Sex differences in CT-FFR of myocardial bridging with or without atherosclerosis: an AI-based quantitative study
Authors: Jingmin Li, Fangying Jia, Haowen Zhang, Tong Pan, Yang Yu, Mengya Li, Caiying Li, Dan Zhang
Published: 2026

ABSTRACT
AbstractBackgroundMyocardial bridging (MB) is a prevalent coronary anomaly with potential links to major adverse cardiac events. While computed tomography-derived fractional flow reserve (FFRCT) offers a non-invasive functional assessment, current evidence predominantly treats MB as a homogeneous entity, overlooking potential sex differences in hemodynamic impact. Existing studies often fail to distinguish between isolated MB and MB with concomitant atherosclerosis, and rarely employ sex-stratified analyses. This study aimed to noninvasively assess sex-specific differences in FFRCT among patients with MB, with or without atherosclerosis, using an artificial intelligence (AI)-based platform to clarify the clinical implications of these differences.MethodsThis retrospective cohort study included 300 left anterior descending artery (LAD)-MB patients, subdivided into an MB group (n=155) and an MB with atherosclerosis (MBLA) group (n=145), along with 104 controls with normal coronary computed tomography angiography (CCTA) findings. Demographic data, clinical symptoms, and risk factors were collected. Morphological parameters of MB were quantitatively analyzed using cardiac function post-processing software, and whole-vessel and local segments (proximal to MB, within MB, and distal to MB) FFRCT values were obtained via an AI-based platform (Shukun-FFRCT). Patients were stratified by FFRCT <0.8, and statistical analyses [t-tests, analysis of variance (ANOVA), Mann-Whitney U tests, and binary logistic regression] were applied to examine sex-based associations with FFRCT abnormalities and influencing factors.Results(I) Demographic analysis revealed significantly higher proportions of male patients in the MB (52.9%) and MBLA (57.2%) groups compared to controls (24.0%, both P<0.05). (II) Sex-stratified analysis within pathological subgroups showed that in the MB group, females had significantly lower distal FFRCT values (FFR3: median 0.92 vs. 0.95, P=0.006) and higher trans-bridge pressure drop (ΔFFR: 0.08 vs. 0.05, P=0.004) than males. No significant sex-based differences in FFRCT were observed in the MBLA group (all P>0.05). (III) Intra-group comparisons of FFRCT between systolic and diastolic phases showed no significant differences in either case or control groups (all P>0.05). (IV) Binary logistic regression indicated that MB length was an independent risk factor for FFRCT abnormalities in both sexes. Each 1-mm increase in MB length raised the risk of FFRCT abnormality by 5.3% in males [odds ratio (OR) =1.053, 95% confidence interval (CI): 1.004–1.104, P=0.033] and 11.3% in females (OR =1.113, 95% CI: 1.024–1.209, P=0.011).ConclusionsThis study innovatively demonstrates through CT-AI quantitative analysis that sex significantly influences FFRCT in LAD-MB patients. Females with isolated MB exhibit more pronounced distal hemodynamic alterations, whereas sex differences diminish when atherosclerosis coexists. Notably, MB length is an independent risk factor for FFRCT abnormalities, with a greater impact observed in females. These findings provide preliminary evidence to inform risk stratification of MB.Keywords: Sex, coronary computed tomography angiography (CCTA), myocardial bridging (MB), fractional flow reserve (FFR), left anterior descending artery (LAD)

Introduction

Myocardial bridging (MB), a common congenital coronary anomaly, is often considered a benign finding; however, growing evidence has established its association with serious cardiovascular events. Coronary computed tomography angiography (CCTA) has emerged as the primary non-invasive imaging modality for detecting MB. Nevertheless, it primarily provides anatomical information and offers limited ability to comprehensively evaluate the hemodynamic alterations induced by MB (1-3).

Although invasive fractional flow reserve (FFR) serves as the gold standard for functional ischemia assessment, its invasive nature limits its widespread clinical application (1,3). In recent years, the non-invasive computed tomography-derived FFR (FFRCT) technique has been developed and validated. This technology utilizes standard CCTA datasets without requiring additional scanning, thereby providing simultaneous anatomical and functional information. Multiple studies have validated the good concordance between FFRCT and invasive FFR in assessing hemodynamically significant coronary stenoses (4,5).

