The various preparations are used on- and off-label for various conditions including upper motor neuron syndrome, hyperhidrosis, various spastic disorders including blepharospasm, hemifacial spasm, neurogenic detrusor over-activity, focal dystonias, strabismus, and vaginismus. BoNT is also used for chronic migraine, benign prostatic hyperplasia, and bruxism, and widely used in cosmetic treatments. Emerging uses for botulinum toxin type A include chronic musculoskeletal pain, vocal cord dysfunction, and allergic rhinitis.

Neither BoNT-A nor BoNT-B has been approved by the US Food and Drug Administration for any gastrointestinal disorder, including achalasia. However, besides achalasia, various other potential off-label uses have been described, including obesity (by purportedly delaying gastric emptying time) [20], and conversely gastroparesis (by improving gastric emptying time), gastric cancer (by possible vagal denervation) [21], sphincter of Oddi dysfunction [22], and anal fissure [23].

The profile of adverse effects for the various available preparations is similar, although BoNT-B may have additional reported systemic autonomic adverse effects. Long-term use did not show additive adverse effects [24]. Serious and/or immediate hypersensitivity reactions have been reported, which include anaphylaxis, serum sickness, urticaria, soft tissue edema, and dyspnea. If such a reaction occurs, further injection of BoNT should be discontinued and appropriate medical therapy should be given immediately. One fatal case of anaphylaxis has been reported in which lidocaine was used as the diluent, thus the causal agent could not be reliably determined. Side effects of BoNT may be seen beyond the site of injection, and may include asthenia, generalized muscle weakness, diplopia, blurred vision, ptosis, dysphagia, dysphonia, dysarthria, urinary incontinence, and breathing difficulties, depending also on site of injection. These side effects have been reported hours to weeks after injection. Swallowing and breathing difficulties can be life threatening and there have been reports of death related to spread of toxin and its effects. The risk of symptoms is probably greatest in children treated for spasticity but symptoms can also occur in adults treated for spasticity and other conditions. BoNT injection is relatively contraindicated in patients with disorders of neuromuscular transmission or with bleeding, and significant risk of side effects has been reported in patients with mitochondrial cytopathies, for example [25]. While various BoNT preparations contain human albumin, their risk for transmitting viral diseases is deemed extremely remote owing to careful screening of donors and manufacturing processes. The potential transmission of Creutzfeldt-Jakob disease (CJD) is also considered extremely remote for the FDA-approved formulations, and no known cases of transmission of viral diseases or CJD have ever been identified.

Side effects specifically after intraesophageal injections include transient non-cardiac chest pain and reflux. Severe complications are rare, and only in isolated case reports. These complications include gastroparesis, mediastinitis, and fatal arrhythmia, which were largely attributed to technical difficulties [26].

Contraindications for use of BoNT include history of allergy/sensitivity, and those with infections at the proposed site of injection. No formal drug interaction studies have been conducted with BoNT for injection. Co-administration of BoNT and aminoglycosides or other agents interfering with neuromuscular transmission (e.g., curare-like compounds) should only be performed with caution because the effect of the toxin may be potentiated. Likewise, the use of anticholinergic drugs after administration of BoNT may potentiate systemic anticholinergic effects. The safety is unknown for using different BoNT products at the same time or within several months of each other before resolution of known effects, but may lead to excessive neuromuscular weakness. Excessive weakness may also be exaggerated by administration of a muscle relaxant before or after administration of BoNT.

No long-term carcinogenicity studies in animals have been performed, and safety and effectiveness in pediatric patients have not been established. Various preparations of BoNT remain FDA Pregnancy Category C, and animal reproduction studies have not been conducted; it is not known whether BoNT can cause fetal harm when administered to a pregnant woman or can affect reproductive capacity. Also studies have not established whether BoNT is excreted in human milk after various sites of administration, thus caution should be exercised when BoNT is administered to a nursing woman.

