Abstract
Purpose:
The purpose of this review was to examine the efficacy of botulinum toxin in the treatment of infantile esotropia and to evaluate the average response of BT and its complication rates.
Methods:
A research was performed in the Latin American and Caribbean Literature on Health Sciences (LILACS), MEDLINE, and Cochrane Central Register of Controlled Trial (CENTRAL). The database was searched between December 28, 2016 and January 30, 2017. The selection was restricted to articles published in English, Spanish, or Portuguese. There were no date restrictions in the search.
Results:
Nine studies were eligible for inclusion. The grouped success rate of BT treatment in infantile esotropia was 76% (95% confidence interval [CI]: 61%–89%). For the success rate, I2 of 94.25% was observed, indicating a high heterogeneity (P < 0.001). The complication rates were also analyzed. The grouped consecutive exotropia (XT) rate was 1% (95% CI: 0%–2%). The grouped ptosis rate was 27% (95% CI: 21%–33%). The grouped vertical deviation rate was 12% (95% CI: 4%–22%). The mean change of the deviation after BT injection was −30.7 (95% CI: −37.7, −23.8), demonstrating a significant improvement in alignment.
Conclusions:
Botulinum toxin injection into medial recti muscles reveals to be a safe procedure and a valuable alternative to strabismus surgery in congenital esotropia, especially in moderate deviations.
Infantile or congenital esotropia is a large angle convergent deviation with onset prior to 6 months of age in neurologically normal children. It is usually not related to refractive errors and often associated with other motor abnormalities, such as inferior oblique overaction, dissociated vertical deviation (DVD), and latent nystagmus.
1,2
The prevalence of esotropia in the population is approximately 1%, while the prevalence of infantile esotropia reported at birth is 27 per 10,000 live births.
1 Children who develop infantile esotropia begin it at approximately 4 months of age. Its etiology is a source of controversy and remains unknown.
1,3
Most investigators agree that early surgical intervention improves both motor and sensory outcomes, even if early surgery reoperation is common and binocular outcomes can be poor.
4,5 Although the treatment of choice for most patients remains surgical, opinions continue to differ as to the best approach, particularly in small and moderate angle deviations.
Alan Scott pioneered the pharmacologic treatment of strabismus in 1980.
6 Botulinum toxin (BT), which is an exotoxin of the bacterium
Clostridium botulinum, was approved by the FDA in 1990. Botulinum toxin A is a large protein molecule that, when injected in the muscle, remains at the nerve terminal, inhibiting the release of acetylcholine and resulting in muscle weakness or paralysis 3 to 5 days after the injection. An irreversible binding occurs and the muscle paralysis lasts approximately 8 to 12 weeks.
3 The chemodenervated muscle lengthens and its antagonist contracts. Sensory mechanisms may also play a role in the realignment after treatment with BT A in some types of strabismus. The visual axes may adopt an alignment that permits binocular single vision; and this may persist or regress, demanding further treatment.
3,6,7 Initial doses of botulinum toxin for individual muscles were initially suggested by Alan Scott (
Table 1).
8
Botulinum toxin injection in the extraocular muscles is an alternative option that has become well established in some cases of adult strabismus, such as sixth nerve palsy. Its use in children is more controversial, as it requires sedation and some complications, as ptosis, may be more frequent in this population.
3,6,7 Nevertheless, many investigators have advocated the use of botulinum toxin as a good alterative in the treatment of infantile esotropia, as it is a less invasive procedure than muscle surgery.
The success of BT injections in infantile esotropia seems to be related to the age that the treatment is performed. There is a critical age when the medial rectus muscle is more sensitive to the chemodenervation induced by BT, allowing the antagonist lateral rectus to regain its function. Besides, morphologic modifications in the structure of the extraocular muscles (i.e., tight medial recti on the forced duction test) are apparently still deficient at early ages. It has been demonstrated that botulinum toxin affects the extraocular muscles in infant monkeys more severely and for longer than those in adult monkeys.
9,10
It could be assumed that early treatment allows a “self-adjusting” sensorimotor mechanism to support stable binocular alignment near orthoposition, once the motor system has been suitably modified. In fact, at early ages, the central connections of the sensory and motor visual systems can still be changed.
The complications reported on the use of BT include secondary ptosis; initial consecutive exotropia (XT) lasting up to 2 to 3 months; and the temporary effect of the BT injection itself.
8,13–18 Also, many studies show that multiple injections may be needed to sustain the alignment effect.
19 On the other hand, the advantage is that BT is a simple treatment, in which injections are performed under minimal anesthesia without intubation. However, some studies show that the motor and sensory outcomes may be significantly worse than surgery even with multiple injections.
20
What is the importance of this review?
Currently there are no guidelines about the use of BT in the treatment of infantile esotropia. In the past decades, several retrospective case series have been published on its use to treat strabismus. The studies were performed in different types of strabismus, using different doses, methods and follow-up times. The outcomes were diverse.
The main purposes of this study were to examine the efficacy of BT in the treatment of infantile esotropia and to evaluate its dose response and complication rates.
Research was performed in the Latin American and Caribbean Literature on Health Sciences (LILACS); MEDLINE; and Cochrane Central Register of Controlled Trial (CENTRAL). The database was searched between December 28, 2016, and January 30, 2017. The selection was restricted to articles published in English, Spanish, or Portuguese. There were no date restrictions in the search.
The strategies used were: botulin* toxin*; botulinum toxin; Clostridium botulin; Botox; strabismus; esotropia; infantile esotropia; congenital esotropia.
We searched for the following elements from the studies:
-
Inclusion and exclusion criteria, presence of a group of patients with infantile esotropia, follow-up period, success criteria.
-
Age at which the procedure was performed, history of previous treatment.
-
Type of BT used, dose, and number of injections.
-
Motor and sensory outcomes after a minimum follow up of 6 months.
-
Effects.
-
Bias.
The following analyses were performed: success rate and possible factors associated to success; association of complications with the dose of BT and complications reported at each study; and mean variation of the deviation after BT injection.
A minimum follow-up of 6 months was required to access the primary outcomes. Motor success was defined as postprocedure deviation within 10 prism diopters (PD) of orthotropia. Sensory success was defined as normal ranges of binocular single vision.
The following adverse outcomes were considered: induced ptosis, induced vertical deviation, and consecutive exotropia.
The heterogeneity between studies was evaluated through the use of the
I2 statistics.
21 The
I2 statistics varies from 0 to 100, and the higher it is, the greater the heterogeneity observed among the rates.
The complications (consecutive exotropia, ptosis, and vertical deviations), and the deviation change were presented using a random model. Meta-regression models were fitted to identify possible sources of heterogeneity between the estimations.
A significance level of 5% was used for all statistical tests. Statistical analyzes were performed using statistical software (STATA 12.0; Stata Corp., College Station, TX, USA).
Due to the substantial heterogeneity observed in the results, the success rate, the complications, and the deviation change were presented using a random model. Subsequently, meta-regression models were fitted to identify possible sources of heterogeneity between the estimations.
The grouped success rate of botulinum toxin treatment in infantile esotropia was 76% (95% confidence interval [CI]: 61%–89%;
Fig. 1). For the success rate,
I2 of 94.25% was observed, indicating a high heterogeneity (
P < 0.001). Studies 1, 7, and 9 presented similar rates (between 66% and 68%), as did studies 3B and 5 (86% and 87%, respectively). On the other hand, study 8 had the lowest success rate (38%) and study 6 (A and B), 100% of success. We attempted to identify the possible causes of heterogeneity by considering the means of the age, initial deviation, mean number of injections, follow-up time, and sample size via meta-regression. Univariate and multivariate models were adjusted and are presented on
Table 4. There was no association between the success rates and the study variables.
Table 4 Meta-Regression Results for Success Rate
Table 4 Meta-Regression Results for Success Rate
Meta-regressions were adjusted for the seven studies that performed the injections with the same dosage of BT (3A, 3B, 4, 5, 6A, 6B, and 9). Due to the number of studies, for the multivariate model we considered the significant predictor variables at 20%.