Nevertheless, research on the application of FFRCT specifically in MB patients remains scarce. Most existing FFRCT-related MB studies focused on validating its diagnostic performance against invasive FFR (6-8), or exploring the association between MB anatomical parameters and FFRCT values, with several limitations: first, nearly all studies analyzed MB patients as a homogeneous group without sex stratification, despite growing evidence of sex differences in coronary physiology (9); second, prior works either exclusively enrolled isolated MB patients or mixed cases with concomitant atherosclerosis (8,10), failing to clarify whether atherosclerosis modifies the sex-specific hemodynamic effect of MB itself; third, no study has systematically compared FFRCT profiles between isolated MB and MB combined with atherosclerosis across sexes.

To address these gaps, our study aims to leverage artificial intelligence (AI)-based CT-FFR to: (I) characterize sex-specific hemodynamic alterations in left anterior descending artery (LAD)-MB patients; (II) compare the magnitude of sex differences between isolated MB and MB + atherosclerosis subgroups; (III) identify sex-specific risk factors for hemodynamically significant MB. Our findings aim to provide evidence for sex-tailored risk stratification of MB. We present this article in accordance with the STROBE reporting checklist (available at https://qims.amegroups.com/article/view/10.21037/qims-2026-0713/rc).

Methods

Study population

This study retrospectively collected data from patients who underwent CCTA examination in the Department of Radiology, The Second Hospital of Hebei Medical University, between January 2019 and June 2022. The flow chart is shown in Figure 1. Based on strict inclusion and exclusion criteria, 404 patients were finally enrolled and divided into three groups: (I) MB group: 155 cases, median age 53 years, including 82 males and 73 females; (II) MB with atherosclerosis (MBLA) group: 145 cases, median age 55 years, including 83 males and 62 females; (III) control group: 104 cases, median age 54 years, including 25 males and 79 females. Inclusion criteria: (I) CCTA-confirmed LAD MB with depth ≥1.0 mm and presence of atherosclerotic plaque (calcified, non-calcified, or mixed) in the LAD, regardless of the degree of luminal stenosis; (II) good image quality in both systolic and diastolic phases without motion artifacts; (III) absence of MB or atherosclerotic plaques in other coronary arteries. Exclusion criteria: (I) complicated with cardiomyopathy and valvular disease; (II) presence of coronary artery aneurysm, coronary termination and origin anomalies; (III) poor CCTA image quality unable to meet diagnostic requirements; (IV) history of coronary artery bypass grafting or interventional therapy; (V) presence of cardiac pacemakers, artificial valves and other implants; (VI) presence of MB or atherosclerosis in non-LAD coronary arteries. Patients’ age, gender, body mass index (BMI), clinical symptoms (angina pectoris, atypical angina, chest tightness/pain, asymptomatic) and cardiovascular disease risk factors (hypertension, hyperlipidemia, diabetes, smoking) were collected.

Flow chart of the study. CABG, coronary artery bypass grafting; CCTA, coronary computed tomography angiography; FFR, fractional flow reserve; LAD, left anterior descending artery; MB, myocardial bridging; MBLA, myocardial bridging and LAD atherosclerosis; PCI, percutaneous coronary intervention.

Instrumentation and measurement methods

CCTA scanning protocol

A 256-slice Philips spiral CT scanner was used for examination. During scanning, patients were placed in supine position. Retrospective electrocardiogram (ECG)-gating technique was used. Patients underwent breathing training before scanning, with single end-expiratory breath-hold scanning performed for 5–7 seconds. The scanning range extended from 0.5 cm below the tracheal bifurcation to the cardiac diaphragm. The aortic threshold was set at 150 Hounsfield unit (HU), and intelligent tracking method was employed, with formal scanning initiating 6 seconds after reaching the threshold. Non-ionic contrast agent iohexol (350 mgI/mL) was injected using a dual-syringe high-pressure injector at a rate of 4–5 mL/s, with a dose of 0.8 mL/kg. Scanning parameters: tube voltage 80–120 kV, tube current 280–370 mAs/rotation, scan field of view 250 mm. Detector collimation 128×0.625, matrix 512×512, pitch 0.18, rotation time 330 ms. The optimal systolic (45%) and diastolic (75%) phases were reconstructed.

Image analysis

LAD-MB was diagnosed on CCTA as a segment of the LAD taking an intramyocardial course. Consistent with established criteria (11) and prognostic validation studies (6), MB was defined by a minimum myocardial thickness (depth) of ≥1.0 mm overlying the vessel during diastole. Patients were strictly stratified into an MB group (no atherosclerotic plaque in the LAD) and an MBLA group (presence of LAD plaque).

