Background. Cisapride is a prokinetic agent that facilitates gastrointestinal motility and is widely used for the treatment of gastroesophageal reflux disease (GERD) in adults and children. However, reports of ventricular proarrhythmia have been noted in patients taking cisapride, particularly in conjunction with other drugs that may inhibit hepatic metabolism of cisapride via the cytochrome P450 3A4 system.
Objective. We designed a prospective, blinded study to evaluate the effect of cisapride on ventricular repolarization in children with GERD.
Methods. We analyzed the electrocardiograms (ECGs) from 35 children (age 0.4 to 18 years, mean 5.2 years) including measurement of the resting QT interval (QTc), JT interval (JTc), as well as QT and JT interlead dispersion markers. Data from these patients were compared with ECGs from a control group of 1000 normal children.
Results. Eleven (31%) of 35 patients receiving cisapride had a prolonged QTc (≥450 ms). The JTc was prolonged ≥360 ms in 16 of 35 patients (46%). The mean QTc in the cisapride group was 428 ± 35 ms and mean JTc was 336 ± 35 ms. An increased QT or JT dispersion (>70 ms) was seen in only 3 of 35 children. Of the 11 children with QTc prolongation, 2 had documented torsades de pointes ventricular tachycardia. Both patients were taking cisapride concomitantly with a macrolide antibiotic. All other patients were treated with either cisapride alone or in conjunction with other GERD agents, such as ranitidine or omeprazole.
Conclusions. Cisapride may cause prolongation of ventricular repolarization in children. There does not appear to be increased heterogeneity of repolarization or delayed depolarization in this small sample. The proarrhythmia may be exacerbated by medications that inhibit cytochrome P450 3A4 hepatic metabolism, overdosage, or mechanisms that result in decreased serum clearance. ECG intervals should be monitored in children maintained on cisapride, particularly when used in combination with other known QT-prolonging medications.
Cisapride (Propulsid) has become a commonly prescribed drug for the treatment of gastroesophageal reflux disease (GERD) in both adults and children. Cisapride is a gastrointestinal prokinetic agent that facilitates gastrointestinal motility by significantly increasing lower esophageal sphincter pressure and improving gastric emptying. The gastrointestinal mechanism of action is thought to be attributable to enhancement of the release of acetylcholine at the myenteric plexus. In vitro studies demonstrate that one of the cardiac mechanisms of action of cisapride is related to an increase in heart rate attributable to its properties as a serotonin-4 receptor agonist.
Recent adult and pediatric case reports have suggested an association of malignant ventricular arrhythmias with administration of cisapride in conjunction with drugs that inhibit cytochrome P450 3A4 metabolism. The rhythm abnormalities include torsades de pointes associated with (and likely secondary to) QT prolongation. Some of these reported cases were fatal.1 Because cisapride is metabolized via the hepatic cytochrome P450 3A4 enzyme system, drugs that inhibit this enzyme, such as ketoconazole, itraconazole, miconazole, clarithromycin, erythromycin, fluconazole, and the nonsedating antihistamines (ie, astemizole and terfenadine) may increase serum levels of cisapride. Any interaction between cisapride and loratadine, certirizine hydrochloride, and fexofenidine is unknown. There is no known prolongation of the QT interval (QTc) when these antihistamines are taking solely. The presumed cardiotoxic mechanism involved results from prolongation of the QTc during treatment with cisapride.2-7
Several reports to the manufacturer have suggested the rare side effect of proarrhythmia may occur in patients taking cisapride.1,8These rhythm changes may have been accentuated by taking cisapride in conjunction with other drugs that may inhibit hepatic metabolism via the cytochrome P450 3A4 enzymatic system. The effects of cisapride on the pediatric electrocardiogram (ECG) have not previously been systematically described. This study was initiated to evaluate the effect of cisapride on ventricular repolarization in children and to prospectively identify children with electrocardiographic abnormalities while taking the drug.
The study was a prospective, blinded study, designed to evaluate the incidence of cisapride-associated ventricular repolarization abnormalities in children. A single 12-lead resting ECG was obtained over an 8-month period beginning in January 1996 through August 1996 on 35 children (ages 0.4–18 years, mean 5.2 years) who were currently being treated with cisapride. The children were primarily followed by the gastroenterology division. Twenty-six (74%) of 35 children were being treated for GERD. The rest of the children were being treated for pseudo-obstruction (2), esophagitis (2), celiac disease (1), Hirschsprung's disease (1) and 1 child had prune-belly syndrome. Only 2 of the 35 children had diagnosed structural heart disease. One child had ventricular septal defect and the other complete atrioventricular canal, both of which had been surgically repaired in the past. There was minimal other testing performed on the study patients. Three of the 35 children had echocardiograms (none with prolonged QT), 2 children had ambulatory Holter monitors (1 with prolonged QT), and 1 (without prolonged QT) had a transtelephonic cardiac event recorder.
