a b s t r a c t

Objectives: Detecting reduced left ventricular ejection fraction (LVEF) by an emergency physician is an im- portant skill. The subjective ultrasound assessment of LVEF by EPs correlates with comprehensive echocardio- gram (CE) results. Mitral annular plane systolic excursion (MAPSE) is an ultrasound measure of vertical movement of the mitral annulus, which correlates to LVEF in the cardiology literature, but has not been studied when measured by an EP. Our objective is to determine whether EP measured MAPSE can accurately predict LVEF

<50% on CE.

Methods: This is a prospective observational single center study using a convenience sample to evaluate the use of a Focused cardiac ultrasound (FOCUS) for patients with possible decompensated heart failure. The FOCUS in- cluded standard cardiac views to estimate LVEF, MAPSE, and E-point septal separation (EPSS). Abnormal MAPSE was defined as <8 mm and abnormal EPSS as >10 mm. The primary outcome assessed was the ability of an abnormal MAPSE to predict an LVEF <50% on CE. MAPSE also was compared to EP estimated LVEF and EPSS. Inter-rater reliability was determined by two investigators performing independent blinded review.

Results: We enrolled 61 subjects, 24 (39%) had an LVEF <50% on a CE. MAPSE <8 mm had a 42% sensitivity (95% CI 22-63), 89% specificity (95% CI 75-97), and accuracy of 71% for detecting LVEF <50%. MAPSE demonstrated lower sensitivity than EPSS (79% sensitivity [95% CI 58-93], and 76% specificity [95% CI 59-88]) and higher spec- ificity than estimated LVEF (100% sensitivity [95% CI 86-100], 59% specificity [95% CI 42-75]). PPV and NPV

for MAPSE was 71% (95% CI 47-88) and 70% (95% CI 62-77) respectively. The ROC for MAPSE <8 mm is 0.79 (95% CI 0.68-0.9). MAPSE measurement interrater reliability was 96%.

Conclusions: In this exploratory study evaluating MAPSE measurements by EPs, we found the measurement was easy to perform with excellent agreement across users with minimal training. A MAPSE value <8 mm had mod- erate predictive value for LVEF <50% on CE and was more specific for reduced LVEF than qualitative assessment. MAPSE had high specificity for LVEF <50%. Further studies are needed to validate these results on a larger scale.

(C) 2023

1. Introduction

An estimated 6.2 million American adults have heart failure, and this population accounts for approximately 800,000 hospital admissions per year. Of those admissions, 55% of patients have a reduced left ventricular ejection fraction (LVEF) [1]. Reliable estimation of LVEF is a critical skill for emergency physicians (EPs) that is used to help manage hemody- namically unstable patients as well as to improve the diagnostic accu- racy for acute decompensated heart failure (ADHF). Often in clinical practice, focused cardiac ultrasound interpretation is com- bined with other information such as physical examination, chest

* Corresponding author.

E-mail address: [email protected] (A.L. Schick).

radiography, laboratory information, and electrocardiograms to help di- agnose acute coronary syndrome or ADHF [2]. Traditionally, EPs per- form this assessment by visually estimating the LVEF on standard cardiac ultrasound views and categorizing it as hyperdynamic (>65%), normal (50-65%), reduced (30-49%), or severely reduced (<30%) [3,4]. This global assessment of LVEF performs fairly well independently, but the range of agreement with comprehensive echocardiogram LVEF measurements varies between 70 and 88% [3-5]. Because of this imper- fect agreement, additional bedside ultrasound assessments of left ven- tricular performance could improve diagnostic accuracy for ADHF in undifferentiated emergency department (ED) patients.

A comprehensive echocardiogram (CE), which is not often indicated or available in the ED setting, includes many other more quantitative ultrasonographic measurements to assess left ventricular function.

https://doi.org/10.1016/j.ajem.2023.03.018

0735-6757/(C) 2023

A quantitative measurement that is sometimes obtained at the bedside in the ED is the E-point septal separation (EPSS) which evaluates the movement of the anterior MItral valve leaflet relative to the cardiac sep- tum during diastole. Increased separation between the leaflet and the in- traventricular septum when the valve is open can suggest left ventricular systolic dysfunction in the absence of valvular disease. A measurement of

>10 mm is considered abnormal and correlates with an LVEF <50%; an EPSS that is <7 mm correlates to a normal ejection fraction [6,7]. Several studies have demonstrated that EPs can reliably and accurately obtain more complex cardiac measurements such as EPSS with FOCUS that could improve diagnostic accuracy of ADHF [8-11]. Mitral valve stenosis, mitral valve repair or replacement, mitral valve tethering, and aortic re- gurgitation can affect the movement of the anterior mitral valve leaflet and result in EPSS values that do not correlate as well with LVEF [8,9].

