Native T1-rho mapping for myocardial scar detection in patients with probable or confirmed coronary artery disease

Introduction

The current gold standard for detecting myocardial fibrosis or scar tissue is late gadolinium enhancement (LGE) imaging, which exploits differences in gadolinium distribution between injured and normal myocardium to generate hyperintense contrast of the focal fibrosis and infarction in cardiac magnetic resonance imaging (MRI) (1). However, the reliance on gadolinium-based contrast agents (GBCAs) poses several limitations, including contraindications in patients with severe renal impairment, concerns about gadolinium deposition, rare but significant allergic reactions, workflow complexity, and added cost (2). These factors have driven the pursuit of non-contrast, quantitative imaging biomarkers capable of detecting fibrotic and ischemic myocardial changes with comparable sensitivity and specificity. Besides, there is growing concern that GBCAs are being released into wastewater, surface water and even drinking water systems, due to their persistence and the fact that conventional wastewater treatment often does not remove them (3).

T1-rho (T1ρ, spin-lattice relaxation in the rotating frame) mapping is an advanced parametric mapping technique that utilizes endogenous tissue contrast does not require exogenous contrast agents (4). T1ρ relaxation time reflects magnetization decay under a spin-lock pulse, enabling sensitivity to slow molecular motion and proton exchange between water and myocardial macromolecules, including collagen, proteoglycans, and glycosaminoglycans, as well as pH-dependent changes during ischemia. In the heart, this contrast arises from dipolar interactions and chemical exchange effects linked to extracellular matrix remodeling. Quantitative T1ρ maps, obtained from varying spin-lock durations (4), offer enhanced sensitivity over native T1 mapping for detecting diffuse fibrosis, microscopic structural changes, and biochemical alterations (5), supporting earlier, non-contrast detection of myocardial injury and ischemia across ischemic, scarred, and otherwise normal myocardial tissue. Previous clinical studies have reported significantly elevated myocardial T1ρ relaxation times in infarcted regions compared to remote myocardium in patients with acute or chronic myocardial infarction (6-8). These findings, though with a limited number of patients, support the hypothesis that T1ρ mapping captures biochemical and microstructural changes in fibrotic or infarcted tissue without contrast agents. Beyond static evaluation, T1ρ reactivity—defined as changes in relaxation time under physiological stress or pharmacologic vasodilation—may offer functional insights into myocardial tissue viability and ischemia (9,10).

The primary objective of this study was to assess the potential of T1ρ mapping in qualitatively and quantitatively identifying myocardial scars in patients with known or suspected coronary artery disease (CAD), compared to the standard LGE method. The secondary objective was to evaluate T1ρ reactivity among three defined groups: ischemic, scarred, and normal cases—to determine its potential as a non-contrast biomarker for ischemic heart disease. We present this article in accordance with the STARD reporting checklist (available at https://qims.amegroups.com/article/view/10.21037/qims-2025-1749/rc).

Methods

Patient population

We conducted a prospective study involving consecutive adult patients with known or suspected CAD who underwent cardiac magnetic resonance (CMR) imaging at Siriraj Hospital between October 2022 and March 2023. Exclusion criteria were: existing heart failure, unable to complete the CMR protocol, inadequate image quality, history of or CMR evidence of hypertrophic cardiomyopathy, myocarditis, non-ischemic cardiomyopathy, or infiltrative diseases, an estimated glomerular filtration rate (eGFR) below 30 mL/min/1.73 m², current airway diseases, atrial fibrillation, and claustrophobia or presence of ferromagnetic materials. This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study protocol was approved by the Siriraj Institutional Review Board [certificate of approval (COA) number Si. 446/2021]. All enrolled patients provided written informed consent. The sample size of 140 is needed to compare the difference of the scarred myocardium with the remote myocardium and non-scarred myocardium based on the data from a previous study (6).

CMR protocol

CMR imaging was conducted utilizing a commercial Ingenia 3.0T MR system (Philips Medical Systems, Best, the Netherlands). Our routine CMR protocol comprised black blood axial plane images; steady-state free precession (SSFP) cine images of standard long-axis, 3-chamber, 4-chamber, and short axis views; and native T1 mapping using modified look-locker inversion sequence.

