Abstract
Introduction: Valvular heart disease is one of the most common heart diseases. It is characterized by abnormal function or structure of the heart valves. There may be no clinical symptoms in the early stages. Clinical symptoms of arrhythmia, heart failure, or thromboembolic events may occur in the late stages of the disease, such as palpitation after activities, breathing difficulties, fatigue, and so on. Aortic valve disease is a major part of valvular heart disease. The main treatment for aortic valve disease is valve replacement or repair surgery, but it is extremely risky. Therefore, a rigorous prognostic assessment is extremely important for patients with aortic valve disease. The global longitudinal strain is an index that describes the deformation capacity of myocardium. There is evidence that it provides a test for systolic dysfunction other than LVEF (left ventricular ejection fraction) and provides additional prognostic information. Method: Search literature published between 2010 and 2023 on relevant platforms and contain the following keywords: “Aortic valve disease,” “Aortic stenosis,” “Aortic regurgitation,” and “longitudinal strain” or “strain.” The data is then extracted and collated for analysis. Results: A total of 15 articles were included. The total population involved in this study was 3,678 individuals. The absolute value of LVGLS was higher in the no-MACE group than in the MACE group in patients with aortic stenosis (Z = 8.10, p < 0.00001), and impaired LVGLS was a risk factor for MACE in patients with aortic stenosis (HR = 1.14, p < 0.00001, 95% CI: 1.08–1.20). There was also a correlation between impaired LVGLS and aortic valve surgery in patients with aortic valve disease (HR = 1.16, p < 0.0001, 95% CI: 1.08–1.25) or patients with aortic valve regurgitation (HR = 1.21, p = 0.0004, 95% CI: 1.09–1.34). We also found that impaired LVGLS had no significant association between LVGLS and mortality during the period of follow-up in patients with aortic valve stenosis (HR = 1.08, 95% CI: 0.94–1.25, p = 0.28), but it was associated with mortality in studies of prospective analyses (HR = 1.34, 95% CI: 1.02–1.75, p = 0.04). Conclusions: Impaired LVGLS correlates with major adverse cardiovascular events in patients with aortic valve disease, and it has predictive value for the prognosis of patients with aortic valve disease.
Introduction
Heart valve disease is one of the most common heart diseases. Its incidence increases with age. Aortic valve disease accounts for a large part of it, mainly divided into aortic regurgitation and aortic stenosis. Statistical analysis showed that 3% of people over 65 years of age were affected by aortic stenosis, and the course of disease accelerated after the onset of symptoms in patients with aortic stenosis [1]. The estimated prevalence of aortic regurgitation was 4.9% and 0.5% for moderate-to-severe regurgitation [2]. The incidence of aortic stenosis has also been previously reported to be 0.2% at age over 50, 3% at age over 65, and 9.8% among people over 80 years old [1, 3]. The prevalence of aortic regurgitation is 4.9% [4]. Aortic regurgitation or stenosis leads to overload of the left ventricular volume or pressure, which can cause damage to the function of the left ventricle and may eventually lead to heart failure. Current studies suggest that left ventricular dysfunction in these patients is associated with ventricular remodeling and hypertrophy due to valvular disease [5]. The main treatment for aortic stenosis or regurgitation is the replacement or repair of the aortic valve. Both procedures are high-risk and difficult; a meta-analysis that included 1,071 patients reported that in-hospital mortality rates for aortic valve repair and replacement were 1.3% and 3.6%, respectively [6]. Therefore, it is important to properly evaluate patients with aortic valve disease. LVEF (left ventricular ejection fraction) has been used as a reliable indicator of left ventricular function in the evaluation of valvular disease, but its measurement depends only on the difference between end-diastolic volume (EDV) and end-systolic volume (ESV), calculated from the position of the ventricular wall at the end of systole and diastole [7]. It does not focus on the activity of the ventricular wall during myocardial motion. There is evidence that LVGLS may be a prognostic indicator of valvular heart disease [8‒11]. In view of this, we conducted this study to assess the prognostic predictive value of LVGLS in aortic valve disease.
Methods
Relevant literature was searched by PubMed, ScienceDirect, and SCIE databases for the following keywords: aortic stenosis, aortic regurgitation, longitudinal strain, strain, GLS. Select and organize the required data, and use the software Review Manager 5.4 to analyze the data. The results are shown in a forest plot.
