Association between Ocular Dominance and Spherical/Astigmatic Anisometropia, Age, and Sex: Analysis of 1274 Hyperopic Individuals

Abstract

Purpose.: To determine the association between ocular dominance and spherical/astigmatic anisometropia, age, and sex in hyperopic subjects.

Methods.: The medical records of 1274 hyperopic refractive surgery candidates were filtered. Ocular dominance was assessed with the hole-in-the-card test. Refractive error (manifest and cycloplegic) was measured in each subject and correlated to ocular dominance. Only subjects with corrected distance visual acuity of >20/22 in each eye were enrolled, to exclude amblyopia. Associations between ocular dominance and refractive state were analyzed by means of t-test, χ ² test, Spearman correlation, and multivariate logistic regression analysis.

Results.: Right and left eye ocular dominance was noted in 57.4 and 40.5% of the individuals. Nondominant eyes were more hyperopic (2.6 ± 1.27 diopters [D] vs. 2.35 ± 1.16 D; P < 0.001) and more astigmatic (−1.3 ± 1.3 D vs. −1.2 ± 1.2 D; P = 0.003) compared to dominant eyes. For spherical equivalent (SE) anisometropia of >2.5 D (n = 21), the nondominant eye was more hyperopic in 95.2% (SE 4.7 ± 1.4 D) compared to 4.8% (1.8 ± 0.94 D; P < 0.001) for the dominant eye being more hyperopic. For astigmatic anisometropia of >2.5 D (n = 27), the nondominant eye was more astigmatic in 89% (mean astigmatism −3.8 ± 1.1 D) compared to 11.1% (−1.4 ± 1.4 D; P < 0.001) for the dominant eye being more astigmatic.

Conclusions.: The present study is the first to show that the nondominant eye has a greater degree of hyperopia and astigmatism than the dominant eye in hyperopic subjects. The prevalence of the nondominant eye being more hyperopic and more astigmatic increases with increasing anisometropia.

Introduction

Functional lateralization occurs in the paired organs of the body, such as hands, legs, eyes, and cerebral hemispheres; but the exact mechanisms resulting in lateralization as well as the strength and quality of lateralization remain obscure. 

In 1903, Rosenbach claimed that most people have a dominant eye, even though each of their two eyes in isolation may provide equal vision and, in unequal vision, the dominant eye is not always the eye with better visual acuity. ¹  

In general, ocular dominance is defined as the tendency to prefer visual input from one eye over the other ^2,3 and may be further characterized as a status whereby one of the eyes commonly dominates or leads the other eye, both in fixation and attention and in perceptive function. ⁴  

Eye dominance, which is commonly determined with the hole-in-the-card test, provides the foundation for a range of clinical decisions, including contact lens wear, ^5,6 cataract surgery, ^7,8 and monovision presbyopic treatment, ^9–11 and is thought to be important in the control of reading. ¹²  

The role of ocular dominance in surgically induced monovision can be assessed by looking at success rates and patient satisfaction after monovision refractive surgery. ^13–15 This clinical practice is based on the assumption that it will be easier to suppress blur in the nondominant eye than in the dominant one. The surgically induced anisometropia should not exceed 2.5 diopters (D), ^11,16 with the dominant eye usually being corrected for distance and the nondominant eye for near vision. The concept of motor and sensory dominance has been developed through the years; however, it seems justified to state that both the importance and basis of eye dominance, be it motor or sensory, is still poorly understood. ¹⁷  

In contrast to previous reports, our recent study on 10,264 subjects reveals that the nondominant eye exhibits a greater myopic and astigmatic refractive error than the dominant eye in anisometropic myopic refractive surgery candidates. ¹⁸  

To the best of our knowledge, no study exists analyzing the association between ocular dominance and refractive state, age, and sex in hyperopic subjects. Therefore, the present study was initiated to explore whether nondominant eyes in anisometropic hyperopic subjects are more ametropic, paralleling and confirming the results in myopic individuals. 

Methods

Study Population and Clinical Measures

The study population comprised 1274 hyperopic individuals attending CARE Vision refractive clinics in Germany (Hamburg, Hannover, Berlin, Cologne, Stuttgart, Frankfurt, Nuremberg, Munich) and Austria (Vienna) between April 2006 and August 2010 for treatment of ametropias. The refractive clinics were not selected according to defined epidemiologic sampling criteria. This resulted in a selection bias with fewer preteen and teenage subjects (<18 years old) and older individuals (>65 years old). Most of the subjects were candidates to undergo refractive surgery with excimer laser, either LASIK or photorefractive keratectomy. Subjects exceeding the range for laser vision correction were candidates for phakic intraocular surgery or clear lens extraction. 

We undertook a detailed examination of the medical records of all cases. In addition to general medical and ophthalmic histories, preoperative measurements included uncorrected distance visual acuity, corrected distance visual acuity (CDVA), manifest refraction, cycloplegic refraction, tonometry, pachymetry, corneal tomography (Orbscan or Pentacam), slitlamp examination of the anterior segment, and funduscopy. We converted all refractive data to minus cylinder form to prevent confusion during analysis. To avoid any effect of amblyopia on ocular dominance, only subjects with CDVA ≥20/22 in each eye were included in the study. Subjects with any of the following diagnoses were excluded from the study: history of ocular surgery, including cataract or refractive surgery; ptosis; any clinically significant retinal pathology; glaucoma; optic neuropathy; optic disc anomalies or other diseases that might affect visual acuity; and the presence of marked facial asymmetry. The study protocol was conducted according to the tenets of the Declaration of Helsinki of the World Medical Association regarding scientific research on human subjects. Informed consent was obtained from the subjects after explanation of the nature and possible consequences of the study. The retrospective analysis of the data was approved by the local ethics committee. 