T1ρ mapping was performed using the adiabatic HS8 pulse, which provides spin-locking robustness against B0 and B1 non-uniformity (11). Spin-lock times (SLTs) of 0, 20, 30, and 40 ms were selected to enable T1ρ mapping with mono-exponential fitting of the relaxation curve. A range of such short to intermediate SLTs was used to capture both faster relaxation components and slower molecular motions linked to fibrosis and ischemia, balancing sensitivity to myocardial biochemical alterations and maintaining adequate signal-to-noise ratio. This selection aligned with recommended cardiac T1ρ protocols reported in prior studies (6-8,12,13). The inclusion of 0 ms provided a baseline signal reference (4). All spin-locks T1ρ images were acquired in 3 breath-holds with a fixed scheme, while the selected 4 SLTs at 0, 20, 30, and 40 ms were from 2 breath-holds, respectively. The details of the pulse sequence diagram are illustrated in Figure 1. Five recovery heartbeats was chosen between each SLT mainly due to pragmatic reason to enhance robustness: it minimizes T1 contamination and acquisition-order effects so each SLT begins from comparable longitudinal magnetization, improving T1ρ fitting and cutoff reproducibility; inserts idle beats to satisfy specific absorption rate (SAR)/duty-cycle limits with high spin-lock amplitudes and longer SLTs, particularly relevant at high field; accommodates heart-rate variability to ensure adequate recovery across diverse RR intervals and prolonged T1 values; and reduces residual magnetization-transfer saturation from prior spin-lock pulses (13,14).

Figure 1 Schematic of cardiac T₁ρ MRI acquisition and processing. (A) ECG-triggered T₁ρ-prepared gradient-echo imaging sequence. Following the ECG trigger, an SL preparation module with a spin-lock pulse of duration SLT (T₁ρ spin-lock time) is applied before image acquisition. The sequence includes a delay, SL preparation (90°-SL-90° pulses), and gradient-echo readouts during the acquisition window. (B) Representation of T₁ρ acquisition during a single breath-hold. Multiple T₁ρ-weighted images are obtained at different SLT values (e.g., 0, 20, 40 ms) within one breath-hold, synchronized to the cardiac cycle. The lower plot shows the recovery of longitudinal magnetization (Mₓ) between heartbeats. (C) Overview of full cardiac T₁ρ mapping protocol. Images with different SLT values are acquired over multiple breath-holds (e.g., 1st: 0–40 ms, 2nd: 0–50 ms, 3rd: 0–60 ms). Post-processing involves voxel-wise fitting of signal decay as a function of SLT to generate quantitative T₁ρ relaxation maps, displayed as grayscale and color-coded maps. AU, arbitrary units; ECG, electrocardiogram; MRI, magnetic resonance imaging; SL, spin-lock; SLT, spin-lock time.

Rest T1ρ images were captured in one short-axis at the mid-level after native T1 imaging. The peak amplitude of the spin-lock (SL) pulse was capped at 500 Hz, ensuring adherence to specific absorption rate guidelines. After intravenous adenosine administration for 2 minutes, stress T1ρ images were acquired in one slice during stress before the first pass perfusion at the mid-level using the same SL pulse sequence. Due to the time limitation that we can only acquire 1 short-axis slice of rest and stress T1 rho images. One slice in the mid-level of the short axis has been used and reported from the previous studies (7,8).

Other imaging parameters were as follows: echo time/repetition time, 1.06/2.2 ms; resolution, 1.3 mm × 1.3 mm; slice thickness, 10 mm; field of view, 265 mm × 298 mm; flip angle, 10 degrees; number of signal averages, 1; shot interval, 5 heartbeats; and SENSE acceleration factor, 2.2.

Ten minutes after the total dose of 0.15 mmol/kg GBCA injection, a T1-weighted three-dimensional inversion recovery fast low-angle shot (3D IR-TFE) sequence was obtained in standard long and short axis views. The imaging parameters for this sequence were as follows: echo time/repetition time, 1.2/3.6 ms; resolution, 1.6 mm × 1.65 mm; slice thickness, 8 mm; field of view, 270 mm × 320 mm; flip angle, 15 degrees; sensitivity encoding (SENSE) acceleration factor, 2.5; and slice thickness, 8 mm. This sequence was used to assess LGE. The summary of image acquisition protocol in this study is illustrated in Figure S1.

