry), tonometry, endothelial cell density (ECD), and slitlamp examinations were obtained. The CXL effect was defined as stabilization or flattening of the central K and maximum K values and stable or improved visual acuity.
Central Corneal Thickness Measurements and Devices The CCT measurements during CXL were performed with a handheld ultrasound (US) pachymeter (Handy Pachymeter, SP-3000, Tomey Corp.). The US CCT was measured with the tip of the pachymeter perpendicular to the cornea as centrally as possible and was performed at the following timepoints during the CXL procedure: (1) directly after epithelial abrasion, (2) directly after 30 minutes of riboflavin instillation, and (3) after 30 minutes of UVA irradiation. The mean of 10 consecutive measurements was used. If the CCT was less than 400 mm at timepoint 2, additional measurements were taken in intervals of 3 extra minutes for a maximum of 12 minutes, with both eyes closed. When the CCT did not increase after 12 minutes, hypoosmolar riboflavin was used and the patient was excluded from analyses. In cases in which the CCT remained less than 400 mm after hypoosmolar riboflavin instillation, the CXL procedure was terminated (ie, no UVA irradiation was applied) and the patient was excluded from the study analysis. Optical pachymetry and keratometry data were determined using a Scheimpflug device (Pentacam HR, Oculus Optikger€ ate GmbH). Corneal ECD was analyzed by specular microscopy (SP3000P, Topcon) preoperatively and at the 6-month follow-up visit. Corneal CXL was performed with the UV-X system (Peschke Meditrade GmbH) (370 nm and 3 mW/cm2).
Surgical Technique After the instillation of local anesthetic eyedrops comprising oxybuprocaine 0.4% and tetracaine 1.0%, an
J CATARACT REFRACT SURG - VOL 40, APRIL 2014
EYELID SPECULUM DURING CXL
eyelid speculum was inserted and a 9.0 mm corneal abrasion was performed using a blunt knife. The cornea was rinsed with a balanced salt solution, and US CCT measurements were performed. After the abrasion, the eyelid speculum remained in place in Group A, while in Group B, the eyelid speculum was removed and patients were instructed to keep their eyes closed. In both groups, isotonic riboflavin 0.1% solution (Medio Cross) was instilled at 3-minute intervals over a 30-minute period. Ultrasound CCT measurements were repeated after rinsing with a balanced salt solution. When the CCT was less than 400 mm at this timepoint, additional measurements were taken at intervals of 3 extra minutes for a maximum of 12 minutes with both eyes closed. When the CCT remained less than 400 mm after 12 minutes, hypoosmolar riboflavin was administered and the patient was excluded from analysis. In Group A, the eyelid speculum remained in place in preparation for the UVA phase. In Group B, the eyelid speculum was reinserted. In both groups, UVA irradiation was performed during 30 minutes, while isotonic riboflavin solution was reapplied to the cornea every 5 minutes. Finally, US CCT measurements were repeated and a bandage lens was placed. Oral pain medication was given for the first day in addition to preservative-free artificial tears (dextran–hypromellose, Duratears Free) for 4 weeks, nepafenac drops (Nevanac) for 1 week, and moxifloxacin hydrochloride eyedrops (Vigamox) for 4 weeks, all 3 times daily. When epithelial healing was complete, the bandage lens was removed and a fluorometholone eyedrop (FML) was applied twice daily for 2 weeks.
Statistical Analysis For pachymetry, a sample size of 24 eyes in each study group in the study indicated a power of 0.88 for a 2-sided test (a Z 0.05, based on literature s Z 37).13 For clinical outcomes (K changes) after CXL, a sample size of 14 eyes in each group was indicated based on the first 3 patients in each group (power of 0.8, a Z 0.05). The mean of 10 consecutive US CCT measurements was used for statistical analysis and compared at different timepoints. The difference in the mean US-measured CCT at different timepoints during CXL, visual acuity, refraction, keratometry, pachymetry, and endothelium were determined with the independent samples t test. Differences in parameters per group between baseline and 6 months after CXL were determined with a paired-samples t test. The Pearson chi-square test was used for the analysis of the extra waiting time difference between Group A and Group B. A P value of less than 0.05 was considered statistically significant. Data collection and analysis were performed using SPSS for Windows software (version 20.0, International Business Machines Corp., SPSS Statistics). The US-measured CCT after the corneal abrasion was considered to be the baseline CCT because CXL treatment was equal in both groups until this timepoint. The CCT differences were evaluated in microns (absolute) as well as in percentages (relative).
577
26 eyes of 26 patients. All were diagnosed with progressive keratoconus and complied with the CXL inclusion criteria. Four eyes, 2 in each group, were excluded from the study because they did not meet the required 400 mm CCT before UVA application during treatment. No complications occurred during the CXL procedure in any case. After exclusions, Group A, in which the eyelid speculum was used throughout the entire CXL procedure, consisted of 17 men and 7 women with a mean age of 23 years G 9 (SD) (range 13 to 49 years). Group B, in which no eyelid speculum was used during the 30-minute riboflavin instillation, consisted of 15 men and 9 women with a mean age of 26 G 7 years (range 18 to 42 years).
Ultrasound Central Corneal Thickness After Epithelial Removal (Baseline Central Corneal Thickness) No statistically significant difference in CCT after corneal abrasion was found between the 2 study groups. The mean CCT was 446 G 35 mm in Group A and 435 G 41 mm in Group B (P Z .329).
Ultrasound Central Corneal Thickness After Riboflavin Instillation A statistically significant difference in absolute and relative CCT reduction was found between Group A and Group B after the riboflavin instillation phase (P ! .001). The mean CCT decreased by 62 G 53 mm (13% G 11%) in Group A and by 11 G 35 mm (2% G 8%) in Group B (Figure 1).
RESULTS Baseline Characteristics This study evaluated 52 eyes of 50 patients who were divided into 2 groups; each group comprised
Figure 1. Central corneal thickness at different timepoints during CXL in Group A and Group B (CI Z confidence interval; UVA Z ultraviolet A).
