Refractive errors are the leading cause of visual impairment throughout the world.1 They cause reduced vision in 63.6% to 94.9% of children between 5 and 15 years old, whereas amblyopia accounts for 1.9% to 7.3%.2–4 Accurate refractive correction is important to improve vision and treat amblyopia in children. Because children have a large accommodative range, the influence of accommodation on the measurement of refractive error is a concern in pediatric ophthalmology.5
The use of cycloplegic agents that reduce the accommodation spasm while determining the “true” manifest refractive error is widespread in clinical practice, especially for infants and children. Pharmaceutical paralysis of accommodation is often necessary to relax the habitual accommodative posture in young children and in patients with high hyperopia or strabismus. Atropine is an organic compound derived from tropic acid and tropine, and it is acknowledged as the “gold standard” for cycloplegic effectiveness. Atropine is the most potent of the cycloplegic agents, with slow onset and a duration of action lasting up to 2 weeks. Ocular or systemic adverse effects can occur with atropine-induced cycloplegia. These include allergic contact dermatitis of the eyelids, allergic conjunctivitis, keratitis, and increased intraocular pressure. Systemic side effects include dryness of secretions, fever, irritability, tachycardia, convulsions, and others.6
In clinical practice, glasses for myopic children are usually prescribed based not only on the cycloplegic refraction, but also on the refraction measured 3 weeks after stopping atropine. At that time, the mydriasis and cycloplegia induced by atropine have worn off. Generally, children need to see their ophthalmolgist three times before getting an accurate prescription. Because of the potential side effects of cycloplegia and the inconvenience of scheduling multiple clinical examinations, autorefractors that do not require cycloplegia have recently become widely used. Autorefraction is useful in obtaining the objective refractive status of children in vision screenings, clinical practice, and/or research settings such as epidemiologic surveys and clinical trials.7–10
It has remained unclear whether and to what extent refractometric results from studies avoiding cycloplegia can be compared with the findings of studies that apply cycloplegia. We conducted this study on preschool (aged 4 to 6 years) and school-aged (aged 7 to 10 years) children to compare the results of the final glasses prescriptions based on cycloplegic refractometry with those of non-cycloplegic refractometry. We also assessed factors associated with the differences between them. Our data may help to interpret the results of population-based studies of refractive error in children when cycloplegia was not applied.
Patients and Methods
This study followed the tenets of the Declaration of Helsinki and was approved by the Hospital Committee for the Protection of Human Subjects. This retrospective study included 11,795 patients from 2007 to 2016 who were prescribed atropine as a cycloplegic agent for refraction examinations in the clinic of the Eye Hospital of Wenzhou Medical University, Wenzhou, Zhejiang, People's Republic of China. The enrolled patients were divided into two groups: 4 to 6 years old (n = 5,320) and 7 to 10 years old (n = 6,475).
Initially, all of the children received routine ophthalmic evaluations including slit-lamp examination, funduscopy, intraocular pressure measured with a non-contact tonometer (CT1; Topcon), and refractive error measurement by autorefraction (KR8900; Topcon) and retinoscopy. For myopic children with retinoscopy-determined refractive errors of −0.50 diopters (D) or less, hyperopic children with retinoscopy-determined refractive errors of 1.50 D or greater, and children with best corrected visual acuity that was worse than normal, cycloplegia by atropine sulfate 1% was prescribed for home use, two times a day for 3 days. On the fourth day, the measurements were repeated under cycloplegia. Atropine use was then stopped, and after 3 weeks, when the cycloplegic effect disappeared, the measurements were repeated.
Glasses were prescribed for children with myopia based on the last retinoscopy results, not on the cycloplegic refraction. The reason for this protocol is derived from the fact that atropine cycloplegic refraction relies on an unnatural state as a result of the full paralysis of the ciliary muscle or paralysis of accommodation and dilation of the pupil.11 After stopping the use of atropine for 3 weeks, the mydriasis and cycloplegia will have worn off, and the natural state with tonic accommodation will have been restored.12 For myopic children, in most cases, the last result based on non-cycloplegic refraction is more minus than the cycloplegic refraction.13 Thus, prescriptions made for eyes treated with atropine will be undercorrected for the natural state without cycloplegia. As many previous studies have shown, the undercorrection will facilitate myopia progression.14–16 Therefore, we prescribed the corrective lenses based on the last retinoscopy results. Children with hyperopia were given approximately 1.50 D smaller plus correction than the cycloplegic refraction as the glass prescription if no esotropia was present.