Inclusion Criteria
- 1.
Left ventricular global longitudinal strain was evaluated by cardiac ultrasound.
- 2.
Articles published from 2010 to 2023.
- 3.
Mean or median follow-up time >6 months.
- 4.
Data of the article is reported with a high degree of credibility and no apparent conflict of interest.
Exclusion Criteria
- 1.
The article does not report the data we need.
- 2.
Case reports, comments, abstracts, and editorials.
- 3.
The reported data is not credible.
Data Extraction and Quality Assessment
Three researchers independently assessed the quality of these articles and extracted relevant data. Since LVGLS is negative, we extract its absolute value for statistical analysis. When some data is not reported directly in the original text but can be calculated from data already reported, the calculated values are extracted. When the data for the full sample is not reported in the original text, but for each subgroup, the mean ± SD is derived by calculating the data for the subgroups or extracted directly from one of the subgroups when only one subgroup reports the data we need. The risk of bias assessment of articles was performed using the Cochrane risk of bias assessment tool.
Statistical Analysis
Review Manager 5.4 software was used to analyze data from the 15 papers selected. The heterogeneity test was performed before the size of the combined effect was calculated, and when there was heterogeneity in the results, a random-effect model was used to calculate the combined values; otherwise a fixed-effects model was used to analyze them. We defined p < 0.05 as statistically significant. The results are shown in the forest plot.
Result
Study Selection
A total of 674 articles published between 2010 and 2023 were retrieved by keyword search, and a total of 659 documents were excluded because they were not relevant to our research project, did not report the data we needed, were not credible, duplicated, or had obvious financial interests. 15 articles were eventually included (Fig. 1).
Characteristics of Selected Studies
A total of 15 papers and 3,678 participants were included. Table 1 shows a summary of the included studies. All studies were published between 2010 and 2023, four of which were multicenter. Seven of these papers are prospective studies. Twelve articles had a study population of patients with aortic stenosis and three with aortic regurgitation among the included articles (Table 1).
Characteristics of selected studies
| Author | Year of publication | Type of study | n | Study population | Mean or median of follow-up | Male | Age | Hypertension | Diabetes | Mean LVEF | Mean LVGLS |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Nagata et al. [12] | 2015 | Prospective/monocentric | 102 | Patients with aortic stenosis | 373 days | 43 | 78±10 | 67 | 21 | 60±5 | 15.8±3.4 |
| Gu et al. [13] | 2018 | Retrospective/multicentric | 218 | Patients with aortic stenosis | 33.4 weeks | 106 | 68.6±14* | 172* | 39* | 65.34±8.20 | 16.27±4.38 |
| Dahl et al. [14] | 2012 | Prospective/monocentric | 125 | Patients with aortic stenosis | 3.8±1.5 years | 79 | 72.22±9.33* | 53 | 11 | 54±7.48 | 15.5±3.7 |
| Kearney et al. [15] | 2012 | Prospective/monocentric | 146 | Patients with aortic stenosis | 2.1 years | 91 | 75+11 | 116 | 38 | 59 + 11 | 15 + 4 |
| Carstensen et al. [16] | 2015 | Prospective/multicentric | 104 | Patients with aortic stenosis | 2.3 years | 71 | 72±9 | 71 | 13 | 59 | 15.1±2.6 |
| Lancellotti et al. [17] | 2012 | Prospective/bicentric | 163 | Patients with aortic stenosis | 20±19 months | 106 | 70±10 | 81 | 27 | 66±9 | 15.7±3.1 |
| Sato et al. [18] | 2014 | Retrospective/multicentric | 98 | Patients with aortic stenosis | 399 days | 72 | 77.23±8.02* | 155 | 75 | 62.79±8.15* | 18.45±3.80* |
| Thellier et al. [19] | 2020 | Prospective/bicentric | 332 | Patients with aortic stenosis | 42 months | 136 | 79 [71; 85] | 235 | 118 | 61 [57; 66] | 14.1 [11.0; 16.7] |
| Dahl et al. [20] | 2016 | Prospective, monocentric | 121 | Patients with aortic stenosis | 5 years | 77 | 72.51±9.13* | 53 | 18 | 54.49±7.13 | 15.48±3.73 |
| Stassen et al. [21] | 2022 | Retrospective/monocentric | 760 | Patients with aortic stenosis | 50 months | 462 | 71.3 6 +11.8 | 539 | 180 | 57.5±10.9 | 15.3±3.9 |
| Lee et al. [22] | 2013 | Retrospective/monocentric | 31 | Patients with aortic stenosis | 2 years | 31 | 72.91±14.53* | 16 | 10 | 71.19±8.75 | − |
| Fries et al. [23] | 2017 | Retrospective/monocentric | 188 | Patients with aortic stenosis | 33 weeks | 85 | 81.4±7.8 | 139 | 75 | 54.2±14.7 | 12.1±4.7 |
| Alashi et al. [24] | 2017 | Retrospective/monocentric | 1,063 | Patients with aortic regurgitation | 6.8±3.0 years | 813 | 53±16 | 589 | 57 | 57±4 | 19.5±0.2 |
| Ewe et al. [25] | 2014 | Retrospective/monocentric | 68 | Patients with aortic regurgitation | 4.2±3.2 years | 40 | 54+17 | 40 | 4 | 61 + 5 | 19.3+2.8 |
| Kusunose et al. [26] | 2014 | Retrospective/monocentric | 159 | Patients with aortic regurgitation | 30±21 months | 124 | 50±15 | 90 | 14 | 58±5 | − |
| Author | Year of publication | Type of study | n | Study population | Mean or median of follow-up | Male | Age | Hypertension | Diabetes | Mean LVEF | Mean LVGLS |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Nagata et al. [12] | 2015 | Prospective/monocentric | 102 | Patients with aortic stenosis | 373 days | 43 | 78±10 | 67 | 21 | 60±5 | 15.8±3.4 |
| Gu et al. [13] | 2018 | Retrospective/multicentric | 218 | Patients with aortic stenosis | 33.4 weeks | 106 | 68.6±14* | 172* | 39* | 65.34±8.20 | 16.27±4.38 |
| Dahl et al. [14] | 2012 | Prospective/monocentric | 125 | Patients with aortic stenosis | 3.8±1.5 years | 79 | 72.22±9.33* | 53 | 11 | 54±7.48 | 15.5±3.7 |
| Kearney et al. [15] | 2012 | Prospective/monocentric | 146 | Patients with aortic stenosis | 2.1 years | 91 | 75+11 | 116 | 38 | 59 + 11 | 15 + 4 |
| Carstensen et al. [16] | 2015 | Prospective/multicentric | 104 | Patients with aortic stenosis | 2.3 years | 71 | 72±9 | 71 | 13 | 59 | 15.1±2.6 |
| Lancellotti et al. [17] | 2012 | Prospective/bicentric | 163 | Patients with aortic stenosis | 20±19 months | 106 | 70±10 | 81 | 27 | 66±9 | 15.7±3.1 |
| Sato et al. [18] | 2014 | Retrospective/multicentric | 98 | Patients with aortic stenosis | 399 days | 72 | 77.23±8.02* | 155 | 75 | 62.79±8.15* | 18.45±3.80* |
| Thellier et al. [19] | 2020 | Prospective/bicentric | 332 | Patients with aortic stenosis | 42 months | 136 | 79 [71; 85] | 235 | 118 | 61 [57; 66] | 14.1 [11.0; 16.7] |
| Dahl et al. [20] | 2016 | Prospective, monocentric | 121 | Patients with aortic stenosis | 5 years | 77 | 72.51±9.13* | 53 | 18 | 54.49±7.13 | 15.48±3.73 |
| Stassen et al. [21] | 2022 | Retrospective/monocentric | 760 | Patients with aortic stenosis | 50 months | 462 | 71.3 6 +11.8 | 539 | 180 | 57.5±10.9 | 15.3±3.9 |
| Lee et al. [22] | 2013 | Retrospective/monocentric | 31 | Patients with aortic stenosis | 2 years | 31 | 72.91±14.53* | 16 | 10 | 71.19±8.75 | − |
| Fries et al. [23] | 2017 | Retrospective/monocentric | 188 | Patients with aortic stenosis | 33 weeks | 85 | 81.4±7.8 | 139 | 75 | 54.2±14.7 | 12.1±4.7 |
| Alashi et al. [24] | 2017 | Retrospective/monocentric | 1,063 | Patients with aortic regurgitation | 6.8±3.0 years | 813 | 53±16 | 589 | 57 | 57±4 | 19.5±0.2 |
| Ewe et al. [25] | 2014 | Retrospective/monocentric | 68 | Patients with aortic regurgitation | 4.2±3.2 years | 40 | 54+17 | 40 | 4 | 61 + 5 | 19.3+2.8 |
| Kusunose et al. [26] | 2014 | Retrospective/monocentric | 159 | Patients with aortic regurgitation | 30±21 months | 124 | 50±15 | 90 | 14 | 58±5 | − |
Risk of Bias
The risk of bias was assessed by the Cochrance risk assessment tool, which showed an unclear risk of blinding and completeness of data (Fig. 2).
Differences in LVGLS between the MACE and No-MACE Groups
A total of 7 studies reported LVGLS values in the MACE and no-MACE groups. Seven papers were studied in patients with aortic stenosis. Heterogeneity test shows that there was no significant heterogeneity, I2 = 28% (Fig. 3, 4), and using a fixed-effects model for analysis, combined effect sizes demonstrate higher absolute LVGLS values in the no-MACE group than in the MACE group in patients with aortic stenosis (Z = 8.10, p < 0.00001) (Fig. 4).
Correlation between LVGLS and MACE
A total of 6 papers reported correlations between LVGLS and MACE in patients with aortic stenosis, and a test for heterogeneity suggested heterogeneity in these studies, so a random-effects model was used, and the combined effect size suggested a relevance between impaired LVGLS and MACE (HR = 1.14, CI 95%: 1.08–1.20, p < 0.00001) (Fig. 5).
LVGLS and Aortic Valve Surgery
A total of 3 articles studied the correlation between LVGLS and aortic valve surgery and reported HR values. Two of the articles studied populations of patients with aortic regurgitation, and one article studied the patients with aortic stenosis. Tests for heterogeneity show that there is no significant heterogeneity in 3 articles and 2 articles about patients with aortic regurgitation, a fixed-effects model was used for analysis, and the combined effect values showed that impaired LVGLS was associated with aortic valve surgery in patients with aortic valve disease (HR = 1.16, 95% CI: 1.08–1.25, p < 0.0001) and patients with aortic valve regurgitation (HR = 1.21, p = 0.0004, 95% CI: 1.09–1.34) (Fig. 6).
Association between LVGLS and aortic valve surgery in patients with aortic valve disease (a) and patients with aortic valve regurgitation (b).
Association between LVGLS and aortic valve surgery in patients with aortic valve disease (a) and patients with aortic valve regurgitation (b).
LVGLS and Mortality
A total of 5 articles reported a correlation between follow-up mortality and impaired LVGLS. Five papers were studied in patients with aortic stenosis. A heterogeneity test suggested that these studies were heterogeneous, so a random-effects model was used for analysis, and the results suggested there was no significant association between LVGLS and mortality during the period of follow-up in patients with aortic valve stenosis (HR = 1.08 95% CI: 0.94–1.25, p = 0.28), but it was associated with mortality in studies of prospective analyses (HR = 1.34, 95% CI: 1.02–1.75, p = 0.04) (Fig. 7).
Association between LVGLS and mortality in patients with aortic stenosis (a) and in subgroup of prospectively analyzed study (b).
Association between LVGLS and mortality in patients with aortic stenosis (a) and in subgroup of prospectively analyzed study (b).
Discussion
The analysis of the 15 included articles, which included a total of 3,678 patients, roughly showed the correlation between impaired LVGLS and MACE, aortic valve surgery, and mortality during follow-up. In the analysis of the correlation between LVGLS and death, there was no significant association between LVGLS and mortality, but it was associated with mortality in studies of prospective analyses.
Heart valve disease is one of the common diseases of the heart and its incidence increases with age. Aortic valve diseases include aortic stenosis and aortic regurgitation. Aortic stenosis is more common than regurgitation [27]. In patients with aortic stenosis, the left ventricle will be exposed to a higher pressure load in the systole, which leads to left ventricular dysfunction and even heart failure. Studies have shown that symptomatic aortic valve stenosis is associated with high mortality [28]. As for aortic regurgitation, the excessive volume load it creates causes the left ventricle to expand until heart failure occurs. Patients with aortic regurgitation also have a higher mortality rate than the general population [29]. Aortic valve disease is therefore often severe and has a major impact on quality of life and survival. One study noted that patients with aortic stenosis have an annual mortality rate of 25% after the onset of symptoms [30]. The annual mortality rate in patients with aortic regurgitation is approximately 25% for NYHA class III-VI patients and 3% for asymptomatic patients [31]. Therefore, a rigorous assessment of the prognosis of patients with aortic valve disease is essential. Left ventricular ejection fraction has long been considered a reliable indicator of cardiac function in clinical practice. However, LVEF is usually preserved in patients with aortic stenosis, making it unable to distinguish between normal myocardial systolic function and subclinical dysfunction [32]. It has also been noted that GLS is associated with all-cause mortality in aortic stenosis and aortic regurgitation [33].
