Decrease of haemoconcentration reliably detects hydrostatic pulmonary oedema in dyspnoeic patients in the emergency department - a machine learning approach.
Francesco Gavelli, Luigi Mario Castello, Xavier Monnet, Danila Azzolina, Ilaria Nerici, Simona Priora, Valentina Giai Via, Matteo Bertoli, Claudia Foieni, Michela Beltrame, Mattia Bellan, Pier Paolo Sainaghi, Nello De Vita, Filippo Patrucco, Jean-Louis Teboul, Gian Carlo Avanzi
{"title":"Decrease of haemoconcentration reliably detects hydrostatic pulmonary oedema in dyspnoeic patients in the emergency department - a machine learning approach.","authors":"Francesco Gavelli, Luigi Mario Castello, Xavier Monnet, Danila Azzolina, Ilaria Nerici, Simona Priora, Valentina Giai Via, Matteo Bertoli, Claudia Foieni, Michela Beltrame, Mattia Bellan, Pier Paolo Sainaghi, Nello De Vita, Filippo Patrucco, Jean-Louis Teboul, Gian Carlo Avanzi","doi":"10.1186/s12245-024-00698-y","DOIUrl":null,"url":null,"abstract":"<p><strong>Background: </strong>Haemoglobin variation (ΔHb) induced by fluid transfer through the intestitium has been proposed as a useful tool for detecting hydrostatic pulmonary oedema (HPO). However, its use in the emergency department (ED) setting still needs to be determined.</p><p><strong>Methods: </strong>In this observational retrospective monocentric study, ED patients admitted for acute dyspnoea were enrolled. Hb values were recorded both at ED presentation (T<sub>0</sub>) and after 4 to 8 h (T<sub>1</sub>). ΔHb between T<sub>1</sub> and T<sub>0</sub> (ΔHb<sub>T1-T0</sub>) was calculated as absolute and relative value. Two investigators, unaware of Hb values, defined the cause of dyspnoea as HPO and non-HPO. ΔHb<sub>T1-T0</sub> ability to detect HPO was evaluated. A machine learning approach was used to develop a predictive tool for HPO, by considering the ability of ΔHb as covariate, together with baseline patient characteristics.</p><p><strong>Results: </strong>Seven-hundred-and-six dyspnoeic patients (203 HPO and 503 non-HPO) were enrolled over 19 months. Hb levels were significantly different between HPO and non-HPO patients both at T<sub>0</sub> and T<sub>1</sub> (p < 0.001). ΔHb<sub>T1-T0</sub> were more pronounced in HPO than non-HPO patients, both as relative (-8.2 [-11.2 to -5.6] vs. 0.6 [-2.1 to 3.3] %) and absolute (-1.0 [-1.4 to -0.8] vs. 0.1 [-0.3 to 0.4] g/dL) values (p < 0.001). A relative ΔHb<sub>T1-T0</sub> of -5% detected HPO with an area under the receiver operating characteristic curve (AUROC) of 0.901 [0.896-0.906]. Among the considered models, Gradient Boosting Machine showed excellent predictive ability in identifying HPO patients and was used to create a web-based application. ΔHb<sub>T1-T0</sub> was confirmed as the most important covariate for HPO prediction.</p><p><strong>Conclusions: </strong>ΔHb<sub>T1-T0</sub> in patients admitted for acute dyspnoea reliably identifies HPO in the ED setting. The machine learning predictive tool may represent a performing and clinically handy tool for confirming HPO.</p>","PeriodicalId":13967,"journal":{"name":"International Journal of Emergency Medicine","volume":null,"pages":null},"PeriodicalIF":2.0000,"publicationDate":"2024-09-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11375861/pdf/","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"International Journal of Emergency Medicine","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1186/s12245-024-00698-y","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"EMERGENCY MEDICINE","Score":null,"Total":0}
引用次数: 0
Abstract
Background: Haemoglobin variation (ΔHb) induced by fluid transfer through the intestitium has been proposed as a useful tool for detecting hydrostatic pulmonary oedema (HPO). However, its use in the emergency department (ED) setting still needs to be determined.
Methods: In this observational retrospective monocentric study, ED patients admitted for acute dyspnoea were enrolled. Hb values were recorded both at ED presentation (T0) and after 4 to 8 h (T1). ΔHb between T1 and T0 (ΔHbT1-T0) was calculated as absolute and relative value. Two investigators, unaware of Hb values, defined the cause of dyspnoea as HPO and non-HPO. ΔHbT1-T0 ability to detect HPO was evaluated. A machine learning approach was used to develop a predictive tool for HPO, by considering the ability of ΔHb as covariate, together with baseline patient characteristics.
Results: Seven-hundred-and-six dyspnoeic patients (203 HPO and 503 non-HPO) were enrolled over 19 months. Hb levels were significantly different between HPO and non-HPO patients both at T0 and T1 (p < 0.001). ΔHbT1-T0 were more pronounced in HPO than non-HPO patients, both as relative (-8.2 [-11.2 to -5.6] vs. 0.6 [-2.1 to 3.3] %) and absolute (-1.0 [-1.4 to -0.8] vs. 0.1 [-0.3 to 0.4] g/dL) values (p < 0.001). A relative ΔHbT1-T0 of -5% detected HPO with an area under the receiver operating characteristic curve (AUROC) of 0.901 [0.896-0.906]. Among the considered models, Gradient Boosting Machine showed excellent predictive ability in identifying HPO patients and was used to create a web-based application. ΔHbT1-T0 was confirmed as the most important covariate for HPO prediction.
Conclusions: ΔHbT1-T0 in patients admitted for acute dyspnoea reliably identifies HPO in the ED setting. The machine learning predictive tool may represent a performing and clinically handy tool for confirming HPO.
期刊介绍:
The aim of the journal is to bring to light the various clinical advancements and research developments attained over the world and thus help the specialty forge ahead. It is directed towards physicians and medical personnel undergoing training or working within the field of Emergency Medicine. Medical students who are interested in pursuing a career in Emergency Medicine will also benefit from the journal. This is particularly useful for trainees in countries where the specialty is still in its infancy. Disciplines covered will include interesting clinical cases, the latest evidence-based practice and research developments in Emergency medicine including emergency pediatrics.