Relationship between coronary microvascular dysfunction (CMD) and left ventricular diastolic function in patients with symptoms of myocardial ischemia with non-obstructive coronary artery disease (INOCA) by cardiovascular magnetic resonance feature-tracking

H Kong; J Cao; J Tian; J Yong; J An; X Song; Y He

doi:10.1016/j.crad.2024.02.007

Relationship between coronary microvascular dysfunction (CMD) and left ventricular diastolic function in patients with symptoms of myocardial ischemia with non-obstructive coronary artery disease (INOCA) by cardiovascular magnetic resonance feature-tracking

Clin Radiol. 2024 Mar 13:S0009-9260(24)00129-6. doi: 10.1016/j.crad.2024.02.007. Online ahead of print.

Authors

H Kong¹, J Cao¹, J Tian², J Yong², J An³, X Song⁴, Y He⁵

Affiliations

¹ Department of Radiology, Beijing Friendship Hospital, Capital Medical University, Beijing, 100050, China.
² Department of Cardiology, Beijing Anzhen Hospital, Capital Medical University, Beijing, China.
³ Siemens Shenzhen Magnetic Resonance, MR Collaboration NE Asia, Shenzhen, China.
⁴ Department of Cardiology, Beijing Anzhen Hospital, Capital Medical University, Beijing, China. Electronic address: songxiantao0929@qq.com.
⁵ Department of Radiology, Beijing Friendship Hospital, Capital Medical University, Beijing, 100050, China. Electronic address: Heyi139@sina.com.

PMID: 38679491
DOI: 10.1016/j.crad.2024.02.007

Abstract

Aim: To investigate whether there was an association between coronary microvascular dysfunction (CMD) and left ventricular (LV) diastolic function in patients with myocardial ischemia with non-obstructive coronary artery disease (INOCA).

Materials and methods: Our study included 115 subjects with suspected myocardial ischemia that underwent stress perfusion cardiac magnetic resonance (CMR). They were divided into non-CMD and CMD two groups. CMR-derived volume-time curves and CMR-FT parameters were used to assess LV diastolic function using CVI42 software. The latter included global/regional LV peak longitudinal, circumferential, radial diastolic strain rate (LDSR, CDSR, RDSR). Logistic regression analysis was performed with CMR-FT strain parameters as independent variables and CMD as dependent variables, and the effect value was expressed as an odds ratio (OR).

Results: Of the 115 patients, we excluded data from 23 patients and 92 patients (56.5% male；52 ± 12 years) were finally included in the study. Of these, 19 patients were included in the non-CMD group (49 ± 11 years) and CMD group included 73patient (52 ± 12 years). The regional CDSR (P=0.019), and regional RDSR (P=0.006) were significantly lower in the CMD group than in non-CMD group. But, regional LDSR in CMD group was higher than non-CMD (P=0.003). In logistic regression analysis, regional LDSR (adjusted β= 0.1, 95%CI 0.077, 0.349, p=0.002) and RDSR (adjusted β= 0.1, 95 % CI 0.066, 0.356, p=0.004) were related to CMD.

Conclusions: LV myocardial perfusion parameter MPRI was negatively correlated with LV diastolic function (CDSR) which needs to take into account the degree of diastolic dysfunction.