A retrospective analysis was conducted of all patients who had undergone unilateral or bilateral MR muscle resection between January 1994 and December 2007 for recurred intermittent exotropia, which had developed after bilateral LR recession.

Fifty-nine patients who had undergone previous bilateral LR recession were included in this study. The subjects were divided into two groups. Patients who had undergone unilateral MR resection were in the unilateral group (38 patients), and those patients who had undergone bilateral MR resection represented the bilateral group (21 patients). The effects of MR resection in both groups were analyzed. When performing unilateral MR resection, the non-dominant eye was selected [5].

Patients who had a lack of cooperation, those with myopia and hyperopia greater than 2 diopters (D), anisometropia above 2 D, and amblyopia with corrected vision worse than 20 / 25 were excluded from this study.

The following parameters were reviewed and analyzed: the amount of LR recession at 1st surgery, deviation before and after MR resection, the amount of MR resection, and the effects of MR resection per millimeter at 1 month postoperative. Alternate prism cover testing was performed at a distance of 6 m with the patient in the primary and secondary gaze positions. Alternate prism cover testing was then performed in the primary gaze position at 30 cm.

Unilateral or bilateral MR resection was performed under general anesthesia. Before pre-place sutures were made, extensive dissection of the connections to the overlying Tenon's capsule and the intermuscular membrane 10 mm or more from its insertion site were performed. Following the pre-place sutures, the muscle was cross-clamped with a hemostat just distal to the suture line, followed by transection of the muscle 1 mm anterior to the suture line. The muscle was disinserted and resected. The resected rectus was then securely attached to the sclera at the original insertion site using a long scleral tunnel suture technique. We performed one muscle resection for less than or equal to 20 PD of exodeviation, two muscle resection for exodeviation of greater than or equal to 30 PD, and one or two muscle resection for exodeviation of 21 to 29 PD.

The angle of deviation was measured 1 month after surgery to estimate the effects of MR resection. This effect was analyzed at 1 month postoperative because the main focus of this study is on the pure effect of medial rectus resection. In the long-term follow up, the degree of binocularity can cause a change in strabismic angle secondarily.

The effect of medial resection for each millimeter was calculated as the value of the angle of preoperative deviation minus postoperative deviation at 1 month postoperative, divided by the total amount of resection. In the bilateral resection group, the total resection amount was calculated by adding each individual resection amount.

The 30 patients among the unilateral resection group with accessible medical records were divided into two groups: patients with previous LR muscles recession greater than or equal to 7 mm (A group, 21 patients) and less than 7 mm (B group, 9 patients).

Statistical analysis was performed using the Mann-Whitney U-test and Pearson's correlation in the SPSS ver. 12.0 (SPSS Inc., Chicago, IL, USA). Differences in the age at MR resection surgery, the mean effects of MR resection per millimeter, and effects of unilateral MR resection according to previous degree of LR recession were analyzed between the two groups.

At the time of the MR resection, the average age of the unilateral group was 12.47 ± 6.74 years (range, 7 to 48 years), and the average age of the bilateral group was 12.38 ± 8.50 years (range, 7 to 47 years). A significant difference was not found (P = 0.252) between the two groups. The average deviation angle at the distal distance of the unilateral group was 20.3 ± 2.90 PD (range, 15 to 25 PD) and the average deviation angle in the bilateral group was 27.0 ± 3.98 PD (range, 25 to 40 PD).

The mean amount of LR recession in each eye in the first surgery was 6.80 ± 0.75 mm (range, 5 to 8 mm). The average amount of medial rectus resection in the unilateral group was 3.98 ± 0.71 mm (range, 2.5 to 6.0 mm) and 6.04 ± 0.84 mm (range, 5 to 8 mm) in the bilateral group.

Table 1 is a summary of the mean postoperative change in deviation and the mean effect per millimeter of MR resection. The average deviation corrected per millimeter of MR resection was 4.15 ± 1.11 PD (range, 2.0 to 6.7 PD). There was no correlation between the average effect per millimeter and the amount of MR resection (Pearson's correlation r = 0.114, P = 0.425).

The average correction per millimeter of MR resection in the unilateral group was 4.21 ± 1.21 PD and 4.05 ± 0.91 PD in the bilateral group. The effects of resection were greater in the unilateral resection group, but there was no statistically significant difference (P = 0.60) between the two groups.

In the analysis of the difference of effects according to the amount of previous LR recession, the average effect per millimeter was 3.99 ± 1.02 PD in group A (LR recession greater than or equal to 7 mm) and 4.15 ± 1.19 PD in group B (LR recession less than 7 mm). The effects of resection were greater in the unilateral B group, but the difference was not statistically significant (P = 0.70).

The amount of MR resection required for recurrent exotropia after bilateral LR recession is difficult to determine, especially in children for whom adjustable surgery is not an option. A useful normogram according to strabismic angle has not yet been defined because the effect is quite variable and resection techniques differ among surgeons.

The chosen degree of resection is based upon an individual surgeon's experience and protocol. Kim and Kim [6] reported that their average deviation corrected was 3.41 ± 0.60 PD/mm, and Chae et al. [7] reported that the mean effects per millimeter of MR resection were 3.53 ± 0.17 PD/mm. Our study found results of 4.15 ± 1.11 PD/mm, but effects ranged from 2.0 to 6.7 PD/mm.

In this study, the average deviation corrected per millimeter of MR resection did not differ significantly between the unilateral and bilateral groups. Possible explanations for the variable effect of MR resection per millimeter and the lack of correlation between the degree and effect of MR resection include variability in the thickness and tonus of the MR. Further study of the relationship between the effects of MR resection and these factors is needed.

In the current study, there was no difference in the effect of MR resection per millimeter between the two previous LR recession groups. Cho and Kim [8] reported that all the new insertionof the 5 to 8 mm recessed LR was anterior to the functional equator. If the insertion of the previously recessed LR was posterior to the functional equator, this would result in a larger difference in MR resection per millimeter because the tonus of the LR would decrease markedly [9]. In the current study, LR tonus change was mild with only 5 to 8 mm of previous LR recession, which was not enough to make a difference between the two LR recession groups.

In conclusion, the average effect of MR resection was 4.15 ± 1.11 mm and there was no significant difference between the two study groups with regard to the laterality of MR resection and the amount of previous LR recession in this study. However, the effects of MR resection were widely varied. The most important conclusion of this study is that surgeons should be aware of the variability of the effects of MR resection.