(Circulation. 1996;94:973-977.)
© 1996 American Heart Association, Inc.
Articles |
the Service de Cardiologie, Hopital Cochin, Universite Paris V (D.A., Z.O., D.D.); Service Hospitalier Frederique Joliot, CEA, Orsay (D.A., P.M., B.M., M. Fiorelli, A.S.); Service de Reanimation Medicale, Hopital Raymond Poincare, Universite Paris V (D.A., P.G.); Unite INSERM U383, Service de Biologie Moleculaire, Hopital NeckerEnfants Malades, Paris (H.R., C.J.); and Unite INSERM U153, Groupe Hospitalier Pitie SalpetrierePavillon Riesler, Paris (B.E., M. Fardeau), France.
Correspondence to Pascal Merlet, MD, Service Hospitalier Frederique Joliot, CEA, 4, place du General Leclerc, 91406 Orsay, France.
| Abstract |
|---|
|
|
|---|
Methods and Results In 6 control subjects and 10 DM patients, we investigated the coronary blood flow reserve using positron emission tomography with 15O-labeled water. Global and regional flow reserves were obtained from myocardial regions of interest manually drawn on a static FDG image encompassing, respectively, the whole left ventricle and the anterior, septal, and lateral walls. The DNA mutation size was determined on circulating lymphocytes in each DM patient. Compared with control subjects, DM patients had decreased global (2.39±0.39 versus 4.00±0.67, P=.00003) and regional (anterior, 2.39±0.64 versus 3.87±0.92, P=.002; septal, 2.60±0.48 versus 4.00±0.70, P=.0003; lateral, 2.26±0.58 versus 4.16±1.11, P=.0005) coronary reserves. In DM patients, the coronary reserve correlated strongly and inversely with the DNA mutation size (r=-.77, P=.009).
Conclusions The study demonstrated that global and regional coronary reserves are impaired, in relation to the DNA mutation size, in symptom-free DM patients with normal ventricular dimensions and function and normal large coronary vessels. We suggest that a gene-related blunted coronary reserve resulting from an impairment of vascular smooth muscle is an early component of DM cardiomyopathy.
Key Words: dystrophy, myotonic genes cardiomyopathy tomography muscle, smooth
| Introduction |
|---|
|
|
|---|
The heart is sooner or later a target organ in DM patients.4 DM seldom results in clinically overt heart failure, but it may be responsible for atrial or ventricular arrhythmias and for abnormalities of the cardiac conduction system. Moreover, patients with DM are reported to have a higher incidence of sudden death than the general population, and the observed deaths are caused by either complete AV block or ventricular arrhythmias.5 6 Indeed, sudden death has also been reported in patients with well-functioning pacemakers, so ventricular arrhythmias seem to play an important role in the genesis of sudden death. Scattered cases of left ventricular dysfunction at rest have been reported in DM patients with a normal coronary angiogram.7 By contrast, exercise radionuclide angiographic studies have shown that during exercise, the LVEF failed to increase or even decreased in the majority of the patients tested.8 9 Although these findings suggest that myocardial ischemia is a potential component of cardiac involvement in DM, the underlying mechanism remains unclear.
We have shown in DM patients, under conditions of glucose loading, a gene-related decrease in myocardial glucose consumption.10 Since glucose is the primary energetic substrate for myocardial anaerobic conditions,11 any alteration in MBF may cause inadequate metabolic adaptation of the heart to stress in DM patients. The incidence of angiographically abnormal coronary arteries is not increased in DM patients compared with an age- and sex-matched population.12 13 However, the microvasculature is often altered at the level of the digital14 and iris15 vascular beds, and Nguyen et al13 found frequent vascular degeneration at the level of the myocardium in DM patients at autopsy. In the latter study, it may not be clarified whether small myocardial vessels were impaired as a result of an end-stage disease or accounted for the development of DM cardiomyopathy.
