Quantitative MRI Values for Atrial and Ventricular Parameters in Altitude‐Dwelling Youth

ABSTRACT

Background

The hypoxic environment on the Qinghai Plateau significantly exacerbates myocardial remodeling, with sex/ethnicity‐specific variations in cardiac parameters among local youth. Current cardiac MRI reference values lack data from physiologically adapted populations on this plateau.

Purpose

To establish sex‐ and ethnicity‐specific normative reference values for cardiac structure and function in native Qinghai Plateau young adults and to analyze their correlations.

Study Type

Prospective.

Subjects

101 healthy young volunteers (15–24 years) residing on the Qinghai Plateau, including 44 males and 57 females; 37 Han people (20 males and 17 females), 20 Tibetan people (3 males and 17 females); the remaining 44 people are from the Hui, Tu, Salar, and Mongolian ethnic groups.

Field Strength/Sequence

Steady state free precession cine sequence at 3.0 T.

Assessment

Circle CVI42 software was used to quantify cardiac chamber structural and functional parameters from cine images. All parameters were normalized to body surface area (BSA).

Statistical Tests

Sex‐ and ethnicity‐specific reference ranges were defined as mean ± 1.96 SD. Sex/ethnicity group comparisons were assessed with independent samples t‐tests. Correlations between parameters were visualized through correlation heatmaps. A p value < 0.05 was considered significant.

Results

Compared to females, males exhibited significantly larger cardiac parameters, including left atrial (LA) and right atrial (RA) diameters (p < 0.05). In the four‐chamber view, the left atrial area was larger in males (end‐systolic 1688.57 ± 326.75 vs. 1562.68 ± 238.01 mm², p < 0.05; end‐diastolic 1178.18 ± 268.09 vs. 1018.81 ± 201.04 mm², p < 0.05). The right atrial area was larger in males (end‐systolic 1835.86 ± 360.73 vs. 1567.98 ± 276.84 mm², p < 0.05; end‐diastolic 1387.75 ± 319.39 vs. 1106.09 ± 228.78 mm², p < 0.05). Males also had significantly higher left ventricular end‐diastolic volume (LVEDV), left ventricular end‐systolic volume (LVESV), stroke volume (LVSV), cardiac output (LVCO), and left ventricular mass (LVM) (p < 0.05). After body surface area (BSA) normalization, males still had higher LVEDVi and LVESVi, while no significant difference was found in left ventricular ejection fraction (LVEF) (p = 0.334). Regarding ethnicity, there were no significant differences between Han and Tibetan participants in left ventricular volumes or systolic function (p > 0.05), but Han individuals had significantly higher left ventricular mass and mass index (LVM 93.37 ± 15.82 vs. 63.72 ± 11.97 g, p < 0.05; LVMI 44.60 ± 7.71 vs. 38.89 ± 4.72 g/m², p < 0.05). Correlation heatmaps showed significant positive correlations (p < 0.05) between body size (height, weight, BSA, BMI) and heart parameters. Left atrial 4‐chamber end‐diastolic area (LA Area 4ch D) was positively correlated with height (R = 0.66, moderate, p < 0.05), weight (R = 0.66, moderate, p < 0.05), BSA (R = 0.65, moderate, p < 0.05), and BMI (R = 0.62, moderate, p < 0.05). Right atrial 4‐chamber end‐diastolic area (RA Area 4ch D) showed similar correlations (R = 0.65–0.68, moderate, p < 0.05). LVEF was positively correlated with height (R = 0.45, weak, p < 0.05) and weight (R = 0.51, moderate, p < 0.05). Left ventricular mass (LVM) strongly correlated with BMI (R = 0.70, moderate, p < 0.05).

Conclusion

The results showed that plateau environment, gender differences, and ethnic differences are all influencing factors of atrioventricular parameters and emphasized the strong correlation between physical parameters and cardiac structure.

Stages of Technical Efficacy

Stage 1.

Evidence Level

Evidence Level 1.

Plain Language Summary

This study performed 3.0 T cardiac MRI on 101 healthy 15–24‐year‐olds living above 2500 m in Qinghai to compare sex and ethnic differences and establish preliminary high‐altitude cardiac MRI reference ranges. Results showed that high‐altitude youths had smaller cardiac volumes, mass, and stroke volume. Men had larger absolute chamber sizes but similar ejection fractions to women. Han participants showed greater left ventricular myocardial mass and mass index than Tibetans. Multiple cardiac parameters were positively correlated with body mass index, while correlations with ejection fractions correlations were weak. The study found that altitude, sex, and ethnicity influence cardiac parameters.

Abbreviations

BMI

Body mass index

BSA

Body surface area

CMR

Cardiovascular magnetic resonance

LA

Left atrial

LV

Left ventricular

LVCI

Left ventricular cardiac index

LVCO

Left ventricular cardiac output

LVEDV

Left ventricular end‐diastolic volume

LVEDVi

Left ventricular end‐diastolic volume index

LVEF

Left ventricular ejection fraction

LVESV

Left ventricular end‐systolic volume

LVESVi

Left ventricular end‐systolic volume index

LVM

Left ventricular msss

LVMi

Left ventricular mass index

LVSV

Left ventricular stroke volume

LVSVi

Left ventricular stroke volume index

RA

Right atrial

RV

Right ventricular

SSFP

Steady‐state free precession

1. Introduction

MRI plays an important role in diagnosing cardiac diseases and in prognostic stratification [1]. It is the reference standard for simultaneously evaluating biventricular volumes and ejection fraction [2]. Therefore, normal values for cardiac structure and function are important for diagnosing cardiac status [3].

Both ischemic and non‐ischemic cardiac injuries trigger pathological alterations in atrial/ventricular chamber size, morphology, and function. This adverse myocardial remodeling process is particularly pronounced under high‐altitude hypoxic conditions [4]. Additionally, while cardiac morphology varies by sex, age, and ethnicity [5], these differences need further exploration. Multiple current studies (including those from the US, Europe, and China) [6, 7, 8, 9, 10, 11] have established cardiac MRI normal reference values based on healthy subjects without cardiovascular disease. However, the cardiac structure and function of populations residing in the high‐altitude Qinghai‐Tibet Plateau continuously experience environmental influences under normal physiological conditions [12]. Sex‐ and ethnicity‐specific normal reference values reflecting ventricular structure and function in this high‐altitude acclimatized state are lacking. These values may be important for clinical practice among Chinese physicians [3] and could also provide data for cardiac MRI diagnosis in other high‐altitude regions.

Thus the aim of this study was to establish sex‐and ethnicity‐specific values for cardiac structure and function in native young adults on the Qinghai Plateau using 3.0 T MRI.

2. Materials and Methods

2.1. Study Population and Sample Size Determination

This study was approved by the Ethics Committee of Qinghai Provincial People's Hospital, and all participants provided written informed consent prior to enrollment. We recruited healthy young individuals who underwent cardiac magnetic resonance imaging (MRI) and met the following inclusion criteria: (1) at least three generations born and permanently residing on the Qinghai Plateau; (2) aged between 15 and 24 years; (3) no history of cardiac, cerebrovascular, or pulmonary diseases; and (4) residing at an altitude greater than 2500 m. Age was recorded via questionnaire, while height, weight, heart rate (HR), systolic blood pressure (SBP), diastolic blood pressure (DBP), and pulse pressure (PP) were measured on site. Body mass index (BMI) was subsequently calculated, and body surface area (BSA) was estimated using the Mosteller formula [13].

The sample size was determined a priori based on the precision requirement for estimating normative cardiac MRI parameters in a healthy population. Using previously reported standard deviations for left ventricular mass index and other key cardiac chamber metrics in young adults, the required sample size was calculated according to the formula:

$$n={\left(\frac{Z_{1-\mathrm{\alpha }/2}\times \sigma }{E}\right)}^2$$

where $Z_{1-\alpha /2}$ is the standard normal deviate for a 95% confidence level (1.96), $\sigma$ is the estimated standard deviation from prior literature, and E is the allowable error set at 10% of the expected mean.

For example, from previous studies, the standard deviation of left ventricular mass indexed to body surface area (LVMi) was approximately 12 g/m² [14]. Setting the allowable error at 2.5 g/m² to ensure sufficient precision for normative reference values, the calculated sample size was approximately 90 subjects. To account for potential exclusions due to image quality or incomplete demographic data, we targeted a slightly larger cohort and enrolled 101 participants. This sample size provides > 80% power to detect moderate effect sizes (Cohen's d ≈ 0.5) at α = 0.05 in subgroup comparisons, which is consistent with prior CMR normative studies [15].

