Mitral Annular Plane Systolic Excursion: An Early Marker of Mortality in Severe COVID-19.

Respiratory failure is a major cause of mortality among hospitalized patients with coronavirus disease 2019 (COVID-19). Previous studies have shown that right ventricular (RV) dilation and reduced RV longitudinal strain are markers of poor outcome in this disease. , COVID-19 can cause direct myocardial injury resulting in left ventricular (LV) systolic dysfunction and heart failure, suggesting that assessment of LV function might also have prognostic value. Reduction of longitudinal systolic function assessed by mitral annular plane systolic excursion (MAPSE) is an early indicator of myocardial disease in various cardiac disorders. In this retrospective study, we investigated the prognostic value of MAPSE in patients admitted with respiratory failure related to COVID-19.

One hundred nine patients were admitted to the intensive care unit at Indiana University Health hospitals with laboratory-confirmed severe COVID-19, defined as oxygen saturation < 94%, respiratory rate ≥ 30 breaths/min, and ratio of partial pressure of oxygen to fraction of inspired oxygen ≤ 300 mm Hg or need for mechanical ventilation. Of these, 68 patients who underwent transthoracic echocardiography on the basis of clinical indications were included in the final analysis. Echocardiograms were obtained using GE S70 and E95 (GE Healthcare, Milwaukee, WI) and Philips EPIQ (Philips Medical Systems, Andover, MA) ultrasound systems. Images were acquired according to a limited COVID-19 protocol, and measurements were performed per American Society of Echocardiography guidelines. LV ejection fraction was measured using the biplane Simpson method. MAPSE was calculated as the average of septal and lateral mitral annular excursion measured in the apical four-chamber view. Thirty-five of the 68 studies (52%) had suboptimal image quality.

Transthoracic echocardiography was performed a median of 1 day (interquartile range, 0–3 days) after hospital admission. Sixty-six patients (97%) required mechanical ventilation during hospitalization. Of these, 52 (79%) were on ventilators at the time of echocardiography. Twenty-two patients (32%) died during hospitalization. Chi-square and Mann-Whitney U tests were used to compare clinical characteristics between survivors and those who died (Table 1 ). P values < .05 were considered to indicate statistical significance. Increased age (P = .018) and shock requiring vasopressors (P = .033) were the only variables associated with mortality. The comparison of echocardiographic variables is shown in Table 2 . Patients who died had significantly increased right atrial (RA) indexed volume (P = .010) and lower MAPSE (P = .006). There were trends toward lower LV ejection fraction and tricuspid annular plane systolic excursion in those who died. Among laboratory parameters, serum creatinine was significantly higher in those who died (P = .032; Supplemental Table 1). Multivariate analysis was conducted using binary logistic regression incorporating variables significant on bivariate analysis. Shock requiring vasopressors (hazard ratio, 9.2; 95% CI, 1.4–59.3), MAPSE (hazard ratio, 1.6; 95% CI, 1.2–2.2), and RA indexed volume (hazard ratio, 1.1; 95% CI, 1.0–1.3) were found to be independently associated with mortality. MAPSE yielded an area under the curve of 0.72, with an optimal threshold of 1.0 cm by receiver operating characteristic curve analysis. The area under the curve for RA indexed volume was 0.61, with an optimal cutoff valve of 22.6 cm3. When used in combination, RA indexed volume and MAPSE enabled the identification of high and lower risk groups. Mortality was 77% in the presence of either reduced MAPSE (<1.0 cm) or increased RA volume index (>22.6 cm3). Survival was 84% in those who had both preserved MAPSE and no significant RA enlargement.

Table 1

Baseline demographics, medical comorbidities, and clinical course of the overall cohort and each study group

Variables	All patients	Alive	Dead	P
	(n = 68)	(n = 46 [68%])	(n = 22 [32%])
Demographic information
Age, y	64 ± 14 (31–87)	61 ± 15 (31–86)	70 ± 10 (47–87)	.018∗
Body mass index, kg/m2	34.9 ± 10.9	33.3 ± 10.3	38.2 ± 11.0	.075
Sex				.503
Male	41 (60)	29 (63)	12 (55)
Female	27 (40)	17 (37)	10 (45)
Race				.944
Black	36 (53)	24 (52)	12 (55)
White	28 (41)	19 (41)	9 (41)
Other	4 (6)	3 (7)	1 (5)
Hypertension	47 (69)	31 (67)	16 (73)	.656
Dyslipidemia	41 (60)	26 (57)	15 (68)	.358
Diabetes mellitus	26 (38)	18 (39)	8 (36)	.826
Tobacco use	22 (32)	16 (35)	6 (27)	.536
Coronary artery disease	18 (26)	12 (26)	6 (27)	.917
Congestive heart failure	13 (19)	10 (22)	3 (14)	.427
Chronic kidney disease	11 (16)	9 (20)	2 (9)	.272
COPD	12 (18)	8 (17)	4 (18)	.936
Asthma	10 (15)	5 (11)	5 (23)	.196
Clinical information
Organ failure
Mechanical ventilation	66 (97)	45 (98)	21 (95)	.588
Shock requiring vasopressor	47 (69)	28 (61)	19 (86)	.033∗
Renal replacement therapy	15 (22)	8 (17)	7 (32)	.180
ICU days	13 ± 8	14 ± 8	10 ± 6	—
Hospital days	18 ± 9	21 ± 8	11 ± 5	—

COPD, Chronic obstructive pulmonary disease; ICU, intensive care unit.

Data are expressed as mean ± SD (range) or as number (percentage).

Statistically significant (P < .05).

