Axial length (AXL) has been well established as an important, critical measurement in studies looking at the progression and control of myopia. It also has clear measures in assessing the risk profile for myopic degeneration; when the axial length reaches and lengthens beyond 26mm the risk of posterior pole complications and potential vision impairment is much higher.1,2 For a look at the clinical use of single versus repeated measures of AXL, and the approach of the International Myopia Institute, read our blog Axial Length Measurement – A Clinical Necessity? Measuring AXL in clinic requires specialty equipment not found in most primary eye care practices. So how crucial is it? Is it required for practicing myopia control?
Question 1: How well does axial length relate to refraction?
There is the simple formula you may have held as truth, that 1mm=3.00D. Easy? Not quite. A variety of papers have shown that this relationship is complex. In the three year MiSight study,1 the correlation became stronger as the children got older – meaning the relationship was more variable in younger ages. Across the three year study, they found that a 0.1mm change in axial length corresponded to a 0.24D change in myopia, in both treatment and control groups, giving a ratio of 2.40D/mm. In the newly published BLINK study,2 this ratio was 1.44D/mm in the high add (+2.50) multifocal contact lens wearing group, 1.55D/mm in the medium add (+1.50) group and 1.63D/mm in the single vision CL group. While we don’t know if those ratios are statistically different, it highlights how much we still need to learn about this ratio. A big part of the complexity is due to the variability in measurement of both refraction and axial length. Axial length measurement is considered the gold standard in the research context when investigating myopia control strategies, and is becoming increasingly important in the clinical setting where reducing axial elongation to reduce lifelong eye health risk is the key goal.3
However even in a research setting, the International Myopia Institute agrees that refractive error should be used in conjunction with axial length measurement to evaluate success of treatments.4 They noted that subjective refraction can be more variable, however axial length is not the only ocular component to change throughout childhood - the cornea and crystalline lens are also key anatomical sites of refractive change and contribution. Refraction provides a summary measurement of all ocular structures, and is routinely and universally measured for all myopes. Despite this, in the case of orthokeratology when the refraction is intentionally altered, the axial length may be the only parameter that can be used as a gauge of myopia control outcomes. Also, since the refraction-to-axial-length ratio seems to come unstuck with low dose atropine – where the refractive outcomes outshine the minimal control of axial length5 – knowledge of axial length progression provides the primary measure of myopia control success.
Question 2: Can we predict future myopia from AXL changes?
Yes, you can, but refraction may be more accurate and simpler – more on that shortly. On axial length, the Comparison of Ocular Component Growth Curves Among Refractive Error Groups in Children paper6 measured the ocular components and axial lengths of 737 children aged from 6 and 14, who were assessed at least three times each. With this data the authors evaluated relationships between the parameters and growth patterns where the end results were either hyperopia, emmetropia or myopia.
Children who became myopic had a similar axial length at age 6 to children who stayed emmetropic, although it quickly showed escalation over time (see the graph below). This highlights the importance of early intervention. Patients who were hyperopic and emmetropic had similar models of growth. The authors developed formulas for predicting changes with age for the children who stayed emmetropic, and compared this to the other refractive error groups for significant differences.
Growth curve for axial length, using the best model derived from emmetropic data and applying it to the other three refractive groups. Comparison of Ocular Component Growth Curves among Refractive Error Groups in Children, Jones et Al. 2005 - Figure 8
Your patient is 5 years and 11 months of age (5.91 years old - the age must include months too. This child is +0.50D and has an axial length of 22.35mm. They are emmetropic and under ten, so their axial length should equal = 20.189+(1.258*Ln(5.91)) = 22.42mm. This data suggests that the axial length is appropriate for this child's age.
When measuring the axial length as a predictive tool, you can theoretically use this equation as a prognostic opportunity that may help guide the initiation and proactivity of your myopia treatment. Interestingly, Rozema et al7 found that in Singaporean children, myopia onset seemed to occur at approximately the same mean axial length, although with large variability: 24.08±0.67mm in boys and 23.69±0.69mm in girls. Future myopes had a higher axial growth rate in the years before onset than children who stayed emmetropic.
