Int J Cardiovasc Imaging (2015) 31:399–407 DOI 10.1007/s10554-014-0562-1
ORIGINAL PAPER
Prediction of adverse cardiac events in dilated cardiomyopathy using cardiac T2* MRI and MIBG scintigraphy Michinobu Nagao • Shingo Baba • Masato Yonezawa • Yuzo Yamasaki Takeshi Kamitani • Takuro Isoda • Satoshi Kawanami • Yasuhiro Maruoka • Yoshiyuki Kitamura • Kohtaro Abe • Taiki Higo • Kenji Sunagawa • Hiroshi Honda
•
Received: 8 September 2014 / Accepted: 23 October 2014 / Published online: 28 October 2014 Ó Springer Science+Business Media Dordrecht 2014
Abstract Iron deficiency and cardiac sympathetic impairment play a role in the worsening of heart failure, and these two conditions may be linked. The present study aimed to clarify the relationship between myocardial iron deficiency, cardiac sympathetic activity, and major adverse cardiac events (MACE) in patients with dilated cardiomyopathy (DCM). Cardiac T2* MRI for iron deficiency and 123I-Metaiodobenzylguanidine (MIBG) imaging for cardiac sympathetic activity were performed in 46 patients with DCM. Myocardial T2* value (M-T2*) was calculated by fitting signal intensity data for mid-left ventricular septum to a decay curve using 3-Tesla scanner. 123I-MIBG washout rate (MIBG-WR) was calculated using a polarmap technique with tomographic data. We analyze the ability of M-T2* and MIBG-WR to predict MACE. MIBGWR and M-T2* were significantly greater in DCM patients with MACE than in patients without MACE. Receiveroperating-characteristics curve analysis showed that the optimal MIBG-WR and M-T2* thresholds of 35 % and 28.1 ms, and the two combination predict MACE with C-statics of 0.69, 0.73, and 0.82, respectively. Patients with MIBG-WR \35 % and M-T2* \28.1 ms had significantly M. Nagao (&) S. Kawanami Department of Molecular Imaging and Diagnosis, Graduate School of Medical Sciences, Kyushu University, 3-1-1 Maidashi, Higashi-Ku, Fukuoka City, Fukuoka 812-8582, Japan e-mail: [email protected] S. Baba M. Yonezawa Y. Yamasaki T. Kamitani T. Isoda Y. Maruoka Y. Kitamura H. Honda Department of Clinical Radiology, Graduate School of Medical Sciences, Kyushu University, Fukuoka City, Japan K. Abe T. Higo K. Sunagawa Department of Cardiovascular Medicine, Graduate School of Medical Sciences, Kyushu University, Fukuoka City, Japan
lower event-free rates than those with MIBG-WR C35 % or M-T2* C28.1 ms (log-rank value = 4.35, p \ 0.05). Cox hazard regression analysis showed that v2 and the hazard ratio were 3.99 and 2.15 for development of MACE in patients with MIBG-WR C35 % or M-T2* C28.1 ms (p \ 0.05). Iron deficiency, expressed by a high M-T2*, and MIBG-WR were both independent predictors of MACE in patients with DCM. The two combination was a more powerful predictor of MACE than either parameter alone. Keywords Dilated cardiomyopathy MIBG scintigraphy T2* MRI Major adverse cardiac event
Introduction Idiopathic dilated cardiomyopathy (DCM) is characterized by ventricular dilation and systolic dysfunction, and generally has a poor prognosis. The 1-year mortality rate of patients with idiopathic DCM can be as high as 25 %, whereas the 5-year mortality ranges from 20 to 50 % [1]. Earlier detection of this disease and the introduction of beta-blocker therapy may have led to an improved prognosis [2–4]. Activation of the sympathetic nervous system is one of the cardinal pathophysiological abnormalities associated with congestive heart failure [5]. Cardiac imaging with 123I-meta-iodobenzylguanidine (MIBG), an analogue of norepinephrine, is a useful tool for detecting abnormalities of the myocardial adrenergic nervous system in patients with congestive heart failure [6, 7], and may serve as a marker of responsiveness to drug interventions [8, 9]. Anemia and iron deficiency have recently been demonstrated to be prognostic markers and therapeutic targets
123
400
in patients with chronic heart failure [10–12]. The autonomic nervous system is stimulated globally in response to an anemia-induced decrease in oxygen delivery to peripheral tissues and is augmented more following heart failure. Reduced kidney function is responsible for impairment of hemopoiesis and renal blood flow, leading to activation of the renin-angiotensin system, which can in turn exacerbate heart failure. These observations strongly suggest a critical interaction among anemia, iron deficiency, and kidney dysfunction in heart failure patients, possibly magnifying the unfavorable effects of increased sympathetic activity on clinical outcomes [13]. Therefore, we hypothesized that myocardial iron deficiency and cardiac sympathetic impairment might be linked, and that the combination of these two factors may predict cardiac mortality in DCM. The aim of this study was to quantify myocardial iron content in patients with DCM utilizing T2* magnetic resonance imaging (MRI), and to investigate the relationship between myocardial iron deficiency, cardiac sympathetic activity as assessed by 123I-MIBG imaging, and major adverse cardiac events (MACE) in DCM.
Materials and methods Patients Forty-six patients with DCM and left ventricular ejection fraction (LVEF) \45 % on echocardiography were admitted to our institution with their first episode of congestive heart failure between June 2010 and February 2014. A detailed history and physical examination were obtained prior to enrollment in this prospective study. All patients underwent the following examinations: chest radiography; 12-lead electrocardiography (ECG); echocardiography; and standard blood tests, including measurements of hemoglobin, serum creatinine, and brain natriuretic peptide (BNP). In 36 out of 46 patients with DCM, myocardial biopsy was performed. The majority of section showed the myocardial tissue with chronic inflammatory infiltrate, interstitial fibrosis, nuclear pleomorphism of myocytes, and variation of myocyte size. These findings were compatible with those of DCM. There is neither evidence of granulomatous lesion, amyloidosis (dylon stain), nor hemochromatosis (iron stain). In addition, coronary angiography or stress perfusion imaging was performed when necessary. The patients were in New York Heart Association (NYHA) functional class II or III at the time of enrollment. Patients were excluded from the study for the following reasons: heart failure NYHA functional class IV (two patients), cardiac sarcoidosis (two patients), amyloidosis (one patient), and active myocarditis (one patient) diagnosed by left ventricular (LV) endomyocardial biopsy. Furthermore,
123
Int J Cardiovasc Imaging (2015) 31:399–407 Table 1 Characteristics of 46 patients with DCM Clinical characteristics Age (years)
44 ± 16
Sex (male/female)
32/14
Height (cm)
166 ± 8
Weight (kg)
64 ± 16
NYHA class II/III
12/34
Hypertension
6 (13)
Diabe mellitus
7 (15)
Hyperlipidemia
7 (15)
Systolic blood pressure (mmHg)
103 ± 18
Heart rate (bpm)
84 ± 25
Hemoglobin (g/dl) eGFR (ml/min/1.73 m2)
14.5 ± 1.8 75 ± 29
BNP (pg/mL)
583 ± 578
I-123 MIBG data Early HM ratio
2.2 ± 0.4
Delayed HM ratio
1.9 ± 0.4
Washout rate (%)
32 ± 10
Cardiac MRI data LVEDV (ml)
259 ± 110
LVESV (ml)
203 ± 102
LVEF (%)
25 ± 13
LGE positive
23 (50)
Myocardial T2* (ms)
26.9 ± 4.5
Parenthesis indicates percent
none of the patients had a history of alcohol abuse, congenital heart disease, or severe liver or kidney disease. After stabilization of their clinical condition, 46 patients who underwent both MRI and 123I-MIBG imaging were enrolled in the study. The study was approved by the Ethical Review Board of our institution, and written informed consent was obtained from each subject. Patient backgrounds are summarized in Table 1. Major adverse cardiac events during follow-up Patients were discharged from the hospital on optimal medical treatment and were examined by cardiologists in the outpatient clinic of our hospital at least every 3 months for a mean follow-up period of 27 months (6–50 months). Medical therapy for heart failure included a beta-blocker (n = 46), an angiotensin-converting enzyme inhibitors or angiotensin receptor blocker (n = 39), and a diuretic (n = 30). Physicians determined the necessity for blood tests, ECG, chest radiography, echocardiography and other examinations. The primary endpoint was major adverse cardiac events (MACE) consisting of cardiac death or heart failure hospitalization for cardiac resynchronization therapy (CRT) [14, 15] or a LV assist device (LVAD) [16, 17]
Int J Cardiovasc Imaging (2015) 31:399–407
401
Patients were injected intravenously with 111 MBq 123IMIBG (Fujifilm RI Pharma Co. Ltd., Tokyo, Japan) while in an upright position under resting fasting conditions after an overnight fast. Single-photon emission computed tomographic (SPECT) images were acquired 15 min after injection and repeated 4 h later. Anterior and lateral planar images were acquired immediately after each SPECT acquisition. SPECT imaging was performed with a dual-head rotating gamma camera (Infinia Hawkeye, GE Medical Systems, Waukesha, WI) equipped with a low-energy, general-purpose collimator. Images were acquired for 35 s each at 60 steps over a 360° orbit and were recorded at a digital resolution of 64 9 64 pixels. A 20 % energy window centered on 159 keV was used [18]. After reconstruction of transverse slices using a filtered backprojection method with a
Butterworth filter, a polar map was generated to calculate washout kinetics of myocardial 123I-MIBG activity by avoiding contamination from background activity. The washout rate of cardiac 123I-MIBG activity was calculated, using SPECT data and a polar map technique, as the percentage of change from early to late images with the following correction for physical decay: washout rate = [100 9 (early 123I-MIBG activity-late 123I-MIBG activity)/early 123I-MIBG activity] [18, 19]. Briefly, depending on patient heart size, 12–16 short-axis slices were selected to generate early and late polar maps. For each pixel in the polar maps, washout rate was calculated using early and late 123I-MIBG data to obtain the mean value that was used as a single washout rate for each patient (MIBG-WR) (Fig. 1). The heart/mediastinum count (H/M) ratio was determined from anterior planar early and delayed 123I-MIBG images, where H was the mean count/pixel in the left ventricle, and M was the mean count/pixel in the upper mediastinum (Fig. 1). The early and delayed H/M ratios were calculated. The MIBG-WR, early H/M ratio, and delayed H/M ratio were used as estimates of cardiac sympathetic activity.
