Review Article

Ocular axial length and its associations: A systematic review

Andrew T. Chikara, Tanya Evans, Solani D. Mathebula

Received: 08 Apr. 2025; Accepted: 12 Aug. 2025; Published: 18 Nov. 2025

Copyright: © 2025. The Author(s). Licensee: AOSIS.
This work is licensed under the Creative Commons Attribution 4.0 International (CC BY 4.0) license (https://creativecommons.org/licenses/by/4.0/).

Abstract

Background: Axial length is a key factor in determining the underlying cause of refractive errors, including myopia. It is essential for assessing myopic progression or risk in patients.

Aim: To systematically evaluate associations between ocular axial length and various ocular biometric, demographic and anthropometric parameters.

Method: A systematic review was conducted on studies highlighting the significance of axial length and its associations with anthropometric factors including height and body mass index; ocular biometry factors including refractive error; corneal thickness and radius; crystalline lens; anterior chamber depth; vitreous chamber depth; retinal, choroidal and scleral thicknesses and demographic factors including age, ethnicity, race and gender. The review includes studies demonstrating the impact of axial length changes on the eye’s refractive status.

Results: Axial length is associated with anthropometric, ocular biometric and demographic metrics, suggesting a complex interplay of factors. Increased axial length is linked to increased anterior chamber depth, vitreous chamber depth, body height and myopia. Axial length is negatively associated with spherical equivalent refractive error, lens and choroidal thickness.

Conclusion: The interplay of age, gender, ethnicity and environmental factors complicates the generalisation of axial length measurements. The strength and direction of these associations vary across studies, indicating complex relationships between factors. Tailored axial length norms based on individual characteristics are necessary, highlighting the need for population-based studies to minimise the generalisation of ocular biometry in clinical practice.

Contribution: This review emphasises the need for population-specific norms in ocular biometry, considering factors like age, gender, ethnicity and stature to enhance the precision and applicability of axial length measurements in clinical practice. The scarcity of instruments to measure axial length in primary and community eye care settings may hinder myopia control treatments.

Keywords: myopia; axial length; emmetropisation; refractive error; environmental factors; ocular biometry.

Introduction

Axial length (AL) is a fundamental ocular parameter that serves as a surrogate predictor of myopia and other ocular abnormalities.^1,2,3 It is measured along the optical axis from the corneal epithelium to the retinal pigmented epithelium and is essential in various ophthalmic procedures, including, but not limited to, cataract surgery and biometric assessments.^4,5,6 Additionally, there is a correlation between AL and retinal shape irregularity measured with optical coherence tomography (OCT), indicating a relationship between eye shape and AL.⁷ The relationship between pupillary responses and AL in children varies depending on the season.⁸ In summer, children with shorter ALs showed faster changes in pupillary responses to blue light stimuli, likely because of light-induced retinal dopamine release.⁸ While numerous investigations have explored the relationship between AL and ocular conditions, the broader interconnections between AL and anthropometric, demographic and ocular biometric metrics need further investigation.^4,5,6,7,8,9

Axial length variations influence refractive errors, ocular health outcomes and surgical complications.^1,2,3,5,10 Existing literature has examined specific aspects of AL, such as, but not limited to, its correlation with corneal spherical aberration, OCT retinal shape, pupillary responses, amblyopia risk factors and retinal ganglion cell layer thickness.^{6,7,8,9,11,12} However, a comprehensive evaluation of AL as a dependent biomarker within a broader physiological and demographic context is modestly addressed in the literature.

This study aims to systematically review the current knowledge surrounding AL and its associations with anthropometric, demographic and ocular biometric metrics. By reviewing existing findings, we aim to refine the understanding of AL as a key variable in myopia studies, providing implications that inform the exploration of novel treatment approaches.

