Effect of 0.025% atropine on ocular biometry changes during accommodation

Abstract

Purpose

Low concentration atropine is an effective treatment to slow myopia progression and axial elongation and also reduces accommodation. On‐axis ocular dimensions of the eye change during accommodation; hence, this study aimed to quantify the effect of 0.025% atropine eye drops on accommodation‐induced changes in ocular biometry.

Methods

Twenty‐eight myopic participants with a mean (SD) age of 17.0 (6.0) years (range: 8.0–25.5 years) and spherical equivalent refraction (SER) of −2.03 (1.05) D (range: −0.75 to −4.38 D) were enrolled. Baseline ocular biometry measurements of the left eye were captured using an optical biometer (Zeiss IOLMaster 700) for 0, 2, 4 and 6 D accommodation stimuli, presented via a Badal optometer. The accommodation response (AR) was determined using wavefront aberrometry (Imagine Eyes irx3) for the same accommodation stimuli and following cycloplegia using 1% tropicamide. Participants instilled 0.025% atropine eye drops nightly for 1 week in both eyes, and ocular biometry measurements were repeated on the day after the final atropine dose.

Results

Anterior chamber depth (ACD) and corrected vitreous chamber depth (cVCD) decreased, and crystalline lens thickness (LT), anterior segment length (ASL), crystalline lens centre position (LCP) and the AR increased significantly during accommodation (all p ≤ 0.009). Accommodation‐induced changes in ACD and LT were reduced following 0.025% atropine use (both p ≤ 0.01), with significant pre‐ and post‐atropine differences for the 4 and 6 D stimuli (all pairwise comparisons, p ≤ 0.004). On average, ACD, ASL and LCP increased, while cVCD, corrected axial length (cAL), and the AR decreased following 1 week of 0.025% atropine use (all p ≤ 0.002).

Conclusions

The AR and on‐axis ocular biometric changes during accommodation were reduced following 1 week of 0.025% atropine use. These findings may have implications for the association between near work and myopia, and atropine's mechanism of action in humans.

Key points.

• One week of nightly 0.025% atropine use reduced on‐axis accommodation‐induced structural changes associated with a reduced accommodation response and a higher lag of accommodation.

• A small posterior displacement of the crystalline lens may contribute to the hyperopic shift following atropine instillation.

• Axial length changes following atropine commencement should be considered in clinical trial design.

INTRODUCTION

Atropine is a non‐selective, anti‐muscarinic agent used to slow myopia progression and axial elongation. ¹ While high concentration atropine eye drops (>0.5%) have been prescribed, ² , ³ clinically significant myopia control can be achieved with much lower concentrations (<0.1%), with increased tolerability due to their substantially reduced side effect profile. ⁴ , ⁵ A concentration of 0.05% offers the best balance between myopia control efficacy and side effects based on randomised clinical trials in East Asian children. ⁶ However, 0.025% offers a clinically acceptable treatment effect and minimal side effects, which has led to its recommendation as the minimum concentration for myopia control. ⁷ This tends to be favoured in Caucasian populations.

Early studies of atropine for myopia control assumed that sustained accommodative force associated with excessive near work drove myopia development and progression. In chicks, atropine is a potent inhibitor of experimentally induced form deprivation myopia development. However, the intraocular muscles of the chick are striated rather than smooth muscle, and therefore, atropine does not produce cycloplegia. ⁸ Atropine has been found to bind to muscarinic and non‐muscarinic retinal receptors in chicks, which initiates a signal cascade involving the retinal pigment epithelium, choroid and sclera, and results in an increase in scleral structural integrity and a slowing of eye growth through scleral remodelling. ¹ , ⁹ Consequently, the mechanism of atropine in avian models is thought to be unrelated to accommodation. ⁸ However, the specific mechanism in humans remains unclear.

Accommodation is the mechanism by which the eye's refractive power changes to enable clear vision from far to near distances. This optical change arises predominantly due to changes in the crystalline lens, specifically the combination of increased thickness, ¹⁰ , ¹¹ , ¹² steepening of the anterior and posterior surfaces ¹³ , ¹⁴ , ¹⁵ and increased refractive index, ¹³ which also reduces the anterior and vitreous chamber depth. ¹⁰ , ¹² Recent studies have demonstrated that small, transient increases in axial length also occur during accommodation in adults ¹⁰ , ¹¹ , ¹⁶ and children, ¹² with some investigations reporting greater changes in myopes than in non‐myopes, ¹¹ , ¹⁷ which may be a potential mechanism underpinning the link between near work and myopia.

The reduction in the amplitude of accommodation due to atropine is well established. ⁴ , ⁵ However, the effect of low concentration atropine on accommodation‐induced ocular structural changes remains unclear. A small study of young myopic adults indicated that the anterior chamber depth shallowing during a 5 D accommodation task was reduced by ~10 μm 1 h after the instillation of one drop of 0.01% atropine. However, no accommodation‐induced changes in any other structures were reported. ¹⁸ No studies have explored the effect of other concentrations of atropine eye drops, other accommodation stimuli or a longer duration of atropine treatment. Therefore, this study aimed to emulate clinical dosing for myopia control and examined the on‐axis structural changes during 0, 2, 4 and 6 D of accommodation before and after 1 week of nightly 0.025% atropine eye drop dosing in myopic children, adolescents and young adults, and explore any age‐related differences in the atropine response.

METHODS

Ethics approval was granted by the Queensland University of Technology Human Research Ethics Committee, and the trial was registered on the Australian New Zealand Clinical Trials Registry (ACTRN1261000563864). All participants ≥18 years of age and parents or guardians of children and adolescents <18 years provided written informed consent, and the children and adolescents gave written informed assent.

