Author + information
- Received September 8, 2015
- Revision received December 30, 2015
- Accepted January 7, 2016
- Published online August 1, 2016.
- Thomas Emil Christensen, MDa,b,∗ (, )
- Lia Evi Bang, MD, PhDb,
- Lene Holmvang, MD, DMScb,
- Dorthe Charlotte Skovgaard, MD, PhDc,
- Ditte Bang Oturai, MSa,
- Helle Søholm, MD, PhDb,
- Jakob Hartvig Thomsen, MDb,
- Hedvig Bille Andersson, MDb,
- Adam Ali Ghotbi, MDa,b,
- Nikolaj Ihlemann, MD, PhDb,
- Andreas Kjaer, MD, DMSca,c and
- Philip Hasbak, MDa
- aDepartment of Clinical Physiology, Nuclear Medicine, and PET, Centre of Diagnostic Investigation, Rigshospitalet, Copenhagen University Hospital, Copenhagen, Denmark
- bDepartment of Cardiology, The Heart Centre, Rigshospitalet, Copenhagen University Hospital, Copenhagen, Denmark
- cCluster for Molecular Imaging, Faculty of Health and Medical Sciences, University of Copenhagen, Copenhagen, Denmark
- ↵∗Reprint requests and correspondence:
Dr. Thomas Emil Christensen, KF 4011, Department of Clinical Physiology, Nuclear Medicine, and PET, Copenhagen University Hospital, Blegdamsvej 9, 2100 Copenhagen E, Denmark.
Objectives The study sought to investigate adrenergic activity in patients with takotsubo cardiomyopathy (TTC).
Background TTC is a specific type of reversible heart failure possibly caused by excessive catecholamine stimulation of the myocardium. Scintigraphic iodine-123–meta-iodobenzylguanidine (mIBG) imaging of the heart and measurement of plasma catecholamines can be used to assess adrenergic activity in vivo. The authors hypothesized that sympathetic nerve activity is increased in the subacute state of TTC, and this study used cardiac mIBG imaging and plasma levels of norepinephrine and epinephrine as markers to assess this hypothesis.
Methods In this study, 32 patients with TTC and 20 controls were examined at admission and again on follow-up with echocardiography, mIBG scintigraphy, and plasma catecholamine measurements.
Results Ejection fraction (EF) was initially 36 ± 9% but increased to >60% (p = 0.0004) in all patients with TTC. In the control subjects EF was initially higher (51 ± 11%; p = 0.0004) than in the patients with TTC. However, EF of the patients with TTC exceeded that of the control subjects on follow-up (56 ± 8%; p = 0.0007). The mIBG imaging showed a lower late (4-h) heart-to-mediastinum ratio (H/Mlate) (2.00 ± 0.38) and a higher washout rate (WR) (45 ± 12%) in the subacute state of TTC, both when compared with follow-up (H/Mlate: 2.42 ± 0.45; p = 0.0004; WR: 33 ± 14%; p = 0.0004) and when compared with the control group in the subacute state (H/Mlate: 2.34 ± 0.60, p = 0.035; WR: 33 ± 19%, p = 0.026). On follow-up, no differences in mIBG parameters were observed between the TTC and control groups (H/Mlate: 2.41 ± 0.51, p = 0.93; WR: 30 ± 13%, p = 0.48) group. In the TTC group, plasma epinephrine levels were elevated in the subacute state (Log2[epinephrine]: 6.13 ± 1.04 pg/ml), both when compared with follow-up (5.25 ± 0.62 pg/ml; p = 0.0004) and when compared with the control group in the subacute state (5.46 ± 0.69 pg/ml; p = 0.044), and these levels remained elevated in the TTC group on follow-up compared with the control group (4.56 ± 0.95 pg/ml; p = 0.014). No significant differences in plasma norepinephrine levels were observed.
Conclusions The present study supports a possible role of adrenergic hyperactivity in TTC.
