Cisapride

General adverse effects and adverse reactions

A comparison of data from prescription event monitoring in over 13 000 recipients of cisapride and from a further 9726 recipients involved in a controlled study showed that diarrhea, in about 2–4% of patients, was the commonest adverse effect reported. Other relatively common adverse effects are headache, abdominal pain, nausea and vomiting, and constipation, all in about 1–1.5% of patients.

A thorough review of the effects of cisapride in patients in intensive care suggested that adverse reactions are not likely to be severe or problematical [ ]. However, cisapride has been withdrawn because of the risk of prolongation of the QT interval and consequent ventricular dysrhythmias.

Observational studies

The effect of cisapride 20 mg bd for 7 days in preventing symptoms of gastro-esophageal reflux disease induced by a provocative meal has been assessed in 122 patients who had had symptoms suggestive of gastro-esophageal reflux for at least 3 months [ ]. Cisapride prevented or reduced the symptoms of heartburn and related symptoms, such as belching and regurgitation. Mild to moderate diarrhea was the main adverse effect.

Comparative studies

In a randomized, double-blind study in 106 patients with non-ulcer dyspepsia, cinitapride 1 mg tds was as effective as cisapride 5 mg tds in relieving symptoms [ ]. Adverse events, mainly gastrointestinal, were transient and did not require drug withdrawal. In another randomized study in 28 children (aged 5–17 years) with diabetic gastroparesis, domperidone 0.9 mg/kg/day for 8 weeks was superior to cisapride 0.8 mg/kg/day in reversing gastric emptying delay and gastric electrical abnormalities, as well as in improving dyspeptic symptoms and diabetic control [ ]. No potentially drug-related adverse effects were reported.

Functional dyspepsia is challenging to manage. Prokinetic drugs have been compared with eradication of Helicobacter pylori in 130 patients in Singapore [ ]. They took either triple therapy of lansoprazole (60 mg/day), clarithromycin (1000 mg/day), and amoxicillin (2000 mg/day) or prokinetic therapy, either cisapride (15 mg/day) or domperidone (30 mg/day). In both arms of this study there was significant symptom improvement at 1 year. Very few adverse events were reported; the most common were minor gastrointestinal disturbances in four patients in the eradication arm and three in the prokinetic arm.

Placebo-controlled studies

The effect of cisapride 10 mg qds for 5 days on the frequency of nocturnal transient lower esophageal sphincter relaxation and esophageal acid exposure has been studied in a double-blind, placebo-controlled, crossover study in 10 patients with gastro-esophageal reflux disease [ ]. Cisapride significantly reduced the frequency of transient lower esophageal sphincter relaxation during sleep and increased lower esophageal sphincter pressure without changing gastric emptying. However, in a larger double-blind, placebo-controlled, crossover study in 30 patients with gastro-esophageal reflux disease, cisapride 20 mg bd for 4 weeks (despite adequate plasma concentrations) had no significant effects on swallow-induced esophageal peristaltic activity, the basal tone of the lower esophageal sphincter, or the number of transient lower esophageal sphincter relaxations induced by gas distension of the stomach [ ]. Adverse effects were similar in the two treatment sequences, although there were a few more episodes of abdominal cramps, nausea, and headache with cisapride.

Cardiovascular

The FDA reported that between September 1993 and April 1996 it received reports of prolonged QT intervals in 23 patients and torsade de pointes or ventricular fibrillation in 34. Some proved fatal [ ].

Several later studies confirmed the finding of QT interval prolongation during cisapride therapy in children [ ]. The effects of cisapride on the QT _c interval, heart rate, and cardiac rhythm were reported in a controlled study of 83 infants aged 2–54 months who received cisapride for a minimum of 4 days for gastro-esophageal reflux and 77 controls, using continuous bipolar limb lead electrocardiography for 8 hours [ ]. The QT _c interval was significantly prolonged by cisapride in infants under 3 months old. There was no significant difference in heart rates and there were no dysrhythmias. None of the infants was receiving drugs that inhibit the hepatic metabolism of cisapride via CYP3A4.

