- Email: [email protected]
Usefulness of Myocardial Strain and Twist for Early Detection of Myocardial Dysfunction in Patients With Autoimmune Diseases
Journal Pre-proof
Usefulness of Myocardial Strain and Twist for Early Detection of Myocardial Dysfunction in Patients with Autoimmune Diseases Fu-Wei Jia MD , Jeffrey Hsu MD , Xiao-Hang Liu MD , Xiao-Jin Feng MD , Hai-Yu Pang PhD , Xue Lin MD , Li-Gang Fang MD , Hua-Xia Yang MD , Wei Chen MD PII: DOI: Reference:
S0002-9149(19)31221-4 https://doi.org/10.1016/j.amjcard.2019.10.035 AJC 24264
To appear in:
The American Journal of Cardiology
Received date: Revised date: Accepted date:
18 July 2019 26 October 2019 30 October 2019
Please cite this article as: Fu-Wei Jia MD , Jeffrey Hsu MD , Xiao-Hang Liu MD , Xiao-Jin Feng MD , Hai-Yu Pang PhD , Xue Lin MD , Li-Gang Fang MD , Hua-Xia Yang MD , Wei Chen MD , Usefulness of Myocardial Strain and Twist for Early Detection of Myocardial Dysfunction in Patients with Autoimmune Diseases, The American Journal of Cardiology (2019), doi: https://doi.org/10.1016/j.amjcard.2019.10.035
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2019 Published by Elsevier Inc.
1 / 22
Usefulness of Myocardial Strain and Twist for Early Detection of Myocardial Dysfunction in Patients with Autoimmune Diseases Fu-Wei Jia, MDa, Jeffrey Hsu, MDa, Xiao-Hang Liu, MDa, Xiao-Jin Feng, MDa, Hai-Yu Pang, PhDb, Xue Lin, MDa, Li-Gang Fang, MDa, Hua-Xia Yang, MDc, Wei Chen, MDa, * a
Department of Cardiology, Peking Union Medical College Hospital, Chinese
Academy of Medical Sciences and Peking Union Medical College, Beijing, China; b
Medical Research Center, Peking Union Medical College Hospital, Chinese
Academy of Medical Sciences and Peking Union Medical College, Beijing, China; c
Department of Rheumatology and Clinical Immunology, Peking Union Medical
College Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China Funding: This work was supported by the National Natural Science Foundation of China, grant number 81470426, Beijing, China and Beijing Natural Science Foundation, grant number 7192156, Beijing, China
*
Corresponding author: Wei Chen, Department of Cardiology, Peking Union Medical
College Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, No. 1. Shuaifuyuan, Dongcheng District, Beijing 100730, China, Tel.: 86 10 6915 5068; fax: 86 10 6915 5068. Email: [email protected]
Abstract Cardiac involvement in autoimmune diseases (AD) is common but underdiagnosed
2 / 22
due to a lack of sensitive imaging methods. We aim to evaluate the characteristics of left ventricular (LV) systolic dysfunction in patients with AD using deformational parameters from two-dimensional speckle-tracking echocardiography (STE). We retrospectively enrolled 86 AD patients and 71 healthy controls. All subjects underwent transthoracic echocardiography and STE to analyze LV strain and twist. A twist-radial displacement loop was constructed to investigate the relationship between LV contractility and dimension. Among AD patients, 68 had preserved LV ejection fraction (EF ≥ 50%), and 18 had reduced LVEF (EF < 50%). The patients with preserved LVEF exhibited significantly lower values of global longitudinal (GLS), circumferential (GCS) and radial strain (GRS) than controls (-19.11 ± 4.18 vs. -21.49 ± 2.53%, -25.17 ± 5.04% vs. -27.37 ± 2.87%, 17.68 ± 5.69% vs. 21.17 ± 6.44%, respectively; all p < 0.01) and a marked attenuation in peak twist (14.24 ± 5.57° vs. 18.10 ± 5.97°, p < 0.01) attributed to impaired apical rotation (9.03 ± 5.17° vs. 12.79 ± 5.99°, p < 0.01). AD patients were more likely to present with abnormal loop types with flat ascending slope and delayed peak twist time. In conclusion, abnormal strain and twist precede deterioration in LVEF, suggesting early myocardial involvement in AD. STE can be used as a good alternative for early detection of myocardial dysfunction in AD patients. Keywords: autoimmune diseases, speckle-tracking echocardiography, strain, twist
3 / 22
Autoimmune diseases (AD) are disorders of tolerance against self-peptide antigens characterized by the involvement of multiple organs and systems.1 Cardiovascular complications are common and remain the main cause of mortality for AD patients.1–4 However, cardiac involvement in the early stage of AD can be asymptomatic with preserved left ventricular ejection fraction (LVEF), and the incidence of cardiac impairment is often underreported.2,5,6 For instance, myocardial involvement can be identified in 8-14% of patients with systemic lupus erythematosus (SLE) through imaging modalities, while myocarditis can be found in as many as 40-50% of cases in autopsy studies.4,5,7 Strain and twist, calculated in two-dimensional speckle-tracking echocardiography (STE), can provide detailed information on global and regional as well as systolic and diastolic myocardial mechanics.3,8–10 We hypothesized that deformational parameters are altered prior to LVEF deterioration and are more sensitive indicators for identifying systolic dysfunction in patients with AD. Methods: We consecutively enrolled 118 adult patients with AD at Peking Union Medical College Hospital between 2014 and 2018. AD patients included patients with SLE, dermatomyositis, Sjogren's syndrome and scleroderma. Patients with congenital heart disease, myocardial infarction and history of thoracic surgery were excluded (7 cases). Subjects were also eliminated due to inadequate imaging quality or lack of view for STE analysis (25 cases). AD Patients were divided into two groups: preserved LVEF (LVEF 50%, LVpEF) and reduced LVEF (LVEF 50%, LVrEF). The control group included 71 sex-and age-matched healthy individuals. The study was approved by the
4 / 22
Peking Union Medical College Hospital’s Institutional Review Board. Transthoracic echocardiography was performed on all subjects using an ultrasound system (Vivid 9, GE Medical Systems, Milwaukee, WI). Images for three consecutive cardiac cycles were recorded at a frame rate of 60 Hz. Left atrial and ventricular dimensions and volumes were measured according to the guidelines from the American Society of Echocardiography.11 Apical four-chamber and two-chamber views were analyzed to evaluate LVEF via Simpson’s biplane method. STE analysis was carried out using EchoPAC software (GE Medical Systems, Milwaukee, WI). Longitudinal strain was assessed on apical four-chamber, two-chamber and long-axis views, while circumferential and radial strains, rotation and twist were quantified on short-axis views at the base, the papillary muscle and the apex. Global strains, including longitudinal (GLS) circumferential (GCS) and radial strain (GRS), were calculated as the average value from 17 segments (Figure 1). LV twist represented the mean longitudinal gradient of the net difference between apical and basal rotation during systole.10,12,13 From the apex, counterclockwise or clockwise rotation was indicated using positive or negative value, respectively. Twist slope was defined as the slope of the twist-time curve and was calculated using the slope function [slope (known_y’s, known_x’s)] in Microsoft Excel. Radial displacement (RD) was the average myocardial movement toward the centroid at the LV apex and base. The twist-radial displacement (T-RD) loop was constructed by plotting twist on the y-axis versus RD on the x-axis during a single cardiac cycle.12,14,15 The statistical analysis was performed with SPSS statistics software (version 23,
5 / 22
