Abstract
Background
Early detection of left ventricular (LV) dysfunction before the onset of Duchenne muscular dystrophy–associated cardiomyopathy (DMDAC) may direct earlier clinical management. Ejection fraction (EF) is the most measured parameter. However, a decline in EF is a late finding in the assessment of heart failure.
Objective
This study aims to measure non-invasive pressure-volume loop-derived ventricular efficiency, contractility, and stroke work (SW) in DMDAC patients and investigate how these measured parameters correlate with EF in patients who develop DMDAC.
Methods
The current study performed a retrospective case-control study of thirty DMDAC patients who underwent two serial CMR imaging from 2016 to 2023. The patients were divided into +DMDAC and –DMDAC based on EF. Brachial pressures from cuff sphygmomanometer and CMR short axis steady-state free-precession images covering the LV were acquired, and non-invasive pressure-volume loop derived efficiency, contractility, and SW were obtained using an elastance based model.
Results
Efficiency and contractility correlated significantly (r = 0.99, p < 0.01) and (r = 0.74, p < 0.01), respectively, with EF for all subjects. Conversely, no significant correlation was found between SW and EF. Furthermore, efficiency and contractility were significantly less in the EF < 55 % group than when compared to EF ≥ 55 % (p < 0.001), but there was no significant difference in SW for both categories.
Conclusion
Efficiency and contractility show a significant correlation with EF, with future work needed to determine if they hold prognostic information toward developing DMDAC.
Highlights
- •
DMDAC patients need better diagnostic endpoints than ejection fraction.
- •
Energy efficiency and contractility could diagnose DMDAC patients non-invasively.
- •
Stroke work may not be a diagnostic index for DMDAC patients.
- •
Energy efficiency and contractility may lead to longitudinal assessment of DMDAC.
1
Introduction
Duchenne muscular dystrophy (DMD) is the most common inherited muscle disease, affecting ∼1 in 10,000 male children [ ]. Loss of functional dystrophin protein results in progressive skeletal muscle weakness, loss of ambulation, and eventually DMD-associated cardiomyopathy (DMDAC) [ , ]. With the onset of adequate respiratory support, left ventricular (LV) dysfunction has been found to be the most common cause of morbidity and mortality in patients with DMD [ , ]. Although cardiomyopathy is inevitably progressive, markers of myocardial dysfunction are necessary for starting therapies and monitoring therapeutic efficacy.
1.1
Noninvasive PV loops using cardiovascular magnetic resonance (CMR) and brachial pressure
Pressure-volume (PV) loops provide significant insight into cardiac function of patients with DMDAC [ ] but are not readily available in clinical procedures. The information provided by PV loops provides a unique, quantitative approach to determining the contractility of the heart, independent of preload and afterload [ ]. Simultaneous ventricular pressure and volume measurements are needed to address the limitation of the influence of loading conditions. Ventricular pressure measurement is invasive and requires cardiac catheterization [ ]. Noninvasive techniques have been proposed [ , ] to derive PV loops without intervention and established to estimate ventricular efficiency, contractility, and other associated quantitative measures. They can detect subclinical signs of LV dysfunction. Seemann et al. [ ] described a novel noninvasive method, based on time-varying elastance, to compute LV PV loops using LV volume curves from CMR imaging and brachial pressure as the input to the algorithm. This method computes LV pressure as the product of elastance from a mathematical formulation and volume from CMR. Their results showed agreement with in vivo PV loop measurements [ ]. The advantage of using CMR, the gold standard, is that it provides accurate intracardiac volumes.
1.2
Application of noninvasive PV loop in DMDAC patients
CMR is a routine modality for cardiac evaluation in DMDAC [ , ]. Hence, noninvasive PV loop-derived contractility, ventricular efficiency, and stroke work obtained from the approach by Seemann et al. [ ] may be helpful in the risk stratification of patients with DMDAC. The ejection fraction (EF) is the most common noninvasively measured parameter in DMDAC patients to evaluate cardiac health. However, a decline in EF is a late finding in the longitudinal assessment of heart failure. Consequently, earlier measures of cardiac function are needed for this population. Early detection of LV dysfunction before the onset of DMDAC may direct earlier clinical management. Acquisition of noninvasive PV loops is yet to be explored for the LV in the DMDAC population. Therefore, this study performed LV PV loop analysis in DMDAC patients. The present study aims to characterize noninvasive LV PV loop-derived energy efficiency, contractility, and stroke work in patients with DMDAC. Additionally, this study intends to investigate how these measured parameters correlate with EF in patients with DMDAC. These new indices include both pressure and volume (discussed in the method section), whereas EF (current gold standard) is a ratio of volumes only.
2
Methodology
The cases and controls in this study were selected retrospectively using random sampling from the DMDAC MRI database. The groups were evaluated based on EF for comparison. The +DMDAC group are patients who demonstrated an absolute EF < 55 %, and –DMDAC group are patients who demonstrated an absolute EF ≥ 55 % on two serial CMR imaging. A total of 50 patients were randomly screened from 2010 to 2023. Patient selection was based on the readable second MRI scan. Using this criteria, patients with MRI scans from 2010 to 2015 had to be excluded when either the first or the second MRI scan could not be read by the medical image analysis software, Segment, v3.2 R9074 (Medviso, Lund, Sweden, segment.heiberg.se ) [ ]. It appears that the version of Segment used in this study was unable to read all MRI scans performed prior to 2014. Out of the 50 DMDAC patients evaluated, 20 were screen failures. Among these 20 screen failures, 14 images could not be opened by Segment due to either MRIs being outdated or the images being inaccessible. Additionally, 3 images had poor-quality MRIs that prevented LV segmentation in Segment, and data for the remaining 3 patients were missing.