- heart failure
- iodine-123–meta-iodobenzylguanidine scintigraphy
- plasma catecholamines
- takotsubo cardiomyopathy
Takotsubo cardiomyopathy (TTC) or transient left ventricular (LV) apical ballooning syndrome is an increasingly recognized cause of acute heart failure. Symptoms and signs are chest pain, dyspnea, abnormalities on the electrocardiogram (ECG), rise in cardiac biomarkers, and clinical signs of heart failure and are thus indistinguishable from the symptoms and signs of acute myocardial infarction. Because of regional wall motion abnormalities with apical and midventricular akinesia and basal hypercontractility, the left ventricle has a characteristic vase shape in end-systole that is known as apical ballooning (1,2). However, acute coronary angiography (CAG) shows no evidence of a culprit lesion. The diagnosis of TTC is usually made on the basis of the 2010 Mayo Clinic Diagnostic Criteria for Tako-Tsubo Cardiomyopathy (3).
The etiology of TTC is not yet completely understood, but several pathophysiological mechanisms have been suggested. Because the onset of TTC is often associated with emotional or physical stress (4) and given that excessive levels of both epinephrine and norepinephrine are known to have cardiotoxic effects (5), one mechanism for the development of TTC is thought to be acute hyperadrenergic myocardial stunning (6).
Increased myocardial sympathetic nerve activity increases exocytosis of norepinephrine from the presynaptic vesicles, and in addition terminal nerve axon reuptake of norepinephrine through the specific uptake-1 transporter is decreased (7). This process results in an increased norepinephrine concentration in the synaptic cleft. Iodine-123–meta-iodobenzylguanidine (mIBG) is a gamma-emitting norepinephrine analogue that shares the same presynaptic uptake, storage, and release mechanisms as norepinephrine, and it is used to visualize cardiac sympathetic nerve activity in vivo (8). Analysis of mIBG images normally includes determination of late heart-to-mediastinum ratio (H/Mlate) and washout rate (WR). H/Mlate offers information about global neuronal function, which is a result of norepinephrine uptake, storage, and release, and WR reflects sympathetic tone (9–11). In patients with heart failure, H/Mlate and WR are independent predictors of prognosis and death (12–15).
Measurement of plasma epinephrine and norepinephrine is an easy and common means of assessing adrenergic drive. This method does have its limitations related to local differences in norepinephrine secretion and clearance and also because single samples do not offer an integral measure of 24-h adrenergic activity (16–18). Plasma catecholamine measurements may be useful, however, if results are interpreted with relevant caution.
Aim and Hypothesis
The aim of this study was to investigate cardiac norepinephrine activity in vivo with mIBG scintigraphy and plasma catecholamine levels in a large, well-characterized population of patients with TTC. Both longitudinal comparisons within the TTC group and cross-sectional comparisons with a control group were made. We hypothesized that cardiac norepinephrine activity and plasma levels of catecholamines would be increased in the subacute state of TTC compared with (A) the TTC group on follow-up and (B) the control group in the subacute state and that (C) no differences between groups would be observed on follow-up.
Included patients were recruited at Rigshospitalet, Copenhagen University Hospital, Copenhagen, Denmark. The TTC group was recruited from patients admitted from December 2010 to July 2014 with suspected acute coronary syndrome (ACS), both ST-segment elevation (STE) and non-STE ACS. Study inclusion criteria were in accordance with the Mayo Clinic TTC criteria (3):
1. Acute onset of symptoms
2. No stenosis >50% of lumen diameter or evidence of a thrombus or vulnerable plaque on acute or subacute CAG
3. Typical apical ballooning on a left ventriculogram (LVG) and/or transthoracic echocardiography (TTE)
4. Elevated levels of both troponin T and creatine kinase MB
5. Normalized LV systolic function on 4-month follow-up
6. Informed patient consent to participate
An expert panel consisting of invasive, imaging, and clinical cardiology specialists reviewed acute and follow-up examinations to verify the TTC diagnosis. The control group consisted of a heterogeneous group of patients with myocarditis or aborted infarction because initial symptoms and signs closely mimicked those of TTC. Control subjects were recruited from May 2012 to February 2014 from patients admitted with suspected acute STE ACS. The inclusion criteria were as follows:
1. Admission ECG fulfilling STE ACS criteria (19)
2. No culprit lesion on acute CAG
3. Absence of typical apical ballooning on LVG and/or TTE
4. Elevated levels of both troponin T and creatine kinase MB
5. Informed patient consent to participate
The study was approved by the Capital Region of Denmark Committee on Health Research Ethics (protocol #H-4-2010-054).