Is prediction of onset by axial length gold standard? Karla Zadnik and co-authors of the 11-year-long Collaborative Longitudinal Evaluation of Ethnicity and Refractive Error (CLEERE) Study evaluated the CLEERE data and found that simple refractive error was the single best predictor of future myopia.8 Children aged six with less than +0.75D were at greatest risk of future myopia, even after adjustment for all other factors.
Tideman and co-authors in 20189 sought to develop a percentile growth chart assessing the axial length, refractive error and other growth parameters of 12,000 children. They found that axial length isn’t the perfect predictor of future myopia, with an axial length increase predicting future myopia only 50% of the time. The chart did provide a degree of prediction for high myopes, but was less effective in predicting low myopia or assessing efficacy of treatment. The authors suggested that growth charts could be a tool for predicting high myopia and identifying children at risk of developing myopia, but not necessarily for tracking myopic growth or comparing efficacy of myopia treatments without addition of many more datasets to improve accuracy.
Question 3: So if I’m going to measure axial length, how?
There is a question of the variability of axial length measurements, both between visits and between different instruments. Most present day studies employ non-contact, interferometry measurement instruments rather than the older generation A-scan applanation ultrasound technique. The IMI Clinical Trials and Instrumentation report4 explains that ultrasonography is limited in resolution to about 0.30D, whereas interferometry measurements have resolution of around 0.03D, making this technique an order of magnitude better as a measure of myopia progression. Compare this also to refraction – cycloplegic autorefraction has a repeatability of ±0.21D.10
Studies which evaluate axial length take multiple measurements and use the average in reporting. How many measurements are needed? Some studies measured 10 times,1others measured 5 times.2,8 The IMI Clinical Trials paper does not make a specific recommendation on this.4
If you are looking at a particular instrument, search the literature and/or ask the company if data has been published on repeatability of that instrument.
Question 4: Is axial length measurement crucial in orthokeratology and atropine treatment?
When the refractive error is intentionally altered to plano or nearabouts, the efficacy of orthokeratology for myopia control can be monitored via axial length measurement, without having to “wash out” the patient from the OrthoK to measure their full refraction. “Wash out” also requires ensuring the cornea has returned to topographical baseline to appropriately judge refraction change, presenting inconvenience to the patient and additional chair time for the practitioner. Clinicians may measure the refraction with the lens-on-eye, but since OK lenses can warp with time this could also prove inaccurate. Research has suggested that OrthoK's induced corneal changes will have minimal influence on repeated axial length measures, when comparing pre- and post-OK.11
There is a similar complication with atropine treatment. When it comes to axial length, the one year LAMP study5 found a 12%, 29% and 51% control effect with 0.01%, 0.025% and 0.05% concentrations respectively, but the refractive results were a much more impressive 27%, 42% and 66% respectively. By comparison, the three year MiSight study showed a strong correlation between axial and refractive control, with similar percentages of just over 50%.1 While comparing percentages across studies is problematic (and that’s a discussion for another blog or two or three!), the evident mismatch between atropine refractive and axial control additionally points to the value of axial length measurement.
Question 5: Will axial length dictate my treatment plan?
It could dictate the level of proactivity needed. A child who is -3.00D and 25mm may need more proactive treatment and closer monitoring than the same -3.00D who is 24mm. We’re still learning if axial length is a better predictor of ocular disease risk than refraction – there are more prevalence studies available which have measured myopia by dioptres rather than axial length.3 One large-scale, multicentre analysis has indicated that axial length is more closely related to myopic complications than refractive error.12
Axial length can also determine other necessary eye health monitoring. For example, it is wise to schedule an annual fundus examination through dilated pupils for a patient if their axial length is over 26mm, regardless of their refraction, due to the increased risk of pathology. Here the data is striking – among people aged 75 years or older, cumulative risk of vision impairment or blindness from any cause increases from 3.8% in eyes less than 26mm to 25% in eyes longer than 26mm; and to more than 90% in eyes longer than 30mm.12
Question 6: Can I practice myopia control without measuring axial length?
