Botulinum toxin in the therapy of gastrointestinal motility disorders


Professor D. Gui, Department of Surgery, Catholic University, L.go A. Gemelli, 8, 00168 Rome, Italy.


Since 1980, botulinum toxin has been employed for the treatment of various voluntary muscle spastic disorders in the fields of neurology and ophthalmology. More recently, botulinum toxin has been proved to be effective in the therapy of dyskinetic smooth muscle disorders of the gastrointestinal tract. Achalasia and anal fissure are the gastrointestinal disorders in which botulinum toxin therapy has been most extensively investigated. Botulinum toxin is the best treatment option for achalasia in patients whose condition makes them unfit for pneumatic dilation or surgery. In anal fissure, botulinum toxin is highly effective and may become the treatment of choice. In the future, botulinum toxin application in the gastrointestinal tract will be extended to many other gastrointestinal disorders, such as non-achalasic motor disorders of the oesophagus, dysfunction of Oddi's sphincter, achalasia of the internal anal sphincter and others. This article describes the mechanism of action, rationale of employment, indications and side-effects of botulinum toxin application in smooth muscle disorders of the gastrointestinal tract, and compares the results of different techniques of botulinum toxin therapeutic application.


Botulinum toxin (Btx) was identified by the German poet and doctor, Justinus Kerner, in 1817 as a ‘fatty acid’ with characteristics of a skeletal muscle and parasympathetic function inhibitor. In his pharmacological text, Kerner assigned the title, ‘About the fatty acid as a possible therapeutic drug’, to one of the chapters, suggesting the future therapeutic use of Btx in clinical settings characterized by spastic phenomena, hyperhidrosis and hypersalivation.1 In 1897, van Ermengen, a Belgian microbiologist, isolated an anaerobic bacterium producing a lethal toxin, which was named Clostridium botulinum.2 One hundred years passed before Alan Scott pioneered the clinical use of Btx for the treatment of ocular strabismus.3

Btx has found numerous fields of application in neurology (spasmodic torticollis,4 laryngeal dystonia,5 writer's cramp,6 tremors,7 hemifacial spasm,8, 9 tics, segmental myoclonus and muscular spasticity7), ophthalmology (blepharospasm,10, 11 strabismus,12 lateral rectus muscle paralysis7, 12 and nystagmus13) and aesthetic medicine.14

However, Btx is also active on smooth muscle, as initially suggested by the gastrointestinal symptoms of botulism, i.e. constipation and vomiting. In 1994, the toxin was proposed for the treatment of achalasia15 and anal fissure,16 which both ensue from smooth muscle spastic conditions of the gastrointestinal tract.

Many studies have been conducted, and the toxin is currently deemed the therapy of choice for subsets of patients with achalasia and anal fissure. Recently, other gastrointestinal applications of the toxin have been proposed. The aim of the present review is to examine the state-of-the-art clinical use of Btx in dyskinetic disorders of the gastrointestinal tract. Data for the study were acquired from systematic literature searches of the Medline database for indexed articles. The grey literature present in the abstracts of the clinical congresses of the American Gastroenterological Association and the American College of Gastroenterology was also reviewed.

Mechanism of action of botulinum toxin

Botulinum neurotoxins produced by the anaerobic bacterium Clostridium botulinum are divided into seven subgroups, identified by the letters A–G. Only subtypes A, B and E have been implicated in human botulinum intoxication.17 Subtype A (Btx-A) is the most frequently employed in clinical practice. The toxin is composed of a heavy protein chain (100 kDa) bound to a light chain (50 kDa) by a disulphide bond.18 The toxin enters the cholinergic nerve endings through a membrane receptor; the light chain is liberated in the cytosol and hydrolyses the SNAP-25 protein which is part of the SNARE complex, a set of proteins docking and fusing the vesicle membrane to the plasmatic membrane (fusion machine) allowing the exocytosis of neurotransmitters.19, 20

The nerve cell remains active, and neuroexocytosis is reactivated in a few weeks after the biosynthesis of new SNAP-25 molecules and/or after the formation of new neuromuscular junctions (sprouting mechanism).

The information available on the mechanism of action of Btx has been obtained from studies carried out on skeletal muscle. Information on the smooth musculature is largely incomplete. In particular, only a few experimental studies have been performed on the smooth muscles of the gastrointestinal tract, demonstrating the presence of SNAP-25 at the myoenteric plexus level (D. Gui, 2002, unpublished data) and the action of Btx on the cholinergic transmission of the ‘enteric nervous system’.21–23

Cricopharyngeal achalasia

The cricopharyngeal muscle, or upper oesophageal sphincter, is the most inferior portion of the inferior constrictor muscle, and separates the hypopharynx from the oesophagus. Cricopharyngeal achalasia can be primary or secondary to a variety of neurological conditions, and the major symptom is proximal dysphagia. Pharyngeal (Zenker's) diverticula are commonly observed. Treatments include cricopharyngeal myotomy and mechanical dilation.24 The success rate for myotomy is 79%,25 whereas for other techniques the data are incomplete.