Post-processing and image analysis

T1ρ maps were generated using in-house software (Interactive Data Language, IDL 6.3; ITT, Boulder, CO, USA) by pixelwise fitting the signal intensity to a monoexponential decay model, M(SLT) = M0·exp(−SLT/T1ρ), where M0 and M(SLT) denote the equilibrium magnetization and T1ρ spin lock prepared magnetization for each SLT, respectively. As five recovery heartbeats were employed to re-equilibrate longitudinal magnetization across SLTs, preparation-history and magnetization transfer effects were minimized and baseline inconsistencies were reduced. Under these conditions and with a limited number of SLTs, the two-parameter model should yield more stable and reproducible fits.

The region-of-interest (ROI) of the mid left ventricular myocardium slice was drawn with adequate margins intended to separate myocardium from the areas that are prone to partial volume averaging, such as the area between the endocardium and blood and the area between the epicardium and the outside tissue. The papillary muscles were excluded as part of the left ventricular myocardium. The image of the mid left ventricular myocardium was equally divided into 6 smaller ROIs to represent anterior wall, anteroseptal wall, inferoseptal wall, inferior wall, inferolateral wall, and anterolateral wall. Two imaging cardiologists interpreted the LGE images and T1ρ mappings, measured the T1ρ value, scored T1ρ images for the overall quality and the ability to detect scars. Image quality and confidence in scar detection were gauged on a 5-tier scale: 1, nondiagnostic with significant artifact; 2, minimally diagnostic with strong artifact; 3, diagnostic with moderate artifact; 4, high diagnostic confidence with minimal artifact; and 5, excellent diagnostic confidence with no artifact (Figure S2). The T1ρ images were qualitatively assessed for the presence or absence of myocardial scarring based on T1ρ hyperintensity by visual assessment. A hypointense area on T1ρ images was also noted.

Two investigating reviewers (N.P. and P.T.) independently interpreted T1ρ images, blinded to the LGE results. The reviewers first completed the analysis of T1ρ images for all 250 cases. Subsequently, they analyzed the combined T1ρ and SSFP cine images for the same cohort. To minimize second-look bias, T1ρ images and the corresponding T1ρ + SSFP cine datasets from a given patient were not evaluated within the same reading session. Specifically, T1ρ values were measured at the septum and for each of the 6 myocardial segments, following the standard segmentation model (15). The reviewers interpreting T1ρ images were also asked to assess abnormalities in wall thickness and motion to determine whether each segment had a scar, before evaluating the T1ρ images. We assessed two approaches for T1ρ interpretation of the myocardial scar: Approach one used T1ρ images without knowing other imaging findings; Approach two used T1rho images in combination with SSFP cine images (wall thickness and wall motion). Reviewer 1 (N.P.) also graded the overall image quality. Sensitivity, specificity, and accuracy was calculated for the diagnostic yield of SSFP alone, T1ρ alone, and SSFP plus T1ρ for the detection of myocardial scar.

Detection of myocardial scar by LGE was based on a hyperintense area by visual assessment and agreement between the two imaging cardiologists, blinded to T1ρ images. The presence of myocardial ischemia was defined as at least 1 myocardial segment had perfusion defect on visual assessment and agreement between the two imaging cardiologists blinded to T1ρ images. Additionally, T1ρ values were extracted from both rest and stress T1ρ mappings. This allowed for quantification of average T1ρ values for all myocardial segments. T1ρ reactivity, which represented the percentage shift between rest and stress average T1ρ values, was compared across three categories: ischemic cases, scarred cases, and normal cases.

Sensitivity analysis was performed for the assessment of quantitative analysis of T1ρ for the detection of myocardial scar. Quantitative analysis of T1ρ values was performed on the scarred area and on the remote area. The drawing of the scarred area on T1ρ images was performed with the guide of the LGE images. For the remote myocardium, the region was drawn in the septal area if the scarred areas were outside the septum.

Statistical analysis

Continuous variables are presented as mean ± standard deviation, while discrete variables are represented as counts and percentages. Independent samples t-test was used to compare continuous variables, and Pearson’s chi-square test was employed for categorical data comparison. The interrater reliability between the two imaging cardiologists was determined using the kappa statistic. ANOVA was utilized to compare T1ρ reactivity among the study groups. Receiver operating characteristic graph was used to determine the sensitivity and specificity of quantitative analysis of T1ρ for the detection of myocardial scar. Any P value under 0.05 was deemed indicative of statistical significance. All data were analyzed using PASW Statistics, version 18 (SPSS Inc., Chicago, IL, USA).

Results

Figure 2 shows a flow diagram of the study population. After enrolling 386 patients, the final analyzed sample comprised 250 patients. The mean age of the patients was 67.2±11.9 years, 44.4% male. Among the 250 patients, 75 (30%) had scars by LGE. Stress-induced perfusion defect indicating ischemia was detected in 47 patients, with 17 having ischemia without any scarring and 30 having ischemia in at least one segment beyond the scar. The baseline characteristics and CMR parameters are presented in Table 1. The mean left ventricular ejection fraction was 69.5%±13.1%.

Figure 2 Flow diagram of the study design. CMR, cardiac magnetic resonance; LGE, late gadolinium enhancement; NICM, non-ischemic cardiomyopathy.

Qualitative assessment of the image quality

Images rated as 1 and 2 were deemed uninterpretable. Of the remaining images, 68 (27%) were rated as 3, 131 (52%) as 4, and 51 (21%) as 5. The primary reason for reduced image quality in the T1ρ mappings was misregistration of the different T1ρ-weighted images, resulting from inconsistent breath holding during scans and arrhythmia.

Comparison of T1ρ and LGE images in the detection of myocardial scarring

As two approaches were used to assess myocardial scar detection in comparison to standard LGE, with only T1ρ images (Approach 1) and with T1ρ plus SSFP cine images (Approach 2), respectively, corresponding results of the diagnostic accuracy are summarized in Table 2.

For Approach 1 using only T1ρ images, the sensitivity of scar detection was 39% for both Reviewer 1 and Reviewer 2. However, the specificity rates varied among different reviewers, with Reviewer 1 achieving 88% and Reviewer 2 achieving 95%. For Approach 2 with combined T1ρ and SSFP cine images, scar detection was guided by wall thickness and regional wall motion abnormalities from the SSFP images in addition to the T1ρ images. This approach increased the sensitivity to 71% for Reviewer 1 and 73% for Reviewer 2. Furthermore, the second approach yielded high specificity rates, with Reviewer 1 at 99% and Reviewer 2 at 98%.

The sensitivity of using SSFP cine alone for the detection of myocardial scar was 46%, in comparison to SSFP + T1ρ 71%. The same trend was found for accuracy, as 83% for SSFP cine only and 90% for SSFP + T1ρ), which demonstrated superior diagnostic benefit of combined SSFP + T1ρ information (Table 3).

An interrater reliability analysis was conducted to assess the reliability of scar detection from T1ρ images of all 250 cases. The results are presented in Table S1.

Quantification of myocardial scars

A significant distinction in the T1ρ value was found between the scarred region (average T1ρ_scarred =78.2±7.4 ms) and the unaffected myocardium (average T1ρ_remote =44.3±5.3 ms, P<0.001). For the group with no scars identified by LGE, T1ρ value was found to be markedly lower (average T1ρ_non-scarred =42.1±4.0 ms, P<0.001). A significant difference was also evident when comparing the T1ρ value of the unaffected myocardium in the positive scar group (T1ρ_remote) against the entirety of the myocardium in the negative scar group (T1ρ_non-scarred, P=0.002). The hyposignal value measured on T1ρ in the hypointense area was 34.6±4.6 ms (n=29). A ROI was drawn to include the low-value area (i.e., the hyposignal area in T1ρ images) and the scarred area (T1ρ_scarred), referred to as T1ρ_combined. Significant differences were observed between T1ρ_scarred, T1ρ_remote, and T1ρ_combined. The T1ρ relaxation time was 42.1±4.0 ms in the T1ρ_non-scarred (n=175), 78.2±7.4 ms in T1ρ_scarred (n=75), 44.3±5.3 ms in T1ρ_remote (n=75), and 55.1±7.6 ms in T1ρ_combined (n=29). These differences between each region (T1ρ_non-scarred, T1ρ_scarred, T1ρ_remote, and hyposignal T1ρ) are depicted in Figure 3. The quantitative measurement of mean T1ρ hypo showed 34.6±4.6 ms, which was significantly lower than T1ρ remote (44.3±5.3 ms) (Figure 3). Figure 4 provides examples of LGE and T1ρ images for negative scar and positive scar cases identified by LGE. The ROIs of different locations of T1ρ measurement are shown in Figure S3.

Figure 3 T1ρ relaxation time comparison across different myocardial regions.

Figure 4 Representative short-axis T1ρ maps corresponding to late gadolinium enhancement images. (A) Non-scarred patient; (B) subendocardial scar of the inferolateral wall; (C) transmural scar of the inferolateral wall. Arrows indicate scarred areas. LGE, late gadolinium enhancement.

Analysis of T1ρ-reactivity

The evaluation of T1ρ reactivity involved comparing three distinct groups: ischemic cases (n=47), scarred cases (n=45), and normal cases (n=158). The recorded T1ρ-reactivity values were −0.3%±7.1% for ischemic cases, 1.5%±6.1% for scarred cases, and 3.4%±7.6% for normal cases. The P values obtained when comparing these groups were 0.007 for ischemic versus normal group, 0.332 for scarred versus normal, and 0.745 for ischemic versus scarred group. Rest-stress images of T1 Rho are shown in Figure S4.

Sensitivity analysis

Sensitivity analysis was performed for the assessment of quantitative analysis of T1ρ for the detection of myocardial scar. Receiver operating characteristic (ROC) graph shows that sensitivity and specificity of quantitative analysis of rest T1ρ images for the detection of myocardial scar were 81.2% and 71.6% with the area under the curve of 0.832 (0.812–0.850, P<0.001; Figure 5 and Table S2). The ROC graph of T1ρ stress for the detection of myocardial scar showed a sensitivity, specificity, and area under the curve of 65.5%, 72.4%, and 0.702 (0.678–0.725) (Figure S5A). We also analyzed the value of native T1 mapping for the detection of myocardial scar. The ROC graph of native T1 for the detection of myocardial scar showed a sensitivity, specificity, and area under the curve of 67.2%, 89.4%, and 0.818 (0.782–0.849; Figure S5B).

Figure 5 Receiver operating characteristic graph shows sensitivity and specificity of quantitative analysis of rest T1ρ images for the detection of myocardial scar. AUC, area under the curve; CI, confidence interval.

Detection of hyposignal intensity of the T1ρ (T1ρ hypo)

In our study, approximately 39% in the positive scar group showed low signal intensity of T1ρ, i.e., a lower T1ρ value (T1ρ hypo). On the other hand, no T1ρ hypo was found in the negative scar group. So apart from the T1ρ scarred, T1ρ non-scarred and T1ρ remote, we measure the T1ρ hypo by manually drawing an ROI at the low signal intensity area (similar to drawing an ROI of the T1ρ scarred). Figure S6 shows images of a patients who had a hyposignal area from T1ρ and had microvascular obstruction (MVO) on the LGE image.

Discussion

This study represents the most extensive investigation of T1ρ imaging in patients with known or suspected CAD. The mean T1ρ value was significantly higher in the scarred area than that in the remote region as well as in the septum of the negative scar group. In terms of diagnostic accuracy, the sensitivity of scar detection was 39%, and the specificity was 83% based on T1ρ mapping alone. Combining T1ρwith short-axis SSFP images significantly improved diagnostic accuracy, raising sensitivity to 71% and specificity to 99%.

The findings from our study supported the use of increased T1ρ relaxation time for myocardial scar detection, in line with the previous study (6). Myocardial scar tissue exhibits elevated T1ρ due to increased collagen, proteoglycans, and altered water-macromolecule exchange, producing slower spin-lock relaxation via dipolar and chemical exchange mechanisms compared to normal myocardium (12,14,16,17). We also showed that T1ρ in the remote myocardium of the patients with scar is slightly higher than those without scar. This finding might be related to a pathology at other segments of the patients with a disease even in the remote myocardium. Previous publications have shown that native T1 or T1ρ can increase in remote segments compared to healthy controls (6,18). The sensitivity of scar detection by T1ρ was only 39% for both reviewers. When T1ρ mapping was used in combination with short-axis SSFP, the sensitivity surged to 71% and 73% for Reviewer 1 and Reviewer 2, respectively, with high specificities of 99% and 98%. The interrater reliability between Reviewer 1 and Reviewer 2 showed excellent agreement for both methods (kappa =0.82 and 0.83, respectively, P<0.001). Furthermore, there was excellent agreement for the intraobserver reliability of Methods 1 and 2 (kappa =0.83 and 0.86, respectively, P<0.001).

This study also represents the first investigation into the usage of T1ρ-reactivity. The T1ρ-reactivity values in our study were 0.3%±7.1% for ischemic cases, 1.5%±6.1% for scarred cases, and 3.4%±7.6% for normal cases (P<0.05 only for ischemic vs. normal). The pronounced variance between ischemic and normal groups may aid in distinguishing between these conditions. Our previous study has shown that T1 reactivity can be used to differentiate myocardial ischemia from normal myocardium (10). However, the difference in the magnitude of changes of T1ρ-reactivity (3.4% in myocardial ischemia and 1.5% in normal myocardium) was less obvious from T1-reactivity (4.8% in myocardial ischemia and 1.4% in normal myocardium). T1 Rho might have potential in the differentiation of myocardial segments with ischemia from normal myocardial segments as we have shown in our previous work on the use of T1 mapping in the differentiation of ischemia and scarred myocardium (10). More studies are needed to validate this finding.

ROC graph of quantitative analysis of rest T1ρ images shows that sensitivity and specificity for the detection of myocardial scar were 81.2% and 71.6% with the area under the curve of 0.832 (0.812–0.850). The area under the curve was similar to the T1 mapping for the detection of myocardial scar and better than the use of T1ρ, which is agreed with the result of the previous report (10).

The detection of low signal intensity in the T1ρ relaxation time (hyposignal T1ρ) was observed in approximately one-third of the positive scar patients in our study. The hyposignal T1ρ might be an early indicator of MVO or artifactual related to motion. A literature review (19) has shown that MVO can be detected using LGE images. When we compared hyposignal T1ρ with MVO, we found a sensitivity of 71%, a specificity of 75% (P=0.001), and an accuracy of 72% for MVO detection. Future studies are needed to confirm the benefit of native T1ρ mapping for the detection of MVO. Despite MVO commonly occurs early after acute myocardial infarction, a previous publication has shown that MVO can persist at 198 days after MI for 14.4% (20). In our study, among 70 patients with a history of myocardial infarction, only 1 patient that performed CMR after 7 days of myocardial infarction when others performed after 90 days after myocardial infarction. The postulation of the hyposignal from T1 Rho image could be related to the MVO detected with the LGE image or the artefact. Future study is needed to explore this finding.

Spin-lock amplitude and duration influence T1ρ sensitivity to dipolar and chemical exchange mechanisms, affecting measured relaxation times and thus diagnostic cutoffs for scar detection. In general, higher amplitudes reduce B0- and B1 sensitivity, and longer durations enhance fibrosis contrast but increase motion and SNR loss (12,13). A direct and more detailed influence of the T1rho sequence parameters to the cutoff value for scar detection is beyond the scope of this clinical evaluation study and needs to be further investigated.

Limitations of study

This study, while demonstrating the potential utility of native T1ρ mapping in detecting myocardial scars in patients with known or suspected CAD, has several limitations that must be acknowledged: (I) clinical applicability. (i) Single-center design: the study was conducted at a single tertiary referral center, which may limit the generalizability of the findings. (ii) Limited diagnostic performance of T1ρ alone: the sensitivity of T1ρ imaging alone for scar detection was relatively low (~39%) compared to LGE, although it improved significantly when combined with SSFP cine images. This suggests that T1ρ alone may not yet be suitable as a standalone replacement for LGE in routine clinical practice. (iii) Lack of external validation. (II) Technical limitations. (i) Limited spatial coverage of T1ρ imaging: due to time and breath-hold constraints, only one mid-ventricular short-axis slice was acquired for both rest and stress T1ρ mapping. (ii) Technological limitations of T1ρ acquisition: the scan protocol was restricted to 4 SLTs due to hardware limitations, potentially affecting the accuracy of T1ρ fitting. Additionally, stress T1ρ imaging was performed only at a single slice level and during the first-pass perfusion phase. (iii) Due to technical limitations, only 250 out of 377 patients had an interpretable T1rho map. (iv) the T1ρ sequence we used in this study was previously also used by Okuaki et al. (11), where the first SLT was reported as 1 ms due to two short 90 degree block pulses for flip down and flip up, but without spin-lock component—to report the scan parameters, for simplicity, should be “0 ms”. (III) Methodological constraints. (i) Qualitative scar detection dependent on reviewer interpretation: the qualitative assessment of myocardial scarring on T1ρ images relied heavily on visual interpretation, introducing subjectivity. (ii) Breath-hold and heart rate limitations: the need for multiple breath-holds and heart rate dependence may impact reproducibility, especially in older or dyspneic patients who may not tolerate breath-holding well. The same applies to multiple breath-holds with no motion correction, which was not available in the baseline software release at the time of the study. This constraint may lead to residual misregistration between T1ρ-weighted images and affect the accuracy of T1ρ measurements. Future clinical studies should incorporate motion correction such as inline or retrospective nonrigid co-registration (21) or explore single-breath-hold approaches to enhance measurement reliability and reproducibility (22).

Conclusions

Native T1ρ mapping offers the potential for contrast-free scar detection in CMR. The T1ρ threshold ≥49.2 ms using the scheme of SL 0, 20, 30 and 40 ms yielded the optimal discrimination of myocardial scar. The sensitivity for scar detection was at 39%, which increased to 71% when used in combination with SSFP cine. Given limited sensitivity, native T1ρ mapping is deemed a complementary rather than a standalone technique at the current stage for clinical applications.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the STARD reporting checklist. Available at https://qims.amegroups.com/article/view/10.21037/qims-2025-1749/rc

Data Sharing Statement: Available at https://qims.amegroups.com/article/view/10.21037/qims-2025-1749/dss

Funding: None.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://qims.amegroups.com/article/view/10.21037/qims-2025-1749/coif). S.B. and P.S. are employees of Philips (Thailand) Ltd. S.Z. is an employee of Philips Medical Systems Nederland. The other 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. This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study protocol was approved by the Siriraj Institutional Review Board [certificate of approval (COA) number Si. 446/2021]. All enrolled patients provided written informed consent.

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. Kim HW, Farzaneh-Far A, Kim RJ. Cardiovascular magnetic resonance in patients with myocardial infarction: current and emerging applications. J Am Coll Cardiol 2009;55:1-16. [Crossref] [PubMed]
2. Ramalho J, Ramalho M, Jay M, Burke LM, Semelka RC. Gadolinium toxicity and treatment. Magn Reson Imaging 2016;34:1394-8. [Crossref] [PubMed]
3. Dekker HM, Stroomberg GJ, Van der Molen AJ, Prokop M. Review of strategies to reduce the contamination of the water environment by gadolinium-based contrast agents. Insights Imaging 2024;15:62. [Crossref] [PubMed]
4. van Oorschot JW, Gho JM, van Hout GP, Froeling M, Jansen Of Lorkeers SJ, Hoefer IE, Doevendans PA, Luijten PR, Chamuleau SA, Zwanenburg JJ. Endogenous contrast MRI of cardiac fibrosis: beyond late gadolinium enhancement. J Magn Reson Imaging 2015;41:1181-9. [Crossref] [PubMed]
5. Han Y, Liimatainen T, Gorman RC, Witschey WR. Assessing Myocardial Disease Using T(1ρ) MRI. Curr Cardiovasc Imaging Rep 2014;7:9248. [Crossref] [PubMed]
6. van Oorschot JW, El Aidi H. Jansen of Lorkeers SJ, Gho JM, Froeling M, Visser F, Chamuleau SA, Doevendans PA, Luijten PR, Leiner T, Zwanenburg JJ. Endogenous assessment of chronic myocardial infarction with T(1ρ)-mapping in patients. J Cardiovasc Magn Reson 2014;16:104. [Crossref] [PubMed]
7. van Oorschot JW, Visser F, Eikendal AL, Vonken EJ, Luijten PR, Chamuleau SA, Leiner T, Zwanenburg JJ. Single Breath-Hold T1ρ-Mapping of the Heart for Endogenous Assessment of Myocardial Fibrosis. Invest Radiol 2016;51:505-12. [Crossref] [PubMed]
8. Bustin A, Toupin S, Sridi S, Yerly J, Bernus O, Labrousse L, Quesson B, Rogier J, Haïssaguerre M, van Heeswijk R, Jaïs P, Cochet H, Stuber M. Endogenous assessment of myocardial injury with single-shot model-based non-rigid motion-corrected T1 rho mapping. J Cardiovasc Magn Reson 2021;23:119. [Crossref] [PubMed]
9. Liu A, Wijesurendra RS, Francis JM, Robson MD, Neubauer S, Piechnik SK, Ferreira VM. Adenosine Stress and Rest T1 Mapping Can Differentiate Between Ischemic, Infarcted, Remote, and Normal Myocardium Without the Need for Gadolinium Contrast Agents. JACC Cardiovasc Imaging 2016;9:27-36. [Crossref] [PubMed]
