Explore the correlation between axial length/corneal radius ratio (AL/CR) and refractive error, and evaluate the accuracy of AL/CR in predicting pre-myopia in children.

The study selected 4544 cases (9088 eyes) of children aged 4 to 12 years old from kindergartens and primary schools in Weihai City and Guan County, Liaocheng City, Shandong Province, who had complete refractive and ocular biological parameter data from the " Shandong Children's Eye Disease Research" project in September to December 2012 as the research subjects. Among them, there were 2451 males (4902 eyes) and 2093 females (4186 eyes); The age range is 4 to 12 years old, with an average age of (8.5 ± 2.2) years. All subjects were examined for refraction, axial length (AL), and corneal curvature radius (CR) before cycloplegia, and auto-refraction was performed after cycloplegia. The data obtained from the detection or calculation of equivalent spherical refraction, axial length, corneal curvature radius, and AL/CR ratio conforms to a normal distribution after normality testing, represented by ±s. The comparison of refractive error, AL, and CR among different age groups was conducted using one-way analysis of variance. The correlation between equivalent spherical (SE) refraction and AL/CR ratio, as well as AL and CR, was analyzed using Spearman correlation analysis. Draw receiver operating characteristic (ROC) curves and calculate the optimal cut off value, sensitivity, specificity, youden index, and area under the curve (AUC). The ROC curve analysis results were drawn using cycloplegic refraction with -0.50 D<SE ≤+ 0.75 D as the diagnostic criteria for pre-myopia and AL/CR ratio as the indicator for diagnosing pre-myopia to evaluate its diagnostic efficacy.

Among all 4544 right eyes (4544 eyes) examined, 1481 cases (1481 eyes) were diagnosed with pre-myopia, accounting for 32.6%; the prevalence of pre myopia in all subjects is within the age range of 4 to 12 years old, and shows an increasing trend with age. Among them, the number of cases and pre-myopia cases (percentage) among 112 cases aged 4 were 21 out of 112, accounting for 18.8%; out of 354 cases aged 569 cases accounted for 19.5%; out of 435 cases aged 696 cases accounted for 22.1%; out of 639 cases aged 7195 cases accounted for 30.5%; out of 739 cases aged 8267 cases accounted for 36.1%; out of 541 cases aged 9, 201 cases accounted for 37.2%; out of 702 cases aged 10, 269 cases accounted for 38.3%; out of 560 cases aged 11, there were cases, accounting for 35.7%; among 462 cases aged 12, there were 163 cases, accounting for 35.3%.The SE of 4-year-old children was (1.33±1.18) D, which develops towards myopia at the age of 12 (-0.94±1.89) D. The AL of 4-year-old children was (22.13±0.72) mm, gradually increasing to (23.93±1.09) mm at the age of 12. The AL/CR ratio of 4-year-old children is (2.83±0.09), gradually increasing to (3.03±0.11) at the age of 12. As age increases, the SE, AL, and AL/CR ratios of the subjects all shew an increase towards myopia, and the differences are statistically significant (F=145.66, 197.75, 257.97; P<0.05); however, there was no significant difference in CR, and the difference was not statistically significant (F=0.977, P>0.05). The SE refraction of all 4544 preschool and primary school children aged 4 to 12 years old was negatively correlated with AL and AL/CR ratio, and the correlation was statistically significant (r=-0.736, 0.851; P<0.05). The correlation between SE and CR was not statistically significant (r=-0.028, P>0.05). The optimal cut-off values, sensitivity, specificity, Youden index, and AUC (95%CI) for children of different age groups were 2.86, 0.77, 0.76, 0.53, and [0.815 (0.716, 0.913), P<0.05] at the age of 4; At the age of 5, the values were 2.85, 0.80, 0.64, 0.45, and [0.799 (0.746, 0.852), P<0.05]. At the age of 6, the values were 2.88, 0.82, 0.72, 0.55, and [0.823 (0.777, 0.869), P<0.05]. At the age of 7, the values were 2.93, 0.66, 0.84, 0.50, and [0.805 (0.769, 0.841), P<0.05]. At the age of 8, the values were 2.92, 0.85, 0.66, 0.51, and [0.840 (0.812, 0.867), P<0.05]. At the age of 9, the values were 2.98, 0.70, 0.90, 0.60, and [0.858 (0.827, 0.889), P<0.05]. At the age of 10, the values were 2.96, 0.78, 0.84, 0.62, and [0.878 (0.853, 0.903), P<0.05]. At the age of 11, the values were 3.00, 0.73, 0.88, 0.61, and [0.874 (0.842, 0.905), P<0.05]. At the age of 12, the values were 3.00, 0.75, 0.93, 0.68, and [0.892 (0.861, 0.923), P<0.05]. As age increases, the optimal cut-off values for diagnosing premyopia gradually increases.

AL/CR is highly correlated with refractive error and has high accuracy in predicting premyopia before cycloplegia. In this study, the optimal cut off values for AL/CR prediction in children of different ages can provide a simple and reliable evaluation method for early screening of high-risk children for myopia, which will help with early intervention for myopia prevention.