Fig. 1 123I-MIBG images for a 25-year-old man who had cardiac death. a Early 123I-MIBG polar map shows that accumulation declines in the septa and inferior wall (left). Delayed 123I-MIBG polar map shows a diffuse accumulation decline (center) and the washout rate polar map demonstrates an increase in washout rate (right). His
MIBG-WR was 50 %. b Calculation of 123I-MIBG heart to mediastinum (H/M) ratio based on an anterior view of the thorax in the early scan (left) and delayed scan (right). ROIs have been drawn over the heart (H) and mediastinum (M). His early and delayed H/M ratios were 1.88 and 1.45, respectively
prior to heart transplantation because of pump failure that was refractory to medical treatment. MIBG imaging and quantification of cardiac MIBG activity
123
402
Int J Cardiovasc Imaging (2015) 31:399–407
Fig. 2 Myocardial T2* image for a 58-year-old man without MACE (left). One large ROI was placed in the mid-left ventricular septum, and the mean myocardial intensity for the ROI was measured by each T2* image with six different echo times (right)
Cardiac MRI All patients underwent 3-Tesla MRI using a scanner (Achieva 3.0T Quarsar Dual; Philips Healthcare, Best, The Netherlands) equipped with dual-source parallel radiofrequency transmission, 32-element cardiac phased-array coils for radiofrequency reception, and a four-lead vector cardiogram used for cardiac gating. Cine balanced turbo field-echo sequences in two-, three- and four-chamber views and a stack of short axis images acquired in parallel to the atrioventricular groove from the base to apex were performed with the following imaging parameters: repetition time, 2.8 ms; echo time, 1.4 ms; flip angle, 45°; slice thickness, 8 mm; field of view, 380 mm; matrix size, 176 9 193; SENSE factor, 2; and 20 cardiac phases/R–R interval on the ECG. A mid-ventricular short axis slice was obtained for T2* measurements. The black blood T2* acquisition used a multi-echo gradient-echo sequence performed during a breath hold (flip angle 30°; matrix size 200 9 123; sample bandwidth 2,275 Hz/pixel; slice thickness 8 mm; and field of view 320 mm). The short-axis images were acquired within a single breath-hold at 6 echo times from 2.9 to 10.3 ms at approximately 1.5-ms increments. A double inversion recovery (DIR) pulse was triggered off of the R-wave, and the inversion time extended into diastole, generating a similar set of six images at increasing echo times [20] (Fig. 2). Subsequently, 0.1 mmol/kg of gadolinium contrast (Magnevist: Bayer Healthcare, Osaka, Japan) was administered via the antecubital vein and flushed with 30 ml of isotonic saline. Images with late gadolinium enhancement (LGE) were obtained with an inversion-recovery threedimensional (3D) T1 turbo field-echo sequence (fast gradient-echo pulse sequence) performed 10 min after contrast injection and acquired in the short-axis images (repetition
123
time 3.9 ms; echo time 1.2 ms; flip angle 15°; field of view 400 mm; matrix size 304 9 179; slice thickness 4 mm; SENSE factor 2). The inversion time was adjusted to a completely null normal myocardium. Cardiac function, myocardial T2*, and LGE analysis Cine images were analyzed with the use of dedicated software (Extended Work Space, Philips Healthcare, Best, The Netherlands). Initially, short-axis images were previewed from the base to the apex in a cinematic mode, then endocardial and epicardial contours for end-diastole and end-systole were manually traced. Delineated contours were used for the quantification of LV volumes and LVEF. For T2* measurements, one large full-thickness region of interest (ROI) was placed in the mid-left ventricular septum, distant from the lungs and cardiac veins, which are otherwise known to cause susceptibility artifacts [21] (Fig. 2). In addition, absolute iron contents and T2* measured in the mid-septum are highly representative of their respective global myocardial value [22]. The signal intensity of these ROIs was plotted against the echo time used for each image. The resulting points form an assumed exponential decay curve, as the image signal decreases with increasing echo time. An exponential function was fitted to the data, using the following equation:
y ¼ KeTE=T2
where K represents a constant, TE represents the echo time, and y represents the image signal intensity (Fig. 3). In our laboratory, the myocardial T2* value (M-T2*) in 50 patients with no history of cardiac disease and LVEF [55 % (mean age 59 years) was 22 ± 3 ms. The short-axis LGE image was assessed using the American Heart Association/American College of Cardiology classification of 17 standardized LV short-axis
Int J Cardiovasc Imaging (2015) 31:399–407
segments. LGE was considered positive if the hyperenhanced area involved [50 % of the wall thickness. Patients with more than one segment with a hyperenhanced area were defined as positive for the presence of LGE; the presence of LGE was used a marker for myocardial fibrosis. Statistical analysis Continuous data are expressed as the mean ± standard deviation (SD). For comparison between DCM patients with and without MACE, the Mann–Whitney U test was used. The correlation between M-T2* and the other parameters was determined by Pearson’s correlation coefficient. Receiver-operating-characteristic (ROC) curve analysis was performed to determine the optimal cutoff of the parameters
403
for the prediction of MACE. After univariate logistic regression analysis, multivariate logistic regression analysis was performed using the variables that were significant in univariate analysis (p \ 0.05) to determine their relationship with MACE. The significance of the coefficients in the logistic regression model was determined by Chi square. Survival curves of patient subgroups were created by the Kaplan–Meier method to clarify the time-dependent, cumulative event-free rate and were compared using the logrank test. The Cox proportional hazards regression analysis was performed to evaluate for factors that were associated with the development of MACE. All statistical tests were two-sided. A p value of \0.05 was considered significant. All analyses were performed using the JMP statistical program package (version 9.0; JMP, Inc.).