To date, 76 patients with cricopharyngeal achalasia treated with Btx-A injection into the cricopharyngeal muscle have been reported using different techniques: percutaneous (guided by electromyography, computed tomography or videofluoroscopy), direct injection during oesophagoscopy (rigid or flexible) or other endoscopic techniques.26–39 The results of the 14 studies present in the literature are reported in Table 1. The average success rate was 76% and no major side-effects were reported. Symptom improvement lasted from 2 months up to 1 year; some authors performed repeated injections with success.

Table 1.  Botulinum toxin (Btx) for cricopharyngeal achalasia — indexed literature
Reference Treatment* Patients (n)AetiologyTechniquesExcellent/good symptomatic improvement (%)
  • EMG, electromyography; HIV, human immunodeficiency virus; CT, computed tomography.

  • *

     In the treatment with Btx-A, two different preparations were used: Botox (Allergan) and Dysport (Ipsen); the dosage in international units is given in parentheses.

  •  Percentage of patients who showed excellent/good symptomatic improvement after endoscopic injections of Btx-A.

Moerman et al.26Botox (100 U)4Post-laryngectomy,  post-radiochemotherapy,  cricopharyngeal  hyperfunctionLaryngopharyngoscopy100
Parameswaran  and Soliman27Botox (10–30 U)11Post-laryngectomy,  post-radiochemotherapy,  cricopharyngeal  hyperfunction,  stroke, amyotrophic  lateral sclerosis,  Parkinson's diseaseEndoscopy73
Restivo et al.28Dysport (30 U)4Parkinson's diseasePercutaneous + EMG100
Haapaniemi et al.29Botox (28–100 U)4Stroke, inclusion body myositis,  peripheral motor neuropathyEndoscopy75
Shaw and Searl30Botox (25–50 U)12Stroke, peripheral neuropathy,  post-laryngectomy/pharyngectomy,  cricopharyngeal hyperfunctionRigid oesophagoscopy  + EMG or flexible  endoscopy  or open surgery67
Ahsan et al.31Botox (40–100 U)5Stroke, HIV infection,  post-radiation therapy,  post-carotid endoarterectomyEndoscopy with microlaryngoscope80
Restivo et al.32Dysport (30 U)1Oculopharyngeal  muscular dystrophyPercutaneous + EMG100
Alberty et al.33Botox (100 U)10Stroke, polymyositis,  cricopharyngeal hyperfunctionRigid oesophagoscopy + EMG70
Brant et al.34Botox (100 U)1StrokeFlexible oesophagoscopy100
Blitzer and Brin3520 U6Stroke, post-laryngectomy,  cricopharyngeal hyperfunctionPercutaneous + EMG83
Atkinson and Rees36Botox (5–20 U)5Stroke, brain tumour,  cricopharyngeal hyperfunctionPercutaneous + CT80
Crary and  Glowasky3725–30 U5Post-laryngectomyPercutaneous + fluoroscopy80
Schneider et al.38Dysport (80–120 U)7Brain tumour, laryngeal tumour,  strokeRigid oesophagoscopy  + EMG71
Dunne et al.39Botox (16 U)1Laryngeal dystonyPercutaneous  + EMG + fluoroscopy100
Btx-treated patients 76  76

The best results were observed in patients with ‘pure’ cricopharyngeal spasm and Zenker's diverticula, whereas a less substantial response was reported in patients with stroke and post-surgical conditions, who presented with more complex swallowing disorders.

Local injection of Btx into the cricopharyngeal muscle is a simple, safe and effective means of improving the associated dysphagia. The toxin may be useful in identifying those patients most likely to benefit from cricopharyngeal myotomy. Indeed, if the toxin fails to improve the symptoms weakening the cricopharyngeal muscle, myotomy is unlikely to be of significant benefit; if the toxin improves the symptoms, the patient will face the possibility of repeated injections approximately every 4 months or myotomy.


Achalasia is a severe neuromuscular disorder of the oesophagus, characterized by the loss of peristalsis and an inability of the lower oesophageal sphincter to reach optimal relaxation.40, 41 This causes functional obstruction of the sphincter with the retention of ingested food and saliva in the oesophageal lumen. The main symptom is dysphagia, which may be accompanied by regurgitation, thoracic pain and a reduction in body weight.

The aetiopathogenesis of achalasia is unknown, but the early alterations involve an inflammation of the myoenteric plexus with a consequent loss of ganglial cells and fibrosis.42 Degenerative modifications (retrograde?) of the vagal nerves and dorsal vagal nuclei have also been described. Inhibitory neurones, which use substances such as vasoactive intestinal polypeptide and nitric oxide for neurotransmission, are selectively and functionally decreased,43, 44 whereas excitatory cholinergic innervation is maintained.45 A disequilibrium between excitatory and inhibitory activity leads to an increase in muscular tone and, consequently, an increase in the lower oesophageal sphincter pressure. The sphincter no longer responds to the physiological stimuli of muscular relaxation, such as swallowing.40, 41 This condition of the absence of peristalsis is probably due to the lack of an inhibitory innervation, although this pathogenetic aspect has not been elucidated fully.