Ocular Dominance

To determine sighting ocular dominance or motor dominance, the hole-in-the-card test (Dolman method) was carried out. The patient holds a card with a hole in the middle using both hands and is asked to view (manifest refraction) a 6-m target through the hole in the card. The observer then occludes each eye alternately to establish which eye is aligned with the hole and the distance target. The selected eye is considered the dominant eye. The process is repeated; the second time, the subject moves the card slowly towards his face without losing the alignment with the fixation point until the hole is over an eye. This is considered to be the dominant eye. This test is a “forced choice” test of dominance, which allows only a right or left eye result. For some subjects, however, it may be difficult to decide which eye deviates first. If we did not observe a clear preference the ocular dominance was classified as “undetermined.” 

Statistical Analysis

After the data were compiled, they were entered into a spreadsheet program (Excel; Hamburg Refractive Data Base) and further statistically analyzed with SPSS software (version 15.0; SPSS, Inc., Chicago, IL). Both right and left eyes were included in statistical analyses. Arithmetic means and standard deviation were used for further descriptive comparison. 

The difference in refractive parameters (spherical equivalent [SE], astigmatism, power vector J0/J45) between dominant and nondominant eyes was compared with a paired Student's t-test and χ² test. For the power vector approach, manifest refractions in conventional script notation ([S]phere, [C]ylinder, axis [α]) were converted to power vector coordinates using the formulas: J0 = (−C/2) cos(2α) and J45 = (−C/2) sin(2α). The axis of the cylindrical component in different age groups was further classified as: with the rule (WTR) if the minus cylinder axis was within ±22.5 of 180°, against the rule (ATR) for minus cylinder axis within ±22.5 of 90°, or oblique (other than WTR or ATR). 

A P value below 0.05 was considered statistically significant. To control the inflation of the α error, P values (see Tables 2 and 4) were adjusted with the Bonferroni correction 1 − (1 − α) ^k . To determine the relationship between ocular dominance and the degree of anisometropic hyperopia, a graph was plotted with the average amount of hyperopia in the left and right eyes on the x-axis versus the amount of anisometropia on the y-axis. Anisometropia was further divided into four SE subgroups (<0.49; 0.5–1.74; 1.75–2.49; ≥2.5) and four astigmatism subgroups (<0.49; 0.5–1.74; 1.75–2.49; ≥2.5) to test if there was a correlation with the likelihood that the nondominant eye was more hyperopic/astigmatic. 

Table 1. 

 

Refractive Parameters in Dominant and Nondominant Eyes of Different SE Anisometropia Groups

	Dominant Eyes		Nondominant Eyes		P subj	P cyclo
	Subjective	Cycloplegia	Subjective	Cycloplegia
All subjects (n subj = 1170; n cyclo = 1274)
SE (D)	2.35 ± 1.16	2.66 ± 1.39	2.6 ± 1.27	2.95 ± 1.52	<0.001*†	<0.001*†
Astigmatism (D)	−1.2 ± 1.2	−1.2 ± 1.3	−1.3 ± 1.3	−1.4 ± 1.3	0.003*	0.003*
J0	0.04 ± 0.85	0.11 ± 0.87	0.09 ± 0.9	0.16 ± 0.9	0.237	0.198
J45	0.0005 ± 0.44	0.0053 ± 0.45	0.015 ± 0.49	0.011 ± 0.49	0.445	0.757
					P subj/P adj‡	P cyclo/P adj‡
Subjects with anisometropic hyperopia ≥2.5 (n subj = 21; n cyclo = 14)
SE (D)	1.8 ± 0.94	1.8 ± 0.98	4.7 ± 1.4	4.88 ± 0.99	<0.001*/<0.001*	<0.001*/<0.001*
Astigmatism (D)	−0.72 ± 0.79	−1.36 ± 0.98	−0.98 ± 0.8	−1.8 ± 1.3	0.295/0.753	0.311/0.775
J0	0.03 ± 0.6	0.42 ± 0.71	0.01 ± 0.49	0.38 ± 0.9	0.92/0.999	0.902/0.999
J45	−0.03 ± 0.14	−0.009 ± 0.32	0.13 ± 0.44	0.22 ± 0.54	0.142/0.458	0.189/0.567
Subjects with anisometropic hyperopia 1.75–2.49 (n subj = 38; n cyclo = 24)
SE (D)	1.8 ± 0.9	2.3 ± 1.3	3.6 ± 1.08	3.9 ± 1.5	<0.001*/<0.001*	<0.001*/<0.001*
Astigmatism (D)	−1.16 ± 1.32	−1.12 ± 1.37	−1.6 ± 1.4	−1.8 ± 1.3	0.107/0.364	0.082/0.289
J0	0.07 ± 0.92	0.17 ± 0.89	0.14 ± 1.07	0.37 ± 1.06	0.767/0.997	0.483/0.928
J45	0.13 ± 0.45	0.03 ± 0.5	−0.04 ± 0.52	−0.02 ± 0.45	0.13/0.427	0.723/0.994
Subjects with anisometropic hyperopia 0.5–1749 (n subj = 473; n cyclo = 270)
SE (D)	2.43 ± 1.13	2.86 ± 1.39	2.8 ± 1.23	3.4 ± 1.47	<0.001*/<0.001*	<0.001*/<0.001*
Astigmatism (D)	−1.26 ± 1.24	−1.43 ± 1.3	−1.5 ± 1.4	−1.6 ± 1.4	0.005*/0.019	0.156/0.492
J0	0.07 ± 0.86	0.196 ± 0.88	0.16 ± 0.9	0.26 ± 0.9	0.121/0.403	0.417/0.884
J45	−0.008 ± 0.46	−0.014 ± 0.54	0.008 ± 0.51	0.035 ± 0.52	0.619/0.979	0.285/0.739
Subjects with anisometropic hyperopia ≤0.5 (n subj = 638; n cyclo = 966)
SE (D)	2.3 ± 1.2	2.6 ± 1.4	2.3 ± 1.2	2.7 ± 1.5	0.556/0.961	0.021*/0.081
Astigmatism (D)	−1.26 ± 1.2	−1.23 ± 1.2	−1.2 ± 1.3	−1.4 ± 1.3	0.34/0.810	0.023*/0.089
J0	0.02 ± 0.86	0.09 ± 0.87	0.03 ± 0.87	0.12 ± 0.9	0.868/0.999	0.348/0.819
J45	−0.006 ± 0.44	0.01 ± 0.41	0.019 ± 0.47	0.002 ± 0.49	0.435/0.898	0.694/0.991

*  Tested with t-test.

†  For controlling the effect of astigmatism, ANCOVA analysis was performed. The covariate, astigmatism, was significantly related to the SE F_subj(1,2310) = 25.060, P < 0.001, r = 0.10. F_{cyclo(1,2545)} = 110.757, P < 0.001, r = 0.204. The significant effect of eye dominance on SE after controlling for effect of astigmatism was F_subj(1,2310) = 28.780, P = < 0.001, r = 0.11. F_{cyclo(1,2545)} = 33.418, P = < 0.001, r = 0.11.

‡  To control the inflation of α error, the P values were adjusted with Bonferroni correction 1 − (1 − α) ⁴ , where exponent 4 means four pairwise comparisons.

Table 2. 

 

Distribution of Dominant and Nondominant Eye Being More Hyperopic in Different SE Anisometropia Subgroups

Anisometropia (SE)	Dominant Eye More Hyperopic		Nondominant Eye More Hyperopic		Total		P *
	Number	%	Number	%	Number	%
≤0.49	222	42.69	298	57.31	520	49.43	<0.001†
0.5–1.74	142	30.02	331	69.98	473	44.96	0.003†
1.75–2.49	2	5.26	36	94.74	38	3.61	<0.001†
≥2.5	1	4.76	20	95.24	21	2.00	0.003†

*  Tested with χ ² test. The overall significance was 0.016 with Spearman correlation 0.088 (P = 0.011).

†  Significant at the 0.05 level.

Table 3. 

 

Refractive Parameters in Dominant and Nondominant Eyes of Different Astigmatism Anisometropia Groups

	Dominant Eyes		Nondominant Eyes		P subj/P adj	P cyclo/P adj
	Subjective	Cycloplegia	Subjective	Cycloplegia
All subjects (n subj = 1170; n cyclo = 1274)
SE (D)	2.35 ± 1.15	2.7 ± 1.4	2.6 ± 1.27	2.95 ± 1.5	<0.001	<0.001
Astigmatism (D)	−1.2 ± 1.2	−1.3 ± 1.3	−1.3 ± 1.3	−1.4 ± 1.3	0.003	0.003
J0	0.04 ± 0.85	0.11 ± 0.87	0.09 ± 0.9	0.16 ± 0.9	0.237	0.198
J45	0.0005 ± 0.44	0.0053 ± 0.45	0.015 ± 0.49	0.011 ± 0.49	0.445	0.757
Subjects with anisometropic hyperopia ≥2.5 (n subj = 27; n cyclo = 30)
SE (D)	1.6 ± 1.2	2.2 ± 1.4	1.9 ± 1.4	2.7 ± 1.8	0.291/0.747	0.258/0.697
Astigmatism (D)	−1.4 ± 1.4	−1.7 ± 1.5	−3.8 ± 1.1	−3.6 ± 1.4	<0.001*/<0.001*	<0.001*/<0.001*
J0	0.3 ± 0.7	0.47 ± 0.8	1.4 ± 0.9	1.39 ± 0.9	<0.001*/<0.001*	<0.001*/<0.001*
J45	0.15 ± 0.7	0.04 ± 0.68	−0.11 ± 1.1	−0.24 ± 1.01	0.289/0.744	0.216/0.622
Subjects with anisometropic hyperopia 1.75–2.49 (n subj = 29; n cyclo = 37)
SE (D)	1.9 ± 1.1	2.1 ± 1.5	2.3 ± 1.2	2.4 ± 1.6	0.24/0.666	0.515/0.945
Astigmatism (D)	−2.17 ± 1.4	−2.08 ± 1.5	−2.98 ± 1.4	−3.07 ± 1.3	0.036*/0.136	0.004*/0.016*
J0	0.7 ± 0.88	0.6 ± 0.9	0.93 ± 0.87	0.92 ± 0.98	0.325/0.792	0.119/0.397
J45	−0.05 ± 0.75	0.06 ± 0.74	0.09 ± 1.1	0.04 ± 0.99	0.631/0.981	0.935/0.999
Subjects with anisometropic hyperopia 0.5–1.749 (n subj = 473; n cyclo = 540)
SE (D)	2.38 ± 1.23	2.6 ± 1.5	2.7 ± 1.4	3.03 ± 1.6	<0.001*/<0.001*	<0.001*/<0.001*
Astigmatism (D)	−1.4 ± 1.3	−1.5 ± 1.3	−1.6 ± 1.3	−1.7 ± 1.3	0.044*/0.165	0.031*/0.118
J0	0.19 ± 0.9	0.24 ± 0.9	0.22 ± 0.9	0.27 ± 0.9	0.731/0.994	0.542/0.955
J45	−0.006 ± 0.48	0.007 ± 0.52	0.03 ± 0.52	0.03 ± 0.55	0.273/0.721	0.381/0.853
Subjects with anisometropic hyperopia ≤0.5 (n subj = 630; n cyclo = 667)
SE (D)	2.3 ± 1.1	2.7 ± 1.3	2.6 ± 1.2	2.9 ± 1.4	0.001*/0.004*	0.005*/0.019*
Astigmatism (D)	−0.98 ± 1.1	−1.04 ± 1.1	−0.99 ± 1.1	−1.05 ± 1.1	0.869/0.999	0.828/0.999
J0	−1.1 ± 0.8	−0.031 ± 0.83	−1.0 ± 0.8	−0.033 ± 0.82	0.878/0.999	0.961/0.999
J45	0.001 ± 0.38	0.005 ± 0.34	0.007 ± 0.36	0.006 ± 0.36	0.793/0.998	0.931/0.999