All CCTA raw data were imported into the post-processing workstation Philips EBW 6.0. Post-processing and analysis of systolic and diastolic images were performed using techniques including axial imaging, multi-planar reformation (MPR), volume rendering (VR) and curved planar reformation (CPR). Measurement of MB anatomical parameters was completed by two CCTA radiologists with more than 5 years of relevant experience, with each parameter measured three times and averaged (Table S1). The position of the MB (distance from the LAD ostium to the proximal end of the MB) and length (distance from the proximal to distal end of the MB) were measured on diastolic-phase CPR images. The depth of the MB (the maximum myocardial thickness overlying the coronary artery during diastole) was measured on vascular cross-sectional images (Figure 2) for subsequent statistical analysis. When the maximum myocardial thickness was ≤1 mm, it was uniformly recorded as 1 mm (11). The relevant calculation formulas are shown below (12-14):

The method to measure FFRCT value and MB anatomic parameter. (A) Measurement method for MB length; (B) measurement of the MB FFRCT; (C) measurement method for MB depth. FFRCT, computed tomography-derived fractional flow reserve; MB, myocardial bridging.

Based on the presence or absence of calcification, plaques were classified into three types: calcified plaques, non-calcified plaques, and mixed plaques.

FFRCT analysis

An AI-based FFRCT analysis software (Shukun-FFRCT, Beijing, China) was employed to measure FFRCT values during both systolic and diastolic phases (Figure 2). This software utilizes an advanced deep learning AI platform for the non-invasive computation of FFRCT. FFR measurements were performed at three specific segments: the 10 mm segment proximal to the MB (FFR1), the MB segment with the most significant stenosis (or the mid-tunnel segment if no significant stenosis was present) (FFR2), and the segment 20–40 mm distal to the MB (FFR3). The ΔFFR was calculated as FFR1 − FFR3. Each FFRCT measurement was performed three times, and the average value was used for the final analysis. An abnormal FFRCT value was defined as ≤0.80 (3). Patients in both the MB group and the MBLA group were further subdivided based on sex into an FFRCT normal group (FFRCT >0.8) and an FFRCT abnormal group (FFRCT ≤0.8) to analyze the factors influencing FFRCT abnormalities across different sexes. The selection of these specific measurement locations (proximal 10 mm for flow acceleration; distal 20–40 mm for pressure recovery) aligns with established hemodynamic principles and prior literature (6).

Ethical statement

The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. This study was approved by the Medical Ethics Committee of The Second Hospital of Hebei Medical University (approval No. W2023-R430). Informed consent was waived in this retrospective study.

Statistical analysis

Statistical analyses were conducted using SPSS software (version 25.0). Measurement data conforming to a normal distribution were expressed as mean ± standard deviation and compared using t-tests or one-way analysis of variance (ANOVA). Non-normally distributed measurement data were expressed as median and interquartile range [M (P25–P75)] and compared using the Mann-Whitney U test. Categorical data were presented as frequencies (percentages) and compared using the Chi-squared (χ2) test. Multifactorial analysis was performed using binary logistic regression to identify factors influencing FFRCT abnormalities. Variables for multivariable regression were selected via a hybrid approach: initial univariate screening (P<0.05) supplemented by a priori inclusion of clinically essential and literature-reported confounders, regardless of screening P values. A two-tailed P value of less than 0.05 (P<0.05) was considered statistically significant.

Results

Comparison of baseline characteristics and FFRCT values among groups

The comparison of baseline characteristics and FFRCT values among the MB group, MBLA group, and control group is presented in Table 1. Compared with the control group, the proportion of male patients was significantly higher in both the MB group and the MBLA group (P<0.05). Additionally, the smoking rate was higher in the MB group (P<0.05), while the incidence of angina pectoris was significantly higher in the MBLA group compared to the control group (P<0.05). No statistically significant differences were observed in other baseline indicators, including age, BMI, other risk factors, and clinical symptoms, between the case groups and the control group (P>0.05). Regarding hemodynamic parameters, both the MB group and the MBLA group showed significantly lower FFR1, FFR2, FFR3, and ΔFFR values compared to the control group during both systolic and diastolic phases (all P<0.01).

