Current therapies for keratoconus include the use of rigid contact lenses and keratoplasty for patients with advanced disease or in whom contact lenses are intolerable.1 However, new treatments such as corneal cross-linking (CXL) have revolutionized the treatment for keratoconus. The combination of riboflavin with ultraviolet light irradiation has been shown to slow or even halt the progression of the disease by increasing corneal strength.2 CXL may result in improved uncorrected (UDVA) and corrected (CDVA) distance visual acuity secondary to flattening and surface regularization. Because of the stiffening effect CXL treatment has on the cornea, laser correction can then be performed on patients with keratoconus, which improves visual acuity.3
There remains a debate about when CXL and photorefractive keratectomy (PRK) should be conducted (ie, during one procedure [simultaneously] or with a period of rest between the two procedures [sequentially]). Kanellopoulos4 conducted a study in which these two methods were compared, suggesting that the simultaneous treatment modality was superior for three reasons. First, a simultaneous procedure reduces patients' time away from work. Second, performing both procedures at the same time appeared to reduce the potential superficial scarring that can occur when using PRK. Finally, when topography-guided PRK was conducted sequentially after CXL it was observed that some or part of the anterior cornea was removed, which therefore reduced the beneficial effect of the CXL.4
However, there are some drawbacks associated with the simultaneous approach, mainly its invasiveness (ie, having two procedures done on the same day). Furthermore, only topography-guided PRK was used in the study conducted by Kanellopoulos.4 This was because the wavefront measurements were difficult in irregular corneas that were included in that study.4
Because there is currently no clarity on this issue, we conducted this review to compare and evaluate the outcomes of the published literature on simultaneous and sequential CXL and excimer laser surface ablation to provide an evidence-based approach, which may result in better and safer outcomes for patients.
Patients and Methods
We conducted a systematic review of studies relevant to the use of surface ablation combined with CXL for the visual rehabilitation of keratoconus cases. We aimed at including randomized controlled trials comparing sequential and simultaneous excimer laser and CXL. In the absence of any randomized controlled trials with direct comparison between the techniques, we decided to include case series in which a minimum of 20 eyes were treated and at least 12 months of follow-up. We accepted peer-reviewed articles of human studies only and included articles in any language. Topography-guided, wavefront-guided, and standard non-topography–guided, non-wavefront–guided excimer laser treatments were included. Studies with epithelium on and off and conventional and rapid CXL were included. Articles published online ahead of print were also included. We planned to classify the results based on these included parameters if we had enough data for comparison. We excluded studies investigating non-keratoconus corneal ectatic pathologies such as pellucid marginal degeneration and post-refractive surgery ectasia. We also excluded studies in which other surgical procedures such as intracorneal segment insertion were a part.
We performed a MEDLINE search for articles published until May 28, 2017 without stipulating any conditions on language of publication. Words used for the literature search were “crosslinking” + “excimer laser,” “crosslinking” + “PRK,” “simultaneous” + “crosslinking” + “PRK,” and “sequential” + “crosslinking” + “PRK.”
We assessed the titles and the abstracts resulting from the searches. We considered full-text copies of all possibly relevant studies to see whether they met the inclusion criteria. We extracted the data using a form developed by us on an Excel 2010 spreadsheet (Microsoft Corporation, Redmond, WA). One author (ASB) entered the data on the spreadsheets. Any disagreements for inclusion or exclusion of the studies were resolved by discussion among the authors.
The data were further divided into three broad groups: CXL followed by sequential excimer laser surface ablation (sequential group); simultaneous excimer laser surface ablation and CXL (simultaneous group); and excimer laser surface ablation alone with no CXL (no CXL group).
Data on preoperative and postoperative logMAR UDVA and spectacle (CDVA), manifest refraction, maximum keratometry (Kmax), thinnest local pachymetry, complications, progression (as per the author's definition in individual studies), need for glasses or contact lenses, and any additional procedures within 1 year of the primary procedure were studied.
We predicted studies to have varying follow-up in each arm and so decided to present the data at their latest follow-up visit and the data on change in logMAR UDVA, spectacle CDVA, change in spherical equivalent (SE) and refractive astigmatism, and change in Kmax. We calculated the pooled mean in each group as a comparative tool.
Statistical analysis was performed with SPSS statistical software (version 17.0; IBM Corporation, Armonk, NY). All visual acuity data were converted to logMAR units if presented in Snellen or decimal formats. A P value of less than .05 was considered statistically significant.
A total of 21 studies fulfilled the inclusion criteria. Figure A (available in the online version of this article) shows the recruitment of studies in this review. There were 3, 11, and 7 studies in the sequential, simultaneous, and no CXL groups, respectively. One of the comparative studies4 had two groups: sequential and simultaneous CXL plus. The study designs of all included studies are described in Table A (available in the online version of this article) All were published in English. The analysis includes a total of 1,410 eyes: 200 eyes in the sequential group, 853 eyes in the simultaneous group, and 357 eyes in the no CXL group. In the simultaneous group, three studies11,12,14 compared simultaneous CXL plus and CXL alone; in total, they compared 73 eyes (CXL plus) to 82 eyes (CXL alone).
Study flow diagram.
UDVA and CDVA Change
Table 1 shows the range of studies categorized in all groups and a comparison of the preoperative and the final postoperative UDVA and CDVA using the logMAR mean and standard deviation. All of the published studies reported improvement in the UDVA and CDVA. In the sequential group, the mean UDVA improved from 0.96 ± 0.035 to 0.27 ± 0.13 logMAR and the mean CDVA improved from 0.34 ± 0.06 to 0.13 ± 0.07 logMAR. In the simultaneous group, the mean UDVA improved from 0.66 ± 0.46 to 0.13 ± 0.06 logMAR and the mean CDVA improved from 0.06 ± 0.08 to 0.04 ± 0.02 logMAR. In the no CXL group, the mean UDVA improved from 1.01 ± 0.069 to 0.30 ± 0.33 logMAR and the mean CDVA improved from 0.20 ± 0.24 to 0.11 ± 0.10 logMAR.
UDVA and CDVA in Included Studies
Figure 1 summarizes changes in UDVA and CDVA, respectively, in all groups of patients. Figure 1A shows comparable change in improvement in UDVA in all three groups; the simultaneous group had a wider range and standard deviation of the preoperative UDVA due to its larger number of studies. In Figure 1B, the sequential group showed greater improvement in CDVA. However, although the mean preoperative CDVA was less in the no CXL group than in the sequential group, there was a large standard deviation suggestive of a wide range of keratoconus from poor to good visual acuity preoperatively.
Mean logMAR (A) uncorrected (UDVA) and (B) corrected (CDVA) distance visual acuity preoperatively and postoperatively. SD = standard deviation; CXL = corneal cross-linking
SE and Refractive Astigmatism
All studies showed reduction of the spherical and cylindrical refractive power after excimer laser, which improved UDVA. In the sequential group, the mean SE decreased from 3.72 ± 0.47 D preoperatively to 1.58 ± 0.85 D at the last follow-up. In the simultaneous group, SE decreased from 2.06 ± 0.53 to 0.33 ± 0.12 D. In the no CXL group, SE decreased from 4.78 ± 1.22 to 1.04 ± 0.91 D. Refractive astigmatism reduced from 3.13 ± 0.32 to 1.74 ± 0.64 D in the sequential group, 1.85 ± 0.54 to 0.85 ± 0.24 D in the simultaneous group, and 3.15 ± 0.90 to 1.60 ± 1.21 D in the no CXL group.
Figure 2 shows a significant reduction of SE after the different interventions in the three groups. SE change was greatest in the no CXL group and the reduction in cylinder was comparable between the sequential group and no CXL groups and less in the simultaneous group.
