Variation of Orthokeratology Lens Treatment Zone (VOLTZ) Study: A 2‐year randomised clinical trial

To compare axial elongation (AE) and treatment zone (TZ) characteristics in children wearing 6 mm or 5 mm back optic zone diameter (BOZD) orthokeratology (ortho‐k) lenses over 2 years.

The Variation of Orthokeratology Lens Treatment Zone (VOLTZ) Study, a randomised clinical trial, was conducted to compare the myopia control efficacy of lenses with smaller BOZD to standard ortho-k lenses over 2 years.As reported previously, ortho-k lenses with a 5 mm BOZD can retard AE by 0.13 mm compared to a 6 mm construction. 7his paper aims to report the 2-year results of the VOLTZ study, including AE and TZ, and the association between AE and various TZ parameters.

Study design
This 2-year VOLTZ study was a randomised, double-masked clinical trial (registered on 19/06/2017, NCT03191942) conducted in Hong Kong, comparing AE and TZ characteristics between children wearing smaller or standard BOZD ortho-k lenses. 7Twenty-four subjects in each group were required to complete the study, providing a 80% power (5% significance) to detect a minimum between-group difference of 0.18 mm in terms of AE, using a reference standard deviation (SD) of 0.25 mm. 7The study was approved by the Ethics Committee of The Hong Kong Polytechnic University (PolyU), and all clinical procedures were conducted following the tenets of the Declaration of Helsinki.

Subjects
The detailed inclusion and exclusion criteria have been reported previously. 7In brief, children (6-11 years old) of Chinese ethnicity, with mild to moderate myopia (−4.00 to −0.75 D) and mild to moderate astigmatism (≥−2.50D withthe-rule), normal general/ocular health and no history of myopia control therapy were recruited.Complimentary care solutions (daily cleaner [O 2 Daily Care Solution, ophte cs.com], saline [Cleadew Dissolving & Rinsing Solution, ophte cs.com], povidone-iodine-based disinfecting solution [Cleadew GP Advanced Care System, ophte cs.com] and unidose artificial tears [Tiare W artificial tears, ophte cs.com]) were provided to enhance compliance.Subjects were required to attend both the ortho-k aftercare visits (first night, first week, first month and every 3 months) as well as the 6 monthly cycloplegic examinations during the study.

Interventions and masking
Subjects were randomly assigned (1:1) (by spreadsheet generator [Excel; Microsoft, micro soft.com]) to wear BE Free lenses (KATT Design Group, Precision Technology Services, ptsop tics.com)having a BOZD of 6 mm (6-mm group) or 5 mm (5-mm group). 7All procedures (subject recruitment, lens fitting, data collection and aftercare) were conducted by the same examiner over the 2 years, except for the primary outcome (axial length [AL] after cycloplegia) which was determined by a masked examiner.

Measurement procedures
Subjects who achieved a successful lens fit 7 were required to attend cycloplegic examinations (two drops of 1% cyclopentolate) at the first month (±2 days), and then six monthly (±2 weeks) thereafter, for data collection.All the procedures were conducted at the Optometry Clinic, School of Optometry of PolyU.All subsequent visits were conducted within ±2 h of the time at the baseline visit to minimise diurnal differences in the measurements.
At the data collection visits, subjective refraction was measured with trial lenses before and after cycloplegia.Unaided (UVA) and best-corrected (BCVA) visual acuity were assessed under normal lighting using high contrast (100%) Early Treatment of Diabetic Retinopathy Study charts (Precision Vision, preci sion-vision.com).Corneal topography (Medmont E300, medmo nt.com) was assessed before cycloplegia.Four topographical maps with similar contour and differences within 0.25 D in central K-readings were obtained. 8Subtractive tangential maps (before and after ortho-k treatment) were selected to evaluate TZ decentration and other TZ parameters, because it provides the most accurate measurement of localised changes after ortho-k. 8Pupil size was estimated from the topography measurement under dark illumination (18.8 lux), but was recorded as the photopic size because of the intrinsic lighting of the Medmont instrument. 6ub-foveal choroidal thickness (ChT) was measured before cycloplegia by the same unmasked examiner conducting the clinical trial using Spectralis SD-OCT (Heidelberg Engineering, Heidelberg, Germany, busin ess-lounge.heidelberg engin eering.com)under enhanced depth imaging mode (six long-scans with 30° separation; each included 30 scans).The scan with the best image quality at baseline was