Pharmacological therapy of achalasia (nifedipine, nitroglycerin, isosorbide dinitrate, anticholinergic agents, beta-adrenergic agonists) has not provided satisfactory results to date.46

Forceful pneumatic dilation of the distal oesophagus is the most frequently applied non-surgical therapy. This technique produces a controlled tear of the lower oesophageal sphincter, resulting in the relief of the functional obstruction. A long-term, good to excellent clinical response is induced in 60–90% of cases, the lower oesophageal sphincter pressure is reduced by 39–68% and the risk of perforation lies in the range 2–3% in different studies.47–49 The perforation rate can be reduced to less than 2% with the use of inelastic balloons.50, 51 Relative contraindications to pneumatic dilation are represented by epiphrenic diverticula, previous oesophageal perforation and aneurysm of the thoracic aorta.52

Heller's cardiomyotomy is the standard surgical therapy, and involves a longitudinal incision on the serosal surface of the oesophagus spanning a few millimetres under the gastro-oesophageal junction to 5–7 cm above this structure.53 This operation can be performed by open surgery or minimally invasive surgery (trans-abdominal or trans-thoracic). The overall results of open surgery include good to excellent symptom improvement in 89% of patients, a reduction in the lower oesophageal sphincter pressure of 74%, mortality in 0.2% and gastro-oesophageal reflux disease in up to 22% when an anti-reflux procedure is not performed simultaneously.50, 51 Most surgeons associate myotomy with some kind of anti-reflux technique. Laparoscopic surgery has shown significant advantages in terms of reduced post-operative pain and a shorter hospital stay. Good to excellent symptom improvement is obtained in an average of 94% of patients (range, 83–100%), the reduction in the lower oesophageal sphincter pressure is 59% and the incidence of gastro-oesophageal reflux disease is 11%. Anti-reflux techniques can be applied simultaneously. The reported mortality approaches zero. The thoracoscopic approach is being abandoned as a result of the high rate of persistent dysphagia (18%) and secondary gastro-oesophageal reflux disease (50%).50, 51, 54, 55

Botulinum toxin therapy in achalasia

Btx-A in the treatment of achalasia was introduced by Pasricha et al.15, 56, 57

The technique involves endoscopic injection with a 5-mm sclerotherapy needle of Btx-A at established doses diluted in 1 mL of saline in each of the four lower oesophageal sphincter quadrants. In order to identify the lower oesophageal sphincter, trans-oesophageal echography has been advocated. However, the best results are obtained at the expense of higher costs and more sophisticated equipment.58 Since 1995, several trials have been carried out to establish the effectiveness of Btx-A therapy and its possible role amongst other therapeutic options. Table 2 reports the data from all 19 articles published in international journals according to the Medline database,57, 59–76 and Table 3 reports the data from 19 abstracts published in Gastroenterology and the American Journal of Gastroenterology between 1995 and 2002.58, 77–94 Of these publications, 12 compare Btx-A with dilation therapy (one also compares it with surgical therapy) and one studies the combination of Btx-A and pneumatic dilation. The Btx-A doses employed vary from 50 to 200 U, but most frequently each quadrant of the lower oesophageal sphincter receives an injection of 20–25 U, for a total dose of 80–100 U. In two studies, instead of Botox (Allergan Inc., Irvine, CA, USA), Dysport (Ipsen, Milan, Italy) was used at doses of 240–250 U (1 U of Botox is approximately as effective as 3 U of Dysport). Grouping all the studies, 831 patients have been treated with Btx-A, 290 with pneumatic dilation and 52 with cardiomyotomy. The fourth column in Table 2 shows the percentage of patients who had already undergone some other form of treatment for achalasia.

Table 2.  Botulinum toxin (Btx) for achalasia — indexed literature
ReferenceTreatment*Patients (n)Pre-treated patients (%)Excellent/good symptomatic improvement (%)Mean reduction of LESP at 1 month (%)§Side-effects (%)
1 month6 months12 months
  • LESP, lower oesophageal sphincter pressure; PD, pneumatic dilation.

  • *

     In the treatment with Btx-A, two different preparations were used: Botox (Allergan) and Dysport (Ipsen); the dosage in international units is given in parentheses. For PD, the dilator used and the diameter of the balloon are given in parentheses.

  •  Percentage of patients who underwent other treatments (PD, cardiomyotomy) before Btx-A injections.

  •  Percentage of patients who showed excellent/good symptomatic improvement 1, 6 and 12 months after endoscopic injections of Btx-A.