CAG was performed by a team of experienced interventional cardiologists. LVG was performed with automated contrast injection through a pigtail catheter in the left ventricle (iodixanol [Visipaque, GE Healthcare, Brøndby, Denmark], 16 ml/s, 55 ml total), with image recording in the right anterior oblique (30°) position. A single, experienced, blinded operator assessed end-diastolic and end-systolic volumes by planimetric tracing on the 2-dimensional images, and left ventricular ejection fraction (LVEF) was subsequently calculated (20).
TTE was done by experienced sonographers using a Philips IE33 Ultrasound System (Philips Healthcare, Eindhoven, the Netherlands) or Vivid e9 (General Electric, Oslo, Norway). Images were transferred to a remote workstation for offline analysis by Philips Xcelera analysis software version 3.1. All images were analyzed by a single investigator blinded to clinical data and angiographic results. LVEF was calculated on the basis of the wall motion score index (21), and LVEF ≥50% was considered normal (22).
The mIBG scintigraphy was carried out as fast as logistically possible after admission. Patients were given 130 and 20 mg potassium iodine 1 h before and 24 h after tracer injection, respectively, to block thyroid iodine uptake. Then 200 MBq of mIBG was injected intravenously, and planar anteroposterior images of the chest were obtained 15 and 240 min after tracer injection by using the following: a Philips Skylight gamma camera with Jetstream Software (Philips Healthcare); a medium-energy collimator; a 256 × 256 matrix; and an acquisition time of 600 s. 123I was imaged with a 15% energy window set symmetrically over the 159-keV photo peak. Image interpretation was done using the Extended Brilliance Workspace NM Application Suite V188.8.131.52140 (Philips Healthcare). Images were assessed by 1 experienced observer: a region of interest was drawn above the heart by following the epicardial contour, and a rectangular region of interest was drawn above the mediastinum on early and late anterior images in accordance with published guidelines (9). Mean count within each region of interest was reported (23). For each patient H/Mlate, and WR were calculated. WR was calculated as follows:where BKG is background; H is heart mean counts per pixel; M is mediastinum mean counts per pixel; e is early; l is late; and 1.21 is the correction factor for 123I decay at 3 h and 45 min (9).
Plasma epinephrine and norepinephrine concentrations were measured in the subacute state and at follow-up. After a 10-min supine rest, venous blood was collected in aprotinin (Trasylol) 6-ml test tubes (K2E K2EDTA Vacuette, Greiner Bio-One GmbH, Kremsmuenster, Austria). After centrifugation, plasma was stored at logged –80° C. Samples were analyzed twice by a single, experienced laboratory technician using highly sensitive enzyme-linked immunosorbent assay (ELISA) single parameter kits (Adrenaline/Noradrenaline Plasma ELISA High Sensitive, LDN Labor Diagnostika GmbH, Nordhorn, Germany) and a multimode microplate reader (FLUOstar Omega, BMG Labtech GmbH, Ortenberg, Germany). The mean value of the 2 observations was reported in picograms per milliliter.
Statistical Analysis and Graphic Presentation SAS 9.4 (SAS Institute, Cary, North Carolina) software was used for statistical analysis. Normal distributed data were presented as mean ± SD unless stated otherwise, and non-normal distributed data were presented as median (interquartile range [IQR]). The skewed plasma catecholamine concentrations were log2 transformed to a normal distribution for comparison. Comparisons were made as paired or unpaired Student t tests as appropriate. The outcome variables were adjusted for difference in independent variables by using linear regression. All p values were corrected for multiple testing with the Benjamini-Hochberg procedure. Statistical significance for all analyses was set at p < 0.05. Graphic presentation was done with GraphPadPrism 6.00 (GraphPad Software, La Jolla, California).
This study group included 32 patients initially suspected of STE ACS (n = 25) or non-STE ACS (n = 7) and 20 patients included as controls (Figure 1). Initial echocardiography and mIBG scintigraphy were carried out as fast as logistically possible, and follow-up echocardiography and mIBG scintigraphy were performed within the same week approximately 4 months after admission.