As shown in
Table 5, only the dose remained significant in the multivariate model (
P = 0.025). Thus, excluding the mean number of injections of the model, we consider the univariate model only with the dose (
P = 0.005). Therefore, the higher the dose of BT injected, the lower the success rate; every 1 IU increased in the mean dose there was a reduction of 0.10% in the success rate.
Table 5 Meta-Regression Results for Success Rate Considering a Pattern Dose of Botulinum Toxin
Table 5 Meta-Regression Results for Success Rate Considering a Pattern Dose of Botulinum Toxin
When correlating the technique used for BT injections and success rates, the injections guided by electromyography revealed higher success rates than the transconjunctival injections with no electromyography (EMG; P = 0.006). When EMG and direct visualization injection were compared, there was no statistic difference (P = 0.89).
The complication rates were also analyzed. The occurrence of permanent consecutive XT, transient ptosis, and vertical deviation were considered complications. For meta-regressions, age and dose were considered predictors, when possible.
According to
Figure 2, only five studies contributed to the analysis of consecutive XT. The grouped consecutive rate of XT was 1% (95% CI: 0%–2%). For this analysis, the
I2 was 0% (
P = 0.490) indicating homogeneity among the estimates. In four of the five studies, no cases of consecutive XT were reported and one study presented a rate of only 3%. Supporting this, there was no association between the consecutive XT rate and age (
P = 0.435) via meta-regression. It was not possible to evaluate the dose-effect as the four studies using a single dose (5, 6A, 6B, and 9) had no consecutive XT cases.
Seven studies contributed to the analysis of the occurrence of ptosis. The grouped ptosis rate was 27% (95% CI: 21%–33%;
Fig. 3). The value of
I2 was 20% (
P = 0.280); therefore, the rates of ptosis were not considered heterogeneous. Only the 6A study presented a rate of ptosis higher than 40%, the other studies presented rates ranging from 17% to 32%. By the meta-regression analysis, there was no association between the rate of ptosis and age (
P = 0.654), as well as the dose of BT injected (
P = 0.581).
When analyzing the occurrence of vertical deviation, the grouped rate was 12% (95% CI: 4%–22%;
Fig. 4). For this analysis,
I2 was 68.1% (
P = 0.020), showing heterogeneity. Study 8 had no case of this complication, differently from the others that presented rates above 15%. There was no association between the rate of vertical deviation and age (
P = 0.069) via meta-regression. It was not possible to assess the effect of the dose as only four studies presented the single dose regardless the dose being smaller or larger.
The mean change of the deviation after BT injection was −30.7 (95% CI: −37.7, −23.8), demonstrating a significant improvement in alignment (
Fig. 5). For this analysis,
I2 was 88.1% (
P < 0.001) denoting that the mean deviation change in the studies was heterogeneous. Adjusting for a meta-regression model and considering age as a predictor variable, it was significant (
P = 0.020), indicating that the older the child at the time of the procedure, the greater the deviation/increase of 0.65 diopters (95% CI: 0.19, 1.11) at every 1-year increase in mean age, reducing the delta in the deviation. It was not possible to establish dose effect information.
Surgical success rates for infantile esotropia have been reported by Scott et al.
22 (65%); von Noorden et al.
23 (66%); and Issaho et al.
24 (59%). Currently most strabismus specialists prefer surgery to other treatments due to the more predictable results in the alignment.
Scott
6 first described the use of botulinum toxin in the treatment of strabismus in 1980 and afterward several authors also reported their outcomes. However, it remains unclear how effective and predictable the BT is in comparison to other options to treat strabismus.
Nine studies in which BT injection was performed to treat infantile esotropia were included in this analysis. Two reports were subdivided for analysis
12,15 since the results were provided separately according to age.
All the authors used botulinum toxin, except for Scott et al.
8,16 who did not specify the type of BT injected. Additionally, some authors used the transconjunctival technique with EMG assistance in the procedures.
8,12,15,16 Others opted for no EMG assistance.
11,13,17,18 Campos et al.
14 performed the injections under direct visualization of the muscle.