  • §

     Percentage reduction in LESP 1 month after endoscopic injections of Btx-A.

  •  Side-effects after Btx-A injections were mild and transient: chest pain; gastro-oesophageal reflux; cutaneous rash.

  • **

     Value at 24 months after treatment with Btx-A.

  • §§

     Value at 25 months after treatment with Btx-A.

  • ¶¶

     Value at 1 week after treatment with Btx-A.

  • ***

     Value at 3 months after treatment with Btx-A.

  • †††

     Value at 22 months after treatment with Btx-A.

  • ‡‡‡

     Oesophageal perforation.

  • §§§

     Randomized trial vs. treatment other than Btx-A.

Neubrand et al.59Botox (100 U)25 064 39§§− 31¶¶
Storr et al.60Botox (100 U)40357777*** − 5510
Allescher et al.61Botox (100 U)       
Dysport (100–300 U)232283− 47
PD (Rigiflex 3.5)14 779− 55
Mikaeli et al.62§§§Dysport (200 U)20 01008560− 240
PD (Rigiflex 3.0)19 0897953− 260
D'Onofrio et al.63Botox (100 U)37 08483†††− 3011
Ghoshal et al.64§§§Botox (60–80 U)72986− 530
PD (Rigiflex)10 080− 6220
Annese et al.65Botox (100–U × 2)38 88481− 305
Botox (50 U)40 77553− 327
Botox (200 U)40 7889357− 3410
Vaezi et al.66§§§Botox (100 U)22 0643632− 6
PD (Rigiflex 3.0)20 0757570− 665‡‡‡
Muehldorfer et al.67Botox (80 U)125875− 44
PD (Pentax 4.0)1266830− 5117
Kolbasnik et al.68Botox (80 U)30277757397
Wehrmann et al.69Botox (100 U)20807520
Prakash et al.70Botox (80 U)42 080419
PD (Rigiflex 3.0–3.5)26 090
Annese et al.71Botox (100 U)57178875**− 55
Cuillière et al.72Botox (80 U)55337565− 3422
Gordon and Eaker73Botox (80 U)16317543
Annese et al.74§§§Botox (100 U)8 0100− 4912
PD (Rigiflex 3.0)8 0100− 7225
Fishman et al.75Botox (80–100 U)605070367
Pasricha et al.57Botox (80 U)31559055− 4523
Rollan et al.76Botox (80–100 U)310067− 6433
Btx-treated patients 517 806754− 3811
Range  0–6664–10036–9320–83− 6 to − 640–33
Table 3.  Botulinum toxin (Btx) for achalasia — grey literature
ReferenceTreatment*Patients (n)Pre-treated patients (%)Excellent/good symptomatic improvement (%)Mean reduction of LESP at 1 month (%)§Side-effects (%)
1 month6 months12 months
  • LESP, lower oesophageal sphincter pressure; Nr, not reported; PD, pneumatic dilation.

  • *

     In the treatment with Btx-A, two different preparations were used: Botox (Allergan) and Dysport (Ipsen); the dosage in international units is given in parentheses. For PD, the dilator used and the diameter of the balloon are given in parentheses.

  •  Percentage of patients who underwent other treatments (PD, cardiomyotomy) before Btx-A injections.

  •  Percentage of patients who showed excellent/good symptomatic improvement 1, 6 and 12 months after endoscopic injections of Btx-A.

  • §

     Percentage reduction in LESP 1 month after endoscopic injections of Btx-A.

  •  Side-effects after Btx-A injections were mild and transient: chest pain; gastro-oesophageal reflux; cutaneous rash.

  • **

     Value at 3 months after treatment with Btx-A.

  • ††

     Value at 24 months after treatment with Btx-A.

  • ‡‡

     Oesophageal perforation.

Anant et al.77Botox (80–100 U)9100− 45 0
+ PD (TTS 2.5)       
Benvenuti et al.78Botox (100 U)14 21 9364− 40
Annese et al.79Botox (100 U)34 88
Dysport (250 U)33 79
Mora et al.80Botox (80 U)14 8847− 24** 0
Nebendahl et al.81Botox (100 U)8 88− 3875
PD (Nr)8 88− 4987
D'Onofrio et al.82Botox (100 U)13 85− 30
Achem et al.83Botox (100 U)10100 707060
Vakil et al.84Botox (100 U)13 9292 8
Kozarek et al.85Botox (50 U)13 3846
Botox (100 U)19 4263
Stein et al.86PD (Nr)141 715‡‡
Botox (Nr)13 54
Cardiomyotomy (Nr)52 8711
Hoffman et al.58Botox (80 U)25 36 8812
Bansal et al.87Botox (80 U)1250− 57
PD (Witzel)989− 45
Greaves et al.88Dysport (240 U)1164**36†† − 3
Gaudric et al.89Botox (80 U)7 43− 32
PD (Rigiflex 3.5)5100− 54
Schroeder et al.90Botox (100 U)11 64 − 927
PD (Rigiflex 3.0)9 67− 68 0
Jin et al.91Botox (80 U)10100 90− 41
Kotfila et al.92Botox (80–100 U)7 71 86 28
Waterfall et al.93Botox (80 U)16  0 676767− 7275
Hills et al.94Botox (80 U)31 84
Btx-treated patients 314  816953− 4032
Range  0–10043–9350–9236–67 − 3 to − 720–75