2.2. Cardiac Magnetic Resonance Imaging

All examinations were performed using a Philips Ingenia Elition 3.0 T MRI scanner (Best, the Netherlands) [16] with a maximum gradient strength of 45 mT/m and a slew rate of 220 mT/m·s. A torso coil was utilized with respiration gating and prospective vectorcardiographic gating techniques. Pre‐scan breath‐holding training was conducted with subjects in the supine position, arms at sides, so that the heart was at the center of the scanner. Standard cardiac imaging planes included: short‐axis, two‐chamber, four‐chamber, three‐chamber, coronal, and right ventricular outflow tract views.

The MRI protocol included: (1) ECG‐gated HASTE dark‐blood sequence: field of view (FOV) 300 × 300 mm², TR 1967 ms, TE 39 ms, slice thickness 8 mm, matrix 152 × 120; and (2) Bright‐blood steady state free precession cine sequence: FOV 300 × 300 mm², TR 2.7 ms, TE 1.37 ms, slice thickness 8 mm, matrix 152 × 187, short axis slices 10–14, phases 25 per cardiac cycle.

2.3. MRI Analysis

All images were analyzed using CircleCVI42 (Version 6.1.1 (4035)), Circle Cardiovascular Imaging Inc., Calgary, Canada [17] post‐processing software. All post‐processing was independently performed by the two observers (X.M.X. and H.X.), who were both clinicians with over 3 and 15 years of cardiac MRI post‐processing, respectively.

1. Cardiac Structural Evaluation: The heart was divided into standardized 2002 American Heart Association defined segments [18]. Measurements were performed using the CVI viewer module (Figure 1). A standard steady‐state free precession (SSFP) [19] sequence was used to measure in four‐chamber (4 ch) view end‐systolic (ES)/diastolic (ED) left atrial (LA) transverse dimension (perpendicular to interatrial septum), longitudinal dimension (perpendicular to mitral valve), and area, plus right atrial (RA) transverse dimension (perpendicular to interatrial septum), longitudinal dimension (perpendicular to tricuspid valve), and area (LA T 4ch S/D, LA L 4ch S/D, LA Area 4ch S/D, RA T 4ch S/D, RA L 4ch S/D, RA Area 4ch S/D); in two‐chamber (2 ch) view at ES/ED LA transverse dimension (perpendicular to interatrial septum), longitudinal dimension (perpendicular to mitral valve), and area (LA T 2ch S/D, LA L 2ch S/D, LA Area 2ch S/D); in three‐chamber (3 ch) view at ES LA anteroposterior dimension (maximal linear distance between anterior/posterior (AP) walls at mid‐atrial level), lateral dimension (perpendicular to AP diameter midpoint), and left ventricular (LV) outflow tract (OT) diameter (1 cm below aortic valve) (LA AP 3ch S, LA LP 3ch S, LVOT 3ch S); in short‐axis maximal level ES/ED LV horizontal/vertical dimensions (LVESD Horizontal, LVESD Vertical, LVEDD Horizontal, LVEDD Vertical) and S/D right ventricular (RV) outflow tract diameter at ES/ED (1 cm below pulmonary valve) (RVOT S, RVOT D); LVOT diameter in the coronal view at ES (LVOT Cor S); finally left ventricular outflow tract at ES (LVOT S) calculated as the average of LVOT 3ch S and LVOT Cor S. All absolute dimensions were normalized to BSA to create index values, with index (abbreviated as i) added to the end of the absolute parameter. The mean LV wall thickness of 16 segments at ES/ED was obtained by specifying the superior and inferior right ventricular insertion points.

2. Cardiac Functional Evaluation: Utilizing CVI short‐axis and multi‐planar long‐axis modules, images were processed with input of height, weight, blood pressure, and heart rate. End‐diastolic and end‐systolic left/right ventricular endocardial and epicardial boundaries (excluding outflow tract levels) were defined through combined automatic detection and manual correction (Figure 2). Papillary muscles were included in ventricular blood pools; interventricular septum was excluded from ventricular mass calculations. The software automatically calculates left ventricular functional parameters, including end diastolic volume (LVEDV), end systolic volume (LVESV), stroke volume (LVSV), ejection fraction (LVEF), cardiac output (LVCO), and left ventricular myocardial mass (LVM). Normalization indices will be calculated based on BSA, including end diastolic volume index (LVEDVi) and end systolic volume index (LVESVi), stroke volume index (LVSVi), cardiac output index (LVCI), and left ventricular myocardial mass index (LVMi).

FIGURE 1.

Measurement of cardiac dimensions: In four‐chamber view end‐systolic (S)/diastolic (D) left atrial (LA) transverse dimension (perpendicular to interatrial septum), longitudinal dimension (perpendicular to mitral valve), and area, plus right atrial (RA) transverse dimension (perpendicular to interatrial septum), longitudinal dimension (perpendicular to tricuspid valve), and area (a, b) (LA T 4ch S/D, LA L 4ch S/D, LA Area 4ch S/D, RA T 4ch S/D, RA L 4ch S/D, RA Area 4ch S/D); in two‐chamber view S/D LA transverse dimension (perpendicular to interatrial septum), longitudinal dimension (perpendicular to mitral valve), and area (c, d) (LA T 2ch S/D, LA L 2ch S/D, LA Area 2ch S/D); in three‐chamber view S LA anteroposterior dimension (maximal linear distance between anterior/posterior (AP) walls at mid‐atrial level), lateral dimension (perpendicular to AP diameter midpoint), and left ventricular (LV) outflow tract (OT) diameter (1 cm below aortic valve) (E) (LA AP 3ch S, LA LP 3ch S, LVOT 3ch S); in short‐axis maximal level S/D LV horizontal/vertical dimensions (f, g) (LVESD Horizontal, LVESD Vertical, LVEDD Horizontal, LVEDD Vertical) and S/D right ventricular (RV) outflow tract diameter (1 cm below pulmonary valve) (h, i) (RVOT S, RVOT D); in coronal view S LVOT diameter (j) (LVOT Cor S); with final LVOT S calculated as the average of LVOT 3ch S and LVOT Cor S.

FIGURE 2.

Comparison of cardiac chamber dimension indices between genders (including absolute values and index values), where solid and dashed black lines represent median and interquartile ranges per group respectively, central boxes indicate overall sample means with width reflecting data density; with * indicates parameters showing significant inter‐gender distribution differences (p < 0.05), i denotes index values, Please refer to Figure 1 for abbreviations.

2.4. Inter‐ and Intra‐Observer Variability

Inter‐ and intra‐observer variability of left ventricular function and left ventricular cardiac dimension measurements was assessed in 20 randomly selected subjects; for intra‐observer variability, the same observer repeated measurements within 4 weeks, while for inter‐observer variability, two independent observers (H.X. with over 15 years of experience, X.M.X. with over 3 years) conducted blinded separate assessments of MRI examinations.

2.5. Statistical Analysis

Statistical analysis and graphing were performed using BM SPSS Statistics 26.0 (IBM, Armonk, NY, USA) [20] and OriginPro2024 software. Quantitative data are presented as mean ± standard deviation. Normal reference ranges for key parameters were determined as mean ± 1.96 SD stratified by sex and ethnicity. Inter‐group differences were assessed using independent samples t‐tests; Pearson correlation analyses evaluated relationships between cardiac structural/functional parameters, with results visualized via heatmaps—correlations were classified as strong (|R| ≥ 0.8), moderate (0.5 ≤ |R| < 0.8), weak (0.2 ≤ |R| < 0.5), or negligible (|R| < 0.2). Inter‐ and intra‐observer agreement was assessed using the Bland–Altman method, intraclass correlation coefficient (ICC), and coefficient of variation (COV). A p < 0.05 was considered statistically significant (Figure 3).

FIGURE 3.

Comparison of cardiac chamber dimension indices across ethnic groups (including absolute values and index values), where solid and dashed black lines represent median and interquartile ranges per group respectively, central boxes indicate overall sample means with width reflecting data density; with * indicating parameters demonstrating significant inter‐ethnic distribution differences (p < 0.05), i denotes index values, Please refer to Figure 1 for abbreviations.

