| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1Centre d'Exploration et de Recherche Médicales par Émission de Positons, 69003 Lyon; 2Centre de Recherche et d'Applications en Traitement de l'Image et du Signal, Unité Mixed de Recherche-Centre National de Recherche Scientifique 5515, 69394 Lyon; 3Hôpital Cardiovasculaire et Pneumologique et Université Claude Bernard, 6934 Cedex 03 Lyon; 4Unité 533, Institut National de la Santé et de la Recherche Médicale (INSERM), Unité de Formation et Recherche de Medecine, Hôpital Laënnec, Nantes 44035; and 5Unité 523, INSERM, Institut de Myologie, Groupe Hospitalier Pitié-Salpétrière, Paris 75013, France
Submitted 27 March 2003 ; accepted in final form 22 May 2003
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCLOSURESREFERENCES |
|---|
long QT syndrome; genetics; positron emission tomography; sympathetic nervous system; [11C]hydroxyephedrine
-subunits of
K+ channels (KCNQ1 for LQT1 and HERG for LQT2), the -subunit
of the Na+ channel associated with Na+ channel current
(INa) (SCN5A for LQT3), and two regulatory subunits of
K+ channels minK (KCNE1 for LQT5 regulating KCNQ1) and MiRP1 (KCNE2
for LQT6 regulating HERG)
(31).
These major genetic findings have temporarily pushed into the background
the three pivotal roles attributed to the sympathetic nervous system in LQTS:
a pathogenetic role, a triggering role, and a role as the therapeutic target,
but a new modifier role has been proposed
(22). Unbalanced sympathetic
innervation was suggested to be the substrate (pathogenetic role)
(21,
25) and a sudden increase in
sympathetic tone to be the trigger for the occurrence of life-threatening
arrhythmia (triggering role)
(17,
21,
36). On the basis of this
working hypothesis, therapy with 
-blockers and left sympathetic
denervation are highly successful in preventing sudden cardiac death (role as
therapeutic target) (17,
23). Because the severity of
the disease could greatly differ among LQTS patients carrying identical gene
defects, differential sympathetic activity [i.e., differences in the amount of
norepinephrine (NE) or in adrenergic receptor distribution] has been suggested
as playing a role leading to modifing the propensity toward life-threatening
arrhythmia (modifier role)
(22).
Given that regional heterogeneity in cardiac sympathetic activity could play a modifier role of arrhythmogenicity in LQTS, several attempts have been performed to noninvasively investigate presynaptic cardiac sympathetic nerve activity by single-photon emission tomography (SPECT) or positron emission tomography (PET). However, [123I]metaiodobenzylguanidine ([123I]MIBG) SPECT studies (11, 19, 34) and [11C]hydroxyephedrine ([11C]HED) PET studies (5) have given contradictory results. None of these studies were performed in genotyped LQTS patients.
Because recent data have emphasized the relationship between phenotype (i.e., symptoms) and genotype in LQTS, the purpose of this study was to evaluate cardiac sympathetic innervation with [11C]HED PET in genotyped LQTS patients.
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCLOSURESREFERENCES |
|---|
Eleven patients and eight controls were studied with their written informed
consent according to the Declaration of Helsinki. LQTS was defined according
to established diagnostic criteria
(24). Patients belonged to
five different families with Romano-Ward syndrome (population characteristics
are shown in Table 1). LQTS
patients having a clinical history of diabetes, hypertension, or
hypercholesterolemia were not included in the study, nor were LQTS patients
with any other cardiac disease. At the time of diagnosis, patients with
syncope, recuperated cardiorespiratory arrest, and torsades de pointes were
considered symptomatic. All patients were asymptomatic at the time of the
study. Eight of eleven patients were receiving -blockers at the time of
evaluation (Table 1). No
patient was receiving any other medication. The QT interval measurements were
done in leads II and V5, and values were corrected for heart rate
(QTc, Bazett's formula). The QTc interval in patients
(Table 1) was unrelated to
electrolyte abnormality or any other cause of QT prolongation.
|
Five of eight controls belonged to four of the five studied families. The absence of cardiac disease was established through detailed history, physical examination, and 12-lead resting ECG. All controls had normal QTc intervals.