Consequently, this retrospective case-control study included 30 patients (m = 30) with DMDAC who underwent two serial CMR imaging (n = 2) at Cincinnati Children’s Medical Center (CCHMC) between 2016 and 2023 based on the readable second MRI scan. Therefore, 60 images ([n <SPAN role=presentation tabIndex=0 id=MathJax-Element-1-Frame class=MathJax style="POSITION: relative" data-mathml='×’>××
×
m] = [2 <SPAN role=presentation tabIndex=0 id=MathJax-Element-2-Frame class=MathJax style="POSITION: relative" data-mathml='×’>××
×
30] = 60) were assessed. The patients were divided into two major categories: +DMDAC and –DMDAC ( Fig. 1 ). The prefix LV is eliminated from efficiency, contractility, and stroke work (SW) from here onward because these parameters were assessed for the LV of DMDAC.
2.1
Study population
The classification of 30 patients for the current study is a) 16 patients have EF < 55 % on serial MRI (i.e., +DMDAC); b) 9 patients have EF ≥ 55 % on serial MRI (i.e., –DMDAC); c) 3 patients have EF < 55 % on the first MRI and EF ≥ 55 % on the subsequent second MRI, and d) remaining 2 patients have EF ≥ 55 % on the first MRI and EF < 55 % on the subsequent second MRI. Therefore, this study had only 2 patients who started with EF ≥ 55 % and showed dysfunction EF < 55 % on the subsequent CMR. With such a limited (2 patients) dataset, a longitudinal comparison cannot be conducted between two MRI scans – initial scan showing EF ≥ 55 % and the latter one showing EF < 55 %. More such patients are needed to predict decrease in LV function. Patients’ characteristics are specified in Table 1 .
| EF ≥ 55 % | EF < 55 % | p-Value | |
|---|---|---|---|
| Age (y) | 14.8 ± 4.8 | 19.4 ± 3.9 | 0.002 ⁎ |
| Heart rate (bpm) | 101.5 ± 17.0 | 90.7 ± 18.2 | 0.03 ⁎ |
| Systolic blood pressure (mm Hg) | 113.9 ± 8.3 | 115.3 ± 7.4 | 0.58 |
| Diastolic blood pressure (mm Hg) | 66.0 ± 8.0 | 64.4 ± 6.6 | 0.40 |
| End diastolic volume (mL) | 81.0 ± 23.3 | 131.6 ± 46.7 | <0.0001 ⁎ |
| End systolic volume (mL) | 33.1 ± 11.3 | 76.6 ± 35.3 | <0.0001 ⁎ |
| Stroke volume (mL) | 47.9 ± 12.6 | 55.0 ± 16.4 | 0.08 |
| Ejection fraction (%) | 59.6 ± 3.5 | 43.0 ± 8.2 | <0.0001 ⁎ |
| Cardiac output (L/min) | 4.7 ± 1.0 | 4.9 ± 1.5 | 0.68 |
Fourteen of the 30 patients had no medication change within the two serial CMR imaging. Of the remaining 16 patients, 10 changed medications before their second CMR imaging. For example, Entresto was increased for two patients, while the remaining eight had either increased or decreased carvedilol, lisinopril, enalapril, and Aldactone. Six patients had no clinical notes between the serial CMRs. Brachial pressures from cuff sphygmomanometer and CMR imaging were acquired for each patient. Approved procedures and protocols by CCHMC regulation were performed in handling patient data. The CCHMC Institutional Review Board approved the study, and informed consent was not required because de-identified data was shared and used. All parameters that are used in this study are from approved standard-of-care procedures only. Therefore, additional consent was not required for this study.
2.2
Data acquisition
2.2.1
CMR protocol
CMR short-axis steady-state free-precession images covering the LV were performed on a 1.5 Tesla magnetic resonance scanner (Ingenia; Philips Healthcare; Best, the Netherlands). Study parameters were temporal resolution of 30 to 40 ms, slice thickness 6‐–8 mm without a gap, and in-plane resolution 1.5 × 1.5 mm. The cine images were obtained during end-expiratory breath-holds using retrospective ECG gating. Images were acquired in the vertical and horizontal ventricular long-axis planes and a stack of slices in a ventricular short-axis plane from the atrioventricular junction through the cardiac apex.
2.2.2
Image analysis
LV endocardial borders were automatically delineated over the entire cardiac cycle (usually 30-time phases), and manual corrections were performed, when necessary, in all time frames. Ventricular segmentations were performed using Segment with the LV end-diastolic pressure fixed to 5 mmHg for all subjects. The PV loops were generated using a formulation based on the time-varying elastance:
Et=PtVt−V0
2.2.3
Hemodynamic parameters
Several hemodynamic parameters can be evaluated with the help of the PV loop. Fig. 2 A shows the noninvasive PV loops of a patient with EF ≥ 55 % and EF < 55 %. The enclosed area in Fig. 2 B shows the stroke work (SW) and potential energy (PE). SW is the energy the heart muscle exerts to expel the stroke volume. PE is the energy that the heart must overcome to eject blood. The slope of the end-systolic pressure-volume relationship (ESPVR) is a metric that quantifies the contractility of the ventricle. It represents the maximum elastance ( Fig. 2 B). The ventricular efficiency ( <SPAN role=presentation tabIndex=0 id=MathJax-Element-4-Frame class=MathJax style="POSITION: relative" data-mathml='η’>𝜂η
η
) can be derived as follows:
η=SWSW+PE
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
