Oscillometry for the diagnosis of asthma in children: a systematic review.
Diagnosing asthma in children and young people (CYP) remains challenging. Oscillometry is a promising tool and is feasible from 2 years of age. European Respiratory Society (ERS) technical standards and bronchodilator response (BDR) oscillometry thresholds have been published, but diagnostic accuracy is not established.
We systematically reviewed studies comparing oscillometry and spirometry in CYP under investigation for asthma. Reference standards were positive BDR or positive methacholine challenge test (MCT). Primary aims were to investigate the sensitivity and specificity of current ERS oscillometry thresholds (>40% decrease in resistance at 5 Hz (R 5), >50% increase in reactance at 5 Hz (X 5) or >80% decrease in the area under the reactance curve); secondary aims were to identify oscillometry threshold values optimising both sensitivity and specificity.
11 studies were included; six (n=992 CYP) utilised BDR and five (n=531 CYP) MCT as reference standard. Meta-analysis was not possible due to heterogeneity of results reported. In two studies using current ERS BDR thresholds, zero sensitivity and high specificity (>85%) were observed. In weighted regression analyses of BDR studies, a 17.0% decrease in resistance at 5-6 Hz had sensitivity and specificity of 71.6% (95% CI 69.7-73.7%); a 20.2% increase in X 5 had sensitivity and specificity of 68.6% (95% CI 66.6-70.8%). Similarly, 27.7% increase in R 5 had sensitivity and specificity of 73.6% (95% CI 71.9-75.3%) for MCT.
Currently recommended ERS thresholds for oscillometry BDR have low sensitivity. Proposed thresholds for defining positive BDR and MCT by oscillometry require prospective validation and adoption of standards for measuring and reporting oscillometry parameters in future diagnostic comparative studies.
We systematically reviewed studies comparing oscillometry and spirometry in CYP under investigation for asthma. Reference standards were positive BDR or positive methacholine challenge test (MCT). Primary aims were to investigate the sensitivity and specificity of current ERS oscillometry thresholds (>40% decrease in resistance at 5 Hz (R 5), >50% increase in reactance at 5 Hz (X 5) or >80% decrease in the area under the reactance curve); secondary aims were to identify oscillometry threshold values optimising both sensitivity and specificity.
11 studies were included; six (n=992 CYP) utilised BDR and five (n=531 CYP) MCT as reference standard. Meta-analysis was not possible due to heterogeneity of results reported. In two studies using current ERS BDR thresholds, zero sensitivity and high specificity (>85%) were observed. In weighted regression analyses of BDR studies, a 17.0% decrease in resistance at 5-6 Hz had sensitivity and specificity of 71.6% (95% CI 69.7-73.7%); a 20.2% increase in X 5 had sensitivity and specificity of 68.6% (95% CI 66.6-70.8%). Similarly, 27.7% increase in R 5 had sensitivity and specificity of 73.6% (95% CI 71.9-75.3%) for MCT.
Currently recommended ERS thresholds for oscillometry BDR have low sensitivity. Proposed thresholds for defining positive BDR and MCT by oscillometry require prospective validation and adoption of standards for measuring and reporting oscillometry parameters in future diagnostic comparative studies.
Authors
Seneviratne Seneviratne, Law-Gleen Law-Gleen, Broomfield Broomfield, Divall Divall, Devani Devani, Ducharme Ducharme, Gaillard Gaillard
View on Pubmed