In order to evaluate the results of therapy, the authors considered symptomatic improvement as a subjective parameter and the percentage decrease in the lower oesophageal sphincter pressure at 1, 6 and 12 months as an objective parameter. In Table 2, good to excellent results in symptomatic improvement were obtained, with an objective reduction in the intensity and frequency of dysphagia, belching and retrosternal pain (expressed using a similar non-standardized scale in different studies).

When the results from the reviewed studies and the congress abstract are combined (Tables 2 and 3, weighted averages), about 80% of the patients treated with Btx-A improved subjectively within 1 month after treatment; this percentage decreased to 67–69% at 6 months and 53–54% at 12 months. In most studies, manometric measurements were registered only 1 month after treatment; at this time, the lower oesophageal sphincter pressure was reduced by about 36–38%. After 6 months, the lower oesophageal sphincter pressure had increased again, with a reduction of about 29% compared to basal values. After 12 months, the data are few and highly variable, and demonstrate a decrease in the lower oesophageal sphincter pressure in the range 12–73%. The side-effects were mild and infrequent (11–32%) and included transitory chest pain, gastro-oesophageal reflux and cutaneous rash. Other single case reports have shown some major side-effects, including gastroparesis,95 urinary retention,96 heart block97 and ulcerative oesophagitis with evidence of gastro-oesophageal reflux;98 the pathogenesis of these side-effects and their link with Btx-A injections are not clear. In general, the results from the reviewed studies and from the ‘grey literature’ (congress abstracts) are remarkably similar. If we consider the four randomized studies (total of 57 patients) comparing Btx-A therapy with pneumatic dilation, the mean percentages of remission at 1, 6 and 12 months are 88%, 61% and 46%, respectively, with a decrease in the lower oesophageal sphincter pressure of 26% at 1 month. Side-effects were present in 4% of patients treated with Btx-A.

From a comparison of the dose–effect relationship in the different studies, it can be concluded that doses of 200 U are not substantially more effective than the usual dose of 80–100 U.

The main advantages of Btx-A therapy are its simplicity, the fewer complications of the technique, the opportunity to treat patients on an out-patient basis and the rapid resumption of oral alimentation.

Patients over 50 years of age and those with a diagnosis of vigorous achalasia appear to respond best to Btx-A.57 Previous pneumatic dilation or cardiomyotomy does not seem to impair the effects of Btx-A treatment,57 but there is still debate as to whether or not previous Btx-A treatment impairs the results of more invasive therapies.99, 100 Recent studies have shown that previous treatment with Btx-A makes the surgical procedure more difficult, even if the clinical results are not worsened.55

Low-risk patients can choose between pneumatic dilation or cardiomyotomy (the surgeon's experience is also important in the choice), but, in the case of failure, Btx-A treatment can be performed. In contrast, Btx-A treatment is considered to be the gold standard in high-risk patients (the elderly, those with co-morbidity and those with contraindications to pneumatic dilation, such as oesophageal perforation or epiphrenic diverticula). Pneumatic dilation is the least expensive therapeutic option ($3111), followed by Btx-A ($3723) and cardiomyotomy ($10 792).101

Non-achalasia oesophageal motility disorders

A therapeutic role of Btx-A has been reported in the literature in other dyskinetic diseases of the oesophagus, such as diffuse oesophageal spasm, isolated lower oesophageal sphincter hypertension and other aspecific disorders. Btx-A was employed by Miller et al. in 15 patients unresponsive to medical therapy; in this study, 20 U of Btx-A was injected into each of the four quadrants of the lower oesophageal sphincter using a sclerotherapy needle. Good or excellent symptomatic improvement was obtained in 73% of cases at 1 month after treatment, with 33% of patients presenting a good to excellent response at the last interview (mean follow-up of 10.6 months).102 Storr et al. injected 100 U of Btx-A diluted in 10 mL of saline into multiple sites along the oesophageal wall of patients suffering from diffuse oesophageal spasm. Eight of the nine treated patients (89%) showed symptomatic improvement after 1 and 6 months. Four of these patients needed reinjections 8, 12, 15 and 24 months after the initial treatment.103 Mathews et al. treated nine patients with a hypertensive lower oesophageal sphincter (lower oesophageal sphincter pressure, > 35 mmHg) with 100 U of Btx-A injected into the lower oesophageal sphincter. Dysphagia resolved in five patients (55%) and, in four of these five, the lower oesophageal sphincter pressure decreased by more than 35%.104

