Methods

The Framingham Heart Study Offspring cohort was initiated in 1971 and consists of children of the original Framingham cohort and their spouses. Offspring undergo comprehensive (“cycle”) examinations every 3 to 5 years. Of Offspring without history or signs of clinical CVD, a subsample without contraindications to CMR was recruited using a random sampling strategy based on equal strata of decade age, gender, and quintile of Framingham Coronary Risk Score, with double sampling of the top quintile. The study was approved by the institutional review boards of the Beth Israel Deaconess Medical Center and Boston University School of Medicine. All participants provided written informed consent.

Clinical covariates were obtained at offspring examination cycle 6. Height and weight were measured with participants in light clothing. Systolic and diastolic blood pressures were determined as the average of 2 readings by a physician. Fasting serum lipid levels, including total and high-density lipoprotein cholesterol, were obtained. Diabetes was defined as being on any antihyperglycemic medication or having a fasting serum glucose ≥126 mg/dl in any cycle examination. Any pharmacologic treatment for hypertension or dyslipidemia was recorded. Smoking was defined as having smoked regularly in the year before the cycle visit. Incident hard CVD events were defined as the first occurrence during follow-up after CMR of myocardial infarction, stroke, first hospital admission for heart failure, or cardiovascular death. Events were adjudicated by a 3-physician panel using previously described criteria.

CMR imaging used a segmented k-space gradient echo cine sequence on a 1.5-T scanner (Gyroscan ACS/NT, Philips Medical Systems, Best, Netherlands) as previously described. Contiguous 10-mm thick slices in the LV short-axis orientation were acquired, 1 per breath-hold. Temporal resolution was 39 ms with an in-plane spatial resolution of 1.25 × 2.0 mm ² .

Image data were analyzed using a commercially available workstation (EasyScil, Philips Medical Systems). LV endocardial borders were manually traced at end-diastole and end-systole. LV epicardial borders were also traced at end-diastole. For consistency in analysis, LV trabeculations and papillary muscles were considered LV cavitary volume. Stroke volume was the difference between LV end-diastolic and end-systolic volumes. LVEF was stroke volume divided by end-diastolic volume. MCF was calculated as LV stroke volume divided by LV myocardial volume. LV mass was myocardial volume multiplied by the mean density of myocardium (1.05 g/ml) and indexed to height, an allometric power of height (height ^2.7 ), and body surface area. The ratio of LV mass to LV end-diastolic volume was used as a measurement of LV concentricity.

Baseline participant characteristics are presented by gender, with continuous variables summarized as mean ± 1 SD. For each participant the gender-specific Framingham Coronary Risk Score was calculated using age, systolic blood pressure and blood pressure treatment status, diabetes, total and high-density lipoprotein cholesterol, treatment status for dyslipidemia, and smoking status. Between-gender differences were assessed by a 2-sample t test for continuous variables and chi-square test for categorical variables. Pearson correlation coefficients were used to assess linear relations between continuous measurements. After confirming the assumption of proportionality, a Cox proportional hazards model adjusting for gender and Framingham Coronary Risk Score was used to estimate the hazard ratio for hard CVD events in each quartile of MCF versus the referent (highest) quartile of MCF. Similar analysis was performed for LVEF. For LV mass the lowest mass quartile was used as the referent. Because MCF is explicitly related to LV mass (as myocardial volume) and implicitly related to LV concentricity, we also determined hazard ratios for MCF after adjustment for LV mass and for LV mass/LV end-diastolic volume. Kaplan–Meier survival plots and log-rank test were used to compare event-free survival between the lowest quartile of MCF and the combination of higher MCF quartiles. Similar analyses were performed for LVEF. All statistical analyses were performed using SAS 8.1 (SAS Institute, Cary, North Carolina).

Results

Of the 318 participants who underwent CMR, images were unevaluable in 14 (4.4%), principally because of difficulty with breath-holding, leaving 304 participants for analysis. Baseline characteristics for each gender are presented in Table 1 . Men were taller, heavier, and had greater body mass index than women. Total and high-density lipoprotein cholesterol were higher in women than in men.

LVEF was ≥0.55 in 301 participants (99%); of the remaining 3 participants, 1 woman had an LVEF of 0.46 and 2 men had LVEFs of 0.52 and 0.54. Overall, LVEF was greater in women than in men (0.72 ± 0.07 vs 0.69 ± 0.09, p = 0.002). MCF was also greater in women than in men (0.58 ± 0.13 vs 0.52 ± 0.11, p <0.01). Men had greater LV mass than women (159 ± 28 vs 110 ± 22 g, p <0.0001) and this difference remained statistically significant after indexation to height (91 ± 16 vs 68 ± 14 g/m, p <0.0001), height ^2.7 (35 ± 6 vs 31 ± 7 g/m ^2.7 , p <0.0001), and body surface area (77 ± 13 vs 62 ± 10 g/m ² , p <0.0001).

Over a median 5.2-year follow-up (interquartile range 4.6 to 5.4), an incident hard CVD event occurred in 31 participant (10%, 17 men) including 5 instances of myocardial infarction and 7 of unstable angina, 13 cerebrovascular events, and 6 cases of heart failure. There were also 4 cardiovascular deaths (all were after 1 of the preceding incident events and, hence, are not included among incident hard CVD events). Overall, participants who developed a hard CVD event during follow-up had significantly lower MCF than those free of hard CVD events ( Table 2 ). LVEF did not differ between those with and without hard CVD events. There was no statistically significant linear correlation between EF and MCF in participants with hard CVD events (r = 0.34, p = 0.07). In gender-specific analyses, participants with hard CVD events had lower MCF than participants without hard CVD events, but this difference was significant only in women ( Table 2 ). LVEF did not differ between participants with and without hard CVD events for men in women. LV mass was consistently greater in participants with hard CVD events than in participants without hard CVD events for men and women, but the significance of these differences varied with method of indexation. Considering the components of MCF, stroke volume, with and without indexation to body surface area, did not differ between participants with and without hard CVD events for men or women (women, p = 0.14; men, p = 0.91), suggesting that decreased MCF in participants with hard CVD events generally was driven by greater LV mass. Framingham Coronary Risk Score was greater in participants with hard CVD events in men and women, as expected.

As presented in Table 3 , the lowest-quartile MCF had increased hazards of hard CVD events even after adjustment for gender and Framingham Coronary Risk Score (hazard ratio 7.11, p = 0.010) compared to the referent (highest) quartile of MCF. LVEF and LV mass were not associated with incident hard CVD events. The highest-quartile indexed LV mass had >4 times the hazard for hard CVD events (hazard ratio 4.34, p = 0.023), which remained unchanged with indexation to height, height ^2.7 , and body surface area. After further adjustment for LV mass, the hazard ratio of lowest-quartile MCF remained significant (hazard ratio 6.09, p = 0.020). Hazard ratios for MCF were no longer significant after adjustment for concentricity (LV mass/LV end-diastolic volume). Hazard ratios for low MCF and LVEF and high LV mass, with and without indexation, are depicted in Figure 1 . Kaplan–Meir survival plots ( Figure 2 ) showed lower event-free survival for lowest-quartile MCF but not for EF. Results did not change with exclusion of the 3 participants with LVEF <0.55.

Table 3

Hazard ratios by quartile for hard cardiovascular events

	HR	95% CI	p Value
Left ventricular mass
Quartile 2	0.84	0.23–3.16	0.80
Quartile 3	1.62	0.53–4.99	0.40
Quartile 4	2.29	0.80–6.51	0.12
Left ventricular mass/height
Quartile 2	1.48	0.33–6.66	0.61
Quartile 3	3.13	0.84–11.65	0.09
Quartile 4	4.34	1.23–15.35	0.023
Left ventricular mass/height 2.7
Quartile 2	1.13	0.23–5.62	0.88
Quartile 3	3.06	0.83–11.37	0.09
Quartile 4	4.68	1.34–16.32	0.016
Left ventricular mass/body surface area
Quartile 2	1.20	0.30–4.81	0.80
Quartile 2	2.74	0.86–8.80	0.09
Quartile 2	3.50	1.12–10.95	0.031
Left ventricular ejection fraction
Quartile 1	0.86	0.35–2.12	0.75
Quartile 3	0.53	0.20–1.41	0.20
Quartile 3	0.29	0.08–1.01	0.051
Myocardial contraction fraction
Quartile 1	7.11	1.60–31.59	0.010
Quartile 2	4.04	0.88–18.58	0.07
Quartile 3	2.11	0.39–11.60	0.39
Myocardial contraction fraction, mass adjusted
Quartile 1	6.09	1.33–27.89	0.020
Quartile 2	3.75	0.81–17.38	0.09
Quartile 3	1.94	0.35–10.68	0.45
Myocardial contraction fraction, concentricity adjusted
Quartile 1	4.76	0.73–30.88	0.10
Quartile 2	3.36	0.68–16.76	0.14
Quartile 3	1.89	0.33–10.64	0.47

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. Impact of Statin Therapy on Late Target Lesion Revascularization After Sirolimus -Eluting Stent Implantation (from the CREDO-Kyoto Registry Cohort-2)
2. Patterns and Loss of Sexual Activity in the Year Following Hospitalization for Acute Myocardial Infarction (a United States National Multisite Observational Study)
3. How the 2008 Stock Market Crash and Seasons Affect Total and Cardiac Deaths in Los Angeles County
4. CHADS 2 and CHA 2 DS 2 -VASc Risk Factors to Predict First Cardiovascular Hospitalization Among Atrial Fibrillation/Atrial Flutter Patients

Stay updated, free articles. Join our Telegram channel

Join