Key points
• Orthokeratology lenses with smaller back optic zone diameters resulted in less axial elongation over 2 years, compared to larger optical zones.• Smaller back optic zone diameter orthokeratology lenses resulted in smaller treatment zone diameters, which was associated with less axial elongation.• Changes in sub-foveal choroidal thickness was unaffected by the treatment zone size over time and hence had limited effect on axial elongation.
selected as the reference and images on subsequent visits were captured with reference to this one to ensure that the same location of retina was scanned.The first three images with >20 dp resolution were exported and auto-segmented using Matlab (mathw orks.com)software (manual adjustments applied where necessary). 9,10The horizontal scans were selected for further analysis.Axial length measurement (IOLMaster 500; Carl Zeiss Meditec, zeiss.com)was performed after cycloplegia, by a masked examiner.Five measurements (signal-to-noise ratio ≥5) with betweendifference ≤0.02 mm were averaged for analysis.

Treatment zone analysis
The best topographical maps were selected at each data collection visit, according to the criteria previously presented. 8Tangential curvature data were exported from the Medmont device and imported into customised software for analysis (Figure 1). 8The TZ was defined as the central flattened zone surrounded by points with zero changes on subtractive tangential maps, which was determined automatically by the software using a best-fit ellipse technique.Treatment zone parameters were determined according to the method described by Guo et al. 8 using the same 30 parameters introduced previously with the addition of TZ area.The analysis included, but was not limited to, TZ decentration, TZ diameter, width and peak of the peripheral steepened zone (PSZ) and the rate of power change with respect to the TZ centre, TZ edge and the deepest point within the TZ.

Statistical analysis
Data from subjects who completed the study (n = 45, right eyes only) were analysed.One 6-mm subject had a poorly formed TZ due to an inadequate response to ortho-k, and was excluded from the TZ analysis.ChT measurements were missing for one subject in each group at baseline and were excluded.

R ESULTS
Of the 82 subjects who were randomised, 70 commenced lens wear (6-mm group: n = 34; 5-mm group: n = 36).After 2 years, twenty-three 6-mm subjects and twenty-two 5-mm subjects completed the study.Eleven (32%) and 14 (39%) subjects from the 6-and 5-mm groups, respectively, dropped out of the study.The reasons for dropout are listed in Figure 2 and none were due to rapid AE.The baseline data of subjects who either completed the study or dropped out are shown in Table 1.Subjects who dropped out from the 6-mm group were more likely to be male with longer baseline AL and a higher refraction (i.e., more myopia and astigmatism); subjects who dropped out from the 5-mm group tended to be older (p = 0.03).No significant difference in the baseline parameters was found between the two BOZD groups (for those who completed the study), except for slightly greater, but clinically insignificant, astigmatism in the 5-mm group (0.25 D, p = 0.04).

Axial elongation
Repeated measures ANOVA revealed significant time (F = 80.53, p < 0.001) and group (F = 14.00, p = 0.001), but no group by time effects (F = 1.20, p = 0.30) on the measurements of AE, which suggested that the between-group difference in AE mainly occurred during the first 6 months (Figure 3).Post-hoc analysis revealed that the 5-mm group had slower AE compared with the 6-mm group over the study period (p < 0.01 after Bonferroni correction), with a difference of 0.20 mm at the 24-month visit (5-mm: 0.15 ± 0.21 mm; 6-mm: 0.35 ± 0.23 mm) (Table 2).Based on this, the power of the study, calculated using G-Power analysis, 11 was 83%.

Visual acuity, refraction and sub-foveal choroidal thickness
Repeated measures analysis showed significant time (F = 205.57,p < 0.001), group (F = 5.58, p = 0.02) and group by time (F = 5.03, p = 0.005) effects on SER.No significant time (F = 1.67, p = 0.18) or group by time (F = 0.59, p = 0.62) effect was observed for BCVA, but a borderline group effect was noted (F = 4.15, p = 0.048).Analysis focusing on changes over time showed no significant time (F = 0.47, p = 0.71) or group by time interaction on SER (F = 0.46, p = 0.71), but a significant group effect (F = 8.71, p = 0.005), suggesting that most of the changes occurred during the first 6 months.None of these effects were found for UVA starting from the 6month visit (F ≥ 0.24, p ≤ 0.87).Table 2 lists changes in post-cycloplegic SER in both groups during the 2-year study.Significantly more change in SER was noted in the 5-mm group at the 6-, 12-and 18-month visits, with clinically significant differences of 0.96, 0.89 and Illustration of reference points for treatment zone (TZ) measurements using customised software, derived from tangential curvatures exported from the Medmont E300 corneal topographer.
0.84D, respectively (p ≤ 0.007), but not at the 24-month visit.There were no significant differences between the groups in terms of residual astigmatism, BCVA and UVA at any visit (p > 0.01 after Bonferroni correction).Repeated measures analysis, excluding the 18-month visit, showed a significant time effect (F = 3.88, p = 0.04) and marginal group effect (F = 4.31, p = 0.047) for sub-foveal ChT, with no significant group by time interaction (F = 0.70, p = 0.46).On the contrary, none of these effects were significant with respect to changes in ChT (F ≥ 0.62, p ≤ 0.44), suggesting that the differences occurred mainly during the first 6 months.Post-hoc analysis showed that betweengroups sub-foveal ChT was not significantly different at the 6-, 12-and 24-month visits (Table 2).Compared to baseline (pooled data analysis), a significant increase in ChT was observed at the 6-month (11.8 ± 19.77 μm, p < 0.001) and 12-month (12.0 ± 23.7 μm, p = 0.004) visits; no such findings were found for the other visits during the study.Using simple linear regression, a significant negative association was found between changes in ChT and AE at each visit (p < 0.001).

Treatment zone characteristics
Comparison between groups at the 24-month visit Table 3 presents comparisons of various TZ parameters between the two groups.An unpaired t-test or its nonparametric equivalent showed that the 5-mm BOZD group exhibited lower measured data for all the following characteristics: TZ diameter (horizontal and vertical, p < 0.001), TZ area (p < 0.001), PSZ power (temporal: p = 0.02; nasal: p = 0.001), nasal myopic defocus (p = 0.01) and nasal rate of refractive power change (to TZ edge, p = 0.02).These seven TZ parameters were analysed using repeated measures analysis and results showed no time or group by time effect (F ≥ 0.01, p ≤ 0.99).Significant group effects were found for TZ area (F = 179.40,p < 0.001), TZ diameters (horizontal: F = 166.93,p < 0.001; vertical: F = 125.20,p < 0.001) and PSZ peak power (temporal: F = 8.69, p = 0.005; nasal: F = 7.63; p = 0.008), while no significant findings were found for nasal defocus (F = 2.21, p = 0.15) or nasal slope to the TZ edge (F = 2.93, p = 0.09).Post hoc analysis, after Bonferroni correction, showed significantly smaller TZ area and TZ diameters in the 5-mm group at all visits (p < 0.001), but smaller nasal PSZ peak power (p = 0.001) and nasal defocus (p = 0.006) only at the 24-month visit (Table 4).

Factors affecting treatment zone parameters
LMM analysis showed that lens design was the major factor affecting all seven TZ parameters mentioned above (β ≤ −0.80, p ≤ 0.03), although nasal defocus, nasal and temporal PSZ peak powers and nasal slope (to TZ edge) were also affected by baseline SER (β ≤ −0.10, p ≤ 0.01).

Factors associated with AE
The seven TZ parameters (at the 6-month visit) were considered as fixed factors in the LMM because they were stable over time, and each parameter was included in one model to avoid high collinearity.Change in sub-foveal ChT was included as continuous data because it was found to increase after treatment using pooled data.Of the seven LMM models, the one including the horizontal TZ diameter (followed by vertical TZ size and TZ area) gave the best fit with the smallest Akaike information criterion score, 12 and was therefore considered as the most relevant to AE in this study.
When analysing the factors affecting AE with LMM using pooled data (Table 5), results showed that only the changes in ChT, TZ diameters and TZ area were significantly associated with AE.Similar results were obtained for the three models including TZ diameters (horizontal and vertical) and TZ area (Models 1-3).Model 1 showed that pooled horizontal TZ diameter at the 6-month visit was positively associated with AE (β = 0.13, p < 0.001), but not in individual 6-mm or 5-mm groups.Changes in ChT in each group, as well as in pooled data, were negatively associated with AE, although weak (β = −0.002 to −0.003).Baseline SER was found to be positively associated with AE using pooled data (β = 0.05, p = 0.048) and baseline age was negatively associated with AE in the 6-mm group only (β = −0.08,p = 0.02).

Ocular adverse events
No significant visual symptoms were reported, except for halos in three 5-mm subjects and one 6-mm subject which subsided within 2 weeks. 7No microbial keratitis was observed in any subject during the study period.Among the 16 dropouts experiencing adverse events, 15 were due to contact lens-related adverse events (Figure 2).One severe (corneal haze), five significant (microcysts, recurrent/ consistent corneal staining and corneal erosion) and nine non-significant adverse events (allergic conjunctivitis and Note: 6-mm-using orthokeratology lenses of BOZD 6 mm; 5-mm-using orthokeratology lenses of BOZD 5 mm; p-probability value of unpaired t-or Mann-Whitney U-tests for difference between subjects who completed or dropped out of the study.p'-probability value of unpaired t or Mann-Whitney U-tests for difference between groups in subjects who completed the study.Abbreviations: BCVA, best-corrected visual acuity; BOZD, back optic zone diameter; SER, spherical equivalent refraction.*Significant differences between subjects who completed or dropped out of the study.sterile infiltrate) were observed during the 2 years (Table 6).
During the study period, 14 subjects (eight in the 5-mm and six in the 6-mm groups) had non-inflammatory corneal infiltrates.All subjects continued lens wear after re-education on lens handling and switching to an multipurpose solution with Progent A & B (Menicon Co., menic on.com),except for the five subjects (details previously published 7 ).Microcysts 13 were observed in 12 and 9 subjects at different visits in the 6-and 5-mm groups, respectively.All but three continued lens wear after temporary discontinuation.

