key: cord-351509-aau3gx6f
authors: Li, Yuman; Li, He; Zhu, Shuangshuang; Xie, Yuji; Wang, Bin; He, Lin; Zhang, Danqing; Zhang, Yongxing; Yuan, Hongliang; Wu, Chun; Sun, Wei; Zhang, Yanting; Li, Meng; Cui, Li; Cai, Yu; Wang, Jing; Yang, Yali; Lv, Qing; Zhang, Li; Xie, Mingxing
title: Prognostic Value of Right Ventricular Longitudinal Strain in Patients with COVID-19
date: 2020-04-28
journal: JACC Cardiovasc Imaging
DOI: 10.1016/j.jcmg.2020.04.014
sha: 
doc_id: 351509
cord_uid: aau3gx6f

Abstract Objectives We aimed to investigate whether right ventricular longitudinal strain (RVLS) was independently predictive of higher mortality in coronavirus disease 2019 (COVID-19) patients. Background RVLS obtained from two-dimensional speckle-tracking echocardiography (2D-STE) has been recently demonstrated to be a more accurate and sensitive tool to estimate RV function. The prognostic value of RVLS in patients with COVID-19 remains unknown. Methods 120 consecutive patients with COVID-19 who underwent echocardiography examination were enrolled in our study. Conventional right ventricular (RV) function parameters, including RV fractional area change (RVFAC), tricuspid annular plane systolic excursion (TAPSE) and tricuspid tissue Doppler annular velocities (S’), were obtained. RVLS was determined by 2D-STE. RV function was categorized by tertiles of RVLS. Results Compared with patients in the highest RVLS tertile, those in the lowest tertile were more likely to have a higher heart rate, D-dimer and C-reactive protein, high-flow oxygen and invasive mechanical ventilation therapy, higher incidence of acute heart injury, acute respiratory distress syndrome (ARDS) and deep vein thrombosis, and higher mortality. After a median follow-up of 51 days, 18 patients died. Compared with survivors, non-survivors displayed enlarged right-heart chamber, diminished RV function, and elevated pulmonary artery systolic pressure. Male, ARDS, RVLS, RVFAC and TAPSE were significant univariate predictors of higher risk of mortality (P < 0.05 for all). The Cox model using RVLS (hazard ratio [HR]: 1.33, 95% confidence intervals [CI]: 1.15～1.53; P < 0.001; Akaike Information Criterion [AIC] =129; C-index = 0.89) was found to predict higher mortality more accurately than that with RVFAC (AIC =142; C-index = 0.84) and TAPSE (AIC = 144; C-index = 0.83). The best cutoff value of RVLS for prediction of outcome was −23% (area under the curve, 0.87; P < 0.001; sensitivity, 94.4%; specificity, 64.7%). Conclusions RVLS is a powerful predictor of higher mortality in patients with COVID-19. Our study supports the application of RVLS to identify higher risk COVID-19 patients.

The coronavirus disease 2019 caused by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) reaches more than 170 countries, resulting in considerable morbidity and mortality. Although currently available studies confirmed the presence of myocardial injury and its association with mortality in patients with COVID-19, 1-2 they lack evidence from echocardiography to determine the features of cardiac injury. As the prevalence of acute respiratory distress syndrome (ARDS) has been reported to be 29% -67% of critically ill patients with COVID-19, 3,4 right ventricular (RV) function is supposed to be more susceptible to impairment due to increased RV afterload. In clinical practice, RV structure and function are mainly evaluated by echocardiography. The conventional echocardiographic parameter have limited diagnostic value, as they may fail to detect early abnormalities of RV systolic function. 5 Recently, two-dimensional speckle-tracking echocardiography (2D-STE) has been introduced, which evaluates myocardial function accurately and reproducibly. 6, 7 Because of its capability to detect subclinical impairment of cardiac function, 2D-STE has been extensively applied to investigate RV function in different clinical settings. [8] [9] [10] Moreover, the RV longitudinal strain (RVLS) derived from 2D-STE has been demonstrated to be of prognostic value. 11 To the best of our knowledge, there are no data regarding the use of RVLS in patients with COVID-19. In addition, confirming the role of RVLS in these patients may be of additional significance, as most cases exhibit the preserved conventional echocardiographic parameters, among whom detection and risk stratification may be challenging. Accordingly, the purpose of this study was to evaluate the prognostic value of RVLS in patients with COVID-19.