Results
Fig. 3 Graph shows methods of calculating the myocardial T2* value. The abscissa represents six different echo times, and ordinate represents natural logarithm (LN) of the mean signal intensity for the ROI. The negative reciprocal of the slope of line from the linear regression equation corresponds to the T2* value. The 25-year-old man with MACE (filled circle) (y = -0.0367 x ? 6.897; r2 = 0.99; T2* = 27.2 ms) is the same subject as in Fig. 1
Fig. 4 Cardiac MRI for a 25-year-old man with cardiac death. The subject is the same as in Fig. 1. The four-chamber cine MR image shows a diffuse thinning of the myocardial wall and dilated left ventricular cavity and left atrium (left). Late gadolinium enhancement
During follow-up in the 46 patients with DCM, heart failure hospitalization was performed in six patients, an LVAD was utilized in three patients, and an intra-aortic balloon pumping was utilized in one patient due to refractory pump failure. Three patients died without LVAD utilization. Representative 123I-MIBG and MRI images are presented in Figs. 1 and 4. MIBG-WR was significantly greater in DCM patients with MACE (n = 13) than in patients without MACE (n = 33) (36.9 ± 6.9 vs. 30.1 ± 10.7 %, respectively, p = 0.04). M-T2* was significantly greater in DCM patients with MACE than in patients without MACE (30.0 ± 4.0 vs. 25.7 ± 4.1 ms, respectively, p = 0.01) (Fig. 5). There was no significant difference in the other
MRI shows a liner enhancement area (arrow) in the mid-ventricular septum (center). The right side shows a myocardial T2* image (echo time 2.9 ms) at the same site of late gadolinium enhancement MRI
123
404
Int J Cardiovasc Imaging (2015) 31:399–407
Fig. 5 Scatterplots of MIBGWR (left) and M-T2* (right) for patients with and without MACE. The MIBG-WR and M-T2* were significantly greater in patients with MACE than those without MACE (*; p \ 0.05)
Table 2 Comparison between patients with and without MACE
Table 3 Pearson correlation coefficient of M-T2*
Parameter
Parameter
Pearson r
p 0.83
With MACE n = 13
Without MACE n = 33
Age
-0.032
Age (years)
44 ± 20
44 ± 15
NS
Systolic blood pressure
-0.068
0.66
Systolic blood pressure (mmHg)
96 ± 16
106 ± 19
NS
Hemoglobin
-0.117
0.44
79 ± 16 14.2 ± 1.4
86 ± 28 14.6 ± 2.0
NS NS
eGFR
0.108
0.48
BNP
0.021
0.89
Heart rate (bpm) Hemoglobin (g/dl) eGFR (ml/min/1.73 m2) BNP (pg/mL)
p
79 ± 37
73 ± 25
NS
623 ± 488
566 ± 617
NS
I-123 MIBG data Early HM ratio
2.1 ± 0.4
2.3 ± 0.4
NS
Delayed HM ratio
1.8 ± 0.3
1.9 ± 0.4
NS
Washout rate (%)
37 ± 7
30 ± 11
\0.05
180 ± 73
143 ± 53
NS
23 ± 13
26 ± 12
NS
Cardiac MRI data LVEDVi (ml/m2) LVEF (%) No. of patient with LGE
7 (54)
Myocardial T2* (ms)
30.0 ± 4.0
16 (48) 25.7 ± 4.1
\0.05
I-123 MIBG data Early HM ratio
-0.22
0.14
Delayed HM ratio
-0.21
0.16
0.24
0.11
0.137 -0.042
0.37 0.78
Washout rate Cardiac MRI data LVEDVi LVEF
showed that v2 and the hazard ratio were 3.99 and 2.15 for development of MACE in patients with MIBG-WR C35 % or M-T2* C 28.1 ms (p \ 0.05) (Table 5).
Parenthesis indicates percent
Discussion parameters between the two groups (Table 2). Furthermore, there was no correlation between the T2* value and any other factor (Table 3). Multivariate logistic regression analysis showed that the v2 and the odds ratio to predict MACE were 2.27 and 1.06 for MIBG-WR (p = 0.132), and 7.31 and 1.27 for M-T2* (p = 0.007). According to univariate logistic regression analysis, no other factors were able to predict MACE (Table 4). ROC curve analysis revealed that the optimal MIBG-WR threshold was 35 % for predicting MACE, with a sensitivity of 69 % and a specificity of 67 %. The optimal M-T2* threshold was 28.1 ms for predicting MACE, with a sensitivity of 54 % and a specificity of 73 %. Furthermore, when the two parameters were combined to predict MACE, the sensitivity was 100 %, and the specificity was 64 % (Fig. 6). Patients with MIBG-WR \35 % and M-T2* \28.1 ms had significantly lower event-free rates than those with MIBGWR C35 % or M-T2* C28.1 ms (Log-rank value = 4.35, p \ 0.05) (Fig. 7). The cox hazard regression analysis
123
The present study quantified myocardial iron content using T2* MRI and cardiac sympathetic activity using MIBG imaging in patients with moderate to advanced DCM. M-T2* and MIBG-WR were significantly associated with the occurrence of MACE in patients treated with optimal medical therapy that included a beta-blocker. On the other hand, we did not find a correlation between M-T2* and any of the 123I-MIBG imaging findings. Further, M-T2* was not correlated with any other biomarker of heart failure (Table 2); therefore, M-T2* is an independent predictive marker of MACE. Finally, the combination of MIBG-WR and M-T2* could predict MACE with a C-statics of 0.82, which was superior to the predictive ability of either parameter alone, and improve risk stratification of DCM patients for MACE. Cardiac sympathetic innervation and norepinephrine kinetics at nerve endings in failing hearts can be quantitatively assessed using 123I-MIBG imaging [23]. Sustained
Int J Cardiovasc Imaging (2015) 31:399–407 Table 4 Relation to risk for MACE in 46 patients with DCM
Parameter
405
Univariate logistic analysis 2
v
Odds ratio
p
Age
0.001
1.001
0.97
Systolic blood pressure
2.93
0.97
0.11
Hemoglobin
0.43
0.88
0.51
eGFR
0.42
1.007
0.52
BNP
0.09
1.0002
0.76
Early HM ratio
2.33
0.255
0.14
Delayed HM ratio
1.81
0.278
0.19
Washout rate
4.64
1.08
0.04
LVEDVi LVEF
3.59 0.86
1.011 0.97
0.07 0.37
T2*
9.68
1.31
0.002
Multivariate logistic analysis v2
Odds ratio
95 % CI
p
2.27
1.06
0.98–1.16
0.132
7.31
1.27
1.06–1.62
0.007
I-123 MIBG data
Cardiac MRI data
Fig. 6 Predictive value of risk for MACE using receiver-operatingcharacteristics analysis. M-T2* (center) could predict MACE with slight higher C-statistics than MIBG-WR (left). The combination of
MIBG-WR and M-T2* (right) could predict MACE with a sensitivity of 100 % and a specificity of 64 %, which was superior to the predictive ability of either parameter alone
Table 5 Cox hazard regression analysis for development of MACE in 46 patients with DCM
Fig. 7 Event-free curves of two groups classified by cutoffs of MIBG-WR C35 % or \35 % and M-T2* C28.1 ms or \28.1 ms. Patients with MIBG-WR \35 % and M-T2* \28.1 ms (in red) had significantly lower event-free rates than those with MIBG-WR C35 % or M-T2* C28.1 ms (in blue)
v2
Hazard ratio
95 % CI
p
MIBG-washout ratio C35 versus 1.45 \35 %
1.61
0.73–3.42
0.23
T2* C28.1 versus \28.1 ms
1.07
1.54
0.66–3.33
0.3
Washout ratio C35 % or T2* C28.1 ms versus washout ratio \35 % and T2* \ 28.1 ms
3.99
2.15
1.01–4.67
0.04
Parameter
excess activation of cardiac sympathetic tone initially plays a compensatory role in patients with mild to moderate heart failure, but there is decreased the efficiency of reuptake,
123
406
turnover, and storage of norepinephrine at presynaptic endings in the myocardium as heart failure progresses [24]. This process leads to an increase in norepinephrine concentration at the sympathetic cleft and desensitization of beta-adrenoceptor, ultimately resulting in loss of norepinephrine content and impairment of sympathetic innervation at an advanced stage of heart failure. Increased norepinephrine spillover, decreased norepinephrine content in nerve endings, and sympathetic denervation in failing hearts can be revealed as an increase in 123I-MIBG washout rate from the heart. Several studies have demonstrated that abnormal 123I-MIBG kinetics are associated with lethal cardiac events in patients with moderate to severe heart failure [25–27]. MIBG-WR was superior to the early and delayed H/M ratios for predicting MACE. This is consistent with observations on 123I-MIBG kinetics in previous studies [25–27]. Particulate intracellular iron causes shortening of the MR relaxation parameter, T2*, on the basis of microscopic magnetic field inhomogeneity. In fact, T2* MRI is an easily quantifiable, clinically robust, and highly reproducible measurement technique [20, 28] that has been used for the assessment of myocardial iron overload in patients with thalassemia. However, the assessment of myocardial iron deficiency in patients with heart failure has not been examined. This is the first report to describe the clinical importance of high M-T2* in DCM. Iron plays a key intracellular role in many cellular processes. In cardiac myocytes, this includes an important role within the mitochondria and in myoglobin [22]. Dietary iron deficiency in rats results in DCM characterized by aberrant mitochondrial and irregular sarcomere organization and an increase in reactive nitrogen species and RhoA expression. Morphological examination showed that ventricular diameters and ventricular volume were significantly enhanced in iron-deficient rats [29]. In our laboratory, the mean M-T2* for 50 subjects with no history of cardiac disease and LVEF [55 % was 22 ms. This value was much lower than that for 46 patients with DCM (mean 26.9 ms). These observations support our hypothesis that myocardial iron deficiency (as expressed by a high M-T2*) is associated with the adverse DCM. Another interesting finding was that M-T2* was a good biomarker for predicting MACE, compared with other biomarkers (such as BNP, systolic blood pressure, hemoglobin, and eGFR). The present study included only a few patients with severe anemia or renal dysfunction. This may be the reason for the lack of an association between hemoglobin or eGFR and MACE. The relationship between M-T2* and serum iron parameters is basically analyzing iron metabolism; however, it is not possible to determine whether patients had absolute iron deficiency (reduced iron stores) or relative iron deficiency (decreased systemic iron availability despite overall normal total/body
123
Int J Cardiovasc Imaging (2015) 31:399–407
iron) from our data. In the present study, serum Fe and ferritin were measured in 16 (35 %) and 13 (28 %) out of 46 patients. Pearson correlation coefficient of T2* showed r = 0.12 for serum Fe and r = -0.35 for ferritin, respectively. T2* might be related to reduced iron stores, expressed as the decline of ferritin. We acknowledge that the present study is limited by a small number of patients at a single center that were followed-up for relatively short time period. Furthermore, M-T2* was measured using just one type of MRI system at a single center. The transferability of the T2* technique is important to increase the utility of M-T2* measurements and would help to assure that results from different MRI scanners at different centers would be comparable.