DISCUSSION
Analysis showed that ortho-k lenses with smaller BOZD can create a smaller TZ diameter, resulting in less AE over 2 years.The most dramatic changes occurred during the first 6 months for TZ diameter, ChT thickening and AE.Further, smaller BOZD ortho-k lenses neither affected visual performance nor caused significant visual sign/ symptoms in children wearing them, compared with conventional construction.No of keratitis were observed during the 2-year study.One participant with diffuse Grade 4 microcysts and corneal haze was observed during the COVID-19 pandemic lockdown, with details reported previously. 14he incidence of non-inflammatory infiltrates observed here was higher than previous findings. 7Earlier studies only reported mild microcysts in adults wearing ortho-k lenses (5% at baseline and 39% after 3 months of ortho-k wear), 15 while no such events were noted in paediatric ortho-k wearers. 16,17In the current study, 21 subjects presented with microcysts at different visits (Grade 1-Grade 4).Except for one severe case with corneal haze 14 and one who stopped lens wear in the early stage of the study, subjects presenting with more than Grade 2 microcysts (n = 10) resumed lens wear after the condition had subsided to a T A B L E 2 Axial elongation (AE), changes in sub-foveal choroidal thickness (ChT), changes in spherical equivalent refraction (SER) and best-corrected visual acuity (BCVA) at 6-, 12-, 18-and 24-month visits (median [range] or mean ± SD), of subjects who completed the study.low to moderate level (≤Grade 2) after temporary lens discontinuation (26-73 days).
The 5-mm group in this study showed less AE (0.20 mm less) compared to the 6-mm group after 2 years.The 6-mm group showed comparable AE with Cho and Cheung 16 (0.35 vs. 0.36 mm, p = 0.87) after 2 years of lens wear, and achieved a similar myopia control effect (in terms of AE) to other investigations performed worldwide. 18,19Compared with a historical control wearing single-vision spectacles, 16 the 6-mm group achieved 0.28 ± 0.07 mm (44%) less AE (p < 0.001) while the 5-mm group showed 0.48 ± 0.06 mm (76%) less AE (p < 0.001).1][22] For example, Chia et al. 20 reported myopia control of 67% with 0.5% atropine compared with 0.01% atropine after 24 months while Shih et al. 21reported an effect of 62.7% with a combined | 1457 0.5% atropine and progressive addition spectacle therapy compared with single-vision spectacles after 18 months.
Although the retardation of AE for 5 mm BOZD ortho-k lenses was still not comparable to results for 1% atropine (87% 23 and 105% 24 ), it may be a better choice as it does not induce side effects such as photophobia and interrupted near vision. 25,26It is important to note that the improved myopia control effect of the smaller BOZD ortho-k lenses occurs mainly during the first 6 months (which may be due to the stabilised TZ diameter) and the betweengroup difference in AE was not significant afterwards.Tan et al. 27 also reported this observation when comparing AE between ortho-k and combined therapy groups (ortho-k and 0.01% atropine).
The changes in sub-foveal ChT were not significantly different between groups over the study period, with a significant thickening during the first 6 months (pooled data).This difference was small compared with the measurement variance of the instrument (10 μm in previous studies 10,28 ).In addition, the ChT change was less than 4% of the 2-year AE; hence, its contribution to the overall axial change was small and negligible.Multivariate analysis showed a negative association between ChT changes and AE; however, the effect size was very small (close to zero).The results of this study suggest that there may be other factors (e.g., optical changes) responsible for the significant AE difference, and further investigations are necessary.
A significant reduction in SER was noted in both groups, with a greater change in the 5-mm group, which stabilised after the first 6 months.However, the changes in SER were not significantly associated with TZ T A B L E 4 Measurement of treatment zone (TZ) parameters (only those significantly different between groups at 24 months) at 6-, 12-, 18-and 24-month visits (median [range] or mean ± SD) for the 44 subjects (22  Note: p-probability value of unpaired t-or Mann-Whitney U-tests for difference between groups, a significant difference requires p < 0.0125 after Bonferroni correction.6-mm-using orthokeratology lenses of back optic zone diameter (BOZD) 6 mm; 5-mm-using orthokeratology lenses of BOZD 5 mm.*Significant differences between groups.
parameters (except for marginal association with the horizontal TZ diameter; β = −0.10,p = 0.047).At the 24-month visit, significantly smaller TZ (area reduced by 4.45 mm 2 ; horizontal/vertical diameters reduced by 1.55/0.77mm) were noted in the 5-mm group, which were comparable with previous studies. 4,5Significantly less PSZ peak powers (temporal and nasal), defocus (nasal) and peripheral rate of power change (nasal) were observed in the 5-mm group.In addition, the PSZ peak power, defocus and rate of power change exhibited no association with AE, suggesting that the TZ diameter, rather than the peripheral corneal changes, is more likely to contribute to myopia control using ortho-k.This study also showed that lens design was the major factor affecting TZ (especially area   and diameters).Thus the decreased TZ (diameter) resulting from smaller BOZD ortho-k lenses may enhance the myopia control effect in ortho-k.Higher refractive error has been shown to be associated with better myopia control using a conventional 6-mm BOZD design in previous reports. 29,30However, these associations were not observed with the 5-mm design.The efficacy of myopia control using smaller BOZD lenses may likely be affected by the amount of defocus created on the retina, as slower AE may result from a greater change in aberrations (especially spherical aberration). 31,32Further investigations are warranted show this potential association between aberration and myopia progression.
A limitation of this study is the lack of a spectacle control group.Since single-vision spectacles have been shown to have no effect on slowing AE in young children, 16,21 it has become unethical to recruit myopic children and prescribe them single-vision spectacles.Parents must be informed if their children were assigned to a control group and they are likely to withdraw from the study or, even if they agree, when the children exhibited rapid AE, they are likely to seek myopia control treatment elsewhere. 33These problems may lead to self-selection (whether to participate or not), which could result in a biased control group, with subjects who are slow (myopia) progressors.However, comparison has been made with the control group from an earlier report conducted at the same centre. 16Since the AE of their ortho-k subjects (at each 6-month visit) were not significantly different from the 6-mm BOZD ortho-k group in the current study, the comparison with the control group is appropriate and valid.
In conclusion, compared to conventional 6-mm BOZD lenses, the 5-mm BOZD lenses achieved similar clinical performance without affecting ocular integrity.The smaller BOZD lenses also created a smaller TZ diameter, leading to a slower AE (0.20 mm smaller) after 2 years of lens wear.ChT changes, however, did not play a significant role in slowing AE.

