Quantitative and semi-quantitative comparison of myocardial blood flow between ^99mTc-tetrofosmin and ^99mTc-sestamibi: based on continuous rapid dynamic imaging technology using cadmium-zinc-telluride cardiac single-photon emission computed tomography

Introduction

Single-photon emission computed tomography (SPECT) myocardial perfusion imaging (MPI) is a noninvasive imaging modality for the diagnosis and risk stratification of patients with suspected or known coronary artery disease (CAD) (1). However, conventional qualitative or semi-quantitative approaches may fail to detect or underestimate balanced multivessel coronary disease. Cardiac-dedicated cadmium-zinc-telluride SPECT (CZT-SPECT) systems enable dynamic tomographic acquisition, thereby allowing absolute quantification of myocardial blood flow (MBF) and assessment of myocardial flow reserve (MFR). This capability addresses key limitations of qualitative or semi-quantitative MPI and further improves the diagnostic performance for CAD (2).

To date, studies have largely focused on standard acquisition protocols and have reported good correlations between MBF and MFR derived from cardiac-dedicated CZT-SPECT and those measured by positron emission tomography (PET) (3-5). The main technetium-99m (^99mTc)-labeled radiotracers used for this technique are ^99mTc-sestamibi (^99mTc-MIBI; MIBI group) and ^99mTc-tetrofosmin (^99mTc-TF; TF group), both of which are widely used in routine MPI practice. However, comparative evidence regarding their performance in dynamic MBF quantification using cardiac-dedicated CZT-SPECT remains limited. Therefore, by retrospectively and consecutively collecting data from patients with no or mild CAD who underwent dynamic MBF quantification on a cardiac-dedicated CZT-SPECT system using either MIBI or TF, this study aimed to investigate and compare tracer-related differences in absolute MBF quantification and semi-quantitative MPI diagnostic indices. We present this article in accordance with the STROBE reporting checklist (available at https://qims.amegroups.com/article/view/10.21037/qims-2026-1-0321/rc).

Methods

Study population

A retrospective consecutive screening was performed for patients who underwent quantitative MBF measurement using cardiac-dedicated CZT-SPECT at TEDA International Cardiovascular Hospital between December 2023 and November 2024. The inclusion criteria were as follows: (I) age ≥18 years; (II) suspected CAD at initial diagnosis (defined as the presence of relevant symptoms without a prior history of myocardial infarction or acute coronary syndrome, and without previous coronary revascularization); (III) visually normal MPI findings; and (IV) availability of coronary anatomical data from computed tomography coronary angiography (CTCA) or invasive coronary angiography (ICA) within 1 week before or after imaging. The exclusion criteria were as follows: (I) presence of stenosis in ≥2 coronary branches of any degree or ≥50% stenosis in a single branch, as determined by CTCA or ICA; (II) failure of dynamic imaging injection curves to meet software-based quality control criteria; (III) MFR values <2.0 in any coronary territory or the global left ventricle (LV), as measured by CZT-SPECT; and/or (IV) coexistence of cardiomyopathy, valvular disease, myocarditis, or congenital coronary artery anomalies. Ultimately, 200 patients were included in the study (see Figure 1): 100 in the TF group (44 men and 56 women; age, 60.00±8.61 years) and 100 in the MIBI group (42 men and 58 women; age, 59.89±9.67 years). The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Institutional Ethics Committee of TEDA International Cardiovascular Hospital, which waived the requirement for individual informed consent due to the retrospective nature of the study.

Figure 1 Diagram of study patient enrollment. MIBI group: ^99mTc-MIBI; TF group: ^99mTc-TF. ^99mTc-MIBI, technetium-99m-sestamibi; ^99mTc-TF, technetium-99m-tetrofosmin; CAD, coronary artery disease; CTCA, computed tomography coronary angiography; ICA, invasive coronary angiography; MFR, myocardial flow reserve; MPI, myocardial perfusion imaging.

Imaging protocol

The MIBI kit (Jiangyuan Pharmaceutical Factory, Wuxi, China) was supplied by Atomic High-Tech Co., Ltd. or Beijing Xinkesida Pharmaceutical Technology Co., Ltd. It was labeled with generator-eluted ^99mTc solution, heated in a boiling water bath for 10 min, and allowed to stand for 10 min before use. The TF kit (GE Healthcare, Oslo, Norway) was supplied by Nanjing Andike. It was labeled with generator-eluted ^99mTc solution at room temperature and allowed to stand for 10 min before use. The radiochemical purity of both MIBI and TF was >95%.

All scans were performed on a cardiac-dedicated CZT-SPECT system (NM 530c, GE Healthcare, Haifa, Israel) equipped with 19 CZT detectors and pinhole collimators. Patients were instructed to strictly avoid coffee, tea, and any foods containing caffeine or theophylline for 24 h before imaging and to discontinue cardiovascular medications.

