FormalPara Key Summary Points

This study analyzed the distribution and characteristics of ocular biometric parameters among a large Chinese population aged 3–114 years.

The aging of the global population emphasizes the importance of assessing ocular biometry parameters in older generations, as these parameters may indicate the development of clinical ocular diseases.

A key strength of the present study is its determination of ocular biometric parameters in subjects aged > 100 years, a group that has not been studied previously.

This study reveals the mean values of biometric parameters, including axial length, anterior chamber depth, corneal keratometry, and keratometric astigmatism, among different age groups and genders of the Chinese population.

Significant inter-gender and inter-age differences were observed in the biometric parameters, including axial length, keratometry, keratometric astigmatism, and anterior chamber depth.

These results contribute to a greater understanding of the inter-gender and inter-age differences in the potential prevalence of ocular diseases.

Introduction

The measurement of ocular biometric parameters is an essential aspect of ophthalmic practice. In addition, determining the means and ranges of the ocular parameters at different ages and gender can help establish reference values and guide the diagnosis of ocular diseases [1,2,3,4,5,6,7,8]. However, since ocular biometric parameters differ by age, gender, and geographic region [9,10,11], there is a need to determine the means and distributions of these parameters among different age groups within a population.

To date, several studies have analyzed the ocular biometric parameters in various populations. For example, the anterior chamber depth (ACD) and axial length (AL) has been reported in German subjects aged 16–87 years [12], and the correlation between ACD and age has been determined among Turkish subjects aged 3–85 years [13]. Moreover, a population-based study including patients from Slovakia aged 19–96 years evaluated the ACD, AL, and corneal keratometry (K) [14]. Another study assessed K among the Canadian population aged 19–84 years [15]. Other studies have reported these ocular biometric parameters in subjects of specific ages and gender. However, few population-based studies have analyzed these ocular parameters among Chinese subjects of all ages.

Therefore, this study aimed to investigate the distribution and characteristics of ocular biometric data across a wide age range (3–114 years) among a large Chinese population.

Methods

Study Design

This retrospective cross-sectional study enrolled patients whose ocular biometric parameters were measured at the ophthalmology clinic of West China Hospital, Sichuan University (Chengdu, China), between November 2011 and December 2019, and whose records were stored in the clinic database. Only biometric measurements recorded from one eye were included in the data analysis to avoid bias.

The study protocol was approved by the Ethics Committee on Biomedical Research of West China Hospital of Sichuan University. Furthermore, the study was conducted in accordance with the Declaration of Helsinki.

Measurements

Ocular biometric parameters, including AL, ACD, K, and keratometric astigmatism (KA), were measured using an IOL Master 500 (Version 7.3, Carl Zeiss Meditec, Jena, Germany) optical biometer featuring distance-independent telecentric keratometry. Patients were termed to have high myopia when the monocular axial length exceeded 26 mm [17]. Mean keratometry was calculated as the average of corneal steep K and flat K, while KA was calculated as the difference between steep K and flat K [18]. The gender of the participants was self-reported [16].

Inclusion and Exclusion Criteria

Raw data retrieved from the clinic database were included in the initial analysis. The data were read and collated by two independent researchers according to the unified quality standard, with results cross-checked. Conflicting views were solved by consensus. Only the initial measurements were included in the analysis for patients with several repeat measurements. Patients with ocular trauma and inconsistent measurement equipment, age, gender, or missing patient data for any of the above indicators were excluded from the analysis. After the data were collated and cleaned, further data analysis was performed (Fig. 1).

Fig. 1
figure 1

The flowchart for data collation and cleaning process

Full size image

Statistical Analysis

All measurement data were recorded in Microsoft Excel (Microsoft Office 2016; Microsoft, Redmond, WA, USA) by two independent researchers. Statistical analyses were performed using SPSS version 26.0 (Chicago, IL, USA) and GraphPad Prism version 8.3 (San Diego, CA, USA) software.