Table 1Basic information of patients and comparison of FFRCTVariableSubgroupControl (n=104)MB (n=155)P1MBLA (n=145)P2Basic informationAge (years)54.0 (45.0–59.0)53.0 (45.0–58.0)0.53455.0 (49.0–59.5)0.288Male25 (24.0)82 (52.9)<0.00183 (57.2)<0.001BMI (kg/m2)24.9 (22.6–27.6)24.5 (22.9–26.1)0.28724.8 (23.0–26.7)0.929Risk factorsHypertension32 (30.8)36 (23.2)0.17643 (29.7)0.850Hyperlipidemia16 (15.4)16 (10.3)0.22526 (17.9)0.597Diabetes 9 (8.7)15 (9.7)0.78118 (12.4)0.347Smoking6 (5.8)27 (17.4)0.00615 (10.3)0.200Clinical symptomsAngina pectoris4 (3.8)13 (8.4)0.14818 (12.4)0.019Atypical angina5 (4.8)10 (6.5)0.5798 (5.5)0.804Chest pain7 (6.7)14 (9.0)0.50619 (13.1)0.105Asymptomatic88 (84.6)118 (76.1)0.097100 (69.0)0.005Diastolic phaseFFR11.0 (1.0–1.0)1.0 (1.0–1.0)0.0050.98 (0.94–1.0)<0.001FFR21.0 (0.99–1.0)0.99 (0.97–1.0)<0.0010.96 (0.91–0.98)<0.001FFR30.97 (0.94–0.98)0.94 (0.88–0.97)<0.0010.89 (0.83–0.95)<0.001△FFR0.03 (0.01–0.06)0.06 (0.03–0.12)<0.0010.07 (0.03–0.11)<0.001Systolic phaseFFR11.0 (1.0–1.0)1.0 (1.0–1.0)0.0240.99 (0.95–1.0)<0.001FFR21.0 (0.99–1.0)0.99 (0.97–1.0)<0.0010.96 (0.90–0.98)<0.001FFR30.96 (0.93–0.98)0.93 (0.88–0.97)<0.0010.89 (0.82–0.93)<0.001△FFR0.04 (0.02–0.06)0.06 (0.03–0.11)<0.0010.07 (0.04–0.12)<0.001Data are presented as median (interquartile range) or n (%). FFR1, FFRCT value at 10 mm proximal to MB; FFR2, the FFRCT value at MB; FFR3, the FFRCT value at 20–40 mm from the far end of MB; △FFR, FFR1 − FFR3; P1, comparison between the MB group and the control group; P2, comparison between the MBLA group and the control group. BMI, body mass index; FFRCT, computed tomography-derived fractional flow reserve; MB, myocardial bridge; MBLA, myocardial bridge combined with left anterior descending artery atherosclerosis.

Basic information of patients and comparison of FFRCT

Data are presented as median (interquartile range) or n (%). FFR1, FFRCT value at 10 mm proximal to MB; FFR2, the FFRCT value at MB; FFR3, the FFRCT value at 20–40 mm from the far end of MB; △FFR, FFR1 − FFR3; P1, comparison between the MB group and the control group; P2, comparison between the MBLA group and the control group. BMI, body mass index; FFRCT, computed tomography-derived fractional flow reserve; MB, myocardial bridge; MBLA, myocardial bridge combined with left anterior descending artery atherosclerosis.

Sex-based differences in FFRCT values among MB patients

Comparisons of FFRCT values between sexes in patients with MB and MBLA are shown in Tables 2,3. During diastolic phase measurements in the MB group, female patients had significantly lower FFR3 values (P<0.05) and significantly higher ΔFFR values than males (P<0.05). No statistically significant differences were found in other FFRCT parameters between sexes (P>0.05).

Table 2Comparison of FFRCT MB patients of different gendersMBSubgroupMale (n=82)Female (n=73) PDiastolic phase FFRCTFFR11.00 (1.00–1.00)1.00 (1.00–1.00)0.997FFR20.99 (0.98–1.00)0.99 (0.97–1.00)0.304FFR30.95 (0.90–1.00)0.92 (0.87–0.96)0.006△FFR0.05 (0.02–0.10)0.08 (0.04–0.13)0.004Systolic phase FFRCTFFR11.00 (1.00–1.00)1.00 (1.00–1.00)0.363FFR20.99 (0.97–1.00)0.99 (0.98–1.00)0.929FFR30.94 (0.89–0.97)0.92 (0.86–0.97)0.120△FFR0.06 (0.03–0.10)0.08 (0.03–0.12)0.080Data are presented as median (interquartile range). FFR1, FFRCT value at 10 mm proximal to MB; FFR2, the FFRCT value at MB; FFR3, the FFRCT value at 20–40 mm from the far end of MB; △FFR, FFR1 − FFR3. FFRCT, computed tomography-derived fractional flow reserve; MB, myocardial bridge.

Comparison of FFRCT MB patients of different genders

Data are presented as median (interquartile range). FFR1, FFRCT value at 10 mm proximal to MB; FFR2, the FFRCT value at MB; FFR3, the FFRCT value at 20–40 mm from the far end of MB; △FFR, FFR1 − FFR3. FFRCT, computed tomography-derived fractional flow reserve; MB, myocardial bridge.

