Answers are grounded in this paper and cite their sources.
Derivatives of 5-aminosalicylic acid (5-ASA) are commonly prescribed for children with inflammatory bowel disease (IBD). Among this class of drugs, mesalamine is currently second only to sulfasalazine in its frequency of use (Procter and Gamble Pharmaceuticals, Inc., Medical Communications Department). In part, the increased role of mesalamine in the treatment of IBD is due to its superior safety profile. Historically, 5-ASA compounds have been associated with several adverse effects reported in the medical literature including: aplastic anemia, arthritis, fever, hepatitis, interstitial nephritis, lichen planus, a lupus-like syndrome, myocarditis, neutropenia, pancreatitis, pericardial effusion, peripheral neuropathy, pneumonitis, thrombo-cytopenia, and a vasculitic rash (1-18). Mesalamine has been causally implicated in the development of a pericardial effusion in four adult cases (10-13). However, the present case illustrates the first reported child to develop this serious complication while taking mesalamine. CASE PRESENTATION A 9-year-old girl was admitted to the pediatric intensive care unit of the Arnold Palmer Hospital for Children and Women with pericarditis and a large pericardial effusion. She had been well until 15 months before her admission, when she experienced abdominal cramping and frequent mucoid bowel movements 1 month after completing a 10-day course of amoxicillin/clavulanate potassium for the treatment of otitis media. She also had received a simultaneous 7-day course of acyclovir for the treatment of concurrent varicella. On evaluation of her abdominal pain and diarrhea, the stool tested positive for occult blood and revealed both white blood cells (WBCs) and the presence of Clostridium difficile toxin. This prompted treatment with a 2-week course of metronidazole 250 mg t.i.d., which afforded some improvement. On the completion of treatment, however, her symptoms of abdominal cramping and mucoid diarrhea gradually returned. Her stool was negative for C. difficile toxin at this time. Therefore, evaluation with complete colonoscopy was performed to evaluate for the presence of IBD, given the patient's history and the past infection with C. difficile(19-21). The endoscopic appearance of the colon demonstrated diffuse friability and granularity, which was both visually and histologically consistent with ulcerative colitis. The terminal ileum was grossly and histologically normal. Consequently, the patient began a 6-month course of olsalazine, which was tolerated well, resulting in a resolution of symptoms. Two months before this admission, olsalazine therapy was discontinued. Shortly afterward, the patient once again began to complain of abdominal cramps, frequent loose bowel movements containing mucous, and hemepositive stools. Thus, the patient was started on mesalamine (Asacol) 400 mg q.d. (15 mg/kg/day) and prednisone 10 mg q.d. She remained symptomatic with as many as 13 bowel movements in the course of a day and systemic symptoms such as malaise and occasional lowgrade fever also emerged on the 13th day after beginning mesalamine therapy. The mesalamine dosage was doubled to 400 mg b.i.d. (30 mg/kg/day) and ultimately increased to 400 mg q.i.d. (60 mg/kg/day) in an effort to alleviate the patient's symptoms; the initial prednisone dose was continued. With the increased dose of mesalamine, the patient began to have well-formed stools and a resolution of her intestinal symptoms. Her systemic symptoms, however, steadily worsened. She began to complain of progressive fevers as high as 39.5°C as well as vague retrosternal discomfort and an occasional cough 21 days after the initiation of mesalamine therapy. Ultimately, 25 days after beginning mesalamine therapy, her general pediatrician noticed a new heart murmur and a friction rub. Echocardiographic evaluation revealed a 2-cm pericardial effusion encroaching on all four chambers of the heart (Fig. 1). The patient was subsequently admitted to the pediatric intensive care unit because of concerns about an impending tamponade. PHYSICAL EXAMINATION On admission to the pediatric intensive care unit of the Arnold Palmer Hospital for Children and Women, the patient's vital signs included a temperature of 37.6°C, pulse rate of 122 beats/min, respiratory rate of 24 breaths/min, a blood pressure of 126/56 mm Hg with no evidence of pulsus paradoxus, and normal anthropometric parameters (50th percentile for weight and 25th percentile for height). In general, she appeared pale, thin, and moderately ill but in no significant distress. Heart examination revealed a muffled S1 and S2, appropriate splitting, tachycardic rate, regular rhythm, and a loud friction rub radiating throughout the chest. Pulses were 2+ with good perfusion. Auscultation of the lungs revealed clear breath sounds and no rales. Abdominal examination revealed benign conditions. LABORATORY FINDINGS A complete blood count showed a peripheral white count of 9,700/ml with a differential of 34% segmented neutrophils, 25% bands, 23% lymphocytes, 15% monocytes, and 3% eosinophils. The hemoglobin and hematocrit were 7.7 g/dl and 24.6%, respectively, with a mean corpuscular volume of 71.9 fl and a platelet count of 681,000/ml. The Westergren sedimentation rate was 85 mm/h. Serum chemistries were within normal limits except for a calcium level of 8.4 mg/dl and a glucose level of 124 mg/dl. An amylase level was 12 U/L. Antinuclear antibody testing gave negative results. Serum titers were negative for coxsackie group A and B viruses, echovirus types 4, 9, 11, and 16, and Mycoplasma pneumoniae. A purified protein derivative was negative with a positive Candida control. Nasopharyngeal and rectal viral cultures showed negative results. A bacterial blood culture gave negative results as well. A chest radiograph showed a markedly enlarged cardiac silhouette with minimal lingular atelectasis. An electrocardiogram demonstrated a normal voltage tracing with no abnormalities. Analysis of the pericardial fluid revealed 552,000 red blood cells/ml and 1,020 WBCs/ml (45% segmented neutrophils, 41% lymphocytes, 10% monocytes, and 4% eosinophils) with a glucose level of 80 mg/dl and a total protein