Methods

This systematic review was conducted per the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 guidelines, ensuring a comprehensive and transparent approach to identifying, selecting and synthesising relevant studies. The inclusion criteria included studies investigating the association between AL and ocular and systemic variables. The population of interest was individuals aged 0 to 44 years to ensure relevance to myopia progression while minimising the influence of age-related ocular changes. Studies published between January 2014 and May 2024 were considered. Any studies outside this criterion were excluded. A comprehensive literature search was conducted from 28 May 2024 to 02 June 2024 using EBSCOhost databases. The following databases were searched: MEDLINE, PubMed, Academic Search Ultimate, Africa-Wide Information, Allied and Complementary Medicine Database (AMED) and E-Journals. The search strategy utilised the Boolean or Phrase Mode in the advanced search settings. The following keywords were used:

• Axial length AND
• (retinal thickness OR choroidal thickness OR scleral thickness OR age OR ethnicity OR race OR vitreous chamber depth (VCD) OR lens thickness (LT) OR crystalline lens thickness OR anterior chamber OR anterior chamber depth (ACD) OR central corneal thickness (CCT) OR corneal thickness OR corneal radius (CR) OR corneal curvature OR corneal radii of curvature OR AL/CR ratio OR spherical equivalent refraction (SER) OR refractive error OR refraction OR body mass index OR body weight OR weight OR body height OR height) AND
• (association OR relationship OR correlation OR link).

One researcher selected the studies based on titles and abstracts for relevance, followed by a full-text review of potentially eligible studies. Two other authors reviewed the selections for verification. The inclusion and exclusion criteria were applied to ensure the studies met the review’s objectives. The extracted data items included detailed study characteristics such as the study design, population demographics, intervention details, primary and secondary outcomes, study settings, sample sizes and study durations. Participant demographics, AL measurements and associated ocular and systemic parameters were also extracted. A narrative synthesis was conducted to summarise the findings. A meta-analysis was not performed because of substantial heterogeneity and a lack of sufficient comparable data among the included studies.

Results

The initial search yielded a total of 3695 studies (see Figure 1) after removing duplicates. A total of 2466 studies were excluded for the following reasons: (1) 92 studies were not written in English and (2) 2374 studies had a study population aged 45 years or older. This left 1229 studies for title and abstract screening. Of these, 1173 studies were excluded based on the following criteria: (1) not relevant to the study, (2) primarily focused on management strategies (such as orthokeratology, atropine and spectacle lenses) and (3) from similar studies. Studies were considered similar if they targeted similar populations, assessed AL as a primary outcome and utilised comparable methodologies; preference was given to the ones published in the last five years to ensure relevance to current and technological practices. Finally, 56 studies met the inclusion criteria and were included in the review.

FIGURE 1: Flowchart of study selection for review.

The included studies investigated the association between AL and various ocular and systemic parameters in individuals aged 0 to 44 years. A narrative synthesis was conducted to summarise the findings. The results indicate that AL is significantly associated with several ocular and systemic parameters, with variations observed across different age groups and ethnicities. However, the findings might not be generalisable for individuals outside the 0 to 44 years old age range.

Discussion: Review findings

Axial length development

Animal models have shown that ocular growth is visually guided.¹³ This theory also applies to the growth of the human eye, which is controlled by an active visual feedback and feedforward mechanism in response to the hypermetropic refractive error present in newborn eyes.¹³ Jackson and Moosajee¹⁴ explored the genetic factors influencing ocular AL growth, highlighting the molecular pathways involved in conditions ranging from microphthalmia to high myopia in childhood. The human eye is, on average, 17 mm long at birth, rapidly growing to about 20 mm by the first year of life.¹⁵ A longitudinal study reported the following findings on AL (mean value ± standard deviation [s.d.]): 19.19 ± 0.69 mm at 3 months, 20.29 ± 0.64 mm at 9 months, 20.71 ± 0.70 mm at 18 months, 21.42 ± 0.68 mm at 3 years, 21.96 ± 0.70 mm at four and a half years and 22.39 ± 0.71 mm at 6.5 years.¹⁶ Empirical evidence has shown a significant association between the rapid decrease in hypermetropia and changes in AL during the early stages of ocular development.^16,17,18 Taken together, these studies suggest that the AL is a crucial factor for emmetropisation in human eyes, and any deviations in this process can lead to ametropias such as hypermetropia and myopia.

