Postural Control in Nonamblyopic Children with Early-Onset Strabismus

Abstract

Purpose.: In healthy subjects, the postural stability in orthostatic position is better when fixating at near than at far. Increase in the convergence angle contributes to this effect. Children with strabismus present a deficit in vergence. We evaluated postural control in children with respect to the vergence angle as they fixated at different depths, thereby engaging in active vergence movements.

Methods.: A TechnoConcept platform was used to record the postural stability of 11 subjects (mean age 11.18 ± 4.02 years) with convergent strabismus and 13 (mean age 11.31 ± 3.54 years) with divergent strabismus in 3 conditions: fixation at 40 cm, at 2 m, and active vergence movements between 20 and 50 cm.

Results.: The mediolateral body sway decreased significantly with proximity for convergent strabismus (from 3.78–2.70 mm) but increased significantly for divergent strabismus (from 3.27–3.97). Relative to fixation, vergence eye movements resulted in a statistically significant increase in mediolateral body sway for convergent strabismus (3.55 vs. 2.70) and a decrease for divergent strabismus (3.11 vs. 3.97, P = 0.047). Vergence eye movements were associated with the least variance of speed (99 mm²/s² for convergent and 117 mm²/s² for divergent strabismus), so less energy was required to control body sway.

Conclusions.: The fixation depth at which postural stability is best is proximal for convergent strabismus and distal for divergent strabismus. Optimal postural stability might be mediated by preponderant eye movement signals related to the angle of strabismus. Reduction of variance of speed in the active vergence condition corroborates our hypothesis.

Introduction

To maintain the body in a stable position, the central nervous system uses multiple sensorimotor inputs that are at once visual, proprioceptive, vestibular, and somesthetic. Paulus et al. demonstrated that postural control improves while fixating at near rather than at far distances. ¹ They attributed what we henceforth will refer to as the stabilizing effect of proximity to the angular size of the retinal drift, such that at near distances the angular size of the retinal slip input resulting from body sway is higher than at far distances. The ensuing afferent signal feeds back into mechanisms of postural control potentially leading to an improvement in body stabilization. 

However, Kapoula and Lê showed that the stabilizing effect that results from the relative proximity of a fixation target also is due to oculomotor signals generated by the vergence angle itself. ² Indeed, the efferent and afferent signals related to the vergence angle are stronger at near distances as a consequence of the relative increase in ocular convergence required to fixate a proximal target as opposed to a distal one. In their study, convergent prisms were used to make the eyes converge while the subjects were fixating at far distances. Despite the conflict with the physical distance, convergence of the eyes resulted in an improvement in postural stability. This stabilizing effect of proximity is present in children ³ and in the elderly. ²  

Bucci et al. showed that children with a vergence deficit who suffered from vertigo still presented this stabilizing effect of proximity with respect to postural control even though they were less stable than control children for conditions at near and far distances. ⁴  

Strabismus is a pathology that affects 3% to 4% of children during the first six years of life. Strabismus eye surgery presently is the principal method of treatment. Children afflicted with this condition do not have single bifoveal binocular vision given that their vergence eye movements are slow and disrupted by multiple saccadic intrusions of poor accuracy. ⁵ One such deficit may be due to an inherent weakness in sensory input, particularly as related to the binocular disparity required for vergence. Studies of postural control in strabismic children are rather scarce and concerned primarily with the impact of eye surgery on body stability. Earlier studies conducted in Sweden have shown that postural stability is greater in children with esotropia than in children with exotropia ^6–8 ; however, these studies failed to include an analysis of body sway and did not control for the children's use of bifocal glasses. More recently, Legrand et al. recorded the postural stability of nine children with strabismus before and after eye surgery. ⁹ Their results indicate that postural stability worsens immediately following surgery (2 weeks) and, indeed, this finding is in keeping with an earlier study conducted by Matsuo et al. on children from 3 to 12 years of age. ¹⁰ Nevertheless, Legrand et al. showed that posture improved as early as eight weeks postoperatively. ⁹ As a secondary observation, the investigators reported the absence of a preoperative stabilizing effect of proximity following an hour of habituation with corrective prisms. Similarly, and regardless of whether or not corrective prisms were worn by the subjects, no postoperative stabilizing effect of proximity was observed for any of the conditions studied in combination with fixation at near and far distances. ¹¹ However, it should be noted that for both studies, the number of children tested was relatively small and, indeed, the strabismic groups themselves appeared clinically and etiologically heterogeneous. 

