## Purpose

To evaluate fusional convergence and associations with control in children with intermittent exotropia (XT).

## Design

Retrospective case series.

## Methods

Sixty-four children (median age 7 years) with intermittent XT were identified with measures of angle of deviation, control (using a previously reported 0-to-5-point scale), and fusional convergence at a single examination. Total convergence was defined as the sum of the angle of deviation plus additional convergence in reserve. Mean values were compared with published normals. A fusion reserve ratio was calculated as “convergence reserve divided by angle of deviation.” Relationships of control score with total amplitude, reserve, recovery, and fusion reserve ratio were assessed using Spearman rank correlations.

## Results

Convergence differed from normals in children with intermittent XT: total convergence was higher at distance (33 prism diopters [pd] vs 17 pd, *P *< .0001) and near (38 pd vs 18 pd, *P *< .0001) whereas convergence reserve was lower at distance (7 pd vs 17 pd; *P *< .0001). There was a strong correlation between fusion reserve ratio and control score at distance (R = −0.75, *P *< .0001) and near (R = −0.66, *P *< .0001).

## Conclusions

Children with intermittent XT have subnormal convergence reserves at distance. The fusion reserve ratio correlates well with control and may be useful in grading the severity of intermittent XT.

Assessing fusional vergences may be important in the clinical evaluation of ocular misalignment because it provides information regarding a patient’s ability to “control” a latent or intermittent deviation. In intermittent exotropia (XT), control of the exodeviation may rely on fusional convergence (positive fusional vergence). The ease with which a patient with intermittent XT maintains control of the exodeviation is currently used as an important indicator of severity and the finding of poor control is considered an indication for surgical intervention. Previous studies of intermittent XT have reported different methods for quantifying a patient’s ability to control, rating persistence of the exotropia or time to recovery of motor fusion. Nevertheless, there are few studies examining associations between fusional convergence and control of the exodeviation in intermittent XT. Convergence has been considered to be normal in children with intermittent XT, which would seem to be inconsistent with the hypothesis of a correlation between control and fusional convergence. In this present study, we investigate total fusional convergence amplitudes, fusional reserves, and fusion recovery in children with intermittent XT, and correlate these measures with exodeviation control.

## Methods

## Children With Intermittent Exotropia

Sixty-four children with intermittent XT (aged 3 to 17, median 7 years) and measures of angle of deviation, control, and fusional convergence at a single office examination were identified in the author’s clinical practices. Visual acuity was required to be 20/40 or better in each eye for inclusion (median 20/20, range 20/15 to 20/32 in each eye). We excluded children with convergence insufficiency–type intermittent exotropia (near angle more than 10 prism diopters greater than distance) and those with developmental delay or sensory or paralytic exotropia. Fifty-three (83%) were female, consistent with previously reported female predominance of intermittent exotropia, and race was reported as “white” for 51 (80%).

## Testing Procedures

All clinical testing was performed with refractive correction if prescribed (21 of 64 children, 33%), which in all cases was the child’s habitual spectacle correction; trial frames were not used. Tests were performed in the following order: control of the exodeviation, angle of deviation, and fusional convergence.

## Assessment of Control and Angle

Control of the exodeviation was quantified using a previously reported 0-to-5-point scale (best to worst control) at distance (3 meters) and then at near (1/3 meter). Levels 3 to 5 on the scale grade the proportion of time any spontaneous exotropia is present during a 30-second period of observation (5 = constant tropia) and levels 0 to 2 grade the speed of recovery after a 10-second dissociation (worst of 3 10-second dissociations; 0 = <1-second recovery). If control was measured more than once over the examination, the worst measure was used for analysis. Median control scores in our cohort of 64 children were 4 (range 0–5) at distance and 1 (range 0–5) at near.

The angle of deviation was assessed using prism and alternate cover testing at distance (3 meters) and then at near (1/3 meter) and recorded in prism diopters. The median angle of deviation by prism and alternating cover test was 25 pd (range 12–45 pd) at distance and 16 pd (range 0-40 pd) at near.

## Assessment of Fusion

Fusional convergence amplitudes were assessed using a 1- to 40-pd fixed horizontal prism bar. An accommodative target was used first at distance (3 meters) and then at near (1/3 meter). To determine the fusional convergence break point, base-out prism was gradually increased, starting with 1 pd, until an exotropia developed with no subsequent recovery of motor fusion. A short time was allowed for spontaneous recovery of fusion after an exotropia occurred. The fusional convergence break point was recorded as the base-out prism strength at which an exotropia developed and could not be re-fused. Base-out prism was then gradually decreased until the induced exotropia could be fused. The prism strength at which re-fusion occurred was recorded as the fusion recovery point. Since children with intermittent XT are often not aware of diplopia when exotropic, convergence break point and recovery point were confirmed by brief cover tests when necessary. Patients who remained fused at 40 pd (8 of 64 patients [13%], at near only) were arbitrarily assigned a break point of 45 pd for the purposes of analysis, but because it is not possible to determine the fusion recovery point for these patients, they were excluded from any analysis of recovery. If there was no ability to fuse the exodeviation at the commencement of fusional convergence testing, break point and recovery point were recorded as 0 (total n = 17: n = 12 at distance only, n = 3 at distance and near, and n = 2 at near only). All patients were able to demonstrate motor fusion at some point during the examination, but were not necessarily able to fuse at the time of fusional convergence testing.

