The seduction of asthma severity categorization.

I n 1997, the National Heart, Lung, and Blood Institute published the seminal and comprehensive treatise “Expert Panel Report II: Guidelines for the Diagnosis and Management of Asthma.”1 The fundamental principle guiding recommendations for asthma management in this handbook was a stepwise approach to pharmacotherapy based on a simple, clinically relevant, standardized method for classifying asthma severity. Adjusting the intensity of pharmacotherapy to the individual patient’s asthma severity seemed to be an intuitively reasonable principle, because it balanced the risks of side effects from more intensive pharmacotherapy (specifically higher doses of inhaled corticosteroids) against the potential benefits in more severe disease. There are three reasons, however, to be concerned that the method proposed in the Expert Panel Report II for classifying asthma severity is flawed. Practical observations suggest that clinicians cannot accurately use the asthma severity categorization method. The basic construct of the model is not supported by recent clinical observations. Severity categorization, as a concept, may not be as valuable as emphasizing asthma control and estimating the probability of certain poor outcomes. Given these concerns, the underlying principle of stepwise asthma pharmacotherapy based on severity categorization seems less appealing than initially suggested. In this issue of CHEST (see page 2156), Baker and colleagues describe a relatively simple study of how pediatric asthma specialists categorize asthma severity based on the Expert Panel Report II method. Eight case summaries were mailed to board-certified pediatric allergists and pulmonologists, along with the Expert Panel Report II asthma severity categorization method. The specialists were asked to complete a multiple choice questionnaire that addressed asthma severity classification and treatment recommendations. There are some difficulties with the methods used in this study. Several of the cases described patients currently receiving treatment with controller medications. Although the authors make the cogent point that asthma specialists usually only see patients already receiving anti-inflammatory treatment, the Expert Panel Report II approach advocates severity categorization based on “clinical features before treatment.” Unfortunately, agreement between the specialists and the “correct” asthma severity categorization, with appropriate explanation from the case summary, was not examined. Despite these weaknesses in the study design, it was obvious that there was poor agreement among the specialists in categorizing asthma severity. This observation should probably have been expected from previous work. Doerschug et al2 tested the content knowledge of residents, fellows, and faculty at the University of Iowa about the Expert Panel II report. In general, knowledge of these guidelines increased with advanced training, but overall was only moderate. Test performance in areas of estimating disease severity was especially disappointing; only 63% of the questions relating to asthma severity were answered correctly by asthma specialists. The results of these studies suggest that asthma specialists cannot reliably and accurately use the asthma severity categorization method proposed in the Expert Panel Report II. The severity classification method divides patients into two major categories, either intermittent or persistent asthma. The distinction between these categories is based on clinical features and acts as a critical threshold for the introduction of long-term controller medications, specifically inhaled corticosteroids. Recent information suggests that using only clinical features to distinguish between intermittent and persistent asthma may not be reliable. Vignola and colleagues3 found evidence of airway inflammation on bronchial biopsy in 24 patients with mild, intermittent asthma. Van Den Toorn et al4 performed bronchial biopsies in 18 young patients with a history of atopic asthma, but who had been in clinical remission for at least 1 year. Clinical remission was defined as the complete absence of symptoms with no use of asthma medication. Despite clinical remission of asthma, the bronchial biopsies confirmed evidence of ongoing airway inflammation. There were elevated levels of major basic protein in the subepithelium and epithelium. Also detected were increased levels in the bronchial wall of interleukin-5, chymase, tryptase, and CD25 cells. This information suggests that mild, intermittent asthma might not exist as a distinct entity if indexes of airway inflammation were considered. Patients with persistent asthma are further categorized into mild, moderate, and severe categories. Limited information at present suggests that proper use of the Expert Panel Report II method will result in most patients being categorized as having severe persistent asthma. In a telephone survey of 1,788 randomly selected patients with asthma, symptom assessments resulted in 77.3% being categorized as having moderate/severe persistent asthma.5 However, patient recall during a telephone survey may not adequately capture symptoms over the previous weeks to months. A retrospective analysis of data