The first goal of our study was to examine the interaction between the postural stability of strabismic patients and shifts in gaze between distal and proximal targets considered as a case in point for the sustained control of vergence during fixations. The patients were tested preoperatively or postoperatively when their condition had relatively stabilized, that is the residual strabismus angle remained the same within this period or frequently a residual microtropia (≤6–8 diopters). To our knowledge, the effects of active vergence eye movements on posture have not yet been examined in strabismic subjects. Given that single binocular vision and vergence eye movements are deficient in this population, ⁵ a stabilizing effect of proximity during fixation at near distances may not always exist in strabismus. Given that vergence has a potentially important role in the stabilizing effect of proximity on postural control, how would abnormal vergence in strabismic patients affect postural control? It should be recalled that vergence is complex and driven by multiple stimuli, which include binocular disparity, the accommodation of the eyes vis-à-vis proximal high level cues and monocular depth cues. The deficit of vergence control in strabismus could be related primarily to the disparity-driven component, but in cases of nonaccommodative strabismus, the accommodative vergence drive may be functioning properly. Given the multitude of cues necessary for vergence, residual vergence abilities certainly are present in all strabismic patients. Even the disparity-driven component can be active based on gross perifoveal or peripheral binocular vision. ¹²  

The second goal of our study was to test if active vergence movements along the median plane can improve postural stability. As mentioned above, Kapoula and Bucci confirmed that for dyslexic children, active vergence eye movements exert a stabilizing effect on posture that has not been observed in non-dyslexic control children. ³ They attributed these results to dyslexic children's greater visual attentional involvement, as well as to the mobilization of attentional resources together with the execution of eye movements. Because dyslexic children also have attentional difficulties, active vergence movements could help to focus their attention by shifting between distal and proximal targets. We refer here to the premotor theory of visual attention according to which eye movements and attention operate in tandem, and therefore are intimately connected and coconstrained. ¹³ Insofar as children with strabismus do not suffer from any specific, associated attentional disorder, we expect them to perform as would children without attentional deficit disorders, that is without gaining any benefit in terms of postural control from active vergence eye movements. We cannot, however, entirely discount the putative postural benefits gained from active eye movements themselves, particularly vergence eye movements, which produce massively changing proprioceptive and efferent vergence signals, and therefore which conceivably may contribute to better postural control. 

Methods

Subjects

A total of 24 adolescents and pre-adolescents with strabismus participated in this study (age range 6–17 years, mean age 11.25 ± 3.69 years). Of the patients 11 presented with convergent strabismus and 13 with divergent strabismus. They were recruited at Necker Hospital's ophthalmology service during their regular follow-up visit. All demonstrated best corrected visual acuity superior to 8/10 for each eye, measured with the Parinaud test for near vision (33 cm) or with the Pigassou drawing test for young children who did not read; for far vision (5 m), the Monoyer scale was used. It is important to emphasize that most of the participants had received regular follow-up examinations at the ophthalmology service since their early childhood to take prophylactic measures against amblyopia. Consequently, almost all of the children had good visual acuity in both eyes. The investigation adhered to the principles of the Declaration of Helsinki and was approved by our institutional human experimentation committee. Informed parental consent was obtained for each subject after the nature of the experimentation had been explained. 