F I G U R E 2
Number of subjects dropping out of the study at different visits (6-mm, 5-mm-wearing orthokeratology lenses having a back optic zone diameter of 6 and 5 mm, respectively).

T A B L E 1
Photopic pupil size (mm)

F I G U R E 3
Changes in axial length over 24 months in orthokeratology subjects wearing lenses of either 6 mm or 5 mm back optic zone diameters.Error bars represent 1 SD.
Note: 6-mm-using orthokeratology lenses of 6 mm back optic zone diameter; 5-mm-using orthokeratology lenses of 5 mm back optic zone diameter.a One case was reported by subject, which subsided before aftercare visit.b Diagnosed by history taking only.c Diagnosed by slit lamp examination and history taking.d Reported by subject.
AC K N O W L E D G E M E N T SOrtho-k lenses, lens material and care solutions were supported by Precision Technology Services (Canada), Paragon Vision Sciences (US) and Ophtecs Corporation (Japan), respectively.FU N D I N G I N FO R M AT I O NB. Guo is supported by the Research Residency Scheme of The Hong Kong Polytechnic University.CO N F L I C T O F I N T E R E S T S TAT E M E N TAll authors, except R. Kojima, have no proprietary interest in any of the products mentioned.R. Kojima is a Clinical Research and Development Director for Precision Technology Services (Vancouver, Canada), a partner in the KATT Design Group (Vancouver, Canada) and a clinical advisor to Medmont International Pty (Nunawading, Australia).D E C L A R AT I O NThe results of this study have been previously presented at the 2022 ARVO meeting and the Vision by Design Poster Session of American Academy of Orthokeratology and Myopia Control in 2022.The results have been reported in the PhD thesis entitled "Variation of Orthokeratology Lens Treatment Zone (VOLTZ) Study".O R C I DBiyue Guo https://orcid.org/0000-0001-6282-8143Pauline Cho https://orcid.org/0000-0002-7866-7513RE F E R E N C E S Treatment zone (TZ) parameters at the 24-month visit (median [range] or mean ± SD) for subjects with definable TZ (44 subjects).
T A B L E 3Note: p-probability value of unpaired t-or Mann-Whitney U-tests for difference between groups.6-mm-using orthokeratology lenses of BOZD 6 mm; 5-mm-using orthokeratology lenses of BOZD 5 mm.Abbreviations: BOZD, back optic zone diameter; PSZ, peripheral steepened zone.*Significant differences between groups.
from each group) with definable TZ.
Statistically significant fixed effects and estimates of influences (β) on axial elongation.Number of adverse events (number of subjects) over 24 months in orthokeratology subjects, presented as eyes per visit/period (right eyes).
T A B L E 5 T A B L E 6