Objective. To evaluate our experience using the antimigraine prophylactic drugs, amitriptyline and cyproheptadine, for the prophylactic management of cyclic vomiting syndrome (CVS) in children.
Methods and Patients. Twenty-seven patients (16 males) ranging in age from 2 to 16 years at diagnosis, fulfilling the diagnostic criteria for CVS and treated prophylactically with either amitriptyline (22) or/and cyproheptadine (6) were identified through retrospective chart review. Individual patient data were corroborated by the attending physician and/or interviews with patients and families. Minimum follow-up time before entry into the study group was 5 months. Patients were stratified according to three treatment outcomes: 1) complete response—no attacks, 2) partial response—50% or greater reduction in frequency of attacks, and 3) no response—less than 50% decrease in frequency of attacks.
Results. Of the 22 patients treated with amitriptyline, 16 (73%) had a complete response while 4 (18%) had a partial response. Of the 6 patients treated with cyproheptadine, 4 (66%) had a complete response and 1 (17%) had a partial response. Thus, 91% of the amitriptyline group and 83% of the cyproheptadine group had at least a partial response to therapy. No patients experienced significant side effects to either medication.
Conclusion. The antimigraine prophylactic drugs, amitriptyline and cyproheptadine, represent effective prophylactic agents for the management of CVS in the vast majority of patients fulfilling the diagnostic criteria for this syndrome.
Cyclic vomiting syndrome (CVS), initially described by Gee1 in 1882, remains poorly understood, and management is often frustrating for clinicians because diagnosis is based on patient presentation of defining symptoms and the exclusion of known neurologic, metabolic, sinus, urologic, endocrinologic, and gastrointestinal disorders that also present with the pattern of CVS.2-4 Children with CVS present with chronic, recurrent, stereotypic episodes of self-limited vomiting often to the point of dehydration separated by varying symptom-free intervals. Associated symptoms may include nausea, abdominal pain, headache, fever, lethargy, photophobia, motion sickness, and diarrhea.5,6
Medical management has focused on supportive measures, abortive strategies, and prophylactic therapies. Supportive and abortive measures include intravenous fluids, a quiet, dark, nonstimulating environment, antiemetic agents (promethazine, ondansetron, granisetron), anxiolytic agents (lorazepam), and antimigraine compounds (sumatriptan, isometheptene mucate).2,6-8 Prophylactic therapies include antimigraine drugs (cyproheptadine, amitriptyline, propranolol, carbamazepine), antiepileptic agents (phenobarbital, carbamazepine), and gastrointestinal prokinetic agents (cisapride, erythromycin).2,9-12
Because patients with CVS have no demonstrable organic pathology and because CVS by its paroxysmal, chronic nature bears striking resemblance to migraine and may represent a migraine equivalent in childhood, we hypothesize that proven antimigraine prophylaxis may also prevent the debilitating symptoms associated with CVS. This study reports our experience using two antimigraine prophylactic therapies in children with CVS: cyproheptadine (Periactin) and amitriptyline (Elavil, Endep).
PATIENTS AND METHODS
A total of 27 patients (16 male) ranging in age from 2 years to 16 years at the time of diagnosis, meeting the diagnostic criteria for CVS13,14 and treated prophylactically with either amitriptyline or cyproheptadine were identified through a retrospective chart review using a computerized search of medical records over the past 13 years at Children's Medical Center of Dallas. All patients were evaluated as clinically indicated to exclude organic causes of CVS. Data were corroborated by the respective attending physicians and/or by telephone interviews with patients and families.
The clinical presentations of individual patients are outlined in Table1. The mean interval between onset of symptoms and diagnosis of CVS was 2 years with a range of 3 months to 8 years. The frequency of vomiting episodes varied from weekly to every 3 months (with most children experiencing vomiting spells every 4 to 6 weeks). For each child the frequency remained consistent, and the spells tended to be stereotypic in both severity and duration. Eleven of the 27 children required frequent hospitalization for dehydration. Five children required visits to the emergency room for infrequent administration of intravenous fluids. The remaining 11 children did not require emergency room or inpatient care for dehydration related to vomiting. Approximately 50% of patients had a parent or sibling with an identified history of migraine. The vast majority of patients had failed numerous previous medical therapies, including antiemetics, prokinetic agents, antibiotics, H2-blockers, antisecretory compounds, anticholinergics, antidepressants, anticonvulsants, and antacids, as well as dietary manipulation. Eight of 27 patients had concurrent, significant medical, or psychiatric conditions, including depression, hyperactivity, and developmental delay.