Mean (A) spherical equivalent (SE), (B) change in SE, and (C) mean refractive astigmatism (RA) preoperatively and postoperatively. SD = standard deviation; CXL = corneal cross-linking
Reduction in the Kmax is a common finding due to the flattening effect of myopic laser treatment in those eyes. Analysis showed a mean Kmax reduction from 48.3 ± 4.25 to 45.8 ± 3.34 D in the sequential group, 47.65 ± 0.58 to 45.73 ± 1.16 D in the simultaneous group, and 48.59 ± 1.73 to 46.03 ± 2.49 D in the no CXL group (Figure 3).
Mean maximum keratometry (Kmax) preoperatively and postoperatively. SD = standard deviation
Keratoconus Progression in Each Group
None of the studies in the sequential group reported progression of keratoconus (longest follow-up: 3 years). In the simultaneous group, one study compared CXL alone with CXL plus and demonstrated progression in 6.7% eyes at 6 months after CXL plus. Progression was defined as an increase in the cone apex keratometry of 0.75 D or myopic shift of 0.75 D in the SE cycloplegic refraction in this study.12 In the no CXL group, one study documented progression in 1.7% over 5 years where only excimer laser was performed. Progression was defined by an increase in the cone apex keratometry of 0.75 D or a change of 0.75 D in the SE refraction in the past 6 months in this study.21 Progression in this group can be explained by the fact that CXL was not performed.
No serious complications were reported by any of the authors. Almost all studies reported some degree of haze in a certain percentage of the cases, as illustrated in Table 2. In the sequential group, all three studies4–6 reported a certain degree of haze. In the simultaneous group, 6 of 11 studies1,4,9,10,12,13 mentioned some degree of haze as a postoperative finding, one study8 mentioned no haze at all, and 4 studies2,7,11,14 did not comment on postoperative haze. In the no CXL group, 4 of 7 studies15,16,18,21 mentioned haze as a complication.
Complications Reported in Included Studies
Safety and Efficacy Indices
The efficacy index is defined as the ratio of the postoperative UDVA to the preoperative CDVA, whereas the safety index is defined as the ratio of the postoperative CDVA to the preoperative CDVA. One study6 in the sequential group measured safety and efficacy indices, which were 1.96 and 1.58, respectively. In the simultaneous group, 3 studies2,8,13 measured safety and efficacy as 1.41 ± 0.32 and 0.91 ± 0.41, respectively. One study16 in the no CXL group reported safety and efficacy indices of 1 and 0.82, respectively.
Our review suggests that sequential protocol studies showed greater improvements in CDVA compared to simultaneous protocol studies. Although the SE and refractive astigmatism reduced in all groups, overall, the simultaneous procedures did not produce changes as great as in the sequential study reports. There was no difference in changes in Kmax between simultaneous and sequential protocol study reports. Haze was reported in all protocols. Of the available data, safety and efficacy indices were lowest in studies where excimer laser was performed with no CXL and highest in studies where CXL was performed before excimer laser ablation.
Although the SE changed in all three groups, it was greatest in the no CXL group and the least in the simultaneous group. This may be because the CXL procedure would induce some change in the refraction that would be better corrected with excimer laser performed as a separate procedure at a later date (sequential) than simultaneously. However, we found that the change in refractive cylinder was similar in the sequential and no CXL groups. Progression was only noted in 1 of 11 studies in the simultaneous group. Although no progression was noted in the sequential group and only 1 of 7 studies21 showed progression in the no CXL group, we cannot comment on the progression parameters with any confidence due to variations in sample size between the three groups. From our study, it also appears that the complication of haze may be more frequent in sequential than in simultaneous protocol studies (Table 2). This may be due to reactivation of keratocytes leading to inflammatory response more than once in sequential protocols compared to simultaneous protocols.
