Low-dose dobutamine stress gated blood pool SPECT assessment of left ventricular contractile reserve in ischemic cardiomyopathy: a feasibility study

The purpose of the present study was to evaluate the feasibility of gated blood pool single-photon emission computed tomography (GBPS) with low-dose dobutamine (LDD) stress test, performed on a single-photon emission computed tomography (SPECT) camera equipped with cadmium-zinc-telluride (CZT) solid-state detectors, in assessing of left ventricle (LV) contractile reserve in patients with ischemic cardiomyopathy (ICM). A total of 52 patients (age 59 ± 7.2 years, 47 men and 5 women) with ICM and a control group of 10 patients without obstructive coronary artery lesion underwent GBPS and transthoracic echocardiography (TTE) at rest and during LDD stress test (5, 10, 15 µg/kg/min). The duration of each GBPS step was 5 min. Stress-induced changes in LV ejection fraction (ΔLVEF), peak ejection rate, LV volumes, and mechanical dyssynchrony (phase histogram standard deviation, phase histogram bandwidth and entropy) obtained with GBPS were estimated. All GBPS indices except end-diastolic volume showed significant dynamics during stress test in both groups. The majority of parameters in ICM patients showed significant changes at a dobutamine dose of 10 µg/kg/min as compared to the rest study. Seventeen percent of ICM patients, but none from the control group, showed a decrease in LVEF during stress, accompanied by a significant increase in entropy. The intra- and inter-observer reproducibility was excellent for both rest and stress studies. There was a moderate correlation (r = 0.5, p = 0.01) between GBPS and TTE, with a mean difference value of − 1.7 (95% confidence interval − 9.8; 6.4; p = 0.06) in ΔLVEF. Low-dose dobutamine stress GBPS performed with high-efficiency CZT-SPECT cameras can be performed for evaluating stress-induced changes in LV contractility and dyssynchrony with lower acquisition time. A dobutamine dose of 10 µg/kg/min can potentially suffice to detect stress-induced changes in patients with ICM during GBPS. ClinicalTrials.gov identifier: NCT04508608 (August 7, 2020).


Introduction
Equilibrium radionuclide angiography (ERNA) is a wellestablished imaging modality for the assessment of cardiac function due to its high accuracy and reproducibility [1]. However, data are lacking regarding the use of ERNA with inotropic pharmacological stress tests. Previously published papers have demonstrated the prognostic value of planar stress ERNA and have shown that it may be used as an alternative to exercise testing for risk stratification [2][3][4]. Salimian et al. have described stress gated blood pool single-photon emission computed tomography (GBPS) in an experimental study on canine models [5]. In this paper, the study duration was of approximately 140 min, which is inappropriate for clinical practice. Introduction of gamma cameras with solid-state semiconductor cadmium-zinc-telluride (CZT) detectors allows to significantly shorten the imaging time [6][7][8], an important feature which could be implemented in GBPS.
The purpose of the present study was to evaluate the feasibility of GBPS with low-dose dobutamine (LDD) stress test, performed on a single-photon emission computed tomography (SPECT) camera equipped with CZT solid-state detectors, in assessing the left ventricle (LV) contractile reserve in patients with ischemic cardiomyopathy (ICM). Although a number of studies have previously addressed the use of CZT cameras for myocardial blood flow measurements [9][10][11][12][13], to the best of our knowledge, this is the first study focused on the assessment of stressinduced changes of contractile function and dyssynchrony in patients with severe heart failure. A few publications have examined the use of CZT GBPS studies addressing the feasibility of low-dose protocols [7] and the validation of ventricular function measured by GBPS with ERNA [6]. Salimian et al. have attempted to detect stress-induced changes in mechanical dyssynchrony (MD) [5] in a canine model using a conventional gamma camera. Present study demonstrates for the first time that CZT GBPS can be used in clinical practice to detect stress-induced changes in contractility and MD. To the best of our knowledge, this is the first study specifically performing stress GBPS using a CZT camera in patients with severely reduced LV systolic function and ejection fraction (LVEF).