3. Results

3.1. Baseline Data of Subjects in Qinghai, China

The study includes 101 healthy young volunteers (15–24 years) residing on the Qinghai Plateau, including 44 males and 57 females; 37 Han people (20 males and 17 females), 20 Tibetan people (3 males and 17 females); the remaining 44 people are from the Hui, Tu, Salar, and Mongolian ethnic groups (due to the scattered ethnic population, a single subgroup comparison could not be formed). No significant sex differences existed in heart rate (HR) (p = 0.157); no significant ethnic differences were observed in HR (p = 0.503) or pulse pressure (PP) (p = 0.726). Males had significantly higher height, weight, BSA, BMI, systolic blood pressure (SBP), diastolic blood pressure (DBP), and PP than females (Table 1) (all p < 0.05); Han participants had significantly greater height, weight, BSA, BMI, SBP, and DBP compared to Tibetans (all p < 0.05).

TABLE 1.

Baseline characteristics of the overall study population and gender/ethnicity.

	Overall (n = 101)	Male (n = 44)	Female (n = 57)	Han (n = 37)	Tibetan (n = 20)	Other ethnic groups (including Hui, Tu, Salar, and Mongolian) (n = 44)	P (Male vs. Female)	P (Han vs. Tibetan)
Age (years)	20.93 ± 1.94	21.68 ± 1.93	20.35 ± 1.76	21.51 ± 1.95	20.45 ± 1.64	20.66 ± 1.98	< 0.001	0.043
Height (cm)	168.69 ± 8.1	176.05 ± 4.82	163.02 ± 4.89	171.3 ± 7.78	164 ± 8.16	168.64 ± 7.5	< 0.001	0.002
Weight (kg)	59.71 ± 12.3	68.27 ± 9.88	53.11 ± 9.65	62.81 ± 12.73	51.6 ± 6.41	60.8 ± 12.57	< 0.001	< 0.001
BSA (m2)	1.67 ± 0.2	1.82 ± 0.14	1.55 ± 0.14	1.72 ± 0.2	1.53 ± 0.12	1.68 ± 0.19	< 0.001	< 0.001
BMI (kg/m2)	20.86 ± 3.32	22.05 ± 3.22	19.94 ± 3.13	21.27 ± 3.34	19.16 ± 1.72	21.29 ± 3.65	0.001	0.003
HR (bpm)	77.47 ± 10.12	75.84 ± 11.81	78.72 ± 8.49	76.92 ± 9.43	78.7 ± 9.69	77.36 ± 11.01	0.157	0.503
SBP (mmHg)	114.6 ± 12.01	121.8 ± 10.54	109.05 ± 10.02	117.59 ± 13.19	108.35 ± 11.06	114.93 ± 10.48	< 0.001	0.01
DBP (mmHg)	71.93 ± 9.28	75.43 ± 9.66	69.23 ± 8.07	75.41 ± 10.13	67 ± 7.21	71.25 ± 8.33	0.001	0.002
PP (mmHg)	42.67 ± 9.01	46.36 ± 9.23	39.82 ± 7.77	42.19 ± 8.7	41.35 ± 8.35	43.68 ± 9.61	< 0.001	0.726

3.2. Normal Reference Values for Gender and Ethnicity in Cardiac Atrioventricular Structure

Table 2 and Figure 2 shows that, compared between sexes, absolute cardiac dimensions were generally significantly larger in males. Specifically, various left atrial measurements demonstrated statistically significant differences, including the two‐chamber apical transverse systolic diameter (LA T 2ch S; 43.79 ± 5.13 mm for males vs. 40.37 ± 5.49 mm for females, p < 0.05), the two‐chamber apical longitudinal systolic diameter (LA L 2ch S; 39.08 ± 5.18 mm vs. 35.37 ± 5.27 mm, p < 0.05), and the corresponding two‐chamber systolic area (LA Area 2ch S; 1606.18 ± 289.63 mm² vs. 1367.35 ± 236.04 mm², p < 0.05). This pattern of males having significantly larger dimensions continued across diastolic measurements. Furthermore, the absolute values for the right atrial transverse systolic diameter (RA T 4ch S; 43.08 ± 4.37 mm vs. 37.61 ± 4.27 mm, p < 0.05) and area (RA Area 4ch S; 1835.86 ± 360.73 mm² vs. 1567.98 ± 276.84 mm², p < 0.05), as well as the short‐axis left ventricular horizontal/vertical diameters in systole and diastole, were also significantly larger in males. However, after adjusting these parameters for Body Surface Area (BSA), the nature of the differences shifted. Several indexed parameters were found to be significantly larger in females, including the indexed two‐chamber apical transverse systolic diameter (LA T 2ch Si; 24.08 ± 2.7 mm/m² for males vs. 26.2 ± 3.53 mm/m² for females, p < 0.05), the indexed left ventricular outflow tract systolic diameter (LVOT Si; 9.73 ± 1 mm/m² vs. 11.2 ± 1.21 mm/m², p < 0.05), and the indexed right ventricular outflow tract systolic diameter (RVOT Si; 9.26 ± 1.14 mm/m² vs. 10.22 ± 1.23 mm/m², p < 0.05). Additionally, other indexed parameters, such as the indexed four‐chamber left atrial systolic area (LA Area 4ch Si; 925.07 ± 154.71 mm²/m² for males vs. 1012.6 ± 139.86 mm²/m² for females, p < 0.05), were also significantly larger in females.

TABLE 2.

Absolute and standardized left and right atrioventricular dimensions in the total group (n = 101), in male (n = 44) and female (n = 57) subgroups, and in Han (n = 37) and Tibetan (n = 20) subgroups.