The present study aims to evaluate the hypothesis of the presence of MBF and coronary reserve alterations in DM patients without clinically detectable cardiac involvement. Myocardial blood flow and coronary reserve were determined noninvasively by use of PET with 15O-labeled water (H215O).16
| Methods |
|---|
|
|
|---|
Study Population
Patients
Patients who met the established clinical and electromyographic criteria for definite DM17 were considered for entry into the study and subsequently underwent the following examinations. (1) For DNA analysis, DNA was extracted from peripheral blood samples as previously reported.18 DNA (10 µg) was digested with EcoRI following manufacturer's instructions and electrophoresed on agarose gel (0.8%) for 48 hours. The DNA fragments were transferred to Hybond N membrane (Amersham). The filter was hybridized to the [
-32P]dCTP-labeled cDNA25 probe3 at 65°C for 18 hours. Before autoradiography, the filter was washed to a final stringency of 0.1xSSC/0.1% SDS at 65°C for 10 to 30 minutes. (2) A 12-lead ECG was performed. (3) A 2D echocardiogram was performed. (4) Radionuclide angiograms were performed in the anterior and left anterior oblique views that best separated the right and left ventricles in the plane of the interventricular septum after administration of 15 to 20 mCi [99mTc]pertechnetate IV labeled to the patient's red blood cells for determination of resting LVEF.
Ten patients (Table 1
) were eligible for the study and met all of the following criteria: (1) they were ambulatory; (2) they had no past history of ischemic heart disease; (3) they had no history of hypertension, diabetes, or Raynaud's syndrome; (4) they had a resting LVEF >45%; and (5) they had taken no medication during the past 3 months.
|
Control Subjects
Six age- and sex-matched subjects volunteered to undergo the same study. None of them had a past history of coronary artery disease, and they had normal clinical, ECG, exercise stress test, and echocardiographic examinations.
PET Imaging
Before the study, an indwelling catheter was inserted in a forearm vein, allowing intravenous injection of tracers and withdrawal of blood samples for biological measurements.
Data Acquisition
Subjects were positioned in the TTVO3 time-of-flight PET scanner (LETI, CEA). The characteristics of this instrument have been described elsewhere.10 Correct positioning was maintained throughout the study by use of laser beams and skin marks placed on the subject's torso.
MBF experiments consisted of a bolus injection of H215O (0.30 mCi/kg IV). The experimental protocol included two injections with a 20-minute delay to allow for oxygen-15 decay: at baseline conditions and 4 to 6 minutes after injection of dipyridamole (0.80 mg/kg IV at a rate of 10 mg/min). Data were acquired in list mode during the 5 minutes after the arrival of the blood radioactivity in the left ventricular cavity. A dynamic series of images (15x4 seconds and 18x10 seconds) was reconstructed by use of a backprojection algorithm and a 0.5-mm-1 cutoff frequency modified Hanning filter. Images were corrected for attenuation, random events, dead-time losses, and scattered radiation as described elsewhere.19
After completion of the MBF experiments, patients underwent an FDG PET study. Patients were given a 50-g glucose load PO 60 minutes before FDG imaging. After the injection of 0.1 mCi/kg FDG IV, data were acquired in list mode over a period of 60 minutes. A 20-minute static image recorded 40 minutes after injection was used to define myocardial ROIs. This image was reconstructed by the same procedure as described for the H215O study.
PET Data Analysis
ROIs encompassing the anterior, septal, and lateral myocardial walls were drawn manually on two to four consecutive slices of the FDG image. Another ROI was drawn in the left ventricular cavity of the FDG PET slice allowing its best visualization. [15O]H2O time-activity curves were generated in each myocardial ROI. With the left ventricular time-activity curve used as an input function,19 a standard model20 with a single tissue compartment and three parameters (K1, k2, and a spillover fraction fv) was fitted to the PET data by minimization of a weighted least-squares criterion. The MBF was estimated as the blood-to-tissue transfer rate constant K1.20 The Marquardt-type optimizer was always provided with the same parameter initial values: K1=1.0 mL·min-1·mL-1, k2=1.0/min, fv=0.20.
Global and regional coronary reserves were estimated by use of the corresponding myocardial ROIs as the ratio of peak (dipyridamole) K1 to baseline K1.