Mitral annular plane systolic excursion (MAPSE) is a cardiac mea- surement that has yet to be studied as performed by an EP at the bed- side. MAPSE evaluates the vertical motion of the mitral valve during the cardiac cycle, which is assessed using the motion mode (M-mode) over the lateral aspect of the valve annulus. In the cardiology literature, MAPSE has a normal range of 12-15 mm [12-15]. Decreased excursion of the mitral valve during systole correlates to left ventricular systolic dysfunction, and it can precede changes in LVEF during myocardial in- farction. MAPSE values can be affected by localized wall motion abnor- malities affecting the myocardium near the mitral valve, severe mitral annular calcification, and mitral valve prosthesis [16,17]. A decreased MAPSE correlates to elevation in pro-brain natriuretic peptide levels and can be seen in patients with isolated diastolic dysfunction [12,16,18-20]. Prior studies in cardiology literature suggest that a MAPSE <8 mm is associated with an LVEF <50% with a sensitivity of 98% and specificity of 82% [13]. MAPSE below 7 mm in those with di- lated cardiomyopathies has a sensitivity of 92% and specificity of 67% for Severe systolic dysfunction (LVEF <30%) [21]. A MAPSE >10 mm has a sensitivity of 90% and specificity of 87% for a preserved systolic function (LVEF >55%) [21,22].

MAPSE was shown to have high reproducibility and accuracy in de-

termining LVEF even when performed by novice echocardiographers or with poor Image quality [17,23,24]. High inter-rater reliability is another advantage of MAPSE, as it is less reliant on endocardial border resolution than standard LVEF estimation techniques and requires a simple one- dimensional measurement [17,23,24]. While MAPSE performed by EPs has not yet been studied, several studies have shown EPs can easily col- lect a similar measurement called tricuspid annular plane systolic ex- cursion (TAPSE) to look for right ventricular dysfunction [25-27].

The primary objective of this study to determine if MAPSE as mea- sured at the bedside by EPs performing focused ultrasonography can re- liably predict decreased LVEF as measured on a CE. We hypothesized that a MAPSE value of less than or equal to 8 mm would be >90% spe- cific for LVEF of <50%. Additionally, the interrater reliability and test performance of MAPSE will be compared to the performance of EPSS measurement and EP estimated LVEF for the detection of LVEF <50% on a CE in the same population.

1. Methods
   1. Study design

This was a single center prospective observational cohort sample using a convenience sample of ED patients with symptoms concerning for heart failure. Patients were enrolled from December 2018 until June 2021. The hospital’s institutional review board approved this study and all patients provided written informed consent prior to participation.

• 1. Study setting and population

All subjects were enrolled in an urban academic emergency depart- ment with approximately 120,000 annual visits. Enrollment occurred

when a trained study investigator was available to perform the FOCUS exam. Subjects were eligible for enrollment if the treating EP was con- cerned about ADHF as an explanation for their presenting symptoms (i.e. shortness of breath, Leg swelling, hypotension, Chest tightness). Study investigators identified potential subjects by scanning through chief complaints in the ED electronic medical record and then asking the primary emergency care provider if the patient was being evaluated for potential ADHF. Patients excluded from enrollment included those under 18, wards of the state, prisoners, and non-English speaking patients.

• 1. Study protocol

Study investigators included one ultrasound fellowship-trained at- tending physician, two third-year emergency medicine residents, one fourth-year emergency medicine resident, and one emergency ultra- sound fellow. All investigators received a 1-h hands-on training to con- firm that images were being obtained in a standardized fashion. At least two supervised FOCUS with MAPSE measurement were performed with the ultrasound fellowship trained attending EP prior to enrolling study subjects. The investigators obtained written informed consent prior to enrollment after a potential subject was identified and screened for eli- gibility. All ultrasounds were obtained with Sonosite Edge or GE Venue R2 machine using the phased array probe (1.1-4.7 MHz frequency).

The study investigator obtained and interpreted the ultrasound im- ages in real time from the FOCUS exam consisting of four cardiac video clips (parasternal long, parasternal short, apical four chamber, and subxiphoid windows), a measurement of MAPSE (Fig. 1), and a measurement of EPSS. Patient position was at the discretion of the study investigator; patients were in the supine position if able, semi- recumbent position if unable, and Left lateral decubitus position if needed. M-mode was used at bedside to measure MAPSE in the apical four chamber view and EPSS in the parasternal long view.

Fig. 1. To measure mitral annual plane systolic excursion (MAPSE) an M-mode cursor is placed over the lateral aspect of the mitral valve annulus in an apical four-chamber view. This will create a wave form with the valve motion during a cardiac cycle; the dis- tance travelled during systole can be measured using calipers placed at the highest and lowest point of the tracing.

MAPSE measurements were obtained by placing the M mode line at the lateral aspect of the mitral valve to measure the vertical excursion dur- ing systole. The M mode image was frozen and calipers were placed at the highest and lowest point of the tracing to obtain the value. The first high quality measurement obtained was recorded. Prior to obtaining EPSS and MAPSE measurements, the investigator denoted their estimated LVEF after obtaining the four cardiac video clips and cat- egorized it as one of the following: hyperdynamic (EF >65%), normal (EF 50-65%), reduced (30-49%), and severely reduced (<30%). An ab- normal MAPSE was defined as values <8 mm, and an abnormal EPSS was a value >10 mm. Patient vital signs and demographics (age, sex, ethnicity, weight) were also recorded at the time of FOCUS exam. MAPSE measurements were not shared with the ED treatment team. If the study investigator identified a finding on ultrasound that could be clinically significant (i.e. a pericardial effusion), this finding was shared with the team.