IBM, Chicago). Categorical variables were expressed as proportions and compared by 2 test. Continuous data were presented as the mean ± SD or as median values and inter-quartile range for normally or nonnormally distributed variables, respectively. The quantitative data were compared via either analysis of variance (ANOVA) or Wilcoxon rank-sum test with Bonferroni’s correction for multiple comparisons. correlation among three types of strains was assessed by liner regression analysis. Receiver operating characteristic (ROC) curve was used to define the optimal cutoff generated from the largest sensitivity and specificity summation. A p value 0.05 was considered statistically significant. Intra- and interobserver variabilities were expressed as the intraclass correlation coefficient (ICC). Results: Our study consisted of 86 consecutive AD patients (12 males, 74 females, mean age 39.2 ± 14.6 years) and 71 healthy subjects (24 males, 47 females, mean age 37.6 12.2 years). The types of AD involved in this study are listed in Table 1. Table 2 summarizes the clinical characteristics and conventional echocardiographic measurements of the study population. Compared with controls, LVpEF patients had a mildly reduced LVEF and a lower systolic mitral annular velocity (s’) (p < 0.001). GLS in LVpEF patients was markedly lower than in controls (p < 0.001) but higher than in LVrEF patients (p < 0.001) (Figure 1). Intriguingly, similar results were also found in terms of GCS and GRS (Table 3). And liner regression analysis indicated GLS was correlated with GCS (r = 0.806, p < 0.001) and GRS (r = 0.502, p < 0.001), respectively (Figure 2). The ROC curve of GLS, at a cutoff of -14.47% for detecting
6 / 22
systolic dysfunction via LVEF measurement, showed a sensitivity of 82.1% and a specificity of 86.7% with an area under the curve of 0.901 (p < 0.001). Rotation analyses were performed to further determine the involved cardiac segments (Figure 3). Compared to controls, both LVpEF and LVrEF groups showed lower peak apical rotation and twist as well as early systolic clockwise twist (all p < 0.01). The peak basal rotation among the three groups was comparable (p = 0.148). Additionally, both LVpEF and LVrEF patients required more time to reach peak twist than controls (p < 0.001), which led to a flatter twist slope (p < 0.001). The T-RD loop is an index that reflects systolic and diastolic force with changes in ventricular deformation during the same heartbeat.16 Four types of loop figures were detected in our study: “eight” (loop 1), “ellipse” (loop 2), “inverse ellipse” (loop 3) and “inverse eight” (loop 4) (Figure 4). Three phases in the T-RD loop were defined: a linear ascending limb throughout systole, a rapid early untwisting limb during the isovolumetric contraction phase and a slow late untwisting limb during the filling period. Loops 3 and 4 (loop 3&4) accounted for greater proportion of the loop figures in AD patients (43%) than in controls (11%). Compared to loops 1 and 2 (loop 1&2), where twist peaked earlier than RD, loop 3&4 exhibited the reverse time sequence of peak twist and maximum RD (RDmax). Though there was no difference in the peak value of twist or RD between groups, loop 3&4 presented markedly delayed peak twist time (48.07 ± 9.18% vs. 43.72 ± 6.67%, p = 0.001) and a smaller slope for the ascending limb than loop 1&2 (2.23 ± 1.99°/mm vs. 2.68 ± 1.33°/mm, p = 0.011). Intra- and interobserver variability for 10 subjects was 0.97 and 0.88 for GLS, 0.94
7 / 22
and 0.82 for twist. Therefore, the data indicated high reproducibility of the measurements performed in our study. Discussion Our study investigated the usefulness of STE measurements of myocardial deformation during systole as a way to detect early myocardial involvement in AD patients. There were three main findings. First, LVpEF patients had decreased strain in multiple directions compared to controls. Second, peak twist was reduced due to decreased peak apical rotation. Last, AD patients were more likely than controls to have abnormal loop types with flat ascending slope and prolonged peak twist time. GLS, generated from subendocardial fibers, is a sensitive indicator of myocardial injury.17,18 In our study, decreased GLS in LVpEF patients suggests subclinical myocardial compromise may be identified even without systolic dysfunction in conventional echocardiography. This is consistent with previous studies of SLE, rheumatic arthritis and granulomatosis with polyangiitis.3,19,20 Reduced early systolic clockwise twist also reflected subendocardial dysfunction.21 Both GCS and GRS, depending on the severity of myocardial involvement, can be spared initially because of relatively unaffected midmyocardial and subepicardial fibers.18 Changes of GCS and GRS in AD remain undetermined and controversial in existing researches. Interestingly, we found the loss of GCS and GRS appeared simultaneously even if LVEF was preserved, which provides pathophysiologic insight into the mechanism of LV transmural impairment. There are no studies that systematically evaluate the changes of rotation and twist
8 / 22
in AD patients. We found that peak apical rotation decreased, while peak basal rotation remained spared, leading to attenuated peak twist, which is probably a consequence of more severe impairment in the apex than in other parts. As the main contributor of twist, decreased apical rotation plays a substantial role in the loss of uniform transmural distribution of deformation along the fiber direction.22 The simultaneous abnormality of strain and twist indicates transmural heterogeneity in myocardial function.23 Thus, the attenuation of multidirectional strains, including GLS, GCS and GRS, and twist implies that myocardial involvement may be transmural, which may be an important characteristic of cardiac dysfunction in AD. Beyar et al first demonstrated that LV twist was directly related to RD throughout the entire cardiac cycle using a T-RD loop.24 As twist is a parameter of contractility and RD is a surrogate for LV dimensions, the T-RD loop can indicate changes in stroke work and in the potential energy stored in myocardial fibers.12,16 Previous studies found that the T-RD loop can be distorted in many diseases, such as hypertrophic cardiomyopathy, chronic mitral regurgitation, myocardial infarction and heart transplantation.12,15,25,26 In our study, we identified four types of T-RD loop and proposed that different types of loop provide valuable insight into myocardial dysfunction in AD. Both loop 1- and loop 2-types are considered normal variants16, while loop 3- and loop 4-types, which perform in the reverse direction, were first presented in our study as abnormal modalities. We found the difference between loop 1&2 and loop 3&4 lies in the time sequence of peak twist and RDmax. A Schematic representation is shown in Figure 5. According to Kydd, the timing of RD can
9 / 22
represent dyssynchrony of myocardial motion, while amplitude reflects contractility. Early or late RDmax in segments even with preserved contractility will not fully contribute to end-systolic function, leading to wasted energy.27 Thus, we inferred that loop 3&4 with late peak twist time and early RDmax time may suggest an imbalance of energy conversion from kinetic energy to elastic potential energy caused by asynchronous myocardial contraction. The abnormal T-RD loop may indicate an altered contractility and dimension relationship. The main limitation is the small number of patients, further study with larger sample size should be performed to confirm our findings. Although deformational parameters are statistically different between groups, it is still difficult to classify an individual patient correctly because of the large standard deviations. Reproducibility of twist parameters may be influenced by imaging difficulties, such as low frame rate of images and small variations in location of image acquisition. Moreover, most patients had accepted glucocorticoid and/or immunosuppressant therapy before enrolling in our study. A recent study demonstrated that silent myocardial impairment can occur in patients with drug-naïve new onset SLE.28 However, whether the aforementioned medicines affect strain or twist is unknown. In conclusion, impaired strain and twist occur early in AD patients prior to LVEF deterioration, suggesting myocardial deformation can be a good indicator for early detection of systolic dysfunction. Our results may assist in improving early diagnosis of cardiac dysfunction and therapeutic strategies in AD patients. Disclosures:
10 / 22
The authors have no conflicts of interest to disclose.
