Do GI or respiratory symptoms rule out this important diagnosis?
Although it has been more than 40 years since Kawasaki disease was first described, it is still something of a medical mystery. Its cause has yet to be determined, and although we know the importance of early treatment— which helps prevent the development of coronary artery aneurysms—timely diagnosis of the acute vasculitis remains difficult.
Symptoms not in the diagnostic criteria can mislead. To make the diagnosis, fever should be present for at least 5 days, along with at least 4 of the following:
• Rash.
• Nonexudative conjunctivitis.
• Oral mucous membrane changes.
• Extremity changes.
• Cervical lymphadenopathy.
However, when other symptoms accompany a prolonged fever, Kawasaki disease may not come to mind, and the window for ideal treatment may be missed.
Study shows associated symptoms are common. New clues that can help clinicians diagnose Kawasaki disease more promptly are a fortuitous by-product of a recent prospective trial, the main purpose of which was to evaluate the effectiveness of corticosteroids as a treatment for the disease.1 The trial included 198 children with Kawasaki disease; the patients were all enrolled after Kawasaki disease had been diagnosed but before treatment had been initiated.
In the course of the trial, Baker and colleagues collected data about symptoms noted at several points around the time of Kawasaki disease diagnosis, including the 10-day period before the patients were enrolled in the study. Among the study’s 198 subjects, 93 different associated symptoms were recorded from parent interviews and primary caregiver observations during the 10 days before diagnosis. Of these 93 associated symptoms, 9 were reported in more than 10% of the patients.
The most common symptoms (other than those included in the Kawasaki disease criteria) were irritability (seen in 50% of patients), vomiting (in 44%), decreased oral intake (in 37%), cough (in 28%), diarrhea (in 26%), rhinorrhea (in 19%), weakness (in 19%), abdominal pain (in 18%), and joint pain (in 15%). When symptoms were grouped together by type, 61% of the patients were found to have 1 or more GI symptoms, and 35% had 1 or more respiratory symptoms.
The authors also examined relationships between associated symptoms and other factors, such as laboratory abnormalities and patient age. They found some correlation between the presence of certain symptoms and abnormal results on such tests as absolute neutrophil count, IgA level, and erythrocyte sedimentation rate. Also, patients with symptoms of irritability and rhinorrhea were younger than those who did not have these symptoms, and children with vomiting, abdominal pain, and joint pain were older than those without these complaints. However, these age-related differences might be explained by the fact that older children are better able to report specific symptoms, while younger children experiencing the same symptoms may simply be described as irritable.
Baker and colleagues concluded that associated symptoms are common in the 10 days preceding a diagnosis of Kawasaki disease. These associated symptoms could be a result of diffuse vasculitis, or they might have an infectious origin. It is also feasible, the study authors note, that these symptoms represent a second, unrelated process; however, they considered this explanation unlikely.
It is worth noting that testing for infectious diseases was not routinely done during the study. Another weakness of the study is that it relied on reported symptoms and did not use standardized definitions of such symptoms as diarrhea. However, when we take a history from a family, we also receive reported symptoms with nonstandardized definitions.
Maintain heightened suspicion in any child with prolonged fever. Still, the Baker study enables us to draw an important conclusion: namely, that simply because a child has GI, respiratory, or other nonspecific symptoms along with persistent fever does not mean that Kawasaki disease can be ruled out. We need to keep this disease in mind whenever we see patients with prolonged fever— even if they have symptoms unrelated to the Kawasaki criteria.