Chart abstraction was performed by trained study investigators blinded to the ultrasound results to obtain the following information: patient disposition from emergency department (intensive care unit, step down unit, floor unit, home, or deceased), CE LVEF, troponin level, and Brain natriuretic peptide level. Subjects were included in the analysis if they had a CE performed within 12 months before or 3 months after study enrollment. Any CE performed during the hospital admission following study enrollment was preferred to a CE performed prior to the study enrollment date. Interrater reliability of the estimated LVEF assessment, MAPSE measurement, and EPSS measurement was calculated for all subjects by two investigators (one being the ultra- sound fellowship trained EP); the identity of the study enroller was blinded when images were being reviewed. All images had a QA review by the ultrasound fellowship trained attending prior to inclusion in the study analysis; all images obtained met the QA requirements for inclu- sion. If there was a discrepancy between the two investigators, then the images were reviewed together and discussed until a consensus was reached.

• 1. Outcome measures

The primary outcome is the test performance of MAPSE <8 mm for detecting reduced LVEF (<50%) on a CE in patients presenting to the ED with symptoms concerning for ADHF. Secondary outcomes include the interrater reliability of MAPSE, the correlation of MAPSE levels to CE measured LVEF, the Predictive performance of MAPSE levels

>10 mm for normal LVEF, and the predictive performance of MAPSE

<8 mm compared to EPSS at two different cut points (>10 mm and >7 mm) and FOCUS estimated reduced LVEF. Other analyses per- formed evaluated the relationship between MAPSE and biomarkers (BNP and troponin).

• 1. Statistical analysis

All data analysis was performed using statistical software (SAS, ver- sion 9.2, Carey, NC). Descriptive statistics were used to report patient and demographic data. Cohen’s Kappa statistic (K) with 95% confidence interval (CI) was calculated to assess the agreement between the two investigators reviewing study ultrasound images. Receiver operator characteristic (ROC) curve analysis employing logistic regression models with Kappa values were conducted for each measurement (i.e. LVEF estimation, MAPSE, and EPSS) to determine the Predictive ability of an abnormal result for identifying confirmed reduced LVEF on CE. The sensitivity, specificity, positive and negative predictive values, and diagnostic accuracy for these indicators alone and in combination with 95% confidence intervals (CI) are reported. An incremental validity as- sessment was performed by comparing the logistic regression models of the individual and combined test measurements. Logistic regression models were created to predict LVEF on CE with each measure indepen- dently (MAPSE, EPSS and LVEF), a combination of two measurements,

and a model exploring all three measurements combined. We deter- mined model fit through an assessment of the Akaike Information Crite- rion (AIC) value, where higher values were indicative of poorer fitting models. To assess optimal cut-point values for MAPSE and EPSS in predicting abnormal LVEF compared to the CE in this sample, the esti- mated Youden’s J Index was used. This statistic assesses the maximum efficiency of different values of a measure to optimize accuracy for quantitative indicators of disease status. The highest Youden’s Index value that is >0.50 corresponds to the optimal sensitivity and specificity combination and are used to identify cut point values [28]. As an explor- atory analysis, scatter plots with best line of fit (including variance ex- plained, R²) were created and show the direct relationship between MAPSE and EPSS with the comprehensive echocardiogram measured LVEF. A similar scatter plot with bivariate correlation was made to model MAPSE results with corresponding troponin I and BNP results. All p-values <.05 were considered to be statistically significant. Also, a box and whisker plot was created to identify the ranges of MAPSE that correspond with the LVEF ranges of hyperdynamic (EF >65%), normal (EF 50-65%), moderately reduced (EF 30-50%), or significantly reduced (EF <30%).

1. Results

A total of 64 subjects were enrolled. Three subjects were excluded from data analysis; one subject was excluded due to inability to obtain necessary ultrasound views, and two subjects were excluded as no CE was available for comparison. Data is reported on 61 subjects; there was no missing data. The ultrasound fellowship trained attending en- rolled 9 subjects, the ultrasound fellow enrolled 11 subjects, and the three resident physicians enrolled the remaining 44 subjects. Detailed patient demographics, presenting symptoms, emergency department disposition, and discharge diagnosis are summarized in Table 1. A total of 24 patients (39%) had a LVEF <50% on CE; of these patients, 17 had an estimated LVEF of 30-50% (moderately reduced) and 7 had an esti- mated LVEF of <30% (severely reduced). Thirty-nine patients (63%) were estimated to have an LVEF <50% on FOCUS. Twenty-five (40%) pa- tients were estimated to have moderately reduced LVEF function, 14 (23%) patients estimated to have severely reduced LVEF function, and one patient was categorized as hyperdynamic. Fourteen patients (22%) had a MAPSE <8 mm; the median MAPSE value was 10 mm (range 5-22, IQR 5). Thirty patients (49%) had an EPSS value >10 mm, and thirty-seven (61%) had an EPSS value >7 mm (median EPSS 9.2, range 0-30, IQR 10).