Bach et al.¹⁹ conducted an extensive study on the development of AL in children, revealing a significant increase within the first 10 months of life, with negligible growth observed after 3 years. This is consistent with existing research indicating a slowdown in axial eye growth during pre-teen and teen years.^1,2,15 These findings are crucial for clinicians and researchers in understanding complex visual pathophysiological processes in children, particularly in conditions where AL plays an essential role, such as myopia, amblyopia risk assessment and refractive development. This underscores the need for further research on early childhood AL changes to refine diagnostic and management approaches for these specific paediatric ophthalmic conditions. However, the study’s methodology, which relied on retrospectively collected data and focused on children examined under anaesthesia because of a positive family history of retinoblastoma or other inherited ocular diseases, may limit the applicability of the findings.¹⁹ Despite these limitations, the study enhances the existing knowledge of the paediatric eye, potentially influencing future research and treatment strategies. Given that the referenced study¹⁹ primarily examined children in America and with a genetic predisposition to ocular diseases, it emphasises the need for further research in diverse populations to provide a more comprehensive understanding of AL development in children and its implications for paediatric ophthalmic diseases, including myopia.

Truckenbrod et al.² noted that children in the 2nd percentile, that is, children with the shortest AL relative to their age group, of AL reached their final length around 13 years of age. In contrast, children in the 50th percentile of AL continued to grow in AL, albeit at a slower rate (0.05 mm/year), until they are 18 years old (end of the observation period). Earlier, Tideman et al.¹ reported in their study that beyond the age threshold of 15 years, changes in AL were confined to the top 50% of the sample population, with the most significant changes occurring within and exceeding the 95th percentile. On the other hand, Tideman and colleagues might have underestimated this trend because of a higher myopia risk in younger cohorts that were not fully accounted for.¹ However, these studies indicate that the growth of AL and, consequently, the risk of myopia can continue into late adolescence beyond the age of 13 years.

The development of AL from infancy to adulthood shows a clear distinction between genders. Boys have higher AL measurements than girls.^1,2 However, the gap in the difference narrows as they reach adulthood. The study by Truckenbrod et al.² provides valuable insights into the growth of AL in children and adolescents, which is a critical factor in the development of emmetropia and ametropias. In their study, Truckenbrod and colleagues observed 3-year-old boys having 0.91 mm longer eyes than girls of the same age; however, the difference decreased to 0.24 mm at 18 years of age.

Anthropometric factors

Axial length and body height

Nilagiri et al.,²⁰ using longitudinal data from the Raine Study Gen2, found a statistically significant association between height and AL, indicating that for every 10 mm increase in (95% confidence interval [CI]: 0.006, 0.019; P < 0.001). height, there was an average increase of 0.012 mm in AL (95% confidence interval [CI]: 0.006, 0.019; P < 0.001). This finding highlights the potential impact of a person’s height on the eye’s dimensions. Although some discrepancies exist,²¹ these results are consistent with those of other studies,^22,23 further emphasising the importance of this relationship in the study of eye growth and height development. Recently, Kuoliene et al.²⁴ reported a significant (Pearson’s correlation coefficient [r] = 0.39, P < 0.001) but weak positive correlation between AL and height in myopic adolescents.

A comparative summary of the study findings is provided in Table 1. However, these studies^20,22,23,24 collectively suffer from limitations that prevent causal inference for this relationship, and the observed correlation may be a result of simultaneous effects from a shared underlying factor, such as genetics. Additionally, the link between AL and body height might be influenced by factors not accounted for in the studies, such as environmental conditions or lifestyle habits.

Axial length and body mass index

The relationship between body mass index (BMI) – a proxy for body fat and overall size – and AL remains inconsistent across studies (Table 1). Terasaki et al.²¹ reported a weak positive correlation (r = 0.23, P = 0.011) in a group of Grade 3 Japanese children (ages 8 to 9 years) with a mean BMI of 16.3 ± 1.9. Roy et al.²⁵ analysed the correlation between AL and BMI across ages of 8 to 70 years, in three groups and by eye. They found correlations of 0.42 and 0.47 in emmetropes, 0.26 and 0.24 in myopes and –0.61 and –0.57 in hyperopes for right and left eyes, respectively. Significance was not provided. The mean BMIs for refractive states, including high to mild myopia, emmetropia and low hyperopia, were all normal (ranging from 20.59 to 22.39); however, the high hyperope group children were overweight (26.5). Aprioku and Ejimadu²⁶ found no significant correlation (r = −0.035, P = 0.467) in a Nigerian adult sample (older than 18 years). However, we note that 63% of their sample was overweight (37%) or obese (27%), which may have contributed to the very weak correlation.