Conclusions Myocardial iron deficiency, expressed by a high M-T2*, and MIBG-WR were both independent predictors of MACE in patients with DCM. However, there was no correlation between M-T2* and sympathetic functional parameters measured by MIBG imaging. The combination of M-T2* and MIBG-WR was a more powerful predictor of MACE than either parameter alone. Acknowledgments This work was supported by the Japan Society for the Promotion of Science (JSPS) KAKENHI (25461831). Conflict of interest
None.
References 1. Sugrue DD, Rodeheffer RJ, Codd MB, Ballard DJ, Fuster V, Gersh BJ (1992) The clinical course of idiopathic dilated cardiomyopathy: a population-based study. Ann Intern Med 117:117–123 2. MERIT-HF Study Group (1999) Effect of metoprolol CR/XL in chronic heart failure: metoprolol CR/XL randomised intervention trial in congestive heart failure (MERIT-HF). Lancet 353:2001–2007 3. Packer M, Bristow MR, Cohn JN et al (1996) The effect of carvedilol on morbidity and mortality in patients with chronic heart failure. U.S. Carvedilol Heart Failure Study Group. N Engl J Med 334:1349–1355 4. Packer M, Coats AJ, Fowler MB et al (2001) Carvedilol prospective randomized cumulative survival study group. Effect of carvedilol on survival in severe chronic heart failure. N Engl J Med 344:1651–1658 5. Thomas JA, Marks BH (1978) Plasma norepinephrine in congestive heart failure. Am J Cardiol 41:233–243 6. Schofer J, Spielmann R, Schuchert A, Weber K, Schluter M (1998) Iodine-123 meta-iodobenzylguanidine scintigraphy: a noninvasive method to demonstrate myocardial adrenergic nervous system disintegrity in patients with idiopathic dilated cardiomyopathy. J Am Coll Cardiol 12:1252–1258 7. Merlet P, Valette H, Dubois-Rande JL et al (1992) Prognostic value of cardiac metaiodobenzylguanidine imaging in patients with heart failure. J Nucl Med 33:471–477
Int J Cardiovasc Imaging (2015) 31:399–407 8. Takeishi Y, Atsumi H, Fujiwara S, Takahashi K, Tomoike H (1997) ACE inhibition reduced cardiac iodine-123-MIBG release in heart failure. J Nucl Med 38:1085–1089 9. Suwa M, Otake Y, Moriguchi A et al (1997) Iodine-123 metaiodobenzylguanidine myocardial scintigraphy for prediction of response to beta-blocker therapy in patients with dilated cardiomyopathy. Am Heart J 133:353–358 10. Tang YD, Katz SD (2006) Anemia in chronic heart failure: prevalence, etiology, clinical correlates, and treatment options. Circulation 113:2454–2461 11. Groenveld HF, Januzzi JL, Damman K et al (2008) Anemia and mortality in heart failure patients: a systematic review and metaanalysis. J Am Coll Cardiol 52:818–827 12. Doi T, Nakata T, Hashimoto A et al (2012) Cardiac mortality assessment improved by evaluation of cardiac sympathetic nerve activity in combination with hemoglobin and kidney function in chronic heart failure patients. J Nucl Med 53:731–740 13. Cohn JN, Levine TB, Olivari MT et al (1984) Plasma norepinephrine as a guide to prognosis in patients with chronic congestive heart failure. N Engl J Med 311:819–823 14. Abraham WT, Fisher WG, Smith AL et al (2002) Cardiac resynchronization in chronic heart failure. N Engl J Med 346:1845–1853 15. Beshai JF, Grimm RA, Nagueh SF et al (2007) Cardiac-resynchronization therapy in heart failure with narrow QRS complexes. N Engl J Med 357:2461–2471 16. Slaughter MS, Rogers JG, Milano CA et al (2009) Advanced heart failure treated with continuous-flow left ventricular assist device. N Engl J Med 361:2241–2251 17. Goldstein D, Zucker M, Pagani FD, Frazier OH (2006) Rotary ventricular assist devices. In: Kirklin JK, Frazier OH (eds) ISHLT monograph series: mechanical circulatory support. Elsevier, New York, pp 77–104 18. Nakata T, Miyamoto K, Doi A et al (1998) Cardiac death prediction and impaired cardiac sympathetic innervation assessed by metaiodobenzylguanidine in patients with failing and non-failing hearts. J Nucl Cardiol 5:579–590 19. VanderVeen BJ, Younis I, DeRoos A, Stokkel M (2012) Assessment of global cardiac I-123 MIBG uptake and washout
407
20.
21.
22. 23.
24.
25.
26.
27.
28.
29.