Table 3Comparison of FFRCT MBLA patients of different gendersMBLASubgroupMale (n=83)Female (n=62) PDiastolic phase FFRCTFFR10.98 (0.92–1.00)0.98 (0.96–1.00)0.819FFR20.95 (0.87–0.98)0.96 (0.93–0.98)0.489FFR30.89 (0.82–0.94)0.91 (0.86–0.95)0.617△FFR0.06 (0.03–0.11)0.07 (0.03–0.10)0.641Systolic phase FFRCTFFR10.99 (0.95–1.00)0.98 (0.95–1.00)0.322FFR20.95 (0.86–0.98)0.96 (0.93–0.98)0.329FFR30.87 (0.81–0.94)0.90 (0.85–0.93)0.469△FFR0.08 (0.03–0.14)0.07 (0.04–0.10)0.430Data are presented as 95% (CI). FFR1, FFRCT value at 10 mm proximal to MB; FFR2, the FFRCT value at MB; FFR3, the FFRCT value at 20–40 mm from the far end of MB; △FFR, FFR1 − FFR3. CI, confidence interval; FFRCT, computed tomography-derived fractional flow reserve; MBLA, myocardial bridge combined with left anterior descending artery atherosclerosis.

Comparison of FFRCT MBLA patients of different genders

Data are presented as 95% (CI). FFR1, FFRCT value at 10 mm proximal to MB; FFR2, the FFRCT value at MB; FFR3, the FFRCT value at 20–40 mm from the far end of MB; △FFR, FFR1 − FFR3. CI, confidence interval; FFRCT, computed tomography-derived fractional flow reserve; MBLA, myocardial bridge combined with left anterior descending artery atherosclerosis.

Analysis of factors influencing FFRCT abnormalities in male MB patients

MB patients were categorized into normal and abnormal FFRCT groups based on their FFRCT values. The comparison of baseline characteristics between these two groups is presented in Table 4. Except for a significant difference in MB length between the two groups (P<0.05), no statistically significant differences were found in age, BMI, cardiovascular risk factors, or clinical symptoms (P>0.05). Binary logistic regression analysis incorporating these variables and other clinically relevant or literature-reported factors (Table 5) revealed that MB length was an independent risk factor for FFRCT abnormalities [odds ratio (OR) =1.053, 95% confidence interval (CI): 1.004–1.104; P=0.033]. This indicates that for every 1 mm increase in MB length, the risk of FFRCT abnormality increases by 5.3% in male patients.

Table 4Comparison of FFRCT normal group and abnormal group in male MB patientsVariableSubgroupFFRCT ≤0.8 (n=34)FFRCT >0.8 (n=48)PBasic informationAge (years)51.50 (40.50–54.50)51.00 (43.00–57.00)0.351BMI (kg/m2)24.54±2.5124.88 (23.06–37.32)0.457Risk factorsHypertension7 (20.6)12 (25.0)0.641Hyperlipidemia2 (5.9)4 (8.3)>0.99Diabetes 3 (8.8)6 (12.5)0.868Smoking9 (26.5)13 (27.1)0.951Clinical symptomsAngina pectoris10 (29.4)6 (12.5)0.057Atypical angina0 (0.0)1 (2.1)>0.99Chest pain22 (64.7)14 (29.2)0.675Asymptomatic10 (29.4)19 (39.6)0.343MB anatomyLength (mm)26.7 (19.8–33.9)20.7 (14.4–32.7)0.028Depth (mm)1.80 (1.00–3.53)1.95 (1.00–3.40)0.958Location (mm)30.7 (28.83–48.03)36.64±11.980.884Systolic compression index0.11 (0.00–0.17)0.08 (0.06–0.14)0.774MBMI42.35 (27.70–119.36)38.44 (18.28–86.02)0.255MB stenosis in diastole 0.36±0.150.32±0.120.394MB stenosis in systole 0.35±0.190.34±0.140.825Data are presented as median (interquartile range), n (%), or mean ± standard deviation. BMI, body mass index; FFRCT, computed tomography-derived fractional flow reserve; MB, myocardial bridge; MBMI, MB muscle index.

Comparison of FFRCT normal group and abnormal group in male MB patients

Data are presented as median (interquartile range), n (%), or mean ± standard deviation. BMI, body mass index; FFRCT, computed tomography-derived fractional flow reserve; MB, myocardial bridge; MBMI, MB muscle index.