The ECG markers measured and analyzed included the QRS duration, QT and JT intervals (QTc and JTc), rate-corrected QTc using Bazett's formula,9 rate-corrected JTc,10 and interlead dispersion markers. Although the QTc has been the standard measurement of ventricular repolarization, it includes both depolarization and repolarization and may not always be a sensitive indicator of the type repolarization abnormalities seen in prolongation of the QT. Intraventricular conduction abnormalities, such as bundle branch block, complicate evaluation of the QTc interval. The JTc is a more accurate measurement of ventricular repolarization, and therefore, may be a more sensitive means of assessing abnormalities.10
The QRS, QT, JT, and sinus cycle length (RR) intervals were measured manually by two experienced blinded observers. Lead II, the lead with the maximum QTc, and the lead with the minimum QTc were each analyzed. Interlead QT and JT dispersion were calculated as the maximum QT (or JT) interval, minus the minimum QT (or JT) interval and compared with and without rate-correction.11,12
ECG data were entered into a database for further statistical analysis. All data are presented as the mean ± 1 SD. Comparisons of data between the study and the control group as well as interobserver variability were assessed by two-tailed Student'st test for unequal or equal variance where appropriate. AP value <.05 was considered statistically significant.
The mean ± standard deviation mg/kg/dose of cisapride was 0.19 ± 0.08 (range, 0.09–0.42) and the mean mg/kg/day of cisapride was 0.67 ± 0.23 (range, 0.30–1.68). The mean (±SD) QTc in the control group (n = 1000) was 420 ms ± 20 ms, and the mean JTc was 320 ms ± 20 ms. The mean (±SD) QTc in the cisapride group was 430 ms ± 37 ms (range, 372 ms–502 ms) and the mean JTc was 337 ms ± 34 ms (range, 282 ms–425 ms). There was no statistically significance difference for either QTc or JTc in the cisapride patients versus the control group (P = NS). However, 11 of the 35 (31%) cisapride patients had prolongation of the QTc (≥450 ms). The JTc was prolonged (≥360 ms) in 16 out of 35 patients (46%). The mean QTc dispersion (QTcd) among 1000 control patients was 30 ms ± 15 ms, and the mean JTc dispersion (JTcd) was 27 ms ± 15 ms. The mean (±SD) QTcd in the cisapride group was 30 ± 25 ms and mean (±SD) JTcd was 26 ± 23 ms. The mean age was 7 years for those children with QTc prolongation, and the mean age was 4 years for those children with normal QTc intervals. An increased QTcd or JTcd (≥70 ms) was documented in 3 out of 35 children (Table). Of these 3 children, 2 also had prolonged QT and JT intervals. The findings were similar when a rate-corrected formula was not used (raw QTd and JTd). Comparison of manual QT and JT measurements varied by 5 to 10 ms, indicating low interobserver variability.
Ten of the 11 children with QTc prolongation were taking cisapride for GERD symptoms, and one for prune-belly syndrome. There were 2 children with structural congenital heart disease; 1 had QTc prolongation and 1 had normal QT and JT intervals. Of the 11 study patients with QTc prolongation, 2 had documented torsades de pointes ventricular tachycardia (VT) (Figure, parts A and B). Both of these children were taking cisapride concomitantly with a macrolide antibiotic—1 with erythromycin, the other with clarithromycin.