11 / 22
References 1. Generali E, Folci M, Selmi C, Riboldi P. Immune-Mediated Heart Disease. Adv Exp Med Biol 2017;1003:145–171. 2. Prasad M, Hermann J, Gabriel SE, Weyand CM, Mulvagh S, Mankad R, Oh JK, Matteson EL, Lerman A. Cardiorheumatology: cardiac involvement in systemic rheumatic disease. Nat Rev Cardiol 2015;12:168–176. 3. Fine NM, Crowson CS, Lin G, Oh JK, Villarraga HR, Gabriel SE. Evaluation of myocardial function in patients with rheumatoid arthritis using strain imaging by speckle-tracking echocardiography. Ann Rheum Dis 2014;73:1833–1839. 4. Knockaert DC. Cardiac involvement in systemic inflammatory diseases. Eur Hear J 2007;28:1797–1804. 5. Lee KS, Kronbichler A, Eisenhut M, Lee KH, Shin JI. Cardiovascular involvement in systemic rheumatic diseases: An integrated view for the treating physicians. Autoimmun Rev 2018;17:201–214. 6. Mavrogeni S, Schwitter J, Rossum A van, Nijveldt R, Aletras A, Kolovou G, Pohost G, Lima J. Cardiac magnetic resonance imaging in myocardial inflammation in autoimmune rheumatic diseases: An appraisal of the diagnostic strengths and limitations of the Lake Louise criteria. Int J Cardiol 2018;252:216–219. 7. Doherty NE, Siegel RJ. Cardiovascular manifestations of systemic lupus erythematosus. Am Hear J 1985;110:1257–1265. 8. Blessberger H, Binder T. NON-invasive imaging: Two dimensional speckle tracking echocardiography: basic principles. Heart 2010;96:716–722.
12 / 22
9. Dedeoglu R, Sahin S, Koka A, Oztunc F, Adrovic A, Barut K, Cengiz D, Kasapcopur O. Evaluation of cardiac functions in juvenile systemic lupus erythematosus with two-dimensional speckle tracking echocardiography. Clin Rheumatol 2016;35:1967–1975. 10. Ayyildiz YO, Vural MG, Efe TH, Ertem AG, Koseoglu C, Ayturk M, Yeter E, Keskin G, Akdemir R. Effect of Long-Term TNF-alpha Inhibition with Infliximab on Left Ventricular Torsion in Patients with Rheumatoid Arthritis. Hell J Cardiol 2015;56:406–413. 11. Anon. Recommendations for Cardiac Chamber Quantification by Echocardiography in Adults: An Update from the American Society of Echocardiography and the European Association of, Cardiovascular Imaging. Eur Hear J Cardiovasc Imaging 2016;17:412. 12. Nawaytou HM, Yubbu P, Montero AE, Nandi D, O’Connor MJ, Shaddy RE, Banerjee A. Left Ventricular Rotational Mechanics in Children After Heart Transplantation. Circ Cardiovasc Imaging 2016;9. 13. Chandrasekaran K, Khandheria BK. Twist Mechanics of the Left Ventricle. JACC Cardiovasc 2008;1:366–376. 14. Nakai H, Takeuchi M, Nishikage T, Kokumai M, Otani S, Lang RM. Effect of aging on twist-displacement loop by 2-dimensional speckle tracking imaging. J Am Soc Echocardiogr 2006;19:880–885. 15. Chang SA, Kim HK, Kim DH, Kim JC, Kim YJ, Kim HC, Sohn DW, Oh BH, Park YB. Left ventricular twist mechanics in patients with apical hypertrophic
13 / 22
cardiomyopathy: assessment with 2D speckle tracking echocardiography. Heart 2010;96:49–55. 16. Maria M V Di, Caracciolo G, Prashker S, Sengupta PP, Banerjee A. Left ventricular rotational mechanics before and after exercise in children. J Am Soc Echocardiogr 2014;27:1336–1343. 17. Kalam K, Otahal P, Marwick TH. Prognostic implications of global LV dysfunction: a systematic review and meta-analysis of global longitudinal strain and ejection fraction. Heart 2014;100:1673–1680. 18. Geyer H, Caracciolo G, Abe H, Wilansky S, Carerj S, Gentile F, Nesser HJ, Khandheria B, Narula J, Sengupta PP. Assessment of myocardial mechanics using speckle tracking echocardiography: fundamentals and clinical applications. J Am Soc Echocardiogr 2010;23:351–355. 19. Bulut M, Acar RD, Acar S, Fidan S, Yesin M, Izci S, Efe SC, Cakir H. Evaluation of torsion and twist mechanics of the left ventricle in patients with systemic lupus erythematosus. Anatol J Cardiol 2016;16:434–439. 20. Miszalski-Jamka T, Szczeklik W, Nycz K, Sokolowska B, Gorka J, Bury K, Musial J. Two-dimensional speckle-tracking echocardiography reveals systolic abnormalities in granulomatosis with polyangiitis (Wegener’s). Echocardiography 2012;29:803–809. 21. Takeuchi M, Otsuji Y, Lang RM. Evaluation of left ventricular function using left ventricular twist and torsion parameters. Curr Cardiol Rep 2009;11:225–230. 22. Carasso S, Yang H, Woo A, Vannan MA, Jamorski M, Wigle ED, Rakowski H.
14 / 22
Systolic myocardial mechanics in hypertrophic cardiomyopathy: novel concepts and implications for clinical status. J Am Soc Echocardiogr 2008;21:675–683. 23. Sengupta PP, Tajik AJ, Chandrasekaran K, Khandheria BK. Twist mechanics of the left ventricle: principles and application. JACC Cardiovasc Imaging 2008;1:366– 376. 24. Beyar R, Yin FC, Hausknecht M, Weisfeldt ML, Kass DA. Dependence of left ventricular twist-radial shortening relations on cardiac cycle phase. Am J Physiol 1989;257:H1119-26. 25. Borg AN, Harrison JL, Argyle RA, Ray SG. Left ventricular torsion in primary chronic mitral regurgitation. Heart 2008;94:597–603. 26. Takeuchi M, Nishikage T, Nakai H, Kokumai M, Otani S, Lang RM. The assessment of left ventricular twist in anterior wall myocardial infarction using two-dimensional speckle tracking imaging. J Am Soc Echocardiogr 2007;20:36–44. 27. Kydd AC, Khan FZ, O’Halloran D, Pugh PJ, Virdee MS, Dutka DP. Radial strain delay based on segmental timing and strain amplitude predicts left ventricular reverse remodeling and survival after cardiac resynchronization therapy. Circ Cardiovasc Imaging 2013;6:177–184. 28. Guo Q, Wu LM, Wang Z, Shen JY, Su X, Wang CQ, Gong XR, Yan QR, He Q, Zhang W, Xu JR, Jiang M, Pu J. Early Detection of Silent Myocardial Impairment in Drug-Naive Patients With New-Onset Systemic Lupus Erythematosus: A Three-Center Prospective Study. Arthritis Rheumatol 2018;70:2014–2024.