In 2007, Kanellopoulos and Binder22 treated one eye of a patient using the sequential approach. They were impressed by the outcome (UDVA was 20/20 and CDVA was 20/15, with a refractive error of plano −0.50 × 150) and validated this approach by comparing with the fellow eye of the same patient that received only CXL with no sequential PRK. Kanellopoulos7 published the simultaneous approach (the Athens protocol) and showed increase in UDVA by +0.38 ± 0.31 logMAR and CDVA by +0.20 ± 0.21 logMAR. He stated that 95% of the cases had at least +0.1 increase in logMAR UDVA and 95% had positive change in CDVA. Kymionis et al.23 studied the simultaneous topography-guided PRK + CXL and found improvement in the preoperative mean logMAR UDVA from 0.99 ± 0.81 to 0.16 ± 0.15 and CDVA from 0.21 ± 0.19 to 0.11 ± 0.15. Sakla et al.8 also studied the simultaneous topography-guided approach in 31 eyes. At 12 months, the mean logMAR UDVA improved to 0.23 ± 0.33 from 0.79 ± 0.36 (P < .001) and the logMAR CDVA improved to 0.06 ± 0.07 from 0.28 ± 0.20 (P < .001). Alessio et al.10 found superior results in the group that received topography-guided PRK together with CXL than with CXL alone. UDVA improved significantly from 0.63 ± 0.36 to 0.19 ± 0.17 logMAR (P < .05) and CDVA from 0.06 ± 0.08 to 0.03 ± 0.06 logMAR (P < .05). We found that the mean CDVA improved more in the sequential group compared to the no CXL group. This may be because the eyes in the sequential group may have more advanced keratoconus, needing CXL first compared to mild and moderate keratoconus in the no CXL group with a wider range of keratoconic eyes involved with larger standard deviation.
In the study by Kymionis et al.,23 mean steepest keratometry was reduced from 48.20 ± 3.40 to 45.13 ± 1.80 D at the last follow-up visit. SE was reduced by a mean of 1.74 D at the last follow-up visit. Sakla et al.8 found flat and steep keratometry showed significant flattening at the latest follow-up. The mean refractive astigmatism decreased from −2.77 ± 1.47 to −0.98 ± 0.76 D (P < .001). Alessio et al.10 found manifest refraction SE and spherical and cylindrical power improved significantly (P < .05). Mean reduction in SE refraction was 1.75 D.
Tuwairqi and Sinjab2 studied safety and efficacy of simultaneous CXL with topography-guided PRK in managing low-grade keratoconus. They noted statistically significant improvement in all study parameters (P < .01). The safety and efficacy indices were 1.6 and 0.4, respectively. A patient satisfaction questionnaire showed that 91% were satisfied. Our review found the least safety and efficacy with protocols where excimer laser ablation was performed alone with no CXL.
All studies that used CXL either simultaneously or sequentially reported using the standard epitheliumoff protocol of 3 mW/cm2 for 30 minutes, except one study in the simultaneous group that reported using the accelerated protocol of 10 mW for 10 minutes.7
In the sequential group, 2 studies4,5 used 50 μm as the maximum allowed ablation depth and the third study used 15% of the thickness at the thinnest point as the maximum allowed ablation depth.6 In the simultaneous group, 8 of 11 studies used the 50 μm rule as the maximum regardless the corneal thickness. In the no CXL group, the ablation varied between the different studies: 2 studies left a residual bed thickness of greater than 300 μm15,17 and another study left a residual bed of 450 μm in all patients.21
There is another debate surrounding the use of topography- versus wavefront-guided laser correction in patients with keratoconus. A wavefront-guided approach is based on the assumption that all anomalies of the eye can be corrected by reshaping the cornea; it is centered on the pupil, takes into account all imperfections of the optical system, and puts them into a treatment profile.24,25 Topography-guided ablations are centered on the corneal vertex and focus on treating problems caused by corneal irregularity.25
Topography-guided treatment has some advantages. First, it can be used in highly irregular corneas that would be beyond the limits of the wavelength measuring device.4 Second, topography maps are relatively easy to interpret and are more intuitive compared to theoretical wavefront maps.25 However, a disadvantage of topography-guided treatment is that it may not produce the desired spherical refractive power outcome. This is because the treatment focuses on normalizing a highly irregular cornea and does not take into account the spherical power of the revised cornea.25 In this review, only 3 studies5,6,16 used wavefront-guided ablation; all other researchers used topography-guided protocols.