Dynamic MBF acquisition was performed using a continuous rapid acquisition protocol. After a pre-injection of 37 MBq of the tracer, patients were positioned supine for cardiac localization. Rest list-mode dynamic acquisition was initiated; 10 s later, the radiotracer (185–296 MBq; 4–5 mL) was injected over 5 s, and data acquisition continued for 10 min. Immediately after completion of the rest acquisition, adenosine was infused for 6 min at 0.14 mg/kg/min. At 2 min 50 s after the start of adenosine infusion, stress list-mode dynamic acquisition was initiated; 10 s later, a radiotracer dose 3.5-fold higher than the rest dose was injected (with the injection volume and flow rate kept constant) and acquisition continued for 10 min.

List-mode acquisition was continuously synchronized with an electrocardiographic gating system to enable simultaneous dynamic and gated imaging. The two list-mode datasets were reconstructed to generate rest and stress MPI datasets. The acquisition parameters were as follows: 8 frames per cardiac cycle, a ±15% heart rate acceptance window, a 140-keV energy peak, and a ±10% energy window. After SPECT acquisition, a low-dose cardiac computed tomography (CT) scan was performed on a PET/CT system (Discovery NM690, GE Healthcare, Waukesha, WI, USA) at 120 kV and 20 mA for attenuation correction of the SPECT dynamic data. All procedural steps were identical between the TF and MIBI groups.

Data processing and quality scoring

Quantitative perfusion parameter estimation

Dynamic list-mode data were transferred to the MyoFlowQ workstation (Full Version: V2.0.1.20230222, Beijing Bailingyun Biotechnology Co., Ltd., Beijing, China) for processing. Attenuation correction, scatter correction, and collimator resolution recovery were applied, and rest MBF (rMBF), stress MBF (sMBF), and MFR were obtained for the left anterior descending (LAD) artery, left circumflex (LCX) artery, right coronary artery (RCA), and global LV. Quantification was performed using a one-compartment kinetic model (6). Based on the time-activity curves (TACs) derived from the left ventricular cavity, atrial blood pool, and myocardial tissue regions, kinetic parameters (K₁ and k₂) were estimated as follows:

Cmyo(t)=FBV⋅Ca(t)+(1−FBV)⋅K1e(−k2t) ⊗Ca(t)[1]

In this model, K₁ (mL/min/g) and k₂ (min⁻¹) represent the tracer uptake rate from blood to myocardial tissue and the washout rate from tissue, respectively. C_myo(t) denotes the measured myocardial activity concentration derived from dynamic images and was assumed to consist of the arterial blood input C_a(t) and true myocardial uptake, with the latter represented as a convolution function of K₁, k₂, and Ca(t). FBV represents the fractional blood volume attributable to the arterial input C_a(t) within the myocardium; that is, the proportion of blood activity in the total myocardial radioactivity concentration [Cm(t)]. In contrast, (1 − FBV) represents the fraction attributable to true myocardial tracer uptake. K₁, k₂, and FBV were estimated using a curve-fitting procedure.

MBF (mL/min/g) was then calculated from K₁ with an additional correction for the myocardial tracer extraction fraction E using the following equation:

K1=MBF⋅E(MBF)=MBF⋅(1−α⋅e(−β/MBF))[2]

where α and β are physiological parameters derived by expressing E as a function of MBF based on the permeability-surface area product (PS, mL/min/g). In the present study, α=0.816 and β=0.267 (7). A representative analysis performed on the MyoFlowQ workstation is presented in Figure 2, illustrating manual delineation, quality assessment, generated TACs, tracer uptake polar maps, and the final automated quantification of regional and global MBF and MFR. The software performed an automated quality assessment (shown in the red box). A red light indicated a failed scan, typically due to multi-peaked curves, abnormal curves, insufficient counts, or obvious interference from extracardiac hot spots. A yellow light indicated fair but acceptable quality, usually caused by patient motion (which can be corrected through frame-by-frame manual motion correction). A green light indicated excellent quality.

Figure 2 Representative analysis of dynamic CZT-SPECT for MBF quantification. The content includes the manual delineation (bottom left), quality assessment (bottom right), generated TACs, tracer uptake polar maps (bottom center), and the final automated quantification of regional and global MBF and MFR (top). BP, blood pool; CZT-SPECT, cadmium-zinc-telluride single-photon emission computed tomography; LAD, left anterior descending; LCX, left circumflex; LV, left ventricle; MBF, myocardial blood flow; MFR, myocardial flow reserve; RCA, right coronary artery; TAC, time-activity curve.

Semi-quantitative perfusion indices and left ventricular function

Dynamic list-mode data were reconstructed into gated MPI datasets on the GE Xeleris 4DR workstation (GE Healthcare, Milwaukee, WI, USA). The rest MPI reconstruction window was 5–10 min post-injection, and the stress MPI reconstruction window was 7–10 min post-injection. The MPI images were then processed using the built-in quantitative perfusion SPECT/quantitative gated SPECT (QPS/QGS) software (Cedars-Sinai Medical Center, Los Angeles, CA, USA) to derive conventional semi-quantitative perfusion indices, including the summed stress score (SSS), summed rest score (SRS), summed difference score (SDS), stress perfusion defect extent (sExtent), rest perfusion defect extent (rExtent), stress total perfusion defect (sTPD), and rest total perfusion defect (rTPD), as well as gated functional parameters, including the stress left ventricular ejection fraction (sLVEF) and rest left ventricular ejection fraction (rLVEF).