Genetic Analysis
Patients (Table 1) and controls were genotyped. Genomic DNA was prepared from peripheral blood lymphocytes by standard methods. PCR amplifications of exonic and juxtaexonic sequences of KCNQ1, HERG, KCNE1, KCNE2, and SCN5A genes were produced using previously published primer sequences (1, 4, 6, 30, 32). Screening of mutations was performed either by single-strand conformational polymorphism analysis (Lyon, France) or denaturing high-performance liquid chromatography (DHPLC) (Paris, France), followed by the sequencing of the abnormal PCR fragments on a ABI377 sequencer. In the Nantes laboratory, mutation analysis was conducted by direct sequencing of the genes on a ABI377 sequencer. One hundred chromosomes from unrelated control subjects were screened for unknown mutations.
PET Acquisitions and Analysis
One hour before data acquisition, patients and controls stayed in the supine position in a quiet ambiance. PET studies were performed on a Siemens/ECAT EXACT HR+ 63-slice whole body tomograph.
Perfusion assessment. H215O (185 MBq) was injected intravenously over 10 s. Data acquisition lasted 5 min, and emission scans were reconstructed in a 128 x 128 matrix using a Hanning filter with a cutoff frequency of 0.15 mm–1. The scan sequence consisted of 22 frames: 10 images x 4 s, 2 images x 10 s, 6 images x 20 s, and 4 images x 30 s.
Neural NE retention assessment. NE retention was measured 20 min after H215O PET scans by using the catecholamine analog [11C]HED. [11C]HED PET scans were acquired after the injection of 167 MBq/kg [11C]HED. Data acquisition lasted 40 min, and emission scans were reconstructed in a 128 x 128 matrix using a Hanning filter with a cutoff frequency of 0.15 mm–1. The scan sequence consisted of 14 frames: 6 images x 30 s, 2 images x 60 s, 2 images x 150 s, 2 images x 300 s, and 2 images x 600 s.
Image processing and analysis. A H215O dynamic series was used to obtain myocardial factor images using a factor analysis of medical image sequences (FAMIS) (2, 7, 13). This method estimates a reduced number of factor curves (contained in the signal intensity curves associated with each pixel) and corresponding factor images. The decomposition is performed under the assumption that each time activity curve may be expressed as a linear combination of the factor curves.
FAMIS computation was performed on slices resulting from the addition of three slices of 2.42 mm. To apply FAMIS to these resulting slices, aggregates were generated as square clusters of 4 x 4 pixels wide. Three factor images were systematically extracted, corresponding to the right and left ventricles and the left ventricular myocardium.
Volumetric sampling and analysis of the spatial tracer retention in the ventricle was done using the Munich Heart Package. Spatial tracer retention was displayed as polar maps that represent the maximal tracer activity over 540 samples resulting from the superposition of 36 elements in the x-direction and 15 elements in the y-direction. The polar maps were divided into 37 sectors (Fig. 1). The apex was excluded from analysis because of the partial volume effect. Thus 36 sectors (each sector = 5–10 g tissue) were analyzed.
|
Myocardial factor images (Fig. 2) were used to construct polar maps representing relative perfusion (H2O%) (2, 13). Mean sectorial H2O% was computed as a percentage of the maximal intensity.
|
[11C]HED absolute retention (HEDRet; the averaged activity from 20 to 40 min/the integral of the 11C activity input function; %/min) and relative [11C]HED retention (HED%Ret; percentage of the maximal retention) were depicted as polar maps (Fig. 1). Mean values of HEDRet and HED%Ret for controls (control databases) and patients were available for each sector. Polar maps of each patient were compared with control databases, and sectors with a value 2SD below the mean for controls were defined as "outside sectors."
Relative [11C]HED retention normalized to perfusion was computed by dividing sectorial values of HED%Ret by values of H2O%. Outside sectors of HED%Ret/H2O% were defined following the same criterion explained above.
Statistical Analysis
Values are presented as means ± SD. Normal distribution of data and
SD were tested with Kolmogorov-Smirnov and equal variance tests, respectively.
Global and sectorial differences between groups were tested by use of ANOVA.
When no equal variance between groups was found, the nonparametric
Kolmogorov-Smirnov test was used. Frequencies and proportions of outside
sectors in both groups were analyzed using the 
2 test for
contingency tables. P < 0.05 was defined as statistically
significant.
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCLOSURESREFERENCES |
|---|
Ion channel gene types, mutations, and phenotype expressions are detailed in Table 1. Mutations were identified in three of the five LQTS families. In family I, the A561T HERG mutation (6, 18) was identified, which affects the S5 domain of the channel. Families III and IV presented distinct mutations in the KCNQ1 gene resulting in amino acid change in domains framing the pore. The A371T missense mutation (family III) previously published by Donger et al. (8) was located in the S6 COOH-terminal region. A novel F296L mutation was identified in family IV, occurring in the S5 domain of the protein. This mutation was not found in the study of the 100 unrelated control subjects.