While most patients continue to respond to repeated BoNT treatments, some become unresponsive possibly as a result of forming neutralizing or blocking antibodies, particularly to the heavy chain of either BoNT-A or -B [27, 28]. Risk factors leading to blocking antibodies is unknown, but is thought to increase with increased dose and frequency. Formation of neutralizing antibodies seems to be a higher risk in younger patients receiving higher doses intramuscularly, as in patients with neuromuscular diseases [29].

Generally, 100 units of BoNT-A marketed as BoTox is used for treatment of achalasia. Although there are reports of similar efficacy using BoNT-B, animal studies generally show longer duration of action for BoNT-A [24]. There are also differences even between preparations of BoNT-A, and a review of head-to-head, randomized controlled trials of onabotulinumtoxinA vs. abobotulinumtoxin suggested that the latter tends to have higher efficacy, longer duration, and higher frequency of adverse effects, albeit depending on use. The actual conversion factor between the two preparations varies in such studies and remains controversial, but a Botox:Dysport conversion ratio of 1:3 may be clinically appropriate, consistent with animal studies, which suggest a conversion ratio of approximately 1:2.5–3.0 [30]. Specifically in treatment of achalasia, however, there were no differences in response rates between onabotulinumtoxinA and abobotulinumtoxin [31].

Botox is diluted in 10 ml preservative-free normal saline just prior to use. Rehydration should be done gently, first by releasing the vacuum from the bottle, slowly introducing 10 ml of preservative-free normal saline, then rotating rather than shaking the vial. Once reconstituted, the solution should be used within a couple hours.

For the injection, a Carr-Locke (US Endoscopy) or equivalent sclerotherapy injection needle is introduced through the accessory channel of the endoscope, and approximately 20–25 units of BoNT-A is injected in four quadrants at or just above (within 1 cm) the squamocolumnar junction just above the z-line under direct visualization. The angle of the needle should be approximately 45 degrees to the surface and care must be taken to confirm that there is significant resistance at the syringe during the injection consistent with intramuscular injection in order to avoid injection outside the esophageal wall or into the superficial layers, which may create a bleb. Use of ultrasound guidance has been proposed, but has not become standard practice [32]. The overall method for injecting BoNT has essentially not changed with different studies, although there is some variation in doses and procedures. For example, one study advocates not a single, but two injections of 100 U of BoNT-A 30 days apart as the most effective therapeutic schedule [33].

The clinical response rate for BoNT is approximately 90 % at 1 month after BoNT treatment, but approximately 30–50 % at 1 year and <5 % at 2 years, according to some studies. However, other observational studies looking particularly at long-term efficacy of BoNT have shown better remission rates of approximately 70–80 % at 1 year, and 50 % at 2 years [18, 34].

Nevertheless, most studies comparing BoNT to pneumatic dilation (PD), show superiority of the latter. For example, one non-randomized study showed that global symptom scores and LES pressures improved significantly in both BoNT (n = 23) and PD (n = 14) groups at 12 months, but at 24 months there was significantly superior response rate of a single PD over BoNT treatment, and at 48 months all BoNT patients had symptomatic relapse while 35 % of patients treated by dilation (and 45 % of patients treated successfully by dilation) remained symptom-free [35]. Similarly, another study looking at response rates particularly in the elderly (>65 years of age) showed that the relief of symptoms was shorter-lived for BoNT compared to dilation: symptom alleviation was 13.8 ± 9.5 months for BoNT and was 48 ± 33 months for myotomy [36]. A Cochrane collaborative meta-analysis compared BoNT to Rigiflex balloon dilation in patients with primary achalasia. In the six reviewed studies involving 178 patients, there was no significant difference in short term (within 4 weeks of injection) responses, but the 6- and 12-month responses were significantly better with dilation, with 74 % failure by BoNT compared to 30 % by dilation at 1 year follow-up [37]. Similarly a large systematic review and meta-analysis of 105 articles involving 7855 patients showed better symptomatic relief with laparoscopic myotomy combined with antireflux procedure than BoNT injection [38]. One randomized control study studied the effect of injecting BoNT 1 month prior to PD, and showed a tendency towards a greater response with pre-dilation BoNT (77 % in remission) compared to dilation alone (62 %) at 1 year follow-up, but this did not reach statistical significance. However, a second dilation upon relapse of symptoms increased the remission rate to 100 % in the BoNT-PD group compared to 85 % for PD alone [39].