A limitation of this study is that the quality of the data was limited to that of the quality of the data in the included studies. There are many variables in different studies, which may lead to biases. The sample sizes of the included studies in each group are distinctly different and hence formal statistical comparison is not possible. The large variation in number between studies using wavefront-guided laser treatment (only 3)5,6,16 and those using topography-guided ablation profiles (17 studies) made it unrealistic to statistically compare outcomes of the two treatment plans. This study points to the need for standardizing the reporting parameters in studies involving keratoconus. Nonetheless, this review helps readers to analyze the existing simultaneous versus sequential protocols in more detail with their outcomes, which will enable them to formulate their own protocols and perhaps set up comparative studies to gain more insight into the combination protocols.
- Mukherjee A, Selimis V, Aslanides I. Transepithelial photorefractive keratectomy with crosslinking for keratoconus. Open Ophthalmol J. 2013;7:63–68. doi:10.2174/1874364101307010063 [CrossRef]
- Tuwairqi W, Sinjab M. Safety and efficacy of simultaneous corneal collagen cross-linking with topography-guided PRK in managing low-grade keratoconus: 1-year follow up. J Refract Surg. 2012;228:341–345. doi:10.3928/1081597X-20120316-01 [CrossRef]
- Kymionis GD, Grentzelos MA, Portaliou DM, Kankariya VP, Randleman JB. Corneal collagen cross-linking (CXL) combined with refractive procedures for the treatment of corneal ectatic disorders: CXL plus. J Refract Surg. 2014;30:566–576. doi:10.3928/1081597X-20140711-10 [CrossRef]
- Kanellopoulos AJ. Comparison of sequential vs same-day simultaneous collagen cross-linking and topography-guided PRK for treatment of keratoconus. J Refract Surg. 2009;25:S812–S818. doi:10.3928/1081597X-20090813-10 [CrossRef]
- Camellin M, Guidotti JM, Arba Mosquera S. Corneal-wavefront guided transepithelial photorefractive keratectomy after corneal collagen cross linking in keratoconus. J Optom. 2017;10:52–62. doi:10.1016/j.optom.2016.02.001 [CrossRef]
- Shaheen MS, Shalaby Bardan AS, Pinero DP, et al. Wave front-guided photorefractive keratectomy using a high resolution aberrometer after corneal collagen cross-linking in keratoconus. Cornea. 2016;35:946–953. doi:10.1097/ICO.0000000000000888 [CrossRef]
- Kanellopoulos AJ, Asimellis G. Keratoconus management: long-term stability of topography-guided normalization combined with high-fluence CCL stabilization (The Athens Protocol). J Refract Surg. 2014;30:88–93. doi:10.3928/1081597X-20140120-03 [CrossRef]
- Sakla H, Altroudi W, Muñoz G, Albarrán-Diego C. Simultaneous topography-guided partial photorefractive keratectomy and corneal collagen cross-linking for keratoconus. J Cataract Refract Surg. 2014;40:1430–1438. doi:10.1016/j.jcrs.2013.12.017 [CrossRef]
- Fadlallah A, Dirani A, Chelala E, Antonios R, Cherfan G, Jarade E. Non-topography-guided PRK combined with CXL for the correction of refractive errors in patients with early stage keratoconus. J Refract Surg. 2014;30:688–693. doi:10.3928/1081597X-20140903-02 [CrossRef]
- Alessio G, L'abbate M, Sborgia C, La Tegola M. Photorefractive keratectomy followed by cross-linking versus cross-linking alone for management of progressive keratoconus: two-year follow-up. Am J Ophthalmol. 2013;155:54–65. doi:10.1016/j.ajo.2012.07.004 [CrossRef]
