A 4-year-old girl is admitted to the ER with high fever and acute torticollis.
GEORGE K. SIBERRY, MD, MPH, SECTION EDITOR
A 4-year-old girl has been admitted to the emergency room, and your care, because of her mother's report of high fever and "a stiff neck." While you take the history from the child's mother, you note that the girl appears ill and irritable. The mother tells you that, two days before admission, her daughter suffered trauma to the posterior neck against the rail of her bed, but did not vomit or lose consciousness afterward. A day later, the child began complaining of difficulty and pain while moving her neck, pain upon swallowing, and generalized weakness. The mother recorded the child's temperature as 39° C at the time.
The child was brought to see her physician, who took a throat culture and prescribed amoxicillin. Because the neck pain persisted, however, she was referred to this hospital to rule out cervical trauma.
Your patient was born to healthy parents who are Sephardic Jews. She was her mother's second full-term pregnancy. Both pregnancy and delivery were uncomplicated; birth weight was 3,150 g. Early growth and development were normal, but the girl had lately been evaluated by an endocrinologist for short stature. Initial tests included a radiograph of the wrist that showed a bone age of 1.5 years and serum levels of T3, T4, and thyroid-stimulating hormone within the normal range. The child had not traveled outside her country in the past few months, and the family does not have cats or other pets at home.
On physical examination, the temperature is 38° C; heart rate, 120/min; respirations, 32/min; and blood pressure, 90/50 mm Hg. You continue to be impressed by how ill and suffering the girl appears to be. You note torticollis, with limited neck movement. Examination of the throat reveals hyperemic tonsils without exudate. Anterior cervical lymph nodes, mostly on the left side, are enlarged and tender on palpation. You find no other lymphadenopathy. Auscultation of the heart reveals normal heart sounds, without murmur. The remainder of your examination is unremarkable.
Based on what you've found, you concentrate on two principal possible explanations for the acute torticollis: neck trauma and cervical lymphadenitis. You request orthopedic consultation; the examination is normal, and lateral and anteroposterior radiographs of the cervical spine are negative.
The patient's history of trauma is probably a red herring, you conclude. Cervical lymphadenopathy, most likely secondary to streptococcal pharyngitis, is now your lead diagnosis.
Your next step is to order initial blood tests, including a complete blood count. The white blood cell count is 28 x 103/µL, with 95% neutrophils; hemoglobin, 11.9 g/dL; platelet count, 376 x 103/µL; and erythrocyte sedimentation rate, 62 mm/h. Urinalysis is normal.
At this point, your clinical impression is one of an ill-appearing preschool girl who has acute torticollis, cervical lymphadenopathy, and leukocytosis. You decide to hospitalize her, order blood and throat cultures, and begin intravenous penicillin for pharyngitis.
The day after admission, your patient continues to be irritable and now complains of abdominal pain. She has developed watery diarrhea and coffee grounds-appearing vomiting. The fever has risen to 39.5° C and the heart rate is 180/min. The left cervical lymph nodes are larger than you found them to be the day before. The abdomen is diffusely tender, with adequate bowel sounds; neither the liver nor spleen is enlarged.
You consider an acute abdomen, but examination by the pediatric surgeons excludes that diagnosis. Could this be mesenteric lymphadenitis, or another condition that mimics acute abdomen? You repeat the blood tests: WBC count, 13 x 103/µL; hemoglobin, 10.5 g/dL; platelet count, 32 x 103/µL. The serum alanine aminotransferase (ALT) level is 100 U/L; aspartate aminotransferase (AST), 295 U/L; alkaline phosphatase, 171 U/L; and lactate dehydrogenase, 572 U/L. You order cultures of blood and stool and serologic studies for Shigella, Salmonella, Yersinia, adenovirus, and hepatitis A, B, and C.
Abdominal ultrasonography reveals a normal liver, pancreas, and spleen; a distended gallbladder without evidence of stones or biliary sludge; enlarged loops of bowel that are filled with fluid; and free fluid in the upper abdomen. Although streptococcal infection is capable of causing gallbladder hydropsby way of toxin production or, perhaps, obstruction of the cystic duct secondary to mesenteric lymphadenopathyyou change the treatment to a broad-spectrum antibiotic, ceftazidime, plus metronidazole because you are now considering abdominal sepsis.
Meanwhile, throat and blood cultures obtained on the day of admission come back negative. A test of the antistreptolysin-O titer is negative.
As the afternoon comes on, the child remains irritable and tachycardic (180/min). Findings of fever, unilateral lymphadenopathy, tachycardia, and hydrops of the gallbladder now alarm your thinkingcall it your diagnostic sixth senseabout the possibility of an unusual or incomplete picture of Kawasaki disease. You promptly order an echocardiogram: The scan reveals a normal heart and valves. Nevertheless, you start intravenous immunoglobulin G (IVIG) at 2 g/kg of body weight and aspirin at 100 mg/kg/day.
The next day, the hemoglobin level is 10.5 g/dL; the WBC count has risen to 14.5 x 103/µL; and the platelet count holds at 32 x 103/µL. AST is 38 U/L; ALT, 80 U/L; albumin, 22 g/dL; and globulin, 34 g/dL. The C-reactive protein level is 49 mg/L. On clinical grounds, you consider a diagnosis of Crohn disease because of the girl's history of short stature, diarrhea, low serum albumin, and reversal of albumin:globulin ratio.
