Cardiac Performance in Mechanically Ventilated COVID-19 Patients

The objective is to determine the prevalence of left ventricle dysfunction and acute cor pulmonale, and its association with respiratory mechanics, in 100 consecutive critically ill COVID-19 patients, who were assessed with critical care echocardiography (CCE) within the first 24 hours of mechanical ventilation.

Material and Methods This prospective, multicenter study was conducted between April and June 2020 in the intensive care unit (ICU) of four university-affiliated hospitals of Santiago, Chile. Local ethics committee of each center approved the study and waived the need to provide written informed consent (protocol ID: 200422002). All participating centers have expertise in CCE on clinical grounds.

The study population Were included systematically all consecutive patients with severe SARS-CoV-2 pneumonia, defined as respiratory failure requiring invasive mechanical ventilation (MV). Patients were included within the first 24 hours of MV. COVID-19 was confirmed by a positive polymerase chain reaction test. The main exclusion criteria were age under 18 years old, severe valvulopathy, poor ultrasound window and do-not-resuscitate status.

Demographic data, Acute Physiology and Chronic Health Evaluation II (APACHE), Sequential Organ Failure Assessment (SOFA) and respiratory system mechanics were recorded.

Echocardiography Transthoracic echocardiography was performed by intensivists trained in echocardiography in according to CCE. Echocardiographic measurements were obtained with a Vivid i echocardiography system (GE Medical Systems, Milwaukee, WI, USA), Philips CX 50 (Philips Healthcare, DA Best, The Netherlands), and Mindray M9 (Bio-Medical Electronics Co., Shenzhen, Chine), as used in each center. All patients were adapted to MV on continuous IV sedation during the echocardiographic assessment. Measurements were acquired at end-expiration and averaged over three consecutive cardiac cycles in according to current recommendations.

Standard echocardiographic views were acquired. Left ventricular systolic function was assessed by the left ventricular ejection fraction (LVEF), as measured with Simpson's modified rule. Based on LVEF patients were categorized in hyperkinetic (LVEF >60%), normokinetic (LVEF between 45% and 60%) and hypokinetic (LVEF <45%).

Cardiac output (CO) was calculated from the left ventricular outflow tract (LVOT) as described by Mclean et al. The diameter of the LVOT was taken from the long parasternal view. Pulsed wave Doppler samples were obtained at the LVOT from the apical view. Doppler velocity curves were manually traced and an average of three measures of velocity time integral (VTI) was calculated. The stroke volume (SV) was calculated as the product of the LVOT area and the VTI. The CO was calculated as the product of the SV and the heart rate obtained during measurement of the aortic VTI. Mitral annular plane systolic excursion (MAPSE) was obtained through an M-mode vector at the level of the mitral annulus at the LV lateral wall. MAPSE was measured in millimeters as described (5). Peak mitral annular myocardial velocity wave (s') was recorded at the level of the mitral annulus at the LV lateral wall with Tissue Doppler Imaging (TDI) as previously described.

Left ventricular diastolic function was assessed by mitral inflow pulsed wave Doppler, to measure early peak velocity (E) and atrial velocity (A). The early diastolic peak velocity (e') of the lateral mitral annulus was also measured with TDI. From these variables E/A and E/e' ratios were calculated. Diastolic function was categorized as normal, grade I to grade III according to current recommendations.

The presence of acute cor pulmonale was assessed through the relation between left and right ventricular end-diastolic areas (LVEDA and RVEDA), and by the presence of paradoxical septum motion. LVEDA and RVEDA were measured from the apical four chambers view, and the RVEDA/LVEDA ratio was calculated. Right ventricle (RV) dilatation was defined as RVEDA/LVEDA ratio > 0.6, ACP was defined as a dilated RV associated with the presence of paradoxical septum motion. Severe ACP was defined as a severely dilated RV (RVEDA/LVEDA ratio >1). Paradoxical septum motion was defined as end-systolic bulging of the interventricular septum toward the left ventricle, while analyzing loops in slow motion. It indirectly assesses RV performance and RV end-diastolic pressures. The paradoxical septum motion was assessed on the four-chamber view. In addition, RV systolic function was assessed by the tricuspid annular plane systolic excursion (TAPSE) obtained through an M-mode vector at the level of the tricuspid annulus.

Hemodynamic assessment The hemodynamic profile was based on cardiac index (CI), patients were categorized as hypodynamic (CI<2 L⋅min-1⋅m2), normodynamic (CI between 2 and 4 L⋅min-1⋅m2) and hyperdynamic (CI>4 L⋅min-1⋅m2). Hemodynamic variables and norepinephrine doses were recorded. Maximum and minimum inferior vena cava (IVC) diameter was measured in M-mode tracing from a longitudinal subcostal view. Fluid responsiveness was assessed by either the respiratory variations of inferior vena cava (IVC), or by the arterial pulse pressure variation (PPV).

The PiCCO (Continuous Cardiac Output Pulse Index) device was used in thirty patients to assess cardiac output (CO). An average of three consecutive thermodilution measurements was used. Other variables such as global end-diastolic index (GEDI), intrathoracic blood volume index (ITBVI), systemic vascular resistance index (SVRi), cardiac function index (CFi), global ejection fraction (GEF), extravascular lung water index (ELWI), and pulmonary vascular permeability index (PVPI), were recorded.

Tissue perfusion was assessed by capillary refill time (CRT) and lactate levels. The peripheral perfusion index (PFI) derived from the photoelectric plethysmography signal of pulse oximetry was also used.

Computer tomography (CT) pulmonary angiography was performed based on the clinical suspicion of PE.

Biomarkers High-sensitive Troponin T and D-Dimer (DD) concentrations were measured simultaneously with echocardiographic assessment. Increases in Troponin T were defined as a value >14ng/l.

Statistical Analysis As the study was descriptive sample size but was arbitrarily set at 100 consecutive patients. Normality was tested by Kolmogorov-Smirnov test. The continuous data is presented as mean ± standard deviation or as median and interquartile ranges, depending on the distribution. Comparisons between groups were analyzed by t-test or Wilcoxon rank-sum (WRS) test according to data distribution. Percentages were analyzed using two-proportion z-test. Categorical variables were compared with chi square test. Pearson or Spearman correlation was performed according to data distribution.

In order to identify independent predictors for mortality, a multivariate regression analysis was carried out including all variables of interest associated with mortality in a univariate analysis (p value < 0.05) and adjusting for the other covariates of the logistic regression model. The discrimination of the model was assessed by the area under the receiver operating characteristic (ROC) curve. Statistical analysis was performed with SPSS (version 22.0, IBM SPSS Inc., Chicago, IL, USA). A p value < 0.05 was considered statistically significant.