Our study reveals that both incisional strabismus surgery and BTX injection are effective in treating AACE. Surgery provided consistent and stable motor alignment over time, while BTX injections led to earlier sensory success and sustained long-term fusion function. More importantly, we identified that AACE patients with anisometropia > 1 D experienced a higher recurrence rate following BTX injections. These findings provide crucial clinical insights, guiding clinicians in selecting the most appropriate therapeutic approach for patients with AACE, particularly those with anisometropia. To the best of our knowledge, for the first time, this prospective comparative study identifies the risk factors for recurrence following BTX injection in AACE patients.

Our results showed no significant difference in motor success rates between chemodenervation and incisional surgery at 1 and 3 months post-treatment. However, by 6 months, the motor success rate in the chemodenervation group reached 69.7%, confirming its efficacy in treating AACE, albeit lower than the 100% success rate in the surgery group. Similar findings were report by Cheung et al. who found no significant differences between the two treatments at 6 and 12 months in an international multi-center retrospective study, but noted a significantly higher success rate for surgery at 24 months [11]. Despite their longer follow-up period, their cohort included only children. Suwannaraj et al. included both adults and children, highlighting the variability in outcomes based on patient demographics and preoperative characteristics [12]. This discrepancy may be attributed to their participants having a preoperative deviation > 40 PD and only one medial rectus muscle being injected in all chemodenervation procedures. Other studies that observed lower efficacy of BTX-A injections also utilized lower doses, administered unilaterally, and conducted follow-up over only a 6-month period [13].

Previous studies comparing the efficacy of chemodenervation and strabismus surgery in treating AACE have generally found comparable results for both methods. In a retrospective analysis involving 49 patients, Wan and colleagues demonstrated that the efficacy of BTX was not inferior to surgery [14]. Further, a recent meta-analysis of nine studies involving 1,100 participants revealed that although surgery was more effective for various types of comitant esotropia, such as infantile and partially accommodative esotropia, chemodenervation proved equal effective in treating AACE [15]. These results support our findings and highlight the potential of chemodenervation therapy for AACE, underscoring the necessity to advance research on chemodenervation strategies for this condition.

Our study also revealed that the chemodenervation group achieved sensory success earlier than the surgery group, with effects persisting up to 6 months post-treatment. This early success may be attributed to several factors. First, BTX temporarily relaxes the medial rectus by inhibiting acetylcholine release, leading to a rapid reduction in muscle overactivity and facilitating early eye alignment, which places images within the fusion range [16]. Second, studies suggest that BTX may induce long-term changes at the neuromuscular junction, potentially promoting sustained muscle relaxation and alignment [17]. Third, rapid eye alignment helps restore binocular visual function by allowing the brain to reintegrate visual information from both eyes, and thus enhance sensory fusion. Additionally, the stretching effect of the relaxed muscle may contribute to stable alignment, further supporting the early recovery of binocular function [18]. Scott et al. pioneered reporting BTX's efficacy in treating strabismus, noting particularly its substantial benefits in adults with minor horizontal deviations and quick restoration of binocular vision [19]. Similarly, Tejedor and Rodriguez observed that chemodenervation rapidly corrected ocular misalignment and facilitated binocular fusion recovery, with increasing effectiveness with subsequent injections [20]. Cheung et al.’s study supported these observations, reporting a sensory success rate of 70.2% at 6 months post-chemodenervation [11], which closely aligns with our findings.

Although the surgery group demonstrated nearly 100% motor success, some patients exhibited inferior stereopsis despite stable ocular alignment. This discrepancy can be attributed to the fact that motor alignment, while essential, does not always guarantee full restoration of sensory fusion or optimal stereoacuity. Factors such as accommodative convergence, the degree of binocularity, and the timing of visual system adaptation post-surgery may influence stereoacuity [21, 22]. While surgical intervention ensures stable alignment, it does not necessarily result in complete sensory fusion or perfect stereopsis, highlighting the complexity of binocular vision recovery. Additionally, the development of stereopsis may be further influenced by visual training and the brain’s ability to integrate the corrected alignment over time. Future studies are needed to better understand the long-term effects of strabismus surgery on stereopsis.