An early non-randomized study comparing BoNT to laparoscopic Heller myotomy showed that both BoNT and myotomy improved dysphagia score and reduced LES nadir pressure at 2 months post BoNT and 6 months post myotomy, but only myotomy significantly reduced LES basal pressure and improved esophageal barium clearance [40]. Subsequently, a randomized control trial comparing BoNT to laparoscopic myotomy showed similar response rates at 6-month follow-up, but marked difference at 1-year follow-up (i.e., 53 % remission in BoNT vs. 90 % remission in myotomy group) [41].

A review of randomized controlled trials comparing different treatment options concluded that endoscopic BoNT should be considered mainly when other treatments are contraindicated [32]. Similarly, guidelines from both the American College of Gastroenterology [42] and the American Gastroenterology Association [43] emphasized that BoNT should not be used as first line treatment for patients who would otherwise be candidates for either PD or surgical/endoscopic myotomy. One concern is that BoNT may increase inflammation in the mucosa and muscle planes, and lead to greater difficulty for subsequent myotomy. Studies in pigs showed fibrotic changes in the LES after BoNT injection [44]. However, most histological studies in humans show that there are no significant changes in muscle [45], and post-surgical studies show no significant differences in outcome after BoNT treatment [46]. One abstract with limited number of cases suggests that a prior BoNT may increase perforation risk of subsequent PD [47]. Other reports suggest that prior BoNT may make surgical myotomy more difficult [48, 49]. While the factors explaining these observational differences remain unclear, BoNT should be reserved for those with significant co-morbidities, and for those who would poorly tolerate medical therapy, dilation, or complication of dilation. However, repeated BoNT treatments can approximate the short-term (2-year) response rates of PD [50].

When patients are given treatment choices upon diagnosis of achalasia, almost a third of patients in one study chose to have no treatment for the first year, and only a minority of these patients underwent treatment ultimately, although these patients generally showed worsening symptom score [51]. Clearly there are no definitive treatments for achalasia, and current therapy options are aimed to improve symptoms and relieve the LES functional obstruction. While most patients prefer non-surgical means, attempts to identify patients who have better response to BoNT injections have had varying results. In general, several studies have shown that elderly patients [52, 53] and those with vigorous achalasia had better responses [33, 54, 55]. There is no specific study addressing which patients are more likely to respond to BoNT injection based on the Chicago Classification of subtypes, although generally type II (with at least 20 % of liquid swallows with a body pressurization > 30 mmHg) has better symptom response to other treatments compared to “classic achalasia” (Type I) and “spastic achalasia” (type III). However, one report finds that no response was seen in patients with IRP <15 mmHg [56].

In light of oral pharmacologic agents falling out of use because of poor efficacy and frequent side effects, BoNT injection offers an appealing alternative to surgery based on ease and safety. However, there is wide variation in reported durability of response, and further studies are required to determine the optimal injection protocol and the best patients for BoNT therapy. The repeatedly demonstrated durability of response with PD and myotomy over BoNT makes the latter reserved largely for frail or elderly patients. However, the role of BoNT or similar injectable agents as treatment for patients who failed PD or myotomy, or as adjuvant therapy to re-dilation remains unclear. Future agents for local injection may include depot formulations of BoNT, other inhibitors of neurotransmission, long-acting nitric oxide donors, and cell-based therapies including engineered stem cells.