Phenotype clustering of hospitalized high-risk patients with COVID-19 - a machine learning approach within the multicentre, multinational PCHF-COVICAV registry.

IMTRODUCTION: The high-risk population of patients with cardiovascular (CV) disease or risk factors (RF) suffering from COVID-19 is heterogeneous. Several predictors for impaired prognosis have been identified. However, with machine learning (ML) approaches, certain phenotypes may be confined to classify the affected population and to predict outcome. This study aimed to phenotype patients using unsupervised ML technique within the International Postgraduate Course Heart Failure Registry for patients hospitalized with COVID-19 and Cardiovascular disease and/or RF (PCHF-COVICAV). MATERIAL AND METHODS: Patients from the eight centres with follow-up data available from the PCHF-COVICAV registry were included in this ML analysis (K-medoids algorithm). RESULTS: Out of 617 patients included into the prospective part of the registry, 458 [median age: 76 (IQR:65-84) years, 55% male] were analyzed and 46 baseline variables, including demographics, clinical status, comorbidities and biochemical characteristics were incorporated into the ML. Three clusters were extracted by this ML method. Cluster 1 (n = 181) represents mainly women with the least number of overall comorbidities and cardiovascular RF. Cluster 2 (n = 227) is characterized mainly by men with non-CV conditions and less severe symptoms of infection. Cluster 3 (n=50) mainly represents men with the highest prevalence of cardiac comorbidities and RF, more extensive inflammation and organ dysfunction with the highest 6-month all-cause mortality risk. CONCLUSIONS: The ML process has identified three important clinical clusters from hospitalized COVID-19 CV and/or RF patients. The cluster of males with severe CV disease, particularly HF, and multiple RF presenting with increased inflammation had a particularly poor outcome.

Heart failure in COVID-19: the multicentre, multinational PCHF-COVICAV registry.

AIMS: We assessed the outcome of hospitalized coronavirus disease 2019 (COVID-19) patients with heart failure (HF) compared with patients with other cardiovascular disease and/or risk factors (arterial hypertension, diabetes, or dyslipidaemia). We further wanted to determine the incidence of HF events and its consequences in these patient populations. METHODS AND RESULTS: International retrospective Postgraduate Course in Heart Failure registry for patients hospitalized with COVID-19 and CArdioVascular disease and/or risk factors (arterial hypertension, diabetes, or dyslipidaemia) was performed in 28 centres from 15 countries (PCHF-COVICAV). The primary endpoint was in-hospital mortality. Of 1974 patients hospitalized with COVID-19, 1282 had cardiovascular disease and/or risk factors (median age: 72 [interquartile range: 62-81] years, 58% male), with HF being present in 256 [20%] patients. Overall in-hospital mortality was 25% (n = 323/1282 deaths). In-hospital mortality was higher in patients with a history of HF (36%, n = 92) compared with non-HF patients (23%, n = 231, odds ratio [OR] 1.93 [95% confidence interval: 1.44-2.59], P < 0.001). After adjusting, HF remained associated with in-hospital mortality (OR 1.45 [95% confidence interval: 1.01-2.06], P = 0.041). Importantly, 186 of 1282 [15%] patients had an acute HF event during hospitalization (76 [40%] with de novo HF), which was associated with higher in-hospital mortality (89 [48%] vs. 220 [23%]) than in patients without HF event (OR 3.10 [2.24-4.29], P < 0.001). CONCLUSIONS: Hospitalized COVID-19 patients with HF are at increased risk for in-hospital death. In-hospital worsening of HF or acute HF de novo are common and associated with a further increase in in-hospital mortality.

Assessment of the relation between mean platelet volume, non-dipping blood pressure pattern, and left ventricular mass index in sustained hypertension.

BACKGROUND: Elevated mean platelet volume may reflect presence of active large platelets, which lead to fatal or non-fatal cardiovascular events. In recent studies, lack of nocturnal blood pressure fall was presented as an independent predictor of poor prognosis in essential hypertension. The relation of raised MPV with left ventricular hypertrophy has also been reported in hypertension. The aim of this study was to investigate the relation between MPV, non-dipping blood pressure pattern, and left ventricular mass index (LVMI) in sustained hypertension. MATERIAL AND METHODS: A total of 2500 patients, whose ambulatory blood pressure (ABP) records had been evaluated retrospectively between January 2010 and December 2012, were included. Patients were divided into 3 groups according to their ABP values: non-dipper hypertensive (n=289), dipper hypertensive (n=255), and normotensive (n=306). The MPV levels and biochemical analyses were recorded from patient files and, LVMI were automatically calculated using a regression equation. RESULTS: The non-dipper and dipper hypertensive groups had significantly higher MPV levels than normotensives (8.4±1 fL, 8.3±1 fL, and 8.1±0.6 fL, respectively, p<0.001). However, there was no difference among the non-dipper and dipper groups in terms of MPV level (p=0.675). Although LVMI was significantly different between non-dipper, dipper, and normotensive groups (p=0.009), no correlation was found between MPV level and LVMI in dipper and non-dipper hypertensive patients (r=-0.080, p=0.142). There was a weak correlation between MPV level and ambulatory 24-h diastolic and systolic blood pressure (r=0.076, p=0.027, and r=0.073, p=0.033, respectively). CONCLUSIONS: We demonstrated that there was no correlation between MPV level, non-dipping pattern of blood pressure, and LVMI in sustained hypertension.