This observational study was performed at the west branch of Union Hospital, Tongji Medical College, Huazhong University of Science and Technology-a designated hospital to treat patients with COVID-19. We included a total of 150 consecutive adult patients with COVID-19 who were diagnosed according to the interim guidance of World Health Organization 12 from Feb 12, 2020 to Mar 15, 2020 . Considering the presence of cardiac damage in patients with COVID-19, bedside echocardiography was performed in all patients from 3 wards managed by the investigators for evaluation of cardiac function. The median time from admission to echocardiography performed was 7 days (interquartile range [IQR] [3] [4] [5] [6] [7] [8] [9] [10] . Patients with known cardiomyopathy, previous myocardial infarction or suboptimal images were excluded. Of these patients, 2 had dilated cardiomyopathy, 4 had old myocardial infarction, and 24 did not have images of sufficient quality for echocardiographic analysis.

The remaining 120 patients were included in our final analysis. The control group consisted of 37 healthy volunteers who had no cardiopulmonary disease based on physical check-up, electrocardiogram, chest X-ray and echocardiography.

The study complied with the edicts of the 1975 Declaration of Helsinki 13 

Patients' demographic characteristics, medical history, laboratory examinations, comorbidities, complication, treatment and outcomes were retrieved from electronic medical records. Cardiac biomarkers measured on admission included hypersensitive troponin I (hs-TNI), creatine kinase muscle-brain (CK-MB) and B-type natriuretic peptide (BNP).

These data were independently reviewed and entered into the computer database by two analysts (C.W. and W.S.) The final follow-up date was April 2, 2020

Bedside transthoracic echocardiographic examinations were performed in all patients with the EPIQ7C ultrasound systems (Philips Medical Systems, Andover, MA, USA). 2D and Doppler echocardiography were acquired based on the guidelines of the American Society of Echocardiography. 14 Images were stored and analyzed by two independent observers (Y.C. and L.C.) blinded to clinical data.

Left ventricular (LV) end-systolic volume (EDV) and end-diastolic volume (ESV), and ejection fraction (EF) were measured by biplane Simpson's method. 15 LV mass was calculated from parasternal view based on Devereux's formula. LV diastolic function was estimated by the early transmitral flow velocity (E) to the late transmitral flow velocity (A) (E/A) ratio, and the transmitral E to the early diastolic medial LV septal tissue velocity (e′) (E/e′) ratio.

Right atrial (RA) and RV size were determined from the apical 4-chamber view. Tricuspid annular plane systolic excursion (TAPSE) was measured as the systolic displacement of the tricuspid lateral annulus, recorded on M-mode. RV end-diastolic and end-systolic area were obtained from the apical 4-chamber view. RV fractional area change (FAC) was calculated as (RV end-diastolic area -RV end-systolic area)/end-diastolic area×100%. Tricuspid lateral annular systolic velocity (S') was assessed by tissue Doppler imaging from the apical 4-chamber view. Estimation of tricuspid regurgitation (TR) incorporated color-Doppler imaging and contour of the jet in continuous-wave Doppler. Moderate to severe TR were defined as moderate, moderate to severe, or severe TR. Pulmonary artery systolic pressure (PASP) was assessed from the peak velocity of the TR jet, using the modified Bernoulli equation plus RA pressure evaluated from the inferior vena cava (IVC) size and its collapsibility.

Echocardiography and European Association of Cardiovascular Imaging. 6 All of the images were analyzed by 2D strain software (Image-Arena; TomTec Imaging Systems, Germany) in the apical four-chamber view at frame rate of 50~70 frames/sec ( Figure 1 ). After tracing the RV endocardial border, the region of interest was automatically generated. And then manual corrections were performed to fit RV myocardial wall thickness. RV free wall was automatically divided into three segments-basal, mid and apical segments. RVLS was calculated as the mean of the strain values in the three segments of RV free wall. If it was not feasible to track one or more segments, the case was excluded. We took the absolute value for a simpler interpretation, as RVLS is a negative value.

Intraobserver and interobserver variability of RVLS were estimated in 20 randomly selected subjects and evaluated by intra-class correlation coefficient (ICC) and Bland-Altman analysis.

