Diagnostic value of the SHAPE questionnaire in recognition of the hyperventilation syndrome in children: a pilot study

• Published in: Archives de pédiatrie : organe officiel de la Société française de pédiatrie 2009-08, Vol.16 (8), p.1118-1123
• Main Authors: Sznajder, M, Stheneur, C, Baranes, T, Fermanian, C, Rossignol, B, Chevallier, B, Bidat, E
• Format: Article
• Language: fre
• Subjects: Adolescent; Asthma - complications; Asthma - diagnosis; Child; Diagnosis, Differential; Female; France; Humans; Hyperventilation - classification; Hyperventilation - diagnosis; Hyperventilation - etiology; Male; Odds Ratio; Pilot Projects; Reference Values; Reproducibility of Results; Surveys and Questionnaires; Syndrome; Wounds and Injuries - complications; Wounds and Injuries - diagnosis; Young Adult
• ISSN: 1769-664X
• DOI: 10.1016/j.arcped.2009.04.013

Determination of a diagnostic scoring method for hyperventilation syndrome (HVS) in children and proposal of a simplified questionnaire. We used the main clinical signs of HVS in children and adolescents identified in a previous study and classified them according to their odds ratios (OR). The intensity of each sign, measured using a visual analogic scale in the previous study, led us to analyze several scoring methods, with a breakdown between major and minor signs according to their OR. Building receiver operating characteristics (ROC) curves allowed us to choose the best diagnostic combination. A sample of 85 children and adolescents aged from 7 to 20 years (mean age, 11.9 years) was tested. This sample comprised 45 children with isolated HVS or HVS associated with asthma and 40 controls, with asthma but without HVS or attending our emergency unit after trauma. Seven respiratory signs and 10 nonrespiratory signs were selected. For each scoring method, a significant difference was observed between HVS and non-HVS patients. The most suitable area under the curve (0.934) and the best combination between specificity and sensitivity (Sp=0.90 and Se=0.82) were obtained when the scoring method was 0,3,6 for major signs, and 0,1,2 for minor signs, according to perceived intensity of each sign ("never or almost never", "sometimes" or "often or very often"). Since no gold standard is available to confirm the diagnosis of HVS in children, we propose using a simplified questionnaire composed of 17 items to compute a diagnostic score. The questionnaire will be validated shortly in a new prospective study.