Inter-rater reliability for all cardiac measurements was high: MAPSE was 96.7% (? = 0.92, 95% CI 0.82-1.00), estimated LVEF EF was 96.7% (? 0.91, 95% CI 0.79-1.00), and EPSS was 93.4% (? 0.71, 95% CI 0.45-0.98).

The ROC for MAPSE was 0.79 (95% CI 0.68-0.90) for estimating abnor- mal LVEF on CE, which was non-inferior to the ROC for EPSS at 0.85 (95% CI 0.75-0.95) and EP estimated abnormal LVEF at 0.85 (95% CI 0.76-0.95) (Fig. 2). Table 2 shows the test characteristics of MAPSE and EPSS to detect abnormal LVEF CE. MAPSE <8 mm had a higher spec- ificity for abnormal LVEF on CE than EPSS or estimated LVEF, but lower sensitivity (Table 2). MAPSE had a sensitivity of 64% (95% CI 31-89), a

specificity of 86% (95% CI 73-94), and accuracy of 82% (95% CI 70-91) for detecting a severely reduced ejection fraction (LVEF <30%). MAPSE

>10 mm had a sensitivity of 62% (95% CI 45-78%) and a specificity of 92% (95% CI 73-99%) for predicting normal LVEF.

The test performance effects of combining of MAPSE, EPSS, and esti- mated LVEF are described in Table 3. The presence of at least two abnor- mal FOCUS findings (MAPSE, EPSS, and estimated LVEF) had the highest accuracy for the detection of abnormal LVEF on CE. The incremental va- lidity of the addition of EPSS and estimated LVEF to MAPSE was assessed using a model fit through the Akaike Information Criterion , where higher values of AIC were indicative of poorer model fits. MAPSE alone had an AIC of 69 for predicting an abnormal LVEF on CE, which

Table 1

Patient demographics and clinical characteristics.

improved when combined with EPSS (AIC of 57) and when combined with both EPSS and LVEF (AIC of 54).

	N (%) or Median (SD)	Figure 3 shows the how MAPSE and EPSS values correlate to mea- sured LVEF on CE. The median LVEF for patients with MAPSE <8 mm
Age	72.64 (17.6)	was 32.5% (range 15%-75%, IQR 23%) compared to 55% among patients
Male sex	32 (52.5%)	with 8 mm or more (range 15%-83%, IQR 23%). The median LVEF for pa-

Race

Caucasian 46 (75.4%)

African American 8 (13.1%)

Asian 1 (1.6%)

Other 6 (9.8%)

Prior Medical Conditions

Congestive Heart Failure 37 (60.7%)

COPD 22 (36.1%)

Myocardial Infarction 21 (34.4%)

Asthma 11 (18.0%)

Pulmonary HTN 11 (18.0%)

end stage renal disease 2 (3.3%)

Pregnancy (current) 1 (1.6%)

Interstitial Lung Disease 1 (1.6%)

Other (pericarditis, heart transplant) 2 (3.3%) Presenting Symptoms

Shortness of breath 46 (75.4%)

Chest pain 11 (18.0%)

lower extremity swelling/edema 7 (11.5%)

Generalized Weakness 4 (6.6%)

Cough 3 (4.9%)

Other 4 (6.6%)

Emergency Department Disposition

ICU 8 (13.1%)

Step Down 4 (6.6%)

Floor 43 (70.5%)

Discharge 5 (8.2%)

Observation Unit 1 (1.6%)

Discharge Diagnosis

Acute decompensated heart failure 38 (62.3%)

COPD or asthma exacerbation 8 (13.1%)

Bacterial or viral pneumonia 7 (11.5%)

Acute Coronary Syndrome 4 (6.6%)

Symptomatic valvular disease 3 (4.9%)

Non-cardiac chest pain 3 (4.9%)

Pulmonary HTN exacerbation 2 (3.2%)

tients with EPSS >10 mm was 35% (range 15%-75%, IQR 22.5%) com- pared to 59% among patients with 10 mm or less (range 34%-83%, IQR 10%). The mean value for MAPSE in patients with an estimated normal LVEF on bedside US was 13.1 mm (95% CI 5.5-20.8), compared to

10.3 mm (95% CI 3.8-16.8) with an estimated moderately decreased LVEF and 7.8 mm (95% CI 4.6-11.1) with an estimated severely reduced LVEF (Fig. 4).

Using the Youden index, different cut-off values for MAPSE and EPSS were tested to determine the optimal cut-off point for accurately detect- ing abnormal LVEF (Supplemental Table A). For MAPSE, the highest Youden index value suggested an optimal cut-point as less than or equal to 8 mm (91.7% sensitivity, 62.2% specificity). For EPSS, the highest Youden index value suggested an optimal cut-point as greater than or equal to 11 mm. The relationship between serum BNP levels and troponin I with measured MAPSE is shown in Supplemental Fig. 1A/1B. Lower MAPSE values correlated with higher troponin I and BNP results.