Data Management and Analysis
The refractive error of the right eye measured by retinoscopy for each patient was used for analysis. Children with astigmatism of greater than 1.50 D and those with esotropia were excluded. Refractive error was expressed as the spherical equivalent (SE) (ie, spherical refractive error plus half of the cylindrical refractive error). Children with a refractive error of less than 0.00 D were assigned to the myopia group, and those with a refractive error of more than 0.50 D were assigned to the hyperopia group. Within the myopia and hyperopia groups, the patients were subdivided into those with mild, moderate, and high myopia based on the dioptric assessments (Table 1). The distribution of the refractive errors was further analyzed by stratifying the study population by two age groups: 4 to 6 years (preschool) and 7 to 10 years (primary school).
Refractive Errors and Differences as Measured by Non-cyloplegic Retinoscopy, Cycloplegic Retinoscopy, and Glasses
The statistical analyses were performed using a commercially available statistical software package (SPSS for Windows version 17.0; IBM-SPSS). Descriptive statistics of continuous variables were calculated as means ± standard deviation. The data in all groups were normally distributed as determined by the Kolmogorov–Smirnov test. Comparisons between measurements were made by independent two-tailed t tests. Correlations between age and refraction with and without cycloplegia were determined by Pearson correlation analysis. All P values were two-sided and considered statistically significant when the values were less than .05.
In children 4 to 6 years old with mild, moderate, and high myopia, the mean differences between the cycloplegic refractive error and the non-cycloplegic refractive error (DIFFC-N) were 1.21 ± 0.90, 0.56 ± 0.65, and 1.08 ± 1.08 D, respectively. For children with mild, moderate, and high hyperopia, the DIFFC-N were 1.58 ± 0.90, 1.65 ± 1.04, and 1.03 ± 0.91 D. For all groups, the DIFFC-N were significant (all P < .05, Table 1). Similarly, for children 7 to 10 years old with mild, moderate, and high myopia, the mean DIFFC-N were 0.41 ± 0.56, 0.38 ± 0.52, and 0.70 ± 0.81 D, respectively. For children with mild, moderate, and high hyperopia, the mean DIFFC-N were 1.25 ± 0.94, 1.33 ± 0.99, and 0.89 ± 0.90 D. For all groups, the DIFFC-N were significant (all P < .05, Table 1).
In the 4 to 6 years group, the DIFFC-N in the groups with mild and high myopia were higher than in the groups with moderate myopia (both P < .05, Table 1). For the 7 to 10 years group, the DIFFC-N were higher in the groups with mild and high myopia than the group with moderate myopia (both P < .05). In that age group, the DIFFC-N of the patients with high myopia was the largest among the three myopia groups (Table 1). In both age groups, the DIFFC-N for the groups with mild and moderate hyperopia were greater than for the group with high hyperopia (all P < .05, Table 1).
Children in the 4 to 6 year age group tended to have larger DIFFC-N than the 7 to 10 year age group. The DIFFC-N was negatively correlated with age (r = −0.356, P < .001, Figure 1).
Correlation between the differences between cycloplegic and non-cycloplegic refractive errors (DFFC-N) and age. DIFFC-N was negatively correlated with age (r = −0.356, P < .001). D = diopters
For children with mild myopia, most of the DIFFC-N were 0.00 to 0.50 D (42%) and 0.50 to 1.00 D (26%) (Figure 2). For children with moderate myopia, most of the DIFFC-N were 0.00 to 0.50 D (39%) and 0.50 to 1.00 D (35%). In contrast, there was little variation (12% to 22%) in the percentage of patients among the different DIFFC-N intervals for the group with high myopia. In children with mild and moderate hyperopia, most (44% for mild hyperopia, 40% for moderate hyperopia) had DIFFC-N of 1.00 to 2.00 D (Figure 2). For the same two groups, 31% each had DIFFC-N of less than 2.00 D. In children with high hyperopia, most of the DIFFC-N were between 0.50 and 1.50 D.