Clinical Characteristics

Clinical characteristics are shown in Table 1 for each subject. Eleven children did not undergo surgery (S1, S4, S10, S12, S13, S15, S18, S22, S24, S30, and S32); seven of them had some stereoacuity (S1, S4, S10, S13, S18, S22, and S30) measured with the TNO test, and six of them had divergent intermittent strabismus. A total of 13 children underwent strabismus surgery, two of them for the second time (S16, S26). We tested all children at least one month after surgery so that the strabismus angle was stable and so that all children adapted to the new ocular deviation. Among the postoperative cases, 9 subjects had the same type of strabismus before and after surgery, five convergent (S6, S14, S21, S26, and S33) and four divergent (S9, S16, S28, and S29). Thus, three of the subjects differed in the type of strabismic condition presented before and after the surgery: S3 and S11 had convergent strabismus before and divergent strabismus after surgery, and S31 had divergent strabismus before and convergent strabismus after surgery. Also, one subject (S5) had preoperative convergent strabismus and postoperative esotropia upon fixation at far distances, and exotropia upon fixation at near distances. Five of them presented some stereoacuity (S14, S16, S28, S29, and S33). For all but two children (S9 and S15) deviation depended on viewing distance. All subjects had good visual acuity in both eyes, except subject 1. During the experiment subjects wore their habitual corrective lenses. 

Table 1. 

 

Clinical Characteristics of Children with Strabismus

Subjects	Age, y	Visual Acuity	Dominant Eye	Refraction		Angle of Strabismus (prism D)	Stereo Acuity	Surgery
				RE	LE
1	6	RE 10/10 LE 10/10	RE	1.00	1.5	ET 45 far ET 55 near		Pre
2	7	RE 10/10 LE 10/10	LE	1.25 (+0.25)	1.5 (+0.5)	XT 35 far IXT 16 near	40″	Pre
3	7	RE 9/10 LE 10/10	LE	4.5 (−1)	4 (−1.5)	XT 10 far XT 4 near		Pre
4	7	RE 10/10 LE 10/10	LE	1.50	0.75	ET 65 far ET 65 near		Pre
5	7	RE 10/10 LE 8/10	RE	4.00	3.50	XT 20 far IXT 45-50 near	100″	Pre
6	7	RE 8-9/10 LE 9/10	RE			ET 35 far ET 55 near	100″	Pre
7	8	RE 9/10 LE 10/10	LE	1.25	1.5	ET 1 far ET 4 near		Post
8	9	RE 9/10 LE 9/10	RE			XT 50 far IXT 35 near	40″	Pre
9	10	RE 10/10 LE 5/10	RE	−1.25	−2.00	ET 25 far ET 20 near		Post
10	10	RE 10/10 LE 9/10	RE	−1.00	2.25 (+0.75)	ET14 far ET 18 near		Post (2)
11	10	RE 10/10 LE 10/10	RE	3.25 (−2.25)	3.25 ( −2.5)	IXT 4 far IXT 6 near	40″	Post
12	10	RE 10/10 LE 10/10	RE	0.75	0.50	IXT 14 far IXT 16 near	40″	Post
13	11	RE 10/10 LE 10/10	RE	2.25 (−1.75)	2.75 (−2.75)	XT 18 far IXT 10-20 near	40″	Post (2)
14	12	RE 10/10 LE 10/10	LE	−1.50	−1.50	IXT 45 far IXT 50 near	240″	Pre
15	12	RE 10/10 LE 10/10	RE	2.75	2.50	X16 far IXT 40 near	40″	Pre
16	13	RE 10/10 LE 8/10	RE	−1.00	−5.5	ET 6 far XT4 near		Post
17	13	RE 10/10 LE 9/10	RE	2 (+0.5)	2 (+0.5)	ET 10 far ET 35 near		Pre
18	15	RE 10/10 LE 10/10	RE	−0.75	−0.50	XT 2 far XT 2 near	100″	Post
19	15	RE 10/10 LE 10/10	LE	4.25	5.75	XT 8 far XT 10 near		Post
20	16	RE 10/10 LE 8/10	RE	2.5	2.5	XT 2 far XT 10 near		Post
21	16	RE 10/10 LE 10/10	RE	−1.25 (+1.75)	1.5 (+2.25)	ET 8 far ET 14 near		Post
22	16	RE 10/10 LE 10/10	RE	0.50	0.50	ET 2 far ET 4 near		Post
23	17	RE 8/10 LE 4/10	RE			XT 35 far IXT 40 near	400″	Pre
24	17	RE 10/10 LE 10/10	RE	5.75	3.25	E 10 far E 12 near	200″	Post

 All subjects had a good visual acuity in both eyes. Stereoacuity was assesssed with the TNO test. XT, exotropia; IXT, intermittent exotropia; ET, esotropia; RE, right eye; LE, left eye.