Data were expressed as mean ± standard deviation (SD). The normal distribution of the data was determined using Kolmogorov–Smirnov (K–S) tests. Unpaired two-tailed t-tests were used to evaluate differences between two groups with normally distributed data, while nonparametric Mann–Whitney tests were used to evaluate differences between two groups with non-normally distributed data. Differences between three or more groups were evaluated using Kruskal–Wallis tests, followed by Dunn’s multiple comparison tests. Correlations between ocular biometric parameters and between these parameters and age were evaluated using Pearson or Spearman correlation tests. A p-value < 0.05 was considered statistically significant.

Results

Distribution and Characteristics of Ocular Biometric Parameters of the Study Population

Data were collected from 146,748 patients who were seen at the ophthalmology clinic of West China Hospital, Sichuan University. However, data from 60,978 patients were excluded due to missing information on age, gender, and ocular biometric parameters. Furthermore, duplicated data were excluded. A total of 85,770 subjects aged 3–114 years (Fig. 2), including 42,218 males (49.22%) and 43,552 females (50.78%), were included in the final analysis.

Fig. 2
figure 2

Age distribution of the study participants

Full size image

The study participants had a mean AL of 24.61 mm, mean ACD of 3.30 mm, mean K of 43.7 D, and mean keratometric astigmatism of 1.19 D (Table 1). The mean AL was 0.11 mm longer in males than in females, while K was steeper in males than in females. However, the mean ACD was shallower in females than in males. The difference in KA between females and males was 0.04 D.

Table 1 Ocular biometric parameters of the study population (N = 85,770)
Full size table

Skewness was used to denote the degree of the skewed distribution of the ocular biometric parameters and was a measure of data asymmetry. The skewness included normal distribution (skewness = 0), right-skew distribution (positive skew distribution, skewness > 0), and left-skew distribution (negative skew distribution, skewness < 0). Kurtosis showed the peakedness of the distribution of the ocular biometric parameters and was divided into mesokurtic, leptokurtic, and platykurtic [19]. The AL (median, 24.02 mm; K–S test, p < 0.0001) was positively skewed, with a kurtosis of 2.815 (Fig. 3). In addition, KA showed a right skew distribution (median, 0.97; K–S test, p < 0.0001) with a kurtosis of 21.14. In contrast, ACD (median, 3.36; K–S test, p < 0.0001) and K (median, 43.81; K–S test, p < 0.0001) showed negative skew distributions with ACD having a kurtosis of 0.097 and K having a kurtosis of 2.756 (Figs. 4 and 5). All the parameters (AL, KA, ACD, and K) showed that the overall data distribution was flat compared with the normal distribution, which was platykurtic.

Fig. 3
figure 3

The distribution of the axial length. The D-value was determined using the Kolmogorov–Smirnov test

Full size image
Fig. 4
figure 4

The distribution of the anterior chamber depths. The D-value was determined using the Kolmogorov–Smirnov test

Full size image
Fig. 5
figure 5

The distribution of keratometry among the study participants. The D-value was determined using the Kolmogorov–Smirnov test

Full size image

Inter-Gender Differences of Ocular Biometric Parameters in Different Age Groups

The World Health Organization defines adolescence as the age group of 10–19 years [20,21,22,23,24], and the other participants were divided into age groups of 10-year intervals [25, 26] stratified by gender to determine age- and gender-related differences: subjects aged 3–9 years (Group A, N = 5191), 10–19 years (Group B, N = 11,208), 20–29 years (Group C, N = 6680), 30–39 years (Group D, N = 4100), 40–49 years (Group E, N = 225), 50–59 years (Group F, N = 10,139), 60–69 years (Group G, N = 18,920), 70–79 years (Group H, N = 16,687), 80–89 years (Group I, N = 5384), and 90–114 years (Group J, N = 1236). The ocular biometric parameters in each age group were further stratified based on gender (Table 2 and Table 3).

Table 2 Ocular biometric parameters in females as stratified by age (N = 43,552)
Full size table
Table 3 Ocular biometric parameters in males as stratified by age (N = 42,558)
Full size table

Axial Length (AL)

In males, the AL was shown to increase with age till the age of 30 years, followed by a slight decline. However, in females, the AL was shown to increase with age till age 40 years, followed by a decline. No significant differences in AL were observed between males and females aged 70–114 years. As shown in Fig. 6, most patients with axial lengths over 26 mm were younger than 30 years. The highest percentage of subjects with axial lengths above 26 mm was observed in Group E among females and Group C among males.