Twenty‐eight participants (18 females and 10 males) with a mean age (standard deviation, SD) of 17.0 (6.0) years (range: 8.0–25.5 years) and in good general health were enrolled. The predominant ethnicity was Caucasian (N = 13; 46%), followed by South Asian (N = 7; 25%), East Asian (N = 4; 14%), Mixed East Asian/Caucasian (N = 2; 7%), Southeast Asian and African (both N = 1, 4%). Most participants had brown irides (N = 16, 57%), followed by blue (N = 9, 32%), hazel/green (N = 2, 7%) and brown/green (N = 1, 4%). Participants were stratified into two age groups, with those <18 years included in the children and adolescent group and those ≥18 years allocated to the young adult group. The proportion of participants of each ethnicity (χ ² = 4.24, p = 0.52) and iris colour (χ ² = 2.87, p = 0.41) was not significantly different between the age groups.

A screening examination was undertaken to confirm study eligibility, including non‐cycloplegic subjective refraction, distance logMAR visual acuity, near logMAR visual acuity at 25 and 50 cm, stereoacuity, push‐up amplitude of accommodation, cover test, ocular health examination, ocular motility function, pupillary responses and intraocular pressure. Enrolled participants had myopia (spherical equivalent refraction, SER ≤−0.50 D), ¹⁹ ≤1.00 D of astigmatism and anisometropia, monocular visual acuity of ≤0.10 logMAR in each eye and an interocular difference of ≤0.10 logMAR, an amplitude of accommodation >8 D, stereoacuity ≤200 arcsec, a van Herick ratio of >0.3:1 and intraocular pressure of ≤21 mmHg using rebound tonometry (IC100, icare‐world.com). None of the participants currently used, or had previously used, atropine eye drops, bifocal spectacles, orthokeratology or spectacle or contact lenses for myopia control, nor did they report a history of amblyopia, strabismus, ocular surgery, significant ocular infection or injury, central nervous system or chromosomal disorders or angle closure glaucoma. No participants reported a known allergy to anti‐cholinergic medications or benzalkonium chloride.

Following the screening assessment, participants watched a video for 10 min at a 6‐m distance through their non‐cycloplegic subjective refraction, to relax accommodation and reduce the potential impact of previous visual activities on biometry measurements. ²⁰ Measurements of the left eye were then captured using an optical biometer (IOLMaster 700, zeiss.com/meditec) (with the right eye patched) at four accommodation stimuli (0, 2, 4 and 6 D) using a Badal optometer, as described previously. ¹² , ¹⁷ For some participants (one adolescent and all young adults), a glass microscope slide was used instead of a longpass filter, due to a reduction in the power of the IOLMaster light source. Since the same optical element was used across all accommodation stimuli for each participant, this is unlikely to have impacted the results significantly.

Before measurements were captured, the Badal optometer was adjusted to present a 0 D accommodation stimulus (AS), and the Badal fixation target was aligned to the IOLMaster internal fixation target (red light). Accommodation stimuli were presented in a semi‐randomised order to minimise any systematic effect of ascending order stimulus presentation, with the first and second stimuli randomised to be either 0 or 4 D, while the 2 and 6 D stimuli were randomised to be either the third or fourth stimuli. Emoticon targets were used to engage accommodation and fixation for the children and were also used for the adolescents and young adults for task consistency. Targets were presented randomly to minimise any systematic bias.

One measurement capture was obtained for each AS at each visit. Each capture included six B‐scans at 30‐degree meridional intervals, which provided on‐axis measurements of central corneal thickness (CCT; anterior to posterior cornea), anterior chamber depth (ACD; posterior cornea to anterior lens), crystalline lens thickness (LT; anterior to posterior lens) and axial length (AL; anterior cornea to retinal pigment epithelium). During each measurement, participants were encouraged to maintain clarity of the target and steady fixation for the capture duration, which was typically 30–60 s. If they reported a blurred target or loss of fixation, the measurement was recaptured, but the participant's data were excluded for that AS if they were again unable to maintain target fixation or clarity.

The accommodation response (AR) was determined based on wavefront aberration data captured using the Hartmann‐Shack aberrometer (irx3, imagine‐eyes.com). Using the same randomisation strategy, 0, 2, 4 and 6 D AS were presented in semi‐randomised order by adjusting the focus of the internal fixation target of the aberrometer (a tumbling E target). The instrument was aligned to the pupil centre for each measurement, and five measurements were captured for each AS for the left eye only, while the right eye was covered with an eye patch. A single drop of 1% tropicamide was instilled into the left eye to induce cycloplegia, and wavefront aberration measurements were repeated after 25 min.

Following the first visit, 0.025% atropine sulphate eye drops were prescribed for each participant. The eye drops were a multidose preparation preserved with 0.01% benzalkonium chloride, prepared by a local sterile compounding pharmacy (Your Solution Compounding Pharmacy, yoursolutioncompounding.com). The eye drops were placed in an aluminium foil bag and delivered directly from the pharmacy to the participant's home using cold chain delivery (i.e., packaged in an insulated box with ice packs) or were collected by the research team and delivered to the participants at their home or at the laboratory. Participants were instructed to instil one drop in each eye every night between 19:00 and 23:00 h (immediately before sleep) for ~7 nights (mean [SD]: 6.5 [1.9] nights; range: 4–10 nights) and text message reminders were sent each evening at 19:00 h, to maximise compliance. Participants were instructed to store the eye drops in the aluminium foil bag and refrigerate the eye drops following drop instillation.