Image quality scoring

MPI tomographic image quality was scored according to a previously published system (8): 0, no or mild extracardiac uptake without any myocardial overlap; 1, mild extracardiac uptake adjacent to the myocardium or minimal overlap that did not interfere with myocardial interpretation, or extracardiac uptake equal to or greater than myocardial uptake without overlap; 2, extracardiac uptake equal to or greater than myocardial uptake with minimal overlap that did not interfere with interpretation, or mild overlap while still allowing normal assessment of myocardial uptake; 3, extracardiac uptake higher than or markedly higher than myocardial uptake with partial or substantial overlap that interfered with myocardial interpretation and required delayed imaging. Higher scores indicated poorer image quality. Image quality was independently rated by two nuclear medicine physicians, each with over 3 years of experience, and the average score was calculated. Representative tomographic slices illustrating the 4-point image quality scoring system [0–3], based on the degree of extracardiac uptake and myocardial overlap, are presented in Figure 3.

Figure 3 Representative examples of the 4-point image quality scoring system. Tomographic image quality was graded from 0 to 3, with higher scores indicating poorer image quality due to adjacent or overlapping extracardiac uptake. 0: no or mild extracardiac uptake without any myocardial overlap. 1: mild extracardiac uptake adjacent to the myocardium or minimal overlap that did not interfere with interpretation. 2: extracardiac uptake with mild overlap that still allowed for normal assessment of myocardial uptake. 3: high extracardiac uptake with substantial overlap that interfered with myocardial interpretation, requiring delayed imaging.

Statistical analysis

Categorical variables were compared between groups using the Chi-squared test. Continuous variables are presented as mean ± standard deviation. Between-group comparisons were performed using the independent-samples t-test for normally distributed variables and the Mann-Whitney U test for non-normally distributed variables, whereas within-group comparisons were performed using the Wilcoxon signed-rank test. For quantitative parameters, within-group coefficients of variation (CVs) were also calculated. A multivariate linear regression model was established to calculate the regression coefficient (β) of the TF group relative to the MIBI group, with corresponding 95% confidence intervals (CIs). Reproducibility was assessed using the intraclass correlation coefficient (ICC). Statistical analyses were performed using SPSS 26.0, and a two-sided P<0.05 was considered statistically significant.

Results

Baseline characteristics are summarized in Table 1, including sex, age, body mass index (BMI), hypertension, hyperlipidemia, diabetes mellitus, arrhythmia, smoking, drinking, and family history of CAD. No significant differences in baseline characteristics were observed between the TF group and MIBI group.

Between-group comparison of MBF and MFR derived from CZT-SPECT

Differences in territory-specific and global MBF and MFR

The results are shown in Table 2. No significant between-group differences were observed in rMBF in the LCX artery, RCA, or global LV (P>0.05). In contrast, significant between-group differences were observed in rMBF in the LAD artery, as well as in sMBF and MFR across all coronary territories and globally (all P<0.05), with higher values consistently observed in the MIBI group than in the TF group.

CVs and linear regression coefficients for MBF and MFR

The results are presented in Table 3. Across all coronary territories and for global values, the within-group CVs of MBF and MFR were consistently lower in the TF group than in the corresponding MIBI group. Moreover, within each group, the CVs of rMBF were lower than those of the corresponding sMBF and MFR. All linear regression coefficients (β) were negative, indicating that the quantitative values were lower in the TF group than in the MIBI group. Consistent with the CV pattern, the β coefficients for rMBF were smaller in magnitude than those for the corresponding sMBF and MFR. Only the between-group difference in LCX artery rMBF was minimal and did not reach statistical significance (P>0.05), whereas the β coefficients for all other quantitative parameters were statistically significant (P<0.05).

Subgroup analysis as an objective preliminary assessment of case inclusion

Within both the TF and MIBI groups, patients were divided into two subgroups (group 1: the first 50 consecutive cases; group 2: the last 50 consecutive cases). Territory-specific and global MBF and MFR values were then compared between the two subgroups, as shown in Table 4. In the TF group, sMBF in the LAD artery differed significantly between subgroups (P<0.05). No other calculated parameters in either the TF group or the MIBI group showed significant between-subgroup differences.

Between-group comparison of semi-quantitative MPI perfusion indices and LVEF derived from CZT-SPECT

The results are shown in Table 5. No significant between-group differences were observed in the SSS, SRS, SDS, sExtent, rExtent, sTPD, rTPD, sLVEF, or rLVEF (all P>0.05).

Between-group and within-group comparisons of MPI image quality

The results are presented in Tables 6,7. Significant between-group differences were observed in image quality scores for both rest MPI and stress MPI (both P<0.05). The MIBI group exhibited higher rest MPI and stress MPI image quality scores than the TF group, indicating superior overall MPI image quality in the MIBI group. Within each group, rest MPI and stress MPI image quality scores also differed significantly (both P<0.05). In both groups, stress MPI image quality scores were higher than rest MPI image quality scores, suggesting that stress MPI images exhibited superior quality compared with rest MPI images, regardless of tracer type. The interobserver correlations were excellent for all image quality scores (ICC: 0.86–0.957; all P<0.05), indicating good interobserver reproducibility of image quality assessment.

Discussion

This study was based on a near-normal population (including patients without CAD and those with mild CAD) and used a continuous rapid dynamic imaging protocol with CZT-SPECT. We found that ^99mTc-TF and ^99mTc-MIBI yielded no meaningful differences in conventional semi-quantitative MPI indices or LV function, while TF provided superior MPI image quality. However, the two tracers still exhibited measurable differences in quantitative perfusion metrics, including rMBF, sMBF, and MFR.

Both TF and MIBI have been used for conventional SPECT MPI for decades, with well-established clinical utility. TF was approved and widely adopted in the United States as early as 1996, slightly later than MIBI [1990] (9). Although the biodistribution mechanisms of TF and MIBI are broadly similar, TF is characterized by faster clearance from the liver and blood. After intravenous administration, TF is rapidly taken up by the myocardium, reaching peak activity within 5 min; its first-pass myocardial extraction fraction is approximately 54%, lower than that of MIBI (65%) (10). TF also clears rapidly from the blood, liver, and lungs, falling below 5% of the injected dose by 10 min and below 4.5% by 60 min. Consequently, TF is less affected by hepatic and pulmonary activity and may be more suitable for early post-injection imaging and 1-day protocols. A previous study (11) reported that MPI image quality acquired 15 min after stress injection of TF was not significantly different from that acquired 40–60 min after stress injection of MIBI using conventional protocols.

At present, imaging protocols for MBF and MFR quantification using cardiac-dedicated CZT-SPECT are predominantly based on 1-day protocols, which typically include the following steps: (I) rest dynamic flow imaging; (II) rest gated MPI; (III) pharmacologic stress testing plus stress dynamic flow imaging; and (IV) stress gated MPI. This workflow requires multiple patient re-positionings on and off the imaging table and multiple image acquisitions (12,13). In a recent study (14), we demonstrated the feasibility of a continuous rapid dynamic acquisition protocol for completing this imaging series, substantially reducing the examination time required for absolute MBF quantification using cardiac-dedicated CZT-SPECT with MIBI. This protocol eliminates conventional gated acquisitions for rest and stress MPI; instead, gated rest and stress MPI datasets are reconstructed from list-mode dynamic acquisition data. As a result, the entire procedure is streamlined into a single continuous sequence—“rest dynamic flow imaging + pharmacologic stress testing + stress dynamic flow imaging”—reducing the total examination time to approximately one-sixth of that required for standard protocols.

The principal limitation of this protocol is the potential interference of extracardiac hot activity with the inferior wall of the LV. Because MPI tomographic images are reconstructed from list-mode dynamic data, they effectively correspond to relatively early post-injection MPI (rest: 5-min post-injection interval with 5 min of reconstructed data; stress: 7-min post-injection interval with 3 min of reconstructed data). If hepatic tracer clearance is slow, extracardiac hotspot activity may overlap with the inferior LV wall and generate artifacts (15). In this continuous dynamic protocol, stress images can be assessed using list-mode reconstructions to determine whether delayed gated imaging at 1 h is required to reduce such interference; however, rest images cannot be similarly improved. Therefore, this protocol imposes higher requirements on tracer clearance characteristics in the liver and blood. In our previous study, 21% of patients underwent delayed MPI to improve the reconstructed stress MPI image quality (14).

Given its rapid hepatobiliary clearance, TF may be better suited to this continuous rapid protocol. A previous study (16) reported good correlations between TF and MIBI for MBF and MFR quantification. However, whether evaluated under standalone or combined MPI protocols, most reported post-injection imaging intervals have ranged from 15 to 45 min, and the findings have been inconsistent (17-19). Only a few studies have used a 5-min interval, and those reported suboptimal results (20).

In the present study, no statistically significant differences were observed between TF and MIBI in the conventional semi-quantitative MPI indices or left ventricular ejection fraction (LVEF). These findings indicate that, under the continuous rapid protocol, the MPI images and LVEF derived from the TF group were comparable to those obtained from the MIBI group. Notably, image quality scores for both rest MPI and stress MPI were lower in the TF group than in the MIBI group. The number of cases requiring delayed imaging due to extracardiac hotspot interference was two in the TF group (both improved after delay) and four in the MIBI group (three improved after delay, and one showed no obvious improvement). These findings suggest that TF yields better image quality in the continuous rapid protocol.

In addition, stress MPI scores were higher than rest MPI scores in both groups, indicating superior rest image quality with either tracer and suggesting that stress images may be more susceptible to extracardiac hotspot interference. This feature is advantageous for the continuous rapid protocol, as it provides an opportunity to improve poor-quality stress images via delayed imaging. Figure 4 presents a representative clinical case in which extracardiac hotspot uptake during the initial stress imaging almost obscured more than two myocardial segments. Following delayed imaging, the extracardiac hotspot was successfully cleared, resulting in a corresponding improvement in image quality.

Figure 4 Improvement of stress MPI image quality following a delayed scan. Representative stress tomographic images from a 55-year-old female patient in the ^99mTc-TF group. The initial reconstructed stress images (bottom row) demonstrate significant extracardiac tracer uptake overlapping with the inferior wall of the LV, which substantially interferes with myocardial interpretation. Following a 30-minute delay, the repeat scan (top row) reveals marked clearance of the extracardiac hotspot, resulting in unhindered visualization of the inferior myocardial wall. ^99mTc-TF, technetium-99m-tetrofosmin; LV, left ventricle; MPI, myocardial perfusion imaging.

Multiple studies (21-23) have demonstrated that integrating quantitative MBF and MFR measurements with conventional tomographic MPI enhances the diagnostic performance and accuracy for CAD, particularly in patients with multivessel coronary disease and microvascular dysfunction. Historically, such quantification has largely relied on PET/CT, limiting its broader clinical adoption (24,25). The advent of cardiac-dedicated CZT-SPECT has overcome key constraints related to PET/CT equipment availability and tracer logistics, enabling MBF and MFR quantification in a manner analogous to PET/CT while markedly reducing technical barriers and costs. This has positioned CZT-SPECT as one of the most actively investigated single-photon approaches for absolute myocardial flow quantification. Previous studies have also reported good correlations between MIBI or TF and nitrogen-13 ammonia (¹³N-NH3) for myocardial flow quantification (4,7).

Based on the literature review, this appears to be the first study to compare quantitative flow differences between MIBI and TF using cardiac-dedicated CZT-SPECT in comparable populations with the same methodology. In the present study, the baseline characteristics were comparable between the TF and MIBI groups, and no obstructive coronary stenosis was observed, indicating that the two cohorts were broadly comparable. Nevertheless, quantitative flow analyses consistently showed higher MBF and MFR values in the MIBI group than in the TF group. This difference was first apparent in the direct comparison of group means. In the multivariate linear regression analysis, quantitative parameters (rMBF, sMBF, and MFR) were treated as dependent variables, tracer type (TF vs. MIBI) as the independent variable, and age, sex, BMI, and cardiovascular risk factors as covariates. The regression coefficient (β) of the TF group relative to the MIBI group was calculated. Negative β values indicated lower values in the TF group compared with the MIBI group, whereas positive β values indicated higher values. In the present results, all β values were negative, confirming that at the statistical model level, rMBF, sMBF, and MFR across all coronary territories and the global LV were lower in the TF group than in the corresponding MIBI group. With the exception of the LCX artery rMBF parameter for which the β value was very small and not statistically significant, the differences for all other parameters were statistically significant, with β values for sMBF and MFR clearly larger than those for rMBF.

We also found that the within-group CVs for all parameters were lower in the TF group than in the MIBI group. Within each tracer group, the CVs of rMBF were lower than those of sMBF and MFR across all coronary territories and for global values. These findings suggest that, in populations with similar characteristics, TF-derived quantitative parameters are more stable than MIBI-derived parameters, and that rMBF is more stable than sMBF and MFR regardless of the tracer. Each group was divided into the first and last 50 consecutive cases and quantitative parameters were compared between these subgroups. In the TF group, a statistically significant difference in LAD artery sMBF was observed; however, this may have been related to individual differences. Some patients had mild (<50%) stenosis in a single coronary artery, mainly in the LAD artery (which is related to the pathological characteristics of CAD). The effects of this mild stenosis may not be consistent across the corresponding cases in each subgroup. However, as no significant differences were observed in the comparisons between the other subgroups, these findings support the overall comparability of the included cases.

Among previous studies, the findings reported by Wieting et al. (26) were generally consistent with our findings. In their subgroup analysis of normal subjects, sMBF was lower in the TF group than in the MIBI group; however, the rMBF values were generally higher than those observed in our study. This discrepancy may be attributed to differences in imaging protocols (e.g., residual effects of pharmacologic stress in stress-first protocols), the use of attenuation correction, and potential variations in post-processing software.

Therefore, in addition to uncertainties related to sample selection, the observed differences in this study may reflect intrinsic differences in myocardial extraction fractions between MIBI and TF. A previous study (27) has shown that due to their nonlinear extraction characteristics, neither TF nor MIBI reflect MBF as directly as ¹⁵O-H₂O, requiring extraction fraction correction. The MyoFlowQ software is based on a one-tissue compartment model that is theoretically closer to PET-based kinetic modeling than the net-retention model, and may provide more accurate MBF estimation; however, it relies heavily on accurate uptake values (K₁) (28). In the present study, identical physiological parameters (α and β) were applied in the Renkin-Crone model for both tracers. However, the first-pass myocardial extraction fraction of MIBI was higher than that of TF, resulting in systematically higher MBF values in the MIBI group, particularly for sMBF. Consequently, the ratio of sMBF to rMBF (i.e., MFR) was also higher in the MIBI group. These findings suggest that the accurate determination of α and β values is critical.

Shrestha et al. (29) reported TF-specific parameters of α=0.91±0.11 and β=0.34±0.20. However, subsequent studies have continued to use α and β values derived from MIBI (4). Most of these studies were based on the Corridor 4DM (INVIA) software, using systems including both NM530c and D-SPECT. We compared Renkin-Crone models with different parameter sets and plotted the inverse transformation curves (see Figure 5). When α=0.91 and β=0.34 were applied for TF, the resulting quantitative values were indeed higher than those obtained in the present study; however, in many cases, they exceeded those of the MIBI group. These findings suggest that α and β values cannot be directly transferred across different software platforms and imaging systems. As noted in previous studies (27), although the Renkin-Crone model is physiologically based, its parameters are empirical and highly dependent on both imaging systems and post-processing algorithms. At present, standardized acquisition protocols and reconstruction/post-processing methods for quantitative MBF measurement using cardiac-dedicated CZT-SPECT with ^99mTc-MIBI or ^99mTc-TF remain lacking (28). These findings highlight the need for standardized acquisition and processing protocols, followed by head-to-head comparative studies between TF-based SPECT and PET to refine tracer-specific physiological parameters.

Figure 5 Renkin-Crone inverse transform curve. The α and β values were set to 0.816 and 0.267 for the blue curve, and 0.91 and 0.34 for the red curve. The curves were generated based on numerical simulation. MBF, myocardial blood flow.

The primary limitation of this study is the lack of a head-to-head comparison of the two tracers within the same patients. Although no significant differences were observed in the baseline characteristics between the two groups, the potential influence of inter-individual variability cannot be completely excluded, and thus the observed differences cannot be attributed solely to tracer-specific properties.

Second, the study population did not comprise a completely normal cohort. Although this reflects real-world clinical practice, it inevitably introduces additional confounding factors. We applied an exclusion criterion of MFR <2.0 to minimize the potential influence of microvascular disease and inadequate adenosine response. Although previous studies have suggested that this cutoff shows good agreement with PET as the reference standard (30), it may have also introduced a degree of selection bias.

Finally, the lack of head-to-head PET quantitative data as a reference standard represents an important limitation. While there is supporting evidence for ^99mTc-MIBI from a previous study (7), comparable studies evaluating ^99mTc-TF against PET using the same methodology are lacking. Therefore, an optimal quantitative model for TF has not yet been established. Even if all other confounding factors were excluded and the observed differences were entirely attributable to tracer characteristics, it would still be unclear which tracer provides more accurate results. This underscores the importance of direct comparative studies between TF-based SPECT and PET quantification, and represents a key limitation of the present study.

Conclusions

Under a continuous rapid dynamic acquisition protocol using cardiac-dedicated CZT-SPECT, TF provides conventional semi-quantitative perfusion indices and LV functional parameters comparable to those obtained with MIBI, while yielding superior MPI tomographic image quality, supporting its practical clinical value. Nevertheless, TF and MIBI still differ in quantitative flow measurements (including rMBF, sMBF, and MFR), underscoring the need for head-to-head comparative studies using PET-derived MBF as the reference standard. Such studies are essential to refine TF-specific quantitative parameters and to establish corresponding tracer-specific reference ranges for MBF.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://qims.amegroups.com/article/view/10.21037/qims-2026-1-0321/rc

Data Sharing Statement: Available at https://qims.amegroups.com/article/view/10.21037/qims-2026-1-0321/dss

Funding: This work was supported by the Tianjin Key Medical Discipline Construction Project (No. TJYXZDXK-3-035C) and the Science and Technology Project of the Health Commission of Tianjin Binhai New Area (No. 2023BWKZ004).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://qims.amegroups.com/article/view/10.21037/qims-2026-1-0321/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. This study was approved by the Institutional Ethics Committee of TEDA International Cardiovascular Hospital, and the requirement for individual informed consent was waived due to the retrospective nature of the study.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Assante R, Zampella E, Cantoni V, Green R, D'Antonio A, Mannarino T, Gaudieri V, Nappi C, Buongiorno P, Panico M, Petretta M, Cuocolo A, Acampa W. Prognostic value of myocardial perfusion imaging by cadmium zinc telluride single-photon emission computed tomography in patients with suspected or known coronary artery disease: a systematic review and meta-analysis. Eur J Nucl Med Mol Imaging 2023;50:3647-58. [Crossref] [PubMed]
2. Cichocki P, Plachcinska A, Blaszczyk M, Adamczewski Z. Higher processing repeatability of myocardial flow reserve calculated using net retention model compared to one compartment model in SPECT studies. Sci Rep 2024;14:16671. [Crossref] [PubMed]
3. Ko KY, Ko CL, Lee CM, Cheng JS, Wu YW, Hsu RB, Chen YS, Wang SS, Yen RF, Cheng MF. Myocardial Flow Assessment After Heart Transplantation Using Dynamic Cadmium-Zinc-Telluride Single-Photon Emission Computed Tomography With (201)Tl and (99m)Tc Tracers and Validated by (13)N-NH(3) Positron Emission Tomography. Circ Cardiovasc Imaging 2023;16:e015034. [Crossref] [PubMed]
4. Giubbini R, Bertoli M, Durmo R, Bonacina M, Peli A, Faggiano I, Albano D, Milan E, Stern E, Paghera B, Rodella C, Cerudelli E, Gazzilli M, Dondi F, Bertagna F, Camoni L. Comparison between N(13)NH(3)-PET and (99m)Tc-Tetrofosmin-CZT SPECT in the evaluation of absolute myocardial blood flow and flow reserve. J Nucl Cardiol 2021;28:1906-18. [Crossref] [PubMed]
5. de Souza ACDAH, Harms HJ, Martell L, Bibbo C, Harrington M, Sullivan K, Hainer J, Dorbala S, Blankstein R, Taqueti VR, Foley Kijewski M, Park MA, Meretta A, Breault C, Roth N, Poitrasson-Rivière A, Soman P, Gullberg GT, Di Carli MF. Accuracy and Reproducibility of Myocardial Blood Flow Quantification by Single Photon Emission Computed Tomography Imaging in Patients With Known or Suspected Coronary Artery Disease. Circ Cardiovasc Imaging 2022;15:e013987. [Crossref] [PubMed]
6. Hsu B, Chen FC, Wu TC, Huang WS, Hou PN, Chen CC, Hung GU. Quantitation of myocardial blood flow and myocardial flow reserve with 99mTc-sestamibi dynamic SPECT/CT to enhance detection of coronary artery disease. Eur J Nucl Med Mol Imaging 2014;41:2294-306. [Crossref] [PubMed]
7. Hsu B, Hu LH, Yang BH, Chen LC, Chen YK, Ting CH, Hung GU, Huang WS, Wu TC. SPECT myocardial blood flow quantitation toward clinical use: a comparative study with (13)N-Ammonia PET myocardial blood flow quantitation. Eur J Nucl Med Mol Imaging 2017;44:117-28. [Crossref] [PubMed]
8. Meyer C, Weinmann P. Validation of early image acquisitions following Tc-99 m sestamibi injection using a semiconductors camera of cadmium-zinc-telluride. J Nucl Cardiol 2017;24:1149-56. [Crossref] [PubMed]
9. Einstein AJ, Pascual TN, Mercuri M, Karthikeyan G, Vitola JV, Mahmarian JJ, et al. Current worldwide nuclear cardiology practices and radiation exposure: results from the 65 country IAEA Nuclear Cardiology Protocols Cross-Sectional Study (INCAPS). Eur Heart J 2015;36:1689-96. [Crossref] [PubMed]
10. Boschi A, Uccelli L, Marvelli L, Cittanti C, Giganti M, Martini P. Technetium-99m Radiopharmaceuticals for Ideal Myocardial Perfusion Imaging: Lost and Found Opportunities. Molecules 2022;27:1188. [Crossref] [PubMed]
11. Duvall WL, Case J, Lundbye J, Cerqueira M. Efficiency of tetrofosmin versus sestamibi achieved through shorter injection-to-imaging times: A systematic review of the literature. J Nucl Cardiol 2021;28:1381-94. [Crossref] [PubMed]
12. Yamamoto A, Nagao M, Ando K, Nakao R, Matsuo Y, Sakai A, Momose M, Kaneko K, Hagiwara N, Sakai S. First Validation of Myocardial Flow Reserve Derived from Dynamic (99m)Tc-Sestamibi CZT-SPECT Camera Compared with (13)N-Ammonia PET. Int Heart J 2022;63:202-9. [Crossref] [PubMed]
13. Do J, Ruddy TD, Wells RG. Reduced acquisition times for measurement of myocardial blood flow with (99m)Tc-tetrofosmin and solid-state detector SPECT. J Nucl Cardiol 2021;28:2518-29. [Crossref] [PubMed]
14. Wang J, Chen Y, Chu H, Pang Z, Hsu B, Li J. Feasibility of myocardial blood flow quantification to detect flow-limited coronary artery disease with a one-day rest/stress continuous rapid imaging protocol on cardiac-dedicated cadmium zinc telluride single photon emission computed tomography. J Nucl Cardiol 2024;34:101825. [Crossref] [PubMed]
15. Johansen A, Lomsky M, Gerke O, Edenbrandt L, Johansson L, Hansen G, Jensen B, Reid MS, Johansson LL, Olofsson C, Minarik D, Nyström K, Wollmer P, Trägårdh E. When is reacquisition necessary due to high extra-cardiac uptake in myocardial perfusion scintigraphy? EJNMMI Res 2013;3:20. [Crossref] [PubMed]
16. Timmins R, Klein R, Petryk J, Marvin B, Wei L, deKemp RA, Ruddy TD, Wells RG. Reduced dose measurement of absolute myocardial blood flow using dynamic SPECT imaging in a porcine model. Med Phys 2015;42:5075-83. [Crossref] [PubMed]
17. Hurwitz GA, McLaughlin DS, Slomka PJ. Comparing the image quality of myocardial perfusion agents in the clinical laboratory: small test groups and large reference populations. Nucl Med Commun 2000;21:907-15. [Crossref] [PubMed]
18. Dizdarevic S, Singh N, Nair S, de Belder A, Ryan N, Aplin M, Vara A, James R, McCready VR, Peters AM. Early gated SPECT adenosine myocardial perfusion imaging may influence the therapeutic management of patients with coronary artery disease. Nucl Med Commun 2015;36:386-91. [Crossref] [PubMed]
19. Turgut B, Unlu M, Cengel A. Dobutamine stress tetrofosmin SPECT; evaluation of short rest-stress protocol and head to head comparison with MIBI in detection of coronary artery disease. Ann Nucl Med 2005;19:115-22. [Crossref] [PubMed]
20. Münch G, Neverve J, Matsunari I, Schröter G, Schwaiger M. Myocardial technetium-99m-tetrofosmin and technetium-99m-sestamibi kinetics in normal subjects and patients with coronary artery disease. J Nucl Med 1997;38:428-32.
21. Di Carli MF. Clinical Value of Positron Emission Tomography Myocardial Perfusion Imaging and Blood Flow Quantification. Heart Fail Clin 2025;21:395-405. [Crossref] [PubMed]
22. Valenta I, Upadhyaya A, Jain S, Schindler TH. PET/CT Assessment of Flow-Mediated Epicardial Vasodilation in Obesity and Severe Obesity. JACC Adv 2024;3:100936. [Crossref] [PubMed]
23. Poitrasson-Rivière A, Moody JB, Renaud JM, Hagio T, Arida-Moody L, Buckley CJ, Al-Mallah MH, Nallamothu BK, Weinberg RL, Ficaro EP, Murthy VL. Integrated myocardial flow reserve (iMFR) assessment: optimized PET blood flow quantification for diagnosis of coronary artery disease. Eur J Nucl Med Mol Imaging 2023;51:136-46. [Crossref] [PubMed]
24. Bailly M, Thibault F, Courtehoux M, Metrard G, Angoulvant D, Ribeiro MJ. Myocardial Flow Reserve Measurement During CZT-SPECT Perfusion Imaging for Coronary Artery Disease Screening: Correlation With Clinical Findings and Invasive Coronary Angiography-The CFR-OR Study. Front Med (Lausanne) 2021;8:691893. [Crossref] [PubMed]
25. Schindler TH, Fearon WF, Pelletier-Galarneau M, Ambrosio G, Sechtem U, Ruddy TD, et al. Myocardial Perfusion PET for the Detection and Reporting of Coronary Microvascular Dysfunction: A JACC: Cardiovascular Imaging Expert Panel Statement. JACC Cardiovasc Imaging 2023;16:536-48. [Crossref] [PubMed]
26. Wieting W, Bengel FM, Diekmann J. Comparison of global and regional myocardial blood flow quantification using dynamic solid-state detector SPECT and Tc-99 m-sestamibi or Tc-99 m-tetrofosmin in a routine clinical setting. Int J Cardiovasc Imaging 2025;41:537-48. [Crossref] [PubMed]
27. Zavadovsky KV, Mochula AV, Maltseva AN, Shipulin VV, Sazonova SI, Gulya MO, Liga R, Gimelli A. The current status of CZT SPECT myocardial blood flow and reserve assessment: Tips and tricks. J Nucl Cardiol 2022;29:3137-51. [Crossref] [PubMed]
28. D'Antonio A, Assante R, Zampella E, Mannarino T, Buongiorno P, Cuocolo A, Acampa W. Myocardial blood flow evaluation with dynamic cadmium-zinc-telluride single-photon emission computed tomography: Bright and dark sides. Diagn Interv Imaging 2023;104:323-9. [Crossref] [PubMed]
29. Shrestha U, Sciammarella M, Alhassen F, Yeghiazarians Y, Ellin J, Verdin E, Boyle A, Seo Y, Botvinick EH, Gullberg GT. Measurement of absolute myocardial blood flow in humans using dynamic cardiac SPECT and (99m)Tc-tetrofosmin: Method and validation. J Nucl Cardiol 2017;24:268-77. [Crossref] [PubMed]
30. Agostini D, Roule V, Nganoa C, Roth N, Baavour R, Parienti JJ, Beygui F, Manrique A. First validation of myocardial flow reserve assessed by dynamic (99m)Tc-sestamibi CZT-SPECT camera: head to head comparison with (15)O-water PET and fractional flow reserve in patients with suspected coronary artery disease. The WATERDAY study. Eur J Nucl Med Mol Imaging 2018;45:1079-90.