Terasaki et al.²¹ proposed a biochemical mechanism whereby diets high in saturated fats might increase circulating insulin-like growth factor, potentially lengthening AL through scleral remodelling. Nevertheless, this hypothesis remains speculative and may not fully account for multifactorial influences shaping the associations between AL and BMI.

Ocular biometry factors

Axial length and refractive error

The relationship between SER and AL has been investigated in several studies, yielding varying results and summarised in Table 2. For example, Dayi et al.²⁷ presented a strong negative correlation (r = −0.826, P < 0.001) while Tang et al.²⁸ reported a moderate negative correlation (r = −0.52, P < 0.001). He et al.³ outlined a strong negative correlation between SER and AL (r = −0.84) and a very strong correlation between SER and AL to CR ratio (AL/CR) (r = −0.911). Similarly, Truckenbrod et al.² found a negative relationship for both boys (r = −0.57, P < 0.001) and girls (r² = 0.37, r = −0.61, P < 0.001), suggesting that as AL increases, there is a corresponding shift towards increased myopia or reduced hyperopia.

Vera-Diaz et al.²⁹ observed a weak, yet statistically significant, negative correlation between SER and AL (r = −0.25, P = 0.01). However, after adjusting for age in children, the correlation between SER and AL was not significant. When the data were stratified by gender, the correlation between SER and AL was significant only in girls (r = −0.29, P = 0.02) but not in boys (r = −0.04, P = 0.82). The weak correlation between SER and AL could be the result of a small sample size (n = 97) in this study and the fact that all participants were emmetropes, which may not represent the general population. Children with longer eyes are more likely to develop myopia although this relationship was weak in German children between 5 and 8 years of age.² These findings collectively suggest a complex relationship between SER and ocular biometrics, potentially influenced by factors such as CR, age, ethnicity and gender.

Axial length and central corneal thickness

The association between AL and CCT may be finite in the literature (Table 2). However, Krishnan et al.³⁰ reported a weak positive correlation (r = 0.211, P = 0.008) in an Indian sample, while Bikbov et al.³¹ (P = 0.18) and Bikbov et al.³² (P = 0.12) reported no significant correlation between these variables in a sample of Russian adults (aged > 40 years) and children (aged 6–18 years), respectively. Furthermore, researchers have indirectly explored this connection within the context of SER.^33,34 A very weak negative correlation between SER and CCT (r = −0.13, P < 0.01)³³ has been observed, suggesting that an increase in myopia (or an extension of AL indicative of myopia) is associated with an increase in CCT, indicative of a thicker cornea.

The reported relationships between AL and CCT, as well as the negative correlation between SER and CCT, may be influenced by various factors such as age,^35,36 gender,^1,2 ethnicity^1,2,11 or biochemistry of the corneal matter.³² These factors could potentially confound the relationships observed. For instance, a positive correlation was observed in a sample from the Indian population.³⁰ However, no significant correlations were identified in either children³² or adults³¹ from a Russian sample. Different methods used to measure CCT may partially explain the variations; Krishnan et al.³⁰ used the ultrasonic pachymeter, while Bikbov et al.^31,32 measured by Scheimflug imaging. Future investigations and cross-validation studies should include diverse populations spanning different age groups and ethnicities to enhance generalisability.

Axial length and corneal radius

Axial length and CR are parameters clinically significantly contributing to the eye’s refractive state. Vera-Diaz et al.²⁹ observed a weak, yet statistically significant, negative correlation between SER and AL/CR (r = −0.36, P < 0.01). The AL/CR ratio has been identified as a predictor of myopia risk, as evidenced by several studies (Table 2).^3,29,37,38 A strong negative correlation has been observed between SER and AL and between SER and the AL/CR, as demonstrated by linear regression analysis.^27,28,37

The area under the receiver operating characteristic (ROC) curve for the AL/CR ratio (0.896) surpasses that of AL alone (0.830), suggesting that the AL/CR ratio possesses a more substantial predictive power for high myopia in adults.³⁷ While AL is typically employed as a proxy for myopia, the AL/CR ratio could serve as an alternative measure for evaluating the risk and degree of myopia, even in situations where cycloplegia is not available, particularly in children.^3,38 Concentrating on the AL/CR ratio could pave the way for the creation of novel preventative approaches for managing myopia. Nonetheless, despite the AL/CR ratio’s predictive power, it may not be a comprehensive measure for evaluating myopia risk because of potential limitations in specific demographic or clinical contexts.

