Identifying adult asthmatic patients with an abnormal chest radiograph in the emergency department

The Global Initiative for Asthma (GINA) guidelines classify asthmatic exacerbations into mild, moderate, severe episodes and those with imminent respiratory arrest. We aimed to identify proportions of adult asthmatic patients in different subgroups with abnormal chest radiograph (CR) results and to explore predictors of them.

In this prospective study, we recruited consecutive adult patients with asthmatic exacerbation over 7 months. Primary outcomes were rates of abnormal CR results among the subgroups. Secondary outcomes included predictors of CR abnormalities, rates of disposition status, and their distribution of abnormal CR results. Logistic regression was applied to identify predictors of an abnormal CR result.

We recruited 201 patients; CR was carried out in 84.6%. Rates of abnormal CR results were 8.1% (nine of 111), 17.6% (nine of 51), 20% (one of five), and 0% (zero of three) in the mild, moderate, severe, and imminent respiratory arrest subgroups. No statistically significant differences existed when the rate of CR abnormalities in each subgroup was compared with the combined rate of the other three subgroups (P values: 0.186-0.943). There were 92 (45.8%) planned discharges, 14 (7%) discharges against medical advice, and 95 (47.2%) admissions. The rates of abnormal CR results in the planned discharged group versus the rest (discharges against medical advice+ admissions) were 3% (two of 67) versus 16.5% (17 of 103) (P=0.006). Logistic regression revealed the female sex to be associated with an abnormal CR result [odds ratio 10.1 (95% confidence interval: 2.2-45.8)].

Abnormal CR results were infrequent in mild exacerbations and planned discharged patients. The female sex was the only predictor of an abnormal CR result.