Purpose
To determine long-term efficacy and safety of riboflavin/ultraviolet A corneal cross-linking (CXL).
Design
Prospective cohort study.
Methods
Thirty-six patients (36 eyes) who underwent epithelium-off CXL at a University Hospital (Guy’s and St Thomas’ National Health Service Foundation Trust) 6–8 years previously were examined. The main outcome measures were refractive error, visual acuity, corneal topographic keratometry, ultrasonic pachymetry, and topography-derived corneal wavefront.
Results
At 7 years compared to preoperative values, mean spherical equivalent refractive error (SEQ) increased by +0.78 diopter (D) ( P < .005) and mean simulated topographic keratometry (SimK) and mean maximum keratometry (Kmax) reduced by −0.74 D ( P < .0001) and −0.91 D ( P < .0001), respectively. Uncorrected distance acuity (UCDA) ( P < .0005) and corrected distance acuity (CDVA) ( P < .0001) had improved and root mean square (RMS) ( P < .0005), coma ( P < .0005), and secondary astigmatism ( P < .005) lessened. At 7 years compared to 1 year, CDVA improved ( P < .05); mean SimK ( P < .0005) and mean Kmax ( P < .005) reduced by −0.45 D and −0.56 D, respectively; and RMS ( P < .0005) and coma ( P < .0005) decreased. At 7 years compared to 5 years, CDVA improved ( P < .05) and trefoil reduced ( P < .05). No treated eyes progressed. In 29 initially untreated fellow eyes mean SimK increased by +0.54 D ( P < .02), mean Kmax by +0.87 D ( P < .05), and refractive astigmatism increased ( P < .0005).
Conclusions
Following corneal cross-linking, improvements in topographic and wavefront parameters evident at 1 year were seen to continue to improve at 5 years and were maintained at 7 years. No treated eyes progressed over the 7-year follow-up period.
Corneal collagen cross-linking (CXL) is the first procedure that may halt the progression of keratoconus and other corneal ectasias. Laboratory studies have exhibited increases in stromal stress-strain measurements, resistance to enzymatic digestion, and thermal damage. Prospective clinical studies have demonstrated stabilization of keratoconus with no alteration in transparency, an improvement in visual performance, and reduction in keratometry. In a randomized, prospective, bilateral study with 18 months follow-up we reported CXL in treated eyes to be effective in halting progression and improving visual, topographic, and wavefront indices, while in untreated eyes 14% progressed. Similarly, Wittig-Silva in 100 eyes with 3 years follow-up demonstrated improvements in topographic and visual parameters in treated eyes with worsening of these parameters in untreated eyes.
While such studies support CXL, there is a paucity of long-term data. The need to repeat CXL, its long-term efficacy, and its safety are uncertain. There are few publications with 5-year follow-up. Hashemi and associates, in 32 patients, found stability of keratoconus with improvement in corrected distance visual acuity (CDVA) and continued reduction in apex elevation between 1 and 5 years. We reported 30 patients at 5 years and found no progression with continued improvements in topographic and wavefront parameters. Only Theuring and associates and Raiskup and associates have published longer-term data at 10 years, reporting sustained improvement of CDVA and keratometry, with 2 patients (10%) progressing and requiring retreatment. The aim of this study was to present further results on the long-term efficacy and safety of CXL beyond 5 years. We report 36 eyes of 36 patients, who underwent epithelium-off CXL with a mean follow-up of 7 years.
Methods
Subjects
Following ethics committee approval (East of Scotland Ethics Committee, reference number 14/ES/0031), subjects who had CXL at our institution 6–8 years previously were invited to return and consented to an ophthalmic follow-up examination. Thirty-six patients from a cohort of 42 were examined (3 were lost to follow-up, 1 had died from an unrelated neoplastic illness, and 2 underwent intrastromal ring segment insertion at 24–36 months after CXL and were excluded from analysis). Mean age at time of CXL was 27.4 years (range 12–46 years, median 25.5 years). Mean follow-up was 80 months (range 69–97 months, median 77.5 months). Thirty-one patients were male and 5 were female.