Axial length and crystalline lens

The crystalline lens goes through substantial optical and biological changes during a person’s lifetime, such as changes in LT, protein clustering and overall flexibility.³⁹ This unique ability of the lens draws attention to its relationship with the AL and refractive status long after the corneal development stabilises. The lens power will be left to keep the balance, which primarily determines the refractive status.

The association between AL and lens power is not straightforward and appears to be different across various refractive statuses.³⁹ In non-myopic eyes, the lens power shows a stronger negative association with AL (regression coefficient [β] = −1.288). This means that as the eye’s AL increases during development and in late adolescence, there is a compensatory reduction in lens power to maintain a state of emmetropia, where the eye can focus light accurately on the retina. However, in myopes, this association is weaker (β = −0.390). This could be because prior to the onset of myopia, with exceptions,⁴⁰ there is an increased rate of lens power loss, which seems to halt after myopia has developed.³⁹ Mechanisms behind this decoupling remain speculative but might be influenced by restricted equatorial growth resulting from myopes’ abnormal growth of the ciliary muscles.³⁹ Myopia develops when the axial growth surpasses the compensatory decrease of the lens power.

Studies on the relationship between AL and LT are consistent, giving a negative correlation (Table 2).^32,41,42 A significant moderate negative correlation was found in a South African study⁴¹ (r = −0.52, P < 0.001), and a weak correlation (r = −0.36, P < 0.001) was found in a Chinese study,⁴² suggesting that an increase in AL is associated with a decrease in LT. While the correlation patterns align between the two studies, variations in effect size suggest the influence of additional confounding variables. Consequently, caution is warranted when extrapolating these findings to diverse populations.

Axial length and anterior chamber depth

The measurements of ACD are critical in the diagnosis of various eye conditions and in determining the power of the intraocular lens before cataract and refractive surgeries. A recent study has unveiled a complex interplay between these two parameters.⁴³ In eyes with an AL of 22 mm or less, an inverse correlation (r = −0.458, P = 0.011) exists between AL and ACD, indicating that an increase in AL corresponds to a decrease in ACD and vice versa. However, in eyes with an AL exceeding 22 mm, no significant correlation has been observed between AL and ACD,⁴³ suggesting the potential influence of other factors on ACD in these eyes. On the other hand, earlier investigations into this relationship by Mashige and Oduntan⁴¹ have revealed contrary evidence suggesting that longer eyes have a deeper ACD (r = 0.66, P < 0.001). Although both studies^41,43 focused on participants of African descent in South Africa and Egypt (Table 2), their divergent findings raise concerns about the generalisability of each study’s results to the broader African population. Consequently, additional population-based research is justified. Furthermore, it is essential to recognise that additional variables beyond the study populations may also impact the observed relationship.

Axial length and vitreous chamber depth

Roy et al.²⁵ found a strong correlation (r = 0.67) and Paritala et al.⁴⁴ found a very strong correlation (r = 0.983, P < 0.001) between AL and VCD, indicating a robust association between these ocular measurements. Vera-Diaz et al.²⁹ observed a weak, yet statistically significant, negative correlation between AL and VCD (r = −0.33, P < 0.01). A summary of the correlations is provided in Table 2.

Paritala et al.⁴⁴ proposed a new concept involving the relationship of the VCD/AL ratio with ocular biometrics.⁴⁴ This ratio demonstrated a positive correlation with both AL (r = 0.811) and VCD (r = 0.827) and was observed to be higher in eyes with severe myopia.^44,45 While the findings from this study are specific to high myopic participants, they imply that the VCD/AL ratio may be a valuable indicator of ocular growth. For example, in myopic conditions characterised by an increased AL, the VCD/AL ratio could provide useful insights into the severity of the condition and could potentially guide therapeutic strategies. However, the relationship of this ratio with refractive status has yet to be investigated.^44,45

