Study Population

The third wave of the Nord-Trøndelag Health Study (HUNT) in Norway was conducted from 2006 to 2008. A total of 93,210 adults aged ≥ 20 years were invited, and 50,839 (54%) participated. Within the third wave of HUNT, the echocardiography study was conducted among participants in predetermined communities. To be eligible, subjects had to be free from known cardiovascular disease, diabetes, or hypertension. A total of 1,296 participants who met the inclusion criteria were randomly selected. In 30 participants, we discovered pathology on echocardiography that could influence the deformation analysis, and these 30 participants were therefore excluded. Validation of the inclusion criteria was performed by the physician echocardiographer (H.D.), who also obtained the medical histories of the participants. The echocardiography study was approved by the Regional Committee for Medical Research Ethics and conducted according to the second Declaration of Helsinki. Written informed consent was obtained from all participants.

Anthropometrics and Laboratory Tests

The HUNT included self-administered questionnaires (medical history and smoking habits), clinical measurements (anthropometry and blood pressure), and blood samples, as previously described. Renal function was assessed by the Modification of Diet in Renal Disease equation to calculate the estimated glomerular filtration rate (eGFR), whereby reduced renal function is usually indicated by values <60 mL/min/1.73 m ² . Smoking information was categorized as never, former, occasional, or current smoking, and the latter three categories were grouped into a category of “ever smoking” in the analyses. Blood pressure was measured three times by trained staff members using a Dinamap 845XT (GE Healthcare, Milwaukee, WI). Measurements were made after 2 min of rest with the arm on a table, and the average of the second and third measurements was used in the analyses. Nonfasting blood samples were collected at study attendance, centrifuged, and placed in a refrigerator before transportation to the IEC 17025–accredited laboratory at Levanger Hospital on the same day. Serum analyses were performed on fresh blood samples using an Architect ci8200 (Abbott Laboratories, Abbott Park, IL). All analyses were performed by photometric methods. The coefficients of variation at the laboratory were 1.4% to 1.7% for glucose, 1.1% to 1.3% for cholesterol, 1.0% to 1.7% for high-density lipoprotein (HDL) cholesterol, and 1.9% to 2.4% for creatinine.

Echocardiographic Acquisition, Analysis, and Reproducibility

All examinations were conducted by one experienced physician echocardiographer (H.D.), and participants were examined in the left lateral decubitus position with a Vivid 7 scanner (version BT06; GE Vingmed Ultrasound AS, Horten, Norway) using a phased-array transducer (M3S and M4S).

Blood flow Doppler recordings were done in four-chamber and five-chamber views as described in the recommendations of the American Society of Echocardiography and the European Association of Echocardiography. Mitral flow was measured with sample volume between the mitral leaflets and analyzed for early (E) and late (A) diastolic filling, and the E/A ratio was calculated. LV internal dimensions were analyzed on parasternal M-mode echocardiograms with the ultrasound beam at the tip of the mitral leaflet. Longitudinal end-systolic strain, defined as the percentage shortening of myocardial segments during contraction, was analyzed semiautomatically by a combination of tissue Doppler and speckle tracking using customized software (GcMat; GE Vingmed Ultrasound AS) that runs on a MATLAB platform (The MathWorks, Inc., Natick, MA). Seven search boxes sized 5 × 5 mm defining segment borders were tracked with tissue Doppler along the ultrasound beam and grayscale speckles perpendicular to the ultrasound beam. Segmental strain was calculated from the variation of segment length. Because the 18-segment model gives too much weight to the apex for global strain measurements, global longitudinal end-systolic strain was analyzed as the global average in a 16-segment model of the left ventricle. Strain values from the inferolateral and anterolateral apical segments, and the inferoseptal (septal) and anteroseptal apical segments were averaged to refer to the lateral and septal apical segments, respectively. Mitral and tricuspid annular systolic and early diastolic velocities were assessed by pulsed-wave tissue Doppler (pwTD) using a Vivid 7 scanner (version BT06). Peak systolic (S′) and peak early diastolic (e′) velocities were measured in the base of the lateral, septal, anterior, and inferior LV wall, and the average of the four myocardial walls is presented as a global measure of LV function. Right ventricular (RV) function was assessed by pwTD by measuring S′ and e′ in the base of the RV free wall at the tricuspid annulus. The localizations used for measurement of S′ and e′ are shown in Figure 1 . All Doppler and tissue Doppler measurements reflect the average of three consecutive cardiac cycles.

