Fever of unknown origin in a 10-year-old boy with allergic rhinitis and asthma

Clinical allergy—immunology rounds Supported by an unrestricted grant from Rhoˆne Poulenc-Rorer

Fever of unknown origin in a 10-year-old boy with allergic rhinitis and asthma Ahmet Zafer Caliskaner, MD*; Mehmet Karaayvaz, MD†; and Nejat Ozanguc, MD‡

CHIEF COMPLAINT Fever for 4 weeks. HISTORY OF PRESENT ILLNESS A 10-year-old boy was admitted to our clinic because of watery nasal discharge, sneezing, nasal itching, and recurrent wheezing and coughing, which was particularly worse at night. Nasal examination revealed moist and pale turbinates. Upon auscultation of the chest, there were inspiratory and expiratory rhonchi and wheezes accentuated during prolonged expiration. An increased number of eosinophils were found on nasal smear. Spirometry revealed lung function impairment with an obstructive pattern, which is improved after bronchodilator administration. Skin prick tests were positive to house dust mites. The patient was diagnosed as having allergic rhinitis and intermittent asthma based on the medical history, characteristics of symptoms, physical examination, pulmonary function tests, and skin pricktest results. Ketotifen 2 mg/day and inhaled salbutamol as needed for symptomatic relief were prescribed and instructions were given for standard environmental control measures. The patient became symptom-free * Specialist in Internal Medicine and Fellowin-Training at Allergy. † Associate Professor, Specialist in Pediatrics and Allergy. ‡ Professor, Specialist in Allergy and Dermatology. Department of Allergy, GATA Gulhane Military Medical Academy, Ankara, Turkey. Received for publication July 28, 1999. Accepted for publication in revised form October 24, 1999.

102

within a week and the treatment has been continued. After 3 weeks, he developed fever, particularly at noon and at night, rising as high as 38.9°C. The accompanying symptom was that of slight weakness, otherwise he appeared well. The patient was evaluated for fever. PHYSICAL EXAMINATION On examination, the patient appeared well although he was febrile. The temperature was 38.5°C, blood pressure was 110/80 mmHg, and the heart rate was 95 beats per minute. Skin examination was normal, no rash or erythema were present. Results of the head, eyes, ears, nose, and throat examination were normal. No pathologic superficial lymph nodes were found on examination of the cervical, axillary, and inguinal regions. The chest movement was normal and the lungs were clear to auscultation. Cardiac examination revealed normal heart sounds. The abdominal examination revealed no visceral organomegaly or free intraperitoneal fluid. The extremities were normal with palpable peripheral pulses. The genital and neurologic examinations were also normal. INITIAL LABORATORY FINDING Erythrocyte sedimentation rate slightly elevated at 25 mm/h. White blood cellcount was 11,000/mm3, with 8% eosinophilia (total eosinophil count was 850/mm3). Peripheral blood smear revealed no abnormal cells suggesting infection or malignancy. Urinalysis was unremarkable. Serum biochem-

istry (including fasting blood sugar, BUN, creatinine, uric acid, total protein, albumin, calcium, phosphate, aspartate aminotransferase, alanine aminotransferase, alkaline phosphatase, gamma glutamyl transpeptidase, direct and indirect bilirubin, amylase, creatinine phosphokinase, lactic dehydrogenase, total cholesterol, and triglyceride) were within normal ranges. Chest radiograph was normal with no evidence of infiltrates or effusion. Sinus radiograph was also normal.

DIFFERENTIAL DIAGNOSIS According to the patient’s medical history, physical examination and initial laboratory investigation, the differential diagnosis was based on the investigation of the clinical entities that may cause fever of unknown origin (FUO). The general categories accounting for the majority of FUO conditions are as follows: 1. 2. 3. 4.

Infections Malignancies Collagen-vascular diseases Drugs reactions

So, which diseases or conditions might be responsible for the fever in this patient? Pulmonary or extra-pulmonary tuberculosis Typhoid or paratyphoid fever Brucellosis Toxoplasmosis Malaria HIV infection Hepatitis B and C infections

ANNALS OF ALLERGY, ASTHMA, & IMMUNOLOGY

Epstein-Barr virus, cytomegalovirus, herpes virus infections Abscesses (dental, intra-abdominal, renal, retroperitoneal, or paraspinal) Infective endocarditis Hodgkin’s disease or non-Hodgkin’s lymphoma Leukemia Solid tumors Churg-Strauss syndrome SLE Still’s disease or Juvenile rheumatoid arthritis Systemic vasculitis A drug-related etiology

FURTHER INVESTIGATIONS Additional laboratory investigations were performed to clarify the diagnosis: Repeated blood cultures. Microscopical examination of sputum, urine and stool. Throat, sputum, urine and stool cultures. Tuberculin-PPD (purified protein derivative) skin test Salmonella hemagglutination test. Brucellosis complement fixation test and serum agglutination test. Serum toxoplasma, Epstein-Barr virus, cytomegalovirus, herpes virus antibodies. Microscopical examination of thick and thin blood smears. Serum anti HIV antibody. Hepatitis markers for HBV and HCV infections. A thorough dental examination. Abdominal ultrasonography. Echocardiogram. Serum rheumatoid factor, ANA, anti DNA and anti-neutrophil cytoplasmic antibodies (ANCA).

DISCUSSION The boy had been in good health until about 4 weeks earlier. There was no history of animal scratches or bites, insect bites, ingestion of raw meat, travel outside of the country, or blood transfusion. The sole clinical manifestation was an unexplained fever, which had its onset after the diagnosis of allergic rhinitis and asthma was made. His clinical findings were not typical for any of the possible causes of FUO; however, some diseases may be subtle or occult when accompanied

VOLUME 85, AUGUST, 2000

with another disease. For example, pulmonary tuberculosis in patients with AIDS is often subtle, and normal X-rays occur in 15% to 30% of cases.1 For this reason, further diagnostic investigations were performed to rule out possible causes of the patient’s fever. Repeated blood, sputum, urine and stool cultures were negative for pathogenic agents. Microscopic examination of thick and thin peripheral blood smears were also negative for either infectious or haematologic causes. Microscopic examination of sputum, and urine and stool samples revealed no abnormal findings. A detailed investigation was performed to rule out a tuberculosis infection. The PPD skin test was within normal ranges in our patient; however, the PPD skin test is positive in fewer than 50% of all patients with TB who present with an FUO.2 For this reason, we performed PCR assay on gastric lavage fluid and blood culture samples. These investigations were negative for tubercle bacillus. Serum serology for viral and bacterial agents was negative. Anti HIV antibody was performed several times and was found negative. Imaging studies were performed to detect occult infection sources and the results were normal. Consequently, at the end of these investigations, no infectious cause was identified. The most frequent occult malignancies that cause fever, are of those of reticuloendothelial origin (eg, lymphoma and leukemia).2 Our patient had no pathologic lymphadenopathy or organomegaly. Acute phase reactants were in normal ranges except for a slightly elevated ESR. He had no marked weakness or weight loss. We did not think the patient was suffering from any solid or haematologic malignancy since these diseases are generally debilitating, and are usually accompanied by many symptoms and are progressive. Another group of disease that may cause FUO, are the collagen-vascular diseases. Systemic onset juvenile rheumatoid arthritis (formerly called Still’s disease) in younger patients, and giant cell arteritis in elderly individuals are

the most common rheumatologic disorders presenting as FUO.2 The hallmark of Still’s disease is the combination of intermittent daily fever greater than 39.5°C and arthritis. In addition, thrombocytosis, elevated ESR and positive rheumatoid factor are other findings.3 Except for the fever, none of these manifestations was present in our patient. In this case, the diagnosis of ChurgStrauss syndrome (CSS) may be considered Because of an absence of specific laboratory tests, the accurate diagnosis of the CSS is usually difficult. The combination of clinical and laboratory findings usually lead to the diagnosis of CSS. In the CSS, the mean age at diagnosis is 50 years, but the systemic vasculitic phase is frequently apparent in late 30s. The lungs are most commonly involved, followed by the skin. Churg-Strauss syndrome, however, can affect any organ system, including the cardiovascular, gastrointestinal, and nervous systems (mononeuropathy or polyneuropathy). Peripheral blood or tissue eosinophilia, normochromic-normocytic anemia, marked elevation in the ESR, leukocytosis, elevated IgE levels are also found at times. Hypergammaglobulinemia, positive rheumatoid factor at low titer and positive P-ANCA are commonly encountered laboratory anomalies of the this disease. Migratory or transient pulmonary opacities are important chest X-ray finding.4,5 In the present patient, clinical manifestations and laboratory findings were insufficient for the diagnosis of CSS. Findings such as weight loss, peripheral neuropathy, skin lesions, radiographic abnormalities, anemia, significantly elevated ESR, serum RF and positive ANCA were not present. Fever, eosinophilia, and atopy alone were not diagnostic, thus CSS was ruled out. Similarly, other systemic vasculitic syndromes are also accompanied by both clinical and laboratory findings. Our patient had not shown any manifestations that suggested a vasculitic syndrome. In addition, serum autoantibodies, that may be indicative of an autoimmune disease, were also negative.

103

Drug fever should be considered in the differential diagnosis of any patient with unexplained fever.6 Drug fever is a disorder, characterized by fever coinciding with the administration of the drug and disappearing after the discontinuation of the drug, when no other cause for the fever is evident after a careful physical examination and laboratory investigation. The only factual way to know whether a patient has a drug fever is by stopping the drug or drugs. Rechallenge with the putative offending agent can confirm the diagnosis if fever recurs.7 DRUG ELIMINATION AND RE-CHALLENGE Since no cause for the fever was revealed, we decided to discontinue salbutamol and ketotifen administration sequentially to rule out drug-induced fever. The patient did not need inhaled salbutamol daily. Thus, this drug was safely stopped and he was closely followed-up. Fever was not resolved during this salbutamol-free period. In the second step, the other drug ketotifen, which was given to treat allergic rhinitis, was discontinued. The result was surprising. The patient became afebrile within 36 hours. During the 2-week ketotifen-free period, the body temperature was normal, and fever did not recur. Since re-challenge is usually safe, particularly when the initial febrile disease is mild,7 ketotifen was given again. Before and after rechallenge, complete blood count, ESR and total eosinophil counts were performed.7 The patient became febrile again, 48 hours after ketotifen re-challenge. The ESR and total white blood count was slightly elevated and total eosinophil count increased 3-fold. The body temperature reached 39°C. Discontinuation of ketotifen resulted in fever resolution. Second re-challenge was performed 3 days later and resulted in fever again. These results were supportive of the role of ketotifen in the febrile disease for this allergic patient.

104

CONCLUSION This case illustrates that whenever an allergic patient presents with prolonged fever, without coexistent disease, fever due to administration of antiallergic drug(s) should be considered. Virtually any drug is capable of causing fever, however, ketotifen-induced fever has never been reported in the English literature to date. Most medications may cause fever, with or without concomitant clinical manifestations. The exact incidence of drug fever remains unknown but it has been estimated to occur in approximately 10% of inpatients. The fever may arise from the pharmacologic action, its effects on thermoregulation, a local complication following parenteral administration, or an idiosyncratic response.7,8 Hypersensitivity is the most common cause of drug fever. Various mechanisms can cause fever, including the formation of circulating antibody-antigen complexes and/or a T cell immune-response provoked by the drug or its metabolites. Any one episode may involve multiple antigenic determinants and mechanisms.6,7,9 –11 Drug fever may have any pattern. It typically occurs after seven to ten days of treatment and usually resolves within 48 hours of discontinuing the drug. Failure to diagnose drug fever may lead to inappropriate and potentially harmful diagnostic and therapeutic interventions. In suspected cases, it is necessary to discontinue administration of all potentially causative medicines, together or sequentially. Rechallenge with the offending agent will usually cause recurrence of fever within a few hours, confirming the diagnosis.8,9,12 Ketotifen is a benzocycloheptathiophene derivative and is marketed throughout the world as an antiallergic drug. Ketotifen has been reported to control symptoms of patients with physical urticaria, chronic idiopathic urticaria, asthma, and mastocytosis. It has a variety of actions in addition to its histamine receptor blocking activity. Several of its effects seem to be the consequence of presumed mast cell