 For controlling the effect of astigmatism ANCOVA analysis was performed. The covariate, astigmatism, was significantly related to the SE F_subj(1,2310) = 25.060, P < 0.001, r = 0.10. F_{cyclo(1,2545)} = 110.757, P < 0.001, r = 0.204. The significant effect of eye dominance on SE after controlling for effect of astigmatism was F_subj(1,2310) = 28.780, P = < 0.001, r = 0.11. F_{cyclo(1,2545)} = 33.418, P = < 0.001, r = 0.11. To control the inflation of α error the P values were adjusted with Bonferroni correction 1 − (1 − α) ⁴ , where exponent 4 means four pairwise comparisons.

*   Tested with t-test.

Table 4. 

 

Distribution of Dominant and Nondominant Eye Being More Astigmatic in Different Cylindrical Power Anisometropia Subgroups

Anisometropia (Astigmatism)	Dominant Eye More Astigmatic		Nondominant Eye More Astigmatic		Total		P *
	Number	%	Number	%	Number	%
≤0.49	176	46.68	201	53.32	377	41.11	0.010†
0.5–1.74	195	40.29	289	59.71	484	52.78	0.374
1.75–2.49	8	27.59	21	72.41	29	3.16	0.118
≥2.5	3	11.11	24	88.89	27	2.94	0.001†

 Tested with χ ² test. The overall significance was 0.494.

*   The overall significance was 0.001 with Spearman correlation 0.109 (P = 0.001).

†   Significant at the 0.05 level.

An additional analysis of covariance (ANCOVA) was performed to control for confounding effects of astigmatism in anisometropia subgroups with a statistically significant effect of eye dominance on SE. The likelihood that the nondominant eye was more hyperopic as a function of the amount of anisometropia was assessed using multivariate logistic regression analysis. 

Finally, a forward stepwise logistic regression approach was applied to identify the relationship between eye dominance and the covariates SE and astigmatism in myopic and hyperopic refraction groups. The dominant eye was entered in the model as the reference category. 

Results

We enrolled 1274 hyperopic individuals who had complete data on ocular dominance and refraction into the statistical analysis. The mean age was 44.39 ± 11.7 years (range, 18–74 years), mean SE refraction was 2.49 ± 1.22, and mean astigmatism was −1.38 ± 1.4 D. There was no statistically significant difference in SE between right (2.5 ± 1.2 D) and left eyes (2.5 ± 1.2 D; P = 0.727). When ocular dominance was determined using the hole-in-the-card test, right eye ocular dominance was present in 57.4% (n = 686) and left eye ocular dominance in 40.5% (n = 484) of the candidates; 2.1% of the subjects (n = 25) were classified as “undetermined.” 

Ocular Dominance and SE Anisometropia

The difference in SE between the dominant eye (+2.35 ± 1.16 D) and the nondominant eye (+2.6 ± 1.27 D) was statistically significant (P < 0.001) for the whole study population. The average cylinder of the dominant eye was −1.2 ± 1.2 D and of the nondominant eye −1.3 ± 1.3 D, with the nondominant eye being more astigmatic (P = 0.003). For subjects with SE anisometropia (subjective refraction) <0.5 D, there was no significant difference in SE power (n = 638; for SE 2.3 ± 1.2; P _adj = 0.96), amount of astigmatism (P _adj = 0.81), and the mean values of the power vectors (for J0, P _adj = 0.99; for J45, P _adj = 0.89) between dominant and nondominant eyes. Calculations based on cyclorefraction revealed a trend of the nondominant eye being more hyperopic and more astigmatic; however, this trend was not statistically significant after Bonferroni adjustment (P _adj = 0.08). The results are summarized in Table 1. 

