Background
Multiple studies have reported echocardiographically determined normal reference values for left ventricular (LV) mass (LVM) derived using fundamental imaging (FI). Modern ultrasound systems now use harmonic imaging (HI) because of the improved LV endomyocardial definition. However, the 2005 American Society of Echocardiography (ASE) recommendations noted that the applicability of the reference values to HI-derived measurements has not been established.
Methods
LVM and LV end-diastolic volume, diameter, and wall thickness were determined using HI in healthy subjects ( n = 251), including normal-weight (NW) (body mass index < 25 kg/m 2 , n = 149, 68% women) and otherwise healthy, overweight (OW) (body mass index ≥ 25 and <30 kg/m 2 , n = 102, 41% women) groups. Measurements were compared with ASE-endorsed reference values. The agreement between FI and HI was determined in a prospective cohort of 51 subjects.
Results
Two-dimensional (2D) derived LV volumes were similar between NW and OW subjects, although M-mode (MM)–derived LV diameters were slightly greater in OW subjects. 2D and MM-derived LVM was greater in OW compared with NW subjects, including after adjustment by height or height 2.7 ; however, indexing to body surface area eliminated these differences. The partition values for abnormal 2D and MM-derived LVM were generally greater in NW and OW subjects of both sexes compared with the ASE-endorsed values (except MM-derived LVM in NW men). However, there were no significant differences in LVM determined by HI compared with FI in a prospectively studied cohort.
Conclusions
Reference values for LVM derived from NW and OW cohorts are generally higher than the ASE-endorsed referenced values. The difference between NW and ASE-endorsed values is unlikely to result from the use of HI rather than FI, because there is excellent agreement between these two imaging modalities. This study emphasizes the need to update normal reference values to reflect modern imaging methods.
Multiple studies have shown that echocardiographically determined measurements of left ventricular (LV) mass (LVM) are predictive of cardiovascular morbidity and mortality. Normal reference values for LV geometry and LVM have been published by several investigators and summarized by the American Society of Echocardiography (ASE); however, these values were derived by use of fundamental imaging (FI). Harmonic imaging (HI) is now routinely used for echocardiography in adults because of superior endomyocardial definition; however, it has been reported to result in smaller LV end-diastolic diameter, which may result in greater measurements of LV wall thickness and LVM compared with FI. The use of HI to determine normal values of LVM and LV end-diastolic volume, diameter, and wall thickness in normal-weight (NW) and in otherwise healthy, overweight (OW) individuals has not been well investigated.
Echocardiographically derived measurements of LVM and LV end-diastolic diameter, volumes, and wall thickness are usually adjusted for body habitus to allow reference to normative values. Echocardiographic measurements of LVM are also important for risk stratification in OW and obese patients, including those without other overt cardiovascular risk factors.
The hypothesis of this study was that measurements of LVM and LV size and wall thickness determined by HI would differ between NW and OW subjects compared with reported normal reference values. This hypothesis was tested by three approaches. First, measurements of LVM and LV end-diastolic volume, diameter, and wall thickness were determined by HI in two groups of asymptomatic subjects without overt cardiovascular disease or cardiovascular disease–associated risk factors: (1) NW subjects (body mass index [BMI] < 25 kg/m 2 ) and (2) OW subjects (BMI, 25–29.9 kg/m 2 ). Second, these measurements were compared with a reference standard published by the ASE. Third, the agreement of LVM and LV end-diastolic volume, diameter, and wall thickness obtained by both HI and FI was prospectively assessed in a subgroup of ambulatory subjects.