J CATARACT REFRACT SURG - VOL 40, APRIL 2014
578
EYELID SPECULUM DURING CXL
Extra Hydration Time After Riboflavin Instillation After riboflavin instillation, 14 (58%) of 24 eyes in Group A (95% confidence interval [CI], 38%-78%) and 5 (21%) of 24 eyes in Group B (95% CI, 4%-37%) required extra hydration time in intervals of 3 minutes to ensure the required safety thickness of 400 mm before the start of UVA irradiation. The difference between the 2 groups was statistically significant (P Z .008, Pearson c2 test). A mean extra hydration time of 4 G 4 minutes was required in Group A and of 2 G 4 minutes in Group B. Statistical analysis of the difference in the mean extra hydration time between Group A and Group B showed an insignificant trend toward a longer interval period in Group A (P Z .067) (Figure 1). Ultrasound Central Corneal Thickness After Ultraviolet-A Irradiation The mean CCT decreased significantly in both groups during the UVA irradiation phase (both P ! .001). The mean decrease was 11% G 8% in Group A and 11% G 6% in Group B (Figure 1). No statistically significant difference in US-measured CCT was found between the 2 groups (P Z .839). Overall Ultrasound Central Corneal Thickness During Corneal Collagen Crosslinking Overall, the mean decrease in US-measured CCT from baseline thickness (ie, after epithelial removal) to the end of CXL treatment (ie, after UVA irradiation) in both groups was statistically significant (both P ! .001). The mean decrease was 15% G 10% in Group A and 12% G 7% in Group B. No statistically significant difference was found between the 2 groups (P Z .201) (Figure 2). However, the extra time needed to reach a minimum thickness of 400 mm after riboflavin instillation was taken into account, meaning that the total CXL time was not equal in all patients. Visual Acuity and Refraction Six months after CXL in both groups, the CDVA improved significantly (in Group A from 0.50 to 0.70 Snellen decimal; in Group B from 0.33 to 0.60 Snellen decimal). The UDVA improved significantly in Group A and remained stable in Group B (Table 1). The UDVA and CDVA improvement was not statistically significant different between the 2 groups. The refractive cylinder increased in both groups, in Group A by a mean of 1.0 G 2 diopters (D) (P Z .111) and in Group B by a mean of 0.6 G 1.7 D (P Z .026). The mean spherical equivalent (SE) values remained stable in both groups. There was no statistically significant difference in refractive cylinder or SE change between the 2 groups.
Figure 2. Overall CCT during CXL in Group A and Group B (CI Z confidence interval; UVA Z ultraviolet A).
Keratometry Six months after CXL, the flattest central K value decreased significantly in Group A; the mean decrease was 0.6 G 1.5 D in Group A and 0.4 G 2.1 D in Group B. The maximum K value showed a statistically significant decrease in Group B; the mean decrease was 0.66 G 1.7 D in Group A and 1.75 G 2.5 D in Group B. There was no statistically significant difference in keratometry change between the 2 groups (Table 1). Pachymetry and Endothelial Cell Count (Endothelium Cell Density) Six months after CXL, a statistically significant decrease in pachymetry at the central cornea, apex, and thinnest corneal location was found in both groups (Table 1). There was no statistically significant difference in the pachymetry decrease between the 2 groups. The ECD remained stable in both groups and between groups (Table 1). Epithelial Healing In 7 (29%) of 24 eyes in Group A and in 5 (21%) of 24 eyes in Group B, the epithelium had not closed at the 1-week checkup after CXL. This difference was not statistically significant (P Z .740, Fisher exact test). In all patients, the epithelium was closed within 2 weeks. Complications Six months after CXL, in Group A, 2 eyes had stromal haze and 1 eye had superficial haze. In Group B, 2 eyes had superficial haze. Six months after CXL, 4 eyes had a maximum increase in the K value of more than 1.0 D (3 in Group A; 1 in
J CATARACT REFRACT SURG - VOL 40, APRIL 2014
579
EYELID SPECULUM DURING CXL
Table 1. Effect of corneal CXL between Group A and Group B after 6 months compared with baseline.
Group/Exam Group A Preop 6 mo postop P value Group B Preop 6 mo postop P value Between groups P value
UDVA (LogMAR)
CDVA (LogMAR)
SE (D)
Kmax (D)
Ksteep (D)
Kflat (D)
Kmean (D)
CTcenter (mm)
0.90 0.65 .001*
0.27 0.18 .012*
1.5 1.3 .544
55.8 55.2 .076
49.4 48.9 .156
46.1 45.5 .038*
47.7 47.1 .078
1.00 0.85 .141
0.47 0.26 .036*
2.4 2.2 .964
59.5 57.8 .004*
52.0 51.6 .080
47.1 46.7 .421
49.4 49.0 .274
.502
.191
.731
.821
.632
.710
.092
CTapex (mm)
CTthin (mm)
ECD (Cells/mm2)
496 479 .002*
489 470 .001*
474 454 !.001*
2800 2870 .333
491 477 .001*
477 464 !.001*
458 444 .001*
2827 2746 .134
.276
.517
.641
.447
CDVA Z corrected distance visual acuity; CT Z corneal thickness (optical pachymetry by Scheimpflug device at the center [CTcenter], apex [CTapex], or thinnest [CTthin] point); ECD Z endothelial cell density; KflatZ flattest central keratometry; Kmax Z maximum keratometry; Kmean Z mean central keratometry; Ksteep Z steepest central keratometry; SE Z spherical equivalent; UDVA Z uncorrected distance visual acuity *Statistically significant
Group B). Two eyes lost more than 2 Snellen lines of CDVA (Group B); 3 eyes lost more than 2 Snellen lines of UDVA (Group B). DISCUSSION This study showed that the avoidance of an eyelid speculum during riboflavin instillation in a CXL procedure prevented corneal thinning before the UVA irradiation phase. This confirms the hypothesis of our recently published case report14 and is in agreement with the report of Schmidinger et al.18 In human corneas thinner than 400 mm, a toxic UVA level is reached at the endothelium during CXL. Therefore, minimizing corneal thinning before UVA irradiation was recommended by Wollensak et al.6 Endothelial damage has been described earlier in young patients with thin corneas below the 400 mm safety limit who were nevertheless treated by standard CXL.11 In the current study, the 400 mm safety limit at the start of UVA irradiation was maintained. However, during the UVA phase, the CCT decreased in both groups. This finding is in agreement with Schmidinger et al.,18 who also found a reduction in CCT, even 10 minutes after the start of UVA. No significant endothelial changes were observed during slitlamp examination or measured 6 months after CXL. In our study, 4 patients were excluded because they eventually had a hypoosmolar CXL treatment. This number is in line with clinical observations in which some eyes also react with disproportional corneal thinning after the application of isotonic riboflavin.9,19 The corresponding sample size in this study was not sufficient to form statistical conclusions about this number of exclusions.
In our study, when the use of an eyelid speculum was avoided during riboflavin instillation (Group B), a mean CCT decrease of merely 2% was measured directly after this timepoint, compared with 13% when the eyelid speculum was used (Group A) during riboflavin instillation. It appeared that in Group A, significantly more patients required extra time after riboflavin instillation to reach a CCT of 400 mm before irradiation, and this extra time leveled the mean CCT before UVA irradiation in both groups. During the intervals of extra time, both eyes remained closed, which contributed to a CCT increase due to rehydration of the cornea. As expected, we found no statistical overall CCT difference after UVA irradiation between the groups because the treatment procedure was equal in both groups during the final irradiation phase. Our results in Group A are a confirmation of those in a noncomparative study by Kymionis et al.13 in which there was a significant CCT decrease (20%) after riboflavin instillation with the use of an eyelid speculum. The explanation for the difference in CCT decrease compared with our study (20% versus 13%) can be sought in a baseline CCT after epithelium removal of 415 mm in their study group compared with 446 mm in our study; thinner corneas seem to dehydrate faster as a result of the stromal swelling pressure minus the hydrostatic compressive effect of the intraocular pressure. Swelling pressure was calculated using the relationships of corneal thickness to hydration.16 When the eyelids remain open, the cornea is known to dehydrate by tear evaporation and endothelial pump mechanisms, leading to thinner CCT values.15,16 In general, CCT values decrease significantly during corneal exposure for about 11 mm/min, or 10% after 5 minutes.20 Furthermore, the ingredient dextran in
J CATARACT REFRACT SURG - VOL 40, APRIL 2014
580
EYELID SPECULUM DURING CXL
riboflavin drops is osmotically active and prevents the cornea from swelling by extracting water from the eye.9 The influence of the eyelid speculum on changes in osmolarity of the riboflavin–dextran solution and the cornea could be an explanation for the difference in CCT results between Group A and Group B. On the other hand, epithelial removal can lead to corneal swelling.21 The total duration of CXL varied among patients due to the extra time intervals after riboflavin instillation in certain cases, which made it impossible to make an overall time-independent comparison of CCT changes between the 2 groups. It was not statistically proven that the CCT decrease during the overall CXL treatment was limited in patients treated without the eyelid speculum during the riboflavin phase (Group B). Another uncertainty of our modified protocol is that it is unknown whether the riboflavin concentration level changes when the eyelids remain closed. During UVA irradiation, the exposed cornea received riboflavin every 5 minutes, which might have neutralized the riboflavin concentration of the first phase. That there was no decrease in ECD in both groups 6 months after CXL might indicate that the concentration level of riboflavin was sufficient during the overall treatment. Longer follow-up is necessary to confirm the absence of differences in long-term ECD loss. The significant decrease in pachymetry that we found in both groups in our study 6 months after CXL is a common phenomenon after CXL.22 No statistically significant difference between the 2 groups was noted, which could mean that the significant CCT decrease in Group A during riboflavin instillation did not necessarily influence post-CXL pachymetry. In another study,23 our group recently reported that advanced keratoconus cases can show a more pronounced CXL-induced flattening effect. There was a difference in the average maximum K decrease between the 2 groups in the current study ( 0.60 D in Group A and 1.75 D in Group B); this could be because Group B contained more eyes with a steeper cornea at baseline (59.5 D in Group B versus 55.8 D in Group A). When comparing the overall 6-month post-CXL effect between the 2 groups, the significant CCT decrease during riboflavin instillation noted in Group A did not seem to significantly influence any clinical parameter after CXL. However, the extra waiting time to reach the required 400 mm could be a confounding factor for differences between the 2 groups, especially in the context of endothelial cell loss, as has been published.11 Concerning the study design and practical execution, all patients were treated at the same institution
by the same practitioners in a relatively short time interval (6 months), minimizing the risk for bias. In addition, because the baseline characteristics in Group A and Group B were similar, the 2 groups were statistically comparable. The absence of a significant difference in CCT change during the similar phases of the CXL treatment (ie, during corneal abrasion and UVA irradiation) was as expected and reconfirmed the comparability of the 2 groups. We used US rather than optical pachymetry measurements during CXL because the literature shows a major sensibility to inhomogeneous structures in crosslinked corneas, leading to underestimation of pachymetry of corneas measured with Scheimpflug imaging (eg, Pentacam).24 The results in the current study are clear and directly implementable in daily clinical use by simply eliminating the use of the eyelid speculum during the riboflavin phase. Another advantage of this technique, in addition to minimizing prolonged corneal dehydration and thus corneal thinning, is increased subjective patient comfort during the procedure. Patients in Group A wore the eyelid speculum during the entire procedure, meaning 65 to 70 minutes in total, and patients in Group B wore the eyelid speculum in time blocks of 5 to 10 minutes (during corneal abrasion) and then 30 minutes (during irradiation).25,26 However, this subjective patient advantage was not objectively evaluated. In conclusion, this study provides evidence that avoidance of the eyelid speculum during riboflavin instillation prevents unwanted perioperative corneal thinning, thereby increasing the chance of sufficient corneal thickness before UVA irradiation. This can reduce the need for hypoosmolar CXL treatments and thus avoid a prolonged and higher risk.
WHAT WAS KNOWN Central corneal thickness does not necessarily remain stable during CXL. Even though there is the possibility of hypoosmolar CXL procedures, unnecessary corneal thinning during a CXL procedure should be avoided. WHAT THIS PAPER ADDS Avoidance of the eyelid speculum during riboflavin instillation prevented unwanted perioperative corneal thinning before UVA irradiation. This outcome can reduce the need for hypoosmolar CXL treatments.
J CATARACT REFRACT SURG - VOL 40, APRIL 2014
581
EYELID SPECULUM DURING CXL
REFERENCES 1. Krachmer JH, Feder RS, Belin MW. Keratoconus and related noninflammatory corneal thinning disorders. Surv Ophthalmol 1984; 28:293–321 2. Rabinowitz YS. Keratoconus. Surv Ophthalmol 1998; 42:297– 319. Available at: http://www.keratoconus.com/resources/ MajorCReview-Keratoconus.pdf. Accessed January 4, 214 3. Spoerl E, Huhle M, Seiler T. Induction of cross-links in corneal tissue. Exp Eye Res 1998; 66:97–103 4. Wollensak G, Spoerl E, Seiler T. Riboflavin/ultraviolet-A– induced collagen crosslinking for the treatment of keratoconus. Am J Ophthalmol 2003; 135:620–627. Available at: http:// grmc.ca/assets/files/collagen_crosslinking_2003_wollensak.pdf. Accessed January 4, 2014 5. Wollensak G. Crosslinking treatment of progressive keratoconus: new hope. Curr Opin Ophthalmol 2006; 17:356–360 €rl E, Reber F, Pillunat L, Funk R. Corneal 6. Wollensak G, Spo endothelial cytotoxicity of riboflavin/UVA treatment in vitro. Ophthalmic Res 2003; 35:324–328 7. Wollensak G, Spoerl E, Wilsch M, Seiler T. Endothelial cell damage after riboflavin–ultraviolet-A treatment in the rabbit. J Cataract Refract Surg 2003; 29:1786–1790 8. Raiskup F, Kißner A, Spoerl E, Pillunat LE. Hornhautvernetzung € sung beim Keratokonus mit mit hypoosmolarer Riboflavin-Lo € nner Hornhaut [Corneal cross-linking with hypo-osmolar ribodu flavin solution for keratoconus with thin corneas]. Ophthalmologe 2011; 108:846–851 9. Hafezi F, Mrochen M, Iseli HP, Seiler T. Collagen crosslinking with ultraviolet-A and hypoosmolar riboflavin solution in thin corneas. J Cataract Refract Surg 2009; 35:621–624 10. Hafezi F. Limitation of collagen cross-linking with hypoosmolar riboflavin solution: failure in an extremely thin cornea. Cornea 2011; 30:917–919 11. Kymionis GD, Portaliou DM, Diakonis VF, Kounis GA, Panagopoulou SI, Grentzelos MA. Corneal collagen crosslinking with riboflavin and ultraviolet-A irradiation in patients with thin corneas. Am J Ophthalmol 2012; 153:24–28 12. Wollensak G, Aurich H, Wirbelauer C, Sel S. Significance of the riboflavin film in corneal collagen crosslinking. J Cataract Refract Surg 2010; 36:114–120 13. Kymionis GD, Kounis GA, Portaliou DM, Grentzelos MA, Karavitaki AE, Coskunseven E, Jankov MR, Pallikaris IG. Intraoperative pachymetric measurements during corneal collagen cross-linking with riboflavin and ultraviolet A irradiation. Ophthalmology 2009; 116:2336–2339 14. Tahzib NG, Soeters N, Van der Lelij A. Pachymetry during crosslinking [letter]. Ophthalmology 2010; 117:2041–2041.e1; reply by GD Kymionis, DM Portaliou 2041–2042 15. Bourassa S, Benjamin WJ, Boltz RL. Effect of humidity on the deswelling function of the human cornea. Curr Eye Res 1991; 10:493–500
16. O’Neal MR, Polse KA. In vivo assessment of mechanisms controlling corneal hydration. Invest Ophthalmol Vis Sci 1985; 26:849–856. Available at: http://www.iovs.org/content/26/6/ 849.full.pdf. Accessed January 4, 2014 17. Spoerl E, Mrochen M, Sliney D, Trokel S, Seiler T. Safety of UVA-riboflavin cross-linking of the cornea. Cornea 2007; 26:385–389 18. Schmidinger G, Pachala M, Prager F. Pachymetry changes during corneal crosslinking: effect of closed eyelids and hypotonic riboflavin solution. J Cataract Refract Surg 2013; 39:1179–1183 € Intraoperative corneal thickness 19. Kaya V, Utine CA, Yılmaz OF. measurements during corneal collagen cross-linking with hypoosmolar riboflavin solution in thin corneas. Cornea 2012; 31:486–490 20. Aurich H, Wirbelauer C, Jaroszewski J, Hartmann C, Pham DT. Continuous measurement of corneal dehydration with online optical coherence pachymetry. Cornea 2006; 25:182–184 21. Maurice DM, Giardini AA. Swelling of the cornea in vivo after the destruction of its limiting layers. Br J Ophthalmol 1951; 35:791– 797. Available at: http://www.pubmedcentral.nih.gov/picrender. fcgi?artidZ1323849&blobtypeZpdf. Accessed January 4, 2014 22. Greenstein SA, Shah VP, Fry KL, Hersh PS. Corneal thickness changes after corneal collagen crosslinking for keratoconus and corneal ectasia: one-year results. J Cataract Refract Surg 2011; 37:691–700 23. Sloot F, Soeters N, van der Valk R, Tahzib NG. Effective corneal collagen crosslinking in advanced cases of progressive keratoconus. J Cataract Refract Surg 2013; 39:1141–1145 24. Mencucci R, Paladini I, Virgili G, Giacomelli G, Menchini U. Corneal thickness measurements using time-domain anterior segment OCT, ultrasound, and Scheimpflug tomographer pachymetry before and after corneal cross-linking for keratoconus. J Refract Surg 2012; 28:562–566 25. Cheng ACK, Young AL, Law RWK, Lam DSC. Prospective randomized double-masked trial to evaluate perioperative pain profile in different stages of simultaneous bilateral LASIK. Cornea 2006; 25:919–922 26. Davis MJ, Pollack JS, Shott S. Comparison of topical anesthetics for intravitreal injections; a randomized clinical trial. Retina 2012; 32:701–705
J CATARACT REFRACT SURG - VOL 40, APRIL 2014
First author: Nienke Soeters, BOptom Utrecht Cornea Research Group, Department of Ophthalmology, University Medical Center Utrecht, Utrecht, the Netherlands
Effect of the eyelid speculum on pachymetry during corneal collagen crosslinking in keratoconus patients Nienke Soeters, BOptom, Erik van Bussel, BSc, Rikkert van der Valk, MD, MSc, PhD, Allegonda Van der Lelij, MD, PhD, Nayyirih G. Tahzib, MD, PhD, FEBOphth
PURPOSE: To compare central corneal thickness (CCT) with and without using an eyelid speculum during corneal collagen crosslinking (CXL). SETTING: University Medical Center Utrecht, Utrecht, the Netherlands. DESIGN: Prospective nonrandomized comparative study. METHODS: Eyes with progressive keratoconus were treated by CXL and consecutively divided into 2 groups. In Group A, an eyelid speculum remained in place throughout the entire CXL procedure. In Group B, the eyelids remained closed during the 30-minute riboflavin instillation. Intraoperative ultrasound pachymetry measurements were performed at different timepoints. The visual acuity, refraction, keratometry, pachymetry, and endothelium were evaluated 6 months after CXL. The main outcome measures were intraoperative CCT measurements and the clinical CXL effect after 6 months. RESULTS: Fifty-two eyes (50 patients) were treated. After riboflavin instillation, a statistically significant difference in CCT reduction was found between the 2 groups (P < .001), with a mean CCT decrease of 62 mm G 53 (SD) (13% G 11%) in Group A and 11 G 35 mm (2% G 8%) in Group B. No statistically significant between-group differences were found after epithelial removal or ultraviolet-A (UVA) irradiation. Six months after CXL, no statistically significant between-group difference was found in the visual acuity, refraction, keratometry, pachymetry, or endothelium. CONCLUSIONS: Avoidance of an eyelid speculum during riboflavin instillation resulted in less CCT reduction during CXL. This finding could increase the chance of attaining the required pachymetry safety margin for applying UVA and thus decrease the chance of premature CXL treatment termination. Financial Disclosure: No author has a financial or proprietary interest in any material or method mentioned. J Cataract Refract Surg 2014; 40:575–581 Q 2014 ASCRS and ESCRS
Keratoconus is a degenerative corneal disorder that presents with corneal thinning and topographic cone formation, leading to irregular corneal shape, astigmatism, and visual loss.1 The disorder affects approximately 1 in 2000 individuals, is generally diagnosed in the late teenage years, and can be progressive in nature. The exact cause of keratoconus is unknown, although different associations and genetic factors have been described.2 Corneal collagen crosslinking (CXL), which is a molecular crosslinking of collagen fibers, has the ability to increase the mechanical corneal rigidity in
Q 2014 ASCRS and ESCRS Published by Elsevier Inc.