We were able to report the primary outcome of improved ocular alignment by a reduction in the mean deviation in all nine studies. Mean preinjection deviation was 38 PD, while mean postinjection deviation was 6.25 PD. The average number of injections was 1.4. The grouped success rate of BT injection was 76%, but we did observe a high heterogeneity in the studies (I2 statistics: 94.25%). It was not possible to evaluate the dose response of BT across a range of different doses used in some of the studies.
An interesting result was the inverse association between the dose of BT and the success rate. We observed that the higher the dose of BT injected, the lower the success rate—every 1 IU increased in the mean dose there was a reduction of 0.1% in the success rate. Probably this is due to the tendency to use higher doses of BT in bigger angles of deviation. Alba Campomanes et al.
11 concluded that BT was most effective for esotropia ranging from 30 to 35 PD, with a success compared to surgery. Success with BT injection when deviation was lower than 30 PD was 51%, while the rate was 39% if deviation was greater than 30 PD. The magnitude of deviation predicted motor outcome when strabismus was treated with BT instead of surgery.
Scott et al.
8 and de Alba Campomanes et al.
11 also analyzed the success rates considering initial angle of deviation (smaller or bigger than 30 PD). They showed better rates in smaller deviations (73% and 59%, respectively) than in bigger esotropias (54% and 36%, respectively).
The average change of the deviation after BT injection was −30.7, revealing a significant improvement in alignment after the intervention, although the mean deviation change in the studies was heterogeneous. This supports the indication of BT injection as a good alternative to treat moderate esotropia (until 30–35 PD of deviation).
Furthermore, the risk of experiencing anesthesia-related complications in a healthy infant is negligible, but this should still be considered. In addition to the procedural simplicity of botulinum toxin injection, performing it under minimal anesthesia without intubation in a baby is another advantage.
25,26
When evaluating the occurrence of complications, the grouped consecutive rate of XT was very low (1%). Hypocorrection is much more frequent than hypercorrection. The overall ptosis rate was 27% with low heterogeneity and vertical deviation rate was 12% with high heterogeneity between groups.
Other less frequent adverse outcomes cited include subconjunctival hemorrhage and diplopia. Inferior oblique overaction and DVD were also observed, but they are naturally associated with infantile esotropia, and not a consequence of the BT injection. Issaho et al.
24 reported frequencies of 38% of inferior oblique overaction and 18% of DVD in a large group of children with infantile esotropia submitted to bilateral medial recti recession. Serious complications such as globe perforation did not occur in any of the studies.
Only two authors attempted to analyze the binocular vision. Scott et al.
8 revealed that of the patients that could be tested for binocularity, several had good motor fusion, but only one third had full stereopsis. Gursoy et al.
18 compared the results of BT injection versus muscle surgery. Binocular vision with Bagolini-striated glasses was detected in 86% of the reliable responders in the BT and 78% in the surgery group. Gross stereopsis was demonstrable in 10 of 15 patients in the botulinum group and 9 of 13 in the surgery group (odds ratio = 3.6,
P = 0.46). Two patients in the botulinum group achieved stereoacuity of 40 arc seconds.
Our results elucidate many questions currently experienced on the use of botulinum toxin to treat infantile esotropia. However, most of published literature consists of retrospective studies, case reviews or cohort studies. Although these provide very useful descriptive information, in order to improve the evidence base for the use of botulinum toxin as an independent management alternative, there is a strong necessity for good quality randomized trials to be performed. Standardization is very important, considering the types of botulinum toxin available and especially the different dosages used as these features are not quite comparable. The evaluation of binocular vision is also an important factor to be considered in future trials.
Supported by São Paulo Research Foundation (FAPESP) Grant 2011/51626-1 and Coordination for the Improvement of Higher Education Personnel (CAPES), Ministry of Education of Brazil. Fellowship support received by CAPES (to DCI), National Postdoctoral Program (PNPD)/CAPES (to LCC-C), Young Researchers at Emerging Centers Program of FAPESP, Grant 2012/15603-0 (to FRdSC), and a research productivity fellowship from the National Council for Scientific and Technological Development (to DdF).
Disclosure: D.C. Issaho, None; F.R.S. Carvalho, None; M.K.U. Tabuse, None; L.C. Carrijo-Carvalho, None; D. de Freitas, None