Short-bowel syndrome (SBS) is a devastating condition characterized by chronically impaired absorption by the small intestine (1–3). The most common etiologies in children are necrotizing enterocolitis, intestinal atresia, volvulus, and gastroschisis (4–6). The incidence of SBS has been estimated to be 24.5/100,000 live births (7). Current treatment includes supportive care with long-term parenteral nutrition (PN) while enteral feeds are carefully advanced. Unfortunately, many patients require prolonged PN due to chronic diarrhea and nutrient malabsorption. Another common reason for prolonged PN is gastrointestinal (GI) dysmotility (8). Symptoms can include vomiting, regurgitation, pain, and bloating. Although long-term PN is a lifesaving therapy, it is associated with multiple complications, including sepsis, liver disease, electrolyte instability, hyperglycemia, and metabolic bone disease. Therefore, a goal of therapy for SBS is to wean PN as quickly as possible.
In patients with SBS with possible GI dysmotility, management begins with evaluating for anatomic obstruction. Other strategies include maximization of antacid medications, use of continuous intragastric tube feeding, placement of a feeding tube distal to the pylorus, treatment of underlying enteropathies including bacterial overgrowth, and use of prokinetic medications such as metoclopramide and cisapride. Clinicians prescribing either of these prokinetic medications must do so with great caution and diligence. In February 2009, the US Food and Drug Administration (FDA) issued a black-box warning regarding the risk of tardive dyskinesia with chronic use of metoclopramide (9). Therefore, clinicians are often reticent about using this medication for a prolonged duration. In 1993, the FDA approved the administration of cisapride (Propulsid) for adults with gastroesophageal reflux disease. Cisapride improves motility by enhancing acetylcholine release at the myenteric plexus. It has been used for treating functional dyspepsia, gastroparesis, chronic constipation, and chronic intestinal pseudoobstruction. In an animal model using rabbit Purkinje fibers, cisapride prolongs the repolarization phase at plasma concentrations expected from therapeutic doses (10). There were reports of an association with its use and prolonged corrected QT interval leading to Torsade de pointes, heart block, and death, especially in elderly patients, but also in children. These events initially prompted the manufacturer to discourage physicians from prescribing concurrent medications that increase serum cisapride levels by inhibiting hepatic cytochrome P450 3A4, such as macrolide antibiotics, antifungals, and cimetidine. In 2000, cisapride was removed from the market after reports of at least 341 cardiac dysrhythmias, including 80 deaths (11).
Under the US Limited Access Program for Cisapride developed through Janssen Pharmaceutica Inc (Titusville, NJ) and the FDA, we obtained approval for the administration of cisapride in children with refractory GI dysmotility, including those with pediatric SBS. A literature search revealed only a single case report describing its use in SBS (12). This patient cohort would seem to be at especially high risk of dysrhythmias while using cisapride because of frequent electrolyte instability. Here we report our 2-year experience using cisapride for enteral intolerance in pediatric SBS.
All of the subjects included in the present study were studied at the Center for Advanced Intestinal Rehabilitation at Children's Hospital Boston between January 2007 and January 2009. This multidisciplinary program includes intestinal failure specialists in pediatric gastroenterology, surgery, nutrition, nursing, pharmacy, and social work (13). All of the subjects were enrolled in the Limited Access Program for Cisapride after obtaining parental informed consent. The protocol was approved by the institutional review board at Children's Hospital Boston.