Patient and study population
Between 2018 and 2019, a total of 77 patients with ICM and a control group of 10 patients without obstructive coronary artery disease were enrolled in the study. Among 77 patients with ICM who underwent LDD GBPS, 7 had no LDD transthoracic echography (TTE) data and 5 had technical issues during the GBPS acquisition. Additional 13 patients had side effects during stress (6 had chest pain, 3 had an excessive hypertensive response, and 4 had heart rhythm disturbances resulting in interrupting the test and image acquisition). As a result, 52 patients with ICM (age 59 ± 7.2 years, 47 men and 5 women) represent the final study population. Among the control group, none had side effects during stress test, and therefore, all 10 patients were enrolled (Fig. 1). The clinical characteristics of the patients are presented in Table 1.
Inclusion criteria for ICM patients were reduced LV systolic function with an LVEF < 40% and an increase in LV volumes demonstrated by TTE, history of myocardial infarction or revascularization (coronary artery bypass grafting or percutaneous coronary intervention), ≥ 75% stenosis of left main or ≥ 75% stenosis in the proximal left anterior descending artery, and/or stenosis of ≥ 75% of two or more epicardial vessels based on invasive coronary angiography results [14]. Inclusion criteria for the control group were the absence of obstructive coronary artery lesion, no history of myocardial infarction or revascularization, and preserved LVEF. Exclusion criteria were patients who had contraindications to the LDD stress test such as acute coronary syndrome, severe aortic valve stenosis, hypertrophic cardiomyopathy, hemodynamic instability, inflammatory myocardial disease, atrial fibrillation, severe hematological or neurological disorders, and patients who could not sign a consent form. The study was approved by the Local Ethical Committee and conformed to the Declaration of Helsinki on Human Research. Written informed consent was obtained from each patient after explanation of the protocol, its aims, and potential risks. This research is part of SciCoRIC study (ClinicalTrials. gov identifier: NCT04508608).

Study design
According to the research protocol, all patients underwent LDD TTE followed by LDD GBPS within 1-3 days of each other.

Patient preparation
The preparation for both LDD TTE and LDD GBSP is similar and was performed according to the guidelines [10]. Prior to the test, an intravenous cannula was inserted in the right cubital vein of each patient.

Stress protocol
Dobutamine was infused into a cubital vein starting at a dose of 5 µg/kg/min to achieve a stable response and then increased to 10 and 15 µg/kg/min [15]. Each stage lasted for 5 min. A syringe pump Braun Perfusor Compact S (Braun, PA) was used. The 12-lead electrocardiogram (ECG) was monitored continuously during the entire examination. Blood pressure was measured at baseline and repeated at each stage of the dobutamine infusion.

Transthoracic echocardiography
TTE was performed in all patients using the ultrasound system Vivid E9 (GE Healthcare) with matrix multi-frequency transducer (1.5-4.6 MHz). 2D M-mode, Doppler echocardiography measurements and quantification were performed according to recommendations of the American Society of Echocardiography [16]. All TTE images were recorded and further analyzed by experienced echocardiographers in at least three cardiac cycles using a dedicated software (GE-Vingmed, Horten, Norway). Echocardiography parameters were recorded initially at rest, at 4-5 min of infusion of each dose of dobutamine, and during the recovery period. Common 2D echocardiography parameters including LV contractility, end-diastolic volume (EDV), and end-systolic volume (ESV), as well as assessment of valvular disease and pulmonary artery pressure, were obtained.