	Overall (n = 101)	Male (n = 44)	Female (n = 57)	Han (n = 37)	Tibetan (n = 20)	P (Male vs. Female)	P (Han vs. Tibetan)
LA T 2ch S (mm)	41.86 ± 5.58 (30.93, 52.79)	43.79 ± 5.13 (33.74, 53.84)	40.37 ± 5.49 (29.61, 51.13)	42.19 ± 5.97 (30.50, 53.89)	39.47 ± 4.04 (31.54, 47.39)	0.002	0.046
LA L 2ch S (mm)	36.99 ± 5.53 (26.16, 47.82)	39.08 ± 5.18 (28.92, 49.24)	35.37 ± 5.27 (25.04, 45.71)	37.58 ± 5.2 6 (27.27, 47.89)	35.15 ± 5.65 (24.08, 46.22)	0.001	0.11
LA Area 2ch S (mm2)	1471.4 ± 285.37 (912.07, 2030.72)	1606.18 ± 289.63 (1038.51, 2173.86)	1367.35 ± 236.04 (904.70, 1830.00)	1508.7 ± 296.94 (926.69, 2090.71)	1327.25 ± 217.93 (900.12, 1754.38)	< 0.001	0.02
LA T 2ch D (mm)	31.92 ± 4.76 (22.59, 41.26)	34 ± 4.39 (25.40, 42.59)	30.32 ± 4.44 (21.62, 39.03)	32.63 ± 4.78 (23.27, 41.99)	29.75 ± 2.75 (24.35, 35.15)	< 0.001	0.006
LA L 2ch D (mm)	31.66 ± 5.46 (20.97, 42.36)	34.41 ± 5.16 (24.30, 44.52)	29.54 ± 4.71 (20.30, 38.78)	32.65 ± 5.63 (21.62, 43.69)	29.08 ± 5.83 (17.66, 40.50)	< 0.001	0.028
LA Area 2ch D (mm2)	1005.03 ± 232.38 (549.57, 1460.49)	1139.59 ± 224.38 (699.80, 1579.38)	901.16 ± 180.44 (547.50, 1254.82)	1058.32 ± 258.06 (552.52, 1564.13)	865.85 ± 176.71 (519.50, 1212.20)	< 0.001	0.004
LA AP 3ch S (mm)	22.19 ± 3.97 (14.41, 29.97)	22.67 ± 4.07 (14.69, 30.64)	21.82 ± 3.89 (14.20, 29.45)	21.98 ± 4.42 (13.31, 30.65)	21.19 ± 3.14 (15.04, 27.34)	0.293	0.485
LA LP 3ch S (mm)	49.96 ± 5.86 (38.48, 61.44)	52.33 ± 5.92 (40.73, 63.93)	48.13 ± 5.15 (38.04, 58.21)	50.39 ± 6.12 (38.40, 62.38)	47.68 ± 6.55 (34.85, 60.51)	< 0.001	0.125
LVOT S (mm)	17.42 ± 1.68 (14.13, 20.71)	17.64 ± 1.5 (14.70, 20.59)	17.25 ± 1.79 (13.73, 20.76)	17.26 ± 1.15 (15.01, 19.51)	17.48 ± 2.31 (12.96, 22.00)	0.24	0.692
RVOT S (mm)	16.22 ± 2.07 (12.15, 20.28)	16.82 ± 2.11 (12.68, 20.96)	15.75 ± 1.94 (11.96, 19.55)	15.58 ± 2.15 (11.37, 19.78)	16 ± 2.06 (11.97, 20.03)	0.01	0.473
RVOT D (mm)	22.22 ± 4.1 (14.19, 30.25)	24.29 ± 4.51 (15.46, 33.13)	20.62 ± 2.9 (14.94, 26.30)	22.89 ± 4.93 (13.22, 32.57)	20.72 ± 2.79 (15.26, 26.17)	< 0.001	0.038
LA T 4ch S (mm)	37.53 ± 5.57 (26.61, 48.44)	38.47 ± 5.32 (28.04, 48.91)	36.8 ± 5.69 (25.64, 47.95)	38.34 ± 6.39 (25.82, 50.87)	35.77 ± 5.17 (25.63, 45.91)	0.134	0.128
RA T 4ch S (mm)	39.99 ± 5.08 (30.03, 49.95)	43.08 ± 4.37 (34.52, 51.64)	37.61 ± 4.27 (29.25, 45.97)	40.49 ± 5.3 (30.11, 50.88)	37.42 ± 3.95 (29.68, 45.15)	< 0.001	0.027
LA L 4ch S (mm)	45.43 ± 6.17 (33.35, 57.52)	47.38 ± 6.68 (34.29, 60.47)	43.93 ± 5.33 (33.49, 54.37)	46.55 ± 6.04 (34.71, 58.40)	42.65 ± 5.05 (32.75, 52.55)	0.005	0.017
RA L 4ch S (mm)	43.94 ± 5.13 (33.89, 53.99)	45.36 ± 5.63 (34.32, 56.40)	42.85 ± 4.45 (34.13, 51.57)	43.82 ± 5.21 (33.62, 54.02)	42.41 ± 5.1 (32.41, 52.41)	0.014	0.33
LA Area 4ch S (mm2)	1617.52 ± 285.6 (1057.74, 2177.31)	1688.57 ± 326.75 (1048.13, 2329.01)	1562.68 ± 238.01 (1096.18, 2029.19)	1662.51 ± 300.86 (1072.84, 2252.19)	1492.5 ± 173.6 (1152.25, 1832.75)	0.035	0.009
RA Area 4ch S (mm2)	1684.68 ± 341.6 (1015.15, 2354.22)	1835.86 ± 360.73 (1128.84, 2542.89)	1567.98 ± 276.84 (1025.38, 2110.58)	1669.3 ± 359 (965.66, 2372.94)	1565.85 ± 238.34 (1098.71, 2032.99)	< 0.001	0.253
LA T 4ch D (mm)	28.64 ± 4.96 (18.92, 38.36)	30.4 ± 5.35 (19.92, 40.88)	27.29 ± 4.2 (19.05, 35.52)	29.17 ± 4.41 (20.53, 37.82)	26.45 ± 4.18 (18.27, 34.63)	0.001	0.027
RA T 4ch D (mm)	36.28 ± 5.98 (24.56, 48.00)	39.21 ± 5.67 (28.09, 50.32)	34.02 ± 5.22 (23.79, 44.26)	37.07 ± 6.3 (24.72, 49.41)	33.31 ± 4.92 (23.66, 42.95)	< 0.001	0.024
LA L 4ch D (mm)	38.01 ± 6.11 (26.03, 49.99)	39.97 ± 6.36 (27.50, 52.44)	36.49 ± 5.5 (25.71, 47.28)	39.22 ± 6.6 (26.27, 52.16)	35.52 ± 5.29 (25.15, 45.88)	0.004	0.035
RA L 4ch D (mm)	36.39 ± 4.97 (26.64, 46.14)	38.7 ± 4.9 (29.11, 48.30)	34.6 ± 4.28 (26.22, 42.99)	37.14 ± 5.07 (27.20, 47.08)	34.58 ± 3.95 (26.85, 42.31)	< 0.001	0.056
LA Area 4ch D (mm2)	1088.24 ± 244.63 (608.76, 1567.72)	1178.18 ± 268.09 (652.73, 1703.63)	1018.81 ± 201.04 (624.77, 1412.84)	1140.92 ± 253.02 (645.01, 1636.83)	988.2 ± 135.18 (723.24, 1253.16)	0.001	0.004
RA Area 4ch D (mm2)	1228.79 ± 304.75 (631.47, 1826.11)	1387.75 ± 319.39 (761.74, 2013.76)	1106.09 ± 228.78 (657.68, 1554.49)	1260 ± 300.79 (670.45, 1849.55)	1106.45 ± 198.57 (717.25, 1495.65)	< 0.001	0.045
SA‐LVEDD (Horizontal) (mm)	43.79 ± 3.54 (36.84, 50.73)	45.21 ± 3.48 (38.38, 52.03)	42.69 ± 3.21 (36.40, 48.98)	44.13 ± 3.82 (36.64, 51.63)	42.35 ± 3.34 (35.81, 48.89)	< 0.001	0.085
SA‐LVEDD (Vertical) (mm)	53.94 ± 4.72 (44.69, 63.18)	55.45 ± 5.06 (45.52, 65.37)	52.77 ± 4.11 (44.72, 60.83)	54.83 ± 4.69 (45.65, 64.02)	52.34 ± 4.45 (43.63, 61.05)	0.004	0.056
SA‐LVESD (Horizontal) (mm)	32 ± 4.41 (23.35, 40.64)	33.29 ± 4.86 (23.77, 42.81)	31 ± 3.78 (23.59, 38.40)	32.2 ± 3.47 (25.39, 39.01)	30.65 ± 4.45 (21.92, 39.37)	0.009	0.15
SA‐LVESD (Vertical) (mm)	33 ± 4.52 (24.15, 41.86)	34.89 ± 4.97 (25.15, 44.64)	31.54 ± 3.53 (24.61, 38.47)	33.41 ± 3.26 (27.02, 39.79)	31.53 ± 3.67 (24.34, 38.71)	< 0.001	0.051
LA T 2ch Si (mm/m2)	25.28 ± 3.35 (18.71, 31.85)	24.08 ± 2.7 (18.78, 29.38)	26.2 ± 3.53 (19.29, 33.12)	24.64 ± 3.44 (17.90, 31.38)	25.8 ± 2.15 (21.58, 30.02)	0.001	0.178
LA L 2ch Si (mm/m2)	22.3 ± 3.09 (16.24, 28.37)	21.47 ± 2.56 (16.46, 26.48)	22.95 ± 3.34 (16.41, 29.48)	21.94 ± 2.8 (16.45, 27.43)	22.97 ± 3.42 (16.27, 29.66)	0.017	0.226
LA Area 2ch Si (mm2/m2)	882.8 ± 137.01 (614.26, 1151.33)	879.75 ± 134.84 (615.46, 1144.05)	885.15 ± 139.81 (611.13, 1159.17)	875.32 ± 133.07 (614.51, 1136.14)	864.84 ± 115.43 (638.58, 1091.09)	0.846	0.768
LA T 2ch Di (mm/m2)	19.29 ± 2.91 (13.59, 24.99)	18.76 ± 2.81 (13.26, 24.26)	19.7 ± 2.94 (13.93, 25.47)	19.07 ± 2.9 (13.38, 24.76)	19.46 ± 1.56 (16.41, 22.52)	0.108	0.576
LA L 2ch Di (mm/m2)	19.03 ± 2.69 (13.76, 24.30)	18.89 ± 2.55 (13.90, 23.88)	19.14 ± 2.81 (13.62, 24.65)	18.96 ± 2.45 (14.15, 23.77)	18.96 ± 3.39 (12.32, 25.61)	0.647	0.999
LA Area 2ch Di (mm2/m2)	601.55 ± 113.34 (379.40, 823.71)	625.59 ± 115.45 (399.30, 851.88)	583 ± 109.09 (369.18, 796.83)	612.4 ± 125.66 (366.10, 858.70)	563.56 ± 96.83 (373.77, 753.34)	0.061	0.137
LA AP 3ch Si (mm/m2)	13.37 ± 2.18 (9.10, 17.64)	12.41 ± 1.82 (8.84, 15.98)	14.12 ± 2.15 (9.90, 18.34)	12.79 ± 2.23 (8.42, 17.16)	13.93 ± 2.35 (9.32, 18.54)	< 0.001	0.077
LA LP 3ch Si (mm/m2)	30.16 ± 3.39 (23.51, 36.82)	28.78 ± 3.1 (22.71, 34.86)	31.23 ± 3.25 (24.86, 37.60)	29.44 ± 3.39 (22.79, 36.08)	31.15 ± 3.64 (24.03, 38.28)	< 0.001	0.081
LVOT Si (mm/m2)	10.56 ± 1.34 (7.94, 13.18)	9.73 ± 1 (7.77, 11.69)	11.2 ± 1.21 (8.83, 13.57)	10.14 ± 1.28 (7.63, 12.66)	11.42 ± 1.26 (8.95, 13.90)	< 0.001	< 0.001
RVOT Si (mm/m2)	9.8 ± 1.28 (7.29, 12.31)	9.26 ± 1.14 (7.01, 11.50)	10.22 ± 1.23 (7.81, 12.63)	9.1 ± 1.22 (6.71, 11.48)	10.45 ± 1.04 (8.42, 12.48)	< 0.001	< 0.001
RVOT Di (mm/m2)	13.36 ± 1.99 (9.46, 17.26)	13.32 ± 2.21 (9.00, 17.65)	13.38 ± 1.82 (9.81, 16.96)	13.25 ± 2.05 (9.23, 17.27)	13.54 ± 1.62 (10.37, 16.71)	0.886	0.587
LA T 4ch Si (mm/m2)	22.72 ± 3.71 (15.45, 29.99)	21.18 ± 2.98 (15.34, 27.03)	23.91 ± 3.8 (16.46, 31.35)	22.4 ± 3.58 (15.38, 29.42)	23.56 ± 4.36 (15.01, 32.10)	< 0.001	0.286
RA T 4ch Si (mm/m2)	24.07 ± 2.22 (19.71, 28.42)	23.69 ± 2.24 (19.30, 28.09)	24.36 ± 2.18 (20.08, 28.64)	23.59 ± 2.38 (18.94, 28.25)	24.47 ± 2.2 (20.15, 28.79)	0.137	0.179
LA L 4ch Si (mm/m2)	27.45 ± 3.84 (19.93, 34.97)	26.02 ± 3.34 (19.47, 32.58)	28.56 ± 3.85 (21.01, 36.11)	27.11 ± 2.7 (21.82, 32.40)	27.91 ± 3.16 (21.72, 34.10)	0.001	0.317
RA L 4ch Si (mm/m2)	26.55 ± 3.12 (20.43, 32.66)	24.9 ± 2.62 (19.76, 30.04)	27.82 ± 2.89 (22.16, 33.47)	25.56 ± 2.38 (20.89, 30.22)	27.86 ± 3.91 (20.20, 35.52)	< 0.001	0.023
LA Area 4ch Si (mm2/m2)	974.46 ± 152.15 (676.26, 1272.67)	925.07 ± 154.71 (621.83, 1228.30)	1012.6 ± 139.86 (738.47, 1286.72)	965.39 ± 132.53 (705.64, 1225.15)	980.49 ± 139.89 (706.32, 1254.67)	0.004	0.689
RA Area 4ch Si (mm2/m2)	1009.43 ± 152.79 (709.95, 1308.90)	1005.54 ± 169.6 (673.13, 1337.95)	1012.42 ± 139.93 (738.16, 1286.69)	965.36 ± 149.38 (672.57, 1258.16)	1024.69 ± 147.29 (736.00, 1313.37)	0.824	0.156
LA T 4ch Di (mm/m2)	17.29 ± 2.84 (11.71, 22.86)	16.74 ± 2.96 (10.94, 22.53)	17.71 ± 2.7 (12.42, 23.00)	17.07 ± 2.72 (11.73, 22.40)	17.35 ± 2.96 (11.56, 23.15)	0.088	0.714
RA T 4ch Di (mm/m2)	21.81 ± 2.92 (16.09, 27.54)	21.54 ± 2.82 (16.02, 27.07)	22.02 ± 3 (16.14, 27.91)	21.54 ± 2.87 (15.92, 27.16)	21.79 ± 3.11 (15.70, 27.89)	0.415	0.756
LA L 4ch Di (mm/m2)	22.93 ± 3.47 (16.12, 29.73)	21.94 ± 3.14 (15.79, 28.09)	23.69 ± 3.55 (16.73, 30.65)	22.83 ± 3.21 (16.53, 29.13)	23.23 ± 3.11 (17.13, 29.33)	0.011	0.654
RA L 4ch Di (mm/m2)	21.92 ± 2.51 (17.01, 26.83)	21.26 ± 2.32 (16.71, 25.81)	22.44 ± 2.54 (17.45, 27.42)	21.63 ± 2.37 (16.99, 26.28)	22.62 ± 2.33 (18.05, 27.19)	0.018	0.136
LA Area 4ch Di (mm2/m2)	652.57 ± 120.77 (415.87, 889.28)	645.11 ± 130.96 (388.44, 901.79)	658.33 ± 113.13 (436.59, 880.06)	662.3 ± 130.54 (406.45, 918.16)	646.76 ± 85.23 (479.72, 813.80)	0.588	0.634
RA Area 4ch Di (mm2/m2)	733.26 ± 135.16 (468.35, 998.17)	759.84 ± 156.17 (453.74, 1065.93)	712.74 ± 113.62 (490.04, 935.44)	727.33 ± 131.8 (469.00, 985.66)	721.84 ± 111.04 (504.20, 939.49)	0.082	0.875
SA‐LVEDDi (Horizontal) (mm/m2)	26.49 ± 2.54 (21.52, 31.46)	24.89 ± 2.07 (20.84, 28.94)	27.72 ± 2.16 (23.49, 31.95)	25.82 ± 2.49 (20.94, 30.70)	27.73 ± 1.99 (23.83, 31.63)	< 0.001	0.005
SA‐LVEDDi (Vertical) (mm/m2)	32.7 ± 3.86 (25.14, 40.26)	30.58 ± 3.44 (23.85, 37.32)	34.33 ± 3.36 (27.75, 40.91)	32.08 ± 2.97 (26.25, 37.90)	34.26 ± 2.66 (29.04, 39.48)	< 0.001	0.008
SA‐LVESDi (Horizontal) (mm/m2)	19.32 ± 2.49 (14.44, 24.19)	18.31 ± 2.53 (13.35, 23.27)	20.09 ± 2.17 (15.85, 24.34)	18.81 ± 1.93 (15.03, 22.59)	19.99 ± 2.19 (15.70, 24.27)	< 0.001	0.041
SA‐LVESDi (Vertical) (mm/m2)	19.94 ± 2.68 (14.69, 25.19)	19.23 ± 2.84 (13.65, 24.80)	20.5 ± 2.43 (15.74, 25.25)	19.54 ± 2.01 (15.59, 23.49)	20.59 ± 1.78 (17.09, 24.08)	0.017	0.057