A second visit was conducted on the day after the participant's final evening dose. Adherence to the dosage regimen, tolerance and adverse events were queried with each participant. Assessments of distance and near visual acuity, cover test, amplitude of accommodation, intraocular pressure and anterior eye examination were repeated, followed by the same data collection procedures as for the first visit (except for the cycloplegic measurement of wavefront aberrations). Both study visits were conducted between 14:00 and 18:00 h to minimise the effect of diurnal variation. ²¹

Ocular biometry images were manually reviewed and measurements from individual B‐scans were removed if alignment or quality was poor. Measurements were averaged from the remaining B‐scans. Each participant's data from the 2, 4 and 6 D accommodation stimuli were reviewed to confirm an active accommodation response. Participants were assumed to be actively accommodating if there was a decrease in ACD (<0 μm) and an increase in LT (>0 μm) of any magnitude. ¹² , ¹⁷ Additional biometric parameters were calculated, including anterior segment length (ASL; anterior cornea to posterior lens = CCT + ACD + LT), vitreous chamber depth (VCD; posterior lens to retinal pigment epithelium = AL − ASL) and crystalline lens centre position (LCP; distance of crystalline lens centre from anterior cornea = CCT + ACD + ½ LT). Due to increased LT during accommodation, the crystalline lens occupies a greater proportion of the eye and increases the average refractive index of the eye used by optical biometers to convert from optical path lengths to geometrical dimensions. Hence, optical biometers have been reported to overestimate VCD and AL measurements during accommodation. ²² Therefore, these parameters were corrected to account for this estimated error (cVCD and cAL), as described previously. ¹⁰ , ¹² , ¹⁷

For each aberrometry measurement, sixth‐order Zernike polynomials for the measured pupil diameter were exported from the irx3 instrument and imported into customised software where the Zernike polynomial was rescaled to a 2.3‐mm pupil using the methods of Schwiegerling ²³ to minimise the influence of spherical aberration on the refractive error determination, ²⁴ and to match the minimum pupil size for open‐field autorefractors often used in accommodation research, such as the Grand‐Seiko WAM‐5500 ²⁵ and Shin‐Nippon NVision‐K 5001. ²⁶ Customised software employing the methods of Iskander et al. ²⁷ was used to determine refractive error measurements for each AS and the cycloplegic measurement, and each measurement was averaged. The AR and accommodation error (AE) for each AS and visit were calculated using Equations (1) and (2) below:

$${\mathrm{AR}}_{\mathrm{AS}}={\mathrm{SER}}_{\mathrm{C}}-{\mathrm{SER}}_{\mathrm{AS}}$$ (1)

$${\mathrm{AE}}_{\mathrm{AS}}=\mathrm{AS}-{\mathrm{AR}}_{\mathrm{AS}}$$ (2)

where ${\mathrm{SER}}_{\mathrm{C}}$ was the SER determined for the cycloplegic measurement and ${\mathrm{SER}}_{\mathrm{AS}}$ was the SER for a particular AS. Although the derivation of SER from wavefront aberrometry differs, the calculation of AR and AE is consistent with the method used in Kaphle et al. ²⁸ Positive and negative AE values as calculated would indicate a lag and lead of accommodation, respectively.

For the AR and each biometric parameter (CCT, ACD, LT, ASL, LCP, cVCD and cAL), a linear mixed model was used (IBM SPSS v29.0.0.0, ibm.com) and included fixed factors of AS (0, 2, 4 and 6 D), age group (children and adolescents, 8.0–15.4 years; young adults, 20.6–25.5 years) and visit (pre‐ and post‐atropine use). Interactions between AS and visit, and AS, visit and age group were also explored. The models used a variance components matrix covariance structure, with the restricted maximum likelihood strategy used for the repeated measures and fixed effects, and individual participant intercepts included as random effects. To compare between AS, post hoc pairwise comparisons with a Sidak correction were used.

For any biometric parameters where there was a significant interaction, the change in the parameter from the 0 D stimulus at each stimulus for each visit was analysed using linear mixed models employing the same strategies, fixed factors and interactions. If the interaction was significant, post hoc pairwise comparisons with a Sidak correction were also used to compare the level of accommodation‐induced change between visits and between age groups for each AS.

RESULTS

Participant characteristics

Two of the child participants (one male and one female) were excluded from all analyses due to poor image quality arising from the reduced IOLMaster light source power. The mean age (SD) of the remaining 26 participants was 17.0 (6.0) years (range: 8.0–25.5 years) and the mean SER (SD) of their left eye was −2.03 (1.05) D (range: −0.75 to −4.38 D).

For each AS, data from a small number of participants were excluded due to a poor AR (i.e., there was an increase in ACD and/or a decrease in LT). Hence, data for fewer participants were included at each AS. The demographic data for the participants included in the analyses are presented in Table 1. Linear mixed models revealed that age and SER did not differ on average, with AS (both p > 0.96) or between visits (both p > 0.61) for the whole cohort; no significant difference in SER was observed between the age groups for any AS (p > 0.98).

TABLE 1.

Number of participants (N), mean (standard deviation, SD) age, spherical equivalent refraction (SER), the accommodation response (AR) and error (AE) and the pupil diameter (PD) for the included participants at each accommodation stimulus (0, 2, 4 and 6 D) for Visit 1 (Pre‐atropine) and 2 (Post‐atropine) for each age group and all participants combined.

