Is cardiorespiratory fitness a determinant of cardiomyopathy in the setting of type 2 diabetes?

Abstract

Objective:

The aim of this study was to determine whether high fitness attenuates the defects in left ventricular (LV) structure, function and triglyceride (TG) content in patients with type 2 diabetes mellitus (T2DM).

Materials and methods:

Patients (n = 74) with T2DM and ≥1 additional cardiac risk factor were recruited to participate in this cross-sectional study. Outcome measures of interest were LV structure and function by magnetic resonance imaging (MRI) and myocardial TG content by ¹H-magnetic resonance spectroscopy (MRS). The primary exposure variable was cardiorespiratory fitness defined by peak oxygen consumption scaled to fat-free mass (FFM; VO_2peak-FFM).

Results:

Mean age was 53.5 years; 42.9% were women and mean glycosylated haemoglobin (HbA_1c) was 8.0% with the mean duration of T2DM 8.2 years. VO_2peak-FFM was crudely associated with both LV end systolic (r = 0.35, p = 0.002) and diastolic volumes (r = 0.32, p = 0.004), but not with ejection fraction (r = −0.15, p = 0.206), myocardial TG (r = −0.04, p = 0.734) or early diastolic peak filling rate (PFR; r = −0.01, p = 0.887). In multiple linear regression analyses, among measures of LV structure/function, VO_2peak-FFM was independently associated only with LV end-diastolic volume (EDV) (β = 1.037, p = 0.038).

Conclusion:

In individuals with T2DM at increased cardiovascular (CV) risk, cardiorespiratory fitness is not associated with LV morphology, function or myocardial TG content.

Keywords

Obesity ectopic fat exercise diabetes mellitus

Introduction

Type 2 diabetes mellitus (T2DM) is an established risk factor for atherosclerotic cardiovascular (CV) disease and congestive heart failure.¹ The increased CV morbidity and mortality associated with T2DM is generally attributed to conventional risk factors including dyslipidemia, hypertension, dysglycaemia, obesity and inflammation.² More recently, our group and others revealed that T2DM is associated with excessive deposition of triglyceride (TG) in the myocardium or cardiac steatosis.^3,4 In turn, steatosis is associated with left ventricular (LV) hypertrophy,⁵ impaired diastolic filling dynamics,⁶ as well as apoptosis contributing to LV systolic dysfunction.⁷ The factors that contribute to ectopic deposition of lipids within the myocardium and the associated changes in LV structure and function remain poorly understood.

Low cardiorespiratory fitness is a robust determinant of CV disease morbidity and mortality,^8,9 independent of chronic disease status.¹⁰ Among individuals with T2DM, low cardiorespiratory fitness is associated with an increased inflammatory state,¹¹ arterial stiffening¹² and reduced life expectancy.¹³ Low cardiorespiratory fitness is associated with an increased risk of ectopic lipid deposition¹⁴ and lower LV diastolic function;¹⁵ it is possible that fitness may contribute to features of diabetic cardiomyopathy, including abnormalities of cardiac structure and function and myocardial steatosis.

In light of these observations, the main objective of this study was to determine whether cardiorespiratory fitness is associated with abnormalities of cardiac structure, function and myocardial steatosis in individuals with T2DM. Our hypothesis was that individuals with T2DM and high cardiorespiratory fitness would have lower levels of myocardial TG and enhanced LV diastolic relaxation compared with individuals with T2DM and low cardiorespiratory fitness.

Materials and methods

Study design and participants

We relied on cross-sectional data collected at baseline in 104 individuals with T2DM enrolled in a therapeutic randomized controlled trial who were selected to undergo baseline cardiac imaging with magnetic resonance imaging (MRI) and ¹H-magnetic resonance spectroscopy (MRS).¹⁶ Participants were recruited from outpatient clinics, research study rosters and public advertisement. Inclusion criteria were age >18 years, medically treated T2DM (previous diagnosis of T2DM or current use of hypoglycaemic medical therapy) and the presence of ≥1 other CV disease risk factors, including smoking, hypertension (≥140/90 mmHg), dyslipidemia (treated, physician diagnosed or self-reported), albuminuria, family history of coronary artery disease or high-sensitivity C-reactive protein >3 mg/L. Exclusion criteria included atherosclerotic disease (i.e. coronary artery disease determined by clinical history and confirmed with an exercise stress test to exhaustion with a 12-lead electrocardiogram (ECG)), stroke, transient ischaemic attack, carotid or peripheral atherosclerosis, determined from patient history, the use of thiazolidinedione treatment within the prior 6 months, prior thiazolidinedione intolerance, prior heart failure, aspartate transaminase or alanine transaminase >3 times the upper limit of normal or inability to perform treadmill exercise. The University of Texas Southwestern Medical Center Institutional Review Board approved the study. All participants provided written informed consent.