Note: All data expressed as mean ± SD, with reference ranges defined as mean ± 1.96 SD; LA denotes left atrium, Area indicates area, L represents longitudinal dimension, T signifies transverse dimension, AP is anteroposterior dimension, LP denotes lateral dimension, RA indicates right atrium, LV represents left ventricle, ES is end systole, ED is end diastole, D stands for diameter, RV signifies right ventricle, 2ch/3ch/4ch/Cor/SA indicate parameters obtained from two‐chamber/three‐chamber/four‐chamber/coronal/short‐axis cine sequences. Index values represent BSA‐normalized raw data (i signifies index values).

Table 2 and Figure 3 shows that, comparing Han and Tibetan groups, significant structural differences were observed, particularly in the left heart and outflow tracts. The Han group generally exhibited larger absolute left atrial dimensions. For example, the two‐chamber apical transverse systolic diameter (LA T 2ch S; 42.19 ± 5.97 mm for Han vs. 39.47 ± 4.04 mm for Tibetan, p < 0.05) and the corresponding systolic area (LA Area 2ch S; 1508.7 ± 296.94 mm² vs. 1327.25 ± 217.93 mm², p < 0.05) were significantly larger in the Han group. This trend was also present in diastolic measurements, where the Han group's two‐chamber diastolic area (LA Area 2ch D; 1058.32 ± 258.06 mm² vs. 865.85 ± 176.71 mm², p < 0.05) was significantly larger. Similarly, the four‐chamber left atrial systolic area (LA Area 4ch S; 1662.51 ± 300.86 mm² vs. 1492.5 ± 173.6 mm², p < 0.05) was larger in the Han group. In contrast, after adjusting for BSA, several key parameters were significantly larger in the Tibetan group. These include the indexed left ventricular outflow tract systolic diameter (LVOT Si; 10.14 ± 1.28 mm/m² for Han vs. 11.42 ± 1.26 mm/m², p < 0.05), the indexed right ventricular outflow tract systolic diameter (RVOT Si; 9.1 ± 1.22 mm/m² for Han vs. 10.45 ± 1.04 mm/m², p < 0.05), and the indexed short‐axis left ventricular horizontal/vertical diameters during diastole (SA − LVEDDi (Horizontal), 25.82 ± 2.49 mm/m² vs. 27.73 ± 1.99 mm/m², p < 0.05; and SA‐LVEDDi (Vertical), 32.08 ± 2.97 mm/m² vs. 34.26 ± 2.66 mm/m², p < 0.05).