Table 5Comparison of FFRCT normal group and abnormal group in female MB patientsVariableSubgroupFFRCT ≤0.8 (n=37)FFRCT >0.8 (n=36)PBasic informationAge (years)53.24±9.2955.97±10.030.137BMI (kg/m2)24.02 (22.05–26.19)24.13 (22.93–25.68)0.873Risk factorsHypertension8 (21.6)9 (25.0)0.733Hyperlipidemia7 (18.9)3 (8.3)0.330Diabetes 3 (8.1)3 (8.3)>0.99Smoking3 (8.1)2 (5.6)>0.99Clinical symptomsAngina pectoris8 (21.6)4 (11.1)0.226Atypical angina4 (10.8)5 (13.9)0.965Chest pain12 (32.4)10 (27.8)0.665Asymptomatic13 (35.1)17 (47.2)0.294MB anatomyLength (mm)23.90 (17.85–39.65)15.55 (12.10–24.65)0.001Depth (mm)1.00 (1.00–2.55)1.00 (1.00–2.08)0.266Location (mm)33.70 (26.65–43.00)40.2±13.120.141Systolic compression index0.10 (0.00–0.15)0.08 (0.05–0.16)0.903MBMI33.46 (20.80–94.04)22.14 (12.98–33.54)0.004MB stenosis in diastole0.31±0.120.27±0.120.102MB stenosis in systole0.34±0.130.25 (0.16–0.37)0.024Data are presented as median (interquartile range), n (%), or mean ± standard deviation. BMI, body mass index; FFRCT, computed tomography-derived fractional flow reserve; MB, myocardial bridge; MBMI, MB muscle index.

Comparison of FFRCT normal group and abnormal group in female MB patients

Data are presented as median (interquartile range), n (%), or mean ± standard deviation. BMI, body mass index; FFRCT, computed tomography-derived fractional flow reserve; MB, myocardial bridge; MBMI, MB muscle index.

Analysis of factors influencing FFRCT abnormalities in female MB patients

The comparison of baseline characteristics between female MB patients with normal and abnormal FFRCT values is shown in Table 5. While no significant differences were observed in age, BMI, risk factors, or clinical symptoms between the two groups (P>0.05), three parameters showed statistically significant differences: MB length, muscle contraction index, and MB systolic stenosis rate (P<0.05). Binary logistic regression analysis that included these variables and other clinically significant factors (Table 6) demonstrated that MB length was an independent risk factor for FFRCT abnormalities in female patients (OR =1.113, 95% CI: 1.024–1.209; P=0.011). This corresponds to an 11.3% increase in FFRCT abnormality risk for every 1 mm increase in MB length among female patients.

Table 6Univariate and multivariate analyses of FFRCT abnormalities in MB patients of different gendersVariableUnivariateMultivariateOR (95% CI)POR (95% CI)PMale Angina pectoris0.343 (0.111–1.061)0.0570.351 (0.109–1.130)0.079 Length (mm)1.053 (1.006–1.102)0.0281.053 (1.004–1.104)0.033Female Age (years)0.970 (0.924–1.019)0.1370.962 (0.911–1.015)0.157 Length (mm)1.078 (1.025–1.133)0.0011.113 (1.024–1.209)0.011 Location (mm)0.971 (0.933–1.010)0.1410.975 (0.930–1.022)0.290 MBMI1.011 (0.999–1.023)0.0040.987 (0.967–1.007)0.190 MB stenosis in diastole20.570 (0.372–1,137.556)0.1020.126 (0.000–248.054)0.593 MB stenosis in systole38.744 (0.992–1,512.925)0.0247.832 (0.007–8,853.104)0.566FFRCT, computed tomography-derived fractional flow reserve; MB, myocardial bridge; MBMI, MB muscle index.

Univariate and multivariate analyses of FFRCT abnormalities in MB patients of different genders

FFRCT, computed tomography-derived fractional flow reserve; MB, myocardial bridge; MBMI, MB muscle index.

Comparison of FFRCT values between cardiac systolic and diastolic phases

Intra-group comparisons of FFRCT values between systolic and diastolic phases in the MB group, MBLA group, and control group are shown in Figure 3. No statistically significant differences in FFRCT values were observed between systolic and diastolic phases in any of the three groups (P>0.05).

Comparison of FFRCT values in the diastolic and systolic cardiac phase. (A) Control group; (B) MB group; (C) MBLA group. FFR1, FFRCT value at 10 mm proximal to MB; FFR2, the FFRCT value at MB; FFR3, the FFRCT value at 20–40 mm from the far end of MB. FFRCT, computed tomography-derived fractional flow reserve; MB, myocardial bridging; MBLA, myocardial bridge combined with left anterior descending artery atherosclerosis.

