RESUMEN
Several clinical scales have been developed to assess the severity of bronchiolitis as well as the probability of needing in-hospital care. A recent systematic review of 32 validated clinical scores for bronchiolitis concluded that 6 of them (Wood-Downes, M-WCAS, Respiratory Severity Score, Respiratory Clinical Score, Respiratory Score and Bronchiolitis risk of admission score) were the best ones regarding reliability, sensitivity, validity, and usability. However, to the best of our knowledge, no study has compared all of them in a clinical scenario. Also, after this review, three more scales were published: BROSJOD, Tal modified, and one score developed by PERN. Our main aim was to compare the ability of different clinical scales for bronchiolitis to predict any relevant outcome. A prospective observational study was conducted that included patients of up to 12 months old attended to, due to bronchiolitis, in the paediatric Emergency Department of a secondary university hospital from October 2019 to January 2022. For each patient, the attending clinician filled in a form with the items of the scales, decomposed, in order to prevent the clinician from knowing the score of each scale. Then, the patient was managed according to the protocol of our Emergency Department. A phone call was made to each patient in order to check whether the patient ended up being admitted in the next 48 h. In the case of those that were impossible to contact by phone, the clinical history was reviewed. For the purpose of the study, any of the following were considered to be a relevant outcome: admission to ward and need for supplementary oxygen, non-invasive ventilation (NIV) or intravenous fluids, and admission to the paediatric intensive care unit (PICU) within the next 48 h or death. For the aim of the study, the area under the curve (AUC) and the odds ratio (OR) for a relevant outcome were calculated in each scale. Also, the best cut-off point was estimated according to the Youden index, and its sensitivity (Sn) and specificity (Sp) for a relevant outcome were calculated. We included 265 patients (52.1% male) with a median age of 5.3 months (P25-P75 2.6-7.4). Among them, 46 (17.4%) had some kind of relevant outcome. AUC for prediction of a relevant outcome ranged from 0.705 (Respiratory Score) to 0.786 (BRAS), although no scale performed significantly better than others. A score ≤ 2 in the PERN scale showed a sensitivity of 91.3% (CI95% 79.7-96.6) for a relevant outcome, with only 4 misdiagnosed patients (only 2 of them needed NIV). Conclusions: There were no differences in the performance of the nine scales to predict relevant outcomes in patients with bronchiolitis. However, the PERN scale might be more useful to select patients at low risk of a severe outcome. What is Known: ⢠Several clinical scales are used to assess the severity of bronchiolitis. Nevertheless, none of them seems to be better than others. What is New: ⢠This is the first study comparing different bronchiolitis scales in a real clinical scenario. None of the nine scales compared performed better than the other. However, the PERN scale might be more useful to select patients at low risk of relevant outcomes.