All of the study subjects had SBS using the standard functional definition of PN dependence for congenital or acquired disease of the GI tract for at least 90 days (1). At the time of enrollment, all of the subjects were making little or no progress in advancing enteral feeds and had failed standard medical and surgical therapies to treat feeding intolerance. Feeding intolerance was defined as vomiting, regurgitation, pain, and bloating directly associated with enteral feeds. We excluded subjects with a history of cardiac disease, prolonged baseline corrected QT interval, family history of prolonged corrected QT interval, electrolyte abnormalities, serious uncontrolled illness, and concomitant use of cardiotoxic medications or medications interfering with cisapride metabolism. History of PN exposure was collected from medical records. As outlined in the Limited Access Program, before treatment with cisapride, all of the participants underwent laboratory tests (complete blood counts, electrolytes, and liver panel), 12-lead electrocardiograms (ECGs) for corrected QT interval, and contrast radiographic studies for anatomic obstruction. If these results were normal and the patient fulfilled inclusion criteria, then we prescribed cisapride at 0.1 mg/kg per dose for 4 doses per day. An ECG was obtained 4 to 6 days after starting cisapride treatment. If ECG remained normal and response to therapy was not observed, we then increased the dose to 0.2 mg/kg per dose or a maximum 10 mg per dose 4 times per day at the next clinic visit. Again, an ECG was obtained 4 to 6 days after starting a new cisapride dose. As outlined by the protocol, patients had scheduled study visits every 2 months for clinical assessments including anthropometrics; ECGs and serum chemistry panels were checked every 4 months. Subjects were instructed to discontinue cisapride temporarily for prolonged corrected QT interval (defined as >450 ms), electrolyte abnormalities, or general anesthesia. Dieticians used recommended daily allowance (RDA) to devise energy intake recommendations and titrated PN prescriptions for appropriate weight gain according to Centers for Disease Control and Prevention growth charts. Whenever possible, enteral nutrition was advanced as much as tolerated according to our standard feeding algorithm (14).
Data collected included patient characteristics, medical and surgical history, and anthropometric data. Our study team recorded adverse effects prospectively. Height and weight z scores were calculated using Centers for Disease Control and Prevention 2000 growth charts. The percentage of enteral energy intake was calculated as the amount of dietary energy intake derived by oral and/or enteral routes divided by the total energy intake (oral, enteral, and parenteral routes) multiplied by 100. Primary endpoints were defined as change in percentage of enteral energy intake and change in corrected QT interval from baseline. We used the Wilcoxon signed rank test to compare change in percentage enteral intake and parenteral energy intake between baseline and follow-up. Secondary endpoints were defined as changes in weight and height z scores, as well as death and any other adverse outcomes. We used a longitudinal regression model to balance effects seen among subjects contributing varying lengths of follow-up. We accounted for within-subject correlation in the outcomes by using the general estimating equation (GEE) technique. Analysis was performed using SAS 9.2 (SAS Institute, Cary, NC).
The clinical characteristics of the 10 subjects enrolled are presented in Table 1. The mean (SD) patient age at enrollment was 30.3 (30.5) months. The ages ranged from 1 month to 7 years. There were a variety of underlying diagnoses leading to SBS, including gastroschisis with and without intestinal atresia, necrotizing enterocolitis, and long-segment Hirschsprung disease. Only 1 girl was enrolled. The median (interquartile range [IQR]) PN exposure before starting cisapride was 23 months (5.6 months–4.9 years). For all of the subjects, this length of PN use reflected nearly an entire lifetime with few interruptions. The median (IQR) duration of follow-up was 8.7 (3.1–14.3) months. The median (IQR) residual bowel length was 102 (85–130) cm. There were 2 subjects whose residual bowel lengths were unknown. Six subjects had previously undergone at least 1 bowel-lengthening procedure, such as serial transverse enteroplasty procedure (STEP) or Bianchi procedure, including 2 subjects who had had both procedures. Four subjects were known to be missing the ileocecal valve. The median (IQR) plasma citrulline level was 14.5 (10.5–31.3) μmol/L. Five subjects had citrulline levels <15 μmol/L, which is prognostic of low likelihood of eventual PN independence (15).
