As we are not aware of any reports on the outcome of ULR for CI-type IXT, we compared ULR and RR in patients with CI-type IXT. We found that ULR was as effective as the RR procedure in treating CI-type IXT in terms of collapse of the N-D difference and the surgical success rate.
The reference value of N-D differences used in the classification remains somewhat controversial across studies: Burian and Spivey 1,8 recommended a 10-PD difference, which has generally been used till date. However, Hardesty et al. 2 used a 5-PD difference as the reference value, and another study used a 15-PD difference 3. While Suh et al. 20 used a 10-PD as the reference value in distance exodeviations of more than 30 PD, in exodeviations of less than 30 PD, the value was defined as one-third of the distance deviation because 10 PD would be a relatively significant difference in a small angle, and the effects of treatments on the change in types might be underestimated in such cases 20,21. In this study, based on Hardesty et al.’s classification, the CI type was defined as exodeviation that was 5 PD larger at near than at distance fixation because only patients with IXT of small to moderate angles, even at near fixation, were included to analyze the surgical results of one muscle surgery (unilateral lateral recession).
CI-type IXT is much less common than other types of IXT and has been reported to occur in only 2.8‒4.2% of the IXT 22,23. Many surgical methods have been used to treat CI-type IXT, but the surgical outcomes are variable and mostly unsatisfactory, with success rates ranging from 18–92% 1,6,8−11.
Under the assumption that MR has the main effect on the near deviation angle and LR plays a role mainly in distance, MR resection has been classically introduced for the treatment of CI-type IXT. However, the success rates of unilateral or bilateral MR resection(s) for CI-type exotropia have been reported to range from 27–67%, which has motivated the development of new surgeries 6,17,24−26. Choi and Rosenbaum 6 performed unilateral or bilateral MR resection with an adjustable suture in 21 consecutive patients with CI-type IXT. The surgical success rate (10 PD esodeviation to 10 PD exodeviation) was 76.2% at the last examination. In 1995, Kraft et al. 5 introduced the revised method for RR for the CI-type X(T) in which LR recession and MR resection were biased to the distance and near deviations, respectively, and MR was strengthened more than LR was recessed. They reported that this surgery had a low risk of creating long-term postoperative esodeviations at distance. In Wang et al.’s 9 prospective study, the surgical results of the revised RR were better than those of unilateral or bilateral MR resection(s), reducing distance and near deviation. However, a high proportion of the patients experienced early postoperative overcorrection.
Raab and Parks reported that correction of N-D exotropia was obtained in only 28% of patients at 6 months after bilateral LR recession 27. Bilateral LR recession augmented to near exodeviation was described in the study by Farid and Abdelbaset 17. In this study, the success rate in the ULR group was 63.3% at the last follow-up, which was higher than that achieved in previous studies of bilateral LR recession: 40% in Raab and Parks’s study 27 and 50% at distance and 27.2% at near in Farid and Abdelbaset’s study 17 after bilateral LR recession.
In our study, the postoperative exodeviation and N-D difference were significantly reduced in both groups; however, there was no statistically significant difference between the ULR and RR groups. In addition, the risk of postoperative esodeviation or diplopia at distance after ULR was lower than that after RR, which might be a concern when performing only LR recession without MR resection. Success rates were not significantly different between the two groups.
This study had several limitations. First, because of the retrospective study design, the surgical method was selected without any specific policy, although the surgeon had no preference for either the ULR or RR procedure. Second, the preoperative near exodeviation angles and N-D difference in the ULR group were smaller than those in the RR group; thus, minor bias could have occurred. However, the collapse of the N-D difference after ULR was achieved with results similar to those obtained after RR, and there was no statistically significant difference between the two groups. Further studies should be conducted with a prospective, randomized design with a larger number of patients to confirm the efficacy of ULR in patients with CI-type IXT.
Notwithstanding these limitations, this study is meaningful in that it provides data suggesting a favorable outcome of unilateral LR recession in CI-type IXT, with comparable results to those of the recess-resect procedure. In CI-type IXT measuring less than 25 PD, there was no significant difference in the amount of N-D difference reduction after surgery or the surgical success rate between ULR and RR.
In conclusion, unilateral LR recession for IXT with CI measuring less than 25 PD is a useful surgical procedure because it produces surgical outcomes similar to those of the unilateral RR.