In our study, LVGLS was statistically different between MACE and non-MACE groups and had higher absolute values in the no-MACE group than in the MACE group. The combined effects model showed a correlation between LVGLS and MACE. Through further analyses, we found that impaired LVGLS was associated with event of aortic valve surgery in patients with aortic valve disease. In our analysis of the association of impaired LVGLS with death, an analysis included 5 articles suggesting that impaired LVGLS had no significant association with death. Further subgroup analysis showed that it was associated with mortality in prospective analysis studies. This may be related to the fact that prospective studies better control bias and confounders. In our opinion, there are still a few published articles that meet the requirements of our study, and the results may not be reliable enough, so we expect meta-analyses that include larger samples will be published when more articles are published in the future.
The normal value of LVGLS is still not clearly defined, but a meta-analysis gave it a mean value of −19.7% [34]. From the studies we included and the analysis of the above results, it is easy to conclude that LVGLS values in patients with aortic valve disease are significantly lower than those in the normal population. This finding is consistent with a previous study by Lafitte et al. [35]. In their study, they found that LVGLS was significantly impaired in patients with aortic stenosis. GLS can be used as an indicator to detect early and subclinical left ventricular dysfunction [36]. We also believe that there is a high prognostic value in patients with aortic valve disease.
Purwowiyoto mentions that LVGLS may be a good indicator because it is sensitive to detecting changes in long axis shortening in his article. The sensitivity comes from the vulnerability of the LVGLS when myocardial damage occurs [37]. In addition, the European Association of Cardiovascular Imaging (EACVI) and the American Society of Echocardiography (ASE) have recognized the superiority of GLS over LVEF [38]. In terms of the predictive value of death, it has also been noted that left ventricular global longitudinal strain is an independent predictor of mortality [39]. Regardless of severity of AS (aortic stenosis), patients with abnormal LVGLS have a higher risk of death [40]. It has also been noted that the assessment of myocardial function by GLS has added value in the prognostic assessment of several valvular diseases [26, 41]. As LVGLS continues to be studied, we believe it has the potential to become another reliable indicator of systolic function and valvular heart disease prognosis.
Conclusion
Through our rigorous evaluations, we determined that there exists a correlation between LVGLS and MACE in patients with aortic valve disease, with LVGLS exhibiting prospective significance in the prognostic evaluation of such patients, particularly those afflicted with aortic stenosis. However, further inclusion into analysis is necessary once additional pertinent articles are published to reach unexplored and validated conclusions.
Statement of Ethics
Since the study we conducted was a meta-analysis and the results obtained were based solely on published literature from 2010 to 2023, no ethical statement applies.
Conflict of Interest Statement
The author has been a student for the past 3 years and has not received financial support from any institution or been involved in any financial endeavors during this period; therefore, the author has no conflict of interest to declare.
Funding Sources
This study was mainly funded by the scientific research fund of the corresponding author Siyuan Yang, which was mainly used to search, download, print, and purchase storage equipment for the article.
Author Contributions
Hongsheng Liao, Siyuan Yang, Xuanyi Hu, and Shaomei Yu completed the conceptualization part of the work, as well as designing the methodology for obtaining and analyzing the data for the work and finally interpreting the results. Xiongwei Meng, Kui Wu, and Hongsheng Liao completed the data collection and review of the important knowledge content. Hongsheng Liao completed the data analysis and article writing.
Data Availability Statement
This is a retrospective analysis, the results are based on published articles, and the data supporting the results of this study can be searched in PubMed, ScienceDirect, and SCIE databases. The corresponding URLs are https://pubmed.ncbi.nlm.nih.gov/, https://www.sciencedirect.com/, https://www.webofscience.com/wos/woscc/basic-search.