15 / 22
Figure Legends:
Figure 1. Representative examples of peak GLS in Bull’s Eye Image during the systolic period. Compared with that in normal controls (A), GLS decreased significantly in both patients with preserved LVEF (LVEF 50%) (B) and patients with reduced LVEF (LVEF 50%) (C). GLS = global longitudinal strain; LVEF = left ventricular ejection fraction.
Figure 2. Scatter plots of the correlation between GLS and GCS (A), between GLS and GRS (B), and between GCS and GRS (C). GLS = global longitudinal strain; GCS = global circumferential strain; GRS = global radial strain.
16 / 22
Figure 3. Comparison of peak apical rotation, basal rotation and twist between controls and AD patients. Apical rotation, basal rotation and twist curves during a cardiac cycle in control (A) versus AD patients with preserved LVEF (LVEF 50%) (B) or with reduced LVEF (LVEF 50%) (C). Peak twist was lower in AD patients than in normal controls, and this was mainly attributed to a relative decrease in peak apical rotation (D). AD = autoimmune diseases; LVEF = left ventricular ejection fraction.
17 / 22
Figure 4. Four types of twist-radial displacement loop in AD patients. Loop 1 had a characteristic appearance similar to a figure eight, and loop 2 was an elliptical loop in the clockwise direction [origin-ascending limb (solid arrow)-peak twist (●)-RDmax (■)-descending limb (dotted arrow)-origin]. In contrast, loop3 and loop4 had inverse directions [origin-ascending limb (solid arrow)- RDmax (■)-peak twist (●)-descending limb (dotted arrow)-origin] similar to an inverse ellipse and an inverse eight, respectively. AD = autoimmune diseases; RD = radial displacement; RDmax, the maximum radial displacement.
18 / 22
Figure 5. Schematic representation of twist generation and T-RD loop formation. (A) When viewed from the apex, subepicardial fibers rotate the apex counterclockwise and the base clockwise (blue arrow) whereas subendocardial fibers cause the opposite rotation in both the apex and base (green arrow). Since the subepicardial fibers with larger radius determine the overall direction, the apical rotation is counterclockwise, while the basal rotation is clockwise (red arrow) during the ejection phase. (B and C) Radial displacement (RD), the average myocardial movement toward the centroid, is enlarged during the systolic period and minimized during the diastolic phase. RDmax is comparable in controls and AD patients as shown in the RD curve. (D, E and F) An apical and basal rotation generate twist during a cardiac cycle. The peak twist time is markedly delayed in AD patients (dotted black line). (G and H) T-RD loop figures are further constructed according to the relationship between peak twist time and RDmax time: one has an earlier time of peak twist than of RDmax (G), which is more common in normal people; the other has the reserve time sequence (H) and is more
19 / 22
frequent in AD patients. AD = autoimmune diseases; RD = radial displacement; RDmax = the maximum of radial displacement; T-RD = twist-radial displacement.
Table 1 Classification of autoimmune diseases in this study Left Ventricular Ejection Fraction Diagnosis
50% (n = 68)
< 50% (n = 18)
Systemic lupus erythematosus
50
9
Dermatomyositis
9
6
Sjogren's syndrome
3
1
Scleroderma
6
2
Table 2 Baseline characteristics and conventional echocardiography indices of the study population Normal Controls
Left Ventricular Ejection Fraction
Variable
(n = 71)
50% (n = 68)
< 50% (n = 18)
Age (years)
37.6 12.2
39.4 ± 15.3
38.5 12.0
Female
47 (66%)
61 (90%)
13 (72%)
Body surface area (m2)
1.64 0.18
1.58 ± 0.16
1.66 0.19
Heart rate (bpm)
65.9 ± 7.9
86.2 ± 15.7 *
80.4 14.4 *
20 / 22
NT-proBNP (pg/ml)
-
441 (129 to 1565)
1845 (842 to 9510) ^
CRP (mg/L)
-
2.76 (1.15 to
1.47 (1.07 to 2.67)
7.62) Hypertension
-
18 (26%)
7 (39%)
NYHA class 2
-
9 (13%)
12 (67%) ^
Pericardial effusion
-
27 (40%)
6 (34%)
-
60 (88%)
12 (67%)
-
52 (76%)
11 (61%)
LVEF (%)
66.42 ± 6.00
63.75 ± 8.30 *
38.39 5.85 *^
Lateral s’ (cm/s)
10.60 ± 2.14
9.15 ± 2.33 *
5.90 1.79 *^
LV mass index (g/m2)
62.26 ± 14.61
81.09 ± 35.04
112.23 59.54 *
E/A ratio
1.56 ± 0.31
1.19 ± 0.40 *
1.24 0.76 *
E/e’ ratio
6.47 ± 1.41
9.98 ± 6.47 *
13.88 6.22 *
IVCT (ms)
60.31 ± 10.32
63.00 ± 13.33
81.33 24.93 *^
ET (ms)
306.04 ± 21.12
265.51 ± 29.20 *
257.60 33.67 *
IVRT (ms)
61.28 ± 9.07
68.73 ± 22.31
93.87 24.63 *^
Medications Glucocorticoid Immunosuppressant
A = peak late diastolic velocity of mitral valve inflow by pulsed-wave Doppler; E = peak early diastolic velocity of mitral valve inflow by pulsed-wave Doppler; ET = ejection time; IVCT = isovolumic contraction time; IVRT = isovolumic relaxation time; NYHA class = New York Heart Association function classification. * adjusted p value < 0.05 when compared to controls
21 / 22
^ adjusted p value < 0.05 when compared to patients with LVEF 50%
Table 3 Comparison of strain and twist variables for controls and patients (mean ± SD)
Variable
Normal controls Left Ventricular (n = 71)
Left Ventricular
Ejection Fraction Ejection Fraction 50% (n = 68)
< 50% (n = 18)
GLS (%)
-21.49 ± 2.53
-19.11 ± 4.18 *
-12.03 ± 3.68 *^
GCS (%)
-27.37 ± 2.87
-25.17 ± 5.04 *
-15.98 ± 4.73 *^
GRS (%)
21.17 ± 6.44
17.68 ± 5.69 *
12.01 ± 3.82 *^
Peak apex rotation (°)
12.79 ± 5.99
9.03 ± 5.17 *
6.33 ± 4.13 *
Peak base rotation (°)
-5.70 ± 2.79
-5.98 ± 2.95
-4.41 ± 2.26
Peak twist (°)
18.10 ± 5.97
14.24 ± 5.57 *
9.72 ± 4.76 *^
-1.42 ± 1.38
-0.92 ± 1.07 *
Early systolic clockwise -1.80 ± 1.24 twist (°) Twist slope (°/s)
63.43 ± 20.51
53.94 ± 24.44 *
33.33 ± 18.85 *^
Peak twist time (%)
39.89 ± 4.45
45.20 ± 6.89 *
50.61 ± 10.48 *
RDmax (mm)
6.04 0.91
6.06 1.43
4.04 1.30 *^
RDmax time (%)
43.32 6.08
45.66 6.45
48.20 8.70 *
Loop 1&2 / loop 3&4
63/8
41/27 *
8/10 *
22 / 22
GLS = global longitudinal strain; GCS = global circumferential strain; GRS = global radial strain. RDmax = the maximum of radial displacement. * adjusted p value < 0.01 when compared to controls ^ adjusted p value < 0.01 when compared to patients with LVEF 50%
Usefulness of Myocardial Strain and Twist for Early Detection of Myocardial Dysfunction in Patients with Autoimmune Diseases Fu-Wei Jia MD , Jeffrey Hsu MD , Xiao-Hang Liu MD , Xiao-Jin Feng MD , Hai-Yu Pang PhD , Xue Lin MD , Li-Gang Fang MD , Hua-Xia Yang MD , Wei Chen MD PII: DOI: Reference:
S0002-9149(19)31221-4 https://doi.org/10.1016/j.amjcard.2019.10.035 AJC 24264
To appear in:
The American Journal of Cardiology
Received date: Revised date: Accepted date:
18 July 2019 26 October 2019 30 October 2019
Please cite this article as: Fu-Wei Jia MD , Jeffrey Hsu MD , Xiao-Hang Liu MD , Xiao-Jin Feng MD , Hai-Yu Pang PhD , Xue Lin MD , Li-Gang Fang MD , Hua-Xia Yang MD , Wei Chen MD , Usefulness of Myocardial Strain and Twist for Early Detection of Myocardial Dysfunction in Patients with Autoimmune Diseases, The American Journal of Cardiology (2019), doi: https://doi.org/10.1016/j.amjcard.2019.10.035
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2019 Published by Elsevier Inc.
