Unlocking the Prognostic Potential of Right Ventricular Strain in Heart Failure Subtypes

Abstract

Objectives:

Right ventricle (RV) function is a recognized independent determinant of morbidity and mortality in heart failure (HF), encompassing various etiologies such as congenital heart disease, pulmonary hypertension, and coronary artery disease. The prognostic significance of RV dysfunction in different HF subtypes – HF with preserved ejection fraction (HFpEF), mid-range EF (HFmrEF), and reduced EF (HFrEF) – continues to gain attention. This study aimed to assess RV dysfunction, quantified by global longitudinal strain (RV GLS), as an early and sensitive predictor of outcomes in HF patients categorized by EF.

Materials and Methods:

A prospective study of 100 HF patients categorized into HFpEF (left ventricular EF [LVEF] >50%, n = 16), HFmrEF (LVEF 40%–50%, n = 47), and HFrEF (LVEF < 40%, n = 37). Echocardiographic assessments, including RV GLS, were performed using a standardized protocol on the Philips EPIC 7C system. RV GLS was analyzed as a sensitive marker for poor outcomes, defined by major adverse cardiac events (MACE) and worsening functional status.

Results:

RV GLS demonstrated significant variation among the groups. Patients in Group C (HFrEF, LVEF < 40%) exhibited markedly lower RV GLS compared to Group A (HFpEF, LVEF > 50%) (−13.5 ± 5.24 vs. −16.8 ± 4.21; P = 0.006). In addition, a significant difference was noted between Group C and Group B (HFmrEF, LVEF 40-50%) (−13.5 ± 5.24 vs. −14.6 ± 4.15; P = 0.047). The predictive value of RV GLS for poor outcomes in HF patients was supported by a sensitivity of 75% and specificity of 69.12%, with a cutoff threshold of − 14.7%.

Conclusions:

RV GLS serves as a dependable marker for predicting poor outcomes in HF patients, with a defined cutoff value of –14.7%. This metric holds substantial clinical promise for effective risk stratification and timely intervention, demonstrating a sensitivity of 75% and a specificity of 69.12%.

INTRODUCTION

Heart failure (HF) is a clinical syndrome caused by structural or functional abnormalities in the heart, typically characterized by elevated natriuretic peptides and signs of pulmonary or systemic congestion.[1] The American Heart Association and the American College of Cardiology define four stages of HF to guide treatment and understand disease progression.[2] Conventionally, the right ventricle (RV) was considered less critical in overall cardiac function; however, recent studies highlight its significant role in predicting morbidity and mortality in HF, pulmonary hypertension, and coronary artery disease.[3] While 2-dimensional echocardiography (2DE) is commonly used to assess RV function, newer imaging techniques such as speckle tracking, tissue Doppler strain, and 3-dimensional echocardiography (3DE) have proven to provide comparable or even superior assessments.[4] Cardiac CT and CMR are also valuable tools, offering precise RV measurements, with CMR considered the gold standard.[5] RV diastolic dysfunction has been linked to poor outcomes and is an independent mortality predictor in chronic HF and pulmonary hypertension.[6] While acute RV pressure overload does not affect diastolic function, chronic overload leads to prolonged relaxation times and increased stiffness, with a biphasic response observed under acute conditions.[7,8]

This study aims to evaluate right ventricular global longitudinal strain (RV GLS) as an early predictor of poor outcomes in HF patients, with a focus on its role across different HF subtypes defined by ejection fraction (EF).

We hypothesize that RV GLS can serve as a reliable predictor of adverse outcomes in HF patients, particularly in those with reduced EF, and may offer a more accurate assessment of RV function than conventional metrics.

MATERIALS AND METHODS

This prospective study enrolled 100 adult patients diagnosed with HF at the Cardiovascular Department, Faculty of Medicine, at our University, from June 2022 to June 2023. The study was approved by the University Institutional Review Board which confirmed that all methods were performed in accordance with the ethical guidelines of the 1975 Declaration of Helsinki as reflected in a priori approval by the institution’s human research. Informed written consent was obtained from all participants.

The sample size was calculated to ensure appropriate statistical power, using a two-sided 95% confidence interval and 80% power. Based on prior studies, the estimated incidence of mortality was 15.8% for HF patients with preserved RV function and 45.4% for those with RV dysfunction. Using these rates, the minimum required sample size was calculated to be 88 patients. With an expected 10% dropout rate, the final sample size was set to 100 patients.