There is no doubt that measurement of axial length will provide the more accurate indicator of myopia progression and control – as described above, optical biometry methods are likely to be 5-10 times more accurate than refraction.4,10 This is why axial length is a required measurement for treatment comparisons in myopia control research, especially to enable evaluation across different studies and patient groups. At this stage, however, it is not a requirement to safely and effectively practice myopia management. In the International Myopia Institute Clinical Management Guidelines, axial length measurement was included as a ‘standard procedure’ but with the caveat that there is currently no established criteria for normal or accelerated axial elongation in a given individual.13
As accessibility to axial length measurement technology becomes easier, and more evidence is developed to support normative and typical myopic patterns for axial length growth across a variety of ethnicities and populations, axial length measurement will become increasingly important in myopia prediction and management. Keep an eye on the research and available equipment, because axial length measurement may become standard of care in our comprehensive myopia management of the future.
Axial length measurement - a clinical necessity?
Axial length measurement in myopia management – how often and how much change is normal? A clinical case study.
When you have low myopia and high axial length - A clinical case study.
How reliable are growth charts for predicting myopia onset and progression?
Cassandra Haines is a clinical optometrist, researcher and writer with a background in policy and advocacy from Adelaide, Australia. She has a keen interest in children's vision and myopia control.
This educational content is brought to you thanks to unrestricted educational grant from
- Chamberlain P, Peixoto-de-Matos SC, Logan NS, Ngo C, Jones D, Young G. A 3-year randomized clinical trial of MiSight lenses for myopia control. Optom Vis Sci. 2019;96:556-567. (link)
- Walline JJ, Walker MK, Mutti DO, et al. Effect of High Add Power, Medium Add Power, or Single-Vision Contact Lenses on Myopia Progression in Children: The BLINK Randomized Clinical Trial. JAMA. 2020;324(6):571–580. (link)
- Haarman AE, Enthoven CE, Tideman JW, Tedja MS, Verhoeven VJ, Klaver CC. The Complications of Myopia: A Review and Meta-Analysis. Invest Ophthalmol Vis Sci 2020;61:49. (link)
- Wolffsohn, JS et al. IMI – Clinical Myopia Control Trials and Instrumentation Report. Invest Ophthalmol Vis Sci 2019;60:M132-M160 (link)
- Yam JC, Jiang Y, Tang SM et al. Low-Concentration Atropine for Myopia Progression (LAMP) Study: a randomized, double-blinded, placebo-controlled trial of 0.05%, 0.025%, and 0.01% atropine eye drops in myopia control. Ophthalmol 2019;126:113-24. (link)
- Jones, LA et al. Comparison of Ocular Component Growth Curves among Refractive Error Groups in Children. Invest Ophthalmol Vis Sci 46, 2317-2327, doi:10.1167/iovs.04-0945 (2005). (link)
- Rozema J, Dankert S, Iribarren R, Lanca C, Saw SM; Axial Growth and Lens Power Loss at Myopia Onset in Singaporean Children. Invest Ophthalmol Vis Sci 2019;60:3091-3099. (link)
- Zadnik, K et al. Prediction of Juvenile-Onset Myopia. JAMA Ophthalmol133, 683-689, doi:10.1001/jamaophthalmol.2015.0471 (2015). (link)
- Tideman, JWL et al. Axial length growth and the risk of developing myopia in European children. Acta Ophthalmol 2018;96:301-309.(link)
- Moore KE, Berntsen DA. Central and peripheral autorefraction repeatability in normal eyes. Optom Vis Sci. 2014;91(9):1106-1112 (link)
- Cheung SW, Cho P. Validity of Axial Length Measurements for Monitoring Myopic Progression in Orthokeratology. Invest. Ophthalmol. Vis. Sci. 2013;54(3):1613-1615.
- Tideman JW, Snabel MC, Tedja MS, et al. Association of Axial Length With Risk of Uncorrectable Visual Impairment for Europeans With Myopia. JAMA Ophthalmol. 2016;134:1355-1363. (link)
- Gifford KL et al. IMI – Clinical Management Guidelines Report. Invest Ophthalmol Vis Sci 60, M184-M203 (2019). (link)