- Labiris G, Sideroudi H, Angelonias D, Georgantzoglou K, Kozobolis VP. Impact of corneal cross-linking combined with photorefractive keratectomy on blurring strength. Clin Ophthalmol. 2016;10:571–576 doi:10.2147/OPTH.S100770 [CrossRef]
- Kontadakis GA, Kankariya VP, Tsoulnaras K, Pallikaris AI, Plaka A, Kymionis GD. Long-term comparison of simultaneous topography-guided photorefractive keratectomy followed by corneal cross-linking versus corneal cross-linking alone. Ophthalmology. 2016;123:974–983. doi:10.1016/j.ophtha.2016.01.010 [CrossRef]
- Sherif AM, Ammar MA, Mostafa YS, Gamal Eldin SA, Osman AA. One-year results of simultaneous topography-guided photorefractive keratectomy and corneal collagen cross-linking in keratoconus utilizing a modern ablation software. J Ophthalmol. 2015;2015:321953. doi:10.1155/2015/321953 [CrossRef]
- Labiris G, Giarmoukakis A, Sideroudi H, Gkika M, Fanariotis M, Kozobolis V. Impact of keratoconus, cross-linking and cross-linking combined with photorefractive keratectomy on self-reported quality of life. Cornea. 2012;31:734–739. doi:10.1097/ICO.0b013e31823cbe85 [CrossRef]
- Cennamo G, Intravaja A, Boccuzzi D, Marotta G, Cennamo G. Treatment of keratoconus by topography-guided customized photorefractive keratectomy: two-year follow-up study. J Refract Surg. 2008;24:145–149.
- Bahar I, Levinger S, Kremer I. Wavefront-supported photorefractive keratectomy with the Bausch & Lomb Zyoptix in patients with myopic astigmatism and suspected keratoconus. J Refract Surg. 2006;22:533–538.
- Alpins N, Stamatelatos G. Customized photoastigmatic refractive keratectomy using combined topographic and refractive data for myopia and astigmatism in eyes with forme fruste and mild keratoconus. J Cataract Refract Surg. 2007;33:591–602. doi:10.1016/j.jcrs.2006.12.014 [CrossRef]
- Tambe DS, Ivarsen A, Hjortdal J. Photorefractive keratectomy in keratoconus. Case Rep Ophthalmol. 2015;6:260–268. doi:10.1159/000431306 [CrossRef]
- Khakshoor H, Razavi F, Eslampour A, Omdtabrizi A. Photorefractive keratectomy in mild to moderate keratoconus: outcomes in over 40-year-old patients. Indian J Ophthalmol. 2015;63:157–161. doi:10.4103/0301-4738.154400 [CrossRef]
- Guedj M, Saad A, Audureau E, Gatinel D. Photorefractive keratectomy in patients with suspected keratoconus: five-year follow-up. J Cataract Refract Surg. 2013;39:66–73. doi:10.1016/j.jcrs.2012.08.058 [CrossRef]
- Chelala E, Rami HE, Dirani A, Fadlallah A, Fakhoury O, Warrak E. Photorefractive keratectomy in patients with mild to moderate stable keratoconus: a five-year prospective follow-up study. Clin Ophthalmol. 2013;7:1923–1928.
- Kanellopoulos AJ, Binder PS. Collagen cross-linking (CCL) with sequential topography-guided PRK: a temporizing alternative for keratoconus to penetrating keratoplasty. Cornea. 2007;26:891–895. doi:10.1097/ICO.0b013e318074e424 [CrossRef]
- Kymionis GD, Kontadakis GA, Kounis GA, et al. Simultaneous topography-guided PRK followed by corneal collagen cross-linking for keratoconus. J Refract Surg. 2009;25:S807–S811. doi:10.3928/1081597X-20090813-09 [CrossRef]
- Mrochen M, Jankov M, Bueeler M, Seiler T. Correlation between corneal and total wavefront aberrations in myopic eyes. J Refract Surg2003;19:104–112.