Patients were categorized into three groups based on left ventricular EF (LVEF): Group A (HF with preserved EF [HFpEF], LVEF > 50%, n = 16), Group B (mid-range EF[HFmrEF], LVEF 40%–50%, n = 47), and Group C (reduced EF [HFrEF], LVEF < 40%, n = 37).

Inclusion criteria comprised adult patients with a confirmed diagnosis of HF as per the American Heart Association and European Society of Cardiology guidelines. HF was defined by clinical symptoms, elevated natriuretic peptides (N-terminal pro-B-type Natriuretic Peptide [NT-proBNP] ≥125 pg/mL), and structural cardiac abnormalities such as left ventricular hypertrophy (LV mass index ≥ 115 g/m² in males and ≥ 95 g/m² in females), left atrial enlargement (>34 mL/m²), or diastolic dysfunction (E/e’ ≥13 and mean e’ septal/lateral wall < 9 cm/s). The study included patients with ischemic, nonischemic cardiomyopathies, or other specific etiologies of HF, provided these met the diagnostic criteria. Atrial fibrillation was not an exclusion criterion and was evaluated as a coexisting condition where applicable.

Exclusion criteria were designed to isolate HF patients whose RV function could be evaluated without confounding variables. These included (1) comorbidities directly impacting RV function, such as hypertension, diabetes mellitus, valvular heart disease, or chronic kidney and liver disease, and (2) patients on medications known to alter RV function, such as amphetamines or phenylephrine.

Echocardiographic evaluations and speckle-tracking echocardiography (STE) for strain analysis, were performed during the same session using a Philips EPIC 7C system (USA), equipped with an “S5-1” matrix array transducer. This ensured uniformity and consistency in image acquisition and analysis across all patients. This ensured no time lapse between standard evaluations and strain measurements, thereby eliminating variability due to temporal differences in patient condition. Standardized imaging protocols were strictly followed to minimize variability. This approach eliminates inter-vendor variability and ensures the homogeneity of 2D and speckle-tracking echocardiographic parameters, strengthening the reliability of our findings. Right ventricular size was assessed by 2D echocardiography in optimized apical four-chamber and subcostal views. RV GLS was measured using 2D STE, with images acquired at a frame rate of 60–80 frames per second. The endocardial borders were manually or semi-automatically delineated, and RV GLS was calculated as the average strain across six segments of the RV [Figure 1].[9]

Figure 1.

Right ventricle global longitudinal strain by 2-dimensional speckle tracking echocardiography was −26.6%

In addition, right ventricular fractional area change (RV FAC) was calculated from end-diastolic (RVEDA) and end-systolic (RVESA) areas using the formula: RV FAC = (RVEDA − RVESA)/RVEDA × 100. The clinical severity of HF was assessed according to the New York Heart Association (NYHA) classification [Figure 2].[10]

Figure 2.

Right ventricle fractional area change in both systole and diastole was 24%

Operator blinding was implemented for all echocardiographic measurements. The operators conducting the measurements were blinded to patient outcomes, minimizing potential bias in the data collection for parameters such as RV free wall longitudinal strain (RV FWLS), RV GLS, and RV FAC.[9]

Definition of poor outcomes

Poor outcomes in this study were defined based on the occurrence of major adverse cardiac events (MACE), including myocardial infarction, HF hospitalization, stroke, and cardiovascular death, as well as a composite measure of worsening functional status as determined by the NYHA classification. Patients were followed for 6 months, with clinical visits conducted every 3 months, and events were verified using medical records and adjudicated by an independent committee. Although the MACE rates were similar across HF subtypes, the study aimed to evaluate RV GLS as an early, sensitive marker for predicting adverse prognostic trends, irrespective of overt event rates during the follow-up period.

Statistical analysis

All statistical analyses were performed using SPSS version 26 (IBM Corp., Armonk, N.Y., USA) and Medcalc software. Descriptive statistics were calculated, and comparisons between groups were made using one-way ANOVA and the Chi-square test. The performance of RV GLS as a predictor of outcomes was assessed using receiver operating characteristic (ROC) curves.

RESULTS

There were no statistically significant differences in age (P = 0.720 and 0.845) and sex (P1 = 0.271 and P2 = 0.995) between the studied groups [Table 1].

Table 1.