keratoconic eyes and to decrease corneal weakening and thinning.3 The standard CXL procedure described by Wollensak et al.4 consists of a corneal abrasion and then 30 minutes of isotonic riboflavin (vitamin B2) instillation followed by 30 minutes of ultraviolet-A (UVA) light irradiation. The riboflavin absorbs the UVA, resulting in a UVA transmission of only 7% across the cornea.5 The same group calculated that in human corneas thinner than 400 mm, a toxic UVA level is reached at the endothelium.6 Therefore, pachymetry measurements should be performed routinely before recommending CXL procedures to
0886-3350/$ - see front matter http://dx.doi.org/10.1016/j.jcrs.2013.08.060
575
576
EYELID SPECULUM DURING CXL
identify unsuitable cases.7 Since 2009, the possibility of treating corneas thinner than 400 mm using a hypoosmolar CXL technique has been studied and has shown good results in stabilizing the cornea without causing stromal scars as long as the stromal corneal thickness is at least 330 mm before administration of the hypoosmolar drops.8–11 In addition to pre-CXL pachymetry measurements, monitoring pachymetry during a CXL procedure is important because pachymetry does not necessarily remain stable during CXL.12 Even though there is now the possibility of hypoosmolar CXL procedures, unnecessary corneal thinning during a CXL procedure should be avoided. Kymionis et al.13 report a significant decrease in central corneal thickness (CCT) directly after riboflavin instillation. Our research group subsequently found stable CCT values after riboflavin instillation in 3 eyes treated using a slightly adjusted CXL technique.14 In this short report, we related our preliminary findings to the fact that no eyelid speculum was used during the 30-minute riboflavin instillation, thereby avoiding prolonged corneal dehydration by tear evaporation and endothelial pump mechanisms.15,16 The current study is an extension of our abovementioned small case study. In this study, we evaluated and compared CCT during CXL in 2 groups with continued and discontinued use of an eyelid speculum during riboflavin instillation to analyze pachymetry during CXL. PATIENTS AND METHODS Study Group and Protocol This prospective nonrandomized study enrolled patients diagnosed with documented progressive keratoconus who were scheduled to have a CXL procedure at the University Medical Center Utrecht (UMCU), the Netherlands. Written informed consent was obtained in accordance with UMCU guidelines, and the study complied with the Declaration of Helsinki. Submitted: May 29, 2013. Final revision submitted: August 20, 2013. Accepted: August 27, 2013. From the Utrecht Cornea Research Group (Soeters, van Bussel, van der Valk, Van der Lelij), Department of Ophthalmology, University Medical Center Utrecht, Utrecht, and Zonnestraal Eye Hospital (Tahzib), Amersfoort, the Netherlands. Supported by the Dr. F.P. Fischer Stichting, Amersfoort, and Stichting Nederlands Oogheelkundig Onderzoek, Rotterdam, the Netherlands (research grant to Dr. Soeters). Corresponding author: Nienke Soeters, BOptom, Utrecht Cornea Research Group, Department of Ophthalmology, University Medical Center Utrecht, HP E03.136, Heidelberglaan 100, 3508 GX Utrecht, the Netherlands. E-mail: [email protected].
Patients were consecutively divided into 2 study groups and were not matched for age, keratometry (K), or pachymetry properties. Group A included eyes that were treated according to standard international CXL protocol17 with continued use of an eyelid speculum throughout the entire procedure. Group B consisted of eyes that were treated by standard UMCU protocol in which no eyelid speculum was used during the 30-minute riboflavin instillation. The eyes remained closed during these 30 minutes and were manually opened for a few seconds every 3 minutes by the person administering the isotonic riboflavin drops. Data in both study groups were prospectively collected from consecutively planned treatments after approval of the Medical Ethics Committee UMCU. Inclusion criteria were defined as a clear central cornea, a minimum CCT of 400 mm before UVA application, the absence of corneal scars, and a documented keratometric progression over 6 to 12 months. Patients who did not meet the inclusion criteria and patients who had a hypoosmolar treatment were excluded. All patients discontinued contact lens wear at least 2 weeks before preoperative examinations and all follow-up measurements. At the pre-CXL examination and follow-up visits, uncorrected (UDVA) and corrected (CDVA) distance visual acuities, refraction, anterior segment imaging (including corneal topography and optical pachymetry), tonometry, endothelial cell density (ECD), and slitlamp examinations were obtained. The CXL effect was defined as stabilization or flattening of the central K and maximum K values and stable or improved visual acuity.
Central Corneal Thickness Measurements and Devices The CCT measurements during CXL were performed with a handheld ultrasound (US) pachymeter (Handy Pachymeter, SP-3000, Tomey Corp.). The US CCT was measured with the tip of the pachymeter perpendicular to the cornea as centrally as possible and was performed at the following timepoints during the CXL procedure: (1) directly after epithelial abrasion, (2) directly after 30 minutes of riboflavin instillation, and (3) after 30 minutes of UVA irradiation. The mean of 10 consecutive measurements was used. If the CCT was less than 400 mm at timepoint 2, additional measurements were taken in intervals of 3 extra minutes for a maximum of 12 minutes, with both eyes closed. When the CCT did not increase after 12 minutes, hypoosmolar riboflavin was used and the patient was excluded from analyses. In cases in which the CCT remained less than 400 mm after hypoosmolar riboflavin instillation, the CXL procedure was terminated (ie, no UVA irradiation was applied) and the patient was excluded from the study analysis. Optical pachymetry and keratometry data were determined using a Scheimpflug device (Pentacam HR, Oculus Optikger€ ate GmbH). Corneal ECD was analyzed by specular microscopy (SP3000P, Topcon) preoperatively and at the 6-month follow-up visit. Corneal CXL was performed with the UV-X system (Peschke Meditrade GmbH) (370 nm and 3 mW/cm2).
Surgical Technique After the instillation of local anesthetic eyedrops comprising oxybuprocaine 0.4% and tetracaine 1.0%, an
J CATARACT REFRACT SURG - VOL 40, APRIL 2014
EYELID SPECULUM DURING CXL
eyelid speculum was inserted and a 9.0 mm corneal abrasion was performed using a blunt knife. The cornea was rinsed with a balanced salt solution, and US CCT measurements were performed. After the abrasion, the eyelid speculum remained in place in Group A, while in Group B, the eyelid speculum was removed and patients were instructed to keep their eyes closed. In both groups, isotonic riboflavin 0.1% solution (Medio Cross) was instilled at 3-minute intervals over a 30-minute period. Ultrasound CCT measurements were repeated after rinsing with a balanced salt solution. When the CCT was less than 400 mm at this timepoint, additional measurements were taken at intervals of 3 extra minutes for a maximum of 12 minutes with both eyes closed. When the CCT remained less than 400 mm after 12 minutes, hypoosmolar riboflavin was administered and the patient was excluded from analysis. In Group A, the eyelid speculum remained in place in preparation for the UVA phase. In Group B, the eyelid speculum was reinserted. In both groups, UVA irradiation was performed during 30 minutes, while isotonic riboflavin solution was reapplied to the cornea every 5 minutes. Finally, US CCT measurements were repeated and a bandage lens was placed. Oral pain medication was given for the first day in addition to preservative-free artificial tears (dextran–hypromellose, Duratears Free) for 4 weeks, nepafenac drops (Nevanac) for 1 week, and moxifloxacin hydrochloride eyedrops (Vigamox) for 4 weeks, all 3 times daily. When epithelial healing was complete, the bandage lens was removed and a fluorometholone eyedrop (FML) was applied twice daily for 2 weeks.