Dysfunction of the sphincter of oddi

The injection of Btx-A in Vater's papilla is useful from a diagnostic and therapeutic viewpoint.105–109

In rare cases, dysfunction of the sphincter of Oddi can cause acute recurrent pancreatitis. It is also deemed to be responsible for 10–20% of recurrent right upper quadrant and epigastric pain occurring in 10% of post-cholecystectomy settings. The hypertonus of Oddi's sphincter may result from stenosis or spasm of the sphincter, obstructing bile drainage. High pressure levels are relieved by endoscopic sphincterotomy, which is regarded as the treatment of choice for this condition; unfortunately, in many patients, the symptoms continue despite sphincterotomy, and this technique carries the risk of pancreatitis, bleeding and perforation. Balloon dilation or stenting involves a higher rate of complications and failure.

Endoscopic manometry to assess Oddi's sphincter pressure is technically complex, produces a considerable number of false positive (17%) and false negative (33%) results and can lead to post-manometry pancreatitis (31%). As the tonus of Oddi's sphincter is influenced by cholinergic innervation, endoscopic injection of Btx (20–100 U) produces a 50% reduction in the basal tonus according to various reports.105–109 This effect results in the long-term relief of biliary symptoms and allows the identification of subjects who will benefit from definitive sphincter ablation, avoiding the specific risks of endoscopic manometry.

Moreover, the symptomatic response to Btx-A injection, which can last for a period of 6 weeks to 6 months, may suggest a therapeutic role of the toxin in the management of acute episodes. Clinical trials are required to investigate this hypothesis.

Anal fissure

Anal fissure is a frequent, greatly incapacitating condition, affecting both sexes in the young, otherwise healthy, population. The prevalence of this disorder is unknown and it is hypothesized to often remain undiagnosed.110 Symptoms include intense post-defecatory pain lasting for many hours, with or without bleeding. The first lesion is a mucosal tear, usually in the posterior commissure. Acute fissures tend to heal spontaneously, but, in one case in three, they evolve into a chronic ulcer that rarely heals spontaneously (< 10%).111 Typical signs of a chronic anal fissure are the hardening of ulcer margins, the exposure of muscular fibres at the level of the internal anal sphincter and a hypertrophic tag (sentinel haemorrhoid).

The pathogenesis of anal fissure has not been elucidated completely.112 Constipation and traumatic voiding are usually considered to be possible causes, with gender, hormonal influences, etc. deemed to contribute as ‘predisposing’ factors.

The mucosal lesion causes intense pain. The reflex action of pain has been hypothesized to cause spastic contraction of the internal anal sphincter, with an anal pressure that remains persistently elevated, leading to compression of the small arterial vessels that run through the muscular fibres of the internal anal sphincter. The blood shortage to the mucosa impairs the healing of the lesion, resulting in a vicious circle of chronic anal fissure. The posterior commissure has been demonstrated to be the least perfused area of the anus.113–115

Current therapies for chronic anal fissure aim at the reduction of the tonus of the internal anal sphincter, interrupting the vicious circle of pain and spasm. Anal dilation techniques have been progressively abandoned because of poor results (healing in 40–95%; recurrence in up to 56.5%; side-effects: 20–39% incontinence or soiling).116 Lateral sphincterotomy is the most frequent surgical technique applied today (success rate, 71–97%; recurrence, 8%; side-effects: risk of permanent defect of anal shape; gas incontinence, 0–36%; soiling, 0–21%; faecal incontinence, 0–9%).117–122

Topically applied nitroglycerin cream (glyceryl trinitrate, 0.2%) may promote the healing of anal fissures by increasing the local blood flow (healing rate, 46–88%; recurrence, 12–45%; side-effects: 33–84% headache).123–135 Isosorbide dinitrate, another nitric oxide releaser, has been topically employed at doses of 2.5–5 mg with a reported healing rate of 80–85%.136–138 Nitrates have shown a brief effect and frequent applications are required; tachyphylaxis also limits their utility. Nifedipine has been prescribed both orally (20 mg b.d., 60% healing rate) and topically (0.2% b.d.).139–141

Botulinum toxin therapy in anal fissure

‘Chemical sphincterotomy’ with Btx-A injected into the internal anal sphincter has the advantage of a long-term effect, minimal invasiveness and complete reversibility, with a healing rate similar to that offered by surgery. We have found 13 studies in the literature reporting Btx-A in the therapy of chronic anal fissure (Table 4) with data from 632 patients in total.16, 142–153 The toxin was injected into the internal anal sphincter (457 patients, nine studies) or the external anal sphincter (175 patients, four studies). The results of three congress abstracts are also reported in Table 4.154–156 Only two of the studies reported in Table 4 are randomized, and the data were often incomplete and could not be analysed separately.