	Visit	Children and adolescents				Young adults				All participants
		0 D	2 D	4 D	6 D	0 D	2 D	4 D	6 D	0 D	2 D	4 D	6 D
N	Pre	13	11	12	12	13	11	13	13	26	22	25	25
	Post		8	10	13		8	12	13		16	22	26
Age (years)	Pre	11.4 (2.0)	11.5 (2.1)	11.5 (2.0)	11.5 (2.0)	22.7 (1.4)	22.7 (1.5)	22.7 (1.4)	22.7 (1.4)	17.0 (6.0)	17.1 (6.0)	17.3 (6.0)	17.3 (5.9)
	Post		11.0 (1.9)	11.4 (1.9)	11.4 (2.0)		22.6 (1.1)	22.4 (1.1)	22.7 (1.4)		16.8 (6.2)	17.4 (5.8)	17.0 (6.0)
SER (D)	Pre	−1.98 (1.20)	−1.72 (1.03)	−1.78 (1.01)	−2.08 (1.2)	−2.08 (0.93)	−2.10 (1.00)	−2.08 (0.93)	−2.08 (0.93)	−2.03 (1.05)	−1.91 (1.01)	−1.94 (0.96)	−2.08 (1.04)
	Post		−1.77 (1.33)	−2.05 (1.18)	−1.98 (1.2)		−2.16 (1.11)	−2.08 (0.97)	−2.08 (0.93)		−1.96 (1.23)	−2.07 (1.04)	−2.03 (1.05)
AR (D)	Pre	0.25 (0.45)	1.00 (0.44)	2.29 (0.90)	3.37 (1.52)	0.44 (0.36)	1.48 (0.34)	3.15 (0.40)	4.86 (0.52)	0.35 (0.41)	1.25 (0.45)	2.76 (0.79)	4.18 (1.31)
	Post	0.24 (0.41)	0.80 (0.36)	2.02 (0.88)	3.24 (1.41)	0.27 (0.25)	1.38 (0.32)	2.69 (0.65)	3.97 (0.89)	0.26 (0.33)	1.09 (0.44)	2.36 (0.83)	3.62 (1.21)
AE (D)	Pre	−0.25 (0.45)	1.00 (0.44)	1.71 (0.90)	2.63 (1.52)	−0.44 (0.36)	0.52 (0.34)	0.85 (0.40)	1.14 (0.52)	−0.35 (0.41)	0.75 (0.45)	1.24 (0.79)	1.82 (1.31)
	Post	−0.24 (0.41)	1.20 (0.36)	1.98 (0.88)	2.76 (1.41)	−0.27 (0.25)	0.62 (0.32)	1.31 (0.65)	2.03 (0.89)	−0.26 (0.33)	0.91 (0.44)	1.64 (0.83)	2.38 (1.21)
PD (mm)	Pre	6.13 (1.46)	5.63 (1.23)	5.88 (1.09)	4.84 (0.92)	6.69 (0.78)	6.28 (0.73)	5.76 (0.81)	5.10 (0.76)	6.41 (1.18)	5.97 (1.03)	5.81 (0.93)	4.98 (0.83)
	Post	6.72 (0.88)	6.53 (1.06)	6.36 (1.02)	5.53 (1.11)	7.30 (0.58)	7.01 (0.64)	6.57 (0.65)	5.84 (0.88)	7.01 (0.79)	6.77 (0.89)	6.46 (0.85)	5.69 (0.99)

The mean (SD) of each biometric parameter for relaxed accommodation (0 D stimulus) and the mean (SD) change of each parameter for each AS are presented in Table 2. The results of the statistical analyses are summarised in Table 3.

TABLE 2.

The mean (standard deviation, SD) measurement of central corneal thickness (CCT), anterior chamber depth (ACD), crystalline lens thickness (LT), anterior segment length (ASL), crystalline lens centre position (LCP), corrected vitreous chamber depth (cVCD) and corrected axial length (cAL) during relaxed accommodation (0 D accommodation stimulus) and the mean change (SD) in these parameters at each accommodation stimulus (2, 4 and 6 D) for Visit 1 (Pre‐atropine) and 2 (Post‐atropine) for each age group and all participants combined. Note the different units used for the measurement of each parameter for the 0 D stimulus (mm) and the mean changes during accommodation (μm).