1 / 22
Usefulness of Myocardial Strain and Twist for Early Detection of Myocardial Dysfunction in Patients with Autoimmune Diseases Fu-Wei Jia, MDa, Jeffrey Hsu, MDa, Xiao-Hang Liu, MDa, Xiao-Jin Feng, MDa, Hai-Yu Pang, PhDb, Xue Lin, MDa, Li-Gang Fang, MDa, Hua-Xia Yang, MDc, Wei Chen, MDa, * a
Department of Cardiology, Peking Union Medical College Hospital, Chinese
Academy of Medical Sciences and Peking Union Medical College, Beijing, China; b
Medical Research Center, Peking Union Medical College Hospital, Chinese
Academy of Medical Sciences and Peking Union Medical College, Beijing, China; c
Department of Rheumatology and Clinical Immunology, Peking Union Medical
College Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China Funding: This work was supported by the National Natural Science Foundation of China, grant number 81470426, Beijing, China and Beijing Natural Science Foundation, grant number 7192156, Beijing, China
*
Corresponding author: Wei Chen, Department of Cardiology, Peking Union Medical
College Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, No. 1. Shuaifuyuan, Dongcheng District, Beijing 100730, China, Tel.: 86 10 6915 5068; fax: 86 10 6915 5068. Email: [email protected]
Abstract Cardiac involvement in autoimmune diseases (AD) is common but underdiagnosed
2 / 22
due to a lack of sensitive imaging methods. We aim to evaluate the characteristics of left ventricular (LV) systolic dysfunction in patients with AD using deformational parameters from two-dimensional speckle-tracking echocardiography (STE). We retrospectively enrolled 86 AD patients and 71 healthy controls. All subjects underwent transthoracic echocardiography and STE to analyze LV strain and twist. A twist-radial displacement loop was constructed to investigate the relationship between LV contractility and dimension. Among AD patients, 68 had preserved LV ejection fraction (EF ≥ 50%), and 18 had reduced LVEF (EF < 50%). The patients with preserved LVEF exhibited significantly lower values of global longitudinal (GLS), circumferential (GCS) and radial strain (GRS) than controls (-19.11 ± 4.18 vs. -21.49 ± 2.53%, -25.17 ± 5.04% vs. -27.37 ± 2.87%, 17.68 ± 5.69% vs. 21.17 ± 6.44%, respectively; all p < 0.01) and a marked attenuation in peak twist (14.24 ± 5.57° vs. 18.10 ± 5.97°, p < 0.01) attributed to impaired apical rotation (9.03 ± 5.17° vs. 12.79 ± 5.99°, p < 0.01). AD patients were more likely to present with abnormal loop types with flat ascending slope and delayed peak twist time. In conclusion, abnormal strain and twist precede deterioration in LVEF, suggesting early myocardial involvement in AD. STE can be used as a good alternative for early detection of myocardial dysfunction in AD patients. Keywords: autoimmune diseases, speckle-tracking echocardiography, strain, twist
3 / 22
Autoimmune diseases (AD) are disorders of tolerance against self-peptide antigens characterized by the involvement of multiple organs and systems.1 Cardiovascular complications are common and remain the main cause of mortality for AD patients.1–4 However, cardiac involvement in the early stage of AD can be asymptomatic with preserved left ventricular ejection fraction (LVEF), and the incidence of cardiac impairment is often underreported.2,5,6 For instance, myocardial involvement can be identified in 8-14% of patients with systemic lupus erythematosus (SLE) through imaging modalities, while myocarditis can be found in as many as 40-50% of cases in autopsy studies.4,5,7 Strain and twist, calculated in two-dimensional speckle-tracking echocardiography (STE), can provide detailed information on global and regional as well as systolic and diastolic myocardial mechanics.3,8–10 We hypothesized that deformational parameters are altered prior to LVEF deterioration and are more sensitive indicators for identifying systolic dysfunction in patients with AD. Methods: We consecutively enrolled 118 adult patients with AD at Peking Union Medical College Hospital between 2014 and 2018. AD patients included patients with SLE, dermatomyositis, Sjogren's syndrome and scleroderma. Patients with congenital heart disease, myocardial infarction and history of thoracic surgery were excluded (7 cases). Subjects were also eliminated due to inadequate imaging quality or lack of view for STE analysis (25 cases). AD Patients were divided into two groups: preserved LVEF (LVEF 50%, LVpEF) and reduced LVEF (LVEF 50%, LVrEF). The control group included 71 sex-and age-matched healthy individuals. The study was approved by the
4 / 22
Peking Union Medical College Hospital’s Institutional Review Board. Transthoracic echocardiography was performed on all subjects using an ultrasound system (Vivid 9, GE Medical Systems, Milwaukee, WI). Images for three consecutive cardiac cycles were recorded at a frame rate of 60 Hz. Left atrial and ventricular dimensions and volumes were measured according to the guidelines from the American Society of Echocardiography.11 Apical four-chamber and two-chamber views were analyzed to evaluate LVEF via Simpson’s biplane method. STE analysis was carried out using EchoPAC software (GE Medical Systems, Milwaukee, WI). Longitudinal strain was assessed on apical four-chamber, two-chamber and long-axis views, while circumferential and radial strains, rotation and twist were quantified on short-axis views at the base, the papillary muscle and the apex. Global strains, including longitudinal (GLS) circumferential (GCS) and radial strain (GRS), were calculated as the average value from 17 segments (Figure 1). LV twist represented the mean longitudinal gradient of the net difference between apical and basal rotation during systole.10,12,13 From the apex, counterclockwise or clockwise rotation was indicated using positive or negative value, respectively. Twist slope was defined as the slope of the twist-time curve and was calculated using the slope function [slope (known_y’s, known_x’s)] in Microsoft Excel. Radial displacement (RD) was the average myocardial movement toward the centroid at the LV apex and base. The twist-radial displacement (T-RD) loop was constructed by plotting twist on the y-axis versus RD on the x-axis during a single cardiac cycle.12,14,15 The statistical analysis was performed with SPSS statistics software (version 23,
5 / 22
IBM, Chicago). Categorical variables were expressed as proportions and compared by 2 test. Continuous data were presented as the mean ± SD or as median values and inter-quartile range for normally or nonnormally distributed variables, respectively. The quantitative data were compared via either analysis of variance (ANOVA) or Wilcoxon rank-sum test with Bonferroni’s correction for multiple comparisons. correlation among three types of strains was assessed by liner regression analysis. Receiver operating characteristic (ROC) curve was used to define the optimal cutoff generated from the largest sensitivity and specificity summation. A p value 0.05 was considered statistically significant. Intra- and interobserver variabilities were expressed as the intraclass correlation coefficient (ICC). Results: Our study consisted of 86 consecutive AD patients (12 males, 74 females, mean age 39.2 ± 14.6 years) and 71 healthy subjects (24 males, 47 females, mean age 37.6 12.2 years). The types of AD involved in this study are listed in Table 1. Table 2 summarizes the clinical characteristics and conventional echocardiographic measurements of the study population. Compared with controls, LVpEF patients had a mildly reduced LVEF and a lower systolic mitral annular velocity (s’) (p < 0.001). GLS in LVpEF patients was markedly lower than in controls (p < 0.001) but higher than in LVrEF patients (p < 0.001) (Figure 1). Intriguingly, similar results were also found in terms of GCS and GRS (Table 3). And liner regression analysis indicated GLS was correlated with GCS (r = 0.806, p < 0.001) and GRS (r = 0.502, p < 0.001), respectively (Figure 2). The ROC curve of GLS, at a cutoff of -14.47% for detecting
6 / 22
systolic dysfunction via LVEF measurement, showed a sensitivity of 82.1% and a specificity of 86.7% with an area under the curve of 0.901 (p < 0.001). Rotation analyses were performed to further determine the involved cardiac segments (Figure 3). Compared to controls, both LVpEF and LVrEF groups showed lower peak apical rotation and twist as well as early systolic clockwise twist (all p < 0.01). The peak basal rotation among the three groups was comparable (p = 0.148). Additionally, both LVpEF and LVrEF patients required more time to reach peak twist than controls (p < 0.001), which led to a flatter twist slope (p < 0.001). The T-RD loop is an index that reflects systolic and diastolic force with changes in ventricular deformation during the same heartbeat.16 Four types of loop figures were detected in our study: “eight” (loop 1), “ellipse” (loop 2), “inverse ellipse” (loop 3) and “inverse eight” (loop 4) (Figure 4). Three phases in the T-RD loop were defined: a linear ascending limb throughout systole, a rapid early untwisting limb during the isovolumetric contraction phase and a slow late untwisting limb during the filling period. Loops 3 and 4 (loop 3&4) accounted for greater proportion of the loop figures in AD patients (43%) than in controls (11%). Compared to loops 1 and 2 (loop 1&2), where twist peaked earlier than RD, loop 3&4 exhibited the reverse time sequence of peak twist and maximum RD (RDmax). Though there was no difference in the peak value of twist or RD between groups, loop 3&4 presented markedly delayed peak twist time (48.07 ± 9.18% vs. 43.72 ± 6.67%, p = 0.001) and a smaller slope for the ascending limb than loop 1&2 (2.23 ± 1.99°/mm vs. 2.68 ± 1.33°/mm, p = 0.011). Intra- and interobserver variability for 10 subjects was 0.97 and 0.88 for GLS, 0.94
7 / 22
and 0.82 for twist. Therefore, the data indicated high reproducibility of the measurements performed in our study. Discussion Our study investigated the usefulness of STE measurements of myocardial deformation during systole as a way to detect early myocardial involvement in AD patients. There were three main findings. First, LVpEF patients had decreased strain in multiple directions compared to controls. Second, peak twist was reduced due to decreased peak apical rotation. Last, AD patients were more likely than controls to have abnormal loop types with flat ascending slope and prolonged peak twist time. GLS, generated from subendocardial fibers, is a sensitive indicator of myocardial injury.17,18 In our study, decreased GLS in LVpEF patients suggests subclinical myocardial compromise may be identified even without systolic dysfunction in conventional echocardiography. This is consistent with previous studies of SLE, rheumatic arthritis and granulomatosis with polyangiitis.3,19,20 Reduced early systolic clockwise twist also reflected subendocardial dysfunction.21 Both GCS and GRS, depending on the severity of myocardial involvement, can be spared initially because of relatively unaffected midmyocardial and subepicardial fibers.18 Changes of GCS and GRS in AD remain undetermined and controversial in existing researches. Interestingly, we found the loss of GCS and GRS appeared simultaneously even if LVEF was preserved, which provides pathophysiologic insight into the mechanism of LV transmural impairment. There are no studies that systematically evaluate the changes of rotation and twist
8 / 22
in AD patients. We found that peak apical rotation decreased, while peak basal rotation remained spared, leading to attenuated peak twist, which is probably a consequence of more severe impairment in the apex than in other parts. As the main contributor of twist, decreased apical rotation plays a substantial role in the loss of uniform transmural distribution of deformation along the fiber direction.22 The simultaneous abnormality of strain and twist indicates transmural heterogeneity in myocardial function.23 Thus, the attenuation of multidirectional strains, including GLS, GCS and GRS, and twist implies that myocardial involvement may be transmural, which may be an important characteristic of cardiac dysfunction in AD. Beyar et al first demonstrated that LV twist was directly related to RD throughout the entire cardiac cycle using a T-RD loop.24 As twist is a parameter of contractility and RD is a surrogate for LV dimensions, the T-RD loop can indicate changes in stroke work and in the potential energy stored in myocardial fibers.12,16 Previous studies found that the T-RD loop can be distorted in many diseases, such as hypertrophic cardiomyopathy, chronic mitral regurgitation, myocardial infarction and heart transplantation.12,15,25,26 In our study, we identified four types of T-RD loop and proposed that different types of loop provide valuable insight into myocardial dysfunction in AD. Both loop 1- and loop 2-types are considered normal variants16, while loop 3- and loop 4-types, which perform in the reverse direction, were first presented in our study as abnormal modalities. We found the difference between loop 1&2 and loop 3&4 lies in the time sequence of peak twist and RDmax. A Schematic representation is shown in Figure 5. According to Kydd, the timing of RD can
9 / 22
represent dyssynchrony of myocardial motion, while amplitude reflects contractility. Early or late RDmax in segments even with preserved contractility will not fully contribute to end-systolic function, leading to wasted energy.27 Thus, we inferred that loop 3&4 with late peak twist time and early RDmax time may suggest an imbalance of energy conversion from kinetic energy to elastic potential energy caused by asynchronous myocardial contraction. The abnormal T-RD loop may indicate an altered contractility and dimension relationship. The main limitation is the small number of patients, further study with larger sample size should be performed to confirm our findings. Although deformational parameters are statistically different between groups, it is still difficult to classify an individual patient correctly because of the large standard deviations. Reproducibility of twist parameters may be influenced by imaging difficulties, such as low frame rate of images and small variations in location of image acquisition. Moreover, most patients had accepted glucocorticoid and/or immunosuppressant therapy before enrolling in our study. A recent study demonstrated that silent myocardial impairment can occur in patients with drug-naïve new onset SLE.28 However, whether the aforementioned medicines affect strain or twist is unknown. In conclusion, impaired strain and twist occur early in AD patients prior to LVEF deterioration, suggesting myocardial deformation can be a good indicator for early detection of systolic dysfunction. Our results may assist in improving early diagnosis of cardiac dysfunction and therapeutic strategies in AD patients. Disclosures:
10 / 22
The authors have no conflicts of interest to disclose.
11 / 22
References 1. Generali E, Folci M, Selmi C, Riboldi P. Immune-Mediated Heart Disease. Adv Exp Med Biol 2017;1003:145–171. 2. Prasad M, Hermann J, Gabriel SE, Weyand CM, Mulvagh S, Mankad R, Oh JK, Matteson EL, Lerman A. Cardiorheumatology: cardiac involvement in systemic rheumatic disease. Nat Rev Cardiol 2015;12:168–176. 3. Fine NM, Crowson CS, Lin G, Oh JK, Villarraga HR, Gabriel SE. Evaluation of myocardial function in patients with rheumatoid arthritis using strain imaging by speckle-tracking echocardiography. Ann Rheum Dis 2014;73:1833–1839. 4. Knockaert DC. Cardiac involvement in systemic inflammatory diseases. Eur Hear J 2007;28:1797–1804. 5. Lee KS, Kronbichler A, Eisenhut M, Lee KH, Shin JI. Cardiovascular involvement in systemic rheumatic diseases: An integrated view for the treating physicians. Autoimmun Rev 2018;17:201–214. 6. Mavrogeni S, Schwitter J, Rossum A van, Nijveldt R, Aletras A, Kolovou G, Pohost G, Lima J. Cardiac magnetic resonance imaging in myocardial inflammation in autoimmune rheumatic diseases: An appraisal of the diagnostic strengths and limitations of the Lake Louise criteria. Int J Cardiol 2018;252:216–219. 7. Doherty NE, Siegel RJ. Cardiovascular manifestations of systemic lupus erythematosus. Am Hear J 1985;110:1257–1265. 8. Blessberger H, Binder T. NON-invasive imaging: Two dimensional speckle tracking echocardiography: basic principles. Heart 2010;96:716–722.
12 / 22
9. Dedeoglu R, Sahin S, Koka A, Oztunc F, Adrovic A, Barut K, Cengiz D, Kasapcopur O. Evaluation of cardiac functions in juvenile systemic lupus erythematosus with two-dimensional speckle tracking echocardiography. Clin Rheumatol 2016;35:1967–1975. 10. Ayyildiz YO, Vural MG, Efe TH, Ertem AG, Koseoglu C, Ayturk M, Yeter E, Keskin G, Akdemir R. Effect of Long-Term TNF-alpha Inhibition with Infliximab on Left Ventricular Torsion in Patients with Rheumatoid Arthritis. Hell J Cardiol 2015;56:406–413. 11. Anon. Recommendations for Cardiac Chamber Quantification by Echocardiography in Adults: An Update from the American Society of Echocardiography and the European Association of, Cardiovascular Imaging. Eur Hear J Cardiovasc Imaging 2016;17:412. 12. Nawaytou HM, Yubbu P, Montero AE, Nandi D, O’Connor MJ, Shaddy RE, Banerjee A. Left Ventricular Rotational Mechanics in Children After Heart Transplantation. Circ Cardiovasc Imaging 2016;9. 13. Chandrasekaran K, Khandheria BK. Twist Mechanics of the Left Ventricle. JACC Cardiovasc 2008;1:366–376. 14. Nakai H, Takeuchi M, Nishikage T, Kokumai M, Otani S, Lang RM. Effect of aging on twist-displacement loop by 2-dimensional speckle tracking imaging. J Am Soc Echocardiogr 2006;19:880–885. 15. Chang SA, Kim HK, Kim DH, Kim JC, Kim YJ, Kim HC, Sohn DW, Oh BH, Park YB. Left ventricular twist mechanics in patients with apical hypertrophic
13 / 22
cardiomyopathy: assessment with 2D speckle tracking echocardiography. Heart 2010;96:49–55. 16. Maria M V Di, Caracciolo G, Prashker S, Sengupta PP, Banerjee A. Left ventricular rotational mechanics before and after exercise in children. J Am Soc Echocardiogr 2014;27:1336–1343. 17. Kalam K, Otahal P, Marwick TH. Prognostic implications of global LV dysfunction: a systematic review and meta-analysis of global longitudinal strain and ejection fraction. Heart 2014;100:1673–1680. 18. Geyer H, Caracciolo G, Abe H, Wilansky S, Carerj S, Gentile F, Nesser HJ, Khandheria B, Narula J, Sengupta PP. Assessment of myocardial mechanics using speckle tracking echocardiography: fundamentals and clinical applications. J Am Soc Echocardiogr 2010;23:351–355. 19. Bulut M, Acar RD, Acar S, Fidan S, Yesin M, Izci S, Efe SC, Cakir H. Evaluation of torsion and twist mechanics of the left ventricle in patients with systemic lupus erythematosus. Anatol J Cardiol 2016;16:434–439. 20. Miszalski-Jamka T, Szczeklik W, Nycz K, Sokolowska B, Gorka J, Bury K, Musial J. Two-dimensional speckle-tracking echocardiography reveals systolic abnormalities in granulomatosis with polyangiitis (Wegener’s). Echocardiography 2012;29:803–809. 21. Takeuchi M, Otsuji Y, Lang RM. Evaluation of left ventricular function using left ventricular twist and torsion parameters. Curr Cardiol Rep 2009;11:225–230. 22. Carasso S, Yang H, Woo A, Vannan MA, Jamorski M, Wigle ED, Rakowski H.
14 / 22
Systolic myocardial mechanics in hypertrophic cardiomyopathy: novel concepts and implications for clinical status. J Am Soc Echocardiogr 2008;21:675–683. 23. Sengupta PP, Tajik AJ, Chandrasekaran K, Khandheria BK. Twist mechanics of the left ventricle: principles and application. JACC Cardiovasc Imaging 2008;1:366– 376. 24. Beyar R, Yin FC, Hausknecht M, Weisfeldt ML, Kass DA. Dependence of left ventricular twist-radial shortening relations on cardiac cycle phase. Am J Physiol 1989;257:H1119-26. 25. Borg AN, Harrison JL, Argyle RA, Ray SG. Left ventricular torsion in primary chronic mitral regurgitation. Heart 2008;94:597–603. 26. Takeuchi M, Nishikage T, Nakai H, Kokumai M, Otani S, Lang RM. The assessment of left ventricular twist in anterior wall myocardial infarction using two-dimensional speckle tracking imaging. J Am Soc Echocardiogr 2007;20:36–44. 27. Kydd AC, Khan FZ, O’Halloran D, Pugh PJ, Virdee MS, Dutka DP. Radial strain delay based on segmental timing and strain amplitude predicts left ventricular reverse remodeling and survival after cardiac resynchronization therapy. Circ Cardiovasc Imaging 2013;6:177–184. 28. Guo Q, Wu LM, Wang Z, Shen JY, Su X, Wang CQ, Gong XR, Yan QR, He Q, Zhang W, Xu JR, Jiang M, Pu J. Early Detection of Silent Myocardial Impairment in Drug-Naive Patients With New-Onset Systemic Lupus Erythematosus: A Three-Center Prospective Study. Arthritis Rheumatol 2018;70:2014–2024.