Distribution of the differences between cycloplegic and non-cycloplegic refractive errors (DIFFC-N) in different refractive error intervals of all children: (A) mild myopia, (B) moderate myopia, (C) high myopia, (D) mild hyperopia, (E) moderate hyperopia, and (F) high hyperopia. D = diopters
When glasses were prescribed 3 weeks after atropine eye drops were discontinued, the differences in refractive error between the prescription and the non-cycloplegic refraction (DIFFG-N) for the 4 to 6 years group were largest in the group with high myopia (0.83 ± 1.15 D, Table 1). For the 7 to 10 years group, the largest DIFFG-N was 0.60 ± 1.47 D for the high myopia group (Table 1). The smallest DIFFG-N were for the children with high hyperopia in both age groups.
In the 4 to 6 years group, there were 905 children with mild myopia before the instillation of atropine (Table 2). In the cycloplegic state, 339 children (37.5%) were still mildly myopic, whereas the remaining 566 (62.5%) were either emmetropic or hyperopic. For the same age group, 629 children with mild myopia had a refractive error of −1.00 D or less before the instillation of atropine (Table 2). Only 88 children (14%) remained mildly myopic after cycloplegia, whereas 541 (86%) were either emmetropic or hyperopic. Among the 7 to 10 years group, 11.3% of the patients with mild myopia and 41.1% of the patients with a non-cycloplegic refractive error of −1.00 D or less became emmetropic or hyperopic under cycloplegia.
Refractive Error State Changes After Cycloplegia
Cycloplegic retinoscopy and subjective refraction remain the gold standards for measuring the refractive state in children. Cycloplegia, typically with either atropine or cyclopentolate, allows estimation of the true refractive error by relaxing accommodation. For preschool and school-aged children examined during their school holidays, atropine is often used as the cycloplegic agent in Chinese clinical practice. For school-aged children examined while school is in session, cyclopentolate is often used to minimize interference with their classroom time. Because all of the cycloplegic medications have side effects, many studies that report the prevalence of myopia screen children under non-cycloplegic conditions.7,8 The large range of accommodation that children have makes the results of refractometry shift in a more myopic direction. We found that after cycloplegia, 62.5% of the children in the 4 to 6 years group and 11.3% of the children in the 7 to 10 years group with mild myopia became emmetropic or hyperopic. However, the lack of cycloplegia is associated with overestimation of the prevalence of myopia and marked errors in the estimation of the prevalence of emmetropia and hyperopia.9,17,18
We found that the refractive errors of glasses prescriptions for myopic children were more minus than the non-cycloplegic refractive errors, especially in children with high myopia. When the prescription for glasses was based solely on non-cycloplegic autorefraction, then the dioptric correction was greater than required. Thus, the children needed to exert a large accommodative effort to overcome the excessive minus correction. Mutti et al19 and Hepsen et al20 have shown that excess accommodation is a possible factor contributing to myopic progression. Therefore, cycloplegia is necessary for all myopic children between 4 and 10 years old, especially for children with high myopia. The practice of using non-cycloplegic autorefraction should be subjected to closer scrutiny, especially if the glasses prescription is made based solely on these readings.
After cycloplegia, the refractive errors changed in the hyperopic children. The DIFFC-N values in the mild and moderate hyperopia groups were greater than in the group with high hyperopia. This is consistent with the accommodation used in daily life. In theory, the amplitude of accommodation is approximately 12.00 D in children, and comfortable reading uses approximately half of the amplitude of accommodation.21,22 If the accommodation needs to be higher, the children stop trying to accommodate and relax. Therefore, children with high hyperopia use less accommodation than children with mild and moderate hyperopia in daily life. Accommodative esotropia is often present in children with mild and moderate hyperopia, but rarely in those with high hyperopia. Children with high hyperopia often have exophoria or external exotropia.23
In the current study, we found that the refractive error of prescription glasses and the non-cycloplegic refractive error were similar in children with high hyperopia; thus, there was no spasm of accommodation in them. Nevertheless, cycloplegia is also needed in children with high hyperopia. Previous clinical studies have shown that, after cycloplegia, children with high hyperopia more easily adapt to the new glasses with high refractive diopter.24 This provides a greater benefit in treating amblyopia. Thus, cycloplegia is necessary for all examinations of children 4 to 10 years old with hyperopia, especially if strabismus or amblyopia are present.