Biometric parameter	Visit	Children and adolescents				Young adults				All participants
		0 D (mm)	Mean (SD) change from 0 D stimulus (μm)			0 D (mm)	Mean (SD) change from 0 D stimulus (μm)			0 D (mm)	Mean (SD) change from 0 D stimulus (μm)
			2 D	4 D	6 D		2 D	4 D	6 D		2 D	4 D	6 D
CCT	Pre	0.540 (0.036)	0 (3)	−1 (3)	−2 (4)	0.530 (0.032)	−1 (2)	−2 (2)	−2 (2)	0.535 (0.034)	−1 (3)	−2 (3)	−2 (3)
	Post	0.538 (0.035)	1 (3)	0 (3)	1 (3)	0.529 (0.032)	0 (2)	−1 (3)	−1 (2)	0.533 (0.033)	0 (3)	−1 (3)	0 (3)
ACD	Pre	3.24 (0.22)	−24 (17)	−68 (45)	−140 (51)	3.10 (0.33)	−52 (37)	−98 (49)	−159 (52)	3.17 (0.29)	−38 (31)	−84 (49)	−150 (52)
	Post	3.24 (0.23)	−31 (30)	−46 (28)	−100 (59)	3.11 (0.32)	−33 (28)	−70 (34)	−111 (52)	3.17 (0.28)	−30 (28)	−59 (33)	−105 (55)
LT	Pre	3.33 (0.22)	35 (23)	86 (45)	186 (67)	3.54 (0.17)	63 (41)	124 (59)	211 (78)	3.43 (0.22)	49 (35)	106 (55)	199 (72)
	Post	3.36 (0.29)	43 (43)	49 (33)	119 (81)	3.55 (0.15)	41 (32)	85 (40)	140 (66)	3.45 (0.25)	39 (36)	69 (41)	129 (73)
ASL	Pre	7.11 (0.19)	11 (13)	17 (12)	44 (23)	7.17 (0.27)	10 (12)	24 (15)	50 (39)	7.14 (0.23)	10 (12)	21 (14)	47 (32)
	Post	7.13 (0.21)	10 (27)	3 (8)	20 (26)	7.18 (0.27)	7 (7)	14 (9)	28 (21)	7.16 (0.24)	10 (19)	9 (10)	24 (23)
LCP	Pre	5.45 (0.17)	−7 (9)	−26 (23)	−49 (22)	5.40 (0.29)	−22 (17)	−38 (21)	−55 (24)	5.42 (0.23)	−14 (16)	−32 (22)	−52 (23)
	Post	5.45 (0.17)	−8 (20)	−22 (13)	−39 (20)	5.41 (0.28)	−11 (14)	−28 (15)	−42 (22)	5.43 (0.23)	−10 (16)	−25 (14)	−40 (21)
cVCD	Pre	17.10 (0.41)	−9 (10)	−17 (12)	−43 (24)	17.40 (0.54)	−7 (14)	−20 (20)	−46 (41)	17.25 (0.49)	−8 (12)	−18 (16)	−44 (33)
	Post	17.08 (0.41)	−7 (19)	−3 (11)	−15 (23)	17.38 (0.56)	−7 (16)	−11 (14)	−26 (30)	17.23 (0.50)	−5 (18)	−7 (13)	−20 (27)
cAL	Pre	24.21 (0.43)	1 (5)	0 (8)	1 (12)	24.57 (0.54)	3 (9)	4 (9)	5 (10)	24.39 (0.51)	2 (7)	2 (8)	3 (11)
	Post	24.22 (0.44)	5 (9)	−1 (5)	4 (7)	24.56 (0.55)	0 (17)	3 (11)	2 (17)	24.39 (0.52)	4 (13)	1 (9)	3 (13)

TABLE 3.

F‐statistics, with degrees of freedom in parentheses and p‐values from the linear mixed model analyses including fixed factors of accommodation stimulus, visit and age group; the interaction between accommodation stimulus and visit and the interaction between accommodation stimulus, visit and age group for biometric parameters and accommodation response. Statistically significant results are presented in bold font.

Parameter	Accommodation stimulus	Visit	Age group	Accommodation by visit	Accommodation by visit by age
CCT	F (3,149.01) = 1.41, p = 0.26	F (1,149.01) = 2.85, p = 0.08	F (1,24.00) = 0.57, p = 0.46	F (3,149.01) = 0.72, p = 0.54	F (7,149.01) = 0.39, p = 0.92
ACD	F (3,148.03)   = 109.27, p < 0.001	F (1,148.03)   = 15.09, p < 0.001	F (1,24.01) = 1.96, p = 0.18	F (3,148.03)   = 3.63, p = 0.01	F (7,148.03) = 1.22, p = 0.30
LT	F (3,148.05)   = 103.84, p < 0.001	F (1,148.05) = 0.94, p = 0.33	F (1,24.00)   = 7.98, p = 0.01	F (3,148.03)   = 5.18, p = 0.002	F (7,148.04) = 1.71, p = 0.11
ASL	F (3,148.02)   = 16.38, p < 0.001	F (1,148.02)   = 21.28, p < 0.001	F (1,24.01) = 0.60, p = 0.45	F (3,148.02) = 2.65, p = 0.05	F (7,148.02)   = 2.14, p = 0.04
LCP	F (3,148.01)   = 61.50, p < 0.001	F (1,148.01)   = 28.96, p < 0.001	F (1,24.00) = 0.31, p = 0.58	F (3,148.01) = 0.93, p = 0.43	F (7,148.01) = 0.64, p = 0.72
cVCD	F (3,148.03)   = 4.04, p = 0.009	F (1,148.03)   = 17.66, p < 0.001	F (1,24.01) = 2.11, p = 0.16	F (3,148.02) = 0.77, p = 0.51	F (7,148.03) = 1.47, p = 0.18
cAL	F (3,148.01) = 0.23, p = 0.87	F (1,148.01)   = 9.77, p = 0.002	F (1,24.00) = 3.20, p = 0.09	F (3,148.01) = 0.40, p = 0.99	F (7,148.01) = 0.77, p = 0.62
AR	F (3,160.31)   = 402.59, p < 0.001	F (1,160.36)   = 10.91, p = 0.001	F (1,23.89)   = 10.94, p = 0.003	F (3,160.12) = 1.40, p = 0.24	F (7,160.22)   = 4.29, p < 0.001
AE	F (3,160.31)   = 182.86, p < 0.001	F (1,160.36)   = 10.91, p = 0.001	F (1,23.89)   = 10.94, p = 0.003	F (3,160.12) = 1.40, p = 0.24	F (7,160.22)   = 4.29, p < 0.001
PD	F (3,160.05)   = 84.53, p < 0.001	F (1,160.06)   = 122.99, p < 0.001	F (1,23.92) = 1.63, p = 0.22	F (3,159.99) = 0.80, p = 0.50	F (7,160.02) = 0.99, p = 0.44

Abbreviations: AR, accommodation response; AE, accommodative error; ACD, anterior chamber depth; ASL, anterior segment length; CCT, central corneal thickness; cAL, corrected axial length; cVCD, corrected vitreous chamber depth; LCP, crystalline lens centre position; LT, crystalline lens thickness; PD, pupil diameter.