Posturography

To measure postural stability, we used a force platform (principle of stain gauge) composed of two dynamometric clogs (Standards by Association Française de Posturologie; Technoconcept, Céreste, France). The excursions of the center of pressure (CoP) were measured during 25.6 seconds; the equipment contained an analog to digital converter of 16 bits. The sampling frequency of the CoP was 40 Hz. 

Procedure

During all posturographic tests, subjects were required to fixate a target (a smiling face, 1° size) placed at eye level in a lighted room. The subjects were required to stand upright and barefoot, with their feet placed side-by-side forming a 30° angle and with their heels separated by 4 cm. The subjects wore their habitual spectacle correction and, when asked, reported clear vision of the target for all conditions. The subjects were asked to maintain quiet stance while fixating the smiling face (i.e., arms held side-by-side, silent with teeth unclenched and normal breathing patterns maintained throughout). Three conditions were performed: fixation of a target at 40 cm, fixation of a target at 200 cm, and active vergence movement between a distal (50 cm) and proximal (20 cm) target (Fig. 1). Insofar as the latter condition is concerned, a ruler was mounted in front of the child along the median plane and the child was asked to alternate fixation between the two markers of the ruler, one at 20 cm and the other at 50 cm. Sustained fixation called for a convergence angle of 9.23° at 40 cm, and 1.86° at 200 cm. Active vergence movements called for a change in the vergence angle of between 18° and 7.40°, and an amplitude of vergence equal to 10.6°. These values are theoretical and indicative given the difference in head size and interpupillary distance that would occur naturally in a population of 6- to 17-year-olds. Subjects were placed carefully in front of the ruler centered along the midline to produce symmetric vergence eye movements. However, vergence eye movements were not measured in this study. 

Figure 1. 

 

Illustration of posturography testing conditions. The subject viewed the smiling face on the screen at eye level, and from distances of 40 and 200 cm, or viewed the ladybug magnet at eye level from 20 to 50 cm away.

Postural Parameters

Postural stability was measured objectively by the size of the surface area of CoP (in mm²) that contains 90% of the closest CoP positions from the central ones; the SD of the lateral and anteroposterior sway (respectively, SdX and SdY, in mm), and the variance of speed of the CoP excursions (in mm²/s²). Surface, SD of mediolateral and anteroposterior body sway values describe postural stability. Lateral and anteroposterior body sways are thought to be related to different muscular strategies. ¹⁴ Indeed, during orthostatic quiet standing, in the frontal plane, a parallelogram is formed between the two sides of the hip and the two ankles. The hip articulation creates an abductor/adductor couple that mostly controls the lateral body sway. In the anteroposterior plane, the postural stability essentially is controlled by the ankle, even though there are two levels of control (hip and ankle). As for the variance of speed parameter, it is believed to reflect the energy needed to maintain postural stability. ¹⁵  

Statistical Analysis

A two-way mixed ANOVA design was performed with the viewing condition at 40 and 200 cm, and the vergence movement as the main factors, and the type of strabismus (convergent or divergent) as the intersubject factor. The post-hoc comparison was done with the LSD test; the effect of a factor was significant when the P value was below 0.05. 

Results

Table 2 shows for each viewing condition the groups' mean values for posturography parameters (surface of CoP, SD of lateral [SdX] and anteroposterior [SdY] sway and the variance of speed). Although an ANOVA was done for all postural parameters (surface, SD of mediolateral and anteroposterior body sway, and variance of speed), significant effects were found only for a few parameters. We delineated these significant observed effects in what follows before demonstrating in the additional “results” section the correlation between postural parameters and angle of strabismus on the one hand, and postural parameters and age on the other. 

Table 2. 

 

Postural Stability Measurements in Quiet Stance

	SdX	SdY	Surface	Variance of Speed
NEAR
Convergent strabismus
Mean	2.70*	4.41	152	136
SE	0.40	0.75	34	63
Divergent strabismus
Mean	3.97	5.32	210	136
SE	0.60	0.62	38	44
FAR
Convergent strabismus
Mean	3.78*	5.43	221	195
SE	0.43	0.89	38	64
Divergent strabismus
Mean	3.27	4.72	237	165
SE	0.59	0.45	54	55
VERGENCE
Convergent strabismus
Mean	3.55	5.60	266	99
SE	0.72	1.05	81	22
Divergent strabismus
Mean	3.11	5.37	233	117
SE	0.48	0.67	51	28

 Means and SEs of SDs of lateral (SdX) and of anteroposterior (SdY) body sway, surface of CoP, and variance of speed for each condition, that is fixation at near (40 cm), fixation at far (200 cm), and active vergence movement (between 20 and 50 cm) for convergent and divergent strabismic children.