GBPS data acquisition and processing
All patients underwent rest and LDD stress GBPS (Discovery NM/CT 570c; GE Healthcare, Haifa, Israel) following in vivo labeling of red blood cells (RBC) with a 99m Tcpertechnetate dose of approximately 9 MBq/kg [1]. Patients were imaged in the supine position with the left arm placed over the head.
The data were acquired with electrocardiogram-gating (16 frames/cycle; acceptance window of ± 15%) for 5 min with a 20% energy window centered at 140 keV. No attenuation correction was used.
The entire examination consisted of three acquisition series. The first series was acquired at rest. The second and third series were acquired at 10 and 15 µg/kg/min of dobutamine, respectively. The acquisition was simultaneous with the dobutamine infusion through the entire duration of the stress.

GBPS data analysis
Raw data were analyzed visually for motion and attenuation artifacts. Quality control was performed according to the European Association of Nuclear Medicine guidelines  [1]. Count statistics were assessed in a region of interest of 100 mm 2 placed in the center of the LV. Images were reconstructed on a dedicated workstation (Xeleris 4.0110; GE Healthcare, Haifa, Israel) using maximum-penalizedlikelihood iterative reconstruction (60 iterations; Green OSL Alpha 0.7; Green OSL Beta 0.3). The Myovation for Alcyone software (GE Healthcare, Haifa, Israel) was used for image reconstruction. The Butterworth post-processing filter (frequency 0.52; order 5) was applied to the reconstructed slices obtained in a 70 × 70 pixels matrix with 57 slices. Reconstructed images were processed using the Quantitative Blood Pool SPECT 2009.0 (Cedars-Sinai Medical Center, Los Angeles, CA, USA) software. The countbased volumes method was used to calculate EDV, ESV, and LVEF [1]. Ventricular contours were adjusted manually when required. The following LV parameters were evaluated in further analysis: EDV (ml), ESV (ml), LVEF (%), and peak ejection rate (PER, expressed as EDV/s). The phase standard deviation (PSD) and histogram bandwidth (HBW, range in which 95% of myocardium initiated contraction) represent the global MD measures and are expressed in degrees. Entropy, which reflects the variability of the phase histogram, is measured in percentage [17]. Changes in GBPS measured LV parameters or in the difference (delta, Δ) between the rest and the different stress test stages were calculated as [the value at stress − rest value]. Differences between peak stress and rest LVEF values obtained by both GBPS and TTE were compared.

Reproducibility
To test the intra-and inter-observer reproducibility, 15 patients included in the ICM group were analyzed by two experienced nuclear medicine physicians blinded to each other's results.

Statistical analysis
Data were assessed for normal distribution using the Shapiro-Wilk test. Continuous variables were expressed as mean ± standard deviation (SD) or as median with interquartile range (IQR, Q25 to Q75). Differences between independent groups were determined by the Mann-Whitney U test and Wilcoxon signed-rank tests were performed for unpaired and paired data, respectively. Friedman ANOVA was used to reveal differences in GBPS parameters measured during the entire study, at rest and for 10 and 15 µg/kg/ min of dobutamine. When the difference was statistically significant, a post hoc paired comparison was applied using the Wilcoxon signed-rank test with Bonferroni correction for multiple comparisons. Each statistical test was 2-sided, and a p-value of < 0.05 was considered statistically significant. The Spearman test was used to estimate the correlation coefficient between quantitative variables. Bland-Altman analysis was applied to determine the magnitude and directions of the intermodal bias for LVEF, as well as intra-and inter-operator bias using limits of agreement (defined as 1.96 SD from the mean difference).

GBPS reproducibility
The intra-and inter-observer reproducibility was excellent for both rest and stress studies with ICC more than 0.9 for all estimated parameters ( Table 2).