The effect of cisapride 0.8 mg/kg/day for 14 days on the QT _c interval has been studied prospectively in 50 infants with feeding intolerance, apnea, and bradycardia episodes secondary to gastro-esophageal reflux and gastrointestinal dysmotility [ ]. In 15 infants there was prolongation of the QT _c interval at some time during the 14 days. Infants with a QT _c interval on day 3 at least two standard deviations above the mean baseline QT _c interval were more likely to develop a prolonged QT _c interval.

Over 30 drugs (for example clarithromycin, erythromycin, and troleandomycin; nefazodone; fluconazole, itraconazole, and ketoconazole; indinavir and ritonavir) and other substances (for example grapefruit juice) can interact with cisapride and enhance its dysrhythmogenic effect. Despite regulatory action, including strengthened warnings in the product information, reports of dysrhythmias have repeatedly appeared. For example, in the UK the Medicines Control Agency received 60 reports of serious cardiac adverse events between 1988 and 2000; five were fatal [ ]. The corresponding worldwide figures were 386 serious ventricular dysrhythmias with 125 deaths and 50 unexplained deaths.

• An 81-year-old woman, who had a permanent pacemaker for complete heart block with symptomatic bradycardia-dependent torsade de pointes, had breakthrough torsade de pointes during therapy with cisapride 10 mg tds for 22 days and paroxetine for 9 days [ ]. She made a good recovery on withdrawal of the drugs.

In some countries therefore the product licence was suspended [ ] and cisapride was subsequently withdrawn from the market by Janssen Cilag in 2003 [ ].

Before its withdrawal in the UK cisapride was specifically contraindicated in premature babies for up to 3 months after birth, because of the risk of QT interval prolongation [ ]. Between 1988 and 2000 the Medicines Control Agency received 64 reports of suspected adverse effects of cisapride in children under 13 years, of which two were cases of QT prolongation and two were sudden unexplained deaths. Another 106 cardiovascular events were reported from other countries, including 30 cases of QT prolongation, six cases of ventricular fibrillation or tachycardia, and four sudden unexplained deaths.

Electrocardiographic changes and predisposition to cardiac dysrhythmias were investigated in 63 children (mean age 29 months) with gastro-esophageal reflux who had taken cisapride 0.2 mg/kg tds for at least 15 days and 57 control children (mean age 27 months) who were hospitalized for other reasons and were not given cisapride or other oral treatment [ ]. All the children had an electrocardiogram performed at inclusion, and 24-hour Holter recording was performed in all children with prolonged QT intervals. When a prolonged QT interval was detected cisapride was withdrawn and a new electrocardiogram was recorded. Five children in the treatment group and six controls had prolonged QT intervals, which normalized in three of the five children after cisapride was withdrawn. Holter recording was normal in all children.

The effect of cisapride 0.2 mg/kg tds for 8 weeks on cardiac rhythm was studied in a placebo-controlled, double-blind trial in 49 children aged 0.5–4 years with gastro-esophageal reflux resistant to other medical therapy [ ]. None had underlying cardiac disease or electrolyte imbalance. Cisapride had no effect on cardiac electrical function. However, in a prospective study cisapride 2 mg/kg qds given for 72 hours to 10 premature infants caused a significant increase in the QT _c interval compared with pretreatment values [ ].

The relation between cisapride plasma concentrations, QT _c interval, and cardiac rhythm was evaluated in a controlled study in 211 infants undergoing routine 8-hour polysomnography [ ]. Cisapride was given for at least 4 days. At comparable doses of cisapride and comparable plasma concentrations, the QT _c was significantly higher in infants below 3 months of age.

In a postmarketing study of the safety of cisapride during 1993–9, 341 patients had cardiac effects, of whom 80 (23%) died, the deaths being directly or indirectly associated with a dysrhythmic event [ ]. The cardiac effects included QT interval prolongation, torsade de pointes, polymorphous ventricular tachycardia, ventricular fibrillation, ventricular tachycardia, cardiac arrest, unspecified “serious” dysrhythmias, and sudden death. In most instances the dysrhythmia occurred in the presence of risk factors such as other drugs or medical conditions.