Effect of accommodation on ocular biometry

Averaged across both visits, a significant change in ACD, LT, ASL, LCP and cVCD was observed with accommodation (all p ≤ 0.009). LT increased, and ACD and LCP decreased with higher levels of accommodation, with all pairwise comparisons significant between AS (all p ≤ 0.02). ASL also increased but was significantly greater only for the 6 D AS compared to all other stimuli (all pairwise comparisons, p ≤ 0.001). However, there was no significant difference between the 0, 2 and 4 D stimuli (all pairwise comparisons, p ≥ 0.46). Compared to the 0 D stimulus, cVCD was significantly shorter for the 6 D stimulus (p = 0.01), but no other pairwise comparisons were significant (all p ≥ 0.07). Both CCT and cAL did not change significantly with accommodation (both p ≥ 0.26).

Effect of 0.025% atropine on ocular biometry

ACD, ASL and LCP were significantly greater on average across all AS following the use of atropine (all p < 0.001). cVCD and cAL were both shorter post‐atropine compared to pre‐atropine (both p ≤ 0.002).

A significant interaction between AS and visit was observed for ACD and LT (both p ≤ 0.01), with each exhibiting a reduction in the amount of accommodation‐induced change following the use of atropine (all p ≤ 0.007) (Figure 1). Post hoc analyses indicated that the accommodation‐induced changes in ACD and LT were significantly reduced for the 4 and 6 D accommodation stimuli (both pairwise comparisons, p ≤ 0.004), but not for the 2 D stimulus (all p ≥ 0.36). The interaction between AS and visit was not significant for CCT, ASL, LCP, cVCD or cAL (all p ≥ 0.05).

FIGURE 1.

The mean change in central corneal thickness (CCT), anterior chamber depth (ACD), crystalline lens thickness (LT), anterior segment length (ASL) and corrected vitreous chamber depth (cVCD) during accommodation for all participants before (solid lines/filled symbols) and after atropine (dashed lines/open symbols). The error bars represent the standard error of the mean, and asterisks indicate statistically significant pairwise comparisons (p < 0.05) between the accommodation‐induced change in the parameter, pre‐ and post‐atropine.

Effect of age

The young adults exhibited a thicker crystalline lens on average compared to the children and adolescents (p = 0.01), but no age group differences were observed for any other biometric variables (all p ≥ 0.09). A significant interaction between AS, visit and age group was observed for ASL (p = 0.04); however, the amount of change in ASL with accommodation during each visit was similar between the age groups (p = 0.77) and no significant interaction was observed for any other biometric parameters (all p ≥ 0.11).

Accommodation response and error

For all participants combined, the AR increased significantly with increasing AS level (p < 0.001), with all between‐stimuli comparisons significant (all p < 0.001). AE also varied significantly with accommodation (p < 0.001), with a small lead of accommodation of −0.35 (0.41) D for the 0 D AS, followed by an increasing lag of accommodation with greater AS, with lags of 0.75 (0.45), 1.24 (0.79) and 1.82 (1.31) D for the 2, 4 and 6 D stimuli, respectively (all pairwise comparisons between stimuli, p < 0.001) (Figure 2). On average across all AS and both visits, the young adults exhibited a greater AR of 2.26 (1.62) D compared to 1.58 (1.44) D for the children and adolescents (p = 0.003), and consequently, the children exhibited a higher lag of accommodation (1.29 (1.38) D) than the young adults (0.72 (0.91) D, p = 0.003).

FIGURE 2.

The mean accommodation response before (solid lines/filled circles) and after atropine (dashed lines/open squares) for all participants. The error bars represent the standard error of the mean, and asterisks indicate statistically significant pairwise comparisons (p < 0.05) between the accommodation‐induced change in the parameter, pre‐ and post‐atropine. The dotted line indicates the line of equality where the accommodation stimulus and response are equal.

Averaged across all AS, a significant decline in the AR (and increase in the AE) was observed following the use of atropine (p = 0.001), with a mean decrease in AR (and increase in AE) of 0.38 (0.22) D. A significant interaction between AS, visit and age group was observed (p < 0.001), with pairwise comparisons revealing a significant reduction in AR (and increase in the AE) for the 4 and 6 D accommodation stimuli following the use of atropine for the young adults (all p ≤ 0.03), but not for the children and adolescents (all p ≥ 0.35). Pairwise comparisons revealed that the young adults produced a significantly greater AR (and reduced lag of accommodation) compared to the children and adolescents for the 4 and 6 D AS, both pre‐ and post‐atropine use (all p ≤ 0.02) and also for the 2 D AS following atropine use (p = 0.04).

Pupil size

Pupil size decreased significantly with increasing accommodation, on average, across both visits (p < 0.001), with significant pairwise comparisons between all AS levels (p ≤ 0.03). There was a significant increase in pupil size following atropine use on average across all accommodation stimuli, with a mean increase of 0.76 (0.64) mm (p < 0.001). There was no interaction for the difference in pupil size between the age groups, nor was there an accommodation by visit interaction or an accommodation by visit by age group interaction (both ≥ 0.22).

DISCUSSION

This study demonstrated that nightly use of 0.025% atropine eye drops in young myopes reduces the magnitude of change in the crystalline lens that occurs during accommodation, consistent with the associated reduction in the AR with atropine. While differences in the AR and the change in the AR following atropine use were observed between the young adults and the children and adolescents, perhaps due to different levels of task engagement between the age groups, no substantial age‐related differences were observed in the accommodation‐induced structural changes. This indicates that the effects of atropine were similar in children, adolescents and young adults in the short term. Previous research has shown axial elongation during accommodation in children and adults, ¹⁰ , ¹¹ , ¹² , ¹⁶ , ¹⁷ yet in this study, no significant accommodation‐induced axial elongation was observed before or after 1 week of 0.025% atropine use, likely related to a reduced AR compared to previous research.