3.3. Gender and Ethnic Distribution of Atrioventricular Function

Table 3 and Figure 4 show that the male group had significantly larger LVEDV, LVESV, LVSV, LVCO, and LVM than the female group (LVEDV 141.36 ± 22.2 vs. 112.42 ± 17.0 mL; LVESV 56.36 ± 11.5 vs. 43.77 ± 9.43 mL; LVSV 85.14 ± 14.5 vs. 68.61 ± 11.6 mL; LVCO 5.37 ± 1.13 vs. 4.39 ± 0.84 L/min; LVM 100.67 ± 21.11 vs. 67.37 ± 15.82 g; all p < 0.05). After indexing to BSA, LVEDVi and LVESVi remained higher in males (77.49 ± 9.81 vs. 72.52 ± 9.08 mL/m², p < 0.05; 30.88 ± 5.53 vs. 28.22 ± 5.32 mL/m², p < 0.05), and LVMI was greater in males (48.39 ± 5.75 vs. 38.0 ± 3.74 g/m², p < 0.05). LVEF did not differ between sexes (60.22% ± 4.62% vs. 61.15% ± 4.90%, p = 0.334), and neither LVSVi (46.61 ± 6.64 vs. 44.28 ± 6.20 mL/m², p = 0.073) nor LVCI (2.94 ± 0.51 vs. 2.84 ± 0.50 L·min⁻¹·m⁻², p = 0.335) showed significant sex differences.

TABLE 3.

Absolute and standardized left ventricular function parameters in the total group (n = 101), in male (n = 44) and female (n = 57) subgroups, and in Han (n = 37) and Tibetan (n = 20) subgroups.

	Overall (n = 101)	Male (n = 44)	Female (n = 57)	Han (n = 37)	Tibetan (n = 20)	P (Male vs. Female)	P (Han vs. Tibetan)
LVEDV (ml)	125.03 ± 24.42 (77.16, 172.90)	141.36 ± 22.02 (98.20, 184.52)	112.42 ± 17.93 (77.27, 147.57)	130.49 ± 28.2 (75.23, 185.75)	113.96 ± 18.7 (77.30, 150.62)	< 0.001	0.625
LVESV (ml)	49.26 ± 12.11 (25.52, 72.99)	56.36 ± 11.55 (33.72, 79.01)	43.77 ± 9.43 (25.28, 62.26)	51.58 ± 13 (26.10, 77.07)	44.8 ± 12.7 (19.91, 69.70)	< 0.001	0.614
LVSV (ml)	75.75 ± 15.12 (46.12, 105.38)	85 ± 14.14 (57.28, 112.71)	68.61 ± 11.64 (45.79, 91.44)	78.91 ± 18 (43.62, 114.19)	69.06 ± 9.14 (51.15, 86.96)	< 0.001	0.739
LVEF (%)	60.74 ± 4.78 (51.37, 70.11)	60.22 ± 4.62 (51.16, 69.27)	61.15 ± 4.9 (51.54, 70.76)	60.49 ± 4.9 (50.89, 70.09)	61.16 ± 5.49 (50.40, 71.93)	0.334	0.639
LVCO (L/min)	4.82 ± 1.07 (2.71, 6.92)	5.37 ± 1.1 (3.21, 7.53)	4.39 ± 0.84 (2.74, 6.04)	5.02 ± 1.25 (2.58, 7.46)	4.5 ± 0.85 (2.82, 6.17)	< 0.001	0.102
LVM (g)	71.81 ± 18.67 (35.21, 108.41)	88.37 ± 13.86 (61.21, 115.52)	59.03 ± 9.68 (40.07, 78.00)	77.82 ± 20.66 (37.32, 118.32)	60.06 ± 11.69 (37.15, 82.97)	< 0.001	< 0.001
LVEDVi (ml/m2)	74.67 ± 9.69 (55.69, 93.65)	77.49 ± 9.81 (58.25, 96.72)	72.5 ± 9.08 (54.69, 90.30)	75.3 ± 11.03 (53.68, 96.92)	73.93 ± 7.94 (58.37, 89.48)	0.01	0.625
LVESVi (ml/m2)	29.38 ± 5.54 (18.51, 40.24)	30.88 ± 5.53 (20.05, 41.71)	28.22 ± 5.32 (17.80, 38.64)	29.7 ± 5.4 (19.11, 40.29)	28.89 ± 6.3 (16.54, 41.25)	0.016	0.614
LVSVi (ml/m2)	45.29 ± 6.47 (32.62, 57.97)	46.61 ± 6.64 (33.60, 59.61)	44.28 ± 6.2 (32.13, 56.43)	45.6 ± 7.89 (30.13, 61.07)	45.03 ± 4.85 (35.53, 54.53)	0.073	0.739
LVCI (L/min/m2)	2.88 ± 0.51 (1.89, 3.88)	2.94 ± 0.51 (1.93, 3.95)	2.84 ± 0.5 (1.86, 3.82)	2.9 ± 0.54 (1.83, 3.96)	2.93 ± 0.51 (1.93, 3.93)	0.335	0.809
LVMi (g/m2)	42.53 ± 6.99 (28.83, 56.23)	48.39 ± 5.75 (37.12, 59.66)	38 ± 3.74 (30.67, 45.33)	44.6 ± 7.71 (29.48, 59.71)	38.89 ± 4.72 (29.63, 48.15)	< 0.001	0.001

Note: All data expressed as mean ± SD, with reference ranges defined as mean ± 1.96 SD. Index values represent BSA‐normalized raw data (i signifies index values).

Abbreviations: CO, cardiac output; EDV, end‐diastolic volume; EF, ejection fraction; ESV, end‐systolic volume; LA, left atrium; LV, left ventricle; LVM, end‐diastolic left ventricular mass; SV, stroke volume.

FIGURE 4.

Comparison of cardiac chamber function between genders (including absolute value and index value), where solid and dashed black lines represent median and interquartile ranges per group respectively, central boxes indicate overall sample means with width reflecting data density; with * indicates parameters showing significant inter‐gender distribution differences (p < 0.05), CI denotes cardiac index, i signifies index values.

Table 3 and Figure 5 show that, compared with the Tibetan group, the Han group had significantly higher LVM and BSA‐indexed LVMi (LVM 77.82 ± 20.66 vs. 60.06 ± 11.69 g; LVMi 44.60 ± 7.71 vs. 38.89 ± 4.72 g/m²; both p < 0.05). The remaining functional and volumetric indices (LVEDV, LVESV, LVSV, LVEF, LVCO, LVEDVi, LVESVi, LVSVi, LVCI) showed no statistically significant differences between groups (all p > 0.05).

FIGURE 5.

Comparison of cardiac chamber function across ethnic groups (including absolute value and index value), where solid and dashed black lines denote group‐specific medians and interquartile ranges respectively, central boxes represent overall sample means with width indicating data density; *Indicates p < 0.05, CI signifies cardiac index, i denotes index values.

Additionally, Figure 6 showed the average LV 16‐segment wall thickness was 8.72 ± 1.56 (5.66, 11.77) mm at ES and 5.44 ± 1.01 (3.46, 7.43) mm at ED. Segments 8 and 9 exhibited the highest wall thickness at LVES: 9.92 ± 1.17 (7.64, 12.21) mm and 10.31 ± 1.36 (7.64, 12.98) mm overall, with males measuring 10.85 ± 0.86 (9.16, 12.53) mm and 11.37 ± 1.11 (9.19, 13.55) mm versus females at 9.21 ± 0.81 (7.61, 10.81) mm and 9.49 ± 0.9 (7.74, 11.25) mm; all 16 segments and their averages showed significantly higher values in males than females at both ED and ES. Tibetan measurements were 9.31 ± 0.52 (8.29, 10.33) mm and 9.41 ± 0.53 (8.38, 10.45) mm versus Han at 10 ± 1.07 (7.90, 12.10) mm and 10.42 ± 1.34 (7.79, 13.06) mm; Han subjects had significantly greater 16‐segment wall thickness and averages than Tibetans at both phases.

FIGURE 6.

Overall, sex‐ and ethnicity‐specific wall thickness of 16 segments (all p < 0.05).