The heart rate was monitored continuously during the examination, and blood pressure was measured every 2 minutes from baseline and during the whole study. The rate-pressure product was determined before dipyridamole infusion and every 2 minutes up to 10 minutes after.
In case of decreased coronary flow reserve, ie, below 3.5 for the ratio of peak K1 to basal K1 (corresponding to the lowest value in normal subjects in our center), the DM patients underwent, within 1 month, a coronary angiogram to assess the normality of large arterial vessels.
Statistical Analysis
All parameters were expressed as mean±SD. Parameters were compared between control subjects and patients by nonparametric tests (Kruskal-Wallis). Two-way ANOVAs were used to compare between-group repeated measures. The correlation between the MBF and coronary reserve, wall thickness, PR and QRS duration, resting LVEF, and the size of the mutation were studied by use of Pearson correlation tests with Bonferroni-adjusted probabilities. The statistical significance level was set to a value of P=.05.
| Results |
|---|
|
|
|---|
All DM patients had large CTG expansions, ranging from 300 to 3500 pb.
PET MBF Measurements
Control subjects and DM patients showed similar rate-pressure product increases at peak effect of dipyridamole (7492±1538 versus 8018±1731 mm Hg·bpm at baseline and 11 977±2090 mm Hg·bpm at peak, respectively). At baseline, DM patients had MBFs similar to those of control subjects. By contrast, global coronary reserve was dramatically altered in DM patients compared with control subjects (2.39±0.39 versus 4.00±0.67, P=.00003) (Fig 1
). Similarly, compared with control subjects, in DM patients anterior regional reserve was significantly decreased (2.39±0.64 versus 3.87±0.92, P=.002), as well as septal (2.60±0.48 versus 4.00±0.70, P=.0003) and lateral (2.26±0.58 versus 4.16±1.11, P=.0005) regional reserves (see Table 2
).
|
|
In DM patients, no significant difference was shown for regional coronary reserve between anterior, septal, and lateral regions of interest.
Relationship Between Flow Measurements and CTG Expansion Length and ECG, 2D Echocardiography, and Radionuclide LVEF
The CTG repeat lengths correlated significantly and inversely with global coronary reserve (r=-.77, P=.009) (Fig 2
) and did not correlate with baseline MBF or regional coronary reserves.
|
The CTG repeat sizes and flow measurements did not correlate, in DM patients, with ECG, 2D echocardiographic parameters, or resting LVEF.
| Discussion |
|---|
|
|
|---|
In humans, the coronary microcirculation can only be investigated indirectly by measurement of MBF in response to various stressors, mainly adenosine, dipyridamole, or papaverine. The concept is to quantify the ability of the microvasculature to dilate, and the coronary reserve is defined as the ratio of "maximal" to baseline blood flow values.21 22 Besides invasive techniques, ie, Doppler catheters, quantitative coronary arteriography, thermodilution, and measurement of the clearance of inert tracer,23 PET with [15O]H2O16 or [13N]ammonia24 provides noninvasive quantitative assessment of global and regional coronary flow reserve.
Among the available PET tracers, [15O]H2O was chosen because it has the major advantage of being freely diffusible and metabolically inert. Thus, its kinetics are related solely to flow and are not altered by changes in myocardial metabolism. Moreover, the short half-life of 15O (2.1 minutes) allows sequential measurements with a low radiation burden for patients.
To evaluate coronary reserve, the dipyridamole dose was chosen to achieve a maximal coronary vasodilatation.25 In this study, the hemodynamic parameters at baseline and after dipyridamole infusion were similar in DM patients and in control subjects. Thus, the observed blunted coronary reserve in DM patients cannot be explained by different baseline hemodynamic parameters or different systemic hemodynamic response to dipyridamole infusion. An altered left ventricular wall motion and/or thickening increases the partial-volume effect and may lead to an underestimation of MBF.26 This was clearly ruled out in our study, since DM patients had normal left ventricular dimensions and function.