Figure 1 depicts the percentage of enteral energy intake after enrollment in study protocol. Enteral tolerance improved in 7 of the 10 subjects during treatment, including 2 subjects who were weaned completely from PN. The median (IQR) percentage of enteral energy intake at baseline was 30.3% (4.4%–52.8%). Feeding route at baseline was intragastric (60%), postpyloric (30%), and a combination of oral and intragastric (10%). Comparing the parenteral energy intake at baseline and at last follow-up, the median (IQR) change was −20.8 (−61.0 to −14) kcal · kg−1 · day−1 (P = 0.01). Comparing percentage enteral energy intake at baseline and at last follow-up, the median (IQR) change was 19.9% (15.4%–29.8%) (P = 0.01). Comparing percentage enteral energy intake at baseline and at last follow-up using cisapride, the median (IQR) change was 18.9% (2.6%–26.65%) (P = 0.03). Using longitudinal regression analysis, we detected a strong positive association between duration of cisapride administration and improved enteral tolerance. The mean percentage of enteral energy intake increased by 2.9% for every month of cisapride treatment (P < 0.0001). Cisapride dose per day was not a significant predictor of enteral tolerance (P = 0.5). Weight and height z scores did not significantly change according to treatment duration (P = 0.08 and P = 0.25, respectively). Median (IQR) weight gain was 7.7 (3.7–17.2) g/day and height velocity was 0.15 (0.1–0.38) cm/week. Five of the 10 subjects discontinued cisapride by the last follow-up.
All of the subjects had documented normal corrected QT intervals before treatment with cisapride. Prolonged corrected QT intervals developed in 2 subjects at 30 days and 11 months after treatment, neither of whom was symptomatic. As a result, 1 of these subjects permanently discontinued cisapride. The other subject restarted cisapride at a lower total daily dose after corrected QT returned to normal. The subject's ECG was repeated again while taking medication and it was normal. Among the whole cohort, we did not detect a relation between length of corrected QT and total daily cisapride dose (P = 0.07) or treatment duration (P = 0.36). The mean change in corrected QT interval from baseline was −0.44 ms (95% CI −1.23 to 0.40).
Other adverse events noted during the study included 3 subjects with GI bleeding, 1 subject with D-lactic acidosis, and 1 death due to presumed sepsis. The data safety and monitoring board concluded that it was doubtful that cisapride was related to the death.
In SBS, the most common obstacles to weaning PN and advancing enteral nutrition are diarrhea and nutrient malabsorption. GI dysmotility, however, is another problem that frequently leads to prolonged PN dependence. Unfortunately, there are few therapeutic medical modalities available, and US federal regulations and advisories have limited the availability of 2 such agents: metoclopramide and cisapride.
In our open-label pilot study among pediatric patients with nearly lifelong PN dependency, we observed a median increment of percentage enteral nutrition of 19.9% (IQR 15.4%–29.8%) during the course of follow-up (P = 0.01). Nearly all of the improvement in enteral tolerance occurred while using cisapride. Two of the 10 subjects were weaned completely from PN support, and 7 had some improvement in enteral tolerance. There was no evidence of growth faltering while PN was weaned in this cohort because weight gain and height velocity were maintained, as were weight and height z scores, before and after the course of cisapride. Therefore, decreasing PN support reflected real improvement in enteral tolerance.
There are few studies of therapies for enteral intolerance in SBS. Modi et al reviewed nutritional outcomes of the STEP registry of 38 patients (16). The percentage of enteral energy intake at baseline and length of follow-up were similar to our study; however, enteral intake improved by 36% following STEP compared with 19.9% in the present study. Both STEP and cisapride have been used in SBS. STEP is a surgical intervention that lengthens and tapers the bowel, thereby improving intestinal function and motility. Cisapride, on the contrary, is a medical treatment that treats only GI motility and has no direct effect on intestinal structure. The majority of our subjects had a history of autologous bowel reconstructive surgery, such as STEP, and remained PN dependent.
We observed corrected QT prolongation in 2 subjects (20%); both subjects were clinically asymptomatic. The corrected QT prolongation was discovered as a result of routine ECG screening. The incidence of corrected QT prolongation is similar to previous reports in the literature. Khongphatthanayothin et al reported an incidence of corrected QT prolongation (defined as >440 ms) of 13% in a prospective study of 101 children treated with cisapride (17). The incidence was only 6% in children without any identifying risk factors, such as drug interaction, heart disease, renal or liver dysfunction, electrolyte abnormalities, or prematurity. Hill et al found an incidence of corrected QT prolongation (defined as >450 ms) of 31% in a prospective study of 35 children treated with cisapride (18).