Statistical Analysis For pachymetry, a sample size of 24 eyes in each study group in the study indicated a power of 0.88 for a 2-sided test (a Z 0.05, based on literature s Z 37).13 For clinical outcomes (K changes) after CXL, a sample size of 14 eyes in each group was indicated based on the first 3 patients in each group (power of 0.8, a Z 0.05). The mean of 10 consecutive US CCT measurements was used for statistical analysis and compared at different timepoints. The difference in the mean US-measured CCT at different timepoints during CXL, visual acuity, refraction, keratometry, pachymetry, and endothelium were determined with the independent samples t test. Differences in parameters per group between baseline and 6 months after CXL were determined with a paired-samples t test. The Pearson chi-square test was used for the analysis of the extra waiting time difference between Group A and Group B. A P value of less than 0.05 was considered statistically significant. Data collection and analysis were performed using SPSS for Windows software (version 20.0, International Business Machines Corp., SPSS Statistics). The US-measured CCT after the corneal abrasion was considered to be the baseline CCT because CXL treatment was equal in both groups until this timepoint. The CCT differences were evaluated in microns (absolute) as well as in percentages (relative).
577
26 eyes of 26 patients. All were diagnosed with progressive keratoconus and complied with the CXL inclusion criteria. Four eyes, 2 in each group, were excluded from the study because they did not meet the required 400 mm CCT before UVA application during treatment. No complications occurred during the CXL procedure in any case. After exclusions, Group A, in which the eyelid speculum was used throughout the entire CXL procedure, consisted of 17 men and 7 women with a mean age of 23 years G 9 (SD) (range 13 to 49 years). Group B, in which no eyelid speculum was used during the 30-minute riboflavin instillation, consisted of 15 men and 9 women with a mean age of 26 G 7 years (range 18 to 42 years).
Ultrasound Central Corneal Thickness After Epithelial Removal (Baseline Central Corneal Thickness) No statistically significant difference in CCT after corneal abrasion was found between the 2 study groups. The mean CCT was 446 G 35 mm in Group A and 435 G 41 mm in Group B (P Z .329).
Ultrasound Central Corneal Thickness After Riboflavin Instillation A statistically significant difference in absolute and relative CCT reduction was found between Group A and Group B after the riboflavin instillation phase (P ! .001). The mean CCT decreased by 62 G 53 mm (13% G 11%) in Group A and by 11 G 35 mm (2% G 8%) in Group B (Figure 1).
RESULTS Baseline Characteristics This study evaluated 52 eyes of 50 patients who were divided into 2 groups; each group comprised
Figure 1. Central corneal thickness at different timepoints during CXL in Group A and Group B (CI Z confidence interval; UVA Z ultraviolet A).
J CATARACT REFRACT SURG - VOL 40, APRIL 2014
578
EYELID SPECULUM DURING CXL
Extra Hydration Time After Riboflavin Instillation After riboflavin instillation, 14 (58%) of 24 eyes in Group A (95% confidence interval [CI], 38%-78%) and 5 (21%) of 24 eyes in Group B (95% CI, 4%-37%) required extra hydration time in intervals of 3 minutes to ensure the required safety thickness of 400 mm before the start of UVA irradiation. The difference between the 2 groups was statistically significant (P Z .008, Pearson c2 test). A mean extra hydration time of 4 G 4 minutes was required in Group A and of 2 G 4 minutes in Group B. Statistical analysis of the difference in the mean extra hydration time between Group A and Group B showed an insignificant trend toward a longer interval period in Group A (P Z .067) (Figure 1). Ultrasound Central Corneal Thickness After Ultraviolet-A Irradiation The mean CCT decreased significantly in both groups during the UVA irradiation phase (both P ! .001). The mean decrease was 11% G 8% in Group A and 11% G 6% in Group B (Figure 1). No statistically significant difference in US-measured CCT was found between the 2 groups (P Z .839). Overall Ultrasound Central Corneal Thickness During Corneal Collagen Crosslinking Overall, the mean decrease in US-measured CCT from baseline thickness (ie, after epithelial removal) to the end of CXL treatment (ie, after UVA irradiation) in both groups was statistically significant (both P ! .001). The mean decrease was 15% G 10% in Group A and 12% G 7% in Group B. No statistically significant difference was found between the 2 groups (P Z .201) (Figure 2). However, the extra time needed to reach a minimum thickness of 400 mm after riboflavin instillation was taken into account, meaning that the total CXL time was not equal in all patients. Visual Acuity and Refraction Six months after CXL in both groups, the CDVA improved significantly (in Group A from 0.50 to 0.70 Snellen decimal; in Group B from 0.33 to 0.60 Snellen decimal). The UDVA improved significantly in Group A and remained stable in Group B (Table 1). The UDVA and CDVA improvement was not statistically significant different between the 2 groups. The refractive cylinder increased in both groups, in Group A by a mean of 1.0 G 2 diopters (D) (P Z .111) and in Group B by a mean of 0.6 G 1.7 D (P Z .026). The mean spherical equivalent (SE) values remained stable in both groups. There was no statistically significant difference in refractive cylinder or SE change between the 2 groups.
Figure 2. Overall CCT during CXL in Group A and Group B (CI Z confidence interval; UVA Z ultraviolet A).
Keratometry Six months after CXL, the flattest central K value decreased significantly in Group A; the mean decrease was 0.6 G 1.5 D in Group A and 0.4 G 2.1 D in Group B. The maximum K value showed a statistically significant decrease in Group B; the mean decrease was 0.66 G 1.7 D in Group A and 1.75 G 2.5 D in Group B. There was no statistically significant difference in keratometry change between the 2 groups (Table 1). Pachymetry and Endothelial Cell Count (Endothelium Cell Density) Six months after CXL, a statistically significant decrease in pachymetry at the central cornea, apex, and thinnest corneal location was found in both groups (Table 1). There was no statistically significant difference in the pachymetry decrease between the 2 groups. The ECD remained stable in both groups and between groups (Table 1). Epithelial Healing In 7 (29%) of 24 eyes in Group A and in 5 (21%) of 24 eyes in Group B, the epithelium had not closed at the 1-week checkup after CXL. This difference was not statistically significant (P Z .740, Fisher exact test). In all patients, the epithelium was closed within 2 weeks. Complications Six months after CXL, in Group A, 2 eyes had stromal haze and 1 eye had superficial haze. In Group B, 2 eyes had superficial haze. Six months after CXL, 4 eyes had a maximum increase in the K value of more than 1.0 D (3 in Group A; 1 in
J CATARACT REFRACT SURG - VOL 40, APRIL 2014
579
EYELID SPECULUM DURING CXL
Table 1. Effect of corneal CXL between Group A and Group B after 6 months compared with baseline.
