Association between axial length elongation and spherical equivalent progression in Chinese children and adolescents

Cutoff values of axial length/corneal radius ratio for determining myopia vary with age among 3-18 years old children and adolescents

PURPOSE

To investigate the effectiveness and cutoffs of axial length/corneal radius (AL/CR) ratio for myopia detection in children by age.

METHODS

Totally, 21 kindergartens and schools were enrolled. Non-cycloplegic autorefraction (NCAR), axial length (AL), horizontal and vertical meridian of corneal radius (CR1, CR2), and cycloplegic autorefraction were measured. Receiver operating characteristic (ROC) curve was used to obtain the effectiveness and cutoff for myopia detection.

RESULTS

Finally, 7803 participants aged 3-18 years with mean AL/CR ratio of 2.99 ± 0.16 were included. Area under the ROC curve (AUC) of AL/CR ratio for myopia detection (0.958 for AL/CR1, 0.956 for AL/CR2, 0.961 for AL/CR) was significantly larger than that of AL (0.919, all P < 0.001), while AUCs of the three were similar with different cutoffs (> 2.98, > 3.05, and > 3.02). When divided by age, the ROC curves of AL/CR ratio in 3- to 5-year-olds showed no significance or low accuracy (AUCs ≤ 0.823) in both genders. In ≥ 6-year-olds, the accuracies were promising (AUCs ≥ 0.883, all P < 0.001), the cutoffs basically increased with age (from > 2.93 in 6-year-olds to > 3.07 in 18-year-olds among girls, and from > 2.96 in 6-year-olds to > 3.07 in 18-year-olds among boys). In addition, boys presented slightly larger cutoffs than girls in all ages except for 16 and 18 years old. For children aged 3-5 years, AL/CR ratio or AL combined with NCAR increased AUC to > 0.900.

CONCLUSION

AL/CR ratio provided the best prediction of myopia with age-dependent cutoff values for all but preschool children, and the cutoffs of boys were slightly larger than those of girls. For preschool children, AL/CR ratio or AL combined with NCAR is recommended to achieve satisfactory accuracy. AL/CR ratio calculated by two meridians showed similar predictive power but with different cutoffs.

Longitudinal analysis of refraction and ocular biometrics in preschool children with early-onset high myopia

We investigated changes in refraction and ocular biometrics in preschool children with early-onset high myopia. Sixty eyes of 60 children with a mean follow-up time of 58.5 months were included in this study. At baseline, mean age of children was 55.6 ± 13.1 months, mean spherical equivalent (SE) was - 8.59 ± 2.66 D, and 25.64 ± 1.16 mm for axial length (AL). The total annual rate of myopic progression and axial elongation were - 0.37 ± 0.39 D/year and 0.33 ± 0.18 mm/year, respectively. During follow-up period, there was a trend toward less myopic progression and axial elongation over time. Of the total participants, 24 children (40%) were in the myopia progression group and the remaining 36 children (60%) were in the myopia stability group. In multiple linear regression analysis, baseline SE and AL were independently associated with myopic progression, while age, sex, and baseline AL-to-CR ratio were not related to myopic progression. According to the model, more myopic SE (β = - 0.186, P = 0.035) and longer AL (β = - 0.391, P = 0.008) at baseline were significantly associated with myopic progression. Myopia progression in preschoolers with high myopia tended to be relatively modest, with 60% of subjects exhibited myopic stability. Higher myopic SE, and longer AL at baseline were associated with myopic progression in preschool children with high myopia.

Normative data for axial elongation in Asian children

AIM

To determine the influence of refractive error (RE), age, gender and parental myopia on axial elongation in Chinese children and to develop normative data for this population.

METHODS

This is a retrospective analysis of eight longitudinal studies conducted in China between 2007 and 2017. Data of 4701 participants aged 6-16 years with spherical equivalent from +6 to -6D contributed to one, two or three annualised progression data resulting in a dataset of 11,262 eyes of 26.6%, 14.8% and 58.6% myopes, emmetropes and hyperopes, respectively. Longitudinal data included axial length and cycloplegic spherical equivalent RE. Axial elongation was log-transformed to develop an exponential model with generalised estimating equations including main effects and interactions. Model-based estimates and their confidence intervals (CIs) are reported.