LDD GBPS results
For each stress stage, stress-induced changes in LV volumes, LVEF, PER, and MD parameters are presented in Figs. 2 and 3.
A dobutamine dose of 10 µg/kg/min led to significant changes in all GBPS parameters except for EDV in both groups, and PSD and entropy in the control group in comparison to rest. A dobutamine dose of 15 µg/kg/min led to significant changes in LVEF, PER, and ESV in both groups and all MD parameters in the control group, but none in the Nine ICM patients (17%) showed a decrease in LVEF on the highest dobutamine dose accompanied by a significant increase in entropy, with no side effects, stress-induced ischemia, and/or arrhythmia documented on electrocardiogram (Fig. 4A). Those patients who demonstrated a decrease in LVEF at a dobutamine dose of 15 μg /kg/min showed an increase in entropy of 2.5% (0; 9) compared to the rest study, while in the remaining ICM patients, the stress test led to a decrease in entropy by − 3% (0; − 6). The comparison of stress-induced changes in GBPS parameters between the ICM and control groups is presented in Figs. 5 and 6. Only ΔLVEF at doses of both 10 and 15 µg/ kg/min dobutamine, as well as ΔPER and Δentropy at 15 µg/ kg/min, showed significant differences between the groups.

Comparison of LDD stress GBPS and TTE
ΔLVEF showed a fair correlation (r = 0.5, p = 0.01) between LDD GBPS and LDD TTE at peak stress. Forty-seven patients showed concordance of LVEF changes during stress between methods and 5 patients (10%) showed a drop in LVEF at peak stress with no concordance to GBPS data.

Discussion
To the best of our knowledge, this is the first study using CZT-SPECT that measures GBPS with LDD stress test in patients with ICM. The main results of present study indicate that LDD GBPS allows for the assessment of the LV contractile reserve and MD changes in patients with ICM and LV dysfunction.
All GBPS parameters showed excellent reproducibility both at rest and stress. In the current study, a dobutamine dose of 10 µg/kg/min led to significant changes in the majority of estimated parameters with the exception of EDV in both groups, and PSD and entropy in the control group. Furthermore, the increase in dobutamine dose did not cause any significant changes of GBPS parameters in the ICM group but led to significant changes of MD parameters in controls. Only in the ICM group 9 patients (17%) showed a drop in the LVEF at peak stress, accompanied by a simultaneous increase in entropy but not in PSD and HBW.
Previous studies have shown that dobutamine stress planar ERNA can be performed in patients with ICM [2,4]. The planar acquisition technique resulted in an underestimation of the LVEF, particularly in patients with LV systolic dysfunction (LVEF < 35%), and also lacked assessment of the LV dyssynchrony [8,19]. SPECT ERNA avoids overlapping activity in cardiac chambers. The complete removal of counts from the left atrium resulted in a 7-10% higher LVEF when using SPECT ERNA as compared to LVEF obtained from planar ERNA and thus increased the accuracy of LV function assessment [8,20].
Performing GBPS on conventional SPECT gamma cameras takes about 20-30 min (32 frames with 40-60 s) per each acquisition [1]. Salimian et al. have applied GBPS using a dual-head gamma camera on a canine model with increasing doses of dobutamine for an infusion duration of 143 min and a high administered dose of the radiopharmaceutical of approximately 1,223 MBq [5]. This approach cannot be applied in clinical practice. CZT cameras allow reducing the GBPS acquisition time to 8 min [21], 5 min [7], or even 4 min [6]. In the present study, we applied a 5-min acquisition protocol in combination with the LDD stress test.