Anterior segment changes during accommodation were diminished following 1 week of 0.025% atropine use. Accommodation‐induced ACD and LT changes were reduced by a mean of 25 and 37 μm for the 4 D stimulus and 45 and 70 μm for the 6 D stimulus, respectively. A reduction in ACD (~10 μm) has been detected previously, 1 h post‐instillation of 0.01% atropine for a 5 D AS in myopic adults, with no changes in the ACD or LT observed up to 24 h post‐instillation. ¹⁸ This 10 μm reduction was significantly less than that observed for the 4 and 6 D stimulus in the current study and likely represents the concentration‐dependent effect of atropine. ⁴ , ⁵ Alternatively, the difference in the response could be related to participant ethnicity and the level of intraocular pigmentation, given that there is some evidence that these individual characteristics can influence the response to atropine. ²⁹ Mitsukawa et al. ¹⁸ did not report the ethnicity of their participants, although it is likely that they were predominantly East Asian based on the study's location, while the participants included in the present study were predominantly Caucasian. There could also be some capacity for accumulation of atropine in the ocular tissues over time with repeated doses. ³⁰ , ³¹ However, this study was not statistically powered to examine the effect of ethnicity or intraocular pigmentation; hence, further research is required to explore these factors.

During accommodation, ciliary muscle contraction leads to an anterior and inward displacement of the ciliary body, resulting in reduced zonular tension which allows the crystalline lens shape to alter. ³² Atropine reduces muscular contraction and causes less anterior and inward ciliary body movement, leading to less zonular tension relaxation and crystalline lens shape alteration; ¹ hence, the observed reduction in the anterior segment changes during accommodation in this study. Ciliary muscle contraction and sequelae during accommodation have been implicated in several accommodation‐induced changes in the eye which may be linked to myopia development and/or progression, including choroidal thinning, ³³ axial elongation ¹⁰ , ¹² , ¹⁷ and scleral thinning. ³⁴ Therefore, although it is generally uncontested that atropine slows myopia progression via a non‐accommodative mechanism, ⁸ the reduction in accommodation‐induced biomechanical forces during atropine use could reduce these structural changes and may contribute to atropine's mechanism of action for slowing myopia and axial elongation. In particular, the reduced anterior segment changes were observed for ≥4 D accommodation stimuli (equivalent to a working distance of ≤25 cm). This is relevant given the association between myopia and the use of short working distances (<30 cm) during near work, particularly by young children. ³⁵ , ³⁶ , ³⁷ Likewise, it is possible that children being treated with atropine may increase their habitual working distance and engage in near work activities at a lower AS due to the reduced AR associated with atropine use, which may also reduce the potential biomechanical stress placed on the eye during accommodation. Therefore, atropine may be particularly effective for those children who engage in excessive near work by acting on both a non‐accommodative and accommodative mechanism in humans; however, longitudinal studies are required to test these hypotheses.

Another novel finding was the increase in ACD, LCP and ASL following the use of 0.025% atropine, which indicates that the crystalline lens undergoes a small posterior displacement of ~15–20 μm based on the estimated marginal means from the linear mixed model. Using the difference in the estimated marginal mean values at each visit for each parameter (CCT, ACD, LT and cVCD), a backwards ray trace through the Le Grand full theoretical eye indicates that this small posterior shift in crystalline lens position results in a slight hyperopic shift of ~0.09 D. This is consistent with previous reports of dose‐dependent hyperopic shifts with varying concentrations of atropine, ranging from ~0 D with 0.01%, ⁴ , ³⁸ ~0.3 D with 0.1%, ~0.4 D with 0.5% ⁴ and >0.50 D with 1% atropine, ³⁸ , ³⁹ and provides some evidence for the mechanism underpinning the hyperopic shift.

Based on the estimated marginal means, a small post‐atropine reduction in cVCD and cAL was observed (cVCD reduction: ~38 μm; cAL reduction: ~18 μm). The reduction in cAL is consistent with the effect of atropine and likely reflects a small choroidal thickening, which has been reported previously following 1 week of 1% atropine use, ⁴⁰ , ⁴¹ and with 0.025% and 0.05% atropine after 4 months of use. ⁴² This choroidal thickening combined with the aforementioned posterior displacement in the crystalline lens likely accounts for the observed cVCD reduction. Previous clinical trials examining the myopia control efficacy of atropine have remeasured cycloplegic refraction after 2 weeks of treatment to account for the short‐term hyperopic shift; however, baseline AL was referenced to pre‐treatment measurements. ⁴ , ⁵ Although small in magnitude, consideration should be given to these short‐term AL changes in the design of myopia control clinical trials using atropine alone or in combination with another treatment, given that AL is often a primary outcome measure. However, it is also possible that the eye adapts to these small, short‐term changes in ACD, LT, cVCD and cAL. Longitudinal studies comprehensively measuring the short‐ and long‐term time course of ocular parameter changes during atropine treatment are required to understand better the significance of these short‐term observations.