No mutations could be identified in families II and V. The ECG profile of family II was characteristic of HERG alteration, but screening of the HERG gene by DHPLC did not display any abnormal profile. Analysis of the other genes by DHPLC (KCNQ1) or SSCP (KCNE1, KCNE2, and SCN5A) gave normal results. Haplotype analyses were performed on family V using microsatellite markers for all six LQT loci. Only LQT2 markers revealed a potential cosegregation of a disease haplotype in affected patients, suggesting that a mutation in HERG could still cause LQTS in this family. A R1047L mutation was found in patient 9 (family V), but it did not cosegregate with the disease. Furthermore, the R1047L mutation was previously reported as a polymorphism in the Finnish population (14). The sequences of exonic regions of SCN5A, KCNQ1, KCNE1, and KCNE2 genes were also normal for patients 9–11 (family V). The failure to identify HERG mutations in families II and V may thus result from incomplete sensitivities of analytic methods or the presence of mutations in promoter or regulatory sequences.
No LQTS-related mutations were identified in controls.
H215O Analysis
Two short-axis factor images of myocardial perfusion from a control and patient are shown in Fig. 2. Polar maps of perfusion were computed from these factor images.
No sectorial differences between groups were found (Table 2). Variability of perfusion among all sectors (288 in controls and 396 in patients) was similar in both groups (variances 140.4 and 126.6, P = 0.35), and global perfusion (i.e., mean perfusion in the whole left ventricle) did not differ between groups (74.8% and 75.9% for controls and patients, P = 0.2).
|
HED Analysis
The mean of HEDRet for all sectors showed a small decrease in patients (12.2 ± 2.8 vs. 14.1 ± 3.3%/min in controls, P < 0.01), without significant differences when looking at the highest values: 15.3 ± 3.2%/min in patients (n = 11) and 17.1 ± 3.7%/min in controls (n = 8) (P = 0.25).
Similar to HEDRet, global HED%Ret was lower in patients related to controls: 81.0 ± 8.8% vs. 83.3 ± 7.9% (P < 0.01).
In patients, three sectors showed a decreased average HED%Ret related to controls (Table 2 and Fig. 3).
|
Furthermore, the number of outside sectors when looking at
HED%Ret was significantly higher in patients (36 sectors
distributed in 9 patients) than in controls (3 sectors belonging to 3
controls; 2 = 16.8 for number of sectors "up" or
"down" in each group, P < 0.01;
Table 2 and
Fig. 3).
HED%Ret Normalized to Perfusion
The average ratio of HED%Ret to H2O% for all sectors was lower in patients (1.102 ± 0.196 vs. 1.151 ± 0.205 in controls, P < 0.01).
The patients showed a reduced averaged ratio of HED%Ret to H2O% in nine sectors compared with controls (Table 2 and Fig. 4). These sectors were localized in the anteroseptal and anterolateral walls (Fig. 4).
|
When compared with the normal database, 2 outside sectors from 2 controls
and 43 outside sectors from 10 patients were found (2 = 23.5,
P < 0.01; Table 2
and Fig. 4). In patients,
outside sectors were localized in the septal, anterior, and lateral walls
(Fig. 4).
The distribution of outside sectors in symptomatic or asymptomatic patients
is shown in Fig. 5. The five
symptomatic patients were more likely to have outside sectors (33 of 180
sectors in symptomatic patients vs. 10 of 216 sectors in asymptomatic
patients, 2 = 14.0, P < 0.01).
|
Discussion
This study showed that most genotyped LQTS patients have a heterogeneous
and decreased cardiac retention of [11C]HED in the septal,
anterior, and lateral left ventricular walls. Moreover, all symptomatic
patients with prior life-threatening arrhythmia showed a large number of
sectors (
3) with decreased [11C]HED retention, suggesting a
possible relationship between the extension of [11C]HED impairment
and the clinical severity of the disease.
Implications Regarding the Pathogenic Role of Cardiac Sympathetic Nerve Terminals in LQTS
As for [123I]MIBG, decreased retention of [11C]HED could be observed when there were abnormalities in myocardial sympathetic nerve terminals (decreased nerve fiber density, impaired uptake-1 function) or increased neuronal firing reflecting augmented sympathetic activity (because increasing NE release leads to competitive inhibition of [11C]HED reuptake). All possibilities implicate disparate levels of NE reaching myocardial cells. Yamanari et al. (34) reported increased washout of [123I]MIBG together with a decreased retention of [123I]MIBG in cardiac segments of LQTS patients studied by SPECT. Because the coexistence of both phenomena indicates increased sympathetic activity (29), these authors hypothesized that there is a relative activation of cardiac sympathetic terminals in LQTS (that is, a rise of NE concentration in the synaptic cleft). However, the reduction in [11C]HED retention could also be caused by decreasing nerve fiber density, and, in this case, a decreased NE concentration is expected. In both cases, increased or decreased sympathetic activity, heterogeneity of [11C]HED retention in LQTS implicates disparate levels of NE reaching myocardial cells. This is the major finding of our study. As suggested by Schwartz et al. (22), this heterogeneity could play a modifier role as a substrate of reentry for life-threatening ventricular arrhythmias.
It is well known that some forms of congenital and acquired LQTS are exquisitely sensitive to adrenergic stimulation (16, 25, 27, 36). Schwartz et al. (26) reported that life-threatening arrhythmia in 670 LQTS patients tended to occur under specific circumstances in a gene-specific manner. LQT1 patients [i.e., those with mutations affecting slow component of the delayed rectifier K+ current (IKs)] experienced the majority of their lethal cardiac events during sympathetic activation characterized by exercise or emotional stress (88%). In contrast, sympathetic activation occurs in only 56% of LQT2 [i.e., those with mutations affecting the rapid component of the delay rectifier K+ current (IKr)] and 33% of LQT3 (i.e., those with mutations affecting INa). Furthermore, a sudden auditory stimulus is the predominant trigger in LQT2 (33).
There is experimental evidence of the differential effect of
-adrenergic agonists and antagonists in LQT1, LQT2, and LQT3 models of
the LQTS (27). Under normal
conditions, the abbreviation of action potential duration (APD) and of QT
interval duration in response to -adrenergic stimulation is due to a
relatively large increase of IKs and
Ca2+-activated Cl– current
[ICl(Ca)] versus Na+/Ca2+
exchange current (INa/Ca). In LQT1 patients, a defect in
IKs, especially in M cells, could account for the failure
of -adrenergic stimulation to abbreviate the APD and QT interval
(3,
20,
28). Among the three
transmural cell types, M cells have an intrinsically smaller
IKs (15).
The disparate effects of -adrenergic stimulation on the three cell types
results in a persistent increase of transmural dispersion of repolarization
and in the development of spontaneous as well as induced life-threatening
arrhythmia (27). In the LQT2
model, isoproterenol transiently prolongs the APD of the M cell, possibly due
to a more rapid increase of INa/Ca than of
IKs. As a consequence, transmural dispersion of
repolarization and the incidence of life-threatening arrhythmia are only
transiently increased
(27).
The above experimental and clinical observations indicate that the "arrhythmogenic substrate" given by genetic mutations in LQTS could be greatly influenced by sympathetic activity and that transmural dispersion of repolarization (i.e., the substrate of reentry) is modified by adrenergic stimulation. In our study, the greater quantity of outside sectors found in symptomatic patients at the time of diagnosis suggests that heterogeneous sympathetic activity could act by amplifying existing differences in transmural dispersion of repolarization in LQTS. In this way, heterogeneous sympathetic activity could be considered as having an amplifier role able to raise the propensity toward life-threatening arrhythmia and could partially explain the distinct severity of the disease within a family.
Comparison With Prior Studies
There have been several studies on cardiac sympathetic innervation assessed by [123I]MIBG SPECT (11, 19, 34) and only one using [11C]HED PET (5) in LQTS patients. However, none of these studies was performed in genotyped patients. The results using [123I]MIBG SPECT (11, 19, 34) showed decreased regional retention in patients. Moreover, the heterogeneity was found in the septal and anterolateral walls. In contrast, the PET study (5) showed no differences in [11C]HED retention between LQTS and controls. These discrepancies with our study could be partially explained by differences in the patient population and employed methods (number of sectors for analysis and perfusion tracers). First, these authors reported negative results in nongenotyped patients. The population of patients in our study is likely K+ channel related, and no patients with Na+ channel mutations were found. Moreover, Calkins et al. (5) drew 9 sectors for analysis on the polar map, whereas 36 sectors were taken into account in our study. The tomograph used in the present study reaches better resolution and sensibility than the Siemens 931 15-slice tomograph employed by these authors, and it allowed us to increase the number of sectors for analysis. Designed sectors on the polar map represent 10-fold volumes that could be discriminated by the camera. Finally, differences from the use of [13N]ammonia or H215O FAMIS to evaluate perfusion could be addressed. In the dysfunctional ischemic myocardium, H215O yields higher absolute myocardial blood flow values those of [13N]ammonia (10). Nevertheless, so far no studies support the idea of perfusion abnormalities in LQTS, which is confirmed by our results and by the clinical history of patients. Thus ischemic territories in LQTS patients are unlikely, and confounding factors from referred discordance are negligible.
Limitations of the Study
The main limitation of this study as well as that of all previous scintigraphic studies (5, 11, 19, 34) is the small number of investigated patients. Comparison to normal database for outside sector detection–a way to partially circumvent this limitation–showed clear differences between controls and patients, which were concordant with sectorial mean differences between both groups. Nevertheless, the small number of patients probably prevents a better separation between patients with and without symptoms. Therefore, we concluded about the existence of heterogeneity in [11C]HED retention in patients, and we suggest that the degree of this heterogeneity is related to the severity of the disease.
During the PET study, 8 of 11 patients and none of the controls were on
-blockers, and so the two populations differed on an important point.
Indeed, cardiac [11C]HED retention could hypothetically be modified
by -blockers. [11C]HED basically traces the presynaptic
reuptake of catecholamines (uptake 1). This mechanism accounts for
70% of NE released from the nerve terminal. Few studies have focused on the
effects of -blockers on uptake 1. Ziegler et al.
(35) studied the plasmatic
clearance of NE and infused isoproterenol in controls and Shy-Drager patients.
By analyzing the effect of propranolol on NE clearance, they suggested that
propranolol acts on extraneural accumulation (uptake 2) rather than
on uptake 1. Nevertheless, one must be cautious when getting
conclusions regarding the plasmatic behavior of NE and isoproterenol, which
could not reflect cardiac uptake 1. Looking at the extraneural
accumulation (uptake 2) and metabolism of NE, it has been proved in
isolated rat hearts that it can be partially inhibited by the addition of
-blockers (9). Because a
small amount of NE enter the cardiac cell by uptake 2, this effect is
likely negligible. Finally, blockage of prejunctional -adrenoceptors,
whose stimulation induces NE release
(12), could modify the amount
of retained [11C]HED. This effect should be restricted to patients
taking nonselective -antagonists, because human prejunctional
-adrenoceptors are predominantly of the 2-subtype
(12). In our population, six
of eight patients were receiving selective -blockers. The three
untreated patients presented a small number of outside sectors when
[11C]HED retention was normalized to perfusion. Unfortunately, the
number of patients off therapy is too small to allow inferences on
-blocker effects. Similarly, in a small population, Calkins et al.
(5) noticed that the results of
[11C]HED imaging were not different in the five LQTS patients
receiving -blockers versus the four untreated patients. In summary, even
if the question remains unanswered, all available information allowed us to
suppose a moderate impact of -blockers on [11C]HED
retention.
In conclusion, genotyped LQTS patients have a heterogeneous and decreased cardiac retention of [11C]HED in the septal, anterior, and lateral left ventricular walls. The larger number of heterogeneous sectors in symptomatic patients suggests that sympathetic function could play an amplifier role in the severity of the disease.
|
DISCLOSURES |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCLOSURESREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
|
REFERENCES |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCLOSURESREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  A. N. Mazzadi, J. Pineau, N. Costes, D. Le Bars, F. Bonnefoi, P. Croisille, R. Porcher, and P. Chevalier Muscarinic Receptor Upregulation in Patients With Myocardial Infarction: A New Paradigm Circ Cardiovasc Imaging, September 1, 2009; 2(5): 365 - 372. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. M. Raffel, W. Chen, P. S. Sherman, D. L. Gildersleeve, and Y.-W. Jung Dependence of Cardiac 11C-meta-Hydroxyephedrine Retention on Norepinephrine Transporter Density J. Nucl. Med., September 1, 2006; 47(9): 1490 - 1496. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
, individual data; 
,
sectorial average values) and patients (
, individual data; 
,
sectorial average values) in anterior (A), lateral (B),
septal (C), and inferior (D) territories. Curve lines,
control means – 2SD. In patients, sectorial values below 2SD were
defined as "outside sectors." *P < 0.05 vs.
patients.