using volumetric quantification of SPECT acquisitions. J Nucl Cardiol 19:752–760 Kirk P, He T, Anderson LJ, Roughton M et al (2010) International reproducibility of single breathhold T2* MR for cardiac and liver iron assessment among five thalassemia centers. J Magn Reson Imaging 32:315–319 Reeder SB, Faranesh AZ, Boxerman JL, McVeigh ER (1998) In vivo measurement of T2* and field inhomogeneity maps in the human heart at 1.5T. Magn Reson Med 39:988–998 Andrews NC (2008) Forging a field: the golden age of iron biology. Blood 112:219–230 Wieland DM, Wu JI, Brown LE, Mangner TJ, Swanson DP, Beierwaltes WH (1980) Radiolabeled adrenergic neuron-blocking agents: adrenomedullary imaging with 131I-metaiodobenzylguanidine. J Nucl Med 21:349–353 Eisenhofer G, Friberg P, Rundqvist B et al (1996) Cardiac sympathetic nerve function in congestive heart failure. Circulation 93:1667–1676 Nakata T, Wakabayashi T, Nagahara D (2005) Assessment of cardiac innervation in heart failure and lethal arrhythmias: therapeutic and prognostic implications. Curr Cardiol Rev 1:29–36 Nishisato K, Hashimoto A, Nakata T et al (2010) Impaired cardiac sympathetic innervation and myocardial perfusion are related to lethal arrhythmia: quantification of cardiac tracers in patients with ICDs. J Nucl Med 51:1241–1249 Jacobson AF, Senior R, Cerqueira MD et al (2010) Myocardial iodine-123 meta-iodobenzylguanidine imaging and cardiac events in chronic heart failure: results of the prospective ADMIRE-HF (AdreView myocardial imaging for risk evaluation in heart failure) study. J Am Coll Cardiol 55:2212–2221 Westwood MA, Anderson LJ, Firmin DN et al (2003) Interscanner reproducibility of cardiovascular magnetic resonance T2* measurements of tissue iron in thalassemia. J Magn Reson Imaging 18:616–620 Dong F, Zhang X, Culver B, Chew HG Jr, Kelley RO, Ren J (2005) Dietary iron deficiency induces ventricular dilation, mitochondrial ultrastructural aberrations and cytochrome c release: involvement of nitric oxide synthase and protein tyrosine nitration. Clin Sci (Lond) 109:277–286
123
ORIGINAL PAPER
Prediction of adverse cardiac events in dilated cardiomyopathy using cardiac T2* MRI and MIBG scintigraphy Michinobu Nagao • Shingo Baba • Masato Yonezawa • Yuzo Yamasaki Takeshi Kamitani • Takuro Isoda • Satoshi Kawanami • Yasuhiro Maruoka • Yoshiyuki Kitamura • Kohtaro Abe • Taiki Higo • Kenji Sunagawa • Hiroshi Honda
•
Received: 8 September 2014 / Accepted: 23 October 2014 / Published online: 28 October 2014 Ó Springer Science+Business Media Dordrecht 2014
Abstract Iron deficiency and cardiac sympathetic impairment play a role in the worsening of heart failure, and these two conditions may be linked. The present study aimed to clarify the relationship between myocardial iron deficiency, cardiac sympathetic activity, and major adverse cardiac events (MACE) in patients with dilated cardiomyopathy (DCM). Cardiac T2* MRI for iron deficiency and 123I-Metaiodobenzylguanidine (MIBG) imaging for cardiac sympathetic activity were performed in 46 patients with DCM. Myocardial T2* value (M-T2*) was calculated by fitting signal intensity data for mid-left ventricular septum to a decay curve using 3-Tesla scanner. 123I-MIBG washout rate (MIBG-WR) was calculated using a polarmap technique with tomographic data. We analyze the ability of M-T2* and MIBG-WR to predict MACE. MIBGWR and M-T2* were significantly greater in DCM patients with MACE than in patients without MACE. Receiveroperating-characteristics curve analysis showed that the optimal MIBG-WR and M-T2* thresholds of 35 % and 28.1 ms, and the two combination predict MACE with C-statics of 0.69, 0.73, and 0.82, respectively. Patients with MIBG-WR \35 % and M-T2* \28.1 ms had significantly M. Nagao (&) S. Kawanami Department of Molecular Imaging and Diagnosis, Graduate School of Medical Sciences, Kyushu University, 3-1-1 Maidashi, Higashi-Ku, Fukuoka City, Fukuoka 812-8582, Japan e-mail: [email protected] S. Baba M. Yonezawa Y. Yamasaki T. Kamitani T. Isoda Y. Maruoka Y. Kitamura H. Honda Department of Clinical Radiology, Graduate School of Medical Sciences, Kyushu University, Fukuoka City, Japan K. Abe T. Higo K. Sunagawa Department of Cardiovascular Medicine, Graduate School of Medical Sciences, Kyushu University, Fukuoka City, Japan
lower event-free rates than those with MIBG-WR C35 % or M-T2* C28.1 ms (log-rank value = 4.35, p \ 0.05). Cox hazard regression analysis showed that v2 and the hazard ratio were 3.99 and 2.15 for development of MACE in patients with MIBG-WR C35 % or M-T2* C28.1 ms (p \ 0.05). Iron deficiency, expressed by a high M-T2*, and MIBG-WR were both independent predictors of MACE in patients with DCM. The two combination was a more powerful predictor of MACE than either parameter alone. Keywords Dilated cardiomyopathy MIBG scintigraphy T2* MRI Major adverse cardiac event
Introduction Idiopathic dilated cardiomyopathy (DCM) is characterized by ventricular dilation and systolic dysfunction, and generally has a poor prognosis. The 1-year mortality rate of patients with idiopathic DCM can be as high as 25 %, whereas the 5-year mortality ranges from 20 to 50 % [1]. Earlier detection of this disease and the introduction of beta-blocker therapy may have led to an improved prognosis [2–4]. Activation of the sympathetic nervous system is one of the cardinal pathophysiological abnormalities associated with congestive heart failure [5]. Cardiac imaging with 123I-meta-iodobenzylguanidine (MIBG), an analogue of norepinephrine, is a useful tool for detecting abnormalities of the myocardial adrenergic nervous system in patients with congestive heart failure [6, 7], and may serve as a marker of responsiveness to drug interventions [8, 9]. Anemia and iron deficiency have recently been demonstrated to be prognostic markers and therapeutic targets
123
400
in patients with chronic heart failure [10–12]. The autonomic nervous system is stimulated globally in response to an anemia-induced decrease in oxygen delivery to peripheral tissues and is augmented more following heart failure. Reduced kidney function is responsible for impairment of hemopoiesis and renal blood flow, leading to activation of the renin-angiotensin system, which can in turn exacerbate heart failure. These observations strongly suggest a critical interaction among anemia, iron deficiency, and kidney dysfunction in heart failure patients, possibly magnifying the unfavorable effects of increased sympathetic activity on clinical outcomes [13]. Therefore, we hypothesized that myocardial iron deficiency and cardiac sympathetic impairment might be linked, and that the combination of these two factors may predict cardiac mortality in DCM. The aim of this study was to quantify myocardial iron content in patients with DCM utilizing T2* magnetic resonance imaging (MRI), and to investigate the relationship between myocardial iron deficiency, cardiac sympathetic activity as assessed by 123I-MIBG imaging, and major adverse cardiac events (MACE) in DCM.
Materials and methods Patients Forty-six patients with DCM and left ventricular ejection fraction (LVEF) \45 % on echocardiography were admitted to our institution with their first episode of congestive heart failure between June 2010 and February 2014. A detailed history and physical examination were obtained prior to enrollment in this prospective study. All patients underwent the following examinations: chest radiography; 12-lead electrocardiography (ECG); echocardiography; and standard blood tests, including measurements of hemoglobin, serum creatinine, and brain natriuretic peptide (BNP). In 36 out of 46 patients with DCM, myocardial biopsy was performed. The majority of section showed the myocardial tissue with chronic inflammatory infiltrate, interstitial fibrosis, nuclear pleomorphism of myocytes, and variation of myocyte size. These findings were compatible with those of DCM. There is neither evidence of granulomatous lesion, amyloidosis (dylon stain), nor hemochromatosis (iron stain). In addition, coronary angiography or stress perfusion imaging was performed when necessary. The patients were in New York Heart Association (NYHA) functional class II or III at the time of enrollment. Patients were excluded from the study for the following reasons: heart failure NYHA functional class IV (two patients), cardiac sarcoidosis (two patients), amyloidosis (one patient), and active myocarditis (one patient) diagnosed by left ventricular (LV) endomyocardial biopsy. Furthermore,
123
Int J Cardiovasc Imaging (2015) 31:399–407 Table 1 Characteristics of 46 patients with DCM Clinical characteristics Age (years)
44 ± 16
Sex (male/female)
32/14
Height (cm)
166 ± 8
Weight (kg)
64 ± 16
NYHA class II/III
12/34
Hypertension
6 (13)
Diabe mellitus
7 (15)
Hyperlipidemia
7 (15)
Systolic blood pressure (mmHg)
103 ± 18
Heart rate (bpm)
84 ± 25
Hemoglobin (g/dl) eGFR (ml/min/1.73 m2)
14.5 ± 1.8 75 ± 29
BNP (pg/mL)
583 ± 578
I-123 MIBG data Early HM ratio
2.2 ± 0.4
Delayed HM ratio
1.9 ± 0.4
Washout rate (%)
32 ± 10
Cardiac MRI data LVEDV (ml)
259 ± 110
LVESV (ml)
203 ± 102
LVEF (%)
25 ± 13
LGE positive
23 (50)
Myocardial T2* (ms)
26.9 ± 4.5
Parenthesis indicates percent
none of the patients had a history of alcohol abuse, congenital heart disease, or severe liver or kidney disease. After stabilization of their clinical condition, 46 patients who underwent both MRI and 123I-MIBG imaging were enrolled in the study. The study was approved by the Ethical Review Board of our institution, and written informed consent was obtained from each subject. Patient backgrounds are summarized in Table 1. Major adverse cardiac events during follow-up Patients were discharged from the hospital on optimal medical treatment and were examined by cardiologists in the outpatient clinic of our hospital at least every 3 months for a mean follow-up period of 27 months (6–50 months). Medical therapy for heart failure included a beta-blocker (n = 46), an angiotensin-converting enzyme inhibitors or angiotensin receptor blocker (n = 39), and a diuretic (n = 30). Physicians determined the necessity for blood tests, ECG, chest radiography, echocardiography and other examinations. The primary endpoint was major adverse cardiac events (MACE) consisting of cardiac death or heart failure hospitalization for cardiac resynchronization therapy (CRT) [14, 15] or a LV assist device (LVAD) [16, 17]
Int J Cardiovasc Imaging (2015) 31:399–407
401
Patients were injected intravenously with 111 MBq 123IMIBG (Fujifilm RI Pharma Co. Ltd., Tokyo, Japan) while in an upright position under resting fasting conditions after an overnight fast. Single-photon emission computed tomographic (SPECT) images were acquired 15 min after injection and repeated 4 h later. Anterior and lateral planar images were acquired immediately after each SPECT acquisition. SPECT imaging was performed with a dual-head rotating gamma camera (Infinia Hawkeye, GE Medical Systems, Waukesha, WI) equipped with a low-energy, general-purpose collimator. Images were acquired for 35 s each at 60 steps over a 360° orbit and were recorded at a digital resolution of 64 9 64 pixels. A 20 % energy window centered on 159 keV was used [18]. After reconstruction of transverse slices using a filtered backprojection method with a
Butterworth filter, a polar map was generated to calculate washout kinetics of myocardial 123I-MIBG activity by avoiding contamination from background activity. The washout rate of cardiac 123I-MIBG activity was calculated, using SPECT data and a polar map technique, as the percentage of change from early to late images with the following correction for physical decay: washout rate = [100 9 (early 123I-MIBG activity-late 123I-MIBG activity)/early 123I-MIBG activity] [18, 19]. Briefly, depending on patient heart size, 12–16 short-axis slices were selected to generate early and late polar maps. For each pixel in the polar maps, washout rate was calculated using early and late 123I-MIBG data to obtain the mean value that was used as a single washout rate for each patient (MIBG-WR) (Fig. 1). The heart/mediastinum count (H/M) ratio was determined from anterior planar early and delayed 123I-MIBG images, where H was the mean count/pixel in the left ventricle, and M was the mean count/pixel in the upper mediastinum (Fig. 1). The early and delayed H/M ratios were calculated. The MIBG-WR, early H/M ratio, and delayed H/M ratio were used as estimates of cardiac sympathetic activity.