Previous studies have demonstrated an excellent reproducibility of the LVEF and volumes obtained with GBPS at rest [22,23]. SPECT ERNA is a very good tool for the assessment of LV MD. Our findings of the stress-induced changes in dyssynchrony parameters are similar to those of Boogers et al. [24], who used a conventional gamma camera and showed good inter-observer reproducibility for both PSD and HBW in patients with ICM with an LVEF ≤ 35%, ICC of 0.88 and 0.93, respectively. The lower ICC values in comparison to our study may be explained by the presence of large perfusion defects (typical in ICM) resulting in incorrect delineation of the myocardium [25]. GBPS avoids this drawback, since it delineates the blood pool and not the myocardium, resulting in more accurate measurements of volumes and function [26]. Moreover, Jensen et al. [27] reported better reproducibility results for rest GBPS parameters (LVEF and volumes) obtained with a CZT device as compared with Anger cameras (both planar and SPECT). Our study demonstrates reproducible results for all parameters, including MD both at rest and stress, using a CZT camera.
GBPS is a widely used imaging technique for assessment of myocardial contractility. This approach has diagnostic and prognostic significance in patients with heart failure including those with arrhythmia (particularly in cardiac resynchronization therapy candidates) [28], pulmonary embolism [29], and cardiotoxicity [20,23,30]. This method was validated on a cardiac phantom [31] and showed good correlation with CMR [32] and TTE [28]. However, this approach is not widely used for assessment of the contractile reserve because of the relatively long study duration and high radiation exposure [8]. The CZT technology allows reducing the administered radiopharmaceutical dose to perform GBPS by up to 40% when the 5-min acquisition protocol is applied or even by up to 50% in case of a 10-min acquisition [8]. In the present study, the mean 99m Tc-RBC administered dose was of 740 ± 111 MBq, which is 40% lower than the dose recommended for stress studies [1], while at the same time, image quality is preserved.
Left ventricular cardiac reserve is an index of global inotropic function increasing during both physical exercise and pharmacological stress test (usually with dobutamine), of well-established prognostic value [4,33]. In patients with heart failure, this parameter is used to assess the likelihood for improvement of the LVEF following treatment, whether medical, surgery [34], or device implantation [4,33]. Contractile reserve has been also shown to correlate with the amount of viable myocardium post infarction [35].
The most accepted criterion of preserved contractile reserve is a 5% absolute increase in LVEF during stress or a relative increase of 20% from the baseline rest LVEF [31,33]. In the present study, the absolute increase in LVEF was of 4.5% (IQR 0; 9) in the ICM group and 14% (IQR 10;19) in the control group, on average.
Our study showed a fair correlation (r = 0.5, p = 0.01) between LDD GBPS and LDD TTE at peak stress. Although stress GBPS and TTE were comparable by Bland-Altman analysis, the quite wide SD of 1.96 for ∆LVEF does not allow an optimal comparison of these methods in practice. However, a trend in estimation of ΔLVEF between GBPS and TTE is clear. These data are partially consistent with those of Raja et al. [36], who showed a fair correlation (r = 0.52) between the values of rest LVEF obtained by TTE and planar ERNA with a mean difference of − 3.3 (1.96 SD − 18.6; 12.0), a trend towards a negative difference for low LVEF and a positive difference for high LVEF. However, stress-induced differences obtained by these two modalities have not been previously compared.