3.4. Correlation Heat Map of Left Atrial and Ventricular Diameter and Function

Figure 7, the correlation heatmap of left atrial/ventricular dimensions and function, shows that most anthropometric measurements have significant positive correlations with cardiac dimensions (all significant results below are statistically significant, p < 0.05). Weight and BSA demonstrate the strongest associations with cardiac structure. BSA is strongly and positively correlated with LVM (R = 0.87) and has a moderately strong positive correlation with LVEDV (R = 0.76). Similarly, weight has a strong correlation with LVM (R = 0.83) and moderate correlations with LVESV (R = 0.61) and LVMI (R = 0.60). BMI also shows widespread weak to moderate positive correlations with cardiac parameters, including LVM (R = 0.61) and LVMI (R = 0.37). Notably, age shows weak or no correlation with multiple parameters, and the correlations between HR and atrioventricular parameters are largely non‐significant (p > 0.05).

FIGURE 7.

Correlation heatmap of left atrial and ventricular dimensions/function, with i representing index values, *Indicates p < 0.05.

Regarding hemodynamics, SBP has non‐significant or weak correlations with linear dimensional parameters (R ≈ −0.03 to 0.35). It shows weak to moderate positive correlations with absolute functional parameters (R ≈ 0.27 to 0.51), excluding LVEF (R = 0.07), with the strongest correlation being with LVM (R = 0.51). The correlations for diastolic blood pressure (DBP) are weaker, with some being non‐significant (p > 0.05). DBP is largely not significantly correlated with linear dimensions, and its most notable significant correlation with absolute functional parameters is a weak positive one with LVM (R = 0.36).

LVEDV is significantly correlated with left atrioventricular linear dimensions, showing weak to moderate positive correlations (R ≈ 0.38–0.72), with the highest correlation observed with LA Area 2ch S (R = 0.72). Similarly, LVESV also demonstrates significant weak to moderate positive correlations with left atrioventricular linear dimensions (R ≈ 0.34–0.73), peaking with SA‐LVESD (Horizontal) (R = 0.73). Interestingly, LVSV also has significant weak to moderate positive correlations with these dimensions (R ≈ 0.27–0.69), with the highest being with LA Area 2ch S (R = 0.69). However, LVEF is not significantly correlated with most left atrioventricular dimensions (p > 0.05). Where significant, the correlations are weakly negative (R ≈ −0.41 to −0.2), with the strongest being with SA‐LVESD (Horizontal) (R = −0.41). LVCO shows weak to moderate positive correlations with the linear dimensions (R ≈ 0.23–0.55), with the highest correlation found with LA Area 2ch S (R = 0.55). LVM has significant weak to moderate positive correlations with the linear dimensions (R ≈ 0.23–0.64), again with the highest being with LA Area 2ch S (R = 0.64).

After BSA adjustment, most cardiac functional parameters retained significant weak‐to‐moderate positive correlations with left atrioventricular dimensions. Specifically, LVEDVi, LVESVi, and LVMi showed R values roughly 0.24–0.67. LVSVi also exhibited a significant weak positive correlation (R ≈ 0.25–0.47), whereas LVCI was generally non‐significant (p > 0.05). Notably, BSA correction removed most associations between functional indices and baseline anthropometrics, except for LVMi, which remained weakly to moderately positively correlated (R ≈ 0.24–0.67).

3.5. Inter‐ and Intra‐Observer Variability in Measurement of Ventricular Function Parameters

Table 4 summarizes intra‐ and inter‐observer variability assessments, with all parameters demonstrating good reproducibility: inter‐observer coefficient of variation (COV) ranged from 4.8% to 11.97% and intraclass correlation coefficients (ICCs) were ≥ 0.85; intra‐observer results demonstrated higher consistency, with COV ranging from 2.59% to 10.15% and ICCs were ≥ 0.91.

TABLE 4.

Intra‐ and inter‐observer repeatability and reproducibility analysis.

	Intra‐observer				Inter‐observer
	Mean ± SD	COV (%)	ICC	95% CI	Mean ± SD	COV (%)	ICC	95% CI
LA T 2ch D	1.02 ± 1.27	4.03	0.95	0.886–0.981	0.37 ± 2.85	9.16	0.89	0.744–0.955
LA L 2ch D	1.32 ± 1.87	5.86	0.95	0.869–0.978	−0.43 ± 3.52	11.32	0.92	0.81–0.968
LA AP 3ch S	0.40 ± 1.01	4.76	0.97	0.93–0.989	0.30 ± 1.71	8.06	0.98	0.95–0.992
LA LP 3ch S	−0.82 ± 1.23	2.59	0.99	0.97–0.995	−0.22 ± 5.62	11.75	0.92	0.81–0.968
LVOT S	0.00 ± 0.70	3.99	0.93	0.831–0.972	−0.02 ± 1.25	7.15	0.91	0.786–0.963
RVOT S	1.33 ± 1.02	6.38	0.91	0.759–0.963	0.18 ± 0.82	5.33	0.95	0.879–0.98
RVOT D	1.44 ± 1.00	4.81	0.97	0.92–0.987	−0.19 ± 2.40	11.97	0.88	0.722–0.951
SA‐LVEDD (Horizontal)	−1.22 ± 3.23	7.55	0.93	0.832–0.972	0.14 ± 3.92	9.04	0.94	0.854–0.976
SA‐LVEDD (Vertical)	−2.02 ± 2.52	4.75	0.99	0.975–0.996	0.54 ± 4.81	8.86	0.91	0.788–0.964
SA‐LVESD (Horizontal)	−0.00 ± 2.50	8	0.97	0.926–0.988	0.61 ± 3.65	11.54	0.94	0.855–0.976
SA‐LVESD (Vertical)	1.51 ± 2.45	7.29	0.96	0.903–0.984	0.21 ± 3.61	10.94	0.85	0.66–0.938
LVEDV (ml)	−0.75 ± 8.59	6.86	0.96	0.898–0.983	0.28 ± 12.63	10.05	0.91	0.722–0.951
LVESV (ml)	−1.20 ± 4.30	8.65	0.97	0.927–0.988	−1.81 ± 2.98	6.04	0.88	0.926–0.988
LVSV (ml)	0.17 ± 6.80	9.04	0.91	0.78–0.962	−1.85 ± 6.00	8.08	0.97	0.926–0.988
LVEF (%)	−0.63 ± 1.78	2.96	0.95	0.888–0.982	0.43 ± 2.91	4.8	0.92	0.81–0.968
LVCO (L/min)	0.02 ± 0.15	3.12	0.99	0.975–0.996	0.13 ± 0.45	9.05	0.91	0.787–0.963
LVM (g)	−3.79 ± 6.93	10.15	0.94	0.861–0.977	0.52 ± 7.62	10.9	0.94	0.851–0.975

Abbreviations: 95% CI, 95% Confidence Interval; COV, coefficient of variation; ICC, interclass correlation coefficient.

4. Discussion

Although previous studies [8, 9, 21, 22, 23] are available for plain‐dwelling Chinese populations, high‐altitude environments induce cardiac adaptations [24]. Our research found that, compared to plain‐dwelling Chinese youth [25], a high‐altitude cohort from Qinghai had a similar LVESD (32.00 ± 4.41 vs. 33.5 ± 3.5 mm) but a significantly smaller LVEDD (43.79 ± 3.54 vs. 48.2 ± 3.9 mm). In a comparison with plain‐dwelling Chinese and Caucasian peers [26], the high‐altitude group showed no significant differences in LVMi (males: 48.39 vs. 52 vs. 55 g/m²; females: 38.03 vs. 39 vs. 44 g/m²), LVESVi (males: 30.88 vs. 31 vs. 38 mL/m²; females: 28.22 vs. 26 vs. 30 mL/m²), or LVEF (males: 60.22 vs. 62 vs. 59%; females: 61.15 vs. 65 vs. 62%). However, the group's LVSVi (males: 46.61 vs. 52 vs. 55 mL/m²; females: 44.28 vs. 49 vs. 49 mL/m²) and LVEDVi (males: 77.49 vs. 82 vs. 92 mL/m²; females: 72.50 vs. 75 vs. 79 mL/m²) were both significantly smaller. The significant reduction in both LVEDD and LVEDVi indicates that the differences are primarily due to restricted diastolic filling rather than a change in systolic function [27], while the lowest LVMi also shows that the heart did not undergo compensatory hypertrophy. Most importantly, despite having smaller cardiac volume, mass, and stroke volume, the LVEF was maintained within the normal range. We propose this as strong evidence that myocardial pumping efficiency remains intact. This differs from the LVEF decline typically seen in heart failure, suggesting that the hearts of this population have successfully adapted to the high‐altitude environment, achieving a new healthy physiological equilibrium within a smaller functional capacity.