Therapy with amitriptyline and cyproheptadine was initiated at the lowest effective dose identified for the prophylactic treatment of migraine: amitriptyline 10 mg po qhs or cyproheptadine 2 mg po bid.2,10,12,15 The medication doses were gradually increased until effective prophylaxis was obtained or adverse medication side effects developed. Amitriptyline blood levels were monitored at doses greater than 50 mg per day, and these patients additionally had an electrocardiogram. In selective patients with refractory symptoms, it was necessary to increase amitriptyline doses to levels sufficient to maintain blood levels within the therapeutic antidepressant range. Generally, these patients were also those individuals identified with a high severity of vomiting (Table 1). Cyproheptadine doses were increased until symptoms were controlled or until the side effect of sedation became unacceptable. Patients were stratified according to three treatment outcomes: 1) complete response—no attacks, 2) partial response—50% or greater reduction in frequency of attacks, or 3) no response—less than 50% decrease in frequency of attacks.
The therapeutic responses for all 27 patients in both the amitriptyline-treated and the cyproheptadine-treated groups are outlined in Table 1. The patients within each drug treatment group are ranked by increasing severity of symptoms. The outcome results of both treatment groups are summarized in Table2. Sixteen of 22 patients (73%) treated with amitriptyline and 4 of 6 patients (66%) treated with cyproheptadine achieved complete remission of vomiting and abdominal pain with follow-up ranging from 5 months to 10 years. Four patients (18%) in the amitriptyline group and 1 patient (17%) in the cyproheptadine group experienced a partial response (at least a 50% reduction in frequency of attacks). Therefore, 91% of the amitriptyline group and 83% of the cyproheptadine group achieved either a complete or a partial reduction in symptoms. All patients failing to respond to either therapy (2/22 treated with amitriptyline and 1/6 treated with cyproheptadine) had severe symptoms requiring repeated hospitalizations for dehydration. Patient 21 failed to respond to amitriptyline but demonstrated partial response to cyproheptadine. Patient 26 was treated successfully with erythromycin with complete response for 6 months. Patient 17 did not respond to any medical therapy. No patients were treated with the abortive drug, sumatriptan.
The response rate was similar for male and female patients in both treatment groups. In addition a family history of migraine was not predictive of response to either drug. Side effects for both medications included sedation and weight gain due to increased appetite. Sedation as a consequence of amitriptyline was managed by giving a single daily dose at bedtime while maintaining therapeutic efficacy. On the other hand, an unacceptable level of sedation in patients receiving cyproheptadine necessitated dosage reduction in one patient. No patient stopped either therapy due to unacceptable weight gain.
The relationship between CVS and migraine is based on the common clinical syndromes of episodic, stereotypic symptoms often occurring in the early morning or at night, often triggered by stress or excitement and frequently associated with a strong family history of migraine.16,17 We did not attempt to separate abdominal migraine as a distinct clinical entity. Based on our hypothesis regarding the relationship between CVS and migraine, we used initially the antimigraine prophylactic agent, amitriptyline, for the prophylactic management of CVS.12 More recently, younger patients with CVS were treated prophylactically with cyproheptadine, also well-recognized for its use in migraine prophylaxis, primarily due to ease of administration because it comes as a liquid preparation and carries fewer potential side effects in the younger child.18
Amitriptyline is a tricyclic antidepressant with complex pharmacologic and clinical effects; the precise mechanisms of antidepressant and antimigraine actions are unclear.18,19 Acting at the neuronal membrane, the drug potentiates the effects of serotonin to a greater extent than epinephrine, thus acting primarily as a serotonin agonist.20 In addition tricyclics exhibit strong anticholinergic activity. The antidepressant activity may be related to any or to all of these actions, although the effects on norepinepherine and serotonin metabolism are thought to be the most important. Tricyclic antidepressants have dose-dependent quinidine-like cardiotoxicity, which, in conjunction with anticholinergic activity and potentiation of norepinepherine, may be associated with cardiovascular disturbances: electrocardiogram changes, tachycardia, and postural hypotension. Common clinical adverse effects include sedation and weight gain due to increased appetite. Maximum antidepressant effects may not occur for 2 or more weeks after the initiation of therapy, but in our experience the onset of effective migraine or CVS prophylaxis is much more rapid, usually a matter of days and at much lower drug doses than required for antidepressant effects. Other clinical indications for amitriptyline involve chronic pain syndromes, including chronic abdominal pain18 and the fibromyalgia syndromes.21
Likewise, cyproheptadine has diverse effects by exhibiting both antihistamine and antiserotonin activities; its effectiveness in the treatment of Cushing's syndrome is generally attributed to its antiserotonin activity.22 In addition the drug has anticholinergic and sedative effects as well as calcium channel-blocking activity.23 Like amitriptyline, cyproheptadine commonly simulates weight gain through increased appetite and has apparently displayed some efficacy in the management of anorexia nervosa.22
Although both of these drugs have established efficacy for antimigraine prophylaxis, their mode of action is unknown. In fact a wide variety of medications have been reported to be of benefit for antimigraine and CVS prophylaxis, including amitriptyline, cyproheptadine, propranolol, carbamazepine, and calcium channel blockers.2,10,12,15,23-25 Erythromycin is reportedly effective for the prevention of CVS.11 Although the modes of action for all these drugs are unknown but based on the diverse nature of the reportedly effective prophylactic agents, one can hypothesize there must be multiple steps in the pathogenesis of both migraine and CVS that may be blocked to prevent clinical expression of the syndromes. Although antimigraine prophylactic therapy has been identified as beneficial for CVS, there are few reports of limited numbers of patients with few details9 and no published series reporting the efficacy of these agents.
Several important points regarding this study and the results should be emphasized. First, the clinical characteristics of our patient population with CVS appear to be very similar to the patients studied and reported by Li2 and Fleisher6 in many respects: about 50% of the patients had a strong family history of migraine, and age of onset and clinical course were also similar. On the other hand, Vanderhoof et al.11 found only a 10% incidence of migraine in their patients. Our patients were about 60% male, while in most other series the sex ratio is reversed.2,6,11
Second, both amitriptyline and cyproheptadine appear to be virtually equally effective for the prophylactic management of CVS regardless of sex, with 91% and 83% of all patients achieving at least 50% reduction in frequency of attacks, respectively. The reported response rate for erythromycin is similar.11 A limited series of a few case reports show similar response rates to CVS prophylaxis with propranolol.10
Third, there appears to be an inverse correlation between the severity of CVS and the responsiveness to these medications. All patients with either mild or moderate symptoms had at least a partial response. The intractable, nonresponsive patients were all in the group of patients requiring repeated hospitalizations. This inverse correlation was not observed in patients treated with erythromycin; patients with severe disease responded as equally well as those with mild symptoms.11
Fourth, most patients responded to both amitriptyline and cyproheptadine at doses reportedly effective for migraine prophylaxis. The reported dosage range for amitriptyline to treat migraine and chronic pain syndromes is 10 to 75 mg/day, while usual antidepressant dosage ranges from 50 to 200 mg/day.2,12,25 Selected patients who had more severe disease necessitating repeated hospitalizations for dehydration (Table 1) required antidepressant doses of amitriptyline to control their symptoms of CVS. Dosages in this range were closely monitored by blood levels to maintain effective antidepressant levels. If blood levels fell for any reason, symptoms rapidly recurred. Therefore, amitriptyline doses were pushed to higher levels in patients with more severe symptoms before abandoning the drug as ineffective. Patient response to cyproheptadine occurred in the recommended dosage range for migraine prophylaxis (0.25 to 0.5 mg/kg/day).2,12,15 Attempts to increase the dosage of cyproheptadine beyond these parameters usually resulted in an unacceptable level of sedation.
Fifth, both amitriptyline and cyproheptadine caused unacceptable sedative effects in some patients and excessive weight gain due to increased appetite in others. The sedative effects were usually mollified with dosage reduction and the passage of time. The daytime sedative effects of amitriptyline were often eliminated by administering the drug as a single daily dose at bedtime while maintaining control of symptoms. No patients discontinued either medication due to excessive weight gain. No other significant side effects were observed; specifically, anticholinergic side effects were not of sufficient severity to stop therapy in any patient.
Sixth, these results are unlikely to represent merely a placebo effect because the vast majority of patients experienced multiple drug trials without benefit before the trial with amitriptyline or cyproheptadine for CVS prophylaxis. Therefore, we conclude that the antimigraine prophylactic drugs, cyproheptadine and amitriptyline, may be very effective agents for the prophylactic management of patients with CVS. These conclusions must be tempered by the fact that these patients were not subjected to a randomized, placebo-controlled trial. In addition, we do not understand the pathophysiology of CVS, nor do we understand the mode of action of the antimigraine prophylactic agents.
- Received December 16, 1996.
- Accepted May 12, 1997.
No reprints available.
- CVS =
- cyclic vomiting syndrome
- Gee S
- Li BUK
- Forbes D
- Korson M
- Fleisher DR
- Pfau BT,
- Li BUK,
- Murray RD,
- Heitlinger LA,
- McClung HJ,
- Hayes JR
- Bentley D,
- Kehely A,
- Al-Bayaty M,
- Michie CA
- Forbes D,
- Withers G
- Vanderhoof JA,
- Young R,
- Kaufman SS,
- Ernst L
- ↵Peroutka SJ. Drugs effective in the therapy of migraine. In: Hardman JG, Limbird LE, Gilman AG, eds. Goodman and Gilman's Pharmacological Basis of Therapeutics. 9th ed. New York, NY: McGraw Hill; 1996:487–502
- Fleisher DR
- Li BUK,
- Spierings ELH
- Symon DNK,
- Russell G
- McEvoy GK,
- ↵Potter WZ, Manji HK, Rudorfer MV. Tricyclics and tetracyclics. In: Schatzberg AF, Nemeroff CB, eds. Textbook of Psychopharmacology. Washington, DC: American Psychiatric Press, Inc; 1995:141–160
- ↵Hollister LE. Drug therapy: tricyclic antidepressants. N Engl J Med. 1978;229:1106–1109, 1168–1171
- McEvoy GK,
- Peroutka SJ,
- Allen GS
- Crouch JR,
- Ziegler DK,
- Hassanein R
- Copyright © 1997 American Academy of Pediatrics