Axial length and choroidal, scleral and retinal thickness

Axial length has been studied in relation to various ocular parameters such as retinal thickness, choroidal thickness and scleral thickness (Table 2).^9,46,47 The association between these factors is crucial in understanding conditions like myopia and other ocular pathologies. Studies have shown that as AL increases, there is a decrease in choroidal and scleral thickness, particularly in the posterior hemisphere of the eye.^46,47,48 This thinning of the choroid and sclera is associated with myopia and is considered to be linked to scleral ischaemia and hypoxia, leading to scleral thinning and weakening, which in turn accelerates the onset of myopia.⁴⁹ Additionally, the thinning of the sclera has been observed to occur with increasing degrees of myopia, particularly in the peripheral regions.⁵⁰ Additional factors beyond AL may contribute to the observed reduction in choroidal and scleral thickness. This thinning alone is not the sole predictor of myopia onset.

In the fundus midperiphery, the retina and retinal pigment epithelium density decrease with longer AL. However, in the macular region, retinal thickness, retinal pigment epithelium cell density and choriocapillaris thickness are not significantly related to AL.⁵¹ This suggests that the impact of AL on retinal thickness may vary across different ocular regions.

Demographic factors

Axial length and age

The association between age and AL, a crucial biometric parameter in the diagnosis and management of myopia, has been a subject of considerable scientific interest. Vera-Diaz et al.²⁹ posit age as a primary determinant of AL, suggesting a natural elongation of the eye with increasing age during development, potentially leading to myopia when other optical components start to fail compensation.^16,52 However, the relationship between age and AL is not unidirectional. Additional research indicates that factors such as gender and CR of curvature are associated with AL.^3,53 Furthermore, genetic factors may also play a pivotal role.

The progression of AL growth is negatively correlated with age in children, exhibiting a decline in growth rates as age advances.^1,2,3,54 Therefore, an indiscriminate aggregation of data from eyes compensated with AL- or myopia-control lenses or uncompensated eyes, irrespective of the age factor, and subsequent evaluation of their AL growth over a specific duration could potentially yield conclusions that are not entirely accurate or representative of the true scenario. Hence, it is imperative to incorporate age as a crucial variable in such evaluations to avoid erroneous interpretations and conclusions.

The AL plays a pivotal role in the preclinical appraisal of intraocular lenses (IOLs), as it directly influences refractive calculations and postoperative visual outcomes. The primary obstacle for IOL calculations in paediatric patients is the prospective alteration in refractive outcome attributable to growth-induced modifications in AL, anterior segment geometry and corneal power. Currently, later-generation algorithms such as the SRK-T, Hoffer-Q and Holladay II, predicated on adult normative data, are employed for IOL power computation.¹⁹ Nevertheless, these could precipitate refractive inaccuracies if utilised for paediatric patients, owing to the presumptions inherent in these formulas.¹⁹ The provision of a fundamental dataset on ocular biometry, particularly AL, for each age bracket in paediatric populations until adulthood could aid cataract surgeons during paediatric preoperative assessments and in the determination of the most suitable IOL, taking into account the alterations in AL.¹⁹ It has been predominantly observed that patients who undergo cataract surgery at an early age typically exhibit long-term refractive errors because of a myopic shift engendered by an augmentation in AL.⁵⁵

Axial length and ethnicity

Ocular biometric parameters appear to be influenced by ethnic and geographic variations. Tideman et al.¹ reported matching CR measurements between 6-year-old Dutch children and 3-year-old Asian children.⁵⁶ Moreover, investigations into ocular biometric parameters, including AL (Table 3), have shown variations with age and ethnicity, particularly in hyperopic children.¹¹

Table 3 presents various AL [mean ± s.d.] from China,³ UK,¹ the Netherlands¹ and Russia.³² Among 9-year-olds, female children from China had shorter AL than their male counterparts, despite having similar SER. A similar pattern was observed in Dutch children, with girls again having shorter AL than boys. The mean AL of 9-year-olds from Russia, for both genders, was comparable to that of Dutch children and Chinese girls. However, the mean ALs from both Russia and the Netherlands were shorter than those of Chinese boys, despite having similar SER. These findings emphasise the influence of ethnicity, cohort effect, age, gender and SER on AL, thereby cautioning against the over-generalisation of these measurements across heterogeneous populations.