Fig. 1 123I-MIBG images for a 25-year-old man who had cardiac death. a Early 123I-MIBG polar map shows that accumulation declines in the septa and inferior wall (left). Delayed 123I-MIBG polar map shows a diffuse accumulation decline (center) and the washout rate polar map demonstrates an increase in washout rate (right). His
MIBG-WR was 50 %. b Calculation of 123I-MIBG heart to mediastinum (H/M) ratio based on an anterior view of the thorax in the early scan (left) and delayed scan (right). ROIs have been drawn over the heart (H) and mediastinum (M). His early and delayed H/M ratios were 1.88 and 1.45, respectively
prior to heart transplantation because of pump failure that was refractory to medical treatment. MIBG imaging and quantification of cardiac MIBG activity
123
402
Int J Cardiovasc Imaging (2015) 31:399–407
Fig. 2 Myocardial T2* image for a 58-year-old man without MACE (left). One large ROI was placed in the mid-left ventricular septum, and the mean myocardial intensity for the ROI was measured by each T2* image with six different echo times (right)
Cardiac MRI All patients underwent 3-Tesla MRI using a scanner (Achieva 3.0T Quarsar Dual; Philips Healthcare, Best, The Netherlands) equipped with dual-source parallel radiofrequency transmission, 32-element cardiac phased-array coils for radiofrequency reception, and a four-lead vector cardiogram used for cardiac gating. Cine balanced turbo field-echo sequences in two-, three- and four-chamber views and a stack of short axis images acquired in parallel to the atrioventricular groove from the base to apex were performed with the following imaging parameters: repetition time, 2.8 ms; echo time, 1.4 ms; flip angle, 45°; slice thickness, 8 mm; field of view, 380 mm; matrix size, 176 9 193; SENSE factor, 2; and 20 cardiac phases/R–R interval on the ECG. A mid-ventricular short axis slice was obtained for T2* measurements. The black blood T2* acquisition used a multi-echo gradient-echo sequence performed during a breath hold (flip angle 30°; matrix size 200 9 123; sample bandwidth 2,275 Hz/pixel; slice thickness 8 mm; and field of view 320 mm). The short-axis images were acquired within a single breath-hold at 6 echo times from 2.9 to 10.3 ms at approximately 1.5-ms increments. A double inversion recovery (DIR) pulse was triggered off of the R-wave, and the inversion time extended into diastole, generating a similar set of six images at increasing echo times [20] (Fig. 2). Subsequently, 0.1 mmol/kg of gadolinium contrast (Magnevist: Bayer Healthcare, Osaka, Japan) was administered via the antecubital vein and flushed with 30 ml of isotonic saline. Images with late gadolinium enhancement (LGE) were obtained with an inversion-recovery threedimensional (3D) T1 turbo field-echo sequence (fast gradient-echo pulse sequence) performed 10 min after contrast injection and acquired in the short-axis images (repetition
123
time 3.9 ms; echo time 1.2 ms; flip angle 15°; field of view 400 mm; matrix size 304 9 179; slice thickness 4 mm; SENSE factor 2). The inversion time was adjusted to a completely null normal myocardium. Cardiac function, myocardial T2*, and LGE analysis Cine images were analyzed with the use of dedicated software (Extended Work Space, Philips Healthcare, Best, The Netherlands). Initially, short-axis images were previewed from the base to the apex in a cinematic mode, then endocardial and epicardial contours for end-diastole and end-systole were manually traced. Delineated contours were used for the quantification of LV volumes and LVEF. For T2* measurements, one large full-thickness region of interest (ROI) was placed in the mid-left ventricular septum, distant from the lungs and cardiac veins, which are otherwise known to cause susceptibility artifacts [21] (Fig. 2). In addition, absolute iron contents and T2* measured in the mid-septum are highly representative of their respective global myocardial value [22]. The signal intensity of these ROIs was plotted against the echo time used for each image. The resulting points form an assumed exponential decay curve, as the image signal decreases with increasing echo time. An exponential function was fitted to the data, using the following equation:
y ¼ KeTE=T2
where K represents a constant, TE represents the echo time, and y represents the image signal intensity (Fig. 3). In our laboratory, the myocardial T2* value (M-T2*) in 50 patients with no history of cardiac disease and LVEF [55 % (mean age 59 years) was 22 ± 3 ms. The short-axis LGE image was assessed using the American Heart Association/American College of Cardiology classification of 17 standardized LV short-axis
Int J Cardiovasc Imaging (2015) 31:399–407
segments. LGE was considered positive if the hyperenhanced area involved [50 % of the wall thickness. Patients with more than one segment with a hyperenhanced area were defined as positive for the presence of LGE; the presence of LGE was used a marker for myocardial fibrosis. Statistical analysis Continuous data are expressed as the mean ± standard deviation (SD). For comparison between DCM patients with and without MACE, the Mann–Whitney U test was used. The correlation between M-T2* and the other parameters was determined by Pearson’s correlation coefficient. Receiver-operating-characteristic (ROC) curve analysis was performed to determine the optimal cutoff of the parameters
403
for the prediction of MACE. After univariate logistic regression analysis, multivariate logistic regression analysis was performed using the variables that were significant in univariate analysis (p \ 0.05) to determine their relationship with MACE. The significance of the coefficients in the logistic regression model was determined by Chi square. Survival curves of patient subgroups were created by the Kaplan–Meier method to clarify the time-dependent, cumulative event-free rate and were compared using the logrank test. The Cox proportional hazards regression analysis was performed to evaluate for factors that were associated with the development of MACE. All statistical tests were two-sided. A p value of \0.05 was considered significant. All analyses were performed using the JMP statistical program package (version 9.0; JMP, Inc.).