Clinical value of stress-induced changes in mechanical dyssynchrony
Numerous studies have shown that a high LV MD value at rest is an independent predictor of all-cause mortality in ICM [37][38][39]. Kano et al. [37] have reported a significant inverse correlation between LV entropy and expression of the SERCA2a mRNA level, which mediates the contraction of cardiomyocytes and the re-entry of Ca 2+ from the cytoplasm into the sarcoplasmic reticulum. Lower amounts of SERCA2a mRNA are usually noticed in the human failing heart and its reduction is associated with Ca 2+ handling impairment and contractile disorders [40].
Exercise or dobutamine stress tests can be used to unmask LV dyssynchrony, which can be absent at rest. Such an approach can help identify high-risk patients [41]. Some studies have reported that in patients undergoing cardiac resynchronization therapy, stress-induced changes in MD show a better correlation to clinical outcome and long-term survival than MD values at rest [42]. Currently, data regarding changes in MD during stress remain controversial. While some studies, using TTE, reported that dyssynchrony is not a dynamic state and does not appear during stress [43,44], other authors claim different patterns of stress-induced MD changes [45,46]. Such controversial results may be due to the high variability of TTE [5].
We observed a significant decrease in all MD parameters during the stress test in both groups of patients and controls, similar to some of previously published results [5,47,48]. In ICM patients, only 10 µg/kg/min of dobutamine caused significant changes in all MD parameters, whereas escalation of the dose did not lead to any significant changes in Orange bar -ICM group; green bar -control group. Δ -mean values of stress-induced changes between rest and stress acquisition; LV, left ventricle; EF, ejection fraction; EDV, end-diastolic volume; ESV, end-diastolic volume; PER, peak ejection rate; ml, milliliter; s, seconds GBPS parameters. On the contrary, in controls, the highest dobutamine dose caused a significant decrease of all MD indices (only HBW decreased significantly at 10 µg/ kg/min). The most significant changes were observed at a dobutamine dose of 10 µg/kg/min by entropy in the ICM group, and HBW in the control group. These findings are partly in accordance with previous data of Salimian et al. [47] who performed MPI-GSPECT on a canine model and showed a significant decrease of entropy at a dobutamine dose of 10 µg/kg/min, whereas PSD and HBW decreased significantly only at 20 µg/kg/min, as compared to rest. Legallois et al. [48] using MPI-GSPECT showed the presence of synchrony reserve in ICM patients at the dobutamine dose of 10 µg/kg/min. Interestingly, we observed a significant increase in entropy but not PSD and HBW in patients with ICM (n = 9) who showed a drop in LVEF at the dobutamine dose of 15 µg/kg/min. None of the control subjects showed such a response. This finding may be due to the fact that ischemia [49,50] can appear during the stress test because of increasing myocardial oxygen demand [51] and may represent an additional sign of the extent of disease, in line with the transient ischemic dilation index [52]. Short episodes of transient ischemia during a stress test may cause contractility dysfunction. This phenomenon called myocardial stunning is characterized by a disruption of the calcium homeostasis and oxygen-free radical injury [53]. The occurrence of this phenomenon is associated with a drop in post-stress LVEF [50,54] as well as an increase in LV MD [55,56]. However, stunning is a time-dependent phenomenon and may resolve by the time perfusion imaging is performed (at least 10-15 min after tracer injection) [57,58]. The prognostic value of entropy has been previously demonstrated [37,59]. It is well recognized that LV entropy may be influenced by LV volume representing the patient's hemodynamic status, LVEF, age, severity of heart failure, and statistical noise   [17]. On the other hand, entropy is less dependent on the shape of the phase-angle distribution histogram and can be a more sensitive parameter for the assessment of LV synchrony, especially in the presence of extended scar areas [17]. Most of the previous investigations were performed with myocardial perfusion imaging, where dyssynchrony parameters can significantly depend on the degree of radiotracer uptake [25,60]. Since GBPS-derived MD parameters are not affected by myocardium tracer uptake, they may be more robust in reflecting myocardial contractile dysfunction.

Limitations
Despite the relatively small number of subjects enrolled in the control group, it seems sufficient considering this to be a feasibility project. In addition, in the present study, we assessed LV contractile function based on the 5-min data acquisition. While there may be some averaging of the contractile function during this acquisition time, its effect, particularly at 10 and 15 μg/kg/min of dobutamine, can be considered as insignificant because of dobutamine saturation at the dose of 5 µg/kg/min. Regional parameters of cardiac contractility and MD were not assessed in this study.

Conclusion
Low-dose dobutamine stress GBPS performed with high-efficiency CZT-SPECT cameras can identify stressinduced changes of LV contractility and MD in patients with ischemic heart failure. Due to the high sensitivity of the new semiconductor detectors, there is an opportunity to perform GBPS with shorter acquisition time and lower radiation exposure to the patients. A dobutamine dose of 10 µg/kg/min was associated with significant changes in the majority of the estimated GPBS parameters, including all MD indices, in ICM patients. These results prove that this dose is sufficient to evaluate MD stress-induced changes. Further large-scale clinical studies are needed to evaluate the potential clinical significance of this approach in patients with severe heart failure.