Methods
Study Population
This was a single-center study of consecutive ambulatory subjects (aged 20–80 years) enrolled in an ongoing research study. Eligible subjects had no histories, symptoms, physical findings, or echocardiographic evidence of cardiovascular disease. A total of 1,293 subjects were evaluated by review of their medical histories, physical examination findings, and fasting blood chemistry obtained at the time of echocardiography. Systolic and diastolic blood pressures were measured after 15 min in the supine position by three consecutive recordings in both arms and averaged. Height and weight were measured using a calibrated stadiometer and scale. NW was defined as a BMI < 25 kg/m 2 and OW as a BMI ≥ 25 and <30 kg/m 2 . The study was approved by the Human Research Protection Office at Washington University; all subjects provided written informed consent.
Exclusion criteria were (1) BMI ≥ 30 kg/m 2 , (2) reported history of a cardiovascular event (i.e., heart failure, hospitalization for unstable angina or myocardial infarction, prior cardiovascular surgery, or cardiac catheterization demonstrating atherosclerosis), (3) reported history of cerebrovascular disease (e.g., stroke or transient ischemic attack), (4) hypertension (i.e., reported history, current antihypertensive drug therapy, or blood pressure ≥ 140/90 mmHg at the time of study), (5) diabetes (i.e., reported history or current antidiabetic drug therapy), (6) fasting glucose ≥ 100 mg/dL or insulin level > 15 μU/dL, or (7) fasting triglyceride level ≥ 150 mg/dL. In addition, subjects with echocardiographic evidence of LV wall motion abnormalities, cardiomyopathy, LV ejection fractions < 50%, or valvular stenosis and/or valvular regurgitation greater than mild were also excluded.
Echocardiography
All subjects were studied by use of an Acuson Sequoia C512 ultrasound system (Siemens Medical Solutions USA, Inc., Mountain View, CA). The echocardiographic examination was performed in all subjects at an HI frequency of 2.75 MHz. LV ejection fraction was derived from two-dimensional (2D) measurements of LV volumes (biplane method of disks in the apical four-chamber and two-chamber views). LV end-diastolic volume was indexed for body surface area (BSA). 2D LVM was determined by the area-length method and indexed for BSA.
2D guided M-mode echocardiography was performed in the cross-sectional view. Measurements included LV end-diastolic diameter and septal and posterior wall thickness by the leading-edge method. M-mode-derived LVM was calculated on the basis of the ASE-corrected method and indexed for BSA, height, and height 2.7 . Relative wall thickness was calculated as 2 × posterior wall thickness/LV end-diastolic diameter. All measurements were made by one observer and represent the average of at least three consecutive cardiac cycles.
A prospective substudy was conducted in an additional 51 consecutive subjects who underwent 2D and M-mode echocardiography using both HI and FI. Images were unpaired and randomly evaluated by a single observer in a blinded fashion for measurement of 2D LVM and M-mode-derived LV end-diastolic diameter, septal and posterior wall thickness, and LVM.
Prior studies from our laboratory have determined that the reproducibility for LVM is excellent, with interclass correlation coefficients of 0.85 to 0.90.
Statistical Analysis
Continuous variables are expressed as mean ± SD. Comparisons of echocardiographic variables between groups (i.e., NW vs OW subjects) were performed using unpaired Student’s t tests. Reference values for the NW and OW groups were determined to be greater than two SDs above the mean in each group. In the substudy of 51 subjects, interclass correlation coefficients were used to determine the agreement between measures derived using HI and FI, and Bland-Altman plots were generated to assess for systematic bias. All statistical analyses were performed using SAS version 9.2 (SAS Institute Inc., Cary, NC). Statistical significant was defined as P < .05.
Results
Of the 1,293 available study subjects, 251 (144 women, 107 men) qualified for the study on the basis of the inclusion and exclusion criteria. The racial composition of the study group was primarily Caucasian (91% of women, 92% of men). A majority of the women were NW ( n = 102 [71%]); OW women ( n = 42 [29%]) had higher systolic and diastolic blood pressures than NW women, although the values for both measurements were within normal limits ( Table 1 ). The majority of men were OW ( n = 60 [56%]) and had higher diastolic blood pressures compared with NW men ( n = 47 [44%]), although the values were within normal limits. Age, height, and heart rate were similar between NW and OW subjects regardless of gender.
| Women ( n = 144) | Men ( n = 107) | |||
|---|---|---|---|---|
| NW | OW | NW | OW | |
| Variable | ( n = 102) | ( n = 42) | ( n = 47) | ( n = 60) |
| Age (years) | 46 ± 14 | 47 ± 12 | 43 ± 16 | 47 ± 13 |
| Height (cm) | 163 ± 6 | 163 ± 6 | 179 ± 7 | 180 ± 7 |
| BSA (m 2 ) | 1.64 ± 0.12 | 1.78 ± 0.11 ∗ | 1.92 ± 0.14 | 2.07 ± 0.13 ∗ |
| BMI (kg/m 2 ) | 22.2 ± 1.8 | 27.1 ± 1.5 ∗ | 22.9 ± 1.7 | 27.4 ± 1.4 ∗ |
| Heart rate (beats/min) | 63 ± 9 | 66 ± 9 | 61 ± 10 | 63 ± 11 |
| SBP (mm Hg) | 109 ± 11 | 112 ± 9 † | 116 ± 9 | 116 ± 8 |
| DBP (mm Hg) | 71 ± 7 | 74 ± 7 † | 75 ± 6 | 77 ± 6 † |
Two-Dimensional and M-Mode Echocardiographic Measurements
2D LV end-diastolic volumes (including correction for BSA) and LV ejection fraction were similar between NW and OW subjects, regardless of sex. 2D LVM was significantly greater in OW women and men compared with the respective NW groups; however, this difference was not significant after adjustment for BSA ( Table 2 ).
| Women ( n = 144) | Men ( n = 107) | |||
|---|---|---|---|---|
| NW | OW | NW | OW | |
| Variable | ( n = 102) | ( n = 42) | ( n = 47) | ( n = 60) |
| 2D | ||||
| EDV (mL) | 76 ± 14 (43–102) | 80 ± 12 (60–109) | 103 ± 22 (70–151) | 107 ± 23 (71–163) |
| EDV/BSA (mL/m 2 ) | 46 ± 8 (29–67) | 45 ± 6 (37–59) | 54 ± 10 (35–76) | 52 ± 10 (35–74) |
| LVM (g) | 125 ± 24 (81–202) | 135 ± 18 ‡ (99–191) | 164 ± 31 (122–240) | 178 ± 30 ‡ (127–281) |
| LVM/BSA (g/m 2 ) | 76 ± 12 (52–111) | 76 ± 9 (61–96) | 86 ± 15 (65–118) | 86 ± 14 (56–130) |
| LVEF (%) | 64 ± 6 (53–78) | 64 ± 4 (52–74) | 62 ± 5 (52–72) | 61 ± 6 (51–75) |
| M-mode | ||||
| EDD (cm) | 4.5 ± 0.4 (3.8–5.6) | 4.7 ± 0.4 † (3.9–5.9) | 5.0 ± 0.4 (4.3–5.9) | 5.2 ± 0.5 ‡ (4.4–6.4) |
| EDD/BSA (cm/m 2 ) | 2.8 ± 0.2 (2.3–3.4) | 2.7 ± 0.2 † (2.1–3.2) | 2.6 ± 0.2 (2.3–3.1) | 2.5 ± 0.2 † (2.2–3.1) |
| VS thickness (cm) | 0.82 ± 0.09 (0.60–1.10) | 0.86 ± 0.10 ‡ (0.73–1.20) | 0.92 ± 0.09 (0.70–1.10) | 0.98 ± 0.12 † (0.76–1.20) |
| PW thickness (cm) | 0.83 ± 0.10 (0.65–1.30) | 0.88 ± 0.09 † (0.7–1.20) | 0.92 ± 0.08 (0.70–1.10) | 0.98 ± 0.10 † (0.76–1.16) |
| LVM (g) | 121 ± 27 (75–205) | 138 ± 23 ∗ (92–199) | 165 ± 29 (115–237) | 190 ± 39 ∗ (123–313) |
| LVM/BSA (g/m 2 ) | 74 ± 14 (48–118) | 78 ± 12 (55–108) | 86 ± 15 (57–118) | 92 ± 18 (61–145) |
| LVM/height (g/m) | 74 ± 15 (48–120) | 85 ± 13 ∗ (59–117) | 95 ± 15 (64–128) | 106 ± 21 ∗ (69–171) |
| LVM/height 2.7 (g/m 2.7 ) | 32 ± 6 (21–49) | 37 ± 6 ∗ (26–51) | 35 ± 6 (25–45) | 39 ± 8 ∗ (26–61) |
| RWT | 0.37 ± 0.05 (0.27–0.59) | 0.38 ± 0.06 (0.27–0.55) | 0.37 ± 0.04 (0.27–0.45) | 0.38 ± 0.05 (0.28–0.47) |
2D M-mode LV end-diastolic diameters were greater in OW women and men compared with NW subjects. However, when LV end-diastolic diameter was corrected for BSA, OW women and men had smaller LV end-diastolic diameters compared with NW subjects. LVM, LVM indexed to height, and LVM indexed to height 2.7 were greater in OW women and men compared with the respective NW groups, whereas LVM indexed to BSA and relative wall thickness were similar between groups.
Two-Dimensional and M-Mode Echocardiographic Measurements by HI Compared with ASE Reference Values
The measurements of LV geometry and LVM derived using HI in both NW and OW subjects were compared with ASE-reported normal reference values ( Table 3 ). The partition values for abnormal measurements were determined as greater than or equal to the mean plus two SDs. The partition values for 2D LV end-diastolic volume were similar in NW and OW women and smaller in NW and OW men compared with the ASE reference values; when indexed by BSA, the resulting values were all smaller than the ASE partition values, regardless of BMI or sex. 2D LVM values, even after adjustment for BSA, were uniformly greater in NW and OW men and women compared with the ASE reference values. The results suggest that 2D LVM by HI using the area-length method may result in high reference values compared with current ASE-reported values.
| Women | Men | |||||
|---|---|---|---|---|---|---|
| Variable | ASE Guidelines | NW | OW | ASE Guidelines | NW | OW |
| 2D | ||||||
| EDV (mL) | ≥105 | ≥104 | ≥104 | ≥156 | ≥147 | ≥153 |
| EDV/BSA (mL/m 2 ) | ≥76 | ≥62 | ≥57 | ≥76 | ≥73 | ≥72 |
| LVM (g) | ≥151 | ≥172 | ≥172 | ≥201 | ≥226 | ≥238 |
| LVM/BSA (g/m 2 ) | ≥89 | ≥100 | ≥94 | ≥103 | ≥115 | ≥114 |
| M-mode | ||||||
| EDD (cm) | ≥5.4 | ≥5.3 | ≥5.5 | ≥6.1 | ≥5.8 | ≥6.1 |
| EDD/BSA (cm/m 2 ) | ≥3.3 | ≥3.3 | ≥3.1 | ≥3.1 | ≥3.1 | ≥2.9 |
| LVM (g) | ≥163 | ≥173 | ≥184 | ≥225 | ≥223 | ≥269 |
| LVM/BSA (g/m 2 ) | ≥96 | ≥101 | ≥101 | ≥116 | ≥114 | ≥127 |
| LVM/height (g/m) | ≥100 | ≥103 | ≥111 | ≥127 | ≥123 | ≥148 |
| LVM/height 2.7 (g/m 2.7 ) | ≥45 | ≥44 | ≥48 | ≥49 | ≥47 | ≥55 |
| RWT | ≥0.43 | ≥0.47 | ≥0.49 | ≥0.43 | ≥0.45 | ≥0.48 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