Of note, the majority of our subjects with SBS had dysmotility due to gastroschisis. In this condition, prenatal amniotic fluid exposure to the serosal surface of the GI tract causes inflammation and bowel wall thickening, leading to disordered postnatal motility. A rat model of gastroschisis suggests that the damage leads to delayed neuronal differentiation and abnormal myenteric plexus organization (19). In infants with gastroschisis, esophageal motor abnormalities and GI dysmotility are common (20–22). Erythromycin is not effective in improving enteral tolerance after primary repair (23). In contrast to other prokinetics such as erythromycin, metoclopramide, and octreotide, cisapride improves ileal contractility in newborn and adult rabbits (24). Our findings suggest that cisapride may be useful in humans with gastroschisis.
Our study had several limitations. First, the manner in which the manufacturer made the study drug available dictated an open-label design. A randomized placebo-controlled study of cisapride in SBS would have been preferable but logistically infeasible. In addition, extensive data on enteral feeding advancement before cisapride treatment were not available for all of the patients, limiting the control data to the period of time immediately before the initiation of cisapride treatment; however, all of the patients had refractory feeding intolerance as reflected by nearly lifelong PN dependence (n = 10), history of autologous bowel reconstructive surgery (n = 6), and low plasma citrulline levels (median 14.5 μmol/L). Taken together, these factors substantiate the severity of intestinal failure in the cohort. A third limitation is the lack of formal manometry testing to corroborate the diagnosis of GI dysmotility. The clinical features of these patients seem sufficient to attribute their symptoms to motility disorders, after anatomic obstruction had been extensively evaluated. In the future, manometry may be helpful in defining which patients may benefit from cisapride treatment, similar to its role in studying erythromycin (25).
In summary, in 10 subjects with SBS and GI dysmotility, cisapride use was associated with a median increment of percentage enteral nutrition of approximately 20%, with 7 patients exhibiting some degree of improvement and 2 patients weaning completely from PN. Corrected QT prolongation occurred in 20% of subjects, which is similar to the reported incidence in other conditions. At the time of study enrollment, these subjects were making little or no progress in advancing enteral feeds and had failed all of the standard medical therapies to treat feeding intolerance with nearly lifelong PN exposure. Among those who tolerated the drug, cisapride use was associated with a modest but statistically significant improvement in enteral tolerance; the mean percentage of enteral energy intake increased by 2.9% for every month of cisapride treatment (P < 0.0001). It is important to continue investigating prokinetic medications for SBS because the incidence of gastroschisis continues to increase steadily (26). Future studies should seek to identify clinical and manometric findings to select patients who are most likely to benefit from cisapride or other prokinetic medical therapy.
We thank the entire CAIR team for their expert care of the patients.
1. O'Keefe SJ, Buchman AL, Fishbein TM, et al
. Short bowel syndrome and intestinal failure: consensus definitions and overview. Clin Gastroenterol Hepatol 2006; 4:6–10.
2. Duro D, Kamin D, Duggan C. Overview of pediatric short bowel syndrome. J Pediatr Gastroenterol Nutr 2008; 47(Suppl 1):S33–S36.
3. Goulet O, Sauvat F. Short bowel syndrome and intestinal transplantation in children. Curr Opin Clin Nutr Metab Care 2006; 9:304–313.
4. Olieman JF, Tibboel D, Penning C. Growth and nutritional aspects of infantile short bowel syndrome for the past 2 decades. J Pediatr Surg 2008; 43:2061–2069.
5. Vargas JH, Ament M, Berquist WE. Long-term home parenteral nutrition in pediatrics: ten years of experience in 102 patients. J Pediatr Gastroenterol Nutr 1987; 6:24–32.