Intraobserver variability was assessed by having one observer remeasure after two weeks.

Interobserver variability was evaluated by a second observer who was blinded to the first observer's measurements.

Continuous numeric variables were expressed as mean ± SD or medians (IQR), and compared with a two-sample t test and one-way analysis of variance (for normally distributed data), or Mann-Whitney test and Kruskal-Wallis test (for non-normal distribution of data).

Categorical variables were expressed as frequency number (%), and compared using χ 2 test or Fisher's exact test. To determine the optimal cutoff value (maximum Youden index) of prognostic RV function parameters for detecting increased mortality, receiver operator characteristic (ROC) curves were used. Survival curves were obtained by the Kaplan-Meier analysis and compared using log-rank test. Estimations of the predictor of mortality were performed using univariate and multivariate Cox regression models. All potential predictors of higher mortality were entered into univariate analyses, including sex, age, cardiac injury and inflammatory marker, LVEF, PASP, RV function echocardiographic parameters, acute respiratory distress syndrome (ARDS), and comorbidities (diabetes mellitus, hypertension, cardiovascular artery disease, malignancy and arrhythmia). Variables with P < 0.05 at univariate analysis entered into multivariate Cox regression models. 

Clinical characteristics of patients with COVID-19 according to tertiles of RVLS were shown in Table 1 .The mean age of COVID-19 patients were 61±14 years, 57(48%) patients were men. Chronic obstructive pulmonary disease was found in 6 patients with COVID-19. None of the patients had the history of pulmonary embolism. No patient was diagnosed with pulmonary embolism. Compared with patients in the highest tertiles, those in the lowest RVLS tertile were more likely to have a higher heart rate, D-dimer and C-reactive protein, high-flow oxygen and invasive mechanical ventilation therapy, higher incidence of acute heart injury, ARDS and deep vein thrombosis, and higher mortality. There were no significant differences in age, gender, systemic arterial pressure, comorbidities, lymphocyte count, CK-MB, hs-TNI, BNP, partial pressure of oxygen: fraction of inspiration oxygen, procalcitonin, and antiviral, antibiotic, glucocorticoid and angiotensin-converting enzyme inhibitors (ACE-I)/ angiotonin II receptor blockers(ARB) use, the number of intensive care unit admission, and acute kidney injury between the tertiles.

Echocardiographic characteristics of patients with COVID-19 according to tertiles of RVLS were described in Tables 2 and 3 . Compared with patients in the highest RVLS tertiles, those in the lowest tertile had similar left atrial and LV size, LV mass, E/A, E/e', LVEDV, LVESV, and LVEF. Distribution of RVLS in patients with COVID-19 was presented in Figure 2 .

Patients in the lowest RVLS tertile exhibited dilated right atrium, lower RVLS, RVFAC and TAPSE, more moderate-severe TR, and higher PASP. While RV, pulmonary artery and IVC dimension or S' did not differ between the tertiles. RVLS was lower in ARDS than non-ARDS patients (21.3±4.6% VS 24.6±4.4%, P < 0.001). Among the patients, when separated by tertiles, mortality was highest in patients with a RVLS ≤ 20.5%, followed by RVLS in the range between 20.6% and 25.4%, and lowest among patients with RVLS ≥ 25.5% (P < 0.001) (Central Illustration). When stratified by cutoff values, a RVLS lower than 23% was associated with higher mortality (P < 0.001) (Central Illustration). It also clearly revealed that survival significantly declined with worsening TAPSE and RVFAC ( Figure 4A and 4B). (Table 4) 

The intraobserver and interobserver reproducibility of RVLS was excellent, as reflected by high ICC (intraobserver: 0.95; interobserver: 0.91). Bland-Altman analysis demonstrated good intraobserver and interobserver agreement, with small bias (intraobserver:-0.33; interobserver: 0.70) and narrow limits of agreement (intraobserver:-2.19～1.54; interobserver: -3.08～4.49).

To our knowledge, this is the first study to comprehensively evaluate the prognostic value of RV function using the conventional echocardiography and 2D-STE in patients with COVID-19. Patients with the greatest degree of RV strain impairment were more likely to have a higher heart rate, high-flow oxygen and invasive mechanical ventilation therapy, higher incidence of acute heart injury, ARDS and deep vein thrombosis, and higher mortality.

