SESSION TITLE: Asthma Diagnosis & Evaluation
SESSION TYPE: Original Investigation Slide
PRESENTED ON: Tuesday, October 29, 2013 at 02:45 PM - 04:15 PM
PURPOSE: The FEV1/FVC ratio can be reduced in asthma, but the degree of reduction from predicted values to diagnose asthma has not been well established. The FEV1/FVC ratio Z-score is the difference between actual and predicted FEV1/FVC ratio, normalized to the standard deviation of predicted ratio from a healthy population. The optimum Z-score depends upon the prevalence of asthma in the population (i.e. pre-test probability). Accordingly, we wanted to estimate the FEV1/FVC ratio Z-score threshold with the highest accuracy for the diagnosis of airflow limitation in a population of asthmatics for different pre-test probabilities.
METHODS: Data for this analysis came from four trials of asthmatics enrolled in trials from the Asthma Clinical Research Centers. Measured and predicted FEV1/FVC ratios were obtained with calculation of the Z-score for each participant. Across a range of hypothetical asthma prevalence and Z-score thresholds, overall diagnostic accuracy was calculated.
RESULTS: 1608 participants were included (median age 39 years, 71% female and 61% Caucasian). The median age of asthma onset was 13 years, with median ACQ and ASUI scores of 1.71 and 0.80, respectively. The median FEV1% predicted was 84%. In a population with 50% prevalence of asthma, the highest accuracy (68%) is achieved with a Z-score threshold of -1.0 (16th percentile), corresponding to a 6% reduction from predicted ratio. If the prevalence of asthma is 5%, then the optimum Z-score is -2.0 (2nd percentile), corresponding to a 12% reduction from predicted ratio.
CONCLUSIONS: Use of a 6% reduction in FEV1/FVC ratio from predicted ratio yields the best accuracy for diagnosing asthma in a symptomatic population with possible asthma (prevalence of 50%). In contrast, when screening at the population level, a 12% reduction in FEV1/FVC ratio would afford the highest accuracy.
CLINICAL IMPLICATIONS: When interpreting spirometry, consideration of the pre-test probability is an important consideration in diagnosis of asthma based on airflow limitation.
DISCLOSURE: The following authors have nothing to disclose: Allison Lambert, M. Bradley Drummond, Christine Wei, Meredith McCormack, Robert Wise
No Product/Research Disclosure Information