- Reinstein DZ, Archer TJ, Gobbe M. Is topography-guided ablation profile centered on the corneal vertex better than wavefront-guided ablation profile centered on the entrance pupil?J Refract Surg. 2012;28:139–143. doi:10.3928/1081597X-20111115-01 [CrossRef]
UDVA and CDVA in Included Studiesa
|Author||Year||Preop UDVA||Postop UDVA||Preop CDVA||Postop CDVA|
| Kanellopoulos4||2009||0.9 ± 0.3||0.49 ± 0.25||0.41 ± 0.25||0.16 ± 0.22|
| Camellin et al.5||2016||1.0 ± 0.6||0.4 ± 0.4||0.4 ± 0.2||0.2 ± 0.2|
| Shaheen et al.6||2016||0.93 ± 0.33||0.14 ± 0.11||0.28 ± 0.24||0.05 ± 0.06|
| Murkherjee et al.1||2013||1.39 ± 0.5||0.27 ± 0.4||0.31± 0.2||0.11 ± 0.13|
| Tuwairqi & Sinjab2||2012||1.72 ± 2.32||0.01 ± 0.072||−0.025 ± 0.077||−0.04 ± 0.063|
| Kanellopoulos & Asimellis7||2014||0.18 ± 0.20||0.59 ± 0.28||0.62 ± 0.23||0.82 ± 0.19|
| Kanellopoulos4||2009||0.96 ± 0.2||0.3 ± 0.2||0.39 ± 0.3||0.11 ± 0.16|
| Sakla et al.8||2014||0.79 ± 0.36||0.23 ± 0.33||0.28 ± 0.20||0.06 ± 0.07|
| Fadlallah et al.9||2014||0.39 ± 0.22||0.12 ± 0.14||0.035 ± 0.062||0.036 ± 0.058|
| Alessio et al.10||2013||0.63 ± 0.36||0.19 ± 0.17||0.06 ± 0.08||0.03 ± 0.06|
| Labiris et al.11||2016||NA||NA||0.14 ± 0.11||0.08 ± 0.08|
| Kontadakis et al.12||2016||0.83 ± 0.54||0.27 ± 0.25||0.26 ± 0.17||0.09 ± 0.10|
| Sherif et al.13||2015||0.83 ± 0.37||0.26 ± 0.25||0.27 ± 0.31||0.08 ± 0.18|
| Labiris et al.14||2012||NA||NA||NA||NA|
|No CXL group|
| Cennamo et al.15||2008||0.82||0.12||0.07||0.03|
| Bahar et al.16||2006||1.3||0.1||0||0|
| Alpins & Stamatelatos17||2007||NA||NA||NA||NA|
| Tambe et al.18||2015||1.1||0.7||0.49 ± 0.27||0.24 ± 0.28|
| Khakshoor et al.19||2015||0.96 ± 0.3||0.02 ± 0.12||0.00 ± 0.07||0.02 ± 0.07|
| Guedj et al.20||2013||NA||0.06 ± 0.26||0.01 ± 0.03||NA|
| Chelala et al.21||2013||0.76 ± 0.34||0.036 ± 0.009||0.018 ± 0.007||0.027 ± 0.003|
Complications Reported in Included Studies
| Kanellopoulos4 (2009)||Mean haze 1.2 ± 0.5|
| Camellin et al.5 (2016)||Mean haze 0.1 ± 0.2; hyperopic shift (n = 2)|
| Shaheen et al.6 (2016)||Grade 1 haze (n = 6)|
| Mukherjee et al.1 (2013)||18% developed haze|
| Tuwairqi & Sinjab2 (2012)||–|
| Kanellopoulos & Asimellis7 (2014)||–|
| Kanellopoulos4 (2019)||Mean haze 0.5 ± 0.3|
| Sakla et al.8 (2014)||Mean haze 0 at 12 months|
| Fadlallah et al.9 (2014)||Mean haze 0.53 ± 0.32|
| Alessio et al.10 (2013)||Grade 0.5–1 haze (n = 11, 7 PRK + CXL, 4 CXL)|
| Labiris et al.11 (2016)||–|
| Kontadakis et al.12 (2016)||Mild clinical haze|
| Sherif et al.13 (2015)||5% mild haze at 12 months|
| Labiris et al.14 (2012)||–|
|No CXL group|
| Cennamo et al.15 (2008)||Mean haze 0.5 in 4 eyes|
| Bahar et al.16 (2006)||7.5% grade 2 haze persisted after 3 months|