Table 4.  Botulinum toxin (Btx) for anal fissure — indexed/grey literature
ReferenceDose (U) and site of injectionPatients (n)Healing rate (%)Re-injection (%) and dose (U)Complete treatment healing rate (%)Recurrence (%)MRRP (%)*MRSP (%)*
1 m2–3 m1 m2 m1 m2 m
  • AM, anterior midline; D, Dysport (where not indicated, Botox was used); EAS, external anal sphincter; GTN, glyceryl trinitrate; IAS, internal anal sphincter; m, months; MRRP, mean reduction in resting pressure; MRSP, mean reduction in squeeze pressure; Nr, not reported; PM, posterior midline.

  • * Decrease in the mean values of anal manometry, carried out at rest and after a maximal voluntary contraction, in comparison with basal values; two studies report an increase in values (‘+’ in the table).

  •  Thirty-three patients received re-injection in the second month.

  •  Group I: 5 U Botox on each side of the anal sphincter close to both lateral midpoints. Group II: an additional 5 U injection below the fissure. Group III: 7 U for each injection.

  • §

     The value refers to the symptomatic improvement rate only.

  • ¶Dose not reported.

  • **

     Randomized trial vs. treatment other than Btx-A.

Brisinda et al.14220, IAS75738910.7 (30 U)100430303.50
30, IAS7587
 (P = 0.04)
964 (50 U)10043234.3157.8
Maria et al.14320, IAS25486024 (25 U)80022.622.44.64.8
 (each side of PM)          
20, IAS25888812 (25 U)100030.931.91.81.8
 (each side of AM)  (P = 0.027)  (P = 0.025)       
Brisinda et al.144**20, IAS2588960026.328.95.74.6
 (each side of AM)  (P < 0.001)  (P = 0.005)       
GTN 0.2%254060016.713.8+ 2.3+ 8.5
Fernandez Lopez  et al.14540, IAS76566745.2 (40 U)67Nr
 (each side of fissure)          
Minguez et al.14610, IAS23 (48)§Nr5283Nr4.9Nr16.9Nr
15, IAS‡27 (74)§Nr3078Nr13.1Nr18.4Nr
21, IAS‡19 (100)§Nr3790Nr16.1Nr34.2Nr
Jost and Schrank  et al.14720, D, EAS25Nr764
 (each side of fissure)          
40, D, EAS25Nr808
 (each side of fissure)          
Madalinski  et al.14820, EAS1384.6NrNrNr15.4
Maria et al.14915, IAS2321.743.58.7 (20 U)86.7027.715.918.212.1
 (each side of fissure)          
20, IAS345067.620.6 (25 U)100027.928.814.311.9
 (each side of fissure)  (P = 0.029) (P = 0.061)       
Maria et al.150**20, IAS1553.373.326.6 (25 U)1006.72725.715.59.9
 (each side of fissure)  (P = 0.05) (P = 0.003)       
Jost et al.1512.5/5, EAS100Nr828Normal limits   
 (each side of fissure)          
Mason et al.152D, IAS5Nr60Nr
Gui et al.1615, IAS10607040 (20 U)901023.97.8+ 0+ 13.1
 (each side of fissure)          
Jost and5, EAS12Nr83.38.3Within normal
Schimrigk153 (each side of fissure)       limits in 3 months   
Btx-treated  patients 63267.179.2 89.34.625.728.411.15.1
Range  21.7–8743.5–954–5278–1001–15.43.9
+ 13.1
Congress abstracts
Siproudhis et al.154200, D, IAS222339
Espi et al.15510, IAS3665 0
15, IAS 81 0
Birkenfeld et al.15610, Nr10 Within

One month after treatment, the studies reported a healing rate of 21.7–88%, the lowest with Btx-A (15 U) injected into the internal anal sphincter (half the dose on each side of the fissure), and the highest with 20 U injected into the internal anal sphincter (each side of the anterior midline, two different studies). Two months after treatment, healing was obtained in 43.5–96% of cases; few studies reported follow-up for more than 2 months.

In seven studies, patients who showed no benefit (4–52%) were re-injected, mostly with a higher toxin dose, in order to maximize the results.

The percentages in column 6 of Table 4 (complete treatment healing rate) refer only to those studies in which a healing rate of 78–100% was obtained. In general, when different doses were compared, the best results were achieved with the higher doses. Ten of 639 patients (9.5%) underwent lateral internal sphincterotomy during the study. The weighted average of the reported recurrence rate is approximately 4.6% (follow-up range, 6–30 months), but recently Minguez et al. reported a late recurrence rate (42-month follow-up) of 41.5%;157 they identified the highest risk of recurrence in patients reporting an anterior location of the fissure, prolonged illness, the need for re-injection and for high doses to achieve healing, and a lower decrease in the maximum squeeze pressure after treatment.