Outcome measures

The primary outcome measure for the present analyses was myocardial TG content, measured non-invasively with ¹H-MRS and expressed as ratio fat/water (%), as previously described.^4,16 Spectroscopy data were collected at end expiration and end systole, triggered by simultaneous respiratory and ECG gating. Spectra were collected using a whole body coil and a 17-cm surface coil for signal reception. Point RESolved Spectroscopy (PRESS) was used for spatial localization and signal acquisition. The interpulse delay T_R was 4 s and the spin echo time (T_e) was 25 ms. Data points (1024) were collected over a 1000-Hz spectral width and averaged >64 spectra. Areas under the curve for resonances of water and methylene groups in fatty acid chains were evaluated with line-fit procedure and commercial software (NUTS; Acorn NMR Inc., Freemont, CA, USA). Signal decay due to a spin–spin relaxation was calculated from a monoexponential relaxation equation¹⁷ using a mean T₂ relaxation time of 40 ms for water and 78 ms for myocardial TG.

The secondary outcome measures included LV filling dynamics, ejection fraction, LV volumes and LV mass quantified using standard cardiac MRI techniques as previously described.¹⁶ Briefly dynamic images were used to quantify LV diastolic and systolic volume using short-axis slices. LV mass was computed as the product of LV end-diastolic volume (EDV) and myocardial density (1.05 g/mL), divided by body surface area. All MRI and spectroscopy measures were obtained using a whole body 1.5 T Gyroscan Intera clinical scanner (Philips Medical Systems, Cleveland, OH, USA) equipped with a spectroscopy package and were analysed by the same technician.^18,19 Systolic function was estimated by ejection fraction and as we did not perform ultrasound in this study, diastolic function was estimated by early LV diastolic peak filling rate (PFR), from MRI cine images, rather than transmitral flow velocities or tissue Doppler-based estimates of tissue relaxation rates.

Primary exposure variable

The primary exposure variable was cardiorespiratory fitness, measured as peak oxygen uptake (VO_2peak-FFM) determined with a graded treadmill exercise test to exhaustion.^20,21 To determine treadmill speed for each participant, a warm-up phase prior to the cardiopulmonary exercise testing was performed. Each participant walked on the treadmill at an initial workload of 1.7 mile/h and 0% grade. Treadmill speed was gradually increased until the participant attained a comfortable stride. Thereafter, further adjustments in speed were made until 60%–70% of the participant’s age-predicted maximal heart rate was achieved and/or until the participant reported a rate of perceived exertion of 11–13 using the Borg scale.²² The participant walked for 4 min at the workload that elicited 60%–70% of the participant’s age-predicted maximal heart rate to become familiar with the pace/effort. Following 5–10 min of rest, the graded treadmill test was performed at the workload needed to elicit 60%–70% of the participant’s age-predicted maximal heart rate while the grade was increased by 2% every 2 min until volitional exhaustion. Expired gases and volumes were quantified breath-by-breath using the MedGraphics^TM Cardiopulmonary Exercise System (CPX/D; Medical Graphics Corporation, St. Paul, MN, USA), at rest and throughout the treadmill exercise test. Maximal values for each of the gas exchange variables were calculated as the average of the two highest consecutive 30-s values for that variable. VO_2peak was scaled to fat-free mass (FFM; VO_2peak-FFM) to control for the potential confounding effect of adiposity on the outcome measures. Additional measures collected at the end of the graded maximal exercise test included the respiratory exchange ratio (RER) = [(rate of CO₂ expired)/(rate of O₂ consumed)], heart rate and the ratio of minute ventilation to oxygen to determine whether the effort met criteria for a maximal exercise test.

Confounding variables

Anthropometric measures and body composition

Body weight was measured to the nearest 0.1 kg on a calibrated scale. Waist circumference was measured to the nearest 0.1 cm with a flexible tape measure parallel to floor at 1 cm above the iliac crest at end expiration. Height was obtained with a standard stadiometer. Body mass index (BMI; kg/m²) was calculated using the standard formula: weight (kg)/height (m²). Skin-fold thickness was measured with calipers prior to each exercise test from four locations in duplicate and averaged, and percent body fat was calculated by the method of Durnin and Womersley.²³

Metabolic profile

The following cardiometabolic risk factors were measured after a 10-h overnight fast using standard clinical lab methods: serum glucose, TGs, high-density lipoprotein (HDL)-cholesterol and urine analysis for albumin to creatinine ratio. Plasma and serum samples were multiply aliquoted and frozen at −70 °C for subsequent analyses, including but not limited to assessment of B-type natriuretic peptide (BNP) using a commercially available assay (Biosite, Inc., San Diego, CA, USA), while glycosylated haemoglobin (HbA_1c) was measured on fresh plasma samples. Resting systolic blood pressure (BP) and diastolic BP were measured in triplicate in a sitting position using a Welch Allyn series 52,000 oscillometric automated BP machine, as recommended by the National Committee on Preventive, Detection, Evaluation and Treatment of High BP.²⁴ The mean of the three different measures was used in the analysis.

Race/ethnicity and smoking status

Race/ethnicity and smoking status (non-smoker, former smoker or current smoker) were self-reported by the participant during the medical interview.

Statistical analysis

Data are presented as mean ± standard deviation for continuous variables and as n (%) for categorical variables. Bivariate Spearman’s correlations were performed to investigate the associations between the exposure variable and primary and secondary outcome measures. Spearman’s correlations were used because the primary and the secondary outcome measures were not normally distributed. Log transformations were performed to normalize the outcome measures for all other group-wise regression analyses. To investigate the independent association between cardiorespiratory fitness and myocardial TG content, LV morphology and function, two strategies were used. First, multiple linear regression models were used to investigate the continuous association between cardiorespiratory fitness and myocardial TG content, LV structure and LV function independent of confounders. Second, the cohort was divided into data-derived tertiles based on VO_2peak-FFM and were used in multiple-regression models using the lowest tertile as the reference group. Statistical analyses were adjusted for age, sex, race/ethnicity, smoking status, systolic BP, dyslipidemia, BMI, diabetes duration and insulin use. Multiple linear regression models forcing all the confounding variables were used based on reported influence of each on the LV structure/function metrics assessed. All analyses were performed with SAS software (SAS Version 9.1.3; Cary, NC, USA) using two-tailed tests at a significance level of 0.05, with no corrections made for multiple comparisons.

Results

General characteristics of the study population

Of the 104 study participants volunteering to undergo cardiac MRI/MRS imaging, 74 participants completed all phases of the imaging protocol with valid measures of LV filling and myocardial TG content. Baseline characteristics for these 74 participants are provided in Table 1 stratified by VO_2peak-FFM tertile. On average, participants were 53.5 ± 9.3 years of age, mostly men (58%), obese (average BMI = 34 ± 6 kg/m²), with modest glycaemic control (HbA_1c = 8.0% ± 2%) and with T2DM duration of 8 ± 6 years. No trends were observed in crude analyses between general characteristics across tertiles of VO_2peak-FFM, with the exception of BMI (p = 0.03). Approximately 35% of participants (26/74) did not reach an RER of 1.1.

Table 1.

General characteristics overall and stratified by cardiorespiratory fitness tertiles.

	Overall	Tertile I VO_2peak 17.85 to 25.93 (mL/FFM/min)	Tertile II VO_2peak 25.98 to 30.14 (mL/FFM/min)	Tertile III VO_2peak 30.44 to 43.97 (mL/FFM/min)	p value
n	74	24	25	25
Age (years)	53.5 ± 9.3	55.5 ± 10.2	53.9 ± 8.0	51.3 ± 9.5	0.15
Men (%)	43 (58.1)	12 (50.0)	12 (48.0)	19 (76.0)	0.06
Race/ethnicity
Caucasian, n (%)	24 (32.4)	9 (37.5)	6 (24.0)	9 (36.0)	0.92
African-American, n (%)	34 (45.9)	13 (54.1)	12 (48.0)	9 (36.0)	0.20
Hispanic, n (%)	15 (20.2)	2 (8.3)	7 (28.0)	6 (24.0)	0.17
Other, n (%)	1 (1.3)	0 (0)	0 (0)	1 (4.0)	0.22
Smoking
Current smoker, n (%)	10 (13.5)	4 (16.6)	1 (4.0)	5 (20.0)	0.71
Former smoker, n (%)	11 (14.8)	4 (16.6)	3 (12.0)	4 (16.0)	0.95
Non-smoker, n (%)	64 (86.4)	16 (66.6)	21 (84.0)	16 (64.0)	0.82
Anthropometrics
Weight (kg)	99.0 ± 19.6	102.8 ± 19.1	99.4 ± 21.2	95.6 ± 18.5	0.13
BMI (kg/m²)	34.0 ± 6.1	35.8 ± 6.8	34.2 ± 6.2	32.1 ± 4.9	0.03
Waist circumference (cm)	106.9 ± 15.1	111.5 ± 17.6	104.9 ± 13.7	105.1 ± 13.3	0.15
Body fat (%)	32.0 ± 7.1	31.4 ± 7.7	33.6 ± 5.7	30.9 ± 7.9	0.54
FFM (kg)	66.6 ± 11.9	69.1 ± 10.9	65.7 ± 14.4	65.2 ± 10.1	0.26
Metabolic profile
Triglycerides (mg/dL)	171.1 ± 126.7	217.5 ± 162.9	142.5 ± 98.6	155.1 ± 102.0	0.14
HDL-cholesterol (mg/dL)	45.7 ± 10.7	46.3 ± 12.8	45.6 ± 11.2	45.3 ± 8.1	0.75
Systolic BP (mmHg)	148.3 ± 24.0	146.8 ± 21.6	155.3 ± 30.9	143.6 ± 18.1	0.58
Diastolic BP (mmHg)	83.1 ± 12.7	82.2 ± 12.5	86.1 ± 15.0	81.2 ± 10.7	0.55
Dyslipidemia, n (%)	46 (62.1)	17 (70.8)	13 (52.0)	16 (64.0)	0.63
Hypertension, n (%)	49 (66.2)	17 (70.8)	19 (76.0)	13 (52.0)	0.15
Diabetes data
HbA_1c (%)	8.0 ± 2.0	8.5 ± 2.2	7.4 ± 1.6	8.1 ± 2.0	0.52
Fasting glucose (mg/dL)	174.8 ± 65.0	185.2 ± 70.0	164.5 ± 68.1	175.1 ± 57.6	0.63
Diabetes duration (years)	8.2 ± 5.8	9.0 ± 6.3	7.3 ± 5.1	8.1 ± 6.1	0.63
Insulin treated (%)	25 (33.7)	12 (50.0)	7 (28.0)	6 (24.0)	0.05

BMI: body mass index; HDL: high-density lipoprotein; BP: blood pressure; HbA_1c: glycosylated haemoglobin; FFM: fat-free mass.

Continuous variables are presented as mean ± standard deviation, while categorical variables are presented as n (%).

LV structure, function and TG content across tertiles of fitness are presented in Table 2. Overall, average LV mass index (80.9 ± 20.7 g/m³), ejection fraction (65.2% ± 9.6%) and LV PFRs (68.7 ± 38.5 mL/s) were within normal ranges, and myocardial TG content was similar to previously reported values for patients with T2DM (0.95% ± 0.5%). Significant crude trends across fitness tertiles were observed for LVEDV (p_trend = 0.002), LV end-systolic volume (ESV; p_trend = 0.01) and ratio of LVPFR:LVEDV (p_trend = 0.01), while no trends were observed for myocardial TG content (p = 0.97) or ejection fraction (p_trend = 0.33). No differences in maximal RER were observed across tertiles of fitness (p_trend = 0.61).

Table 2.

Characteristics of LV structure, function and myocardial triglyceride content overall and stratified by cardiorespiratory fitness tertiles.

	Overall (N = 74)	Tertile I VO_2peak 17.85 to 25.93 (mL/FFM/min), N = 24	Tertile II VO_2peak 25.98 to 30.14 (mL/FFM/min), N = 25	Tertile III VO_2peak 30.44 to 43.97 (mL/FFM/min), N = 25	p _trend
Exercise test variables
Absolute VO_2peak (L/min)	1.9 ± 0.4	1.5 ± 0.3	1.8 ± 0.3	2.3 ± 0.4	<0.01
Relative VO_2peak (mL/kg/min)	19.7 ± 4.6	15.6 ± 2.3	18.8 ± 1.7	24.4 ± 3.8	<0.01
Relative VO_2peak (mL/FFM/min)	28.9 ± 6.1	22.7 ± 2.7	28.1 ± 1.2	35.7 ± 4.1	<0.01
RER (VCO₂/VO₂)	1.14 ± 0.09	1.13 ± 0.10	1.15 ± 0.07	1.14 ± 0.09	0.61
Cardiac structure and function
LVEDV (mL)	120.5 ± 25.3	110.3 ± 23.3	119.1 ± 23.3	131.5 ± 25.6	<0.01
LVESV (mL)	42.5 ± 17.4	39.0 ± 21.7	40.8 ± 13.2	47.5 ± 15.9	0.01
LV mass (gm²)	80.9 ± 20.7	78.8 ± 22.0	83.9 ± 23.4	79.7 ± 16.8	0.41
Ejection fraction (%)	65.2 ± 9.6	65.9 ± 11.0	65.3 ± 10.4	64.3 ± 7.5	0.33
PFR (mL/s)	68.7 ± 38.5	66.2 ± 29.3	79.6 ± 43.8	60.1 ± 39.5	0.48
PFR/EDV	1.4 ± 0.7	1.5 ± 0.6	1.6 ± 0.7	1.1 ± 0.8	0.01
Myocardial triglycerides (%)	0.95 ± 0.5	0.94 ± 0.5	0.97 ± 0.5	0.93 ± 0.6	0.97

FFM: fat-free mass; LV: left ventricular; PFR: peak filling rate; PFR/EDV: peak filling rate/end-diastolic volume.

Bivariate associations between continuous VO_2peak-FFM and various measures of LV structure and function are presented in Table 3. LVEDV (r = 0.35, p = 0.002), LVESV (r = 0.32, p = 0.004) and the ratio of LVPFR:LVEDV (r = −0.24, p = 0.037) were significantly associated with VO_2peak-FFM, while ejection fraction (r = −0.14, p = 0.206), absolute LV PFR (r = −0.01, p = 0.887) and myocardial TG (r = −0.04, p = 0.734) were not associated with VO_2peak-FFM. In addition, RER was positively correlated with myocardial TG (r = 035, p = 0.006), while no such association was observed with any other outcome variables.

Table 3.

Association between body composition, metabolic variables, myocardial steatosis, left ventricular structure and function.

	Myocardial steatosis	LVEDV	LVESV	LVMass	Ejection fraction	PFR	PFR/EDV
Age (years)	−0.01	−0.27*	−0.36*	−0.02	0.30*	−0.14	0.10
BMI (kg/m²)	−0.03	0.15	0.05	−0.06	−0.01	0.20	0.17
Body fat (%)	−0.01	−0.01	−0.19	−0.39*	0.19	0.04	0.19
Total FFM (kg)	0.13	0.34*	0.25*	0.37*	−0.11	0.29*	0.09
RER (VCO₂/VO₂)	0.35*	0.18	0.14	0.20	−0.08	0.06	−0.04
VO_2peak (mL/FFM/min)	−0.04	0.35*	0.32*	0.06	−0.14	−0.01	−0.24*
Triglycerides (mg/dL)	−0.08	−0.08	−0.01	0.10	0.06	0.03	0.09
HDL-cholesterol (mg/dL)	−0.06	−0.18	−0.14	−0.30*	0.05	−0.08	0.01
Systolic blood pressure (mmHg)	0.16	0.08	−0.01	0.38*	0.01	−0.04	0.12
Diastolic blood pressure (mmHg)	0.24	0.20	0.22	0.43*	−0.17	0.01	0.11
Fasting glucose (mg/dL)	−0.14	−0.02	0.07	0.05	−0.13	−0.08	0.14
HbA_1c (%)	−0.20	0.01	0.10	0.08	−0.15	−0.04	−0.09

LVEDV: left ventricular end-diastolic volume; LVESV: left ventricular end-systolic volume; LVMass: left ventricular mass; BMI: body mass index; FFM: fat-free mass; HDL: high-density lipoprotein; HbA_1c: haemoglobin A_1c; PFR/EDV: peak filling rate/end-diastolic volume.

Data are presented as Rho.

p < 0.05.

Results from a series of multivariable regression models are presented in Table 4. After adjusting for all selected confounding variables, VO_2peak-FFM was positively and independently associated only with LVEDV (β = 1.037, p = 0.038). LVESV (β = 0.306, p = 0.410), LV mass (β = −0.648, p = 0.080), ejection fraction (β = −0.075, p = 0.723), LV PFR (β = 0.477, p = 0.562), the ratio of LV PFR (β = −0.011, p = 0.482) and myocardial TG (β = −0.027, p = 0.128) were not independently associated with VO_2peak-FFM.

Table 4.

Independent association between VO_2peak-FFM and measures of cardiac structure and function.

	β	p value
Myocardial steatosis
Model 1	−0.020	0.189
Model 2	−0.020	0.215
Model 3	−0.026	0.145
Model 4	−0.027	0.128
Left ventricular EDV
Model 1	0.804	0.081
Model 2	1.269	0.006
Model 3	1.123	0.023
Model 4	1.037	0.038
Left ventricular end-systolic volume
Model 1	0.142	0.657
Model 2	0.182	0.586
Model 3	0.274	0.450
Model 4	0.306	0.410
Left ventricular mass
Model 1	−0.434	0.258
Model 2	−0.391	0.300
Model 3	−0.631	0.081
Model 4	−0.648	0.080
Ejection fraction
Model 1	0.052	0.792
Model 2	0.095	0.641
Model 3	−0.023	0.912
Model 4	−0.075	0.723
Peak filling rate
Model 1	−0.326	0.693
Model 2	0.260	0.758
Model 3	0.461	0.566
Model 4	0.477	0.562
PFR/EDV
Model 1	−0.016	0.305
Model 2	−0.007	0.632
Model 3	−0.010	0.514
Model 4	−0.011	0.482

PFR/EDV: peak filling rate/end-diastolic volume; BP: blood pressure; BMI: body mass index; FFM: fat-free mass.

Data are presented as unstandardized β and p values.

Model 1 adjusted for age, gender and race/ethnicity. Model 2 adjusted for age, gender, race/ethnicity, smoking, continuous BP, dyslipidemia and BMI. Model 3 adjusted for age, gender, race/ethnicity, smoking, continuous BP, dyslipidemia, BMI and diabetes duration. Model 4 adjusted for age, gender, race/ethnicity, smoking, continuous BP, dyslipidemia, BMI, diabetes duration and insulin used.

Multiple linear regression models comparing measures of LV structure, function and TG content across tertiles of fitness after adjusting for confounding are presented in Table 5. Compared with patients in the highest tertile of VO_2peak-FFM, LVEDV (β = −14.85, p = 0.032) was significantly lower in patients within the lowest tertile of VO_2peak-FFM, after adjusting for confounders. Tertiles of VO_2peak-FFM were not independently associated with any other measure of cardiac structure, function or myocardial TG content (p > 0.05).

Table 5.

Independent association between myocardial triglyceride content, hearth structure and function with tertiles of cardiorespiratory fitness.

Models	Myocardial steatosis		LVEDV		LVESV		LVMass		Ejection fraction		PFR		PFR/EDV
	β	p value	β	p value	β	p value	β	p value	β	p value	β	p value	β	p value
Model 1
Tertile 1	0.074	0.747	−13.06	0.046	−2.539	0.579	4.609	0.393	−0.621	0.825	11.53	0.314	0.368	0.104
Tertile 2	0.313	0.197	−5.15	0.423	−1.544	0.734	9.375	0.083	−0.795	0.776	24.24	0.035	0.480	0.034
Tertile 3	Reference		Reference		Reference		Reference		Reference		Reference		Reference
Model 2
Tertile 1	0.172	0.465	−17.78	0.005	−2.629	0.573	3.269	0.529	−1.332	0.641	6.03	0.605	0.277	0.222
Tertile 2	0.425	0.097	−8.70	0.190	−1.564	0.750	7.795	0.157	−1.501	0.618	15.39	0.213	0.351	0.144
Tertile 3	Reference		Reference		Reference		Reference		Reference		Reference		Reference
Model 3
Tertile 1	0.238	0.344	−15.84	0.021	−3.023	0.551	6.420	0.203	−0.412	0.889	−0.407	0.971	0.227	0.291
Tertile 2	0.497	0.065	−5.99	0.400	0.115	0.982	7.225	0.175	−1.699	0.586	1.628	0.890	0.086	0.703
Tertile 3	Reference		Reference		Reference		Reference		Reference		Reference		Reference
Model 4
Tertile 1	0.241	0.343	−14.85	0.032	−3.367	0.513	6.487	0.207	0.174	0.952	−0.461	0.967	0.241	0.272
Tertile 2	0.487	0.074	−5.95	0.402	0.101	0.984	7.227	0.179	−1.676	0.587	1.626	0.891	0.086	0.703
Tertile 3	Reference		Reference		Reference		Reference		Reference		Reference		Reference

LVEDV: left ventricular end-diastolic volume; LVESV: left ventricular end-systolic volume; LVMass: left ventricular mass; BMI: body mass index; PFR/EDV: peak filling rate/end-diastolic volume.

Data are presented as β and p values. Model 1 adjusted for age, gender and race/ethnicity. Model 2 adjusted for age, gender, race/ethnicity, smoking, dyslipidemia, continuous blood pressure and BMI. Model 3 adjusted for age, gender, race/ethnicity, smoking, dyslipidemia, continuous blood pressure, BMI and diabetes duration. Model 4 adjusted for age, gender, race/ethnicity, smoking, dyslipidemia, continuous blood pressure, BMI, diabetes duration and insulin used.

Conclusion

The results from this cross-sectional study reveal that among individuals with T2DM with prevalent or increased risk of atherosclerotic CV disease, cardiorespiratory fitness is not a determinant of myocardial structure, function or TG content. In contrast with other studies,^25,26 we did not find an association between LV diastolic performance and fitness determined by VO_2peak-FFM. Furthermore, we did not observe an association between fitness and myocardial TG content or LV mass.

Fitness, cardiac structure and function in T2DM

Cardiorespiratory fitness is an important modifiable prognostic indicator for CV disease-related and all-cause mortality.^8–10,13 The association between fitness and mortality is evident across several chronic diseases, including T2DM.^13,27 Individuals with high fitness levels for their age and sex have a significantly reduced risk of premature mortality from all causes, including T2DM.^28,29 The mechanisms to explain the protective effect of high fitness against CV disease, especially in patients with T2DM, remain poorly understood. There is some evidence that this protection may be mediated through enhanced cardiac performance and vascular function.^12,28,30 For example, chronic exercise training interventions that lead to 9%–13% improvements in peak oxygen uptake can improve both LV systolic and diastolic functions,^30–33 and in some cases normalize diastolic filling in individuals with well-controlled T2DM.³⁰ The results of this study do not support observations from previous observational studies or experimental trials,³⁴ as cardiorespiratory fitness was not associated with LV filling parameters or ejection fraction in this study. The discrepancy between this and previous studies may be attributable to the very narrow range of fitness observed in this study cohort, at the lower end of the continuum (16.60–22.40 mL/kg/min), which was significantly lower than fitness levels reported in other studies (33.9–43.8 mL/kg/min).^28,35 It is possible that fitness levels may be a determinant of cardiac structure, function and TG content at higher ranges of fitness, or perhaps fitness may have less influence on such measures among older patients with a longer duration of T2DM. Studies in larger samples of patients with broader ranges of fitness are needed to determine the relationship between fitness and alterations in cardiac structure and function in patients with long-standing diabetes.

Cardiac steatosis and T2DM

Studies by our group and others have demonstrated that myocardial steatosis is increased in states of impaired glucose tolerance^4,36 and T2DM.^3,20 The importance of this excessive myocardial accumulation of lipid has been a significant topic of debate.^37–39 Some clinical studies of younger patients with newly diagnosed T2DM suggest that higher TG content in the myocardium is associated with impaired LV diastolic filling.⁴⁰ However, this is not a universal finding, as one experimental trial reported a significant improvement in cardiorespiratory fitness and cardiac function without any detectable change in cardiac steatosis in patients with T2DM.⁴¹ To the best of our knowledge, only one other study has examined the association between fitness and myocardial TG content in humans.³⁶ Among older, obese, normoglycaemic women, cardiorespiratory fitness was weakly and negatively associated with myocardial TG content, independent of BMI, age, insulin sensitivity and serum TG (β = −0.27, p = 0.04).³⁶ This supports findings from experimental trials demonstrating that exercise training at doses sufficient to enhance fitness and cardiac function have little effect on myocardial TG content in patients with T2DM.⁴¹ The data presented here also support the concept that a higher level of cardiorespiratory fitness is not a protective factor against cardiac steatosis in patients with long-standing T2DM. Specifically, cardiac TG content was approximately two-fold higher in the patients studied here relative to previously published values in otherwise healthy, normal weight adults. This difference was also evidence in the myocardial TG content in patients within the highest tertile of fitness and published norms from previous studies by our group and others (0.4%–0.6% vs 0.93% ± 0.60%).^4,36,42 The results may again reflect the narrow range of fitness levels within this cohort or the heterogeneity of lipid deposition within the myocardium of these patients. Larger experimental exercise intervention trials restricted to patients with established myocardial steatosis are needed to determine the influence of increasing fitness on myocardial TG content.

Strengths and limitations

To the best of our knowledge, this is the first study evaluating the associations between cardiorespiratory fitness and myocardial structure, function and steatosis in patients with T2DM. The study is strengthened by the use of gold standard measures of cardiac structure, function and steatosis coupled with a direct measurement of cardiorespiratory fitness. Despite these strengths, the study suffers from several limitations that should be acknowledged. First, the small sample size limits the power to detect small but possibly clinically relevant differences in the cardiac measures assessed, challenging the external validity of the study findings. Second, the methods used to assess LV diastolic function and LV relaxation relative to preload are not as sensitive as more sophisticated methods (i.e. ultrasound-derived strain rates or cardiac MRI tagging methods), therefore we may not have been able to detect subtle abnormalities in LV diastolic performance associated with low fitness levels. Third, the cross-sectional nature of the study design limits any causal interpretations of the associations noted here. Despite these limitations, the data provide novel insight into the associations between fitness and cardiac structure, function and steatosis in an understudied segment of the population of patients with T2DM.

In conclusion, in adults with long-standing T2DM and prevalent CV disease or risk factors, cardiorespiratory fitness is not associated with LV structure, function or myocardial TG content. These data suggest that factors other than LV function and TG accumulation contribute to the cardioprotective effects of high aerobic fitness in individuals with T2DM.

Footnotes

Acknowledgements

We thank all the participants of the study as well as all staff member who collected the data.

Declaration of conflicting interests

No potential conflicts of interest relevant to this article were reported. The other authors have nothing to disclose.

Funding

We would like to thank the Manitoba Health Research Council, The Canadian Institute of Health Research and the Integrated and Mentored Pulmonary and Cardiovascular Training that supports the salary and the training of Dr Sénéchal. Dr McGavock is supported by a New Investigator Salary Award from the Canadian Institutes of Health Research. Dr McGuire has received research support from Daiichi Sankyo, Eli Lilly, Merck, Bristol-Myers Squibb, Boehringer Ingelheim, Genentech, F. Hoffmann-La Roche, Eisai, Omthera, Takeda, GlaxoSmithKline and Novo Nordisk and is a consultant for Janssen Pharmaceuticals, Boehringer Ingelheim, Regeneron, Genentech, Merck, Sanofi-Aventis and Takeda.

References

Laakso

. Cardiovascular disease in type 2 diabetes from population to man to mechanisms: the Kelly West Award Lecture 2008. Diabetes Care 2010; 33: 442–449.

Zimmerman

. Cardiovascular disease and risk factors in law enforcement personnel: a comprehensive review. Cardiol Rev 2012; 20: 159–166.

Rijzewijk

van der Meer

Smit

. Myocardial steatosis is an independent predictor of diastolic dysfunction in type 2 diabetes mellitus. J Am Coll Cardiol 2008; 52: 1793–1799.

McGavock

Lingvay

Zib

. Cardiac steatosis in diabetes mellitus: a 1H-magnetic resonance spectroscopy study. Circulation 2007; 116: 1170–1175.

Feuvray

Darmellah

. Diabetes-related metabolic perturbations in cardiac myocyte. Diabetes Metab 2008; 34(Suppl. 1): S3–S9.

Sharma

Adrogue

Golfman

. Intramyocardial lipid accumulation in the failing human heart resembles the lipotoxic rat heart. FASEB J 2004; 18: 1692–1700.

Van de Weijer

Schrauwen-Hinderling

Schrauwen

. Lipotoxicity in type 2 diabetic cardiomyopathy. Cardiovasc Res 2011; 92: 10–18.

Blair

Brodney

. Effects of physical inactivity and obesity on morbidity and mortality: current evidence and research issues. Med Sci Sports Exerc 1999; 31: S646–S662.

Blair

Kampert

Kohl

. Influences of cardiorespiratory fitness and other precursors on cardiovascular disease and all-cause mortality in men and women. JAMA 1996; 276: 205–210.

10.

Myers

Prakash

Froelicher

. Exercise capacity and mortality among men referred for exercise testing. N Engl J Med 2002; 346: 793–801.

11.

McGavock

Mandic

Vonder Muhll

. Low cardiorespiratory fitness is associated with elevated C-reactive protein levels in women with type 2 diabetes. Diabetes Care 2004; 27: 320–325.

12.

Tanaka

Dinenno

Monahan

. Aging, habitual exercise, and dynamic arterial compliance. Circulation 2000; 102: 1270–1275.

13.

Wei

Gibbons

Kampert

. Low cardiorespiratory fitness and physical inactivity as predictors of mortality in men with type 2 diabetes. Ann Intern Med 2000; 132: 605–611.

14.

Wittmeier

Wicklow

MacIntosh

. Hepatic steatosis and low cardiorespiratory fitness in youth with type 2 diabetes. Obesity 2012; 20: 1034–1040.

15.

Levine

Lane

Buckey

. Left ventricular pressure-volume and Frank-Starling relations in endurance athletes. Implications for orthostatic tolerance and exercise performance. Circulation 1991; 84: 1016–1023.

16.

McGuire

See

Abdullah

. The effect of rosiglitazone on integrated cardiovascular performance, cardiac structure, function and myocardial triglyceride: trial design and rationale. Diab Vasc Dis Res 2009; 6: 43–50.

17.

Reingold

McGavock

Kaka

. Determination of triglyceride in the human myocardium by magnetic resonance spectroscopy: reproducibility and sensitivity of the method. Am J Physiol Endocrinol Metab 2005; 289: E935–E939.

18.

Szczepaniak

Dobbins

Metzger

. Myocardial triglycerides and systolic function in humans: in vivo evaluation by localized proton spectroscopy and cardiac imaging. Magn Reson Med 2003; 49: 417–423.

19.

Felblinger

Jung

Slotboom

. Methods and reproducibility of cardiac/respiratory double-triggered (1)H-MR spectroscopy of the human heart. Magn Reson Med 1999; 42: 903–910.

20.

McGuire

Levine

Williamson

. A 30-year follow-up of the Dallas Bedrest and Training Study: I. Effect of age on the cardiovascular response to exercise. Circulation 2001; 104: 1350–1357.

21.

McGuire

Levine

Williamson

. A 30-year follow-up of the Dallas Bedrest and Training Study: II. Effect of age on cardiovascular adaptation to exercise training. Circulation 2001; 104: 1358–1366.

22.

Borg

. Perceived exertion: a note on ‘history’ and methods. Med Sci Sports 1973; 5: 90–93.

23.

Durnin

Womersley

. Body fat assessed from total body density and its estimation from skinfold thickness: measurements on 481 men and women aged from 16 to 72 years. Br J Nutr 1974; 32: 77–97.

24.

Chobanian

Bakris

Black

. Seventh report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure. Hypertension 2003; 42: 1206–1252.

25.

Poirier

Bogaty

Garneau

. Diastolic dysfunction in normotensive men with well-controlled type 2 diabetes: importance of maneuvers in echocardiographic screening for preclinical diabetic cardiomyopathy. Diabetes Care 2001; 24: 5–10.

26.

Arbab-Zadeh

Dijk

Prasad

. Effect of aging and physical activity on left ventricular compliance. Circulation 2004; 110: 1799–1805.

27.

Myers

Kaykha

George

. Fitness versus physical activity patterns in predicting mortality in men. Am J Med 2004; 117: 912–918.

28.

Wei

Kampert

Barlow

. Relationship between low cardiorespiratory fitness and mortality in normal-weight, overweight, and obese men. JAMA 1999; 282: 1547–1553.

29.

Lee

Artero

Sui

. Mortality trends in the general population: the importance of cardiorespiratory fitness. J Psychopharmacol 2010; 24: 27–35.

30.

Brassard

Legault

Garneau

. Normalization of diastolic dysfunction in type 2 diabetics after exercise training. Med Sci Sports Exerc 2007; 39: 1896–1901.

31.

Hill

Olson

. Cardiac plasticity. N Engl J Med 2008; 358: 1370–1380.

32.

Baggish

Wood

. Athlete’s heart and cardiovascular care of the athlete: scientific and clinical update. Circulation 2011; 123: 2723–2735.

33.

Bronstad

Tjonna

Rognmo

. Aerobic exercise training improves right- and left ventricular systolic function in patients with COPD. COPD 2013; 10: 300–306.

34.

McGavock

Eves

Mandic

. The role of exercise in the treatment of cardiovascular disease associated with type 2 diabetes mellitus. Sports Med 2004; 34: 27–48.

35.

Loimaala

Groundstroem

Rinne

. Exercise training does not improve myocardial diastolic tissue velocities in Type 2 diabetes. Cardiovasc Ultrasound 2007; 5: 32.

36.

Utz

Engeli

Haufe

. Myocardial steatosis, cardiac remodelling and fitness in insulin-sensitive and insulin-resistant obese women. Heart 2011; 97: 1585–1589.

37.

Bedford

. The story of fatty heart. A disease of Victorian times. Br Heart J 1972; 34: 23–28.

38.

Amad

Brennan

Alexander

. The cardiac pathology of chronic exogenous obesity. Circulation 1965; 32: 740–745.

39.

McGavock

Victor

Unger

. Adiposity of the heart, revisited. Ann Intern Med 2006; 144: 517–524.

40.

Korosoglou

Humpert

Ahrens

. Left ventricular diastolic function in type 2 diabetes mellitus is associated with myocardial triglyceride content but not with impaired myocardial perfusion reserve. J Magn Reson Imaging 2012; 35: 804–811.

41.

Schrauwen-Hinderling

Meex

Hesselink

. Cardiac lipid content is unresponsive to a physical activity training intervention in type 2 diabetic patients, despite improved ejection fraction. Cardiovasc Diabetol 2011; 10: 47.

42.

Van der Meer

Hammer

Smit

. Short-term caloric restriction induces accumulation of myocardial triglycerides and decreases left ventricular diastolic function in healthy subjects. Diabetes 2007; 56: 2849–2853.