There are some potential limitations in our study. First, because this study was retrospective in nature, it is possible that some children did not use the atropine in their home as expected. Second, multiple clinicians were involved with making the refractive measurements. This could have introduced some human error or variability into the measurements. Third, only Asian children were included in this study, and there may be racial factors that could affect the amount of difference between cycloplegic and non-cycloplegic refractions. Thus, application of these findings may not necessarily be consistent across all children of all races.25 Further studies are needed to evaluate this. Fourth, for the children with hyperopia to more easily to adapt to the new high refractive error glasses, the prescriptions were often ordered just after the cycloplegia treatment. Therefore, most children with hyperopia had approximately 1.50 D smaller plus correction than the cycloplegic refraction as the glasses prescription, and there was no examination at 3 weeks after cycloplegia. Fifth, our study only observed the cycloplegic effects on children between 4 and 10 years old. Older children should be included in a future study. We found that the DIFFC-N were negatively correlated with age; therefore, older children may be less affected by cycloplegia.
The refractive errors before and after cycloplegia were different in children 4 to 10 years old with both myopia and hyperopia. Without cycloplegia, autorefraction tends to overestimate the refractive error of children with myopia and underestimate it in children with hyperopia. For accurate glasses prescriptions, cycloplegia should be used for children between 4 and 10 years old, especially those with high myopia.
- Pascolini D, Mariotti SP. Global estimates of visual impairment: 2010. Br J Ophthalmol. 2012;96(5):614–618. doi:10.1136/bjophthalmol-2011-300539 [CrossRef]
- Kumah BD, Ebri A, Abdul-Kabir M, et al. Refractive error and visual impairment in private school children in Ghana. Optom Vis Sci. 2013;90(12):1456–1461. doi:10.1097/OPX.0000000000000099 [CrossRef]
- He M, Zeng J, Liu Y, Xu J, Pokharel GP, Ellwein LB. Refractive error and visual impairment in urban children in southern china. Invest Ophthalmol Vis Sci. 2004;45(3):793–799. doi:10.1167/iovs.03-1051 [CrossRef]
- Naidoo KS, Raghunandan A, Mashige KP, et al. Refractive error and visual impairment in African children in South Africa. Invest Ophthalmol Vis Sci. 2003;44(9):3764–3770. doi:10.1167/iovs.03-0283 [CrossRef]
- Anderson HA, Stuebing KK. Subjective versus objective accommodative amplitude: preschool to presbyopia. Optom Vis Sci. 2014;91(11):1290–1301. doi:10.1097/OPX.0000000000000402 [CrossRef]
- Rosenbaum AL, Bateman JB, Bremer DL, Liu PY. Cycloplegic refraction in esotropic children. Cyclopentolate versus atropine. Ophthalmology. 1981;88(10):1031–1034. doi:10.1016/S0161-6420(81)80032-2 [CrossRef]
- Maul E, Barroso S, Munoz SR, Sperduto RD, Ellwein LB. Refractive Error Study in Children: results from La Florida, Chile. Am J Ophthalmol. 2000;129(4):445–454. doi:10.1016/S0002-9394(99)00454-7 [CrossRef]
- Pokharel GP, Negrel AD, Munoz SR, Ellwein LB. Refractive Error Study in Children: results from Mechi Zone, Nepal. Am J Ophthalmol. 2000;129(4):436–444. doi:10.1016/S0002-9394(99)00453-5 [CrossRef]
- Zhao J, Mao J, Luo R, Li F, Pokharel GP, Ellwein LB. Accuracy of noncycloplegic autorefraction in school-age children in China. Optom Vis Sci. 2004;81(1):49–55. doi:10.1097/00006324-200401000-00010 [CrossRef]
- Zhao J, Pan X, Sui R, Munoz SR, Sperduto RD, Ellwein LB. Refractive Error Study in Children: results from Shunyi District, China. Am J Ophthalmol. 2000;129(4):427–435. doi:10.1016/S0002-9394(99)00452-3 [CrossRef]
- Hoefnagel D. Toxic effects of atropine and homatropine eyedrops in children. N Engl J Med. 1961;264(264):168–171. doi:10.1056/NEJM196101262640403 [CrossRef]
- Zadnik K, Mutti DO, Kim HS, Jones LA, Qiu PH, Moeschberger ML. Tonic accommodation, age, and refractive error in children. Invest Ophthalmol Vis Sci. 1999;40(6):1050–1060.