Additional studies are necessary to further investigate the implications of AL studies.^1,2,3,32 It is also crucial to acknowledge that while the studies offer insightful perspectives, they do not consider all potential influencing factors, such as genetics, lifestyle and environmental factors, which could affect the development of myopia.

There is a need for extensive ocular biometry studies in various populations, including those of African descent. Such research efforts could enhance our comprehension of ocular biometry variations and facilitate the development of more population-based tailored eye care strategies.

Implications and recommendations

This systematic review demonstrates varying degrees of associations between AL and the studied factors. Height, gender, SER, LT and VCD present strong consistent correlations and associations with AL, whereas CCT and choroidal thickness show only weak correlations. Associations with age, ethnicity and ACD are inconsistent among the different studies. Following this, individualised norms for AL that consider demographic and anthropometric factors such as age, gender, ethnicity and body stature are recommended. This tailored approach will enhance the accuracy of refractive error predictions and improve the management of myopia in different populations.

Incorporating AL measurements into primary and community eye care settings is crucial for the early detection and treatment of myopia and other refractive errors. This integration can help bridge the gap in myopia control treatment. Furthermore, there is a need for large-scale, population-based longitudinal studies to establish normative data for AL and other ocular biometric parameters across diverse demographic groups. In addition, conducting longitudinal studies will help clarify the causal relationships between AL and various anthropometric, biometric and environmental factors over time.

The included studies exhibited heterogeneity across multiple variables, including study design, population characteristics (age ranges and geographic settings), environmental and lifestyle exposures and measurement techniques, which likely introduced variability into the findings and made a quantitative meta-analysis unjustified. Additionally, the reliance on published literature means there is a potential for publication bias, where studies with significant results are more likely to be published than those with null findings. Future research should address these limitations by considering a broader range of influencing factors.

Conclusion

The multifaceted associations of AL with various factors were explored. The findings indicate that an increase in AL is predominantly linked to increased VCD and a shift towards more negative refractive error values, leading to myopia. The growth and state of AL are influenced by a complex interplay of age, gender, ethnicity, body stature and biometrics, challenging the generalisation of AL measurements and ocular biometry. Given these findings, it is evident that AL baselines or norms should be tailored to account for different predisposing factors. This approach underscores the necessity for population-based studies on ocular biometry that incorporate all relevant factors to provide more accurate and personalised measurements, thereby minimising the generalisation of ocular biometry in clinical practice. Notably, variations across studies, stemming from differences in age distribution, sex, ethnicity, location, measurement instruments and statistical approaches, contribute to inconsistencies in reported associations.

Acknowledgements

This article is partially based on the author’s dissertation (in progress) entitled ‘Axial length growth and development in school children of African descent, South Africa’ towards the degree Master’s in Health Sciences (Optometry) in the Department of Optometry, University of Johannesburg, South Africa with supervisors Tanya Evans and Solani D. Mathebula.

The authors declare that they have no financial or personal relationships that may have inappropriately influenced them in writing this article. S.D.M. serves as an editorial board member of African Vision and Eye Health. The peer review process for this submission was handled independently, and S.D.M. had no involvement in the editorial decision-making process for this manuscript. S.D.M. has no other competing interests to declare.

This review was written as part of a higher degree by A.T.C., under the guidance of T.E. and S.D.M.; A.T.C. conceived the idea and was responsible for data collection and analysis. The theory and concept were developed by A.T.C.; T.E. led the data examination and verification of the review analysis with S.D.M. All authors participated in the final review of the article and discussed the findings.

An application for full ethical approval was made to the University of Johannesburg, Faculty of Health Sciences Research Ethics Committee and ethics consent was received on 25 July 2024. The ethics approval number is REC-2198-2023.

This research received no specific grant from any funding agency in the public, commercial or not-for-profit sectors.

The data supporting the findings of this systematic review are available from the corresponding author, A.T.C., upon reasonable request. This includes the template data collection forms, data extracted from included studies and any other materials used in the review. No meta-analysis was conducted, so there are no analytic codes or datasets related to meta-analyses.

The views and opinions expressed in this article are those of the authors and are the product of professional research. They do not necessarily reflect the official policy or position of any affiliated institution, funder, agency or that of the publisher. The authors are responsible for this article’s results, findings and content.

References