The 32 patients in the TTC group (n = 2 men, 6%) had a median age of 68 years (63 to 73 years). Twenty-two patients (69%) had experienced a physically or emotionally stressful event before admission. In 25 patients (78%), the initial ECG fulfilled conventional STE myocardial infarction criteria (19). In addition to the acute CAG, an LVG was performed in 28 patients (88%). The remaining 4 patients had apical ballooning confirmed by TTE. LV systolic function was impaired, planimetric LVEF was 32 ± 2%, and cardiac biomarkers were increased: troponin Tpeak value was 668 ng/l (IQR: 439 to 953 ng/l) (reference <50 ng/l), and creatine kinase MBpeak value was 21 μg/l (IQR: 13 to 31 μg/l) (reference <4 μg/l). Compared with the TTC group, the 20 controls (n = 13 men, 65%) were generally younger, with a median age of 56 years (48 to 63 years). In addition to the acute CAG, LVGs were performed in 8 patients (40%), whereas the absence of apical ballooning was confirmed by TTE in the remaining patients. Cardiac biomarker levels were higher in the controls: troponin Tpeak value was 906 ng/l (range 520 to 1,810 ng/l) (reference <50 ng/l), and creatine kinase MBpeak value was 44 μg/l (20 to 113 μg/l) (reference <4 μg/l) (Figure 2, Tables 1 and 2).
TTE was performed during admission on day 1 (IQR: 1 to 2 days) in the TTC group and on day 1 (IQR: 0 to 1 days) in the control group. In the TTC group, initial TTE showed apical ballooning and basal normokinesia or hyperkinesia. LVEF was reduced to 36 ± 9%. LVEF was initially higher in the control group (51 ± 11%; p = 0.0004). Follow-up TTE was performed 111 days (IQR: 99 to 127 days) and 121 days (IQR: 104 to 134 days) after initial admission in the TTC and control groups, respectively. In the TTC group, LVEF was >60% and LV wall motion was normal in all patients. In the control subjects, LVEF had also increased but was lower than in the patients with TTC (56 ± 8%; p = 0.0007).
In the TTC group, subacute and follow-up examinations were performed 8 days (range 6 to 10 days) and 109 days (IQR: 97 to 129 days) after admission, respectively. In the control subjects, examinations were performed 6 days (IQR: 6 to 9 days) and 121 days (IQR: 112 to 130 days) after admission. In the TTC group, H/Mlate in the subacute state was 2.00 ± 0.38. Thus, it was significantly decreased both when compared with follow-up (2.42 ± 0.45; p = 0.0004) and when compared with the control subjects in the subacute state (2.34 ± 0.60; p = 0.035). Correspondingly, WR was 45 ± 12% in the subacute state, and it was thus increased when compared with follow-up (33 ± 14%, p = 0.0004) and with the control subjects in the subacute state (33 ± 19%, p = 0.026). No difference in mIBG parameters was observed between the TTC and control groups on follow-up. H/Mlate was 2.42 ± 0.45 and 2.41 ± 0.51 (p = 0.93) and WR was 33 ± 14% and 30 ± 13% (p = 0.48) in the TTC and control groups, respectively (Figure 3). We tested the null hypothesis that the difference in WR between the TTC and the control group in the acute state was independent of the difference in LVEF between the 2 groups. Group and LVEF were independent variables, and WR was the dependent variable. The p value was insignificant (p = 0.28), and thus the null hypothesis was verified.
Blood samples for analysis of plasma epinephrine and norepinephrine were obtained 3 days (range 2 to 4 days) and 105 days (range 94 to 128 days) after admission in the TTC group and 3 days (range 2 to 4 days) and 117 days (range 105 to 127 days) after admission in the control group. In the TTC group, plasma epinephrine levels were elevated in the subacute state (Log2[epinephrine]: 6.13 ± 1.04 pg/ml), both when compared with follow-up (Log2[epinephrine]: 5.25 ± 0.62 pg/ml; p = 0.0004) and when compared with the control subjects in the subacute state (Log2[epinephrine]: 5.46 ± 0.69 pg/ml; p = 0.044). In the TTC group, plasma norepinephrine levels in the subacute state (Log2[norepinephrine]: 10.12 ± 0.90 pg/ml) were not different from those observed during follow-up (9.89 ± 0.62 pg/ml; p = 0.24) or in the control group in the subacute state (Log2[norepinephrine]: 9.74 ± 0.74 pg/ml; p = 0.24). The TTC group still had increased plasma epinephrine levels (Log2[epinephrine]: 5.25 ± 0.62 pg/ml) on follow-up when compared with the control group (Log2[epinephrine]: 4.56 ± 0.95 pg/ml; p = 0.014). No difference was observed for norepinephrine between the TTC group (Log2[norepinephrine]: 9.89 ± 0.62) and the control group (Log2[norepinephrine]: 9.44 ± 0.1.00 pg/ml; p = 0.1) on follow-up (Figure 3, Table 2). The p value for the null hypothesis, that difference in plasma epinephrine between the TTC and the control groups in the acute state was independent of the difference in LVEF between the 2 groups, was insignificant (p = 0.89).