- Kothari M, Hussain A. Is post mydriatic test necessary in children having compound myopic astigmatism?J Clin Ophthalmol Res. 2015;3(2):77–79. doi:10.4103/2320-3897.156587 [CrossRef]
- Chung K, Mohidin N, O'Leary DJ. Undercorrection of myopia enhances rather than inhibits myopia progression. Vision Res. 2002;42(22):2555–2559. doi:10.1016/S0042-6989(02)00258-4 [CrossRef]
- Vasudevan B, Esposito C, Peterson C, Coronado C, Ciuffreda KJ. Under-correction of human myopia—is it myopigenic?: a retrospective analysis of clinical refraction data. J Optom. 2014;7(3):147–152. doi:10.1016/j.optom.2013.12.007 [CrossRef]
- Li SY, Li SM, Zhou YH, et al. Effect of undercorrection on myopia progression in 12-year-old children. Graefes Arch Clin Exp Ophthalmol. 2015;253(8):1363–1368. doi:10.1007/s00417-015-3053-8 [CrossRef]
- Choong YF, Chen AH, Goh PP. A comparison of autorefraction and subjective refraction with and without cycloplegia in primary school children. Am J Ophthalmol. 2006;142(1):68–74. doi:10.1016/j.ajo.2006.01.084 [CrossRef]
- Fotedar R, Rochtchina E, Morgan I, Wang JJ, Mitchell P, Rose KA. Necessity of cycloplegia for assessing refractive error in 12-year-old children: a population-based study. Am J Ophthalmol. 2007;144(2):307–309. doi:10.1016/j.ajo.2007.03.041 [CrossRef]
- Mutti DO, Mitchell GL, Moeschberger ML, Jones LA, Zadnik K. Parental myopia, near work, school achievement, and children's refractive error. Invest Ophthalmol Vis Sci. 2002;43(12):3633–3640.
- Hepsen IF, Evereklioglu C, Bayramlar H. The effect of reading and near-work on the development of myopia in emmetropic boys: a prospective, controlled, three-year follow-up study. Vision Res. 2001;41(19):2511–2520. doi:10.1016/S0042-6989(01)00135-3 [CrossRef]
- Sterner B, Gellerstedt M, Sjöström A. The amplitude of accommodation in 6–10-year-old children—not as good as expected!Ophthalmic Physiol Opt. 2004;24(3):246–251. doi:10.1111/j.1475-1313.2004.00201.x [CrossRef]
- Castagno VD, Vilela MA, Meucci RD, et al. Amplitude of accommodation in schoolchildren. Curr--Eye Res. 2017;42(4):604–610. doi:10.1080/02713683.2016.1220586 [CrossRef]
- Kassem IS, Rubin SE, Kodsi SR. Exotropia in children with high hyperopia. J AAPOS. 2012;16(5):437–440. doi:10.1016/j.jaapos.2012.06.003 [CrossRef]
- Wutthiphan S. Guidelines for prescribing optical correction in children. J Med Assoc Thai. 2005;88(suppl 9):S163–S169.