1. Discussion

This prospective study is the first to show that MAPSE measure- ments performed at the bedside by EPs can predict abnormal LVEF on CE. The measured MAPSE values did correlate with measured LVEF. MAPSE has a higher specificity and lower sensitivity for abnormal LVEF on CE function than visually estimated LVEF or EPSS >10 mm with a similar diagnostic accuracy. MAPSE measurements had a high inter-rater reliability and were collected without difficulty for all but one enrolled subject (difficult to obtain apical four-chamber cardiac ul-

trasound view). One strength of our study includes the robust review

Other (pre-eclampsia, anemia, pleural effusion, pyelonephritis, cardiogenic shock)

5 (8.2%)

process where all subject images were reviewed by two study investiga- tors after enrollment who were blinded to the study enroller. An abnor-

COPD: chronic obstructive pulmonary disease, HTN: hypertension, ICU: intensive care unit.

mal MAPSE was better correlated to abnormal LVEF in the patient population with severely reduced ejection fraction (LVEF <30%). Similar to EPSS, an advantage of MAPSE is that only one cardiac view is

Fig. 2. A. ROC Model for MAPSE Predicting Abnormal LVEF on Comprehensive Echocardiogram. B: ROC Model for EPSS Predicting Abnormal LVEF on Comprehensive Echocardiogram.

Table 2

Test Characteristics of MAPSE, EPSS, and Estimated LVEF Predicting Abnormal LVEF on Comprehensive Echocardiogram.

Measurement	Sensitivity %, 95% CI	Specificity	PPV	NPV	PLR	NLR	Accuracy
		%, 95% CI	%, 95% CI	%, 95% CI	#, 95% CI	#, 95% CI	%, 95% CI
MAPSE <8 mm	42 (22-63)	89 (75-97)	71 (47-88)	70 (62-77)	3.9 (1.4-10.9)	0.65 (0.46-0.93)	71 (58-82)
EPSS >10 mm	79 (58-93)	76 (59-88)	68 (53-79)	85 (71-93)	3.3 (1.8-6.0)	0.28 (0.12-0.61)	77 (65-87)
EPSS >7 mm	92 (73-99)	60 (42-75)	60 (49-69)	92 (74-98)	2.3 (1.5-3.4)	0.14 (0.04-0.54)	72 (59-83)
LVEFa	100 (86-100)	59 (42-75)	62 (52-70)	100	2.5 (1.7-3.6)	0.00	75 (62-86)

Abnormal LVEF on comprehensive echocardiogram defined as <50%. MAPSE: mitral annular plane systolic excursion, EPSS: E-point septal separation, LVEF: left ventricular ejection fraction.

^a LVEF: Measures considered abnormal when the estimated LVEF value is <50% (either estimated to be moderately reduced [30-49%] or severely reduced [<30%]).

Table 3

Test Characteristics of Combining MAPSE, EPSS, and Estimated LVEF for Prediction of Abnormal LVEF on Comprehensive Echocardiogram.

Measurement	Sensitivity %, 95% CI	Specificity	PPV	NPV	PLR	NLR	Accuracy
		%, 95% CI	%, 95% CI	%, 95% CI	#, 95% CI	# 95% CI	%, 95% CI
1 or more abnormal	100 (86-100)	50 (33-67)	58 (50-66)	100	2.0 (1.4-2.8)	0.00	71 (57-82)
2 or more abnormal	80 (59-93)	86 (71-95)	90 (63-90)	86 (74-93)	5.8 (2.5-13.3)	0.23 (0.1-0.5)	84 (72-92)
All three abnormal	44 (24-65)	94 (81-99)	85 (57-96)	71 (63-78)	7.9 (1.9-32.7)	0.59 (0.42-0.85)	74 (61-84)

Abnormal LVEF on comprehensive echocardiogram defined as <50%. MAPSE: mitral annular plane systolic excursion, EPSS: E-point septal separation, LVEF: left ventricular ejection fraction.

MAPSE abnormal if <8 mm, EPSS abnormal if >10 mm, and estimated LVEF considered abnormal when the estimated LVEF value is <50% (either estimated to be moderately reduced [30-49%] or severely reduced [<30%]).

necessary for measurement, which can be helpful in the assessment of cardiac function patients with limited ultrasound windows. MAPSE measurements in this study were performed by providers with varying expertise in ultrasound, which supports generalizability of this technique.

The test performance of MAPSE in our patient population mirrors the cardiology data. MAPSE values of <8 mm had high specificity for low LVEF and MAPSE values >10 mm had high specificity for normal LVEF. Our Youden index analysis did show that values of 8 mm was the optimal cut point for MAPSE data, which is in line with the published literature. EPSS literature has a varying diagnostic cut-off values that have been reported in the literature ranging from 7 to 10 mm [7,9,29,30]. When testing two different cut off points in our population, EPSS >7 mm had a higher sensitivity and EPSS >10 mm had a higher specificity and higher overall accuracy. Our Youden index EPSS cut-off of greater than or equal to 11 mm supports the use of the higher cut off value in the ED patient population.

The sensitivity of estimated LVEF and EPSS were higher than MAPSE; it is likely that a using a combination of FOCUS evaluations will provide the most comprehensive assessment of cardiac function. Our incremen- tal validity assessment did show that adding EPSS and EP estimated LVEF to MAPSE did increase the test performance in this study. Addi- tional quantitative cardiac measurements validated for use in the ED will help EPs to have a more nuanced ability to detect subtle depressed cardiac function or to detect Cardiac abnormalities in patients with lim- ited FOCUS ultrasound views.

MAPSE could have important future implications for evaluation of patients with suspected atherosclerotic cardiac disease in the ED. As shown in this data, MAPSE values can correlate with BNP levels and tro- ponin I levels [19]. In patients with significant coronary artery stenosis, MAPSE was significantly decreased when compared to controls. This ef- fect correlated to the severity and extent of the coronary stenosis [31]. In one study assessing LV regional wall motion (the usual echocardiogra- phic evidence of coronary artery disease), those that had a depressed

Fig. 3. A: Scatter Plot of Comprehensive Echocardiogram LVEF (%) to MAPSE (mm), p < .001. B:. Scatter Plot of MAPSE Measurement (%) to EPSS (mm), p < .001.

Fig. 4. Box and Whisker Plot showing mean, median, IQR, and range of MAPSE values for subjects with varying EP estimated LVEF categories: hyperdynamic (LVEF >65%), normal (LVEF 50-65%), reduced (LVEF 30-49%), or severely reduced (LVEF <30%).

MAPSE were found to have evidence of severe cardiovascular disease (i.e. prior myocardial infarction, coronary artery disease, Uncontrolled hypertension). Willenheimer et al. showed that only 4% of the patients with reduced MAPSE did not have evidence of cardiovascular disease [32]. MAPSE also can help to risk stratify patients after myocardial in- farction and with heart failure. Those who have MAPSE measurements over 9 mm have better 10-year survival compared to those with MAPSE <5 mm [16,33]. In summary, this study shows that MAPSE mea- surements in the ED are reliable and correlate to LVEF on CE using the same cut-off values used in the cardiology literature. Further studies on the future application of MAPSE measurements for ED patients with suspected cardiac disease are needed to delineate any role MAPSE could have in Cardiovascular risk stratification in the ED.

1. Limitations

This study has a few limitations related to study design. As this was a convenience sample, it is possible that this study does not represent the entire population of patients that are being evaluated for ADHF in the ED. It is possible that given our non-systematic approach to screening continuously could contribute to selection bias. Patients were enrolled during all days of the week and all hours of the day, but only when a trained study provider was available to collect ultrasound images. As our primary goal was to try to validate the use of MAPSE in the ED pop- ulation by comparison to CE results, we hope that any effect that selec- tion bias would exert was minimal. Given the lack of prior data on MAPSE in the ED setting, sample size calculations were not able to be performed. Due to this factor and our small sample size, it is possible that this study was underpowered to detect test characteristic differ- ences between different FOCUS cardiac assessments. As this is a popula- tion that was being evaluated for possible ADHF, a high percentage of patient (39%) had an abnormal LVEF on comprehensive echocardio- gram. It is possible that if MAPSE were applied to a broader range of ED patients that the test characteristics would change.

Some patients did not receive a new CE during the admission associ- ated with study enrollment. Patients could be included if they had a CE result in the 12 months prior to or 3 months after admission. A 12 month pre-enrollment range was chosen as patients with chronic heart failure will have follow-up echocardiograms performed on an an- nual basis, and therefore results within 12 months are considered to be recent [34]. However, if there were any significant change in a subject’s

LVEF from the study performed at a different time than study enroll- ment could change the test characteristics of all FOCUS measurements. The CE performed most proximally to study enrollment was used to minimize this effect. As most in this study were not diagnosed with acute coronary syndrome, the study investigators felt that the LVEF re- sults from the CEs obtained at times other than the hospitalization asso- ciated with enrollment were likely a valid representation of the subject’s baseline cardiac function.

The evaluation of MAPSE results for subgroups of EP estimated LVEF function (Fig. 4) should be regarded as exploratory, as confirming the mapping of MAPSE values to LVEF ranges would require a substantially larger sample to be confirmed and replicated.

1. Conclusions

MAPSE is a cardiac measurement that can be reliably and easily ob- tained by EPs with varying ultrasound experience with minimal addi- tional training. In the ED setting, A MAPSE value <8 mm had moderate predictive value and specificity for LVEF <50% on comprehen- sive echocardiogram and had similar performance to LVEF estimation and EPSS. MAPSE values >10 mm had high specificity for LVEF >50%. Adding MAPSE to a multimodal FOCUS assessment in the ED could in- crease diagnostic accuracy for reduced LVEF. Further studies are needed to validate these results on a larger scale.

Author contributions

AS: study concept and design, data acquisition, data analysis, manu- script writing, manuscript revision.

JK: study concept and design, data acquisition, manuscript revision. NA: study design, data acquisition, manuscript revision.

TL: statistical expertise, data analysis.

JB: study design, statistical expertise, data analysis. KB: data acquisition.

KD: study concept and design, data acquisition, data analysis, manu- script writing, manuscript revision.