Prolongation of the QTc may be congenital in nature or acquired. Congenital or idiopathic prolonged QTc may be sporadic or inherited. The most common type of hereditary prolonged QT syndrome is Romano-Ward syndrome, characterized by autosomal dominant transmission, prolongation of the QTc on ECG, ventricular tachyarrhythmias, in particular torsades de pointes, syncope, and sudden death.13,14
Acquired long QTc may be attributable to electrolyte abnormalities such as hypocalcemia, hypokalemia, hypomagnesemia, metabolic disturbances, or drug-induced. Typically these proarrhythmic medication side effects are seen with drug overdose, liver disease, or concomitant administration of a drug that interferes with hepatic cytochrome P450 3A4 enzymatic metabolism (ie, ketoconazole, macrolide antibiotics, cimetidine).2-7 It has been previously shown that other noncardiac medications such as the nonsedating antihistamines (terfenadine, astemizole) have significant actions on the cardiac potassium channels.7 The molecular or ionic mechanism involved with the use of cisapride on ventricular repolarization and the action on potassium or other ion channels have recently been studied in isolated guinea pig hearts, and cisapride was shown to be a potent inhibitor of the rapid delayed rectifier potassium current (Ikr).15
Further work at the ion channel level, whole animal electrophysiology model,16 and human myocyte studies will help to elucidate the exact mechanisms of cisapride-associated proarrhythmia and repolarization delay. Although the efficacy of cisapride administration in children with GERD has been established, the safety attributable to potential malignant ventricular arrhythmias has recently been questioned. Lewin et al8 reported QTc prolongation as well as 2 to 1 atrioventricular conduction, and T-wave alternans, during cisapride therapy in a 2-month-old infant. Cisapride-induced arrhythmias have been reported in several other recent pediatric and adult case reports over the past several years, including some fatal events.17-21
The results of this present study suggest that cisapride may cause prolongation of ventricular repolarization in children. Many of these patients may have been entirely asymptomatic without signs of arrhythmia or cardiotoxicity, despite the QT prolongation. There does not appear to be increased heterogeneity of repolarization (QT dispersion) or delayed depolarization (QRS prolongation) in this small sample. The proarrhythmia may be exacerbated by medications that inhibit cytochrome P450 3A4 hepatic metabolism, overdosage, or other mechanisms that may result in decreased serum clearance. Cisapride may not, therefore, be the drug of choice in treating GERD in children who have impairment of hepatic metabolism attributable to liver disease, or with concomitant administration of drugs metabolized via the cytochrome P450 3A4 enzymatic system. In addition, caution should be used in patients who have ventricular repolarization abnormalities and in patients who are maintained on class III antiarrhythmic agents and other medications that prolong the repolarization phase of the cardiac action potential. Potential risks versus benefits should be evaluated carefully, and we suggest that ECG intervals be monitored routinely in children maintained on cisapride, particularly when administered in combination with other medications that affect cytochrome P450 3A4 metabolism or ventricular repolarization. The nature of the QTc prolongation, although unknown, may be related to dosing and serum levels, and/or an idiosyncratic reaction in children who may have a partial channel defect that is unmasked by the drug.
As with other medications, there may be a lower therapeutic index in children compared with adults on cisapride, particularly regarding its apparent repolarization prolonging action. Given the results of this study, as well as previous case reports of proarrhythmia associated with QT prolongation, further clinical, animal, and molecular research is warranted on the effect of cisapride on cardiac ion channels and receptor sites.
A potential limitation of this study is the small sample size, particularly of patients on multiple medications. However, the high incidence of QT prolongation remains surprising and unexpected. Patients in the study group may not have been as healthy as normal controls and may have greater likelihood of electrolyte abnormalities and metabolic disturbances secondary to GERD. Hepatic and renal dysfunction and/or metabolic disturbances were not clinically suspected, although during the course of this study liver enzymes, electrolytes, blood urea nitrogen, and creatinine were not systematically assessed. It is also suggested that 12 lead resting ECGs be obtained before and after initiation of cisapride therapy in future larger-scale prospective studies.
We thank Marga Rivera for her assistance with data collection, and the patients and families involved in this study.
- Received December 8, 1997.
- Accepted March 19, 1998.
Reprint requests to (C.I.B.) Department of Cardiology, Harvard Medical School, Children's Hospital, Boston, 300 Longwood Ave, Boston, MA 02115.
Presented in part at the 18th Annual Scientific Session of the North American Society of Pacing and Electrophysiology in New Orleans, LA, May 1997.
- GERD =
- gastroesophageal reflux disease •
- QTc =
- QT interval •
- ECG =
- electrocardiogram •
- JTc =
- JT interval •
- QTcd =
- QTc dispersion •
- JTcd =
- JTc dispersion •
- VT =
- ventricular tachycardia
- ↵Janssen Pharmaceutica, Inc. Safety Advisory. February 1995
- Berul CI,
- Morad M
- Day CP,
- Mc Comb JM, Campbell RWF
- Romano C,
- Gemme G,
- Pongiglione R
- Drolet B,
- Khalifa M,
- Daleau P,
- Hamelin BA,
- Turgeon J
- Berul CI,
- Aronovitz MJ,
- Wang PJ,
- Mendelsohn ME
- Olsson S,
- Edwards IR
- Copyright © 1998 American Academy of Pediatrics