For this present study of convergence in intermittent XT, we used the following terms and definitions: 1) Patients able to control or motor fuse the exodeviation in free space were assumed to have convergence fusion at least equal to the magnitude of the measured *angle of deviation *; 2) Additional convergence fusion (superimposed on ability to fuse the exodeviation) was measured by determining the fusion break point and termed c *onvergence reserve *; 3) The sum of the angle of deviation and the convergence reserve was termed the *total convergence amplitude *.

## Analysis

Mean and median total fusional convergence amplitude, fusional convergence reserve, and fusion recovery were calculated for children with intermittent XT. Values in children with intermittent XT were then compared to previously reported mean values for normal, nonstrabismic children using *t *tests. Because of low convergence reserves possibly confounding analysis of fusion recovery, we also analyzed recovery in a sub-group of patients with at least 10 pd of convergence fusion reserve. The proportion of subjects recovering fusion within 5 pd of the break point were calculated for distance and near. Fusion values for visually normal children were derived from published data from subjects with orthophoria or very small degrees of heterophoria and therefore the reported break point was used to represent both the total fusional amplitude and the fusion reserve.

For children with intermittent XT, correlations of control score with total convergence amplitudes, convergence reserve, and convergence recovery were calculated using Spearman rank correlations. To assess the effect of the underlying angle of deviation, a fusion reserve ratio was calculated as fusional convergence reserve divided by prism alternating cover test (eg, fusional convergence reserve = 20; angle of deviation = 15; fusion reserve ratio = 1.3). It has been suggested that the fusion in reserve should be twice the magnitude of the angle of deviation (Sheard’s criterion), which would correspond to a fusion reserve ratio of 2.0. Spearman rank correlations between fusion reserve ratio and control score were calculated and control scores for patients meeting Sheard’s criterion were evaluated. Correlations between angle of deviation and control score were also analyzed.

## Results

## Children With Intermittent Exotropia Compared With Visually Normal Children

At both distance and near, the mean total convergence amplitude was significantly greater for children with intermittent XT than for visually normal children (distance: 33 ± 11 pd vs 17 ± 7 pd, P < .0001; near: 38 ± 13 pd vs 18 ± 8, *P *< .0001, Figure 1 ). In contrast, mean convergence reserves were significantly lower at distance fixation for children with intermittent XT than for visually normal children (7 ± 8 pd vs 17 ± 7 pd, *P *< .001, Figure 2 ), but at near fixation there was no difference between children with intermittent XT and visually normal children (21 ± 14 pd vs 18 ± 8 pd, *P *= .054, Figure 2 ). Mean fusion recovery point was significantly lower at distance for intermittent XT children than for visually normal children (2 ± 5 pd vs 11 ± 6 pd, *P *< .0001, Figure 3 ) but there was no difference between children with intermittent XT and visually normal children for mean fusion recovery at near fixation (11 ± 10 vs 13 ± 6, *P *= .2, Figure 3 ). Of the 22 patients with >10 pd of convergence reserve at distance, only 8 (36%) recovered within 5 pd of the break point. At near, of the 41 with >10 pd of convergence reserve at near (and a measured recovery point), only 12 (29%) recovered fusion within 5 pd of the break point.

## Correlations Between Fusional Convergence and Control Score in Children With Intermittent Exotropia

There was a strong inverse correlation between convergence reserve and control score (ie, higher reserves associated with lower [better] control scores and vice versa) at distance (R = −0.75, *P *< .0001) and a moderate correlation at near (R = −0.57, *P *< .0001, Table ).

Total Fusion Amplitude | Fusion Reserve (Break Point) | Fusion Recovery | Fusion Reserve Ratio | |
---|---|---|---|---|

Distance control score | R = −0.54 ^{a }P < .0001 |
R = −0.75 P < .0001 |
R = −0.57 P < .0001 |
R = −0.75 P < .0001 |

Distance angle of deviation | R = 0.63 P < .0001 |
R = 0.005 P = .97 |
R = −0.21 P = .09 |
R = −0.17 P = .2 |

Near control score | R = −0.30 P = .02 |
R = −0.57 P < .0001 |
R = −0.58 P < .0001 |
R = −0.66 P < .0001 |

Near angle of deviation | R = 0.03 P = .8 |
R = −0.54 P = .0001 |
R = −0.46 P = .0003 |
R = −0.82 P < .0001 |