6. Cole CR, Hansen N, Higgins RD, et al
. Very low birth weight preterm infants with surgical short bowel syndrome: incidence, morbidity and mortality, and growth outcomes at 18 to 22 months. Pediatrics 2008; 122:e573–e582.
7. Wales PW, de Silva N, Kim J, et al
. Neonatal short bowel syndrome: population-based estimates of incidence and mortality rates. J Pediatr Surg 2004; 39:690–695.
8. Phillips JD, Raval MV, Redden C, et al
. Gastroschisis, atresia, dysmotility: surgical treatment strategies for a distinct clinical entity. J Pediatr Surg 2008; 43:2208–2212.
9. Puiseux FL, Adamantidis MM, Dumonteir BM, et al
. Cisapride-induced prolongation of cardiac action potential and early afterdepolarizations in rabbit Purkinje fibers. Br J Pharmacol 1996; 117:1377–1379.
10. Kenney C, Hunter C, Davidson A, et al
. Metoclopramide, an increasingly recognized cause of tardive dyskinesia. J Clin Pharmacol 2008; 48:379–384.
11. US Food and Drug Administration. Withdrawal of troglitazone and cisapride. JAMA
12. Puntis JW, Booth IW, Buick R. Cisapride in neonatal short gut. Lancet 1986; 2:108–109.
13. Modi BP, Langer M, Ching YA, et al
. Improved survival in a multidisciplinary short bowel syndrome program. J Pediatr Surg 2008; 43:20–24.
14. Utter S, Duggan C. Short bowel syndrome. In: Hendricks K, Duggan C, eds. Manual of Pediatric Nutrition
, 4th ed. Hamilton, Canada: BC Decker; 2005:718–35.
15. Fitzgibbons S, Ching YA, Valim C, et al
. Relationship between serum citrulline levels and progression to parenteral nutrition independence in children with short bowel syndrome. J Pediatr Surg 2009; 44:928–932.
16. Modi BP, Javid PJ, Jaksic T, et al
. First report of the international serial transverse enteroplasty data registry: indications, efficacy, and complications. J Am Coll Surg 2007; 204:365–371.
17. Khongphatthanayothin A, Lane J, Thomas D, et al
. Effects of cisapride on QT interval in children. Pediatrics 1998; 133:51–56.
18. Hill SL, Evangelista JK, Pizzi AM, et al
. Proarrhythmia associated with cisapride in children. Pediatrics 1998; 101:1053–1056.
19. Vannucchi MG, Midrio P, Flake AW, et al
. Neuronal differentiation and myenteric plexus organization are delayed in gastroschisis: an immunohistochemical study in rat model. Neurosci Lett 2003; 339:77–81.
20. Jadcherla SR, Gupta A, Stoner E, et al
. Neuromotor markers of esophageal motility in feeding-intolerant infants with gastroschisis. J Pediatr Gastroenterol Nutr 2008; 47:158–164.
21. Oyachi N, Lakshmanan J, Ross MG, et al
. Fetal gastrointestinal motility in a rabbit model of gastroschisis. J Pediatr Surg 2004; 39:366–370.
22. Cheng G, Langham MR, Sninsky CA, et al
. Gastrointestinal myoelectric activity in a child with gastroschisis and ileal atresia. J Pediatr Surg 1997; 32:923–927.
23. Curry JL, Lander AD, Stringer MD. A multicenter, randomized, double-blind, placebo-controlled trial of prokinetic agent erythromycin in the postoperative recovery of infants with gastroschisis. J Pediatr Surg 2004; 39:565–569.
24. Langer JC, Bramlett G. Effect of prokinetic agents on ileal contractility in a rabbit model of gastroschisis. J Pediatr Surg 1997; 32:605–608.
25. Jadcherla SR, Berseth CL. Effect of erythromycin on gastroduodenal contractile activity in developing neonates. J Pediatr Gastroenterol Nutr 2002; 34:16–22.
26. Keys C, Drewett M, Burge DM. Gastroschisis: the cost of an epidemic. J Pediatr Surg 2008; 43:654–657.