It is important to note that the accommodation‐induced changes prior to atropine use in this study are less than previously reported in adults and children. For the 6 D AS before atropine was instilled, there was an ACD reduction of 150 μm and an LT increase of 199 μm, which was ~40% less than previously reported for myopic young adults ¹⁰ and children ¹⁷ (ACD reduction of 285 μm and 238 μm and LT increase of 346 μm and 326 μm, respectively). This reduced response to the same accommodation stimuli also resulted in a reduction in accommodation‐induced changes in cVCD and cAL. Based on correlations between the change in ACD and LT and the change in AL during accommodation in myopic children, ¹⁷ it seems that, had the participants accommodated similarly to previous cohorts, the AL increases would have been ~8 μm greater on average than the pre‐atropine mean cAL change, which may have altered the findings.

It appears likely that the reduced biometry changes were due to the participants exhibiting a reduced AR compared to previous studies, although the reason for this is unclear given that the experimental set‐up, conditions and instrumentation in this study were similar to previous investigations. ¹⁰ , ¹⁷ It is also important to note that the AR and accommodation‐induced structural changes were measured separately, using different fixation targets, and hence, it is possible that individual participants accommodated differently during measurements using the two instruments.

The children and adolescents exhibited an AR that was generally poorer than the young adults both pre‐ and post‐atropine. This was most likely due to a shorter attention span on the part of the children and adolescents resulting in less task engagement; even though the accommodation targets were designed to maximise engagement, the task was of short duration and participants were encouraged to maintain clear focus during the task. Due to their overall reduced AR, the children and adolescents' AR was minimally affected by the use of atropine, while the post‐atropine AR for the young adults was significantly reduced compared to their pre‐atropine AR. Measuring accommodation‐induced biometric changes and the AR simultaneously would facilitate greater understanding of the impact of atropine on these changes, in particular, accommodation‐induced axial elongation.

The 0.025% atropine concentration used in this study required formulation by a sterile compounding pharmacy. Given that inconsistencies in atropine formulations, labelling and storage instructions provided by different compounding pharmacies have been reported, ⁴³ there may be differences in the atropine preparations in this study compared to previous research using the same concentration. Although prepared by the same pharmacy, it is possible that there were discrepancies in the chemical characteristics (e.g., atropine and impurity concentration, pH, viscosity) of the atropine eye drops used by different participants, ⁴⁴ which may have influenced atropine stability during the study and therefore the results. However, participants were instructed to store the drops in an aluminium foil bag in the refrigerator, to which they all reportedly complied. Hence, it is unlikely that atropine stability will have varied substantially over the short duration of use in this study (4–10 nights). ⁴⁵ , ⁴⁶

Adherence to the dosing regimen was queried at the second study visit, with all participants reporting 100% compliance except for one young adult participant, who missed one dose on night 4 of 7. Additional evidence of good compliance with the dosing regimen was the significant mean increase in pupil size of ~0.76 mm between the visits, which is consistent with previous research using 0.025% atropine eye drops. ⁵ Hence, it is unlikely that poor compliance influenced the results.

While the repeated‐measures study design is a strength, it is possible that there was a learning effect associated with greater task familiarity at the second visit, which may have resulted in false enhancement of the post‐atropine changes or may partially explain the reduced AR and expected pre‐atropine biometry changes due to less task familiarity, particularly in the children and adolescents. Future studies could explore the eye's response to accommodation during atropine treatment by considering measurement of choroidal thickness, choroidal blood flow and ciliary muscle and body changes during accommodation. It would be of particular interest to determine whether the reduction in these accommodation‐induced changes influenced the efficacy of atropine to slow myopia progression and axial elongation over time.

In conclusion, ~1 week of nightly 0.025% atropine instillation reduced the anterior segment structural changes that accompany accommodation. AL elongation during accommodation was not detected, likely due to reduced AR by the study participants. These findings suggest that atropine inhibits the amount of structural change during accommodation, which may indicate a reduction in the internal ocular biomechanical forces within the myopic eye associated with accommodation. These findings may have implications for our understanding of the association between near work and myopia, and the ocular effects and mechanism of atropine eye drops.

AUTHOR CONTRIBUTIONS

Rohan P. J. Hughes: Conceptualization (lead); data curation (lead); formal analysis (lead); funding acquisition (lead); investigation (lead); methodology (lead); project administration (lead); resources (lead); validation (lead); visualization (lead); writing – original draft (lead); writing – review and editing (supporting). Emily C. Woodman‐Pieterse: Formal analysis (supporting); methodology (supporting); supervision (supporting); visualization (supporting); writing – review and editing (equal). Scott A. Read: Formal analysis (supporting); methodology (supporting); supervision (supporting); visualization (supporting); writing – review and editing (equal). Stephen J. Vincent: Formal analysis (supporting); funding acquisition (supporting); methodology (supporting); supervision (supporting); visualization (supporting); writing – review and editing (equal). Michael J. Collins: Formal analysis (supporting); funding acquisition (supporting); methodology (supporting); project administration (supporting); resources (supporting); supervision (lead); visualization (supporting); writing – review and editing (equal).

FUNDING INFORMATION

This research received funding from the Children's Hospital Foundation in the form of an Early Career Fellowship awarded to Dr. Hughes.

CONFLICT OF INTEREST STATEMENT

The authors report no conflicts of interest and have no proprietary interest in any of the materials mentioned in this article.

CLINICAL TRIAL REGISTRATION

This study was registered on the Australian New Zealand Clinical Trials Registry (ACTRN1261000563864).

Hughes RPJ, Woodman‐Pieterse EC, Read SA, Vincent SJ, Collins MJ. Effect of 0.025% atropine on ocular biometry changes during accommodation. Ophthalmic Physiol Opt. 2025;45:865–876. 10.1111/opo.13485

DATA AVAILABILITY STATEMENT

Data may be available from the corresponding author upon reasonable request.