For subjects with SE anisometropia_subj of 0.5 to 1.749 D (n = 473), both SE (2.8 ± 1.2 vs. 2.3 ± 1.1; P _adj < 0.001) and amount of astigmatism (−1.5 ± 1.4 vs. −1.26 ± 1.2; P _adj = 0.019) were statistically significantly higher in nondominant eyes compared to dominant eyes, but no significant difference was seen regarding the mean values of J0 and J45 power vectors. Calculations based on cyclorefraction confirmed the nondominant eye to be more hyperopic (3.4 ± 1.5 vs. 2.86 ± 1.4; P _adj < 0.001), but the trend of the nondominant eye being more astigmatic (−1.6 ± 1.4 vs. −1.4 ± 1.3) was not statistically significant (P _adj = 0.49). 

Nondominant eyes were more hyperopic in all subgroups with SE anisometropia of >0.5 D, regardless of the method of refraction (manifest and cycloplegic). An additional ANCOVA analysis revealed that the observed effect of ocular dominance on SE (nondominant more hyperopic) persisted when the confounding effects of astigmatism on ocular dominance were taken into account. Multivariate logistic regression analysis showed that for subjects in whom SE anisometropia increased by 1.0 D, the odds ratio (OR) for the nondominant eye being more hyperopic was 2.026 (95% confidence interval: 1.69–2.428). 

For SE anisometropia <0.5 D (n = 520), the nondominant eye was more hyperopic in 57.3% of the subjects and the dominant eye was more hyperopic in 42.7% of subjects. The prevalence of the nondominant eye being more hyperopic increased steadily as the amount of anisometropia increased, as shown in Table 2. For anisometropia 0.5 to 1.74 D (n = 473), nondominant eyes were more hyperopic in 70% of subjects; and for anisometropia of >2.5 D (n = 21), nondominant eyes were more hyperopic in 95.2% of subjects. 

In contrast, the percentage of undetermined subjects decreased with increasing anisometropia; 2.4% of the subjects with anisometropia of <0.5 D did not demonstrate a clear ocular dominance (undetermined), whereas none of the subjects with anisometropia of >2.5 D was classified as undetermined. This observed tendency was not statistically significant (P = 0.27). 

Figure 1 shows the scatter plots of the average amount of hyperopia versus the amount of SE anisometropia for the dominant/nondominant eye being more hyperopic. For anisometropia of >2.5 D the nondominant eye was always more hyperopic. 

Figure 1. 

 

Scatter plots of the amount of anisometropia versus the average amount of hyperopia in each subject. The average amount of hyperopia was calculated as the mean SE of the right and left eyes. The dominant eye was more hyperopic in 367 (31.4%) subjects (circles), whereas the nondominant eye was more hyperopic in 685 (58.5%) subjects (crosses). Note that for SE anisometropia of >2.5 D (horizontal line) the nondominant eye was always more hyperopic.

Ocular Dominance and Astigmatic Anisometropia

In 53.9% of the subjects (n = 630), astigmatic power of the two eyes measured within <0.5 D of each other, and the interocular difference in astigmatic power between dominant (−0.98 ± 1.1 D) and nondominant eyes (−0.99 ± 1.1 D) was not statistically significant (P = 0.87; P _adj = 0.99). 

When studying the demographics and statistical distribution of refractive error in the human population, a full description of refractive state, including the magnitude and axis of astigmatism, is needed. Therefore, we decided to analyze refractive state using the power vector approach. There was again no significant difference in the mean values of the power vectors (J0 and J45) between dominant and nondominant eyes for astigmatic anisometropia of <0.5 D. Only SE showed a statistically significantly higher hyperopic value in nondominant (2.6 ± 1.2 D) compared to dominant eyes (2.3 ± 1.1 D; P _adj = 0.004). 

For subjects with astigmatic anisometropia of >1.75 D, there was no statistically significant difference in SE of nondominant and dominant eyes. Although nondominant eyes of anisometropic subjects with >0.5 D tended to be more astigmatic, this trend was statistically significant after Bonferroni adjustment only for anisometropia of >2.5 D with a mean astigmatic value in nondominant eyes of −3.8 ± 1.1 D compared to −1.4 ± 1.4 D in dominant eyes (P _adj < 0.001). The observed tendency of the nondominant eye to have a higher J0 power vector value became statistically significant in the >2.5 D anisometropia group (nondominant 1.4 ± 0.9 D vs. dominant 0.3 ± 0.7 D; P _adj < 0.001). The results for the analysis based on subjective refraction and cyclorefraction revealed similar trends and are summarized in Table 3. 

Figure 2 shows the scatter plots of the average amount of astigmatism versus the amount of astigmatic anisometropia for the dominant/nondominant eye being more astigmatic. For astigmatic anisometropia of >3.25 D (horizontal line) the nondominant eye was always more astigmatic. 

Figure 2. 

 

Relationship between interocular difference in astigmatism and mean astigmatism. The dominant eye was more astigmatic in 382 (32.6%) subjects (circles), whereas the nondominant eye was more astigmatic in 535 (45.7%) subjects (crosses).

The prevalence of the nondominant eye being more astigmatic increased with higher amounts of astigmatic anisometropia, as shown in Table 4. For anisometropia of 0.5 to 1.74 D (n = 484), nondominant eyes were more astigmatic in 59.7% of subjects; and for astigmatic anisometropia of >2.5 D (n = 27), nondominant eyes were more astigmatic in 88.9%. 

