Avaliação do desempenho dos escores prognósticos ADHERE e GWTG-HF em população internada por insuficiência cardíaca descompensada em hospital quaternário no Rio de Janeiro

Abstract

Heart failure (HF) is a clinical syndrome with high prevalence in the world and Brazil. Decompensations are frequent in HF and have elevated morbimortality. Risk stratification is an important step on clinical management and risk scores could help. Hence, to be applied on a population other than the original one they should be externally validated. ADHERE (Acute Decompensated Heart Failure Registry) and GWTG-HF (Get With the Guidelines – Heart Failure) were developed to stratify inhospital mortality (IHM) risk for Decompensated heart failure (DHF) patients, while the MAGGIC (Meta-Analysis Global Group in Chronic Heart Failure) score was validated to stratify long term mortality of nonhospitalized patients. Objectives: The primary end-point of this study was to validate IHM risk scores in patients admited for DHF in the Hospital Universitário Pedro Ernesto (HUPE), Rio de Janeiro, Brazil. Secondary end-points were: to evaluate the risk factors related to IHM and long-term post-discharge mortality, to evaluate the accuracy of MAGGIC score in DHF patients, and to analyze the adequacy of the pharmacological HF prescription. Methodology: A retrospective longitudinal single-center study was conducted. The population consisted of patients hospitalized for DHF in HUPE, between 2019 and 2021. Validation of the scores was performed by calculating the discriminatory index (DI) and analysis of classification capacity. Multivariated analysis with logistic regression was performed to evaluate the factors related to inhospital and long term mortality. Results: 304 patients with median age of 63 years were included. ADHERE and GWTG-HF DI were moderate, with an AUC of 0,69 and 0,65, respectivelly. Although the IHM observed was higher in all subgroups of patients compared to the original cohort, the GWTG-HF score stratified the risk groups that had patients included in the cohort. On the other hand, the ADHERE score was not adequate to classify patients. Admission Blood Urea Nitrogen (BUN) was the only prognostic factor associated with IHM on multivariated analysis (OR 1,03; CI 95% 1,01-1,05; p<0,001). Admission BUN (1,013; CI 95% 1,003-1,024; p= 0,013) and right ventricular dysfunction were associated to post-discharge mortality (1,79; IC 95% 1,112-2,881; p= 0,016) in multivariated analysis. MAGGIC risk score was also associated with post-discharge mortality (1,061; CI 95% 1,025-1,098; p= 0,001). Concerning pharmacological prescprition before admission, 78,9%, 58%, 39% e 9% of the patients had beta-blockers (BB), renin-angiotensin-aldosterone system inhibitors (RAASi), mineralocorticoids receptors inhibitors (MRI) and sodium-glucose cotransporter type 2 inhibitors prescription, respectively. Ambulatorial prescription was not associated with IHM. Long term mortality was not associated with pre-discharge prescription of BB (p=0.065) and RASSi (p=0.079) on the entire cohort. Conclusion: GWTG-HF score can be used to predict IHM in the cohort evaluated. For full validation, recalibration would be required. Admission BUN was able to select patients at higher risk of IHM. MAGGIC score, the presence of VD dysfunction and elevated BUN levels on admission were associated with post-discharge mortality