Echocardiographic measurements and study population. (A) Mitral annular velocity curve from the septum by pwTD echocardiography in the apical four-chamber view. The sample volume is positioned close to the septal insertion of the mitral leaflets, and peak systolic velocity (S′), early diastolic velocity (e′), and late diastolic velocity (a′) were measured at the maximum of the solid Doppler spectrum with low gain setting. (B) Age distribution of the participants. (C) Strain curves from one myocardial wall. End-systolic strain (S _es ) is marked by blue arrows. Ecg , Electrocardiogram.

Reproducibility of the echocardiographic analyses has previously been described. Briefly, the interobserver mean error (the absolute difference between two measurements on separate recordings divided by the mean of the measurements) of the indices of cardiac function was 4% to 8% in the left ventricle and 12% to 14% in the right ventricle. The intraobserver mean error was 2% and 4%, respectively.

Other Measures of LV Systolic Function

Association of cardiac risk factors with peak systolic longitudinal strain rate, peak systolic mitral annular velocities by color tissue Doppler, and EF are shown in the Supplementary Data only. However, except for EF, these data were in line with those presented in the “Results” section.

Statistical Analysis

Because all cardiac function indices were mutually correlated (all Pearson’s R values ≥ 0.20, all P values < .001), associations between cardiovascular risk factors and cardiac function were estimated using multivariate linear regression analyses, with the different cardiac function measures as dependent variables. The cardiac function measures were log transformed, and the regression coefficients are presented as the percentage difference in cardiac function per standard deviation higher risk factor level with the corresponding 95% confidence intervals. In the analyses, age was included as a continuous covariate, and in the analyses of lipids and glucose, we also included time since last meal as a covariate, because the blood samples were nonfasting. We used fractional polynomial regression analyses to model associations of the risk factors with cardiac function ( Figures 2 and 3 , Supplementary Figures 1–3 ). In tables and illustrations, we have used the absolute value of negatively denominated echocardiographic indices of cardiac function. Statistical analyses were performed using SPSS for Windows version 16.0 (SPSS, Inc., Chicago, IL) and Stata for Windows version 10 (StataCorp LP, College Station, TX).

Association of BMI with longitudinal LV end-systolic strain. Age-adjusted fractional polynomial regression plot of global longitudinal LV end-systolic strain by BMI. Estimated mean ( line ) and 95% confidence interval ( shadow ) are displayed. Blue lines refer to men and red lines to women.

Association of diastolic blood pressure with longitudinal LV end-systolic strain. Age-adjusted fractional polynomial regression plot of global longitudinal LV end-systolic strain by diastolic blood pressure. Estimated mean ( line ) and 95% confidence interval (CI) ( shadow ) are displayed. Blue lines refer to men and red lines to women.

Associations of Risk Factors with Conventional Measures of Cardiac Function

Table 2 shows that mitral inflow E/A ratio was lower with older age, higher BMI, higher systolic and diastolic blood pressure, higher non-HDL cholesterol, and lower HDL cholesterol and was also lower among smokers ( P = .18 for HDL cholesterol in women, P = .13 for male smokers, and all other P values ≤ .07). Indexed LV mass was higher with older age, higher blood pressure, and higher BMI. Age influenced all measurements, but for LV outflow tract velocity-time integral and mitral peak E velocity, the associations with other risk factors were weak and not consistent. There were no clear associations of any risk factor with cardiac function, as measured by fractional shortening.

Associations of Risk Factors with New Measures of Cardiac Function

In both men and women, older age, higher BMI, higher systolic and diastolic blood pressure, and higher non-HDL cholesterol were all associated with reduced indices of LV systolic and diastolic function ( Table 3 ). Ever smoking was associated with reduced indices of LV systolic and diastolic function in women but not in men, whereas lower eGFR was associated with reduced indices of LV systolic and diastolic function in men but not in women. In both men and women, higher HDL cholesterol was associated with better LV systolic and diastolic function. Thus, the overall results showed that LV strain was reduced by approximately 5% per 5 kg/m ² increase in BMI and reduced by approximately 4% per 10 mm Hg increase in diastolic blood pressure. The corresponding reductions in early diastolic LV function were approximately 7% for both BMI and diastolic blood pressure. In multivariate analyses, we explored whether these associations could be influenced by potentially confounding factors, including heart rate, blood pressure, and body surface area, but the results were not substantially altered after multivariate adjustment (results not tabulated). In a sensitivity analysis of an extremely selected population, we excluded participants with systolic blood pressure ≥160 mm Hg, BMIs ≥30 kg/m ² , nonfasting glucose ≥7.8 mmol/L, and indexed LV mass ≥109 g/m ² (women) and 132 g/m ² (men), leaving 974 individuals (514 women and 463 men). The results of these analyses ( Supplementary Table 2 ) did not substantially differ from those presented for the total population in Tables 2 to 4 .

Table 4

Percentage difference in RV performance per standard deviation difference in common cardiac risk factors

Covariate	Gender	n	Covariate SD	Percentage difference (95% confidence interval) in RV function per standard deviation difference in covariates by different echocardiographic indices
				Systolic annular velocity		Early diastolic annular velocity
Age (y) ∗	Women	663	13.6	−3.4 (−4.6 to −2.2)	P < .001	−10.3 (−12.0 to −8.7)	P < .001
	Men	603	13.7	−1.7 (−3.1 to −0.2)	P = .02	−10.6 (−12.7 to −8.5)	P < .001
BMI (kg/m 2 )	Women	647	4.2	−0.3 (−1.5 to 0.9)	P = .58	0.0 (−1.4 to 1.5)	P = .98
	Men	594	3.5	−3.1 (−4.8 to −1.4)	P < .001	−2.1 (−4.2 to 0.0)	P = .05
Systolic blood pressure (mm Hg)	Women	610	17.4	1.2 (−0.2 to 2.6)	P = .09	−0.8 (−2.7 to 1.2)	P = .43
	Men	565	13.9	2.8 (1.2 to 4.4)	P < .001	0.6 (−1.7 to 2.9)	P = .61
Diastolic blood pressure (mm Hg)	Women	610	10.4	−0.6 (−1.8 to 0.7)	P = .37	−2.5 (−4.2 to −0.7)	P = .006
	Men	565	10.2	−2.3 (−3.8 to −0.7)	P = .004	−5.3 (−7.5 to −3.1)	P < .001
Non-HDL cholesterol (mmol/L) †	Women	640	1.08	−1.5 (−2.9 to −0.1)	P = .04	−2.4 (−4.4 to −0.4)	P = .02
	Men	591	0.95	−1.3 (−2.9 to 0.2)	P = .09	−1.5 (−3.8 to 0.7)	P = .18
HDL cholesterol (mmol/L) †	Women	640	0.34	1.3 (0.1 to 2.6)	P = .04	2.0 (0.2 to 3.8)	P = .03
	Men	591	0.30	1.0 (−0.6 to 2,5)	P = .22	2.1 (−0.1 to 4.4)	P = .06
Smoking ‡	Women	339/280	Ever vs. never	−2,8 (−5,2 to −0,3)	P = .03	−5.7 (−9.1 to −2.3)	P < .001
	Men	278/283	Ever vs. never	−3.8 (−6.7 to −0.8)	P = .01	−4.4 (−8.7 to −0.2)	P = .04
eGFR (mL/min/1.73 m 2 )	Women	495	11.2	−1.0 (−2.5 to 0.6)	P = .22	−0.6 (−2.9 to 1.6)	P = .57
	Men	477	11.5	0.0 (−1.8 to 1.9)	P = .97	0.7 (−2.1 to 3.5)	P = .62
Glucose (mmol/L) ∗	Women	640	0.81	0.5 (−0.8 to 1.7)	P = .44	−2.0 (−3.8 to −0.2)	P = .03
	Men	591	1.39	−0.1 (−1.9 to 1.7)	P = .93	−2.1 (−4.7 to 0.6)	P = .12

Only gold members can continue reading. Log In or Register to continue

Share this:

• Click to share on Twitter (Opens in new window)
• Click to share on Facebook (Opens in new window)

Related

Related posts:

1. Co-Publication—What You Always Wanted to Know but Have Been Afraid to Ask!
2. Pitfalls in Echocardiographic Diagnosis of Patent Foramen Ovale by Transesophageal Echocardiography
3. “The Patient and the Image: The Role of Cardiovascular Ultrasound in Clinical Decision Making”
4. Continuing Education and Meeting Calendar

Stay updated, free articles. Join our Telegram channel

Join