Compared with survivors, non-survivors presented enlarged right-heart chamber, diminished RV function, and elevated PASP. More importantly, RVLS was able to predict higher risk of mortality in COVID-19 patients, independently of, and incrementally to other echocardiographic parameters. Therefore, comprehensive assessment of RV function using 2D-STE may be essential for risk stratification in COVID-19 patients.

It is significant to recognize patients with COVID-19 at higher risk for poor outcomes who might benefit from vigilant monitoring. Several risk factors of poor prognosis have been identified in SARS-CoV-2 infection. 16, 17 The role of the previously described prognostic markers, in particular ARDS and male sex, was confirmed in the present study. Furthermore, our study revealed important additional prognostic value of RV dysfunction. Most notably, the additional prognostic value of RVLS was substantial independent of LV systolic functional index, which failed to predict mortality in patients with COVID-19. This is consistent with a previous study of LV performance in severe acute respiratory syndrome (SARS), showing that only subclinical LV diastolic impairment was observed in patients with SARS. 18 Nevertheless, the prognostic value of RV function was not yet explored in the study of SARS.

RV dysfunction is related to significant morbidity and mortality in a variety of cardiovascular diseases. [9] [10] [19] [20] Indeed, in the present study, non-survivors displayed RV dilation and dysfunction. It has been reported that SARS-CoV-2 infection could cause both pulmonary and systemic inflammations, which may contribute to RV failure through RV overload and direct damage to cardiomyocyte. 17, 21 Similarly, in 42 patients with moderate-to-severe ARDS, Lazzeri et al demonstrated that troponin release can be related to RV dysfunction, thus emphasizing the clinical role of RV function. 22 Moreover, although sharing considerable similarities, myocardial injury and cardiac insufficiency were more frequently reported in patients with COVID-19 than in those with SARS. 17, 23 Therefore, assessment of RV function and recognition its prognostic significance is necessary in patients with COVID-19. RV dysfunction is not only a sign of increased pulmonary pressures, but also directly contribute to cardiac insufficiency.

Although cardiac magnetic resonance (CMR) imaging remains the gold standard for quantifying RV function, 24 in our study, the highly contagious nature of Covid-19 and patients' inability to hold breath for a short time limit the application of CMR.

Echocardiography is more widely used in daily practice. In clinical practice, it is routine to evaluate RV function using conventional echocardiogram parameters recommended by guidelines, 14 

Our data demonstrated that RVLS was a powerful and independent predictor of higher mortality, providing additive predictive value over other echocardiographic parameters in patients with COVID-19. Accordingly, the present study revealed the important clinical implication of RVLS, as it can be easily obtained from bedside echocardiography. This suggests that evaluation of RV function by conventional echocardiography measurements (ie, FAC or TAPSE) need to be complemented by LS analysis to identify patients at higher risk of poor outcome.

Although our study was a cohort of homogenous and consecutive patients who were hospitalized with COVID-19, it is limited by being a single-center study with a relatively limited sample size. Our hospital is one of hardest-hit hospitals by the COVID-19 in Wuhan, the patients we included may not represent the population in other areas. Moreover, we enrolled patients who were hospitalized with COVID-19, and the asymptomatic patients who had not been admitted to hospital were not included. Considering the wide clinical spectrum of SARS-CoV-2 infection, our findings may not be applicable to the entire COVID-19 population. Therefore, future studies with multi-center and a larger sample size studies are needed to determine the prognosis value of RVLS in COVID-19 patients. Beside, we excluded 24 patients owing to poor image quality precluding strain analysis, which limits the generalizability of our results. And the results pertain only to the software used in our study and may not apply to other software algorithms because the 2D-STE parameters are hampered by the intervendor variability. we used RV free-wall LS instead of global RV strain because a simultaneous echocardiographic-catheterization study demonstrated that free-wall LS might better reflect RV function than global RV strain. 27 Moreover, the relationship between LV and RV function could not be explored in the present study, as RV function could be affected by subclinical LV dysfunction. Furthermore, obesity patients were uncommon in our study, as obesity is preliminarily considered as a high risk factor.