Ocular Dominance and Age

The SE and amount of astigmatism were statistically significantly higher in nondominant eyes of subjects younger than 29 years and 40 to 59 years of age. No statistically significant difference in astigmatic power vector values J0 and J45 between dominant and nondominant eyes was observed in the analyzed age groups, with only one exception: the values for J45 were higher in nondominant eyes for subjects aged 40 to 49 years when calculation was based on cycloplegic refraction (0.05 ± 0.49 vs. −0.007 ± 0.44; P = 0.047). For subjects older than 60 years, no statistically significant difference was found between dominant and nondominant eyes for any of the analyzed refractive parameters (SE, astigmatism, J0, and J45). 

No statistically significant difference was observed in each age group with regard to the distribution of WTR, ATR, and oblique astigmatism between dominant and nondominant eyes. The share of WTR astigmatism in dominant eyes decreased from 65.9% in subjects younger than 29 years to 29.5% in subjects older than 60 years, whereas the percentage of ATR astigmatism increased from 7.2% in subjects younger than 29 years to 46.6% in subjects older than 60 years. The summary of refractive parameters (subjective and cycloplegic) in dominant and nondominant eyes of different age groups is listed in Table 5. 

Table 5. 

 

Refractive Parameters in Dominant and Nondominant Eyes of Different Age Groups

	Dominant Eyes		Nondominant Eyes
	Subjective	Cycloplegia	Subjective	Cycloplegia	P subj*	P cyclo*
Subjects with age ≤29 (n subj = 179; n cyclo = 203)
SE (D)	2.4 ± 1.3	2.8 ± 1.6	2.7 ± 1.4	3.0 ± 1.6	0.036	0.048
Astigmatism (D)	−1.95 ± 1.4	−2.05 ± 1.3	−2.3 ± 1.5	−2.3 ± 1.4	0.038	0.05
J0	0.59 ± 0.8	0.65 ± 0.9	0.68 ± 0.93	0.71 ± 0.92	0.33	0.466
J45	0.04 ± 0.64	0.04 ± 0.6	−0.01 ± 0.7	−0.02 ± 0.68	0.551	0.353
WTR	118 (50.9%)	140 (50.7%)	114 (49.1%)	136 (49.3%)	0.658	0.670
ATR	13 (43.3%)	16 (43.2%)	17 (56.7%)	21 (56.8%)	0.445	0.389
Oblique	48 (50.0%)	47 (50.5%)	48 (50.0%)	46 (49.5%)	1.000	0.906
Subjects with age 30–39 (n subj = 161; n cyclo = 203)
SE (D)	2.5 ± 1.4	2.8 ± 1.8	2.8 ± 1.37	3.2 ± 1.9	0.052	0.103
Astigmatism (D)	−1.8 ± 1.4	−1.68 ± 1.29	−1.7 ± 1.3	−1.76 ± 1.26	0.895	0.541
J0	0.36 ± 0.9	0.35 ± 0.86	0.31 ± 0.9	0.35 ± 0.88	0.701	0.957
J45	0.04 ± 0.6	0.03 ± 0.56	0.015 ± 0.55	−0.004 ± 0.55	0.733	0.517
WTR	94 (51.4%)	123 (50.4%)	89 (48.6%)	121 (49.6%)	0.574	0.839
ATR	27 (50.0%)	31 (47.0%)	27 (50.0%)	35 (53.0%)	1.000	0.591
Oblique	40 (47.1%)	49 (51.0%)	45 (52.9%)	47 (49.0%)	0.527	0.815
Subjects with age 40–49 (n subj = 418; n cyclo = 460)
SE (D)	2.4 ± 1.2	2.8 ± 1.4	2.8 ± 1.2	3.1 ± 1.5	<0.001	0.001
Astigmatism (D)	−1.1 ± 1.2	−1.3 ± 1.3	−1.3 ± 1.3	−1.4 ± 1.4	0.041	0.061
J0	0.07 ± 0.8	0.16 ± 0.9	0.12 ± 0.9	0.20 ± 0.9	0.436	0.445
J45	−0.009 ± 0.42	−0.007 ± 0.44	0.05 ± 0.49	0.05 ± 0.49	0.054	0.047
WTR	219 (50.8%)	262 (51.4%)	212 (50.8%)	248 (48.6%)	0.628	0.353
ATR	96 (50.0%)	96 (50.3%)	96 (50.0%)	95 (49.7%)	1.000	0.935
Oblique	103 (48.4%)	102 (46.6%)	110 (51.6%)	117 (53.4%)	0.578	0.246
Subjects with age 50–59 (n subj = 324; n cyclo = 321)
SE (D)	2.17 ± 1.0	2.3 ± 1.0	2.4 ± 1.23	2.6 ± 1.29	0.007	0.002
Astigmatism (D)	−0.71 ± 0.76	−0.73 ± 0.8	−0.86 ± 0.92	−0.87 ± 0.98	0.019	0.046
J0	−0.3 ± 0.7	−0.27 ± 0.68	−0.2 ± 0.7	−0.21 ± 0.73	0.204	0.237
J45	−0.02 ± 0.26	−0.01 ± 0.26	−0.008 ± 0.37	−0.009 ± 0.38	0.561	0.904
WTR	135 (51.5%)	143 (52.8%)	127 (48.5%)	128 (47.2%)	0.522	0.231
ATR	120 (52.2%)	118 (53.9%)	110 (47.8%)	101 (46.1%)	0.412	0.157
Oblique	69 (44.2%)	60 (39.5%)	87 (55.8%)	92 (60.5%)	0.098	0.003
Subjects with age ≥60 (n subj = 88; n cyclo = 87)
SE (D)	2.4 ± 0.8	2.5 ± 0.8	2.5 ± 0.8	2.6 ± 0.8	0.254	0.297
Astigmatism (D)	−0.63 ± 0.61	−0.58 ± 0.5	−0.63 ± 0.47	−0.62 ± 0.5	0.973	0.609
J0	−0.5 ± 0.6	−0.5 ± 0.5	−0.5 ± 0.5	−0.5 ± 0.5	0.79	0.767
J45	−0.01 ± 0.22	−0.002 ± 0.21	−0.03 ± 0.24	−0.03 ± 0.21	0.631	0.386
WTR	26 (54.2%)	27 (50.0%)	22 (45.8%)	27 (50.0%)	0.498	1.000
ATR	41 (51.2%)	39 (51.3%)	39 (48.8%)	37 (48.7%)	0.762	0.760
Oblique	21 (43.8%)	21 (47.7%)	27 (56.2%)	23 (52.3%)	0.310	0.727

*   Mean differences were tested with t-test. Percentage differences were tested with χ ² test.

Ocular Dominance and Sex

There was an approximately equal number of male (n = 583) and female (n = 587) subjects. Men had a higher rate of right eye dominance (60.2%) than women (57.1%), but the difference between the sexes was not statistically significant (P = 0.276; χ²). 

Logistic Regression Model for Myopia and Hyperopia

Logistic regression was performed to identify the relationship between eye dominance and the covariates SE and astigmatism. The dominant eye was entered into the model as the reference category. Our previously analyzed myopic subjects (n = 10,264) ¹⁸ were compared to the hyperopic refraction group. Analyzing the myopic subjects within a forward stepwise model, we were able to detect a stronger effect of astigmatism in discriminating between the dominant and nondominant eye (Step 1: χ² score = 24.680; −2 Log likelihood = 27,654.073; b = −0.080; OR = 0.923; P < 0.001) compared to SE (Step 2: χ² score = 4.323; −2 Log likelihood = 27,649; b = −0.014; OR = 0.986; P = 0.038). According to the model, the chance of being the dominant eye increased as SE or astigmatism increased toward zero. The same model for hyperopic subjects showed a stronger discriminatory power of SE (Step 1: χ² score = 30.726; −2 Log likelihood = 3213.202; b = 0.189; OR = 1.208; P < 0.001) compared to astigmatism (Step 2: χ² score = 16.252; −2 Log likelihood = 3196.950; b = 0.216; OR = 1.241; P = 0.038). For hyperopic subjects, the increase in both SE and astigmatism reflects a better prediction of eyes being nondominant. 

Discussion

We present here, to the best of our knowledge, the first detailed and largest study on the association between ocular dominance and refractive asymmetry in hyperopic subjects. Our study cohort only comprises refractive surgery candidates who had decided to attend a laser clinic to correct refractive errors. The clinical selection of subjects is thus biased toward persons with refractive errors, mainly at the expense of emmetropes. 

Paralleling our recent findings in myopes (nondominant eyes were more myopic and more astigmatic than dominant eyes in anisometropic subjects), the present study showed that the nondominant eye has a greater degree of hyperopia and astigmatism than the dominant eye in anisometropic hyperopes. The prevalence of the nondominant eye being more hyperopic and more astigmatic increased with increasing anisometropia, again paralleling our observations in myopic refractive surgery candidates for whom the prevalence of the nondominant eye being more myopic and more astigmatic increased with increasing anisometropia. ¹⁸  

In summary, nondominant eyes showed a symmetrical V-shaped pattern with a higher refractive (SE and astigmatism) error compared to dominant eyes when diverging from zero to both the myopic and hyperopic direction. A logistic regression model was applied to identify the relationship between eye dominance and the covariates SE and astigmatism in both the myopic and hyperopic refraction groups. In our previous study, myopic subjects showed a stronger effect of astigmatism in discriminating between dominant and nondominant eye (OR = 0.923; P < 0.001), compared to SE (OR = 0.986; P = 0.038). According to this model the chance of being the dominant eye increases as the astigmatism or SE increases toward zero. Applying the same model to our hyperopic subjects revealed a stronger discriminatory power of SE (OR = 1.208; P < 0.001) compared to astigmatism (OR = 1.241; P = 0.038). Thus, in myopic subjects, astigmatism and, in hyperopic subjects, SE appear to be the strongest determinants on ocular dominance in our stepwise logistic regression model. 

In contrast to our results in myopic refractive surgery candidates, ¹⁸ with a higher rate of right eye dominance in males (70%) than females (65%), we did not find a statistically significant difference in right eye dominance between males (60.2%) and females (57.1%) (P = 0.28) in hyperopic subjects. 

The strengths of our study include a large sample size, the homogeneity of the method of refraction (subjective and cycloplegic refraction were available), and a strict exclusion of ocular pathologies and amblyopia as a result of a detailed ophthalmological examination of the subjects. The clinical selection of our cohort contains a bias, mainly at the expense of emmetropes, toward persons with refractive errors, as it only includes individuals who had elected to attend a laser clinic to correct these errors. 

Our study does contain some limitations and drawbacks. First, no relevant data were available about the hand dominance of the subjects in our study. However, Cheng et al. ¹⁹ and Mansour et al. ²⁰ find no correlation between handedness and refraction. Second, the reliability of ocular dominance tests in general should be addressed. Here we used the hole-in-the-card test since it is simple to use and only rarely produces undetermined results. ¹³ On the other hand, it is difficult to characterize the magnitude and reliability of ocular dominance with the hole-in-the-card test alone. To characterize ocular dominance further, Seijas et al. ¹³ recommend confirming ocular dominance with a second test or a panel of test methods to select those individuals with strong or clear dominance. 

Using a motor test, the extent of ocular dominance can only be indirectly approximated by retrieving the percentage of “undetermined” subjects. Cheng et al. ¹⁹ suggest that the extent of ocular dominance may vary among individuals, and that those with stronger ocular dominance might eventually develop higher amounts of anisometropia, whereas those with less dominance might not. 

In agreement, we observed a decreasing prevalence of undetermined subjects with increasing anisometropia: 2.4% undetermined subjects for anisometropia of <0.5 and 0% of undetermined subjects for anisometropia of >2.5 D. However, this trend was not statistically significant (P = 0.27). 

Since recent studies conclude that eye dominance is not determined by a more faithful input from one eye ^1,21 or more efficient cortical processing of one eye's input, ^22,23 there remains the possibility that its basis lies in the nature of the interaction that occurs between the eyes when both eyes are operating together (i.e., when both eyes are contributing to a fused, stable percept) as is the case in everyday viewing. ¹⁷ Although monovision creates an asymmetric visual correction with reduced binocular function, the reported subjective satisfaction with pseudophakic and Excimer induced monovision is high. ^8,15  

The role of ocular dominance in monovision success might be attributed to a fused and stable percept despite an anisometropic refractive state. Excimer monovision in hyperopic subjects adds to the overall laser correction and requires more tissue removal in nondominant eyes, which limits the amount of correction and may add to the unpredictability of the outcome. Since our study revealed a higher refractive error in nondominant eyes of hyperopic anisometropic subjects (for anisometropia of >2.5 D, nondominant eyes were more hyperopic in 95.2% of subjects), the number of potential hyperopic monovision candidates is restricted. Therefore, preoperative contact lens (CL) trials are strongly recommended in hyperopic laser-induced monovision, since a rejection of monovision and correction to distance vision would require reversing some of the original corrections in the near eye, causing decreased predictability and a possible increase in high order aberrations. ¹⁵ Braun et al. ²⁴ show that every hyperopic patient, regardless of ocular dominance or crossed versus conventional treatment, selected the more hyperopic eye for near vision correction. This strong sighting preference for distance vision with the clearer, less hyperopic eye may contribute to decreased interocular blur suppression and difficulty accepting monovision. According to our study, the vast majority (95.2%) of anisometropic hyperopes select the more ametropic eye to be the nondominant eye. 

In future studies, it may be more appropriate to carry out more than one test, or to develop novel methods, to evaluate the magnitude of ocular dominance, rather than just determining right or left eye ocular dominance. ¹³ Handa et al. provide some evidence of equivalence of sighting eye dominance, identified by the hole-in-the-card test, and sensory eye dominance, determined by binocular rivalry technique. ²⁵ In addition, functional imaging studies and correlation to sensory ocular dominance might help to further characterize the strength and complexity of motor ocular dominance. A concerted effort has been mounted to explain the development and function of ocular dominance columns. Although progress has been made over the decades, the riddles posed by ocular dominance columns continue to elude solutions. ²⁶  

Epidemiological studies like ours analyzing the association between ocular dominance and refractive state may help to identify contributing factors in the determination process of ocular dominance. 

The cross-sectional nature and selection of refractive surgery candidates precludes any definite conclusions on the causality or temporal relationship between ocular dominance and hyperopia in our study. 

Due to the increasing numbers of surgically induced monovision procedures performed in presbyopic patients, the role of ocular dominance and its central performance in monovision trials is gaining more attention. From the intervention point of view, Cheng et al. ¹⁹ surmise that it will be interesting to determine whether different treatment modalities (e.g., eye drops or surgical correction of refractive errors) in the dominant and nondominant eyes could effectively slow the progression of myopia in both eyes, particularly in patients with higher amounts of anisometropia. Future longitudinal studies are warranted to elucidate the mutual clinical impact of ocular dominance and refractive asymmetry. 

Conclusion

To the best of our knowledge, this is the first study on the association between ocular dominance and refractive asymmetry in hyperopic subjects. We have demonstrated that nondominant eyes exhibit a greater hyperopic SE power and a higher amount of astigmatism in subjects with anisometropia of >0.5 D. Supplementary information about the complexity and the influence of determinative factors on ocular dominance and the mutual impact of sensory and motor ocular dominance might be gained by implementing functional imaging techniques in characterizing ocular dominance. 

Acknowledgments

The authors thank the staff and patients of CARE Vision for their support in establishing the anonymized refractive data collection (Hamburg Refractive Data Base). The authors are particularly grateful to Vasyl Druchkiv for supporting and supervising the database and expert statistical analysis. 

Footnotes

 Disclosure: S.J. Linke, None; J. Baviera, None; G. Richard, None; T. Katz, None