In the present study, the control values for resting MBF and coronary reserve values were closely similar to those previously reported in normal subjects.16 27 The DM patients had normal resting MBF, as previously reported.10 The decrease in flow reserve homogeneously affected the anterior, septal, and lateral myocardial segments, suggesting, in the context of angiographically normal arteries, a diffuse alteration of coronary microvessels or vasomotor tone. Microangiopathy is a known feature, in DM patients, at the level of digital14 and iris15 arteries. Abnormal muscle capillaries were reported in a series of 18 patients.28 Autopsy studies in DM patients commonly showed predominant involvement of tissue conduction with fibrosis and fatty infiltration,29 and one study in a series of 12 patients reported vascular degeneration in the myocardium, in addition to myocyte hypertrophy and fibrosis.13
One explanation for the dysfunction of the microvasculature may be an involvement of smooth muscle, as previously found for the gastrointestinal tract, gall bladder, urinary bladder, ureter, uterus, and eyes.30 The nice inverse correlation between global flow reserves and the length of the CTG expansion may depict a gene-related involvement of smooth muscles at the level of myocardial microvessels. This finding is concordant with the observed correlation between the mutation and the severity of disease31 32 and between the mutation and cardiac involvement10 33 in DM patients. Obviously the determination of the CTG expansion in the lymphocyte is not as accurate as the determination in muscle. Nonetheless, the correlation between the DNA mutation size and the blunted coronary reserve would have been stronger. Obviously, obtaining smooth muscles from microcoronary vessels raises ethical and technical problems in ambulatory DM patients. The size of the genetic defect is closely correlated with global coronary reserve but not with regional coronary reserve. It should be pointed out that global flow reserve was obtained directly from the time-activity curve within an ROI encompassing the whole left ventricular wall and was not calculated as the mean of regional flow reserve. Thus, it is likely that, as underlined by the larger variance for regional flow reserve, static-countrelated errors were less when flow reserve was measured within the largest ROI (eg, global flow reserve) than within the regional ROIs (eg, anterior, lateral, and septal segments).
In our study, blunted coronary reserve was observed in patients without clinical cardiac involvement, suggesting that the microvascular dysfunction is an early component of DM cardiomyopathy. These patients with limited exercise capability resulting from skeletal muscle weakness showed no evidence of ischemia during daily activities. Thus, the low coronary reserve may still be adequate to meet the usual demands of routine stress testing. However, the combination of a microvascular dysfunction and an inability to utilize glucose may lead to ischemic phenomena potentially responsible for cardiac pump insufficiency and ventricular arrhythmias.
Our results are in line with the rather common finding of blunted coronary reserve in different cardiomyopathies of genetic origin (dilated or hypertrophic). In patients with hypertrophic cardiomyopathy, the coronary flow reserve is reduced in both hypertrophic and nonhypertrophic myocardium, suggesting that the alteration of flow reserve is independent of the severity of the hypertrophy.34 Recently, the observation of a blunted coronary reserve in a patient having the mutation of the MYH7 gene without clinically detectable hypertrophic cardiomyopathy suggested that the noninvasive quantification of coronary flow reserve by PET may be an additional marker for familial hypertrophic cardiomyopathy.35
Clinical Implications
The present data suggest that a careful follow-up should be recommended in patients with large mutation size, even though they show no clinical cardiac involvement. In DM patients, any treatment that would improve myocardial perfusion during stress could protect the myocardium from ischemic injury, which is enhanced in the presence of altered glucose utilization, and might therefore delay the occurrence of cardiac events in DM patients. Whether therapeutic vascular smooth muscle relaxation may delay cardiac involvement in DM patients needs further study.
| Selected Abbreviations and Acronyms |
|---|
|
Received August 3, 1995; revision received February 9, 1996; accepted March 4, 1996.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
F. Muntoni Cardiac Complications of Childhood Myopathies J Child Neurol, March 1, 2003; 18(3): 191 - 202. [Abstract] [PDF] |
||||
![]() |
D Babuty, L Fauchier, D Tena-Carbi, P Poret, J Leche, M Raynaud, J P Fauchier, and P Cosnay Is it possible to identify infrahissian cardiac conduction abnormalities in myotonic dystrophy by non-invasive methods? Heart, November 1, 1999; 82(5): 634 - 637. [Abstract] [Full Text] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Circulation Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 1996 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |