The relationship between the left ventricular geometric remodeling and clinical worsening in patients with pulmonary arterial hypertension

doi:10.11958/20261200

Abstract

Abstract:

Objective To investigate the association between left ventricular (LV) geometric remodeling derived from short-axis cine cardiac magnetic resonance (CMR) images and clinical worsening (CW) in patients with pulmonary hypertension (PH). Methods A total of 194 patients with PH confirmed by right heart catheterization and who underwent CMR were retrospectively included. Patients were divided into the CW group (n=61) and the non-CW group (n=133) based on follow-up outcome. LV eccentricity, circularity and axis ratio were measured from short-axis cine CMR images at end-diastole (ED) and end-systole (ES). A composite left ventricular geometric index (LVGI) was subsequently constructed. Differences in geometric indices between the two groups were compared. Cox regression was used to evaluate the association between LVGI and clinical worsening. Prognostic discrimination was assessed using the Harrell C-index, 60-month time-dependent receiver operating characteristic (ROC) analysis and Kaplan-Meier analysis. In addition, Spearman correlation analysis was performed to assess the association of LVGI with selected hemodynamic and CMR-derived cardiac functional parameters. Results Compared with the non-CW group, the LV eccentricity and long-to-short axis ratio increased and circularity decreased at both ED and ES in the CW group (all P<0.05). After integration of these geometric parameters, both ED and ES LVGI were significantly higher in the CW group (both P<0.05). In univariable Cox analysis, ED LVGI was significantly associated with clinical worsening (HR=1.684, 95%CI: 1.353-2.107, P<0.001). After multivariable adjustment, ED LVGI remained independently associated with clinical worsening (HR=1.370, 95%CI: 1.028-1.827, P=0.032). The 60-month time-dependent AUC of ED LVGI for predicting clinical worsening was 0.690, and the Harrell C-index was 0.663.Spearman correlation analysis showed that ED LVGI was positively correlated with mean pulmonary arterial pressure, pulmonary vascular resistance and right ventricular end-diastolic volume index (r_s=0.436, 0.421, and 0.399, respectively), and negatively correlated with right ventricular ejection fraction and left ventricular cardiac index (r_s=-0.247 and -0.227, respectively). After stratification by the median ED LVGI, the clinical-worsening-free survival was higher in the high-LVGI group than that of the low-LVGI group (Log-rank P<0.001). Conclusion LV geometric remodeling is more pronounced in PH patients with clinical worsening. CMR-derived ED LVGI is independently associated with CW and may serve as a complementary imaging marker reflecting ventricular interaction and disease progression.

Key words: pulmonary arterial hypertension, magnetic resonance imaging, ventricular remodeling, ventricular function, disease progression

CLC Number:

R445.2，R544.16

Figures/Tables 9

References 18

[1]	Kovacs G, Bartolome S, Denton C P, et al. Definition,classification and diagnosis of pulmonary hypertension[J]. Eur Respir J, 2024, 64(4):2401324. doi:10.1183/13993003.01324-2024.
[2]	阿里旦·艾尔肯, 李鸣远, 米娜瓦尔·胡加艾合买提. 右心室重构特征与肺动脉高压患者治疗反应性的相关性[J]. 中华全科医学, 2026, 24(3):403-406.
	Aierken A D, Li M Y, Hujiaaimaiti M. Correlation between right ventricular remodeling characteristics and treatment responsiveness in patients with pulmonary hypertension[J]. Chin J Gen Pract, 2026, 24(3):403-406. doi:10.16766/j.cnki.issn.1674-4152.004406.
[3]	McCoy D, Movahed M R. Review of D-shape left ventricle seen on magnetic resonance imaging(MRI)or computed tomography(CT),similar to the Movahed sign seen on cardiac gated single-photon emission computed tomography(SPECT)as an indicator for right ventricular overload[J]. J Clin Med, 2025, 14(17):6041. doi:10.3390/jcm14176041.
[4]	Celant L R, Wessels J N, Marcus J T, et al. Toward the implementation of optimal cardiac magnetic resonance risk stratification in pulmonary arterial hypertension[J]. Chest, 2024, 165(1):181-191. doi:10.1016/j.chest.2023.07.028.
[5]	Humbert M, Kovacs G, Hoeper M M, et al. 2022 ESC/ERS guidelines for the diagnosis and treatment of pulmonary hypertension[J]. Eur Heart J, 2022, 43(38):3618-3731. doi:10.1093/eurheartj/ehac237.
[6]	Weatherald J, Fleming T R, Wilkins M R, et al. Clinical trial design,end-points,and emerging therapies in pulmonary arterial hypertension[J]. Eur Respir J, 2024, 64(4):2401205. doi:10.1183/13993003.01205-2024.
[7]	Namazi M, Eftekhar S P, Mosaed R, et al. Pulmonary hypertension and right ventricle:a pathophysiological insight[J]. Clin Med Insights Cardiol, 2024, 18:11795468241274744. doi:10.1177/11795468241274744.
[8]	武才鑫, 闫彦, 邓瑗琳, 等. 心脏磁共振评估结缔组织病伴重度肺动脉高压的价值[J]. 天津医药, 2024, 52(7):691-695.
	Wu C X, Yan Y, Deng Y L, et al. Value of cardiac magnetic resonance in evaluating severe pulmonary hypertension associated with connective tissue disease[J]. Tianjin Med J, 2024, 52(7):691-695. doi:10.11958/20231770.
[9]	Fang H, Wang J, Shi R, et al. Biventricular dysfunction and ventricular interdependence in patients with pulmonary hypertension:a 3.0-T cardiac MRI feature tracking study[J]. J Magn Reson Imaging, 2024, 60(1):350-362. doi:10.1002/jmri.29044.
[10]	Ahmad A, Zhang S, Ren Q, et al. Transthoracic echocardiography of left ventricular underfilling improves risk stratification in pulmonary arterial hypertension[J]. Sci Rep, 2025, 15(1):45718. doi:10.1038/s41598-025-28206-z.
[11]	Lindholm A, Kjellström B, Rådegran G, et al. Right ventricular dyssynchrony predicts outcome in pulmonary arterial hypertension when assessed in multiple cardiac magnetic resonance views[J]. J Cardiovasc Magn Reson, 2024, 26(2):101103. doi:10.1016/j.jocmr.2024.101103.
[12]	Chiou F R, Sung S H, Chiu J H, et al. Cardiac magnetic resonance parameters for prognostic stratification in pulmonary hypertension:a longitudinal study[J]. J Chin Med Assoc, 2025, 88(11):858-867. doi:10.1097/JCMA.0000000000001292.
[13]	Li W, Zhang X C, Qian Y, et al. Prognostic value of cardiac MRI-derived left ventricular layer-specific strain in pulmonary arterial hypertension[J]. Radiol Cardiothorac Imaging, 2025, 7(4):e240334. doi: 10.1148/ryct.240334.
[14]	Burkett D A, Patel S S, Mertens L, et al. Relationship between left ventricular geometry and invasive hemodynamics in pediatric pulmonary hypertension[J]. Circ Cardiovasc Imaging, 2020, 13(5):e009825. doi:10.1161/CIRCIMAGING.119.009825.
[15]	Haarman M G, Coenraad I, Hagdorn Q A J, et al. Cardiac magnetic resonance derived left ventricular eccentricity index and right ventricular mass measurements predict outcome in children with pulmonary arterial hypertension[J]. Children(Basel), 2023, 10(4):756. doi:10.3390/children10040756.
[16]	Kallianos K, Brooks G C, Mukai K, et al. Cardiac magnetic resonance evaluation of left ventricular myocardial strain in pulmonary hypertension[J]. Acad Radiol, 2018, 25(1):129-135. doi:10.1016/j.acra.2017.07.009.
[17]	Guo J, Wang J, Wang L, et al. Left ventricular underfilling in PAH: a potential indicator for adaptive-to-maladaptive transition[J]. Pulm Circ, 2023, 13(4):e12309. doi:10.1002/pul2.12309.
[18]	Ahmad A, Zou Y, Zhang P, et al. Non-invasive imaging techniques for early diagnosis of bilateral cardiac dysfunction in pulmonary hypertension: current crests,future peaks[J]. Front Cardiovasc Med, 2024, 11:1393580. doi:10.3389/fcvm.2024.1393580.