One eye of each patient was selected for analysis. In 18 patients, only 1 eye was treated, but in 11 patients the contralateral eye underwent CXL 18–95 months after treatment of the first. In these individuals the first eye treated was selected for analysis. The initially untreated 29 eyes had a mean follow-up of 68 months (18–95 months) and acted as a control. In 3 cases the fellow eye had undergone deep anterior lamellar keratoplasty and in 2 intrastromal ring segment insertion; these were excluded from analysis. In 2 patients bilateral simultaneous CXL had been performed. In these individuals, 1 eye was selected randomly for analysis, using an independent observer with a shuffled envelope system.
Prior to CXL treatment keratoconus/ectasia was confirmed in all eyes by an experienced corneal surgeon (D.O’B.). All had abnormal corneal topography with inferior/paracentral inferior steepening, bow-tie asymmetry of >1.5 diopters (D), and skewed astigmatic axes. They had keratoconic features on the Cone Location and Magnitude Index (CLMI) software on the Keraton Scout Corneal Analyzer (Optikon 2000, Rome, Italy), which locates the cone center by finding the steepest curvature with a 2.0 mm circle within the central 8.0 mm zone, then subtracting the area corrected average of all points outside the circle from the area corrected average of the points within the 2.00 mm circle. The area corrected average of the points within the 2.00 mm circle is compared to a 2.00 mm circle 180 degrees away and the results are used to determine if the steep area represents a cone. In addition, Orbscan II scans (Bausch and Lomb, Bridgewater, New Jersey, USA) showed changes consistent with keratoconus, with localized areas of elevation posteriorly and anteriorly at the same location as the anterior curvature steepening. Thirty-four eyes had early to moderate keratoconus with mean simulated keratometry (SimK) readings less than 56 D, absence of corneal scarring, and minimum corneal thickness of greater than 400 μm. Only 1 eye had advanced keratoconus with a preoperative Kmax value greater than 58 D. All eyes had evidence of progression with reduced uncorrected distance (UDVA) or CDVA by >1 line and/or worsening of refractive or corneal astigmatism, SimK, or maximum keratometry (Kmax) by 0.75 D over the 12–24 months prior to CXL. One patient had progressive, post–laser in situ keratomileusis (LASIK) iatrogenic ectasia.
Patients were asked to refrain from rigid lens wear for 3 weeks and soft lens wear for 1 week prior to all examinations. Objective and subjective refraction, Snellen decimal equivalent UDVA and CDVA, slit-lamp biomicroscopy, tonometry, ultrasonic central corneal pachymetry (Pachmate, DGH55; DGH Technology Inc, Pennsylvania, USA) and mydriatic funduscopy were performed preoperatively. Corneal topography was performed with scanning-slit corneal topography (Orbscan II) and/or Placido-disc videokeratography (Keraton Scout Corneal Analyzer). For Orbscan examinations, 2 scans for each eye were taken and the average of the keratometry values calculated for analysis. For Keraton Scout examinations, 4 scans for each eye were taken and the highest-quality scan closest to the average keratometry values was selected for wavefront analysis. Endothelial counts were not undertaken in this cohort of patients.
Surgical Procedure
We have described the procedure, based on that of Wollensak, previously. Following informed consent, tetracaine 1% and chloramphenicol 0.5% were instilled. A 9.00-mm area of central epithelium was removed using an epithelial spatula (Malosa Medical, Elland, UK). Five drops of riboflavin 0.1% in dextran 20% were instilled and reapplied 5 minutes later. A period of 10 minutes elapsed before ultraviolet A (UVA) exposure. UVA exposure was for 30 minutes with 370 nm UVA radiation at 3 mW/cm 2 with a beam diameter of 8.00 mm. During UVA exposure, riboflavin 0.1% drops were administered every 3 minutes and tetracaine 1% drops if the patient reported discomfort. Care was taken not to irradiate the limbus, by monitoring and centering the UVA beam over the axial cornea. Intraoperative pachymetry was not performed, nor was intraoperative slit-lamp examination, as this was not part of the original treatment protocol.