- Manny RE, Fern KD, Zervas HJ, et al. 1% cyclopentolate hydrochloride: another look at the time course of cycloplegia using an objective measure of the accommodative response. Optom Vis Sci. 1993;70(8):651–665. doi:10.1097/00006324-199308000-00013 [CrossRef]
Refractive Errors and Differences as Measured by Non-cyloplegic Retinoscopy, Cycloplegic Retinoscopy, and Glasses
|Group||Dioptric Criteria (D)||No.||Non-cycloplegic Refractive Error (D)||Cycloplegic Refractive Error (D)||Glasses Prescription||DIFFC-N(D)||DIFFG-C(D)||DIFFG-N(D)|
|4 to 6 years|
| Mild myopia||< 0.00 to > −3.00||905||−0.88 ± 0.72||0.34 ± 1.36||−0.53 ± 1.00||1.21 ± 0.90||−0.87 ± 0.69||0.35 ± 0.63|
| Moderate myopia||≤ −3.00 to > −6.00||153||−4.13 ± 0.83||−3.57 ± 1.07||−3.90 ± 1.00||0.56 ± 0.65||−0.33 ± 0.54||0.24 ± 0.58|
| High myopia||≤ −6.00||99||−8.93 ± 3.01||−7.85 ± 2.92||−8.10 ± 2.91||1.08 ± 1.08||−0.25 ± 0.83||0.83 ± 1.15|
| Mild hyperopia||> 0.50 to < 3.00||2,610||0.93 ± 0.75||2.51 ± 1.24||1.37 ± 1.30||1.58 ± 0.90||−1.14 ± 0.74||0.44 ± 0.95|
| Moderate hyperopia||≥ 3.00 to < 6.00||827||4.33 ± 0.85||5.98 ± 1.25||5.01 ± 1.39||1.65 ± 1.04||−0.98 ± 0.81||0.67 ± 1.21|
| High hyperopia||≥ 6.00||726||7.54 ± 1.31||8.57 ± 1.43||7.51 ± 1.46||1.03 ± 0.91||−1.06 ± 0.79||−0.03 ± 1.06|
|7 to 10 years|
| Mild myopia||< 0.00 to > −3.00||3,136||−1.60 ±0.68||−1.19 ±0.97||−1.40 ±0.78||0.41 ±0.56||−0.21 ±0.50||0.19 ±0.45|
| Moderate myopia||≤ −3.00 to > −6.00||634||−3.74 ±0.72||−3.36 ±0.88||−3.51 ±0.86||0.38 ±0.52||−0.14 ±0.55||0.24 ±0.51|
| High myopia||≤ −6.00||75||−8.38 ±2.16||−7.68 ±2.05||−7.78 ±2.23||0.70 ±0.81||−0.10 ±1.09||0.60 ±1.47|
| Mild hyperopia||> 0.50 to < 3.00||1,617||0.98 ±0.78||2.23 ±1.30||1.21 ±1.55||1.25 ±0.94||−1.02 ±1.25||0.23 ±1.30|
| Moderate hyperopia||≥ 3.00 to < 6.00||607||4.37 ±0.87||5.70 ±1.21||4.64 ±1.35||1.33 ±0.99||−1.06 ±0.83||0.27 ±1.12|
| High hyperopia||≥ 6.00||406||7.57 ±1.31||8.46 ±1.48||7.44 ±1.46||0.89 ±0.90||−1.02 ±0.67||−0.13 ±0.91|
Refractive Error State Changes After Cycloplegia
|Group||Total (n)||Cycloplegia||Glasses Prescription|
|Remained Myopic (n [%])||Became Emmetropic or Hyperopic (n [%])||Remained Myopic (n [%])||Became Emmetropic or Hyperopic (n [%])|
|4 to 6 years|
| Mild myopia||905||339 (37.5%)||566 (62.5%)||622 (68.7%)||283 (31.3%)|
| Non-cycloplegic refractive error ≥ −1.00 D||629||88 (14%)||541 (86%)||353 (56.1%)||276 (43.9%)|
|7 to 10 years|
| Mild myopia||3,136||2,781 (88.7%)||355 (11.3%)||2931 (93.5%)||205 (6.5%)|
| Non-cycloplegic refractive error ≥ −1.00 D||764||450 (58.9%)||314 (41.1%)||600 (78.5%)||164 (21.5%)|