Our study demonstrates that RVLS is an independent determinant of outcomes in patients with COVID-19. Importantly, this index may have an additional predictive value over other echocardiographic parameters. Therefore, evaluation of RV function should be implemented by investigation of RVLS for risk stratification in COVID-19 patients.

longitudinal strain is a powerful and independent predictor of higher mortality, providing additive predictive value over other echocardiographic parameters in patients with COVID-19. Our study demonstrated comprehensive assessment of RV function by 2D-STE may be essential for risk stratification in COVID-19 patients.

Future studies are needed to verify the value of 2D-STE using different software in multicenter design for risk stratification of COVID-19 patients. 

Cardiovascular implications of fatal outcomes of patients with coronavirus disease 2019 (COVID-19)

Association of cardiac injury with mortality in hospitalized patients with COVID-19 in Wuhan, China

Clinical features of patients with 2019 novel coronavirus in Wuhan, China

Clinical course and outcomes of critically ill patients with

China: a single-centered, retrospective, observational study

Prognostic value of right ventricular dysfunction in heart failure with reduced ejection fraction: superiority of longitudinal strain over tricuspid annular plane systolic excursion

Current and evolving echocardiographic

Role of two-dimensional speckle-tracking echocardiography strain in the assessment of right ventricular systolic function and comparison with conventional parameters

Impaired right and left ventricular function in asymptomatic children with repaired tetralogy of Fallot by two-dimensional speckle tracking echocardiography study

The effect of right ventricular myocardial remodeling on ventricular function as assessed by two-dimensional speckle tracking echocardiography in patients with tetralogy of Fallot: A single center experience from China

Impaired RV global longitudinal strain is associated with poor long-term clinical outcomes in patients with acute inferior STEMI

Clinical management of severe acute respiratory infection when novel coronavirus (nCoV) infection is suspected

World Medical Association Declaration of Helsinki: ethical principles for medical research involving human subjects

Guidelines for the echocardiographic assessment of the right heart in adults: a report from the American Society of Echocardiography endorsed by the European Association of Echocardiography and the Canadian Society of Echocardiography

Recommendations for cardiac chamber quantification by echocardiography in adults: an update from the American Society of Echocardiography and the European Association of Cardiovascular Imaging

Epidemiological and clinical characteristics of 99 cases of 2019 novel coronavirus pneumonia in Wuhan, China: a descriptive study

Clinical characteristics of 113 deceased patients with coronavirus disease 2019: retrospective study

Left Ventricular Performance in Patients with Severe Acute Respiratory Syndrome: a 30-Day Echocardiographic Follow-Up Study

Incremental prognostic utility of left ventricular

Additional prognostic value of 2D right ventricular speckle-tracking strain for prediction of survival in heart failure and reduced ejection fraction: A comparative study with cardiac magnetic resonance

Prevalence and impact of cardiovascular metabolic diseases on COVID-19 in China

Serial Measurements of Troponin and Echocardiography in Patients with Moderate-to-Severe Acute Respiratory Distress Syndrome

Clinical and laboratory features of severe acute respiratory syndrome vis-a-vis onset of fever

Cardiac magnetic resonance normal reference values of biventricular size and function in male athlete's heart

The prognostic value of right ventricular deformation imaging in early arrhythmogenic right ventricular cardiomyopathy

Association of right ventricular longitudinal strain with mortality in patients undergoing transcatheter aortic valve replacement

Right ventricular longitudinal strain and right ventricular stroke work index in patients with severe heart failure: left ventricular assist device suitability for transplant candidates

FAC=fractional area change; IQR=interquartile range; IVC=inferior vena cava; LA=left atrial; LV=left ventricular; LVEDV=left ventricular end diastolic volume; LVESV=left ventricular end systolic volume; LVEF=left ventricular ejection fraction; RA=right atrial; RV=right ventricular; RVLS=right ventricular longitudinal strain; PA=pulmonary artery; PASP=pulmonary artery systolic pressure; TAPSE=tricuspid annular plane systolic excursion

AIC=Akaike Information Criterion; ACE-I=angiotensin-converting enzyme inhibitors

ARB=angiotonin II receptor blockers; ARDS=acute respiratory distress syndrome; CI=confidence interval; CK-MB=creatine kinase muscle-brain

HR=hazard ratio; hs-TNI=hypersensitive troponin I; LVEF=left ventricular ejection fraction; TAPSE=tricuspid annular plane systolic excursion; TR=tricuspid regurgitation