Postoperative Treatment and Assessment
Following CXL, ofloxacin 0.3% and chloramphenicol 1% were administered and the eye patched. Oral analgesics, ibuprofen 400 mg pro re nata (PRN) 3 times a day and codeine phosphate 30–60 mg PRN 4 times a day, were prescribed. Four vials of benoxinate 0.4% were given, with instructions to be administered if the postoperative pain was severe, with a maximum dosage of 1 drop only every 2 hours for a maximum of 48 hours. Ofloxacin 0.3% eye drops were administered 4 times a day for 1 week and chloramphenicol 1% ointment at night for 2 weeks. Postoperative examinations were conducted at 1 week; 1, 3, 6, and 12 months; and 5 and 7 years and included Snellen UDVA, CDVA, refraction, ultrasonic central corneal pachymetry, slit-lamp biomicroscopy, tonometry, and mydriatic funduscopy. All 36 patients attended for the 1 week, 3 and 6 month, and 7 year follow-up visits, while 34 attended at 1 and 5 years. Placido-disc topography and scanning-slit tomography were undertaken for the examinations at 1, 3, 6, and 12 months. At 5 and 7 years, Placido-disc corneal topography was performed using the Keraton Scout system, while tomography was undertaken by the Pemtacam Scheimpflug camera system (Oculus Optikgeräte GmbH, Wetzlar, Germany). To investigate astigmatic change in the refraction, vector analysis was performed according to the system described by Retzlaff and associates.
Prior to statistical analysis all measured variables were investigated to determine that they were normally distributed and did not have significant skew, with all having values less than 1.21 around the mean. Paired Student t tests were used to compare pre- and postoperative outcomes within the treated and untreated groups. Visual acuity results were converted into decimal before averaging and statistical calculation. Results of P < .05 were considered significant.
Postoperative Treatment and Assessment
Following CXL, ofloxacin 0.3% and chloramphenicol 1% were administered and the eye patched. Oral analgesics, ibuprofen 400 mg pro re nata (PRN) 3 times a day and codeine phosphate 30–60 mg PRN 4 times a day, were prescribed. Four vials of benoxinate 0.4% were given, with instructions to be administered if the postoperative pain was severe, with a maximum dosage of 1 drop only every 2 hours for a maximum of 48 hours. Ofloxacin 0.3% eye drops were administered 4 times a day for 1 week and chloramphenicol 1% ointment at night for 2 weeks. Postoperative examinations were conducted at 1 week; 1, 3, 6, and 12 months; and 5 and 7 years and included Snellen UDVA, CDVA, refraction, ultrasonic central corneal pachymetry, slit-lamp biomicroscopy, tonometry, and mydriatic funduscopy. All 36 patients attended for the 1 week, 3 and 6 month, and 7 year follow-up visits, while 34 attended at 1 and 5 years. Placido-disc topography and scanning-slit tomography were undertaken for the examinations at 1, 3, 6, and 12 months. At 5 and 7 years, Placido-disc corneal topography was performed using the Keraton Scout system, while tomography was undertaken by the Pemtacam Scheimpflug camera system (Oculus Optikgeräte GmbH, Wetzlar, Germany). To investigate astigmatic change in the refraction, vector analysis was performed according to the system described by Retzlaff and associates.
Prior to statistical analysis all measured variables were investigated to determine that they were normally distributed and did not have significant skew, with all having values less than 1.21 around the mean. Paired Student t tests were used to compare pre- and postoperative outcomes within the treated and untreated groups. Visual acuity results were converted into decimal before averaging and statistical calculation. Results of P < .05 were considered significant.