Disclosures

AS: none. JK: none.

NA: none. TL: none. JB: none. KB: none. KD: none.

Acknowledgments

none.

Declaration of Competing Interest

We confirm that the authors have no conflicts of interest to disclose.

Appendix A. Supplementary data

Supplementary data to this article can be found online at https://doi. org/10.1016/j.ajem.2023.03.018.

References

1. Virani SS, Alonso A, Benjamin EJ, Bittencourt MS, Callaway CW, Carson AP, et al. Heart disease and stroke statistics-2020 update: a report from the American Heart Association. Circulation. 2020;141:e139-596.
2. Kuo DC, Peacock WF. Diagnosing and managing acute heart failure in the emergency department. Clin Exp Emerg Med. 2015;2:141-9. https://doi.org/10.15441/ceem.15. 007.
3. Moore CL, Rose GA, Tayal VS, Sullivan DM, Arrowood JA, Kline JA. Determination of left ventricular function by emergency physician echocardiography of hypotensive patients. Acad Emerg Med. 2002;9:186-93. https://doi.org/10.1111/j.1553-2712. 2002.tb00242.x.
4. Unluer EE, Karagoz A, Akoglu H, Bayata S. visual estimation of bedside echocardiog- raphic ejection fraction by emergency physicians. West J Emerg Med. 2014;15: 221-6. https://doi.org/10.5811/westjem.2013.9.16185.
5. Randazzo MR, Snoey ER, Levitt MA, Binder K. Accuracy of emergency physician as- sessment of left ventricular ejection fraction and central venous pressure using echocardiography. Acad Emerg Med. 2003;10:973-7. https://doi.org/10.1111/j. 1553-2712.2003.tb00654.x.
6. Petti MA, Viotti R, Armenti A, Bertocchi G, Lococo B, Alvarez MG, et al. Predictors of heart failure in chronic chagasic cardiomyopathy with asymptomatic left ventricular dysfunction. Rev Esp Cardiol. 2008;61:116-22.
7. D’Cruz IA, Lalmalani GG, Sambasivan V, Cohen HC, Glick G. The superiority of mitral E point-ventricular septum separation to other echocardiographic indicators of left ventricular performance. Clin Cardiol. 1979;2:140-5. https://doi.org/10.1002/clc. 4960020210.
8. Secko MA, Lazar JM, Salciccioli LA, Stone MB. Can junior emergency physicians use E- point septal separation to accurately estimate left ventricular function in acutely dyspneic patients? Acad Emerg Med. 2011;18:1223-6. https://doi.org/10.1111/j. 1553-2712.2011.01196.x.
9. McKaigney CJ, Krantz MJ, La Rocque CL, Hurst ND, Buchanan MS, Kendall JL. E-point septal separation: a bedside tool for emergency physician assessment of left ventric- ular ejection fraction. Am J Emerg Med. 2014;32:493-7. https://doi.org/10.1016/j. ajem.2014.01.045.
10. Fellow Corner: E-point septal separation in the patient with congestive heart failure

n.d. (accessed August 16, 2021). https://www.acep.org/how-we-serve/sections/ emergency-ultrasound/news/dece/fellow-corner-e-point-septal-separation-in-the- patient-with-congestive-heart-failure/.