Manometry (in nine studies, all injecting the internal anal sphincter) showed a significant weighted average reduction (25.7% of the basal value at the first month and 28.4% at the second) of the resting anal pressure, which is a measure of the internal anal sphincter activity; the squeeze pressure (i.e. strength of internal + external anal sphincter) was reduced by 11.1% in the first month and 5.1% in the second month. Complications were remarkably rare (0.8% overall) and minor: haemorrhoidal thrombosis, circumscribed subcutaneous infection, small haematoma of the anal region. Systemic side-effects were not reported, even with higher doses. Episodes of transitory gas incontinence and soiling were described in 3.4% of patients. A recent extensive prospective study of the side-effects of Btx-A for benign anal disorders has shown similar results.158 Re-injection is possible in the case of failure or recurrence of chronic anal fissure. No sedation or anaesthesia is necessary.16, 142–156

At present, Btx-A injection is not a fully standardized procedure for the treatment of chronic anal fissure; indeed, the studies identified used different doses injected at different sites. The best results seem to be associated with two injections of 15 U of toxin (Botox) near the anterior midline of the anal sphincter (one each side of the midline). The association between Btx-A and topical nitrates has been debated.159–161 The preliminary results of therapy with Btx-B do not seem encouraging.162 At present, the pressure levels of the internal anal sphincter are not used as an indication for the treatment or dose setting. Although, from a pathogenetic viewpoint, chronic anal fissure is associated with spasm of the internal anal sphincter, two studies in which the toxin was injected into the external anal sphincter reported comparable results. This effect means that the role of muscular relaxation of the external anal sphincter is unclear.

One week after Btx treatment, a large number of patients reported symptomatic relief. These data indicate that Btx injection may be considered as an effective method for the treatment of acute lesions in the future.

Other applications in the gastrointestinal tract

Btx-A has been used successfully to resolve prolonged pyloric spasm after duodeno-cephalo-pancreatectomy with pylorus preservation; a case of pyloric obstruction syndrome after pyloroplasty for gastric ulcer has been treated, with resulting incomplete resolution at endoscopic controls.163

Gastroparesis is a disorder of gastric motility that results in delayed gastric emptying, early satiety, post-prandial fullness, epigastric pain, nausea, vomiting and weight loss. Recently, Btx injection (80–200 U, 20–50 U/mL) into the pyloric sphincter has been shown to improve gastric emptying and reduce symptoms in patients with idiopathic and diabetic gastroparesis.164–169 Post-Nissen's fundoplication gastroparesis was treated with intrapyloric injections, with a reduction in symptoms and a decrease in the gastric emptying time.170 However, Btx-A was used in two infants with infantile hypertrophic pyloric stenosis without success.171

Proctalgia fugax, whose pathophysiology seems to be related to anal sphincter spasm, has been successfully treated in one case using Btx-A injections.172

In constipation due to pelvic floor dysfunction, Btx seems to relieve the contraction of the puborectalis muscle, allowing the resumption of normal rectal voiding without adverse systemic effects.173–177


In the gastrointestinal tract, Btx-A has been applied to the treatment of oesophageal achalasia and anal fissure. Symptoms are relieved in both disorders, although the temporary nature of benefit in some situations (e.g. in the treatment of achalasia) remains a drawback. However, Btx-A is the treatment of choice in achalasia patients who are poor candidates for surgery or pneumatic dilation and in elderly patients with co-morbid illnesses associated with a reduced life expectancy. In anal fissure, many authors consider Btx-A as the therapy of choice as a result of its efficacy, safety and lack of invasiveness, but further studies are necessary in order to elucidate the role of the external anal sphincter in the pathophysiology of the disease. The incidence of the failure of Btx therapy varies widely in the literature. This is an aspect that remains to be resolved, as repeated injections of Btx may lead to Btx-blocking antibody formation. This phenomenon, which has been documented in neurological diseases, has not been studied in the application of Btx in the gastrointestinal tract.

Btx-A appears to be a promising therapeutic agent for a variety of gastrointestinal disorders which depend on dyskinesia or muscle hypertonus as causative agents; some treatment options are extremely exciting.

Large controlled trials are needed to demonstrate the efficacy of Btx. However, physiological studies must also be performed, because the mechanism of action of the toxin on the gastrointestinal tract has not been elucidated; this has been determined previously in the voluntary musculature. In the gastrointestinal tract, the neuronal pattern is much more complex. The toxin may impair the neuromuscular transmission of acetylcholine, but other neuromediators may be involved and it is not known whether there is a dose–response effect. The presence of the large ganglial plexus between the muscle layers (Auerbach) and in the submucosa (Meissner) is peculiar of the gastrointestinal tract, and it can be assumed that toxin injected into the muscle diffuses to the plexus, possibly exerting a neuro-nervous block at various levels (excitatory, inhibitory?). What is observed after injecting Btx into the gastrointestinal wall could be the final result of many, somewhat unpredictable, concentration-related effects on various components of the nerve and muscular tissues.