The present study prospectively evaluated cardiac norepinephrine activity with mIBG scintigraphy and plasma catecholamine samples in 32 patients with TTC and 20 control subjects. The most important findings were evidence of myocardial sympathetic hyperactivity and concomitantly increased plasma epinephrine in the subacute phase of TTC, both when compared with follow-up and when compared with the control group in the subacute state. On follow-up, complete remission of LV function had occurred in the TTC group, and there was no difference in mIBG parameters when compared with the control group. However, plasma epinephrine levels remained elevated in the TTC group. Our large study made cross-sectional comparisons of the TTC mIBG imaging data, and the diagnosis of TTC was established with care by thorough acute and follow-up examinations.
TTC is primarily seen in postmenopausal women (24), and it normally has a good prognosis with complete clinical remission within weeks (25). TTC is presumably caused by regional catecholaminergic stunning (26), although other mechanisms also have been suggested, including lysed infarction (27) and multivessel spasm (28). Plasma catecholamine levels have been found elevated in patients with TTC compared with patients with acute myocardial infarction and heart failure (Killip class III) (29). A suggested mechanism of the regional LV akinesia is that an excessive epinephrine level in the blood causes a switch from stimulation to inhibition of adenylate cyclase in the intracellular signaling pathway of the ß2-adrenergic receptor that results in a net negative inotropic effect on the cardiomyocytes (30). In a canine model, ß2-adrenergic receptor density increased from the basal toward the apical region (31), a finding that may explain why akinesia characteristically occurs in the midventricular and apical regions. Excessive norepinephrine stimulation has a cardiotoxic effect possibly mediated by the α1-adrenergic receptor (32,33).
Abnormalities of norepinephrine activity in TTC were previously investigated, with mIBG imaging normally showing impaired tracer uptake in the akinetic LV region (34–36). Akashi et al. (37) examined 8 patients with TTC within the first 3 days of admission who had increased WR and decreased H/Mlate in the acute state compared with follow-up, although the latter difference was not statistically significant. These investigators suggested a causal role of sympathetic nerve hyperactivity in TTC (37). Burgdorf et al. (38) found similar results.
Norepinephrine reuptake is blocked by high circulating levels of epinephrine (39), a finding that may account for the low H/Mlate and high WR on mIBG imaging in the acute state of TTC (40). Other studies have also found abnormalities of norepinephrine metabolism that may reflect the hyperadrenergic state in TTC (29,41). The basal region of the left ventricle has a high density of sympathetic nerve endings (42), and hypercontractility causing LV outflow tract obstruction is occasionally present in this region (43). Thus, perhaps TTC should be thought of as 2 separate entities of heart failure with different causes: an epinephrine-induced ß2-adrenergic receptor−mediated regional akinesia of the LV apex and midventricular region; and an LV outflow tract obstruction caused by norepinephrine-induced hyperkinesia of the basal region. Because it can be observed in the subacute state of TTC, sympathetic hyperactivity must be prolonged. This may have clinical implications because pharmacological reduction of the sympathetic hyperactivity may accelerate remission. Given that activation of the β2-receptor subtype is thought to play a key role in the development of the syndrome, nonselective β-blockers may be appropriate in the treatment of TTC. TTC is associated with an anxious and worried personality type (44–46), and persons with this personality type show increased sympathetic reactivity in response to stress (47,48). The increased epinephrine level found in patients with TTC on follow-up when compared with control subjects could thus reflect a transient hyperadrenergic state caused by the stress of re-examination.