Table 2

Echocardiographic parameters of the entire cohort and comparison between patients surviving and those who died

Echocardiogram parameter	Cohort	Alive	Dead	P
	(n = 68)	(n = 46 [68%])	(n = 22 [32%])
LV diastolic volume (mL)	103 ± 34	107 ± 38	95 ± 24	.186
LV systolic volume (mL)	48 ± 24	48 ± 24	49 ± 24	.908
LV ejection fraction (%)	55 ± 14	56 ± 12	50 ± 16	.089
LVOT VTI (cm)	18 ± 4	18 ± 4	17 ± 5	.263
IVS thickness (cm)	1.1 ± 0.2	1.1 ± 0.2	1.1 ± 0.2	.886
LVPW thickness (cm)	1.0 ± 0.2	1.0 ± 0.2	1.1 ± 0.2	.719
LV diastolic diameter (cm)	4.6 ± 0.8	4.7 ± 0.8	4.6 ± 0.8	.699
LV systolic diameter (cm)	3.2 ± 0.8	3.2 ± 0.8	3.3 ± 1.0	.993
LV basal fractional shortening	0.30 ± 0.11	0.31 ± 0.10	0.28 ± 0.14	.407
LA volume (indexed to BSA; mL/m2)	21.7 ± 9.7	21.4 ± 7.7	22.2 ± 12.9	.767
LA AP diameter (cm)	3.6 ± 0.7	3.7 ± 0.7	3.4 ± 0.7	.057
RV basal diameter (cm)	3.9 ± 0.8	3.8 ± 0.7	4.0 ± 0.9	.484
TAPSE (cm)	2.0 ± 0.4	2.1 ± 0.4	1.9 ± 0.4	.084
RV diastolic area (indexed to BSA; cm2/m2)	10.2 ± 3.1	10.1 ± 2.8	10.6 ± 3.6	.566
RV systolic area (indexed to BSA; cm2/m2)	6.4 ± 2.5	6.2 ± 1.9	7.0 ± 3.2	.260
RV FAC (%)	37.7 ± 9.2	38.8 ± 8.8	35.8 ± 9.9	.268
RA volume (indexed to BSA; mL/m2)	19.6 ± 10.9	17.0 ± 7.0	24.3 ± 14.8	.010∗
RVOT VTI (cm)	14.3 ± 4.1	14.4 ± 3.9	14.0 ± 4.8	.778
E/e′ lateral	8.9 ± 5.0	9.3 ± 5.5	8.1 ± 3.5	.395
E/e′ septal	10.2 ± 3.7	10.2 ± 3.3	10.3 ± 4.4	.951
E/e′ (average; cm/sec)	9.7 ± 4.3	10.0 ± 4.6	9.1 ± 3.8	.441
TR jet velocity (cm/sec)	255 ± 52	256 ± 49	254 ± 56	.933
RAP (mm Hg)	8.7 ± 5.4	7.8 ± 5.3	10.3 ± 5.2	.103
RVSP (mm Hg)	34.0 ± 12.0	31.3 ± 13.2	37.1 ± 9.9	.213
PVR (WU)	2.2 ± 0.9	2.1 ± 0.9	2.5 ± 0.8	.274
MAPSE (cm)
Lateral	1.3 ± 0.3	1.3 ± 0.3	1.1 ± 0.3	.004∗
Septal	1.0 ± 0.3	1.1 ± 0.2	0.9 ± 0.2	.008∗
Averaged	1.1 ± 0.3	1.2 ± 0.3	1.0 ± 0.3	.006∗

AP, Anteroposterior; BSA, body surface area; FAC, fractional area change; IVS, interventricular septal; LA, left atrial; LVOT, LV outflow tract; LVPW, LV posterior wall; PVR, pulmonary vascular resistance; RAP, RA pressure; RVOT, RV outflow tract; RVSP, RV systolic pressure; TAPSE, tricuspid annular plane systolic excursion; TR, tricuspid regurgitation; VTI, velocity-time integral; WU, Wood units.

Data are expressed as mean ± SD.

Statistically significant (P < .05).

MAPSE emerged as the only left heart parameter independently associated with increased mortality. Other parameters of left heart size and function were not significantly different between groups, although there was a trend toward lower ejection fraction in those who died. This suggests that MAPSE may enhance the early assessment of LV myocardial injury in patients with severe COVID-19. MAPSE is a validated measure of LV longitudinal function that likely correlates reasonably with longitudinal strain. Thus, in critically ill patients in whom image quality is insufficient for accurate assessment of longitudinal strain, MAPSE may be a reasonable alternative. Additionally, RA enlargement was independently associated with mortality. In the setting of respiratory failure from COVID-19, RA enlargement may result from the combined effect of increasing pulmonary pressures, RV systolic dysfunction, positive pressure ventilation, and fluid resuscitation. It is therefore not surprising that the development of RA enlargement early in the hospital stay is a marker of later mortality. Although our study did not confirm the association between mortality and RV dilation or dysfunction, there was a trend toward lower tricuspid annular plane systolic excursion in patients who died.

There has been significant effort to identify early predictors of mortality from COVID-19, including using modalities such as point-of-care ultrasound. , In our study population of critically ill patients with COVID-19, the presence of either reduced MAPSE or RA enlargement was associated with high risk for in-hospital mortality, while those who had preserved MAPSE and no RA enlargement had a high rate of in-hospital survival. This suggests that in combination, MAPSE and RA size can successfully risk-stratify patients with severe COVID-19 early in the hospital course and may enable early institution of aggressive therapies for high-risk patients.

The retrospective nature, small sample size, and possible bias related to the selective use of echocardiography are limitations of our investigation. Our findings suggest that the assessment of longitudinal systolic function of the left heart using MAPSE in addition to evaluation of the right heart may yield accurate information for risk stratification of patients with severe COVID-19 early in the hospital course. Larger, prospective studies are needed to further define parameters that predict outcomes in patients with this disease.