Discussion

This study, through CT-based AI quantitative analysis, innovatively and systematically reveals the significant role of sex differences in the FFRCT assessment of patients with MB of the LAD. The results not only confirm the substantial impact of MB on coronary hemodynamics but, more importantly, uncover sex-specific manifestations under different pathological conditions-namely, isolated MB versus MBLA. This provides a new theoretical foundation for the precise diagnosis and treatment of MB.

The most critical finding of this study is the first systematic identification of sex-based differences in CT-FFR assessment of LAD-MB, which expands on prior literature in three aspects: first, compared with Zhou et al. (6) and Zhang et al. (7) that primarily validated the feasibility of FFRCT in MB evaluation, we for the first time demonstrated that sex is an effect modifier of MB-induced hemodynamic impairment: isolated MB led to significantly lower distal FFRCT and larger trans-bridge pressure drop in females, a phenomenon unreported in previous MB functional studies. Second, distinct from Sun et al. (8) that mixed isolated MB and MBLA in analysis, we further found that this sex difference was completely masked when MB coexisted with atherosclerosis, indicating that plaque-related fixed stenosis overrides the dynamic hemodynamic effect of MB itself—a finding that explains the inconsistent sex-related results in prior MB studies with heterogeneous populations. Third, our sex-stratified regression analysis revealed that MB length exerted a significantly stronger effect on FFRCT abnormality risk in females (11.3% per mm) than in males (5.3% per mm), providing the first quantitative evidence for female-specific vulnerability to MB anatomical variations. This finding aligns with prior evidence that atherosclerotic stenosis dominates hemodynamic impairment (7), and further clarifies for the first time that this dominance eliminates the sex disparities observed in isolated MB.

The sex differences in the hemodynamic effects of MB may stem from multiple interacting factors. Firstly, hormonal variations may lead to differences in vascular reactivity; the protective effect of estrogen on vascular endothelial function could make females more sensitive to hemodynamic changes induced by MB (9). Secondly, sex-based differences in coronary microcirculation may affect the compensatory capacity for flow disturbances caused by MB (15). Additionally, the finding that MB length has a greater impact on females (an 11.3% increase in risk per mm) suggests that female coronary arteries may have lower tolerance for anatomical changes.

The results indicate that when MB is accompanied by atherosclerosis, sex differences are no longer significant, likely because atherosclerosis itself becomes the predominant hemodynamic influencing factor. Consistent with our multivariable analysis in this cohort, the loss of significance for MB anatomy after adjusting for plaque characteristics underscores a pathophysiological hierarchy: fixed atherosclerotic stenosis overrides the dynamic compression effect of the bridge. This highlights the importance of distinguishing between isolated MB and MBLA in clinical assessment. As highlighted by Fogante et al. (16), combining CCTA and stress computed tomography perfusion (CTP) provides a comprehensive assessment. Although FFRCT quantifies dynamic compression, CTP captures myocardial perfusion, which may be particularly relevant in atherosclerotic cohorts where MB anatomy is overshadowed by plaque burden. Notably, while atherosclerosis typically spares the tunneled segment, preferentially affecting the proximal LAD, our definition of “MB with atherosclerosis” included any plaque within the LAD. Future studies should investigate how plaque location relative to the MB influences hemodynamics.

The AI-based FFRCT analysis technology used in this study offers significant methodological advantages. Recent prospective comparisons of various CT-FFR interpretation methods have validated the robustness and diagnostic accuracy of such computational platforms (17). While traditional FFR measurement is invasive, the non-invasive FFRCT technique employed here maintains diagnostic accuracy while avoiding the risks associated with invasive procedures. By systematically measuring FFRCT values at different segments (proximal, within, and distal to the MB), this study provides a comprehensive assessment of the impact of MB on coronary blood flow. Furthermore, dual-phase analysis ensured result reliability, and the finding of no significant difference between systolic and diastolic FFRCT values further confirms the persistent effect of MB.

Compared to the study by Zhou et al. (6), which primarily validated the value of FFRCT in MB assessment, our research further reveals the influence of sex factors. Previous studies (7,8) mostly focused on the anatomical characteristics of MB or isolated functional assessments. This study innovatively incorporates sex factors into the analytical framework, offering a new perspective for understanding the heterogeneity of MB. The sex differences identified particularly in the isolated MB group provide important clues for understanding the diverse clinical manifestations of MB.