组别	n		年龄/岁	女性		身高/cm		体质量/kg
非CW组	133		40（31，57）	107（80.5）		162（160，165）		60（55，67）
CW组	61		51（37，60）	46（75.4）		162（160，170）		62（53，69）
Z或χ²			2.121^＊	0.372		0.354		0.635
组别		WHO-FC （Ⅰ/Ⅱ/Ⅲ/Ⅳ）			6MWD/m		BNP/（ng/L）
非CW组		5/85/39/4			382（298，457）		102.50（4.32，332.00）
CW组		1/23/30/7			238（43，378）		508.50（8.83，931.00）
χ²或Z		15.691^＊＊			5.282^＊＊		4.431^＊＊

组别	n		年龄/岁	女性		身高/cm		体质量/kg
非CW组	133		40（31，57）	107（80.5）		162（160，165）		60（55，67）
CW组	61		51（37，60）	46（75.4）		162（160，170）		62（53，69）
Z或χ²			2.121^＊	0.372		0.354		0.635
组别		WHO-FC （Ⅰ/Ⅱ/Ⅲ/Ⅳ）			6MWD/m		BNP/（ng/L）
非CW组		5/85/39/4			382（298，457）		102.50（4.32，332.00）
CW组		1/23/30/7			238（43，378）		508.50（8.83，931.00）
χ²或Z		15.691^＊＊			5.282^＊＊		4.431^＊＊

组别	n	mRAP/mmHg	sPAP/mmHg
非CW组	133	5.00（3.00，7.00）	70.00（55.25，84.00）
CW组	61	7.00（5.00，11.00）	88.50（69.25，99.50）
Z		4.592＊＊	3.933＊＊
组别	dPAP/mmHg	mPAP/mmHg	PVR/WU
非CW组	24.00（19.00，30.00）	41.00（35.00，52.00）	8.83（6.34，12.50）
CW组	31.00（25.00，39.75）	53.00（43.00，63.00）	13.69（10.25，18.60）
Z	4.159^＊＊	4.767^＊＊	4.725^＊＊

组别	n	mRAP/mmHg	sPAP/mmHg
非CW组	133	5.00（3.00，7.00）	70.00（55.25，84.00）
CW组	61	7.00（5.00，11.00）	88.50（69.25，99.50）
Z		4.592＊＊	3.933＊＊
组别	dPAP/mmHg	mPAP/mmHg	PVR/WU
非CW组	24.00（19.00，30.00）	41.00（35.00，52.00）	8.83（6.34，12.50）
CW组	31.00（25.00，39.75）	53.00（43.00，63.00）	13.69（10.25，18.60）
Z	4.159^＊＊	4.767^＊＊	4.725^＊＊

组别			n			RVEDVI/（mL/m²）					RVESVI/（mL/m²）
非CW组			133			81.64（62.36，102.22）					48.62（37.14，62.86）
CW组			61			120.06（96.70，147.40）					82.38（65.49，107.26）
Z						6.003^＊＊					7.109^＊＊
组别		RVSVI/（mL/m²）						RVEF/%				RVMMI/（g/m²）
非CW组		31.54（22.88，38.68）						38.44±8.93				16.56（12.15，17.33）
CW组		33.50（27.78，42.37）						28.51±8.01				32.66（25.69，43.87）
Z或t		1.463						7.733^＊＊				4.282^＊＊
组别					RVCI/[L/（min·m²）]				LVEDVI/（mL/m²）
非CW组					2.60（2.13，3.10）				67.63（55.62，77.99）
CW组					2.13（1.59，2.89）				62.79（50.53，72.41）
Z					2.892^＊＊				1.731
组别	LVESVI/（mL/m²）						LVSVI/（mL/m²）						LVMMI/（g/m²）
非CW组	25.75（20.93，33.38）						38.49±12.04						45.86（40.13，52.85）
CW组	27.66（20.57，34.17）						35.79±10.37						39.81（35.14，47.46）
Z或t	0.171						1.726						2.861^＊＊
组别				LVCI/[L/（min·m²）]						LVEF/%
非CW组				2.77（2.53，3.48）						59.80（52.80，64.00）
CW组				2.47（2.31，2.90）						56.00（44.10，63.00）
Z				2.474^＊						1.833