Although most clinical guidelines suggest that the clinical effects of BTX gradually diminish after 3 to 4 months, animal studies have shown that temporary weakness induced by BTX injections into unilateral horizontal muscles can last from 2 weeks to 8 months [23]. Our study found that, despite a recurrence of strabismus at 6 months post-treatment, patients retained a certain level of binocular visual function, indicating that BTX enhances control for patients with AACE. While the surgical group demonstrated complete alignment from the first month, the BTX-A group exhibited comparable initial correction, which gradually diminished over time as some patients developed residual esotropia. This fluctuation in alignment within the BTX-A group explains the lack of sustained improvement in stereopsis, despite similar initial corrections between the two groups. The success rate of chemodenervation could potentially be higher with multiple injections. Shi and colleagues proposed that treatment can be considered successful after up to three repeat BTX injections [24]. Similarly, Tejedor and Rodriguez reported that the motor success rate significantly improved with the number of injections: 52.9% with one injection, 70.6% with two injections, and 88.2% with three injections in cases of acquired esotropia [25]. Our study focused solely on patients receiving initial treatment; subsequent research has indicated that repeated injections can further improve control of eye position in AACE. (unpublished data).

Previous studies predominantly employed retrospective designs and rarely included measurements of refractive status [11, 12, 14, 26]. This study is the first to incorporate cycloplegic refraction results, identifying anisometropia as a potential risk factor for recurrence in AACE patients treated with BTX. The underlying mechanism involves an imbalance in the binocular accommodative-convergence function. Patients with anisometropia experience different refractive states between their eyes, requiring varied accommodative efforts during visual fixation [27]. This disparity can decouple accommodative and convergence processes, particularly in cases of pronounced anisometropia, which exacerbate instability within the binocular visual system [28]. Consequently, while BTX injections may initially correct ocular misalignment, they often fail to sustain long-term binocular fusion. The misalignment between accommodation and convergence likely contributes to strabismus recurrence as the neuromuscular blockade from BTX diminishes [29]. It has been reported that anisometropia disrupts convergence responses, adversely affecting the long-term outcomes of strabismus treatment [30]. Therefore, clinicians should carefully consider the degree of anisometropia when selecting treatment strategies for AACE patients to optimize long-term therapeutic outcomes.

Additional potential risk factors that may affect the efficacy of BTX-A treatment include the prolonged duration from the onset of esotropia to treatment, patient age, and large preoperative deviation angles. Sheth and colleagues found that patients with acute comitant esotropia who underwent surgery 20.6 months after onset had a lower success rate compared to those operated on at 10.9 months post-onset [31]. However, their study lacked data on stereopsis. Yagasaki et al. discovered that patients undergoing strabismus surgery more than 12 months after onset have poorer stereoacuity [32]. Previous research also indicated that larger deviation angles, especially at near, are more challenging to correct with BTX [33]. Scott et al. reported that the maximum correction range for strabismus treated with BTX is 40 PD [19]. However, our study did not identify other risk factors affecting the efficacy of BTX treatment beyond anisometropia due to the smaller sample size and the timely interventions administered.

There are several limitations within this study. The small sample size is insufficient for generalizing findings to diverse populations or varying severities of AACE, and thus further studies with larger cohorts are necessary for conclusive clinical evidence. Additionally, the follow-up period was limited to 6 months, preventing an assessment of the long-term sustainability of treatment effects. Treatment allocation based on patient preference rather than random assignment can introduce selection bias, affecting the internal validity of the study. The lack of standardized dosage guidelines for BTX may result in variability in treatment efficacy, thereby complicating comparisons of its effectiveness in a broader clinical context. These factors emphasize the necessity for future studies with larger samples, extended follow-up periods, and more rigorous controls to thoroughly evaluate the long-term benefits and comparative effectiveness of these treatments for AACE.