Group/Exam Group A Preop 6 mo postop P value Group B Preop 6 mo postop P value Between groups P value
UDVA (LogMAR)
CDVA (LogMAR)
SE (D)
Kmax (D)
Ksteep (D)
Kflat (D)
Kmean (D)
CTcenter (mm)
0.90 0.65 .001*
0.27 0.18 .012*
1.5 1.3 .544
55.8 55.2 .076
49.4 48.9 .156
46.1 45.5 .038*
47.7 47.1 .078
1.00 0.85 .141
0.47 0.26 .036*
2.4 2.2 .964
59.5 57.8 .004*
52.0 51.6 .080
47.1 46.7 .421
49.4 49.0 .274
.502
.191
.731
.821
.632
.710
.092
CTapex (mm)
CTthin (mm)
ECD (Cells/mm2)
496 479 .002*
489 470 .001*
474 454 !.001*
2800 2870 .333
491 477 .001*
477 464 !.001*
458 444 .001*
2827 2746 .134
.276
.517
.641
.447
CDVA Z corrected distance visual acuity; CT Z corneal thickness (optical pachymetry by Scheimpflug device at the center [CTcenter], apex [CTapex], or thinnest [CTthin] point); ECD Z endothelial cell density; KflatZ flattest central keratometry; Kmax Z maximum keratometry; Kmean Z mean central keratometry; Ksteep Z steepest central keratometry; SE Z spherical equivalent; UDVA Z uncorrected distance visual acuity *Statistically significant
Group B). Two eyes lost more than 2 Snellen lines of CDVA (Group B); 3 eyes lost more than 2 Snellen lines of UDVA (Group B). DISCUSSION This study showed that the avoidance of an eyelid speculum during riboflavin instillation in a CXL procedure prevented corneal thinning before the UVA irradiation phase. This confirms the hypothesis of our recently published case report14 and is in agreement with the report of Schmidinger et al.18 In human corneas thinner than 400 mm, a toxic UVA level is reached at the endothelium during CXL. Therefore, minimizing corneal thinning before UVA irradiation was recommended by Wollensak et al.6 Endothelial damage has been described earlier in young patients with thin corneas below the 400 mm safety limit who were nevertheless treated by standard CXL.11 In the current study, the 400 mm safety limit at the start of UVA irradiation was maintained. However, during the UVA phase, the CCT decreased in both groups. This finding is in agreement with Schmidinger et al.,18 who also found a reduction in CCT, even 10 minutes after the start of UVA. No significant endothelial changes were observed during slitlamp examination or measured 6 months after CXL. In our study, 4 patients were excluded because they eventually had a hypoosmolar CXL treatment. This number is in line with clinical observations in which some eyes also react with disproportional corneal thinning after the application of isotonic riboflavin.9,19 The corresponding sample size in this study was not sufficient to form statistical conclusions about this number of exclusions.
In our study, when the use of an eyelid speculum was avoided during riboflavin instillation (Group B), a mean CCT decrease of merely 2% was measured directly after this timepoint, compared with 13% when the eyelid speculum was used (Group A) during riboflavin instillation. It appeared that in Group A, significantly more patients required extra time after riboflavin instillation to reach a CCT of 400 mm before irradiation, and this extra time leveled the mean CCT before UVA irradiation in both groups. During the intervals of extra time, both eyes remained closed, which contributed to a CCT increase due to rehydration of the cornea. As expected, we found no statistical overall CCT difference after UVA irradiation between the groups because the treatment procedure was equal in both groups during the final irradiation phase. Our results in Group A are a confirmation of those in a noncomparative study by Kymionis et al.13 in which there was a significant CCT decrease (20%) after riboflavin instillation with the use of an eyelid speculum. The explanation for the difference in CCT decrease compared with our study (20% versus 13%) can be sought in a baseline CCT after epithelium removal of 415 mm in their study group compared with 446 mm in our study; thinner corneas seem to dehydrate faster as a result of the stromal swelling pressure minus the hydrostatic compressive effect of the intraocular pressure. Swelling pressure was calculated using the relationships of corneal thickness to hydration.16 When the eyelids remain open, the cornea is known to dehydrate by tear evaporation and endothelial pump mechanisms, leading to thinner CCT values.15,16 In general, CCT values decrease significantly during corneal exposure for about 11 mm/min, or 10% after 5 minutes.20 Furthermore, the ingredient dextran in
J CATARACT REFRACT SURG - VOL 40, APRIL 2014
580
EYELID SPECULUM DURING CXL
riboflavin drops is osmotically active and prevents the cornea from swelling by extracting water from the eye.9 The influence of the eyelid speculum on changes in osmolarity of the riboflavin–dextran solution and the cornea could be an explanation for the difference in CCT results between Group A and Group B. On the other hand, epithelial removal can lead to corneal swelling.21 The total duration of CXL varied among patients due to the extra time intervals after riboflavin instillation in certain cases, which made it impossible to make an overall time-independent comparison of CCT changes between the 2 groups. It was not statistically proven that the CCT decrease during the overall CXL treatment was limited in patients treated without the eyelid speculum during the riboflavin phase (Group B). Another uncertainty of our modified protocol is that it is unknown whether the riboflavin concentration level changes when the eyelids remain closed. During UVA irradiation, the exposed cornea received riboflavin every 5 minutes, which might have neutralized the riboflavin concentration of the first phase. That there was no decrease in ECD in both groups 6 months after CXL might indicate that the concentration level of riboflavin was sufficient during the overall treatment. Longer follow-up is necessary to confirm the absence of differences in long-term ECD loss. The significant decrease in pachymetry that we found in both groups in our study 6 months after CXL is a common phenomenon after CXL.22 No statistically significant difference between the 2 groups was noted, which could mean that the significant CCT decrease in Group A during riboflavin instillation did not necessarily influence post-CXL pachymetry. In another study,23 our group recently reported that advanced keratoconus cases can show a more pronounced CXL-induced flattening effect. There was a difference in the average maximum K decrease between the 2 groups in the current study ( 0.60 D in Group A and 1.75 D in Group B); this could be because Group B contained more eyes with a steeper cornea at baseline (59.5 D in Group B versus 55.8 D in Group A). When comparing the overall 6-month post-CXL effect between the 2 groups, the significant CCT decrease during riboflavin instillation noted in Group A did not seem to significantly influence any clinical parameter after CXL. However, the extra waiting time to reach the required 400 mm could be a confounding factor for differences between the 2 groups, especially in the context of endothelial cell loss, as has been published.11 Concerning the study design and practical execution, all patients were treated at the same institution
by the same practitioners in a relatively short time interval (6 months), minimizing the risk for bias. In addition, because the baseline characteristics in Group A and Group B were similar, the 2 groups were statistically comparable. The absence of a significant difference in CCT change during the similar phases of the CXL treatment (ie, during corneal abrasion and UVA irradiation) was as expected and reconfirmed the comparability of the 2 groups. We used US rather than optical pachymetry measurements during CXL because the literature shows a major sensibility to inhomogeneous structures in crosslinked corneas, leading to underestimation of pachymetry of corneas measured with Scheimpflug imaging (eg, Pentacam).24 The results in the current study are clear and directly implementable in daily clinical use by simply eliminating the use of the eyelid speculum during the riboflavin phase. Another advantage of this technique, in addition to minimizing prolonged corneal dehydration and thus corneal thinning, is increased subjective patient comfort during the procedure. Patients in Group A wore the eyelid speculum during the entire procedure, meaning 65 to 70 minutes in total, and patients in Group B wore the eyelid speculum in time blocks of 5 to 10 minutes (during corneal abrasion) and then 30 minutes (during irradiation).25,26 However, this subjective patient advantage was not objectively evaluated. In conclusion, this study provides evidence that avoidance of the eyelid speculum during riboflavin instillation prevents unwanted perioperative corneal thinning, thereby increasing the chance of sufficient corneal thickness before UVA irradiation. This can reduce the need for hypoosmolar CXL treatments and thus avoid a prolonged and higher risk.
WHAT WAS KNOWN Central corneal thickness does not necessarily remain stable during CXL. Even though there is the possibility of hypoosmolar CXL procedures, unnecessary corneal thinning during a CXL procedure should be avoided. WHAT THIS PAPER ADDS Avoidance of the eyelid speculum during riboflavin instillation prevented unwanted perioperative corneal thinning before UVA irradiation. This outcome can reduce the need for hypoosmolar CXL treatments.
J CATARACT REFRACT SURG - VOL 40, APRIL 2014
581
EYELID SPECULUM DURING CXL
REFERENCES 1. Krachmer JH, Feder RS, Belin MW. Keratoconus and related noninflammatory corneal thinning disorders. Surv Ophthalmol 1984; 28:293–321 2. Rabinowitz YS. Keratoconus. Surv Ophthalmol 1998; 42:297– 319. Available at: http://www.keratoconus.com/resources/ MajorCReview-Keratoconus.pdf. Accessed January 4, 214 3. Spoerl E, Huhle M, Seiler T. Induction of cross-links in corneal tissue. Exp Eye Res 1998; 66:97–103 4. Wollensak G, Spoerl E, Seiler T. Riboflavin/ultraviolet-A– induced collagen crosslinking for the treatment of keratoconus. Am J Ophthalmol 2003; 135:620–627. Available at: http:// grmc.ca/assets/files/collagen_crosslinking_2003_wollensak.pdf. Accessed January 4, 2014 5. Wollensak G. Crosslinking treatment of progressive keratoconus: new hope. Curr Opin Ophthalmol 2006; 17:356–360 €rl E, Reber F, Pillunat L, Funk R. Corneal 6. Wollensak G, Spo endothelial cytotoxicity of riboflavin/UVA treatment in vitro. Ophthalmic Res 2003; 35:324–328 7. Wollensak G, Spoerl E, Wilsch M, Seiler T. Endothelial cell damage after riboflavin–ultraviolet-A treatment in the rabbit. J Cataract Refract Surg 2003; 29:1786–1790 8. Raiskup F, Kißner A, Spoerl E, Pillunat LE. Hornhautvernetzung € sung beim Keratokonus mit mit hypoosmolarer Riboflavin-Lo € nner Hornhaut [Corneal cross-linking with hypo-osmolar ribodu flavin solution for keratoconus with thin corneas]. Ophthalmologe 2011; 108:846–851 9. Hafezi F, Mrochen M, Iseli HP, Seiler T. Collagen crosslinking with ultraviolet-A and hypoosmolar riboflavin solution in thin corneas. J Cataract Refract Surg 2009; 35:621–624 10. Hafezi F. Limitation of collagen cross-linking with hypoosmolar riboflavin solution: failure in an extremely thin cornea. Cornea 2011; 30:917–919 11. Kymionis GD, Portaliou DM, Diakonis VF, Kounis GA, Panagopoulou SI, Grentzelos MA. Corneal collagen crosslinking with riboflavin and ultraviolet-A irradiation in patients with thin corneas. Am J Ophthalmol 2012; 153:24–28 12. Wollensak G, Aurich H, Wirbelauer C, Sel S. Significance of the riboflavin film in corneal collagen crosslinking. J Cataract Refract Surg 2010; 36:114–120 13. Kymionis GD, Kounis GA, Portaliou DM, Grentzelos MA, Karavitaki AE, Coskunseven E, Jankov MR, Pallikaris IG. Intraoperative pachymetric measurements during corneal collagen cross-linking with riboflavin and ultraviolet A irradiation. Ophthalmology 2009; 116:2336–2339 14. Tahzib NG, Soeters N, Van der Lelij A. Pachymetry during crosslinking [letter]. Ophthalmology 2010; 117:2041–2041.e1; reply by GD Kymionis, DM Portaliou 2041–2042 15. Bourassa S, Benjamin WJ, Boltz RL. Effect of humidity on the deswelling function of the human cornea. Curr Eye Res 1991; 10:493–500
16. O’Neal MR, Polse KA. In vivo assessment of mechanisms controlling corneal hydration. Invest Ophthalmol Vis Sci 1985; 26:849–856. Available at: http://www.iovs.org/content/26/6/ 849.full.pdf. Accessed January 4, 2014 17. Spoerl E, Mrochen M, Sliney D, Trokel S, Seiler T. Safety of UVA-riboflavin cross-linking of the cornea. Cornea 2007; 26:385–389 18. Schmidinger G, Pachala M, Prager F. Pachymetry changes during corneal crosslinking: effect of closed eyelids and hypotonic riboflavin solution. J Cataract Refract Surg 2013; 39:1179–1183 € Intraoperative corneal thickness 19. Kaya V, Utine CA, Yılmaz OF. measurements during corneal collagen cross-linking with hypoosmolar riboflavin solution in thin corneas. Cornea 2012; 31:486–490 20. Aurich H, Wirbelauer C, Jaroszewski J, Hartmann C, Pham DT. Continuous measurement of corneal dehydration with online optical coherence pachymetry. Cornea 2006; 25:182–184 21. Maurice DM, Giardini AA. Swelling of the cornea in vivo after the destruction of its limiting layers. Br J Ophthalmol 1951; 35:791– 797. Available at: http://www.pubmedcentral.nih.gov/picrender. fcgi?artidZ1323849&blobtypeZpdf. Accessed January 4, 2014 22. Greenstein SA, Shah VP, Fry KL, Hersh PS. Corneal thickness changes after corneal collagen crosslinking for keratoconus and corneal ectasia: one-year results. J Cataract Refract Surg 2011; 37:691–700 23. Sloot F, Soeters N, van der Valk R, Tahzib NG. Effective corneal collagen crosslinking in advanced cases of progressive keratoconus. J Cataract Refract Surg 2013; 39:1141–1145 24. Mencucci R, Paladini I, Virgili G, Giacomelli G, Menchini U. Corneal thickness measurements using time-domain anterior segment OCT, ultrasound, and Scheimpflug tomographer pachymetry before and after corneal cross-linking for keratoconus. J Refract Surg 2012; 28:562–566 25. Cheng ACK, Young AL, Law RWK, Lam DSC. Prospective randomized double-masked trial to evaluate perioperative pain profile in different stages of simultaneous bilateral LASIK. Cornea 2006; 25:919–922 26. Davis MJ, Pollack JS, Shott S. Comparison of topical anesthetics for intravitreal injections; a randomized clinical trial. Retina 2012; 32:701–705
J CATARACT REFRACT SURG - VOL 40, APRIL 2014
First author: Nienke Soeters, BOptom Utrecht Cornea Research Group, Department of Ophthalmology, University Medical Center Utrecht, Utrecht, the Netherlands
![[Corneal collagen crosslinking in keratoconus].](/assets/img/hold.png)