*  Statistically significant differences.

Mediolateral Body Sway

There was no main effect for the condition (F _(2,42) = 0.15, P = 0.86), neither was there an effect with respect to the type of strabismus (F _(1,21) = 0.028, P = 0.87), but there was an interaction between the condition and the type of strabismus (F _(2,42) = 3.30, P = 0.047, Fig. 2A). Figure 2A shows that the mediolateral body sway for convergent strabismus was lower at near than at far distances, while the opposite was true for divergent strabismus. It also should be noted that, where convergent strabismus was concerned, the mediolateral body sway increased for the distal fixation condition relative to the near fixation condition during active vergence. In contrast, for divergent strabismic children, the mediolateral body sway decreased significantly during active vergence and at fixation at far distances relative to the near fixation condition. 

Figure 2. 

 

Effect of visual condition on postural parameters. Means of the SD of lateral (SdX) body sway (A) and of the variance of speed (B) for convergent and divergent strabismic adolescents. Error bars: SE. Asterisks indicate significant differences (P <0.05).

The post-hoc test shows a significant difference for the near fixation condition as the children with divergent strabismus were significantly more unstable than children with convergent strabismus (P = 0.027). Thus, the effect of a given condition on postural control, be it fixation at far distances, fixation at near distances, or vergence eye movements, depends on the specific nature of the child's strabismic condition. 

Variance of Speed

A marginal effect was observed for all conditions relative to the variance of speed (F _(2,42) = 2.97, P = 0.06, Fig. 2B). The strabismic children invariably exhibited a larger variance of speed during the fixation at far distance condition than during either the active vergence or fixation at near distance conditions. 

Additional Results

In this section, we will examine the effects of the angle of strabismus and of age on different postural parameters. The significant effects are presented below: 

Correlation between the Anteroposterior Body Sway and the Angle of Strabismus.

For either viewing distance and for all children, the anteroposterior body sway increased with the angle of strabismus (absolute value), that is convergent or divergent (Fig. 3). The larger the angle of strabismus, the more difficult it was to maintain small body sway in the anteroposterior axis. No other parameter for postural control (surface, SdX, SdY, and variance of speed) was correlated with the strabismus angle. 

Figure 3. 

 

Effect of the angle of strabismus on postural parameters. Correlation between the angle of strabismus and the SD of anteroposterior (SdY) body sway at far. P = 0.05 indicates significant correlation.

Correlation between the Postural Parameters and Age.

The anteroposterior body sway also decreased with age for all children with strabismus; note, however, that negative correlations of this kind were found only for fixation at far distances and not for fixation at near distances (Fig. 4A). Figure 4A also shows that the youngest children had variable data, but after age 7 years, the variance was the same for all. Insofar as the variance of speed decreased with age for near and far fixation distances, we can adduce that with age, strabismic children used less energy to stabilize their bodies (Fig. 4B). 

Figure 4. 

 

Effect of the age on postural parameters. Correlation between the age and the SD of anteroposterior (SdY) body sway (A) near and far, and correlation between the age and the variance of speed (B) at near and at far. P = 0.05 indicates significant correlation.

Comparison with Non-Strabismic Children.

One also might ask whether postural performances in general were of lesser quality in strabismic children than in non-strabismic children. Control subjects were not examined in this study; however, we performed a comparison with age-matched children coming from another study using the same methodology. ⁵ The results showed no difference for any of the postural stability parameters (surface, SdX, SdY). The only differences found were in terms of the variance of speed (t _(37,1) = 1.99 at near and t _(37,1) = 2.59 at far distances for controls versus convergent strabismus and t _(39,1) = 2.1 at near distances and t _(39,1) = 1.18 at far distances for controls versus divergent strabismus). Thus, more energy was required to maintain postural control in strabismic, nonamblyopic children than in non-strabismic children. 