Results
Fig. 3 Graph shows methods of calculating the myocardial T2* value. The abscissa represents six different echo times, and ordinate represents natural logarithm (LN) of the mean signal intensity for the ROI. The negative reciprocal of the slope of line from the linear regression equation corresponds to the T2* value. The 25-year-old man with MACE (filled circle) (y = -0.0367 x ? 6.897; r2 = 0.99; T2* = 27.2 ms) is the same subject as in Fig. 1
Fig. 4 Cardiac MRI for a 25-year-old man with cardiac death. The subject is the same as in Fig. 1. The four-chamber cine MR image shows a diffuse thinning of the myocardial wall and dilated left ventricular cavity and left atrium (left). Late gadolinium enhancement
During follow-up in the 46 patients with DCM, heart failure hospitalization was performed in six patients, an LVAD was utilized in three patients, and an intra-aortic balloon pumping was utilized in one patient due to refractory pump failure. Three patients died without LVAD utilization. Representative 123I-MIBG and MRI images are presented in Figs. 1 and 4. MIBG-WR was significantly greater in DCM patients with MACE (n = 13) than in patients without MACE (n = 33) (36.9 ± 6.9 vs. 30.1 ± 10.7 %, respectively, p = 0.04). M-T2* was significantly greater in DCM patients with MACE than in patients without MACE (30.0 ± 4.0 vs. 25.7 ± 4.1 ms, respectively, p = 0.01) (Fig. 5). There was no significant difference in the other
MRI shows a liner enhancement area (arrow) in the mid-ventricular septum (center). The right side shows a myocardial T2* image (echo time 2.9 ms) at the same site of late gadolinium enhancement MRI
123
404
Int J Cardiovasc Imaging (2015) 31:399–407
Fig. 5 Scatterplots of MIBGWR (left) and M-T2* (right) for patients with and without MACE. The MIBG-WR and M-T2* were significantly greater in patients with MACE than those without MACE (*; p \ 0.05)
Table 2 Comparison between patients with and without MACE
Table 3 Pearson correlation coefficient of M-T2*
Parameter
Parameter
Pearson r
p 0.83
With MACE n = 13
Without MACE n = 33
Age
-0.032
Age (years)
44 ± 20
44 ± 15
NS
Systolic blood pressure
-0.068
0.66
Systolic blood pressure (mmHg)
96 ± 16
106 ± 19
NS
Hemoglobin
-0.117
0.44
79 ± 16 14.2 ± 1.4
86 ± 28 14.6 ± 2.0
NS NS
eGFR
0.108
0.48
BNP
0.021
0.89
Heart rate (bpm) Hemoglobin (g/dl) eGFR (ml/min/1.73 m2) BNP (pg/mL)
p
79 ± 37
73 ± 25
NS
623 ± 488
566 ± 617
NS
I-123 MIBG data Early HM ratio
2.1 ± 0.4
2.3 ± 0.4
NS
Delayed HM ratio
1.8 ± 0.3
1.9 ± 0.4
NS
Washout rate (%)
37 ± 7
30 ± 11
\0.05
180 ± 73
143 ± 53
NS
23 ± 13
26 ± 12
NS
Cardiac MRI data LVEDVi (ml/m2) LVEF (%) No. of patient with LGE
7 (54)
Myocardial T2* (ms)
30.0 ± 4.0
16 (48) 25.7 ± 4.1
\0.05
I-123 MIBG data Early HM ratio
-0.22
0.14
Delayed HM ratio
-0.21
0.16
0.24
0.11
0.137 -0.042
0.37 0.78
Washout rate Cardiac MRI data LVEDVi LVEF
showed that v2 and the hazard ratio were 3.99 and 2.15 for development of MACE in patients with MIBG-WR C35 % or M-T2* C 28.1 ms (p \ 0.05) (Table 5).
Parenthesis indicates percent
Discussion parameters between the two groups (Table 2). Furthermore, there was no correlation between the T2* value and any other factor (Table 3). Multivariate logistic regression analysis showed that the v2 and the odds ratio to predict MACE were 2.27 and 1.06 for MIBG-WR (p = 0.132), and 7.31 and 1.27 for M-T2* (p = 0.007). According to univariate logistic regression analysis, no other factors were able to predict MACE (Table 4). ROC curve analysis revealed that the optimal MIBG-WR threshold was 35 % for predicting MACE, with a sensitivity of 69 % and a specificity of 67 %. The optimal M-T2* threshold was 28.1 ms for predicting MACE, with a sensitivity of 54 % and a specificity of 73 %. Furthermore, when the two parameters were combined to predict MACE, the sensitivity was 100 %, and the specificity was 64 % (Fig. 6). Patients with MIBG-WR \35 % and M-T2* \28.1 ms had significantly lower event-free rates than those with MIBGWR C35 % or M-T2* C28.1 ms (Log-rank value = 4.35, p \ 0.05) (Fig. 7). The cox hazard regression analysis
123
The present study quantified myocardial iron content using T2* MRI and cardiac sympathetic activity using MIBG imaging in patients with moderate to advanced DCM. M-T2* and MIBG-WR were significantly associated with the occurrence of MACE in patients treated with optimal medical therapy that included a beta-blocker. On the other hand, we did not find a correlation between M-T2* and any of the 123I-MIBG imaging findings. Further, M-T2* was not correlated with any other biomarker of heart failure (Table 2); therefore, M-T2* is an independent predictive marker of MACE. Finally, the combination of MIBG-WR and M-T2* could predict MACE with a C-statics of 0.82, which was superior to the predictive ability of either parameter alone, and improve risk stratification of DCM patients for MACE. Cardiac sympathetic innervation and norepinephrine kinetics at nerve endings in failing hearts can be quantitatively assessed using 123I-MIBG imaging [23]. Sustained
Int J Cardiovasc Imaging (2015) 31:399–407 Table 4 Relation to risk for MACE in 46 patients with DCM
Parameter
405
Univariate logistic analysis 2
v
Odds ratio
p
Age
0.001
1.001
0.97
Systolic blood pressure
2.93
0.97
0.11
Hemoglobin
0.43
0.88
0.51
eGFR
0.42
1.007
0.52
BNP
0.09
1.0002
0.76
Early HM ratio
2.33
0.255
0.14
Delayed HM ratio
1.81
0.278
0.19
Washout rate
4.64
1.08
0.04
LVEDVi LVEF
3.59 0.86
1.011 0.97
0.07 0.37
T2*
9.68
1.31
0.002
Multivariate logistic analysis v2
Odds ratio
95 % CI
p
2.27
1.06
0.98–1.16
0.132
7.31
1.27
1.06–1.62
0.007
I-123 MIBG data
Cardiac MRI data
Fig. 6 Predictive value of risk for MACE using receiver-operatingcharacteristics analysis. M-T2* (center) could predict MACE with slight higher C-statistics than MIBG-WR (left). The combination of
MIBG-WR and M-T2* (right) could predict MACE with a sensitivity of 100 % and a specificity of 64 %, which was superior to the predictive ability of either parameter alone
Table 5 Cox hazard regression analysis for development of MACE in 46 patients with DCM
Fig. 7 Event-free curves of two groups classified by cutoffs of MIBG-WR C35 % or \35 % and M-T2* C28.1 ms or \28.1 ms. Patients with MIBG-WR \35 % and M-T2* \28.1 ms (in red) had significantly lower event-free rates than those with MIBG-WR C35 % or M-T2* C28.1 ms (in blue)
v2
Hazard ratio
95 % CI
p
MIBG-washout ratio C35 versus 1.45 \35 %
1.61
0.73–3.42
0.23
T2* C28.1 versus \28.1 ms
1.07
1.54
0.66–3.33
0.3
Washout ratio C35 % or T2* C28.1 ms versus washout ratio \35 % and T2* \ 28.1 ms
3.99
2.15
1.01–4.67
0.04
Parameter
excess activation of cardiac sympathetic tone initially plays a compensatory role in patients with mild to moderate heart failure, but there is decreased the efficiency of reuptake,
123
406
turnover, and storage of norepinephrine at presynaptic endings in the myocardium as heart failure progresses [24]. This process leads to an increase in norepinephrine concentration at the sympathetic cleft and desensitization of beta-adrenoceptor, ultimately resulting in loss of norepinephrine content and impairment of sympathetic innervation at an advanced stage of heart failure. Increased norepinephrine spillover, decreased norepinephrine content in nerve endings, and sympathetic denervation in failing hearts can be revealed as an increase in 123I-MIBG washout rate from the heart. Several studies have demonstrated that abnormal 123I-MIBG kinetics are associated with lethal cardiac events in patients with moderate to severe heart failure [25–27]. MIBG-WR was superior to the early and delayed H/M ratios for predicting MACE. This is consistent with observations on 123I-MIBG kinetics in previous studies [25–27]. Particulate intracellular iron causes shortening of the MR relaxation parameter, T2*, on the basis of microscopic magnetic field inhomogeneity. In fact, T2* MRI is an easily quantifiable, clinically robust, and highly reproducible measurement technique [20, 28] that has been used for the assessment of myocardial iron overload in patients with thalassemia. However, the assessment of myocardial iron deficiency in patients with heart failure has not been examined. This is the first report to describe the clinical importance of high M-T2* in DCM. Iron plays a key intracellular role in many cellular processes. In cardiac myocytes, this includes an important role within the mitochondria and in myoglobin [22]. Dietary iron deficiency in rats results in DCM characterized by aberrant mitochondrial and irregular sarcomere organization and an increase in reactive nitrogen species and RhoA expression. Morphological examination showed that ventricular diameters and ventricular volume were significantly enhanced in iron-deficient rats [29]. In our laboratory, the mean M-T2* for 50 subjects with no history of cardiac disease and LVEF [55 % was 22 ms. This value was much lower than that for 46 patients with DCM (mean 26.9 ms). These observations support our hypothesis that myocardial iron deficiency (as expressed by a high M-T2*) is associated with the adverse DCM. Another interesting finding was that M-T2* was a good biomarker for predicting MACE, compared with other biomarkers (such as BNP, systolic blood pressure, hemoglobin, and eGFR). The present study included only a few patients with severe anemia or renal dysfunction. This may be the reason for the lack of an association between hemoglobin or eGFR and MACE. The relationship between M-T2* and serum iron parameters is basically analyzing iron metabolism; however, it is not possible to determine whether patients had absolute iron deficiency (reduced iron stores) or relative iron deficiency (decreased systemic iron availability despite overall normal total/body
123
Int J Cardiovasc Imaging (2015) 31:399–407
iron) from our data. In the present study, serum Fe and ferritin were measured in 16 (35 %) and 13 (28 %) out of 46 patients. Pearson correlation coefficient of T2* showed r = 0.12 for serum Fe and r = -0.35 for ferritin, respectively. T2* might be related to reduced iron stores, expressed as the decline of ferritin. We acknowledge that the present study is limited by a small number of patients at a single center that were followed-up for relatively short time period. Furthermore, M-T2* was measured using just one type of MRI system at a single center. The transferability of the T2* technique is important to increase the utility of M-T2* measurements and would help to assure that results from different MRI scanners at different centers would be comparable.