Our data were obtained after the onset of TTC, and mIBG scintigraphy was further delayed by limited tracer availability. It is therefore not possible from our results to conclude that increased adrenergic activity has a causal role in TTC. We have demonstrated only that increased cardiac norepinephrine activity is associated with the subacute state of TTC, and it could be argued that this simply represents secondary adaptation in acute heart failure. We have tried to address this issue by adjusting for the difference in LVEF between our TTC and control groups, by showing that the difference in LVEF is unlikely to account for the difference in WR. Because single photon emission computed tomography was not added, reliable predictions about regional tracer distribution in the heart cannot be made. The study was designed to yield a control group in which the disease pattern mimicked TTC as closely as possible. TTC is a syndrome diagnosis, and consequently the demarcation between it and other disease entities is blurred, posing a small risk of inclusion of atypical TTC in the control subjects.
In addition, the age and sex ratio of our prospectively included control subjects differed from that of our study group. Some studies have found no effect of age or sex on mIBG parameters (49), whereas others have found that older age and male sex decrease H/Mlate and increase WR (50,51). Thus, the older age could tend to increase WR and decrease H/Mlate in the study group, whereas the higher proportion of men could tend to increase WR and decrease H/Mlate in the control group.
A 24-h urine collection and analysis for catecholamine metabolites demand a much more complicated setup than the convenient plasma sampling. However, such an analysis would have offered a more robust measure of overall adrenergic activity.
LVGs were performed less often in the control group. Although the relative sensitivity and specificity are unknown, baseline echocardiography is normally regarded as equal to an LVG to define apical ballooning (52).
In the present study, we found that mIBG imaging in the subacute state of TTC was characterized by decreased H/Mlate and increased WR, as well as by increased levels of plasma catecholamines. Thus, the present study supports an assumed role of adrenergic hyperactivity in TTC.
COMPETENCY IN MEDICAL KNOWLEDGE: The presence of adrenergic hyperactivity in the subacute state of TTC has now been documented in a large and carefully examined patient cohort.
TRANSLATIONAL OUTLOOK: That adrenergic hyperactivity plays a role in the pathogenesis of TTC may have important implications for the clinical management of such patients.
All authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- acute coronary syndrome
- coronary angiography
- late heart-to-mediastinum ratio
- left ventricular
- left ventriculogram
- ST-segment elevation
- takotsubo cardiomyopathy
- transthoracic echocardiography
- washout rate
- Received September 8, 2015.
- Revision received December 30, 2015.
- Accepted January 7, 2016.
- American College of Cardiology Foundation
- Hurst R.T.,
- Prasad A.,
- Askew J.W. III.,
- Sengupta P.P.,
- Tajik A.J.
- Caldwell J.H.,
- Link J.M.,
- Levy W.C.,
- Poole J.E.,
- Stratton J.R.
- Kioka H.,
- Yamada T.,
- Mine T.,
- et al.
- Nakata T.,
- Nakajima K.,
- Yamashina S.,
- et al.
- Ogita H.,
- Shimonagata T.,
- Fukunami M.,
- et al.
- Wakabayashi T.,
- Nakata T.,
- Hashimoto A.,
- et al.
- Grassi G.,
- Mark A.,
- Esler M.
- van Berkel A.,
- Lenders J.W.,
- Timmers H.J.
- Thygesen K.,
- Alpert J.S.,
- Jaffe A.S.,
- et al.
- Echocardiographic Normal Ranges Meta-analysis of the Left Heart Collaboration
- Yamada T.,
- Shimonagata T.,
- Fukunami M.,
- et al.
- Paur H.,
- Wright P.T.,
- Sikkel M.B.,
- et al.
- Mori H.,
- Ishikawa S.,
- Kojima S.,
- et al.
- Downing S.E.,
- Lee J.C.
- Akashi Y.J.,
- Nakazawa K.,
- Sakakibara M.,
- Miyake F.,
- Musha H.,
- Sasaka K.
- Pessoa P.M.,
- Xavier S.S.,
- Lima S.L.,
- et al.
- Kastaun S.,
- Schwarz N.P.,
- Juenemann M.,
- et al.
- Summers M.R.,
- Lennon R.J.,
- Prasad A.
- Kalk N.J.,
- Nutt D.J.,
- Lingford-Hughes A.R.
- Nutt D.J.
- Tsuchimochi S.,
- Tamaki N.,
- Tadamura E.,
- et al.
- Sakata K.,
- Shirotani M.,
- Yoshida H.,
- Kurata C.