Fig. 6
figure 6

The percentage of study participants with an axial length above 26 mm, anterior chamber depth below 2.8 mm, and average corneal keratometry above 48 diopters, as stratified by gender and age groups

Full size image

Corneal Keratometry (K)

As presented in Tables 2 and 3, the corneal keratometry was relatively stable, with fewer variations among all age groups in both males and females. However, there was a slightly increasing trend with fluctuations in females and males. Figure 6 also displays the data on the average keratometry over 48 diopters in each age and gender group to show the potential prevalence of suspect keratoconus in all age groups. Most study participants with K over 48 D were aged 30–39 years, with a proportion of 0.68% in females and 0.70% in males.

In both males and females, KA showed a gradual decrease with age up to 60–79 years, followed by a slight increase until the age of 114 years (Tables 2 and 3).

Anterior Chamber Depth (ACD)

The ACD showed an increasing trend until the age of 20–29 years, followed by a decline in both males and females. Furthermore, the mean ACD differed significantly among Groups A to G. However, there were no significant differences between Groups H, I, and J. There was a declining trend in ACD values with increasing age in males and females. As shown in Fig. 6, females were more likely than males to have a shallower ACD (ACD < 2.8 mm) before the age of 60 years. Moreover, the older population were more likely to have an anterior chamber depth of less than 2.8 mm than the younger population.

Correlation of Ocular Biometric Parameters with Age in Males and females

Table 4 presents the correlation of ocular biometric parameters with age in females among all age groups. Age was negatively correlated with AL (r = −0.256, p < 0.0001), ACD (r = −0.551, p < 0.0001), and KA (r = −0.270, p < 0.0001). However, age was positively correlated with K (r = 0.215, p < 0.0001). In addition, positive correlations were observed between AL and ACD (r = 0.424, p < 0.0001), AL and KA (r = 0.081, p < 0.0001), ACD and KA (r = 0.025, p < 0.0001), and between K and KA (r = 0.110, p < 0.0001). Whereas K was negatively correlated with AL (r = −0.246, p < 0.0001) and ACD (r = −0.070, p < 0.0001).

Table 4 Correlation of ocular biometric parameters with age in females
Full size table

Table 5 presents the correlation of ocular biometric parameters with age in males. The results revealed that age was negatively correlated with AL (r = – 0.301, p < 0.0001), ACD (r = – 0.546, p < 0.0001), and KA (r = – 0.312, p < 0.0001). However, age was positively correlated with K (r = 0.234, p < 0.0001). Furthermore, positive correlations were observed between AL and ACD (r = 0.418, p < 0.0001), AL and KA (r = 0.090, p < 0.0001), ACD and KA (r = 0.044, p < 0.0001), and between K and KA (r = 0.116, p < 0.0001). In contrast, K was negatively correlated with AL (r = – 0.243, p < 0.0001) and ACD (r = – 0.076, p < 0.0001). The correlation coefficients for K, AL, and ACD were almost similar between males and females.

Table 5 Correlation of ocular biometric parameters with age in males
Full size table

Discussion

To the best of our knowledge, this hospital-based population study is the first to analyze ocular biometric parameters among males and females aged 3–114 years living in western China. Ocular biometric parameters differ between males and females. However, most previous studies did not stratify the ocular biometric parameters based on gender [27]. This study stratified the ocular biometric parameters based on gender and age. Knowing the magnitude of age-related changes in ocular biometric parameters can help in diagnosing ocular diseases. Furthermore, the findings of this study can be used to establish a database to monitor age-associated changes in ocular biometry in the Chinese population.

This study showed a positive skewed distribution of AL, consistent with previous studies [14, 28, 29]. The standard value of AL in humans is considered to be 24 mm, regardless of gender or ethnicity [30, 31]. However, AL tends to be longer in myopic and shorter in hypermetropic eyes. The present study showed a mean AL of 24.61 mm, similar to that recorded among South Korean subjects aged 29–95 years [32]. However, the AL recorded in this study was longer than that recorded among Germans aged 16–87 years [12] and Slovaks aged 19–95 years [14] (Supplementary material Table 1). Furthermore, the present study showed that the mean AL was slightly longer in males (24.67 mm) than in females (24.56 mm), consistent with other studies [14, 29, 33, 34].

The present study also showed that the mean AL tends to increase from infancy to adulthood [35]. Consistent with this evidence, we found an increasing trend of AL with age in Groups A and B in both males and females. According to previous studies, Chinese children and adolescents have a longer AL than other ethnic groups [10, 36,37,38], which could partly explain the higher prevalence of myopia in Asia than in other geographic regions [6, 39]. The AL observed in this study among Groups H, I, and J did not differ significantly with the results observed among Mongolians and Latinos [34, 40].

Globally, myopic maculopathy is one of the leading causes of visual impairment and blindness [41]. Furthermore, older age and longer AL are risk factors for myopic maculopathy [42, 43], a corollary to this is that early screening of people with long AL can potentially prevent or retard the onset or progression of myopic maculopathy. The present study showed a high prevalence of long AL (> 26 mm) among study participants aged 20–40 years, suggesting an increased risk for high myopia. Therefore, this population should be advised to undergo regular ocular follow-ups, and prevent any ocular diseases induced by high myopia, which will probably worsen when they get older, with more risk factors related to these ocular dysfunctions.

Studies evaluating the effects of age on ocular biometric parameters, especially AL, have yielded conflicting results. Some previous studies reported that AL changes significantly with age [14, 29, 44], while another previous study [34] reported that AL did not change significantly with age. These discrepancies could be attributed to differences in sample size, the age distribution of the populations, and the ocular measurements. The present study reported a negative correlation of AL with age in males and females aged 3–114 years, consistent with other previous studies [10, 14, 36,37,38, 45]. Moreover, that the lens thickness tends to increase with age is worth mentioning, which may influence the accuracy of AL measurement.

The present study showed a negative skewness of ACD, with a mean kurtosis of 0.097. The mean ACD was 3.30 mm, which was deeper than that recorded among Slovakian [14], Turkish (3.03 mm) [13], and Iranian (2.79 mm) [26] populations with similar age distribution, but shallower than that recorded among South Korean [32] and German (3.36 mm) [12] populations. The present study also showed significant gender differences in ACD, with the ACD being 0.05 mm shallower in females than males, consistent with previous studies [14, 33, 34]. This study showed that the mean ACD (3.66 mm) among participants under 18 years was shallower than that of their Australian counterparts [46] but deeper than that of Egyptian children [47]. In addition, in the present study, ACD among Chinese subjects aged over 40 years was shallower than that in Latin American [40], Portuguese [48], and Cuban [28] populations, but was almost equal to the ACD in the South Korean population [49]. The differences in optical biometric parameters could have been due to differences in ethnicity [49]. Consistent with previous studies, the present study revealed that ACD decreased with increasing age in males and females [25, 44, 50, 51]. Furthermore, the ACD is estimated to decrease by 0.007 mm per year [52]. Age-related changes in ACD are likely the result of rapid lens thickening with age [53, 54]. However, further studies are required to determine the exact mechanisms.

In Asians, especially in China [55], Singapore [56], and India [57], primary acute angle-closure glaucoma (PAACG) is the most common type of acute ocular hypertensive disease leading to low vision, poor visual field, and even blindness [58, 59]. The high prevalence of PAACG in Asia could be due to the biological characteristics of the Asian eye [58, 60]. According to a previous study, shallow anterior chamber depth is a common feature in patients with glaucoma [61]. Furthermore, patients with shallower ACD are more likely to have glaucomatous optic neuropathy [62]. In addition, patients with an ACD of less than 2.80 mm tend to have angle closure glaucoma, compared with patients with an ACD of at least 3 mm [63]. The present study showed that the older population were more likely than younger population to have an ACD of less than 2.80 mm, suggesting an increased risk of glaucoma with age, consistent with a previous study[64].

