Tejedor J, Rodríguez JM. Management of Nonresolving Consecutive Exotropia Following Botulinum Toxin Treatment of Childhood Esotropia. Arch Ophthalmol. 2007;125(9):1210-1213. doi:10.1001/archopht.125.9.1210
To study motor and sensory results of surgery for exotropia following botulinum toxin A injection to correct childhood esotropia.
The medical records of 2445 patients treated with botulinum toxin bimedial injection were retrospectively reviewed to select patients operated on after 1 year of consecutive exotropia. We recorded age at onset of deviation, retinoscopic refractive error, visual acuity and age and dosage at the time of botulinum toxin injection. Retinoscopic refractive error, visual acuity, deviation angle, and stereoacuity before surgery for consecutive exotropia were also recorded. Families were contacted to obtain retinoscopic refraction, visual acuity, deviation angle, Bagolini test, vectography, and stereoacuity data 1 to 8 years after surgery.
A total of 5 children with acquired esotropia and 2 with infantile esotropia were included. A high dose of the toxin per injection might increase the risk of consecutive esotropia. Preoperative mean exotropic deviation was 15.42 prism diopters (PD) (range, 10-25 PD), and stereoacuity was not measurable before surgery. Postoperative mean deviation was 6 PD (range, 4-8 PD), and mean stereoacuity was 447.14 arc seconds. In 2 patients, suppression of the nondominant eye was detected. Three children had poor stereoacuity.
Surgery for exotropia following botulinum toxin injection in children is effective from a motor and sensory point of view.
Consecutive exotropia due to overcorrection of esotropia following botulinum toxin A injection is a rarely reported complication.1 Botulinum toxin injection of medial recti usually induces transient exotropia of variable duration, frequently less than 3 months, at the end of which realignment is achieved. When exotropia lasts more than 3 months, most practitioners wait until a decision about eventual surgery is reached, because the possibility of recovery is contemplated. The timing of surgery depends on the size of the angle, direction, and amount of change. In some children who developed postinjection exotropia after treatment for esotropia in our hospital, about 1 year elapsed before realignment to orthotropia was observed. In a few of the patients with exodeviation, realignment had not occurred after 1 year of exotropia and the deviation angle increased or remained stable for the second half of this period. We report the characteristics and results of surgical treatment of exotropia in these children, in the context of botulinum toxin–treated childhood esotropia.
We retrospectively reviewed the medical records of children treated for esotropia with botulinum toxin between June 1990 and December 2005, to identify those with consecutive exotropia who did not respond to conservative treatment. This treatment included alternate occlusion, near-point orthoptic exercises, and minus overcorrection. We defined exotropia lasting more than 1 year after injection, despite conservative treatment, as nonresolving consecutive exotropia. Preinjection variables recorded included estimated age at onset of deviation, retinoscopic refractive error (determined 30 minutes after instillation of 1% cyclopentolate), visual acuity when possible (Teller Acuity Cards [Vistech Consultants Inc, Dayton, Ohio], Cardiff Acuity test [Keeler Ltd, Windsor, England], or logMAR Crowded Glasgow Acuity Cards [Keeler Ltd, London, England]), deviation angle as measured by the simultaneous prism and cover test or Krimsky test with refractive correction, age at treatment, and dosage of botulinum toxin. All children were treated with bimedial injection of botulinum toxin A (Botox; Allergan Pharmaceuticals [Ireland] Ltd, Inc, Westport, County Mayo, Ireland) using dosages of previously published guidelines,2- 4 under nitrous oxide inhalation, ketamine intramuscular injection, or sevoflurane.
After more than 1 year of persistent exotropia, children were treated with surgery following informed consent of parents. Data recorded preoperatively were retinoscopic refractive error under cycloplegia, visual acuity (logMAR Crowded Glasgow Acuity Cards or Early Treatment Diabetic Retinopathy Study Lighthouse Chart Illumination Unit; Sussex Vision, West Sussex, England), deviation angle (Krimsky test or simultaneous prism and cover test), and stereoacuity (House fly or Randot circles; Stereo Optical Company, Chicago, Illinois). In the records of the 6-month postoperative examination, stereoacuity measurements were not available in 4 of the 7 operated on patients. Therefore, we contacted families and examined children again. Retinoscopic refraction under 1% cyclopentolate with subjective refining, visual acuity determination (Glasgow Acuity Cards or Early Treatment Diabetic Retinopathy Study chart), simultaneous prism and cover test, Bagolini striated glasses, vectography, and Randot circles stereoacuity (House fly or Randot circles) were performed at this visit, occurring between 1 and 8 years after surgery for consecutive exotropia.
In a series of 2445 children treated for esotropia, 7 patients were identified who, after more than 1 year of persistent exotropia, were treated with surgery. To facilitate placing these patients in the context of all children treated for childhood esotropia with botulinum toxin injection, we summarize their characteristics in Table 1.
In 5 of the 7 children with nonresolving exotropia (patients 1-5), the diagnosis of acquired esotropia had been made before botulinum therapy, based on late onset and absence of features commonly associated with essential infantile esotropia (dissociated vertical deviation and latent nystagmus). In the 2 remaining patients (patients 6 and 7), the pretreatment diagnosis was essential infantile esotropia, based on early onset, relatively large angle of deviation, and coexistent latent nystagmus. The deviation angle before botulinum toxin treatment was 30.57 prism diopters (PD) on average (range, 24-40 PD). Amblyopia, defined as at least 2 logMAR lines of difference in visual acuity between the 2 eyes, was not detected in any of the 7 patients immediately before botulinum toxin treatment, but 3 of them had undergone occlusive treatment before botulinum toxin injection. The characteristics of children with nonresolving exotropia are listed in Table 2. All patients received only 1 botulinum toxin injection of both medial recti. The mean total botulinum toxin dose was 6.07 IU, although in most cases 5 IU were injected in total (2.5 IU in each muscle). Surgery after botulinum toxin treatment was aimed at the correction of an exotropic distance angle averaging 15.42 PD (range, 10-25 PD). Stereopsis could not be detected in any of the operated on patients preoperatively. The surgical procedure consisted of unilateral recession of the lateral rectus muscle in the nondominant deviated eye in 2 patients and bilateral recession of lateral recti muscles in 5 patients. Table 3 summarizes the data of children before surgery.
After surgical correction of exotropia, the mean distance deviation was 6 PD (esotropia range, 4-8 PD), and mean stereoacuity was 447.14 arc seconds (without prisms but with correction of refractive error). In 2 patients (patients 3 and 7), suppression of the nondominant eye was detected. These children, and another child (patient 6), also had poor stereoacuity. The postoperative recorded variables are displayed in Table 4.
Overcorrection of esotropia with botulinum toxin is an uncommon previously reported complication of this therapy.1 We have analyzed the motor and sensory results of surgical treatment of this complication. It has been argued that the sensory outcome of primary botulinum toxin therapy might be worse than that of surgery, because of delayed alignment resulting from frequently necessary multiple injections, particularly in infantile esotropia.5 The motor results of the patients with nonresolving consecutive exotropia, after toxin injection and realignment surgery, were satisfactory (within 8 PD of esotropia at distance in all of them). Regarding sensory outcome, it was considered to be fair on average (447 arc seconds), but patients with the initial diagnosis of infantile esotropia had poor results (800 and 1000 arc seconds), whereas the results of those in the acquired or late-onset esotropia subgroup were good (mean stereoacuity, 266 arc seconds). Considering all the children treated with botulinum toxin whose records included posttreatment stereoacuity measurement (n = 1864), the average stereoacuity was 469 arc seconds, and decreased to 634 arc seconds when only patients with essential infantile esotropia (n = 794) were analyzed.
In 3 of the children with nonresolving consecutive exotropia (patients 2, 4, and 7), the initially nondominant eye was the exodeviated eye after injection of the toxin, but in the remaining 4 children there was no apparent fixation preference for any of the 2 eyes during exodeviation.
Overcorrection of esotropia using botulinum toxin might be related to a small deviation angle or a high dose of the toxin. Although some of the described patients had small-angle esotropia, the mean deviation in this group was about the same as (or slightly lower than) the mean deviation in the total number of children treated with botulinum toxin for esotropia (33.42 diopters). This treatment is indicated particularly in small–deviation angle esotropia. The total dose used in these children is the same on average as that used in other children, in agreement with previously published guidelines.2- 4 A point of interest is that the children with nonresolving consecutive exotropia received only 1 botulinum toxin injection, whereas all children received, on average, 1.6 injections, which means that the dose administered per injection was higher in the exotropia subgroup. A dose closer to the upper limit of the recommended dosage per injection might increase the risk of consecutive exotropia. On the other hand, in 4 of the 7 patients in the exotropia subgroup, the exodeviation increased with time after botulinum toxin injection, and in 3 of them, it remained constant after reaching a maximum. Of all the children treated with the toxin, 1143 additional children developed posttreatment exotropia that decreased after reaching a maximum between 15 and 35 days postinjection. Increasing exodeviation after 2 to 3 months postinjection is a possible sign of nonresolving exotropia.
A matter of debate is whether a small consecutive exotropia, as is the case in 2 of the patients described (patients 4 and 7), should be treated with surgery or could be left without treatment. We decided to operate on these children, who had a constant exodeviation for at least 1 year. In our experience, most exodeviations remain the same or increase with time. Stereopsis could be detected in children postoperatively (small esodeviation), but not preoperatively (while exodeviated), which supports the contention that sensory results of corrected infantile esotropia are better for small residual esotropia than for small overcorrected exotropia,6 and probably as well in acquired esotropia.
A 0.28% prevalence of esotropia overcorrection by botulinum toxin in children, which could be inferred from the present study, is not completely reliable because some patients with consecutive exotropia may have not been followed up or treated by us. The present study protocol was not a systematic longitudinal follow-up but a retrospective account of data. The reported value also depends on the criterion used to define overcorrection. We used a restrictive criterion to include all patients with exodeviation, but those in the lower end of the exodeviation angle range could have been left untreated for some time or indefinitely.
There is a low potential risk of overcorrection when botulinum toxin is indicated for the treatment of childhood esotropia, which might increase when a higher dose per injection of the toxin is used, and conservative measures may be used for 1 year before surgical correction is decided. However, surgical treatment of this complication is usually effective.
In summary, we have reported good motor and sensory outcomes of patients treated with surgery for nonresolving consecutive exotropia as a result of botulinum toxin injection. A 1- or 2-muscle procedure is usually sufficient to provide a stable motor outcome, and some degree of stereopsis, in children with no measurable stereoacuity before surgery.
Correspondence: Jaime Tejedor, MD, PhD, Department of Ophthalmology, Hospital Ramón y Cajal, Carretera Colmenar km 9100, Madrid 28034, Spain (email@example.com).
Submitted for Publication: November 7, 2006; final revision received January 29, 2007; accepted January 29, 2007.
Author Contributions: Dr Tejedor had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Financial Disclosure: None reported.
Funding/Support: This study was supported by Hospital Ramón y Cajal.