level of 5.8 g/dl. Bacterial, viral, fungal, and acid-fast cultures of the pericardial fluid showed no growth. Cytological examination of the pericardial fluid revealed no malignant cells. Pathologic examination of the pericardium detected a thickened, fibrotic pericardium with evidence of fibrinous pericarditis. There was a lymphocytic infiltrate present in the pericardium, but only scattered polynuclear cells with no prominent eosinophilic infiltration. CLINICAL COURSE Mesalamine therapy was discontinued, and an investigation was begun to identify a cause for the patient's condition. On the cessation of mesalamine, however, there was a clear resolution of the patient's fever, chest pain, and cough. Unfortunately, the pericardial effusion showed no evidence of change. At the recommendation of the pediatric cardiothoracic surgery department, a subxiphoid pericardial window was created. At the time of surgery, 400 ml of serosanguinous fluid was recovered. As already noted, the fluid and disease on the pericardium itself was unrevealing. The patient did well post-operatively and was discharged home 3 days after her surgery. Since discharge, the patient's ulcerative colitis has been controlled with low-dose prednisone, and her clinical course has been uneventful. DISCUSSION In our patient, the administration of mesalamine had a direct temporal relationship with the report of progressing systemic symptoms in a dose-related fashion. Furthermore, when the administration of mesalamine was halted, a striking decline in clinical symptoms was witnessed. These observations, the failure of a thorough investigation to isolate a cause, the fact that the pericardial fluid's composition suggested more of a transudative than an actively exudative process, and that a review of the literature showed mesalamine's adverse effects to include the development of a pericardial effusion in adults offer evidence that our patient experienced a drug-related complication. Compounds containing a 5-ASA moiety have been used in a variety of disease processes. For example, antituberculic compounds (e.g., para-aminosalicylic acid and its salts, isoniazid and streptomycin) have been reported to have several similar side effects with chemically related pharmaceuticals prescribed for patients with IBD (e.g., mesalamine and sulfasalazine) (22-26). Among the serious adverse effects seen with 5-ASA antituberculin medications, myocarditis and pericardial effusion have been reported (25-26). The exact mechanism of mesalamine-induced pericarditis is not well described. The possible explanations include a direct toxic effect of the drug, an IgE-mediated allergic reaction, a cell-mediated hypersensitivity reaction or a humoral antibody response. The disease of the pericardium in this patient revealed no evidence of cell death or necrosis. This would tend to exclude the possibility of mesalamine having a direct toxic effect on the pericardium. An IgE-mediated allergic reaction would also seem unlikely secondary to the lack of a peripheral eosinophilia and the absence of an eosinophilic infiltrate within the pericardium, although an IgE level was not measured in our patient. A drug-induced cell-mediated hypersensitivity reaction seems rather unlikely because of the localized nature of the patient's inflammatory process but could be argued for based on the lymphocytic infiltration seen within the pericardium. Therefore, a humoral antibody response, possibly with cross-reactive antimesalamine antibodies, seems the most likely mechanism. The reported side effects of 5-ASA derivatives bear striking similarities to findings in autoimmune disease. In fact, sulfasalazine has been associated with the development of a drug-induced lupus syndrome characterized by the production of anti-DNA antibodies (27). These patients classically develop pleuropericarditis and arthritis that resolve on stopping the drug and also respond to corticosteroid therapy. Our patient did not express anti-DNA antibodies and formally does not meet the criteria for a drug-induced lupus syndrome, but she still could have had a humoral antibody response to mesalamine. Given the patient's apparent tolerance of olsalazine therapy, a specific antimesalamine antibody with cross-reactive properties against the pericardium would be a viable explanation for this patient's condition. Although cardiac involvement is an accepted extraintestinal manifestation of IBD, pericarditis is rarely reported. Hinterreiter et al. (28) described 39 reported cases of cardiac involvement believed to be related to IBD during the past 30 years. Of these 39 patients, 28 were taking sulfasalazine at the time they developed pericarditis. Therefore, a 5-ASA-induced cause of pericarditis cannot totally be excluded in these patients. In a study of the extraintestinal manifestations of IBD by Greenstein et al. (29) involving 700 participants, there were no patients identified with either pericarditis or a pericardial effusion. Another review of the extraintestinal manifestations of IBD in 241 patients by Young revealed only one patient who had pericarditis (30). In all, a review of the literature failed to produce a report of a patient <18 years of age with a pericardial effusion attributable solely as an extraintestinal manifestation of IBD. In addition, nearly 80% of IBD-associated pericarditis and pericardial effusions are treated with corticosteroids (28). The patient in this case, however, developed a pericardial effusion despite taking prednisone for >3 weeks before her presentation. With the use of mesalamine, four adult patients have been reported in the literature to have developed a pericardial effusion in association with use of the drug (10-13). Our patient, however, represents the first pediatric case of a severe pericardial effusion associated with the administration of mesalamine. Given the common use of mesalamine in the treatment of IBD, the practitioner should be aware of the possibility of this serious complication.FIG. 1.: Echocardiogram reveals a 2-cm pericardial effusion (PE), indicated by the arrows, surrounding both ventricles.
No takes yet. Share an insight, caveat, or question.
Kaiser et al. (1997) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: