OBJECTIVES—To measure left ventricular mass (LVM), left ventricular volumes, and left ventricular function (LVF) in a cohort of type 1 diabetic patients and to correlate measures of imaging to NH2-terminal pro-brain natriuretic peptide (NT-proBNP).
RESEARCH DESIGN AND METHODS—In a cross-sectional study, all patients with type 1 diabetes underwent cardiovascular magnetic resonance imaging. We included 63 patients with diabetic nephropathy and 73 patients with normoalbuminuria.
RESULTS—All patients had normal global LVF. LVM was increased in patients with diabetic nephropathy compared with patients with persistent normoalbuminuria. Patients with nephropathy had smaller left ventricular volumes and increased levels of NT-proBNP. Linear regression analysis in patients with diabetic nephropathy showed that NT-proBNP and creatinine were associated with LVM.
CONCLUSIONS—Increased LVM is identified in asymptomatic type 1 diabetic patients with nephropathy compared with normoalbuminuric patients. Elevated levels of NT-proBNP were associated with increased LVM, which are both markers of increased cardiovascular risk.
Cardiovascular disease remains the leading cause of mortality and morbidity in diabetic patients. The increased mortality in type 1 diabetic patients is primarily due to a poor prognosis in patients with diabetic nephropathy. Since ∼40% of type 1 diabetic patients with diabetic nephropathy develop cardiovascular disease (1), it may be beneficial to diagnose cardiac disease early in this population. We have already shown a higher coronary plaque burden in asymptomatic individuals with type 1 diabetes and diabetic nephropathy than in patients with persistent normoalbuminuria (2).
RESEARCH DESIGN AND METHODS—
We randomly included 136 patients with type 1 diabetes: 63 with diabetic nephropathy and 73 with persistent normoalbuminuria. All patients were without symptoms or clinical history of cardiovascular disease. The selection of patients and clinical measurements have previously been described (2).
NH2-terminal pro-brain natriuretic peptide (NT-proBNP) was analyzed by Roche Diagnostics with an immunoassay (3). All 136 patients underwent cardiovascular magnetic resonance imaging (CMR) (Intera 1.5 T MR; Philips, Best, Netherlands). Left ventricular volumes and left ventricular mass (LVM) were measured using a steady-state free procession breath-hold cine sequence. Left ventricular wall stress at end-systole was calculated (4).
To estimate left ventricular filling pressures by CMR, transmitral flow and myocardial tissue velocities were measured (5). To evaluate coronary and aortic plaque burden, subjects underwent black-blood vessel wall imaging according to previously validated protocols (2,6,7).
In analysis of correlations with LVM, variables used in all linear regression analyses were those that were significantly associated with LVM in a univariate analysis: sex, age, creatinine, smoking, systolic blood pressure, diastolic blood pressure, total cholesterol, BMI, hemoglobin, A1C, and NT-proBNP. Univariate correlations of clinical and CMR variables to NT-proBNP were described with r values.
RESULTS
Clinical data, NT-proBNP, and CMR parameters are shown in Table 1. NT-proBNP was significantly increased in patients with nephropathy than in normoalbuminuric patients.
All patients had normal global left ventricular function (LVF). LVM, LVM index, and heart rate were higher in patients with nephropathy. Left ventricular end-diastolic and end-systolic volumes tended to be smaller in those with diabetic nephropathy. Consequently, left ventricular end-systolic wall stress was smaller in patients with nephropathy than in patients with normoalbuminuria. The noninvasively estimated filling pressures (peak flow velocity/peak velocity muscle) were similar in both groups.
In patients with diabetic nephropathy, LVM correlated with NT-proBNP (r = 0.42; P = 0.01). No other CMR measures or plaque burden measures correlated to NT-proBNP. NT-proBNP was positively correlated with systolic blood pressure, urinary albumin excretion, and creatinine and negatively correlated to glomerular filtration rate, BMI, and hemoglobin (P < 0.01).
A1C was correlated with heart rate variability (r = −0.28; P = 0.001), with higher A1C showing a smaller heart rate variation. Furthermore, A1C was correlated with heart rate (r = 0.28; P = 0.001), stroke volume (r = −0.34; P < 0.0001), left ventricular end-systolic volume (r = −0.28; P = 0.001), and left ventricular end-diastolic volume (r = −0.38; P < 0.0001).
In multiple regression analysis, NT-proBNP was correlated with LVM in patients with diabetic nephropathy (with an estimated 13.5-g increase in LVM per 10-fold increase in NT-proBNP; P = 0.04). Creatinine was also correlated with LVM (with an estimated 41.5-g increase in LVM per 10-fold increase in creatinine; P = 0.024). A negative correlation was found between LVM and a 10-year increase in age (P = 0.02) and between LVM and sex (P < 0.0001); thus, older-age individuals and women had lower LVM. In patients with persistent normoalbuminuria, sex (P = 0.043) was associated with smaller LVM and older age was associated with lower LVM (P = 0.009).
In the analysis of all patients together, increased LVM was correlated with increased NT-proBNP and creatinine, whereas increased A1C, older age, and sex were correlated with decreased LVM (P < 001).
CONCLUSIONS—
The present study showed that asymptomatic type 1 diabetic patients with diabetic nephropathy have larger LVM and higher levels of NT-proBNP than patients with normoalbuminuria. All patients had normal LVF and normal filling pressures in accordance with their clinical status. LVM was correlated with NT-proBNP, creatinine, age, and sex in multiple regression analysis in patients with diabetic nephropathy. NT-proBNP was correlated with well-known cardiovascular risk factors.
No patients had left ventricular hypertrophy according to normal ranges for LVM (8). LVM indexes in normoalbuminuric patients were slightly lower (47.4 ± 8.8 ml/m2) than measurements considered normal values in nondiabetic subjects (64.7 ± 9.3 in men and 52.0 ± 7.4 ml/m2 in women) and may reflect differences between the populations (8). In the present study, blood pressure was not correlated with LVM, most likely due to intensive antihypertensive treatment. Patients with nephropathy had considerably lower blood pressure than patients of earlier studies and would be less likely to have LVH. Our study more accurately reflects cardiovascular function in asymptomatic type 1 diabetic patients on contemporary reno- and cardioprotective medication, emphasizing the beneficial effect of blood pressure reduction in diabetic patients. We found a negative correlation between A1C and LVM showing that patients with high levels of A1C had smaller LVM. A possible explanation is that patients with a higher A1C tend to have a higher degree of autonomic neuropathy (9) resulting in a relatively high heart rate with smaller left ventricular dimensions as a consequence to keep the same cardiac output. In accordance, our data showed that higher A1C was correlated with less heart rate variability. Furthermore, high levels of A1C showed an increased heart rate with smaller left ventricular dimensions. Thus, autonomic neuropathy in this normotensive population seems to induce negative left-ventricular remodeling.
Clinical parameters, NT-proBNP, and CMR parameters in 136 type 1 diabetic patients with and without diabetic nephropathy
. | Normoalbuminuria . | Diabetic nephropathy . | P . |
---|---|---|---|
Sex (male/female) | 43/30 | 36/27 | 0.837 |
Age (years) | 52 ± 9 | 48 ± 9 | 0.011 |
Duration of diabetes (years) | 31 ± 7 | 34 ± 9 | 0.017 |
Duration of nephropathy (years) | 0 | 15 ± 7 | — |
BMI (kg/m2) | 24.7 ± 2.7 | 24.4 ± 2.9 | 0.624 |
A1C (%) | 8.0 ± 0.9 | 8.8 ± 1.1 | <0.0001 |
Glomerular filtration rate (ml/min per 1.73m2) | 75 ± 12 | — | — |
Urinary albumin excretion rate (mg/24 h)* | 8 (2–29) | 209 (3–6,375) | <0.0001 |
S-creatinine (μmol/l) | 89 (61–116) | 119 (68–825) | <0.0001 |
Systolic blood pressure (mmHg) | 129 ± 15 | 135 ± 18 | 0.065 |
Diastolic blood pressure (mmHg) | 71 ± 6 | 71 ± 10 | 0.582 |
Total cholesterol (mmol/l) | 4.8 ± 0.8 | 4.7 ± 0.8 | 0.944 |
HDL cholesterol (mmol/l) | 1.9 ± 0.5 | 1.8 ± 0.5 | 0.117 |
LDL cholesterol (mmol/l) | 2.5 ± 0.6 | 2.3 ± 0.8 | 0.250 |
Triglycerides (mmol/l) | 0.9 ± 0.5 | 1.2 ± 0.7 | 0.007 |
Hemoglobin (mmol/l) | 8.5 ± 0.7 | 7.9 ± 0.8 | <0.0001 |
Smoking (%) | 17 (23) | 15 (24) | 0.943 |
NT-proBNP (ng/l) | 44 (6–621) | 77 (5–3,718) | 0.016 |
Statins (%) | 20 (27) | 36 (57) | <0.0001 |
Aspirin (%) | 19 (26) | 45 (71) | <0.0001 |
RAAS-blockade (%)† | 27 (37) | 58 (92) | <0.0001 |
β-Blockers (%) | 2 (3) | 5 (8) | 0.01 |
Average number of AHT drugs | 0.8 | 2.3 | <0.0001 |
CMR parameters | |||
LVM unadjusted (g) | 90.1 ± 20.8 | 105.0 ± 29.8 | 0.001 |
LVMI adjusted for BSA (g/m2) | 47.4 ± 8.8 | 56.5 ± 14.3 | <0.0001 |
LVM adjusted for height (g/m2.7) | 19.9 ± 3.1 | 23.9 ± 5.6 | <0.0001 |
Ejection fraction (%) | 68 ± 8 | 69 ± 9 | 0.555 |
Cardiac output (l/min) | 6.3 ± 1.4 | 6.3 ± 1.4 | 0.868 |
Heart rate (beats/min) | 75 ± 12 | 81 ± 11 | 0.003 |
E/Ea ratio | 7.1 ± 2.7 | 7.1 ± 3.6 | 0.931 |
LVEDV/BSA (ml/m2) | 66.4 ± 13.2 | 60.8 ± 11.6 | 0.010 |
LVESV/BSA (ml/m2) | 21.2 ± 8.4 | 18.6 ± 7.4 | 0.054 |
LVWS (N/m2) | 18.2 ± 0.9 | 14.4 ± 0.9 | 0.002 |
Aortic plaque burden | |||
n | 68 | 60 | |
Thoracic plaque frequency | 0 | 2 (3) | 0.28 |
Abdominal plaque frequency | 11 (16) | 13 (22) | 0.7 |
Coronary plaque burden | |||
n | 33 | 21 | |
RCA VW mean thickness (mm) | 1.3 ± 0.2 | 1.7 ± 0.3 | <0.001 |
RCA VW max thickness (mm) | 1.6 ± 0.3 | 2.2 ± 0.5 | <0.001 |
RCA plaque detected | 5 (15) | 16 (76) | <0.001 |
. | Normoalbuminuria . | Diabetic nephropathy . | P . |
---|---|---|---|
Sex (male/female) | 43/30 | 36/27 | 0.837 |
Age (years) | 52 ± 9 | 48 ± 9 | 0.011 |
Duration of diabetes (years) | 31 ± 7 | 34 ± 9 | 0.017 |
Duration of nephropathy (years) | 0 | 15 ± 7 | — |
BMI (kg/m2) | 24.7 ± 2.7 | 24.4 ± 2.9 | 0.624 |
A1C (%) | 8.0 ± 0.9 | 8.8 ± 1.1 | <0.0001 |
Glomerular filtration rate (ml/min per 1.73m2) | 75 ± 12 | — | — |
Urinary albumin excretion rate (mg/24 h)* | 8 (2–29) | 209 (3–6,375) | <0.0001 |
S-creatinine (μmol/l) | 89 (61–116) | 119 (68–825) | <0.0001 |
Systolic blood pressure (mmHg) | 129 ± 15 | 135 ± 18 | 0.065 |
Diastolic blood pressure (mmHg) | 71 ± 6 | 71 ± 10 | 0.582 |
Total cholesterol (mmol/l) | 4.8 ± 0.8 | 4.7 ± 0.8 | 0.944 |
HDL cholesterol (mmol/l) | 1.9 ± 0.5 | 1.8 ± 0.5 | 0.117 |
LDL cholesterol (mmol/l) | 2.5 ± 0.6 | 2.3 ± 0.8 | 0.250 |
Triglycerides (mmol/l) | 0.9 ± 0.5 | 1.2 ± 0.7 | 0.007 |
Hemoglobin (mmol/l) | 8.5 ± 0.7 | 7.9 ± 0.8 | <0.0001 |
Smoking (%) | 17 (23) | 15 (24) | 0.943 |
NT-proBNP (ng/l) | 44 (6–621) | 77 (5–3,718) | 0.016 |
Statins (%) | 20 (27) | 36 (57) | <0.0001 |
Aspirin (%) | 19 (26) | 45 (71) | <0.0001 |
RAAS-blockade (%)† | 27 (37) | 58 (92) | <0.0001 |
β-Blockers (%) | 2 (3) | 5 (8) | 0.01 |
Average number of AHT drugs | 0.8 | 2.3 | <0.0001 |
CMR parameters | |||
LVM unadjusted (g) | 90.1 ± 20.8 | 105.0 ± 29.8 | 0.001 |
LVMI adjusted for BSA (g/m2) | 47.4 ± 8.8 | 56.5 ± 14.3 | <0.0001 |
LVM adjusted for height (g/m2.7) | 19.9 ± 3.1 | 23.9 ± 5.6 | <0.0001 |
Ejection fraction (%) | 68 ± 8 | 69 ± 9 | 0.555 |
Cardiac output (l/min) | 6.3 ± 1.4 | 6.3 ± 1.4 | 0.868 |
Heart rate (beats/min) | 75 ± 12 | 81 ± 11 | 0.003 |
E/Ea ratio | 7.1 ± 2.7 | 7.1 ± 3.6 | 0.931 |
LVEDV/BSA (ml/m2) | 66.4 ± 13.2 | 60.8 ± 11.6 | 0.010 |
LVESV/BSA (ml/m2) | 21.2 ± 8.4 | 18.6 ± 7.4 | 0.054 |
LVWS (N/m2) | 18.2 ± 0.9 | 14.4 ± 0.9 | 0.002 |
Aortic plaque burden | |||
n | 68 | 60 | |
Thoracic plaque frequency | 0 | 2 (3) | 0.28 |
Abdominal plaque frequency | 11 (16) | 13 (22) | 0.7 |
Coronary plaque burden | |||
n | 33 | 21 | |
RCA VW mean thickness (mm) | 1.3 ± 0.2 | 1.7 ± 0.3 | <0.001 |
RCA VW max thickness (mm) | 1.6 ± 0.3 | 2.2 ± 0.5 | <0.001 |
RCA plaque detected | 5 (15) | 16 (76) | <0.001 |
Data are means ± SD, median (range), and n (%) unless otherwise indicated. Systolic and diastolic blood pressure values are the mean of 24-h measurements.
Some patients with previously persistent albuminuria receiving antihypertensive medication had a urinary albumin excretion <300 mg/24 h. Values are the mean of two 24-h urine collections.
Blockade of the renin-angiotensin-aldosteron system. AHT, antihypertensive; BSA, body surface area [(height in cm × weight in kg)/3,600]1/2; E/Ea ratio, peak flow velocity (E) divided by peak velocity muscle (Ea); LVEDV, left ventricular end-diastolic volume; LVESV, left ventricular end-systolic volume; LVMI, LVM index; LVWS, left ventricular wall stress [3 × LVESV/(LVM/1.05) + 1] × systolic blood pressure; RCA VW, right coronary artery vessel wall.
References
Published ahead of print at http://care.diabetesjournals.org on 5 February 2008. DOI: 10.2337/dc07-1536.
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.