Binocular Vision &

Eye Muscle Surgery Qtrly°

Major Review: Intermittent Exotropia; Basic and Divergence Excess Type

Summer of 1993 Volume 8 (No.3): 185-216

in DEX(T), i.e., 4/1 (38). Basic X(T)s are thought to be affected equally at distance and near by occlusion and usually demonstrate minimal effects. Thus, slow vergence mechanisms are weaker in basic X(1) as opposed to DEX(T).

Niederker & Scott (57) occluded a group of X(T)s, to determine the minimal length of time which would elicit the maximum deviation. Occlusion was for 45 minutes, 1 day, and 5 days. They reported that 60% of the X(T)s increased the angle of deviation after occlusion. This is similar to the percentage reported by Burian & Franceschetti (24). They found no difference between 45 minutes and 1 day, occlusion; however, five days occlusion increased the deviation in some and decreased it others. Small vertical deviations (2-3d) not present before occlusion were present in 20% of the patients after occlusion. Five days of occlusion increased the vertical deviation to 4-5k.

The application of prism to neutralize the angle, followed and by remeasurement 30 minutes later, and reneutralization of the angle with additional prism is known as the prism adaptation test (PAT). The PAT elicits results similar to those seen with occlusion (58,59). This is expected since prism neutralization, like occlusion, decreases the disparity vergence signal, thus eliminating slow vergence or vergence aftereffects. Since prism and occlusion produce similar results in DE they may be used interchangeably.

Brown (33) suggested that +3.00 D lenses interposed at near could be used to differentiate simulated from true DEX(T). He suggested that +3.00 D would increase the near deviation in simulated DEX(T), indicating the presence of a latent deviation, while there would be no change in the deviation in true DEX(T). However, both von Noorden (16) and Cooper (1), and others, have previously pointed out that a significant change in the near angle with +3.00 D OU is due to. relaxation of accommodative convergence secondary to a high AC/A ratio. High AC/As are “more” consistent ‘with true DEX(T) compared to simulated DEX(T), since simulated DEX(T) has normal distance near AC/A ratios after occlusion (40).

If one performs both monocular occlusion tests and +3.00 D lens tests, most X(T)s would be classified as pseudo or simulated DEX(T), since the former test identifies patients with a large slow vergence mechanism and the latter test

identifies patients with high AC/As.

Accommodation dynamics in DEX(T) have been studied by Schneider et al (60) using an infrared optometer. They found the latency for accommodation for true DEX(T) was significantly increased by 40-100 msec while overshoot frequency was decreased. On the other hand, simulated DEX(T) had a greater magnitude and frequency of overshoots of accommodation which is thought to be related to a robust slow vergence response, i.e., vergence overdriving the accommodative system.

Kushner (41), basing his results on our earlier findings (40), categorized 83 basic and DEX(T)s by measuring distance and near deviation with a cover test, gradient AC/A with -2.00 D at distance, and +3.00 D gradient AC/A at 33 cm both before and after one hour of occlusion. Occlu.sion results at near were used to differentiate simulated from true DEX(T).

Table 1 below is a rearrangement for analysis of Kushner’s data. Distance results are reported in the next to last column. All of Kushner’s patients, who had surgery, received bilateral symmetric lateral rectus recession.

Ku.shner’s type la or true DEX(T), have a high AC/A with both distance near and gradient methods. They account for 11% of X(T)s. Neither proximal convergence nor vergence aftereffects are strong in these patients. Traditional bilateral lateral rectus recessions on patients with true DEX(T) and truly high Ar/A ratios, according to Kushner, result in consecutive esotropia at near with good alignment at

distance. Kushner advises that if the measured gradient AC/A is high, the pa-tient should be advised of the possibility of consecutive esotropia at near and/or the surgical goals should be modified.

Kushner’s type ib, true DEX(T), has a high AC/A using the distance-near relationship but a normal gradient AC/A and accounts for 9% of the DE. The discrepancy between distance-near AC/A and gradient AC/A is most likely due to abnormally high proximal convergence since minimal vergence aftereffects are demonstrated by an increase in deviation at distance after occlusion. If the distance deviation had increased after occlusion then the patients would also have strong vergence aftereffects.

Simulated DEX(1) type 2a and type 2b both have normal gradient AC/A after occlusion. Prior to occlusion, type 2a demonstrates an abnormal AC/A with a +3.00 lens at near but not with a -2.00 at distance. This finding has no physiological explanation other than indicating the variability of stimulus gradient AC/A ratios depicted in Figure 2. Thus, Kushner’s two types of simulated DEX(T) may, in fact, be the same entity which represents large variability in the measurement of a stimulus AC/As. The normal AC/A and the smaller near finding are probably a result of a robust slow vergence mechanism.

Kushner’s data suggest that all simulated DEX(T) and most true DEX(T) have a normal AC/A The discrepancy between distance-near AC/A and gradient AC/A may be explained by a robust slow vergence system in the simulated DEX(T) and a robust proximal convergence in the true DEX(T).

193