RESULTS

Annual axial elongation decreased significantly with increasing age, with the rate of decrease specific to the RE group. Axial elongation in myopes was higher than in emmetropes and hyperopes but these differences reduced with age (0.58, 0.45 and 0.27 mm/year at 6 years and 0.13, 0.06 and 0.05 mm/year at 15 years for myopes, emmetropes and hyperopes, respectively). The rate of elongation in incident myopes was similar to that in myopes at baseline (0.33 vs. 0.34 mm/year at 10.5 years; p = 0.32), while it was significantly lower in non-myopes (0.20 mm/year at 10.5 years, p < 0.001). Axial elongation was greater in females than in males and in those with both parents myopic compared with one or no myopic parent, with larger differences in non-myopes than in myopes (p < 0.01).

CONCLUSIONS

Axial elongation varied with age, RE, gender and parental myopia. Estimated normative data with CIs could serve as a virtual control group.

Baseline metrics that may predict future myopia in young children

PURPOSE

We used baseline data from the PICNIC longitudinal study to investigate structural, functional, behavioural and heritable metrics that may predict future myopia in young children.

METHODS

Cycloplegic refractive error (M) and optical biometry were obtained in 97 young children with functional emmetropia. Children were classified as high risk (HR) or low risk (LR) for myopia based on parental myopia and M. Other metrics included axial length (AXL), axial length/corneal radius (AXL/CR) and refractive centile curves.

RESULTS

Based on the PICNIC criteria, 46 children (26 female) were classified as HR (M = +0.62 ± 0.44 D, AXL = 22.80 ± 0.64 mm) and 51 (27 female) as LR (M = +1.26 ± 0.44 D, AXL = 22.77 ± 0.77 mm). Based on centiles, 49 children were HR, with moderate agreement compared with the PICNIC classification (k = 0.65, p < 0.01). ANCOVA with age as a covariate showed a significant effect for AXL (p < 0.01), with longer AXL and deeper anterior chamber depth (ACD) (p = 0.01) in those at HR (differences AXL = 0.16 mm, ACD = 0.13 mm). Linear regression models showed that central corneal thickness (CCT), ACD, posterior vitreous depth (PVD) (=AXL - CCT - ACD-lens thickness (LT)), corneal radius (CR) and age significantly predicted M (R = 0.64, p < 0.01). Each 1.00 D decrease in hyperopia was associated with a 0.97 mm elongation in PVD and 0.43 mm increase in CR. The ratio AXL/CR significantly predicted M (R = -0.45, p < 0.01), as did AXL (R = -0.25, p = 0.01), although to a lesser extent.

CONCLUSIONS

Although M and AXL were highly correlated, the classification of pre-myopic children into HR or LR was significantly different when using each parameter, with AXL/CR being the most predictive metric. At the end of the longitudinal study, we will be able to assess the predictability of each metric.

Comparing the effects of highly aspherical lenslets versus defocus incorporated multiple segment spectacle lenses on myopia control

To compare spectacle lenses with highly aspherical lenslets (HAL) versus defocus incorporated multiple segments (DIMS) on myopia progression control in 1 year. This retrospective cohort study involved data from children prescribed HAL or DIMS spectacle lenses in Guangzhou Aier Eye Hospital, China. To address the discrepancy that some children followed up at less than or more than 1 year, the standardized 1-year spherical equivalent refraction (SER) and axial length (AL) changes from baseline were calculated. The mean differences in the changes between the two groups were compared with linear multivariate regression models. Age, sex, baseline SER/AL, and treatment were included in the models. A total of 257 children who qualified for the inclusion criteria were included for the analyses (193 in the HAL group and 64 in the DIMS group). After controlling baseline variates, the adjusted mean (standard error, SE) of the standardized 1-year changes in SER for HAL and DIMS spectacle lens users were - 0.34 (0.04) D and - 0.63 (0.07) D, respectively. HAL spectacle lenses reduced myopia progression by 0.29 D (95% confidence interval [CI] 0.13 to 0.44 D) at 1 year compared to DIMS lenses. Accordingly, the adjusted mean (SE) ALs increased by 0.17 (0.02) and 0.28 (0.04) mm for children wearing HAL lenses and DIMS lenses, respectively. HAL users had 0.11 mm less AL elongation (95% CI - 0.20 to - 0.02 mm) than DIMS users. Age at baseline was significantly associated with AL elongation. Chinese children wearing spectacle lenses designed with HAL had less myopia progression and axial elongation than those wearing spectacle lenses designed with DIMS.