Results
Spherical Equivalent Refractive Error
Preoperatively, the mean spherical equivalent refractive error (SEQ) in the 36 CXL study eyes was −1.39 D (range +1.00 to −5.25 D). At 1 year, it reduced to −0.92 D (range +2.50 to −4.625 D) ( P < .005). At 5 years, it reduced to −0.67 D (range +2.25 to −3.875 D) ( P < .0005). At 7 years it was −0.61 D (range +2.00 to −4.75 D), representing a mean +0.78 D hyperopic shift (range +4.00 to −1.00 D) compared to preoperative measurements ( P < .005). There was no difference in SEQ between values at 1, 5, and 7 years ( Table 1 , Figure 1 ). In 8 eyes (22%) hyperopic shift was +2.00 D or greater and in 4 (11%) it was +3.00 D or more. In 29 initially untreated fellow eyes, at first examination the mean SEQ was −1.66 D. At 18–95 months (mean 68 months) it was −1.72 D ( P = .8) ( Table 2 ).
| Parameter (Mean Values) | Preoperative N = 36 | 1 Year N = 34 | 5 Years N = 34 | 7 Years N = 36 | P Value a | P Value b | P Value c |
|---|---|---|---|---|---|---|---|
| SEQ | −1.39 ± 1.9 D | −0.92 ± 1.74 D | −0.67 ± 1.67 D | −0.6 ± 1.81 D | <.005 | .5 | .8 |
| Refractive cyl | −3.33 ± 2.51 D | −3.37 ± 2.55 D | −3.16 ± 2.34 D | −3.13 ± 2.3 D | .4 | .08 | .7 |
| UDVA (SDE) | 0.32 ± 0.26 | 0.41 ± 0.33 | 0.39 ± 0.30 | 0.46 ± 0.5 | <.0005 | .5 | .09 |
| CDVA (SDE) | 0.85 ± 0.25 | 0.92 ± 0.24 | 0.92 ± 0.2 | 0.96 ± 0.17 | <.0001 | <.05 | <.03 |
| Pachymetry | 488 ± 37 μm | 483 ± 35 μm | 492 ± 39 μm | 484 ± 40 μm | .4 | <.04 | .06 |
| Mean SimK | 46.3 ± 3.49 D | 45.98 ± 3.23 D | 45.59 ± 3.27 D | 45.57 ± 3.32 D | <.0001 | <.0005 | .6 |
| Mean Kmax | 48.23 ± 3.49 D | 47.79 ± 3.97 D | 47.37 ± 3.77 D | 47.32 ± 3.85 D | <.0001 | <.005 | .9 |
| Topographic astigmatism | 3.75 ± 2.52 D | 3.68 ± 2.36 D | 3.51 ± 1.92 D | 3.38 ± 2.06 D | .1 | .9 | .07 |
a P values at 7 years compared to preoperative values.
b P values at 7 years compared to 1-year values.
| Parameter | First Examination | Final Follow-up (18–95 Months, Mean 68 Months) | P Value a |
|---|---|---|---|
| SEQ | −1.66 ± 2.51 D | −1.72 ± 2.27 D | .8 |
| Refractive cyl | −2.57 ± 2.51 D | −3.45 ± 3.22 D | <.0005 |
| UDVA (SDE) | 0.56 ± 0.4 | 0.43 ± 0.37 | .4 |
| CDVA (SDE) | 0.91 ± 0.28 | 0.92 ± 0.29 | .9 |
| Pachymetry | 493 ± 35 μm | 495 ± 36 μm | .06 |
| Mean SimK | 45.56 ± 2.99 D | 46.06 ± 3.28 D | <.02 |
| Mean Kmax | 47.01 ± 3.54 D | 47.87 ± 4.43 D | <.05 |
| Topographic astigmatism | 2.85 ± 1.6 D | 3.58 ± 3.06 D | .2 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