The findings of this study hold significant clinical translational value. Firstly, increased attention should be given to female MB patients, especially those with greater MB length, who may require closer follow-up. Secondly, clinical assessment of MB should distinguish between cases with and without concomitant atherosclerosis, as management strategies may differ. Additionally, FFRCT, as a non-invasive examination tool, can serve as an important method for the long-term follow-up of MB patients.

Summary of novel contributions

This study provides three novel insights into MB pathophysiology and management: (I) it is the first to report sex disparities in FFRCT profiles of LAD-MB, challenging the traditional perception of MB as a “sex-neutral” condition; (II) it clarifies that atherosclerosis co-existence eliminates MB-related sex differences, highlighting the necessity of stratifying MB patients by plaque burden in clinical assessment; (III) it identifies MB length as a sex-specific risk factor for hemodynamic impairment, supporting more aggressive risk stratification for females with long MB. These hemodynamic alterations carry direct clinical implications.

The observed FFRCT differences are clinically meaningful. Since an FFRCT ≤0.80 signifies hemodynamically significant ischemia, the lower distal FFRCT in females with isolated MB indicates a greater ischemic burden. This supports closer surveillance in this subgroup. While our cross-sectional design cannot establish causality for clinical outcomes, prior studies associate ischemic FFRCT in MB with adverse events (18-20). Future prospective studies are needed to confirm if these sex-specific FFRCT patterns predict long-term risk.

Some limitations existed in our study. First, the retrospective design inevitably introduced selection bias regarding sex distribution (MB and MBLA groups: 52.9–57.2% male vs. control: 24%). Although this aligns with clinical epidemiology-where females more frequently undergo CCTA for chest pain but often have normal findings-we performed multivariable adjustment and interaction analysis. While the interaction term did not reach statistical significance (P=0.459), likely due to limited power, the consistent trend of higher ORs in females supports our findings. Second, the single-centre design, modest cohort size, and inclusion of only Chinese patients may limit generalizability. Third, the limited number of events within specific subgroups resulted in extremely wide CIs for certain variables (systolic MB stenosis in females: OR =7.832, 95% CI: 0.007–8,853.104), reflecting statistical instability. Fourth, we did not account for the potential impact of nitroglycerin administration, which has been shown to significantly alter machine-learning FFRCT values in CCTA (21). Future prospective studies should consider standardizing vasodilator protocols to control for this variable.

Clinical Implications and management recommendations

Based on our findings, we propose sex-specific strategies for managing LAD-MB: risk stratification: due to the 11.3% increased risk per mm, women with isolated MB >20 mm warrant regular monitoring, including clinical reassessment and consideration of repeat FFRCT within 1–2 years. Diagnostic optimization: for females with angina but non-obstructive CCTA plaques, utilize FFRCT to differentiate ischemia driven by dynamic compression (MB) versus fixed stenosis (atherosclerosis). Therapeutic tailoring: symptomatic females with isolated MB and FFRCT ≤0.80 should receive beta-blockers or referral for surgical unroofing. Conversely, management for MBLA should align with standard coronary artery disease (CAD) guidelines (statins/anti-platelets), focusing on plaque rather than the bridge.

Conclusions

In conclusion, utilizing AI-based CT-FFR technology, this study demonstrates that sex significantly influences the hemodynamic assessment of LAD-MB, particularly in isolated MB. While atherosclerosis appears to mask these sex differences, MB length emerged as a potent risk factor, especially in females. These findings provide preliminary evidence to inform risk stratification and long-term management. Future integration with modalities like stress CTP may offer a more comprehensive assessment of MB.

Supplementary

The article’s supplementary files as

Acknowledgments

We would like to thank the Department of Radiology at The Second Hospital of Hebei Medical University for their support and assistance during data collection. Additionally, we acknowledge that the data collection for this study was completed during the corresponding author’s doctoral studies at The Second Hospital of Hebei Medical University.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. This study was approved by the Medical Ethics Committee of The Second Hospital of Hebei Medical University (approval No. W2023-R430). Informed consent was waived in this retrospective study.

Footnotes

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://qims.amegroups.com/article/view/10.21037/qims-2026-0713/rc

Funding: This study was supported by Medical Science Research Project of the Health Commission of Hebei Province (title: “Research on functional assessment of patients with myocardial bridge based on artificial intelligence technology”; ID: 20230739).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://qims.amegroups.com/article/view/10.21037/qims-2026-0713/coif). The authors have no conflicts of interest to declare.

Data Sharing Statement

Available at https://qims.amegroups.com/article/view/10.21037/qims-2026-0713/dss

References

Associated Data

Supplementary Materials

The article’s supplementary files as

Data Availability Statement

Available at https://qims.amegroups.com/article/view/10.21037/qims-2026-0713/dss