| Alpins & Stamatelatos17 (2007)||–|
| Tambe et al.18 (2015)||3 months: 20 eyes grades 1–2 haze, 6 eyes grades 3–4 haze; latest follow-up: 15 eyes grade 1 and 1 eye grade 2|
|Khakshoor et al.19 (2015)||–|
|Guedj et al.20 (2013)||–|
|Chelala et al.21 (2013)||1 month: 8 eyes had grade 2; 6 months: 1 eye had grade 2; 5 years: no eyes with haze|
|Author||Year||Study Design||Follow-up (mo)||No. of Eyes||CXL Protocol||PRK Platform|
| Kanellopoulos4||2009||Retrospective case series||36||127||Epi-off 3 mW/cm2 for 30 minutes||Topography guided|
| Camellin et al.5||2016||Retrospective case series||10 ± 8||39||Epi-off 3 mW/cm2 for 30 minutes||Wavefront guided|
| Shaheen et al.6||2016||Prospective case series||12||34||Epi-off 3 mW/cm2 for 30 minutes||Wavefront guided|
| Mukherjee et al.1||2013||Prospective case series||12||22||Epi-off 3 mW/cm2 for 30 minutes||Standard|
| Tuwairqi & Sinjab2||2012||Prospective case series||12||22||Epi-off 3 mW/cm2 for 30 minutes||Topography guided|
| Kanellopoulos & Asimellis7||2014||Prospective case series||36||231||Accelerated high fluence 10 mW for 10 minutes||Partial topography guided|
| Kanellopoulos4||2009||Retrospective case series||36||198||Epi-off 3 mW/cm2 for 30 minutes||Topography guided|
| Sakla et al.8||2014||Retrospective cohort||12||31||Epi-off 3 mW/cm2 for 30 minutes||Topography guided|
| Fadlallah et al.9||2014||Retrospective case series||24||140||Epi-off 3 mW/cm2 for 30 minutes||Standard|
| Alessio et al.10||2013||Prospective case series||24||34||Epi-off 3 mW/cm2 for 30 minutes||Topography guided|
| Labiris et al.11||2016||Prospective comparative||12||32 (13)||Epi-off 3 mW/cm2 for 30 minutes||Topography guided|
| Kontadakis et al.12||2016||Prospective comparative||36||60 (30)||Epi-off 3 mW/cm2 for 30 minutes||Topography guided|
| Sherif et al.13||2015||Prospective case series||12||20||Epi-off 3 mW/cm2 for 30 minutes||Topography guided|
| Labiris et al.14||2012||Prospective case series||12||32(13)||Epi-off 3 mW/cm2 for 30 minutes||Topography guided|
|No CXL group|
| Cennamo et al.15||2008||Prospective case series||24||25||No CXL||Topography guided|
| Bahar et al.16||2006||Retrospective case series||40||40||No CXL||Wavefront guided|
| Alpins & Stamatelatos17||2007||Retrospective case series||12||45||No CXL||Topography guided|
| Tambe et al.18||2015||Retrospective case series||84 (median)||28||No CXL||Topography guided|
| Khakshoor et al.19||2015||Prospective case series||30||38||No CXL||Standard|
| Guedj et al.20||2013||Retrospective case series||60||62||No CXL||Standard|
| Chelala et al.21||2013||Prospective case series||60||119||No CXL||Standard|