In summary, the main results include the following: (1) The SD of lateral body sway decreased with distance for divergent strabismus, but increased for convergent strabismus. (2) Relative to near fixation, the SD of lateral body sway decreased with active vergence movements for divergent strabismus, but increased for convergent strabismus. (3) Regardless of the strabismic disorder in question, the variance of speed was greater when fixating at far distances than it was when fixating at either near distances or when engaging in active vergence. (4) The anteroposterior body sway increased with the size of the strabismus angle. (5) The anteroposterior body sway and the variance of speed decreased with age. These results will be discussed below. 

Discussion

The main goal of our study was to determine the postural stability of children with strabismus, either convergent or divergent, while fixating at different sustained vergence angles or while actively converging and diverging between different depths. 

Distance Effect-Fixation Far versus Near

Our findings highlight the significant and rather complex effect of proximity on postural control in strabismic children. Children with convergent strabismus are more stable (in terms of mediolateral body sway) when fixating at near distances than when fixating at far distances; however, the pattern is reversed in children with divergent strabismus. In other words, only children with convergent strabismus, as opposed to children with divergent strabismus, behave as did the controls we investigated in a prior study. ⁵ Although the above observations merit further confirmation in conjunction with a larger pool of strabismic children, we proposed the following explanation: Given the behavior of convergent strabismus at far distances, the angle of vergence is particularly inappropriate. Conversely, the angle of vergence for divergent strabismus is inappropriate at near distances. Thus, strabismic children show less postural stability while fixating at the depth for which they have difficulty adopting the appropriate vergence angle. Convergence for divergent strabismus and divergence for convergent strabismus correspond to the depth at which their postural stability is difficult to achieve in terms of mediolateral body sway. Our results are not in keeping with those reported by Legrand et al., who did not observe distance effects of any kind. ¹¹ However, it should be noted that their study subject population was comparatively small with groups of convergent and divergent strabismic children represented unequally. Moreover, the SD for body sway was not included in their analysis. Our data consolidated further previous research and demonstrated the impact of vergence angle control on postural stability. Bucci et al. reported a convergent drift of the eyes during fixation for convergent strabismus and a divergent drift of the eyes for divergent strabismus. ¹⁶ These observations indicate continuous strong innervations, and related efferent and proprioceptive ocular motor signals directed toward the angle of strabismus. The postural data demonstrate how body control during quiet stance can betray a bias corresponding to the nature of strabismus. For fixation at near, subjects with convergent strabismus would put less effort and vice versa, and this would lead to the improvement in postural control observed. The explanation we proposed is in term of motor effort regardless of presence or absence of stereoacuity. Indeed, a further analysis dividing the subjects in two subgroups, with and without stereoacuity, shows no main effect of stereoacuity (F _(1,19) = 0.54, P = 0.47) nor interaction with the other factors (F _(1,19) = 0.61, P = 0.44). This analysis was done on the SdX and variance of speed parameters. Perhaps this bias originates from the continual action of privileged proprioceptive and efferent inputs, as well as from convergent versus divergent motor commands acting on the postural control system. Therefore, we are led to conclude that the privileged space for optimal postural control in strabismic patients corresponds to proximal space for convergent strabismus and distal space for divergent strabismus. Recall that convergent and divergent oculomotor systems enjoy partial, physiologic independence, such that the cell populations of the mesencephalic reticular formation that fire during convergence and divergence are distinct from each other. ¹⁷ In other words, esotropic subjects use predominantly convergent oculomotor signals, and exotropic subjects use divergent oculomotor signals to control their body sway in the mediolateral direction. It is important to note that the difference between convergent and divergent strabismus is relevant only for lateral body sway, which is believed to reflect a hip control strategy distinctive from the ankle strategy subtending anteroposterior body sway. ¹⁴ Legrand et al. reported a change in postural control that depended on the amount of time that had elapsed since surgery. ¹¹ In our study, all of the children tested were given sufficient time to adjust to the new postoperative angle of strabismus, which had relatively stabilized following surgery, as well as to all of the adaptive mechanisms required. Even though changes in strabismus angle can occur beyond one month after surgery, at the time of our test all but 4 subjects presented a strabismus angle of the same type as before surgery (see Methods). Further studies with larger populations preoperatively convergent and/or divergent would be of interest to confirm our observation. 