Conclusions Myocardial iron deficiency, expressed by a high M-T2*, and MIBG-WR were both independent predictors of MACE in patients with DCM. However, there was no correlation between M-T2* and sympathetic functional parameters measured by MIBG imaging. The combination of M-T2* and MIBG-WR was a more powerful predictor of MACE than either parameter alone. Acknowledgments This work was supported by the Japan Society for the Promotion of Science (JSPS) KAKENHI (25461831). Conflict of interest
None.
References 1. Sugrue DD, Rodeheffer RJ, Codd MB, Ballard DJ, Fuster V, Gersh BJ (1992) The clinical course of idiopathic dilated cardiomyopathy: a population-based study. Ann Intern Med 117:117–123 2. MERIT-HF Study Group (1999) Effect of metoprolol CR/XL in chronic heart failure: metoprolol CR/XL randomised intervention trial in congestive heart failure (MERIT-HF). Lancet 353:2001–2007 3. Packer M, Bristow MR, Cohn JN et al (1996) The effect of carvedilol on morbidity and mortality in patients with chronic heart failure. U.S. Carvedilol Heart Failure Study Group. N Engl J Med 334:1349–1355 4. Packer M, Coats AJ, Fowler MB et al (2001) Carvedilol prospective randomized cumulative survival study group. Effect of carvedilol on survival in severe chronic heart failure. N Engl J Med 344:1651–1658 5. Thomas JA, Marks BH (1978) Plasma norepinephrine in congestive heart failure. Am J Cardiol 41:233–243 6. Schofer J, Spielmann R, Schuchert A, Weber K, Schluter M (1998) Iodine-123 meta-iodobenzylguanidine scintigraphy: a noninvasive method to demonstrate myocardial adrenergic nervous system disintegrity in patients with idiopathic dilated cardiomyopathy. J Am Coll Cardiol 12:1252–1258 7. Merlet P, Valette H, Dubois-Rande JL et al (1992) Prognostic value of cardiac metaiodobenzylguanidine imaging in patients with heart failure. J Nucl Med 33:471–477
Int J Cardiovasc Imaging (2015) 31:399–407 8. Takeishi Y, Atsumi H, Fujiwara S, Takahashi K, Tomoike H (1997) ACE inhibition reduced cardiac iodine-123-MIBG release in heart failure. J Nucl Med 38:1085–1089 9. Suwa M, Otake Y, Moriguchi A et al (1997) Iodine-123 metaiodobenzylguanidine myocardial scintigraphy for prediction of response to beta-blocker therapy in patients with dilated cardiomyopathy. Am Heart J 133:353–358 10. Tang YD, Katz SD (2006) Anemia in chronic heart failure: prevalence, etiology, clinical correlates, and treatment options. Circulation 113:2454–2461 11. Groenveld HF, Januzzi JL, Damman K et al (2008) Anemia and mortality in heart failure patients: a systematic review and metaanalysis. J Am Coll Cardiol 52:818–827 12. Doi T, Nakata T, Hashimoto A et al (2012) Cardiac mortality assessment improved by evaluation of cardiac sympathetic nerve activity in combination with hemoglobin and kidney function in chronic heart failure patients. J Nucl Med 53:731–740 13. Cohn JN, Levine TB, Olivari MT et al (1984) Plasma norepinephrine as a guide to prognosis in patients with chronic congestive heart failure. N Engl J Med 311:819–823 14. Abraham WT, Fisher WG, Smith AL et al (2002) Cardiac resynchronization in chronic heart failure. N Engl J Med 346:1845–1853 15. Beshai JF, Grimm RA, Nagueh SF et al (2007) Cardiac-resynchronization therapy in heart failure with narrow QRS complexes. N Engl J Med 357:2461–2471 16. Slaughter MS, Rogers JG, Milano CA et al (2009) Advanced heart failure treated with continuous-flow left ventricular assist device. N Engl J Med 361:2241–2251 17. Goldstein D, Zucker M, Pagani FD, Frazier OH (2006) Rotary ventricular assist devices. In: Kirklin JK, Frazier OH (eds) ISHLT monograph series: mechanical circulatory support. Elsevier, New York, pp 77–104 18. Nakata T, Miyamoto K, Doi A et al (1998) Cardiac death prediction and impaired cardiac sympathetic innervation assessed by metaiodobenzylguanidine in patients with failing and non-failing hearts. J Nucl Cardiol 5:579–590 19. VanderVeen BJ, Younis I, DeRoos A, Stokkel M (2012) Assessment of global cardiac I-123 MIBG uptake and washout
407
20.
21.
22. 23.
24.
25.
26.
27.
28.
29.
using volumetric quantification of SPECT acquisitions. J Nucl Cardiol 19:752–760 Kirk P, He T, Anderson LJ, Roughton M et al (2010) International reproducibility of single breathhold T2* MR for cardiac and liver iron assessment among five thalassemia centers. J Magn Reson Imaging 32:315–319 Reeder SB, Faranesh AZ, Boxerman JL, McVeigh ER (1998) In vivo measurement of T2* and field inhomogeneity maps in the human heart at 1.5T. Magn Reson Med 39:988–998 Andrews NC (2008) Forging a field: the golden age of iron biology. Blood 112:219–230 Wieland DM, Wu JI, Brown LE, Mangner TJ, Swanson DP, Beierwaltes WH (1980) Radiolabeled adrenergic neuron-blocking agents: adrenomedullary imaging with 131I-metaiodobenzylguanidine. J Nucl Med 21:349–353 Eisenhofer G, Friberg P, Rundqvist B et al (1996) Cardiac sympathetic nerve function in congestive heart failure. Circulation 93:1667–1676 Nakata T, Wakabayashi T, Nagahara D (2005) Assessment of cardiac innervation in heart failure and lethal arrhythmias: therapeutic and prognostic implications. Curr Cardiol Rev 1:29–36 Nishisato K, Hashimoto A, Nakata T et al (2010) Impaired cardiac sympathetic innervation and myocardial perfusion are related to lethal arrhythmia: quantification of cardiac tracers in patients with ICDs. J Nucl Med 51:1241–1249 Jacobson AF, Senior R, Cerqueira MD et al (2010) Myocardial iodine-123 meta-iodobenzylguanidine imaging and cardiac events in chronic heart failure: results of the prospective ADMIRE-HF (AdreView myocardial imaging for risk evaluation in heart failure) study. J Am Coll Cardiol 55:2212–2221 Westwood MA, Anderson LJ, Firmin DN et al (2003) Interscanner reproducibility of cardiovascular magnetic resonance T2* measurements of tissue iron in thalassemia. J Magn Reson Imaging 18:616–620 Dong F, Zhang X, Culver B, Chew HG Jr, Kelley RO, Ren J (2005) Dietary iron deficiency induces ventricular dilation, mitochondrial ultrastructural aberrations and cytochrome c release: involvement of nitric oxide synthase and protein tyrosine nitration. Clin Sci (Lond) 109:277–286
123