membrane stabilization: it inhibits histamine and leukotriene release in experimental systems. It also inhibits platelet activating factor activity in pro-inflammatory cells such as eosinophils. Inhibition of activated eosinophils and T-cell recruitment into the airway have been considered as the beneficial effects of ketotifen in asthma.13–16 Ketotifen is virtually non-toxic. Its primary side effects seem to be temporary drowsiness, mouth dryness (and other mucous membranes), appetite stimulation, and weight gain. A reversible fall in the thrombocyte count in patients receiving ketotifen concomitantly with oral antidiabetic agents has been observed in rare cases.17 SUMMARY We believe this case represents a clear example of drug fever, and it appears to be the first report to implicate ketotifen as the responsible agent, confirmed with double rechallenge. The recognition of drug fever is clinically important. Failure to recognize the etiologic relationship between the drug and fever has unnecessary consequences, including extra testing, empiric therapy, and longer hospital stays. We suggest that ketotifen should be considered as a possible cause of fever in allergic patients receiving this drug. REFERENCES 1. Greenberg SD, Frager D, Suster B, et al. Active pulmonary tuberculosis in patients with AIDS: spectrum of radiographic findings (including a normal appearance). Radiology 1994;193: 115–119. 2. Arnow PM, Flaherty JP. Fever of unknown origin. Lancet 1997;350:575. 3. Fujikawa S, Okuni M. Clinical analysis of 570 cases with juvenile rheumatoid arthritis: results of a nationwide retrospective survey in Japan. Acta Paediatr Jpn 1997;39(2):245–249. 4. Lanham JG, Elkon KB, Pusey CD, Hughes GR. Systemic vasculitis with asthma and eosinophilia: A clinical approach to the Churg-Strauss syndrome. Medicine(Balt) 1984;63:65– 81. 5. Schmitt WH, Csernok E, Kobayashi S, et al. Churg-Strauss syndrome: serum

ANNALS OF ALLERGY, ASTHMA, & IMMUNOLOGY

6. 7. 8. 9. 10. 11.

markers of lymphocyte activation and endothelial damage. Arthritis Rheum 1998;41(3):445– 452. Hanson MA. Drug fever. Remember to consider it in diagnosis. Postgrad Med 1991;89(5):167–70, 173. Johnson DH, Cunha BA. Drug fever. Infect Dis Clin North Am 1996;10(1): 85–91. Lipsky BA, Hirschmann JV. Drug fever. JAMA 1981;27;245(8):851– 854. Roush MK, Nelson KM. Understanding drug-induced febrile reactions. Am Pharm 1993;NS33:39. Tabor PA. Drug-induced fever. Drug Intell Clin Pharm 1986;20:413– 420. Mackowiak PA, LeMaistre CF. Drug fever: critical appraisal of conven-

VOLUME 85, AUGUST, 2000

12. 13.

14. 15.

tional concepts. Ann Intern Med 1987; 106:728 –733. Mackowiak PA. Drug fever: mechanisms, maxims and misconceptions. Am J Med Sci 1987;294:275–286. Hoshino M, Nakamura Y, Shin Z, Fukushima Y. Effects of ketotifen on symptoms and on bronchial mucosa in patients with atopic asthma. Allergy 1997;52(8):814 – 820. Egan CA, Rallis TM. Treatment of chronic urticaria with ketotifen. Arch Dermatol 1997;133:147–149. Bustos GJ, Bustos D, Bustos GJ, Romero O. Prevention of asthma with ketotifen in preasthmatic children: a three-year follow-up study. Clin Exp Allergy 1995;25(6):568 –573.

16. Cordoba H, Fernandez M, Santos F, Oehling A. Variations of intracellular histamine basophil levels after treatment with ketotifen. J Investig Allergol Clin Immunol 1992;2(4):187–190. 17. Rackham A, Brown CA, Chandra RK, et al. A Canadian multicenter study with Zaditen (ketotifen) in the treatment of bronchial asthma in children aged 5 to 17 years. J Allergy Clin Immunol 1989;84(3):286 –296. Request for reprints should be addressed to: Dr Mehmet Karaayvaz GATA Gulhane Military Medical Academy Department of Allergy Ankara 06018 Turkey

105