Introduction

NIDCM is a severe form of primary myocardial disease characterized by dilatation and dysfunction of the left or both ventricles. Despite the recent progress in medical care, progressive HF with a poor prognosis is related with NIDCM [1]. Current diagnostic methods rely on a combination of clinical assessment, family history, and genetics alongside imaging methods [2]. NIDCM often features an extended sub-clinical phase–a period where symptoms and myocardial dysfunction may not be present [3], this can potentially lead to a delay in accurate diagnosis and timely prescription of treatment. Thus, more sensitive tools are needed for detailed risk evaluation.

CMR has become the most useful and accurate non-invasive technique for assessing cardiac structure and function [4]. CMR has recently been proven as a reliable method for determining functional myocardial deformation parameters using FT, and may offer greater clinical value compared to LVEF measurements alone [5]. Previous small and large multicenter studies with long term follow up (up to 4–6 years) demonstrated the independent and incremental predictive value of CMR derived LV GLS for both ischemic and nonischemic dilated cardiomyopathy patients [6, 7]. Despite the surge in reports about this issue, most of the studies that investigated patients with NIDCM only limited analysis to standalone LVEF, LV, or RV deformation parameters [7, 9, 11]. Regarding risk stratification using the FT-derived mechanics for patients with NIDCM, there are no studies that analyzed whole-heart strain parameters. In this context, we aimed to perform a study assessing the relationship between CMR-derived whole-heart myocardial strain and early primary outcomes in NIDCM patients.

Methods

Study population

This was a prospective study involving patients with NIDCM. The diagnosis of NIDCM was made according to the latest European Society of Cardiology (ESC) document [2]. The exclusion criteria for the study were:

  • an ischemic coronary disease (more than 50% coronary artery occlusion by invasive or cardiac computed tomography coronary angiography);

  • primary valvular heart disease;

  • chronic severe kidney disease;

  • poor CMR quality;

  • inflammatory myocardial disease;

  • tachycardia-induced HF;

  • previous pulmonary embolism;

  • peripartum cardiomyopathy;

  • toxic damage;

  • over 18 years of age.

of 110 patients with a diagnosis of NIDCM, we excluded 12 because their CMR image quality was inappropriate for analysis, resulting in a final sample size of 98 patients. Basic clinical characteristics, laboratory results, and electrocardiography were recorded in standard form. All participants gave written informed consent before enrollment. The study was approved by the local institutional ethics committee.

Follow-up and endpoints

The study consisted of two phases: during the first phase, patients were enrolled and examined for the first time and diagnosed with NIDCM (patients without chronic or worsening HF); during the second phase, early primary outcomes were evaluated after 1-year follow-up. During follow-up, all patients were treated with an optimal HF treatment according to chronic HF guidelines [12]. Information about the presented adverse events was collected from medical records or patients were invited for a follow-up visit. The early primary outcomes were cardiac death, heart transplantation, and hospitalization for WHF at one year. WHF was defined according to current recommendations by the American College of Cardiology (patients admitted to hospital with decompensated HF requiring treatment with intravenous HF drugs) [13].

CMR protocol and strain analysis

The CMR analysis was conducted with a 3.0-T magnetic resonance imaging scanner (MAGNETOM Skyra, Siemens Healthcare, Erlangen, Germany) using an 18-channel cardiac coil and electrocardiogram gating. Cine images were acquired using standard balanced steady-state free precession (bSSFP) sequences in long axes (2-, 3-, and 4-chamber) and the stack of short axis (covering entire ventricles) views during an expiratory breath hold. The CMR images were transferred to an off-line workstation with CMR post-processing software Medis Suite 3.1 (Medis Medical Imaging, Leiden, The Netherlands), and the data were analyzed by a trained, blinded observer.

LV and RV volumes, LVEF, right ventricular ejection fraction (RVEF), and LV mass were evaluated from short-axis cine images using standard volumetric techniques [14]. CMR strain analysis was performed using the FT application QStrain (Medis Suite 3.1, Medis Imaging, Leiden, The Netherlands). The endocardial borders of all cardiac chambers were traced manually in end-systole and an automated tracking algorithm was used, further manual adjustments were performed when needed [15].

LV GLS and GCS were derived by averaging the peak strain values of individual segments using a 17 segment model. The LV GLS was calculated from 2-, 3-, and 4-chamber long-axis views (Fig. 1).

Fig. 1
figure 1

Example of left ventricular global longitudinal strain assessment using feature tracking software. Endocardial borders were delineated on long-axis two-(A), three-(B), and four-chamber (C) SSFP cine images in end-systole and end-diastole. The final automatic calculation was performed by the software: the schematic picture shows minimal myocardium movement and the average GLS of all 17 cardiac segments in this case was − 5.09 (D)

Full size image

Mean T1 map values were measured on T1 mapping short-axis images by drawing a region of interest (ROI) in the septum at LV basal, mid, and apical planes. This was performed both before and after gadolinium-based contrast agent administration.

For myocardial extracellular volume (ECV) evaluation, a ROI in the center of the blood pool in the native and the post-contrast T1 mapping images was drawn, excluding papillary muscles and trabeculae. Hematocrit level was determined for each patient from a venous blood sample less than 24 h before the CMR examination.

The LV GCS was automatically calculated from previously traced points in QMass on short-axis views at the base, mid, and apex planes during end-systole and end-diastole (Fig. 2).

Fig. 2
figure 2

Left ventricular global circumferential strain assessment using feature tracking software

Full size image

RV and right atrial (RA) GLS were calculated from 4-chamber, while left atrial (LA) GLS was calculated from 2-chamber long-axis views (Fig. 3).

Fig. 3
figure 3

Cardiac magnetic resonance imaging-derived whole-heart myocardial mechanical parameters assessment using feature tracking software. Global longitudinal strain of the left atrium A right atrium B and right ventricle C was assessed by tracing endocardial borders on long-axis SSPF cine images in end-systole and end-diastole

Full size image

Statistical analysis

The results were presented as means ± standard deviations (SD) or as absolute numbers and percentages. The study population was divided into two groups according to the presence of outcomes (with or without the early primary outcomes). The student’s t‐test was used to compare normally distributed parameters, and the Mann–Whitney U‐test for abnormally distributed parameters. The point-biserial coefficient of correlation R2 was used to assess the relationship between myocardial mechanics and morphometrics, and the presence of worse prognosis. Binary logistic regression analysis was used to determine the potential predictors of a worse prognosis. Intra- and interobserver variability was tested by using the intraclass correlation coefficient. A p-value of < 0.05 was considered statistically significant. Analyses were performed using SPSS version 22 (IBM, Chicago, IL, USA).

Results

The baseline characteristics of the entire study cohort are described in Table 1. There were no significant differences in mean age, sex, risk factors (arterial hypertension, dyslipidemia, smoking etc.), and clinical characteristics (New York Heart Association (NYHA) class, atrial fibrillation etc.) in both groups (p > 0.05). The brain natriuretic peptide (BNP) and C-reactive protein concentrations were higher in patients with worse prognosis (1858.1 ± 963.2 vs. 791.9 ± 398.2 ng/l, p = 0.01; 10.9 ± 5.8 vs. 8.3 ± 5.0 g/l, p = 0.03). The presence of early primary outcomes in NIDCM patients was not linked to genetically identified pathogenic variants. During a 1-year follow-up period, cardiovascular (CV) outcomes occurred in 32 (32,6%) patients: 30 (93.8%) have been hospitalized for HF worsening, including 8 (25%) deaths and 3 (9.4%) heart transplants.

Table 1 Baseline characteristics for NIDCM patients with and without early primary outcomes
Full size table

BSA—body surface area; ACE-I—angiotensin-converting enzyme inhibitor; ARB—angiotensin receptor blocker; CCB—calcium channel blocker; VT—ventricular tachycardia; LBBB—left bundle branch block; HF—heart failure; CA—coronary artery; NYHA—New York Heart Association; 6MWT—6-min walk test; Hs-CRP—high sensitivity C-reactive protein; BNP—brain natriuretic peptide; NLR—Neutrophil to Lymphocyte Ratio; TnI—troponin I. The data from the MRI parameters in NIDCM patients are summarized in Table 2. The patients with worse CV outcomes had more dilated LV (p = 0.016), LA area (p = 0.005), and RV (p = 0.005). There was no significant difference in measurements of RA area between the groups (p > 0.05). Whole-heart chambers strain parameters were significantly better in the patient group without early primary outcomes (p < 0.001). Patients with better prognosis had significantly higher LVEF (22.7 ± 8.7 vs. 33.56 ± 10.4, p < 0.001, respectively). T1 mapping and late gadolinium enhancement did not show significant connections with NIDCM patient outcomes (p < 0.05).

Table 2 MRI parameters in NIDCM patients with and without early primary outcomes
Full size table

LVEDD—left ventricular end-diastolic diameter; LVEDV—left ventricular end-diastolic volume; LVESV—left ventricular end-systolic volume; IVS—interventricular septum; PW—posterior wall; LVGLS—left ventricular global longitudinal strain; LVGCS—left ventricular global circumferential strain; LVEF—left ventricular ejection fraction; RVEDV—right ventricular end-diastolic volume; RVESV—right ventricular end-systolic volume; LAA—left atrial area; RAA—right atrial area; RVLS—right ventricular longitudinal strain; LALS—left atrial longitudinal strain; RALS—right atrial longitudinal strain; ECV—extracellular volume.

Whole-heart myocardial mechanics showed weak-moderate but significant correlations with early primary outcomes (p < 0.05). The parameters of left ventricular end-diastolic diameter, left ventricular end-systolic volume, right ventricular end-systolic volume, and LA area were also related with patients’ worse prognosis (p < 0.05). LVEF and LV GLS showed the strongest correlation with poor prognosis compared with other parameters (rs 0.457, p < 0.001, and rs − 0.420, respectively). However, multivariate regression analysis (Table 3) revealed that the LV GLS was an independent predictor of an early worse prognosis such as cardiac death, HT or hospitalization for WHF after one year (OR 0.787, CI 95% 0.697–0.890, p < 0.001).

Table 3 Binary logistic regression analysis for the MRI parameters related to the presence of adverse events
Full size table

LV—left ventricular; LVESVi—left ventricular end-systolic volume index; LAAi—left atrial area index; RV—right ventricular; GLS—global longitudinal strain; LALS—left atrial longitudinal strain; RALS—right atrial longitudinal strain; CI—confidence interval; OR—odds ratio.

Discussion

In this study, we evaluated the early primary outcomes of NIDCM and whole-heart myocardial mechanics in patients first diagnosed with NIDCM. We found that LVEF and whole-heart myocardial strain parameters were significantly better in the patient group without negative outcomes. However, the LV GLS was the only indep endent predictor of early primary outcomes after one year. Findings from this study confirmed that LV GLS offers additional prognostic insights and may have significant implications for management, clinical treatment decisions and identifying optimal parameters for assessing the clinical status during follow-up.

Traditionally NIDCM research primarily focused on assessing the mechanics and prognostic significance of individual components of the heart. However, this paper recognizes the need for a more thorough approach and represents a significant change from the usual methodology. In an effort to increase our understanding of NIDCM, this study uniquely attempts to evaluate the combined influence of all of the heart chambers on the early prognosis of this disease.

In the area of cardiac imaging, CMR has emerged as a reliable and non-invasive method for assessing myocardial tissue, providing valuable insights into cardiac structure and function. Our study did not find significant results related to T1 mapping or ECV. However, recent interest has shifted towards T1 mapping and quantifying ECV due to their potential to offer valuable insights into the presence of subtle interstitial myocardial fibrosis, diffuse changes to myocardial structure, and altered function [16,17,18]. Recently a growing focus appeared on the independent and incremental prognostic value of CMR FT-derived myocardial mechanics across diverse cardiac conditions [4, 5, 19]. The most attention is focused on LV GLS, showing it as a significant independent predictor for adverse outcomes in NIDCM patients. Recent studies have demonstrated the association of GLS with adverse outcomes [20, 21]. Romano et al. [6] found that the GLS was associated with mortality in both ischemic and non-ischemic dilated cardiomyopathy, independently of the LVEF and late gadolinium enhancement. The recent study by Buss et al. [7] also showed that FT-derived LV GLS was an independent predictor for a combined outcome for cardiac death and heart transplantation. These results emphasize the essential role of LV GLS and its utility in predicting a spectrum of adverse cardiovascular outcomes in patients with non-ischemic dilated cardiomyopathy. Our study results confirmed the importance of GLS.

However, it is important to mention that we evaluated whole-heart myocardial mechanics. To expand our viewpoint the examination of other heart chamber functions becomes crucial. The fact that NIDCM affects the myocardium of both ventricles and might contribute to the onset of RV dysfunction has been noted in previous studies. These studies have shown RV dysfunction as an important prognostic predictor in HF [22] as well as in NIDCM patients [23]. In our study, RV strain parameters were significantly better in the patient group without early primary outcomes, which indicates a potential prognostic value of comprehensive strain analysis. However, our study results did not show independent connections between RV measurements and outcomes.

The LA also plays a central role in cardiac performance. Its functions are intricately linked to LV function and encompass reservoir, conduit, and booster roles during different phases of the cardiac cycle. The assessment of LA longitudinal strain through various techniques, particularly CMR, has recently gained significance and has been shown to be an independent predictor of prognosis in patients with HF [24, 25]. Results of one study have revealed that lower LA reservoir and conduit strain values provided valuable prognostic insights as independent predictors of adverse clinical outcomes [26]. The often overlooked RA function has been recently explored in a study by Yangjie Li et al. The study investigated the significance of impaired RA function in patients with NIDCM [27] and RA reservoir strain, and conduit strain emerged as independent predictors of all-cause mortality even after adjusting for covariates. In our study, the assessment of LA and RA deformation parameters provided valuable insights into cardiovascular outcomes, indicating a complex link between atrial function and early prognosis in patients with severely reduced LV systolic function.

In conclusion, a comprehensive understanding of cardiac function encompasses assessments of various heart chamber strain parameters. Our study contributes to the evolving landscape of NIDCM research by emphasizing the clinical value of assessing the entire heart’s mechanics rather than individual components. We confirmed the CMR-derived LV GLS as a valuable prognostic tool, offering additional insights for clinical management and treatment decisions in patients with NIDCM.

Conclusions

Within this study, LV GLS was found to be an independent and incremental predictor of worse outcome, which exceeded LVEF in patients with optimally treated dilated cardiomyopathy. This indicates the need to routinely include GLS in the CMR follow‐up of dilated cardiomyopathy.