But a repeat physical examination offers more evidence of Kawasaki disease: Strawberry tongue and nonpurulent conjunctivitis have appeared. Slit-lamp examination shows no uveitis.
On the seventh day of hospitalization, repeat echocardiography reveals new mitral valve regurgitation. The girl still has a high fever and continues to appear ill. A second dose of IVIG is administered. Two days later, the child complains of arthralgia. Physical examination reveals tender and swollen proximal interphalangeal joints.
Toward the 10th day of hospitalization, you note a gradual improvement in the patient's general condition: Both the fever and abdominal pain disappear. She develops periungual desquamation. (All serologic tests and cultures have returned negative.) Laboratory tests on the 10th day reveal an elevated ESR (>100 mm/h) and extreme thrombocytosis (>1,000 x 103/µL).
After two weeks of hospitalization, your patient is discharged. Although she is now asymptomatic, repeat echocardiography reveals a new aneurysm of the left coronary artery. Dipyridamole is added to aspirin and dexamethasone.
Kawasaki disease is one of the most common systemic vasculitides and the most common cause of coronary vasculitis in childhood.1 Incomplete and atypical cases have been reported, particularly in children younger than 6 months2,3 but also in older ones.
Failure to diagnose the disease is a concern, mostly because of its cardiac complications, which might be prevented with IVIG treatment. IVIG reduces both the duration of fever and the prevalence of coronary artery aneurysm when given within 10 days of onset.4 On the other hand, overdiagnosis and unnecessary treatment, which is costly and has the potential for adverse effects, can be prevented by adherence to established clinical criteria for diagnosis.
This girl had an unusual presentation of Kawasaki disease: fever, cervical lymphadenopathy, and abdominal pain. All her signs and symptoms are well described in Kawasaki disease,5 but at admission she had not yet exhibited a total of any four of five criteria for diagnosis that are required in addition to fever (cervical lymphadenopathy; injected conjunctivae; rash; peripheral skin changes such as erythema or edema; and various oral manifestations). In fact, the prompt visit paid to her physician, and then her admission on only the second day of the course, came earlier in this phasic disease than is typical because of an unrelated precipitating eventthe trauma to her neckand thus brought you onto the scene at a diagnostically challenging time. The other clinical features and laboratory tests completed the puzzle several days into treatment with IVIG and aspirin.
Severe abdominal pain, as your patient developed, often associated with diarrhea, is reported in approximately 20% of patients in the first days of Kawasaki disease and may suggest an acute abdomen.5 Hypoalbuminemia, also common, is associated with prolonged fever.
Although it took your sixth sense, so to speak, to make an early diagnosis and begin appropriate treatment with IVIG and aspirin, your patient nonetheless had a prolonged course. In fact, approximately 10% of children treated with IVIG have persistent or recrudescent fever.6 This child required two courses of IVIG, but inflammation was not well controlled. Upon follow-up one month after discharge, she was afebrile and active, and continues to be treated with prednisone, aspirin, cimetidine, and dipyridamole.
Early reports linked streptococci, as well as other pathogens, to Kawasaki disease.5 In one study, evidence of recent streptococcal infection was found in as many as 25% of cases.7 Streptococcal infection was your initial clinical diagnosis, but it could not be confirmed by laboratory testing.*
Because the etiology of the disease is unknown, it is considered to be a final common pathway for a number of pathologic processes, an important aspect of which is polyclonal T cell activation by toxins from a variety of organisms acting as superantigens.9
This case emphasizes the importance of, first, suspecting Kawasaki disease even when its diagnostic criteria are not strictly met and, second, questioning the possibility of a different but related disease that has a pathogenesis similar to that of Kawasaki. For this patient, one plus one did equal six. And that made sense after all!
*We have pondered alternate developments in this case and come to the conclusion that a positive throat culture would not have prevented us from treating this patient with IVIG and aspirin; we agree with others who have proposed that Kawasaki disease should not be excluded because of concurrent group A streptococcal infection.8
1. Kato H, Inoue O, Akagi T: Kawasaki disease: Cardiac problems and management. Pediatr Rev 1988;9:209
2. Avner JR, Shaw KN, Chin AJ: Atypical presentation of Kawasaki disease with early development of giant coronary artery aneurysms. J Pediatr 1989;114:605
3. Burns JC, Wiggins JW, Toews WH, et al: Clinical spectrum of Kawasaki disease in infants younger than 6 months of age. J Pediatr 1986;109:759
4. Newburger JW, Takahashi M, Burns JC, et al: The treatment of Kawasaki syndrome with intravenous gamma globulin. N Engl J Med 1986;315:341
5. Melish ME: Kawasaki syndrome. Pediatr Rev 1996;17:153
6. Newburger JW, Takahashi M, Beiser AS, et al: A single intravenous infusion of gamma globulin as compared with four infusions in the treatment of acute Kawasaki syndrome. N Engl J Med 1991;324:1633
7. Dhillon R, Newton L, Rudd PT, et al: Management of Kawasaki disease in the British Isles. Arch Dis Child 1993;69:631
8. Hoare S, Abinum M, Cant AJ: Overlap between Kawasaki disease and group A streptococcal infections. Pediatr Infec Dis J 1997;16:633
9. Arav-Boger R, Assia A, Jurgenson J, et al: The immunology of Kawasaki disease. Adv Pediatr 1994;41:359
Pediatric Puzzler: Rising temperature, stiff neck.