Corneal curvature, an indicator of corneal health, is measured by keratometry [18]. The present study showed that the corneal curvature was negatively skewed, with a skewness of −0.633 and a kurtosis of 2.756, consistent with other studies [28, 44, 45, 48]. The mean value of K was 43.76 D, steeper than that recorded among a Turkish population (43.20 D) [13] and an Iranian population aged 6–90 years (43.48 D) [18]. However, the mean K among the western Chinese population was flatter than that of the Slovakian (44.03 D) [14] and Canadian populations (43.95 D) [15]. Gender stratification showed that K was flatter in males than females, with a mean difference of 0.08 D, in agreement with previous results [13,14,15,16, 18, 38, 48]. The difference in corneal keratometry between males and females may be due, at least in part, to physiological and anatomical differences [13, 18].

The present study showed that K in subjects aged 3–18 years was flatter (43.04 D) than that in similar populations in Germany [10], Iran [18], and Egypt [47]. In addition, K was positively correlated with age in males and females, consistent with previous findings [14, 16, 18]. However, the differences were less significant in subjects over 60 years, suggesting that physiological effects and corneal biomechanisms may be responsible for age-associated alterations in K [65].

Keratoconus is a corneal disease characterized by thinning of the cornea. However, the pathophysiological mechanisms behind keratoconus remain poorly understood [66]. Most subjects are diagnosed with keratoconus in their mid-twenties and early thirties (mean age at diagnosis, 28 years) [67]. Consistent with these findings, the present study showed a high percentage of subjects aged 30–39 years with K values of above 48 D. Since the assessment of keratoconus progression is based on corneal curvature measurements [68], abnormal values in corneal keratometry could be used as an early keratoconus warning.

Consistent with previous studies [14, 29, 69,70,71,72], the present study showed that K was negatively correlated with AL. The negative correlation may be associated with emmetropization, suggesting that age-associated physiological changes in corneal structure could affect AL [73,74,75,76]. Generally, ACD is considered to be positively correlated with keratometry [46]. However, this study showed a negative correlation between ACD and keratometry in both males and females. This difference could possibly be attributed to variations in the refractive states of the study participants, and it is likely the result of mechanical influences on the peripheral cornea, which makes the cornea less prolate as the anterior chamber elongates in response to axial length elongation [77]; also, the changes of ACD can be influenced by many other characteristics, including the intraocular pressure [78], glaucoma, cataracts, and other ocular diseases.

The KA in the present study was higher than that recorded among Iranian [18, 79], Slovakian [14], Canadian [15], and Portuguese [48] populations. However, it was lower than that recorded in a population in Turkey [13], and similar to that recorded among the Korean population [80, 81]. Therefore, it is hypothesized that different ethnicities could have different KAs. Similar to previous findings [13, 14, 18, 79], the present study showed between-gender differences in KA. However, the difference was not statistically significant. These results may underestimate the role of variation among ethnic groups and measurements [48, 49].

This study had several limitations. First, this was a cross-sectional and retrospective hospital-based study. Therefore, this study could not show changes in ocular biometric parameters over time in individual subjects. Secondly, the central corneal thickness, the type of keratometric astigmatism, horizontal corneal diameter, lens thickness, and refraction states were not recorded because of the measurement instruments and incomprehensive recording limitations. Thirdly, although data on age and gender were available, other demographic characteristics, such as smoking history and other systemic diseases, were unavailable for all subjects.

Conclusions

This hospital-based study is the first to report the distribution and characteristics of ocular biometric parameters among a large western Chinese population of subjects aged 3–114 years. A vital strength of the present study was the determination of ocular biometric parameters in subjects aged > 100 years, a group that had not been studied previously.

Mean AL, ACD, K, and KA were 24.61 mm, 3.30 mm, 43.76 D, and 1.19 D, respectively. AL was positively correlated with ACD and KA, and negatively correlated with age and K. In addition, ACD was negatively correlated with age. Significant inter-gender and inter-age differences were observed in the ocular parameters, which may contribute to a greater understanding of the inter-gender and inter-age differences in the prevalence of several ocular diseases.