15 / 22
Figure Legends:
Figure 1. Representative examples of peak GLS in Bull’s Eye Image during the systolic period. Compared with that in normal controls (A), GLS decreased significantly in both patients with preserved LVEF (LVEF 50%) (B) and patients with reduced LVEF (LVEF 50%) (C). GLS = global longitudinal strain; LVEF = left ventricular ejection fraction.
Figure 2. Scatter plots of the correlation between GLS and GCS (A), between GLS and GRS (B), and between GCS and GRS (C). GLS = global longitudinal strain; GCS = global circumferential strain; GRS = global radial strain.
16 / 22
Figure 3. Comparison of peak apical rotation, basal rotation and twist between controls and AD patients. Apical rotation, basal rotation and twist curves during a cardiac cycle in control (A) versus AD patients with preserved LVEF (LVEF 50%) (B) or with reduced LVEF (LVEF 50%) (C). Peak twist was lower in AD patients than in normal controls, and this was mainly attributed to a relative decrease in peak apical rotation (D). AD = autoimmune diseases; LVEF = left ventricular ejection fraction.
17 / 22
Figure 4. Four types of twist-radial displacement loop in AD patients. Loop 1 had a characteristic appearance similar to a figure eight, and loop 2 was an elliptical loop in the clockwise direction [origin-ascending limb (solid arrow)-peak twist (●)-RDmax (■)-descending limb (dotted arrow)-origin]. In contrast, loop3 and loop4 had inverse directions [origin-ascending limb (solid arrow)- RDmax (■)-peak twist (●)-descending limb (dotted arrow)-origin] similar to an inverse ellipse and an inverse eight, respectively. AD = autoimmune diseases; RD = radial displacement; RDmax, the maximum radial displacement.
18 / 22
Figure 5. Schematic representation of twist generation and T-RD loop formation. (A) When viewed from the apex, subepicardial fibers rotate the apex counterclockwise and the base clockwise (blue arrow) whereas subendocardial fibers cause the opposite rotation in both the apex and base (green arrow). Since the subepicardial fibers with larger radius determine the overall direction, the apical rotation is counterclockwise, while the basal rotation is clockwise (red arrow) during the ejection phase. (B and C) Radial displacement (RD), the average myocardial movement toward the centroid, is enlarged during the systolic period and minimized during the diastolic phase. RDmax is comparable in controls and AD patients as shown in the RD curve. (D, E and F) An apical and basal rotation generate twist during a cardiac cycle. The peak twist time is markedly delayed in AD patients (dotted black line). (G and H) T-RD loop figures are further constructed according to the relationship between peak twist time and RDmax time: one has an earlier time of peak twist than of RDmax (G), which is more common in normal people; the other has the reserve time sequence (H) and is more
19 / 22
frequent in AD patients. AD = autoimmune diseases; RD = radial displacement; RDmax = the maximum of radial displacement; T-RD = twist-radial displacement.
Table 1 Classification of autoimmune diseases in this study Left Ventricular Ejection Fraction Diagnosis
50% (n = 68)
< 50% (n = 18)
Systemic lupus erythematosus
50
9
Dermatomyositis
9
6
Sjogren's syndrome
3
1
Scleroderma
6
2
Table 2 Baseline characteristics and conventional echocardiography indices of the study population Normal Controls
Left Ventricular Ejection Fraction
Variable
(n = 71)
50% (n = 68)
< 50% (n = 18)
Age (years)
37.6 12.2
39.4 ± 15.3
38.5 12.0
Female
47 (66%)
61 (90%)
13 (72%)
Body surface area (m2)
1.64 0.18
1.58 ± 0.16
1.66 0.19
Heart rate (bpm)
65.9 ± 7.9
86.2 ± 15.7 *
80.4 14.4 *
20 / 22
NT-proBNP (pg/ml)
-
441 (129 to 1565)
1845 (842 to 9510) ^
CRP (mg/L)
-
2.76 (1.15 to
1.47 (1.07 to 2.67)
7.62) Hypertension
-
18 (26%)
7 (39%)
NYHA class 2
-
9 (13%)
12 (67%) ^
Pericardial effusion
-
27 (40%)
6 (34%)
-
60 (88%)
12 (67%)
-
52 (76%)
11 (61%)
LVEF (%)
66.42 ± 6.00
63.75 ± 8.30 *
38.39 5.85 *^
Lateral s’ (cm/s)
10.60 ± 2.14
9.15 ± 2.33 *
5.90 1.79 *^
LV mass index (g/m2)
62.26 ± 14.61
81.09 ± 35.04
112.23 59.54 *
E/A ratio
1.56 ± 0.31
1.19 ± 0.40 *
1.24 0.76 *
E/e’ ratio
6.47 ± 1.41
9.98 ± 6.47 *
13.88 6.22 *
IVCT (ms)
60.31 ± 10.32
63.00 ± 13.33
81.33 24.93 *^
ET (ms)
306.04 ± 21.12
265.51 ± 29.20 *
257.60 33.67 *
IVRT (ms)
61.28 ± 9.07
68.73 ± 22.31
93.87 24.63 *^
Medications Glucocorticoid Immunosuppressant
A = peak late diastolic velocity of mitral valve inflow by pulsed-wave Doppler; E = peak early diastolic velocity of mitral valve inflow by pulsed-wave Doppler; ET = ejection time; IVCT = isovolumic contraction time; IVRT = isovolumic relaxation time; NYHA class = New York Heart Association function classification. * adjusted p value < 0.05 when compared to controls
21 / 22
^ adjusted p value < 0.05 when compared to patients with LVEF 50%
Table 3 Comparison of strain and twist variables for controls and patients (mean ± SD)
Variable
Normal controls Left Ventricular (n = 71)
Left Ventricular
Ejection Fraction Ejection Fraction 50% (n = 68)
< 50% (n = 18)
GLS (%)
-21.49 ± 2.53
-19.11 ± 4.18 *
-12.03 ± 3.68 *^
GCS (%)
-27.37 ± 2.87
-25.17 ± 5.04 *
-15.98 ± 4.73 *^
GRS (%)
21.17 ± 6.44
17.68 ± 5.69 *
12.01 ± 3.82 *^
Peak apex rotation (°)
12.79 ± 5.99
9.03 ± 5.17 *
6.33 ± 4.13 *
Peak base rotation (°)
-5.70 ± 2.79
-5.98 ± 2.95
-4.41 ± 2.26
Peak twist (°)
18.10 ± 5.97
14.24 ± 5.57 *
9.72 ± 4.76 *^
-1.42 ± 1.38
-0.92 ± 1.07 *
Early systolic clockwise -1.80 ± 1.24 twist (°) Twist slope (°/s)
63.43 ± 20.51
53.94 ± 24.44 *
33.33 ± 18.85 *^
Peak twist time (%)
39.89 ± 4.45
45.20 ± 6.89 *
50.61 ± 10.48 *
RDmax (mm)
6.04 0.91
6.06 1.43
4.04 1.30 *^
RDmax time (%)
43.32 6.08
45.66 6.45
48.20 8.70 *
Loop 1&2 / loop 3&4
63/8
41/27 *
8/10 *
22 / 22
GLS = global longitudinal strain; GCS = global circumferential strain; GRS = global radial strain. RDmax = the maximum of radial displacement. * adjusted p value < 0.01 when compared to controls ^ adjusted p value < 0.01 when compared to patients with LVEF 50%