Demographic data of the studied groups

Variables	Group A (n=16)	Group B (n=47)	Group C (n=37)	F	P1	P2
Age (years)
Mean±SD	55.3±11.22	57±12.07	56.5±11.11	0.120	0.720	0.845
Range	34–77	37–78	38–82
Sex, n (%)
Male	8 (50)	33 (70.21)	26 (70.27)	2.49	0.271	0.995
Female	8 (50)	14 (29.79)	11 (29.73)

F=ANOVA test, P1=P value between Group C and Group A, P2=P value between Group C and Group B, SD=Standard deviation

In addition, no significant differences were observed in heart rate, systolic blood pressure, diastolic blood pressure, or body mass index (BMI) across the groups [Table 2].

Table 2.

Vital signs of the studied groups

Variables	Group A (n=16)	Group B (n=47)	Group C (n=37)	F	P1	P2
Heart rate (beats/min)
Mean±SD	92.8±12.52	89.6±12.02	89.4±11.81	0.493	0.349	0.939
Range	70–112	72–112	70–109
Systolic blood pressure (mmHg)
Mean±SD	118.1±7.93	116.1±7.8	119.7±8.81	2.08	0.535	0.051
Range	105–130	100–130	100–130
Diastolic blood pressure (mmHg)
Mean±SD	75±5.16	75.5±4.33	74.3±5.29	0.645	0.657	0.256
Range	70–85	70–85	60–85
BMI (kg/m2)
Mean±SD	27.4±2.72	27.5±3.58	28.1±3.09	0.376	0.437	0.478
Range	23.12–32.21	20.53–34.16	22.04–33.95

F=ANOVA test, P1=P value between Group C and Group A, P2=P value between Group C and Group B, BMI=Body mass index, SD=Standard deviation

A highly significant difference in LVEF was found between Group C (HFrEF) and Group A (HFpEF) (P1 < 0.001), with a decrease in LVEF in Group C (35.9 ± 2.39) compared to Group A [Table 3]. Similarly, a significant difference was observed between Group C and Group B (HFmrEF) (P2 < 0.001), with a decrease in LVEF in Group C compared to Group B (35.9 ± 2.39) [Table 3].

Table 3.

Echocardiogram of the studied groups

Variables	Group A (n=16)	Group B (n=47)	Group C (n=37)	F	P1	P2
LV EF (%)
Mean±SD	54.2±4.4	43.7±2.98	35.9±2.39	208.3	<0.001	<0.001
Range	50–66	35–49	32–40
RV FAC (%)
Mean±SD	33.6±8.59	36.2±8.61	34.1±10.01	0.786	0.968	0.609
Range	21–52	18–57	17.4–54
TAPSE (mm)
Mean±SD	19.8±3.1	18.4±4.72	17.6±4.22	1.45	0.141	0.547
Range	14–25	0.9–28	9–27
RV FWLS (%)
Mean±SD	−20.7±8.55	−18.8±9.67	−15.9±7.61	2.05	0.168	0.248
Range	−32.6–−1.3	−36.2–−0.1	−31.7–−0.4
RV GLS (%)
Mean±SD	−18.1±5.46	−17.5±7.34	−13.5±5.24	6.04	0.006	0.047
Range	−25.3–−5.3	−28.3–−1.1	−21.9–−2.3
ePASP (mmHg)
Mean±SD	29.8±4.1	29.8±3.55	30.2±3.66	0.098	0.957	0.917
Range	24–38	20–36	22–39
RVD-mid cavity (mm)
Mean±SD	2.6±0.39	2.6±0.5	2.8±0.56	1.643	0.542	0.206
Range	2.1–3.5	1.4–3.4	1.7–4
RVD-basal (mm)
Mean±SD	3.3±0.44	3.4±0.54	3.5±0.7	1.02	0.519	0.681
Range	2.4–3.8	2.3–4.8	2.1–5.3
Base-apex (mm)
Mean±SD	6.5±0.7	6.1±0.64	6.4±0.7	3.52	0.537	0.151
Range	5.1–7.7	4.6–7.6	5.2–8.4

P1=P value between Group C and Group A, P2=P value between Group C and Group B, F=ANOVA test, LV EF=Left ventricle ejection fraction, RV FAC=Right ventricle fractional area of change, TAPSE=Tricuspid annular plane systolic excursion, RV FWLS=Right ventricle free-wall longitudinal strain, RV GLS=Right ventricular global longitudinal strain, ePASP=Pulmonary artery systolic pressure estimated by echocardiography, RVD=Right ventricular systolic dysfunction, ANOVA=Analysis of variance

No statistically significant differences were found in RV FAC, tricuspid annular plane systolic excursion (TAPSE), or RV FWLS across the groups. However, a highly significant difference was noted in RV GLS between Group C and Group A (P1 = 0.006), with an increase in RV GLS in Group C compared to Group A (−13.5 ± 5.24) [Table 3]. A significant difference was also observed between Group C and Group B (P2 = 0.047), with Group C showing higher RV GLS compared to Group B (−13.5 ± 5.24) [Table 3].

There were no significant differences in estimated pulmonary artery systolic pressure (ePASP), RV dilatation at mid-cavity, basal cavity, or base-apex dimensions across the groups. This finding emphasizes the utility of RV GLS as a more sensitive marker of RV dysfunction compared to conventional metrics. The formula used for ePASP estimation was derived from the tricuspid regurgitation jet velocity (TRV) and right atrial pressure (RAP), with RAP estimated from inferior vena cava (IVC) dimensions and collapsibility: ePASP = 4(TRV) ² + RAP. IVC dimensions and collapsibility were assessed to refine the estimation of RAP, an essential factor in cases of HF with RV dysfunction. These parameters provide valuable insights into the hemodynamic status of patients and may complement GLS measurements.

There were no significant differences in ePASP, right ventricular dilatation at mid-cavity (RVD-mid cavity), basal cavity (RVD-basal), or base-apex dimensions across the groups [Table 3].

Furthermore, no significant differences were observed in the NYHA classification, MACE, or mortality rates between the groups [Table 4].

Table 4.

New York heart association classification and outcome of the studied groups

Variables	Group A (n=16), n (%)	Group B (n=47), n (%)	Group C (n=37), n (%)	χ 2	P1	P2
NYHA classification
II	5 (31.25)	27 (57.45)	17 (45.95)	6.435	0.191	0.302
III	10 (62.5)	17 (36.17)	13 (35.14)
IV	1 (6.25)	3 (6.38)	6 (16.22)
MACE	5 (31.25)	11 (23.4)	16 (43.24)	3.749	0.607	0.089
Mortality	2 (12.5)	4 (8.51)	3 (8.11)	0.289	0.615	0.947

χ²=Chi-square test, P1=P value between Group C and Group A, P2=P value between Group C and Group, B NYHA=New York Heart Association, MACE=Major adverse cardiac event

In additional analyses [Tables 5-7], no significant differences were found in TAPSE, RV FAC, RV GLS, or RV FWLS across the studied groups.

Table 5.

Relation pulmonary artery systolic pressure estimated by echocardiography and (tricuspid annular plane systolic excursion, right ventricle fractional area of change, right ventricle global longitudinal strain and RV free-wall longitudinal strain) in Group A

Variables	ePASP <25 (n=3)	ePASP ≥25 (n=13)	t-test	P
TAPSE (mm)
Mean±SD	20.67±2.08	19.62±3.33	0.52	0.614
Range	19–23	14–25
RV FAC
Mean±SD	33.67±15.89	33.54±7.08	0.02	0.982
Range	24–52	21–47
RVGLS
Mean±SD	−19.83±3.86	−19.13±6.09	0.019	0.852
Range	−23.5–−15.8	−25.3–−5.3
RVFWLS
Mean±SD	−21.83±8.18	−20.49±8.93	0.24	0.815
Range	−27–−12.4	−32.6–−1.3

TAPSE=Tricuspid annular plane systolic excursion, RV FAC=Right ventricle fractional area of change, RV GLS=Right ventricular global longitudinal strain, RV FWLS=Right ventricle free-wall longitudinal strain, ePASP=Pulmonary artery systolic pressure estimated by echocardiography, SD=Standard deviation

Table 7.

Relation pulmonary artery systolic pressure estimated by echocardiography and (tricuspid annular plane systolic excursion, right ventricle fractional area of change, right ventricular global longitudinal strain and RV free-wall longitudinal strain) in Group C

Variables	ePASP <25 (n=2)	ePASP ≥25 (n=35)	t-test	P
TAPSE (mm)
Mean±SD	19±4.24	17.52±4.27	0.48	0.637
Range	16–22	9–27
RV FAC (%)
Mean±SD	40.5±0.71	33.71±10.18	0.93	0.358
Range	40–41	17.4–54
RV GLS (%)
Mean±SD	−9.95±2.19	−13.31±5.43	0.86	0.395
Range	−11.5–−8.4	−21.9–−2.3
RV FWLS (%)
Mean±SD	−16.3±9.05	−15.87±7.68	0.08	0.940
Range	−22.7–−9.9	−31.7–−0.4

TAPSE=Tricuspid annular plane systolic excursion, RV FAC=Right ventricle fractional area of change, RV GLS=Right ventricular global longitudinal strain, RV FWLS=Right ventricle free-wall longitudinal strain, ePASP=Pulmonary artery systolic pressure estimated by echocardiography, SD=Standard deviation

Table 6.

Relation pulmonary artery systolic pressure estimated by echocardiography and (tricuspid annular plane systolic excursion, right ventricle fractional area of change, right ventricular global longitudinal strain and RV free-wall longitudinal strain) in Group B

Variables	ePASP <25 (n=4)	ePASP ≥25 (n=43)	t-test	P
TAPSE (mm)
Mean±SD	21.75±3.86	18.04±4.7	1.53	0.134
Range	18–27	0.9–28
RV FAC (%)
Mean±SD	33±10.13	36.49±8.53	0.77	0.444
Range	26–48	18–57
RV GLS (%)
Mean±SD	−10.13±4.55	−16.85±6.86	1.91	0.062
Range	−14.5–−5.4	−27.1–−1.1
RV FWLS (%)
Mean±SD	−13±5.08	−19.38±9.85	1.27	0.210
Range	−18.3–−6.1	−36.2–−0.1

TAPSE=Tricuspid annular plane systolic excursion, RV FAC=Right ventricle fractional area of change, RV GLS=Right ventricular global longitudinal strain, RV FWLS=Right ventricle free-wall longitudinal strain, ePASP=Pulmonary artery systolic pressure estimated by echocardiography, SD=Standard deviation

Finally, the sensitivity and specificity of RV GLS as a predictor of poor outcomes in HF patients were 75% and 69.12%, respectively, with a cutoff value of − 14.7% [Table 8 and Figure 3].

Table 8.

Role of right ventricular global longitudinal strain in prediction of poor outcome among heart failure patients

Cutoff	Sensitivity (%)	Specificity (%)	PPV (%)	NPV (%)	AUC	P
>−14.7	75	69.12	53.3	85.5	0.819	<0.001

PPV=Positive predictive value, NPV=Negative predictive value, AUC=Area under the curve

Figure 3.

Receiver operating characteristic curve of right ventricle global longitudinal strain in the prediction of poor outcome among heart failure patients

DISCUSSION

In this study, no statistically significant differences were observed regarding age, sex, heart rate, and systolic and diastolic blood pressure, or BMI, between the studied groups. This can be attributed to the cross-matching of patients. Similarly, Delepaul et al.[11] conducted a study involving 482 HF patients (258 with HFrEF, 115 with HFmrEF, and 109 with HFpEF) and found no significant differences in BMI across groups. Bonsu et al.,[12] in a study of 1,488 HF patients (345 with HFrEF, 265 with HFmrEF, and 878 with HFpEF), reported similar findings, with no significant differences in age, sex, BMI, heart rate, or blood pressure between the groups.

In contrast, Tsuji et al.[13] conducted a study on 3,480 HF patients with echocardiographic data and found that HFpEF patients had significantly higher age, female sex predominance, BMI, systolic and diastolic blood pressure, and lower heart rate compared to the HFmrEF and HFrEF groups. These discrepancies may stem from differences in study design, sample size, or population characteristics.

Regarding LVEF, our study found a significant decrease in LVEF in Group C (HFrEF) compared to Group A (HFpEF) and Group B (HFmrEF). In contrast, no significant differences were found in RV FAC, TAPSE, RV FWLS, ePASP, or RVD dimensions between the groups. Our findings are consistent with those of Lejeune et al.,[14] who reported a negative correlation between RV GLS and RV EF. This is also supported by Bonsu et al.,[12] Tsuji et al.,[13] and Delepaul et al.,[11] who showed that LVEF was significantly lower in HFrEF patients compared to HFmrEF and HFpEF groups. In addition, Motoki et al.[15] found that worse RV GLS was associated with reduced LVEF, which may reflect more severe myocardial injury in the HFrEF group.