In conclusion, strabismic children do have a privileged distance for optimal postural control, which is proximal for esotropia and distal for exotropia, that is equal in both instances to the distance for which the vergence angle corresponds to the strabismus angle. 

Active Vergence Movement versus Sustained Fixation

The postural results obtained for the active vergence condition are indistinctive from those obtained for the fixation at far distance condition; however, they do provide significant differences relative to the fixation at near distance condition: divergent strabismus improved during active vergence whereas convergent strabismus deteriorated further (Fig. 2). These differences appear once again with respect to mediolateral body sway. Importantly, the vergence eye movement condition also produced a marginal effect on the variance of speed parameter. Regardless of a given child's strabismic condition, be it divergent or convergent, the variance of speed increased during the fixation at far distance condition; however, a decrease in the variance of speed was noted during the active vergence movement condition relative to the sustained fixation at near distance condition. Thus, the condition providing the smallest variance of speed is the active vergence condition for both groups of children. The variance of speed is believed to reflect the energy needed to stabilize the body. ¹⁵ These results illustrate that no matter the strabismic condition, children use less energy to control their body sway when actively engaging in vergence eye movements than when fixating at near and far distances, with fixation at far distances being the most energy consumptive. These results are in line with previous work on non-strabismic children. ^3,4 Thus, active vergence movements can be particularly useful for actively maintaining postural stability with less effort. This observation highlights the importance of vergence eye movements, even when deficient. Perhaps it is essential that the eyes be kept in motion to minimize the effort required to stabilize the body. Another study has shown that saccadic eye movements can reduce anteroposterior body sway in healthy young adults. ¹⁸ Thus, active eye movements potentially could prove beneficial and translate into an improvement in postural control. In a recent study dealing with dyslexic and non-dyslexic children, Kapoula and Bucci postulated the existence of an underlying attentional mechanism that is activated by eye movements and that also might contribute to improved postural performance in dyslexics. ³ A similar attentional mechanism, observable only in terms of the variance of speed parameter, also might be at work in strabismic patients, albeit in a more subtle way. 

Effect of the Angle of Strabismus and of the Age

Considering convergent and divergent strabismus together, the absolute value of the angle of strabismus was correlated positively with the anteroposterior body sway. The larger the angle of strabismus, the greater the anteroposterior body sway. Note also that the anteroposterior body sway decreased with age and, since all of the subjects had either convergent or divergent strabismus, the presence of strabismus cannot account for the change in behavior with increasing age. Two other explanations are more likely: that older children are able to focus more on the task with less postural instability, or that the control of posture matures or improves with age via a better multisensory integration. 

Finally, the variance of speed, which is another parameter subject to developmental change, also decreased with age. Our observations are in line with a study conducted by Rival et al., who demonstrated a linear decrease in the variance of speed parameters for children 6 to 10 years of age during a quiet stance task that was performed with eyes closed. ¹⁹  

Conclusions

In conclusion, our study aimed to evaluate the quality of postural stability in strabismic, nonamblyopic children. Emphasis was placed on a comparison between convergent and divergent strabismus in relation to the quality of postural control achieved while the subjects either fixated at near and far distances or engaged in vergence eye movements in depth. Children with strabismus demonstrated a quality of postural control comparable to that of controls, as represented in the literature; however, an increase in energy expenditure was required of the strabismic child to achieve such results. 

Importantly, the stabilizing effect of proximity as observed and reported in our study and as confirmed in the literature is specific to convergent strabismus, while the opposite pattern is seen for divergent strabismus. Finally, active vergence movements represented the condition for which all strabismic children control their posture with the least effort. Anteroposterior body sway and variance of speed decreased with age, but anteroposterior body sway increased with the size of strabismus. 

Acknowledgments

Gabiola Lipede helped correct our English. 

Footnotes

 Disclosure: C. Gaertner, None; C. Creux, None; M.-A. Espinasse-Berrod, None; C. Orssaud, None; J.-L. Dufier, None; Z. Kapoula, None