1. Adhikari S, Fiorello A, Stolz L, Amini R, Gross A, O’Brien K, et al. Can emergency phy- sicians accurately identify complex abnormalities on Point-of-care echocardiogram? Ann Emerg Med. 2013;62:S78. https://doi.org/10.1016/j.annemergmed.2013.07. 034.
2. Hu K, Liu D, Herrmann S, Niemann M, Gaudron PD, Voelker W, et al. Clinical impli- cation of mitral annular plane systolic excursion for patients with cardiovascular dis- ease. Eur Heart J Cardiovasc Imaging. 2013;14:205-12. https://doi.org/10.1093/ ehjci/jes240.
3. Simonson JS, Schiller NB. Descent of the base of the left ventricle: an echocardiogra- phic index of left ventricular function. J Am Soc Echocardiogr. 1989;2:25-35. https:// doi.org/10.1016/s0894-7317(89)80026-4.
4. Alam M, Hoglund C, Thorstrand C, Philip A. Atrioventricular plane displacement in severe congestive heart failure following dilated cardiomyopathy or myocardial in- farction. J Intern Med. 1990;228:569-75. https://doi.org/10.1111/j.1365-2796.1990. tb00281.x.
5. Alam M. The atrioventricular plane displacement as a means of evaluating left ven- tricular systolic function in acute myocardial infarction. Clin Cardiol. 1991;14: 588-94. https://doi.org/10.1002/clc.4960140711.
6. Bergenzaun L, Ohlin H, Gudmundsson P, Willenheimer R, Chew MS. Mitral annular plane systolic excursion (MAPSE) in shock: a valuable echocardiographic parameter in intensive care patients. Cardiovasc Ultrasound. 2013;11:16. https://doi.org/10. 1186/1476-7120-11-16.
7. Adel W, Roushdy AM, Nabil M. Mitral annular plane systolic excursion-derived ejec- tion fraction: a simple and valid tool in adult males with left ventricular systolic dys- function. Echocardiography. 2016;33:179-84. https://doi.org/10.1111/echo.13009.
8. Grue JF, Storve S, Dalen H, Salvesen O, Mjolstad OC, Samstad SO, et al. Automatic measurements of mitral annular plane systolic excursion and velocities to detect left ventricular dysfunction. Ultrasound Med Biol. 2018;44:168-76. https://doi.org/ 10.1016/j.ultrasmedbio.2017.09.002.
9. Elnoamany MF, Abdelhameed AK. Mitral annular motion as a surrogate for left ven- tricular function: correlation with brain natriuretic peptide levels. Eur J Echocardiogr. 2006;7:187-98. https://doi.org/10.1016/j.euje.2005.05.005.
10. Ozer PK, Govdeli EA, Demirtakan ZG, Nalbant A, Baykiz D, Orta H, et al. The relation of echo-derived lateral MAPSE to left heart functions and biochemical markers in pa- tients with preserved ejection fraction: short-term prognostic implications. J Clin Ul- trasound. 2022. https://doi.org/10.1002/jcu.23173.
11. Alam M, Hoglund C, Thorstrand C. Longitudinal systolic shortening of the left ventri- cle: an echocardiographic study in subjects with and without preserved global func- tion. Clin Physiol. 1992;12:443-52. https://doi.org/10.1111/j.1475-097x.1992. tb00348.x.
12. Silva JA, Khuri B, Barbee W, Fontenot D, Cheirif J. Systolic excursion of the mitral an- nulus to assess septal function in paradoxic septal motion. Am Heart J. 1996;131: 138-45. https://doi.org/10.1016/s0002-8703(96)90062-9.
13. Stenberg Y, Wallinder L, Lindberg A, Wallden J, Hultin M, Myrberg T. Preoperative point-of-care assessment of left ventricular systolic dysfunction with transthoracic echocardiography. Anesth Analg. 2021;132:717-25. https://doi.org/10.1213/ANE. 0000000000005263.
14. Matos J, Kronzon I, Panagopoulos G, Perk G. Mitral annular plane systolic excursion as a surrogate for left ventricular ejection fraction. J Am Soc Echocardiogr. 2012;25: 969-74. https://doi.org/10.1016/j.echo.2012.06.011.
15. Daley J, Grotberg J, Pare J, Medoro A, Liu R, Hall MK, et al. Emergency physician per- formed tricuspid annular plane systolic excursion in the evaluation of suspected pul- monary embolism. Am J Emerg Med. 2017;35:106-11. https://doi.org/10.1016/j. ajem.2016.10.018.
16. Dwyer KH, Rempell JS, Stone MB. Diagnosing centrally located pulmonary embo- lisms in the emergency department using point-of-care ultrasound. Am J Emerg Med. 2018;36:1145-50. https://doi.org/10.1016/j.ajem.2017.11.033.
17. Daley JI, Dwyer KH, Grunwald Z, Shaw DL, Stone MB, Schick A, et al. Increased sen- sitivity of focused cardiac ultrasound for pulmonary embolism in emergency depart- ment patients with Abnormal vital signs. Acad Emerg Med. 2019;26:1211-20. https://doi.org/10.1111/acem.13774.
18. Ruopp MD, Perkins NJ, Whitcomb BW, Schisterman EF. Youden index and optimal cut-point estimated from observations affected by a lower limit of detection. Biom

J. 2008;50:419-30. https://doi.org/10.1002/bimj.200710415.

1. Hanson EW, EPSS Lowson. Anesth Analg. 1998;86:62SCA. https://doi.org/10.1097/ 00000539-199804001-00062.
2. Massie BM, Schiller NB, Ratshin RA, Parmley WW. Mitral-septal separation: new echocardiographic index of left ventricular function. Am J Cardiol. 1977;39: 1008-16. https://doi.org/10.1016/s0002-9149(77)80215-4.
3. Rydberg E, Willenheimer R, Erhardt L. Left atrioventricular plane displacement at rest is reduced in relation to severity of coronary artery disease irrespective of prior myocardial infarction. Int J Cardiol. 1999;69:201-7. https://doi.org/10.1016/ s0167-5273(99)00036-4.
4. Willenheimer R, Rydberg E, Stagmo M, Gudmundsson P, Ericsson G, Erhardt L. Echo- cardiographic assessment of left atrioventricular plane displacement as a comple- ment to left ventricular regional wall motion evaluation in the detection of myocardial dysfunction. Int J Card Imaging. 2002;18:181-6. https://doi.org/10. 1023/a:1014664825080.
5. Svealv BG, Olofsson EL, Andersson B. Ventricular long-axis function is of major im- portance for long-term survival in patients with heart failure. Heart. 2008;94: 284-9. https://doi.org/10.1136/hrt.2006.106294.
6. Celutkiene J, Spoletini I, Coats AJS, Chioncel O. Left ventricular function monitoring in heart failure. Eur Heart J Suppl. 2019;21:M17-9. https://doi.org/10.1093/eurheartj/ suz218.