Avitabile et al.[16] showed no association between TAPSE and RVEF, while Anastasiou et al.[17] found that HF patients with LVEF < 45% exhibited worsening in RV FWLS, which may explain some of the variations observed in RV function between the groups in our study.

Regarding NYHA classification, major adverse cardiac events (MACE), and mortality, our study found no significant differences across the groups. This is in line with the findings of Ishiguchi et al.,[18] who studied 142 HF patients and found comparable rates of MACE between HFpEF and systolic HF patients. Similarly, Lyu et al.[19] confirmed that MACE was similar in HFmrEF and HFpEF groups, although HFmrEF showed a better prognosis compared to HFrEF. Rickenbacher et al.[20] also reported no significant differences in mortality or NYHA class ≥ III between HFmrEF, HFrEF, and HFpEF groups. Supporting our results, Delepaul et al.[11] found no significant differences in mean NYHA class or mortality between these groups.

Conversely, Bonsu et al.[12] reported that HFrEF patients were more symptomatic compared to HFmrEF and HFpEF patients, which may be due to differences in sample size or geographical population. Tsuji et al.[13] observed significant differences in NYHA class among the HF subtypes, with HFpEF patients having a better functional status compared to HFrEF and HFmrEF patients.

In relation to ePASP and its correlation with RV parameters, our study found no significant differences between ePASP < 25 and ePASP ≥ 25 groups regarding TAPSE, RV FAC, RV GLS, or RV FWLS. Meng et al.[21] reported a weak association between 3D-RVFWLS and 2D-RVFWLS with PASP, while Zhang et al.[22] found no association between TAPSE and PASP. In contrast, Vymetal et al.[23] reported significant associations between PASP and RV GLS, as well as between PASP and FAC.

Regarding RV GLS as a predictor of poor outcomes, our study found that RV GLS had a sensitivity of 75% and specificity of 69.12%, with a cutoff value of − 14.7%. This result is consistent with studies by Gaggin et al.,[24] who found that NT-proBNP was a strong predictor of clinical outcomes in HF patients, with a cutoff of < 900 pg/mL showing high sensitivity (90%) and specificity (85%). Similarly, Yılmaz Öztekin et al.[25] found elevated NT-proBNP levels to be associated with adverse outcomes in chronic HF patients, supporting our findings.

In line with our results, Salah et al.[26] demonstrated that discharge NT-proBNP levels predicted outcomes in both HFpEF and HFrEF patients, and Linssen et al.[27] showed that NT-proBNP at discharge was a powerful, independent predictor of all-cause mortality. Kang et al.[28] found high NT-proBNP levels to be significantly associated with poor outcomes in HF patients, while Bettencourt et al.[29] reported that NT-proBNP variations correlated with hospital readmission and death within 6 months.

Furthermore, Bay et al.[30] found that NT-proBNP was a predictor of reduced LVEF at a cutoff value of 357 pmol/L, with a sensitivity of 73%, specificity of 82%, and a high negative predictive value (98%). These findings further reinforce the predictive value of NT-proBNP in assessing HF prognosis.

RV GLS and NT-proBNP are both established predictors of poor outcomes in HF, yet they provide complementary insights into disease pathophysiology. NT-proBNP reflects neurohormonal activation and ventricular wall stress, serving as a global biomarker of HF severity. In contrast, RV GLS specifically quantifies myocardial deformation, offering a direct assessment of RV function and mechanical performance.

The prognostic superiority of RV GLS lies in its ability to detect subclinical myocardial dysfunction, particularly in patients with preserved or moderately reduced EF (HFpEF and HFmrEF), where NT-proBNP levels may not always correlate with disease progression. Lejeune et al.[14] demonstrated that RV GLS independently predicts mortality and HF rehospitalization in HFpEF, underscoring its utility in this subgroup. Similarly, our study found RV GLS to be a sensitive marker of poor outcomes with a cutoff of − 14.7%, providing prognostic value even in the absence of significant differences in NT-proBNP levels across groups.

While NT-proBNP remains an invaluable tool for guiding HF management, its levels can be influenced by factors such as age, renal function, and obesity, potentially limiting its specificity. RV GLS, being less susceptible to these confounders, offers a direct measure of RV mechanical dysfunction. The integration of RV GLS with NT-proBNP could enhance risk stratification by combining structural and biomarker-driven approaches, as proposed by Iacoviello et al.[31] Future studies should explore synergistic models incorporating both metrics to refine prognostication in HF [Figure 4].

Figure 4.

Graphical abstract

Previous studies, such as Dalen et al.,[2] have reported normal GLS values ranging from − 18% to − 22% in healthy individuals, varying by age and sex. These benchmarks are critical for contextualizing our findings and establishing the prognostic cutoff of − 14.7% observed in our HF cohort.

Although this study focuses on RV GLS, the role of 3DTTE EF in assessing RV function and prognosis in HF has been well-established. Nagata et al.[32] demonstrated that 3DTTE EF provides superior volumetric accuracy and predictive value compared to 2D methods. Incorporating 3DTTE alongside GLS may enhance RV function assessment and risk stratification.

Incorporating IVC dimensions and collapsibility into the estimation of ePASP is critical in evaluating RV dysfunction, particularly in HF. These measures provide a noninvasive assessment of RAP, which plays a pivotal role in determining pulmonary artery systolic pressure. A recent review by Fusini et al.[33] underscores the ongoing importance of these noninvasive techniques, emphasizing their clinical relevance in estimating right atrial, RV, and pulmonary pressures. Fusini et al. also highlight the integration of advanced imaging metrics, such as GLS, which may further enhance the diagnostic and prognostic capabilities of echocardiographic assessments.

The differences in RV GLS observed across the HF subgroups (HFpEF, HFmrEF, and HFrEF) can be attributed to distinct pathophysiological mechanisms affecting the right ventricle. In HFpEF, the preserved LVEF and relatively lower RV afterload may contribute to better RV function and more favorable GLS values. The observed RV dysfunction in HFpEF may reflect subtle, early alterations in myocardial mechanics due to increased left atrial pressure and pulmonary venous congestion, even in the absence of overt pulmonary hypertension.

In HFmrEF, the intermediate LVEF suggests a transitional state between HFpEF and HFrEF, with moderate RV impairment. The decline in RV GLS in this group may result from progressive RV myocardial injury and increased wall stress, secondary to the combined effects of volume and pressure overload.

In HFrEF, the significantly impaired LVEF and elevated pulmonary pressures impose a greater hemodynamic burden on the RV, leading to increased RV dilatation, wall stress, and myocardial fibrosis. These factors likely explain the most pronounced reduction in RV GLS observed in this group. Moreover, RV-arterial uncoupling, which occurs due to chronic afterload mismatch, further exacerbates RV dysfunction and contributes to worse outcomes.

These findings highlight the importance of RV GLS as an integrative marker that reflects the interplay between RV mechanics and systemic and pulmonary vascular load. The progressive decline in RV GLS across HF subtypes underscores its potential as a sensitive indicator of RV dysfunction and a prognostic tool for stratifying risk in HF patients.

The similar MACE rates observed across groups highlight the challenge of relying solely on overt clinical events for risk stratification in HF patients. GLS of the right ventricle, as a subclinical marker of RV dysfunction, provides valuable prognostic insights beyond traditional metrics. RV GLS may identify subtle myocardial impairments that precede clinical deterioration, offering an opportunity for early intervention before overt MACE. This underscores its utility as an integrative biomarker of ventricular function and hemodynamic stress.

Study limitations

There are several limitations to our study that should be acknowledged. First, this was a single-center study with a relatively small sample size, which may limit the generalizability of the findings. Future multicenter studies with larger sample sizes are needed to validate the role of RV GLS in predicting outcomes in HF patients. In addition, our study was observational, meaning we cannot draw definitive conclusions about causality. Longitudinal studies with follow-up data are needed to determine whether RV GLS can serve as an independent predictor of long-term outcomes in HF.

Another limitation is that we did not assess the impact of RV GLS on other clinical outcomes, such as quality of life or exercise capacity. Future studies should explore these aspects to provide a more comprehensive understanding of how RV GLS correlates with overall patient well-being.

CONCLUSIONS

RV GLS is a reliable predictor of poor outcomes in HF patients, with a cutoff value of − 14.7%. At this threshold, RV GLS demonstrates a sensitivity of 75% and specificity of 69.12%, highlighting its potential as an important marker for risk stratification in HF management.

Conflicts of interest

There are no conflicts of interest.

Funding Statement

Nil.