Sex is a key determinant of cardiac structure and function. While men naturally have larger absolute cardiac dimensions, such as LVEDV and LVM [28, 29], normalizing these parameters for body surface area (BSA) reveals a more complex landscape of sex‐based differences. A key finding in this study is that even after BSA correction, male LVEDVi, LVESVi, and LVMi remained significantly higher. This is consistent with research in specific populations, like athletes, where indexed male cardiac parameters also remain greater than those of females despite common training stimuli [29]. However, a noteworthy discrepancy emerges when this finding is placed in the broader context of reference value studies, particularly those involving healthy Asian populations. Le et al. [22] and Zhuang et al. [28], in their respective 3 T CMR studies on healthy Singaporean and mainland Chinese populations, both found that BSA normalization could significantly “attenuate” or even completely eliminate the vast majority of sex‐based differences in cardiac parameters. This discrepancy between our study and these large‐scale Asian cohort studies may arise from a confluence of factors, including ethnic background and subtle methodological differences (e.g., in image post‐processing) [30]. These complex and at times seemingly contradictory findings precisely highlight the profound biological mechanisms underlying cardiac sex dimorphism, the causes of which extend far beyond the intuitive factor of males' generally larger body size [28]. While some sex‐dimorphic features are visually apparent [31] (e.g., body size differences, where males are typically larger, constitute one of the primary factors leading to sex differences in absolute cardiac parameter values), many “invisible” factors play an equally, if not more, important role. Firstly, the regulatory effects of sex hormones are crucial; estrogen and testosterone exert opposing effects on the process of myocardial remodeling, shaping the response of the male and female heart to various physiological and pathological stimuli at a molecular level [32]. Secondly, the contribution of genetic background is not to be overlooked; large‐scale genomic studies have confirmed a genetic basis for cardiac functional traits, and these genetic effects may exhibit sex specificity [33]. Furthermore, differences in body composition, such as the muscle‐to‐fat ratio, also influence cardiac morphology, particularly in pathological states like hypertension [34]. In summary, the systematic differences in cardiac structure and function between males and females are the ultimate manifestation of a complex interplay between body size, sex hormones, genetic background, and body composition. Therefore, in clinical practice, it is imperative to use appropriate normalization methods, such as BSA, and to interpret the results in the context of sex‐specific normal reference ranges to make an accurate assessment of cardiac structure and function.

Ethnicity is a key determinant of cardiac structure and function, profoundly reflecting the unique evolutionary adaptive pathways forged by populations under different environmental pressures. Consistent with numerous previous studies based on CMR [3, 35], significant differences in cardiac morphology exist among various ethnic groups. In the present study, all significant (p < 0.05) absolute indices of cardiac structures, including LA Area 2ch S, RA T 4ch S, and the critically important LVM, were significantly greater in the Han population than in the Tibetan population [3]. However, after standardizing these parameters by BSA, the pattern of differences underwent a complex and profound shift. Specifically, while the LVMi remained significantly higher in Han individuals post‐correction, a notable change occurred: parameters that were not significant in their absolute values became significant after BSA correction. Specifically, LVOT Si, RVOT Si, and SA‐LVEDDi were, in contrast, significantly larger in the Tibetan population than in the Han population. This finding aligns with other research; for instance, a meta‐analysis indicated that East Asian populations (including Chinese) generally have lower left ventricular mass than Caucasians [2], which is consistent with the trend of lower LVMi we observed in the Tibetan group. However, the “relative advantage” of Tibetans in specific indexed internal diameters is a unique adaptive feature not commonly reported in comparisons of lowland populations, highlighting the fine‐tuned sculpting of cardiac geometry by long‐term high‐altitude selective pressures. These ethnic differences stem from the interplay of multiple factors. The most direct is the difference in hemodynamic load: as migrants to the plateau, the Han population exhibits a stronger hypoxic pulmonary vasoconstriction (HPV) response. This leads to a long‐term increase in right ventricular afterload, forcing the heart to undergo high‐energy‐consumption compensatory remodeling by increasing its overall mass and size [36]. A deeper reason lies in the decisive role of genetic background. Unique genetic variants in Tibetans (such as in the EPAS1 and EGLN1 genes) fundamentally blunt their HPV response, alleviating the pressure load on the heart and laying the foundation for developing a low‐mass, high‐efficiency cardiac structure [37]. Furthermore, differences in metabolic efficiency are crucial. Research has confirmed that the myocardial cells of Tibetans (Sherpas) possess higher oxygen utilization efficiency—for instance, by downregulating high‐oxygen‐cost fatty acid oxidation—thereby shaping a more economical operational model for the heart in a hypoxic environment at the molecular level [38].

In addition, the most basic and core relationship presented in the correlation heat map is the strong positive correlation between anthropometric indicators (especially BSA and body weight) and absolute left ventricular size (LVM, LVEDV). In any population, this reflects a basic physiological principle: the size of the heart must match the individual's physique to meet the metabolic needs of the whole body. This is also the fundamental reason why BSA must be used for standardization in clinical practice [7, 28]. However, in the context of the plateau population in this study, we believe that the significance of this relationship goes far beyond the simple logic of “big body, big heart”. It should be interpreted as a fundamental adaptive expansion: in order to compensate for the decline in blood oxygen carrying capacity, the heart must increase its LVM and LVEDV to increase stroke volume, thereby ensuring tissue perfusion in a hypoxic environment [39]. Therefore, this strong correlation with physique is essentially the morphological basis of physiological, non‐pathological cardiac remodeling in plateau populations. At the same time, the weak correlation (or even no correlation) between LVEF and absolute heart size indicates that LVEF, as an indicator of pumping efficiency, is independent of heart size [40]. In the plateau population, this phenomenon is the core evidence of successful adaptation. It shows that while the heart significantly increases its volume and mass, its intrinsic pumping efficiency is not impaired, and it successfully avoids systolic decompensation caused by overexpansion. This is the key to distinguishing physiological “athlete's heart” remodeling from pathological heart failure [41]. In addition, we also found a more special phenomenon: LVMi still has a strong correlation with weight and BMI after BSA correction, which suggests that BSA cannot fully explain the independent effect of body composition such as obesity on myocardial mass [42]. In the plateau population, this residual correlation may have a more special meaning. It may suggest that a specific body composition (such as higher muscle mass rather than fat mass) is more conducive to the heart in carrying out effective adaptive remodeling in the plateau environment, and this effect cannot be fully covered by the simple body surface area.

5. Study Limitations

This study has several limitations. First, the limited random sample size and gender imbalance within ethnic subgroups (particularly among Tibetan men) weakened the statistical power of interethnic comparisons. Second, the large number of ethnic groups and the small number of participants within each ethnic group precluded independent subgroup comparisons. Third, the narrow age range (15–24 years) limits the generalizability of the reference values. Finally, the lack of a contemporaneous control group from the plains makes the observed differences between the plateau and plains essentially correlational, potentially subject to confounding by unmeasured factors such as lifestyle.

6. Conclusions

This study provides preliminary evidence for establishing normal cardiac magnetic resonance imaging reference ranges for adolescents of different genders and ethnicities living on the Qinghai Plateau, confirming that plateau environment, gender, ethnicity, and body size are key factors influencing cardiac structure. The study also found that adolescents living on the plateau exhibit a “small size, high efficiency” physiological adaptation pattern; that gender differences in cardiac volume and mass are greater in males; and that Han adolescents have a higher left ventricular mass index, reflecting different ethnic adaptation strategies.

Author Contributions

X.M.X. was in charge of data acquisition, analysis, and interpretation. H.X. critically reviewed and revised the manuscript and supervised the study. H.X. and X.M.X. made a contribution to study conception and design. All authors contributed to the manuscript preparation, read, and approved the final text.

Ethics Statement

The study has been performed in accordance with the Declaration of Helsinki. The study was approved by the Ethics Committee of Qinghai Provincial People's Hospital, with all participants providing informed consent for CMRI.

Consent

The authors have nothing to report.

Conflicts of Interest

The authors declare no conflicts of interest.

Xiao X. and Xu H., “Quantitative MRI Values for Atrial and Ventricular Parameters in Altitude‐Dwelling Youth,” Journal of Magnetic Resonance Imaging 63, no. 3 (2026): 828–845, 10.1002/jmri.70132.

Data Availability Statement

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.