Premium
A clinical risk score for pulmonary artery thrombosis during acute chest syndrome in adult patients with sickle cell disease
Author(s) -
Winchenne Anaïs,
Cecchini Jérôme,
Deux JeanFrançois,
De Prost Nicolas,
Razazi Keyvan,
Carteaux Guillaume,
Galacteros Frederic,
Habibi Anoosha,
Bartolucci Pablo,
Melica Giovanna,
Khellaf Mehdi,
Michel Marc,
Maitre Bernard,
Mekontso Dessap Armand
Publication year - 2017
Publication title -
british journal of haematology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.907
H-Index - 186
eISSN - 1365-2141
pISSN - 0007-1048
DOI - 10.1111/bjh.14914
Subject(s) - medicine , acute chest syndrome , framingham risk score , receiver operating characteristic , acute coronary syndrome , cardiology , thrombosis , confidence interval , logistic regression , retrospective cohort study , disease , sickle cell anemia , myocardial infarction
Summary Pulmonary artery thrombosis ( PAT ) is involved in lung vascular dysfunction during acute chest syndrome ( ACS ) complicating sickle cell disease ( SCD ). No clinical score is available to identify patients eligible for multi‐detector computed tomography ( MDCT ) angiography during ACS . This retrospective study aimed to develop a risk score for PAT during ACS ( PAT ‐ ACS risk score). Patients with SCD were investigated by MDCT during ACS . A logistic regression was performed to determine independent risks factors for PAT and to build the PAT ‐ ACS risk score. A total of 43 episodes (11·9%) of PAT were diagnosed in 361 episodes of ACS . Multivariate analysis identified four risk factors, which were included in the PAT ‐ ACS risk score: a baseline haemoglobin >82 g/l, the lack of a triggering factor for ACS , a platelet count >440 × 10 9 /l and a Pa CO 2 <38 mmHg at ACS diagnosis. The area under the receiver operating characteristic curve for the PAT ‐ ACS risk score was 0·74 (95% confidence interval [ CI ] 0·69–0·79) and differed from that of the revised Geneva score (0·63 (95% CI 0·58–0·69); P = 0·04). The negative predictive value of a PAT ‐ ACS risk score ≥2 was 94%. In conclusion, we propose a simple clinical risk score to identify SCD patients at high risk of PAT during ACS .