10. Yimcharoen S, Zhang S, Kaolawanich Y, Tanapibunpon P, Krittayaphong R. Clinical assessment of adenosine stress and rest cardiac magnetic resonance T1 mapping for detecting ischemic and infarcted myocardium. Sci Rep 2020;10:14727. [Crossref] [PubMed]
11. Okuaki T, Takayama Y, Nishie A, Ogino T, Obara M, Honda H, Miyati T, Van Cauteren M T. (1ρ) mapping improvement using stretched-type adiabatic locking pulses for assessment of human liver function at 3T. Magn Reson Imaging 2017;40:17-23. [Crossref] [PubMed]
12. Witschey WR, Zsido GA, Koomalsingh K, Kondo N, Minakawa M, Shuto T, McGarvey JR, Levack MM, Contijoch F, Pilla JJ, Gorman JH 3rd, Gorman RC. In vivo chronic myocardial infarction characterization by spin locked cardiovascular magnetic resonance. J Cardiovasc Magn Reson 2012;14:37. [Crossref] [PubMed]
13. Bustin A, Pineau X, Sridi S, van Heeswijk RB, Jaïs P, Stuber M, Cochet H. Assessment of myocardial injuries in ischaemic and non-ischaemic cardiomyopathies using magnetic resonance T1-rho mapping. Eur Heart J Cardiovasc Imaging 2024;25:548-57. [Crossref] [PubMed]
14. Wáng YX, Zhang Q, Li X, Chen W, Ahuja A, Yuan J. T1ρ magnetic resonance: basic physics principles and applications in knee and intervertebral disc imaging. Quant Imaging Med Surg 2015;5:858-85. [Crossref] [PubMed]
15. Cerqueira MD, Weissman NJ, Dilsizian V, Jacobs AK, Kaul S, Laskey WK, Pennell DJ, Rumberger JA, Ryan T, Verani MSAmerican Heart Association Writing Group on Myocardial Segmentation and Registration for Cardiac Imaging. Standardized myocardial segmentation and nomenclature for tomographic imaging of the heart. A statement for healthcare professionals from the Cardiac Imaging Committee of the Council on Clinical Cardiology of the American Heart Association. Int J Cardiovasc Imaging 2002;18:539-42.
16. Lemainque T, Pridöhl N, Huppertz M, Post M, Yüksel C, Siepmann R, Radke KL, Zhang S, Yoneyama M, Prescher A, Kuhl C, Truhn D, Nebelung S. Two for One-Combined Morphologic and Quantitative Knee Joint MRI Using a Versatile Turbo Spin-Echo Platform. Diagnostics (Basel) 2024;14:978. [Crossref] [PubMed]
17. Lemainque T, Pridöhl N, Zhang S, Huppertz M, Post M, Yüksel C, Yoneyama M, Prescher A, Kuhl C, Truhn D, Nebelung S. Time-efficient combined morphologic and quantitative joint MRI: an in situ study of standardized knee cartilage defects in human cadaveric specimens. Eur Radiol Exp 2024;8:66. [Crossref] [PubMed]
18. Brodov Y, Matetzky S, Konen E, Di Segni M, Younis A, Massalha A, Berkovitch A, Beytelman A, Chernomordik F, Goitein O. Elevated Native T1 Values in the Remote Myocardium Supplied by Obstructive Non-Infarct Related Coronary Arteries in Post-STEMI Cardiac Magnetic Resonance. Cardiology 2023;148:106-13. [Crossref] [PubMed]
19. Abbas A, Matthews GH, Brown IW, Shambrook JS, Peebles CR, Harden SP. Cardiac MR assessment of microvascular obstruction. Br J Radiol 2015;88:20140470. [Crossref] [PubMed]
20. Bodi V, Gavara J, Lopez-Lereu MP, Monmeneu JV, de Dios E, Perez-Sole N, Bonanad C, Marcos-Garces V, Canoves J, Minana G, Nunez J, Moratal D, Chorro FJ, Rodríguez-Palomares JF, Freixa A, Borrás R, Ortiz-Pérez JT, Rios-Navarro C. Impact of Persistent Microvascular Obstruction Late After STEMI on Adverse LV Remodeling: A CMR Study. JACC Cardiovasc Imaging 2023;16:919-30. [Crossref] [PubMed]
21. Zhang S, Le TT, Kabus S, Su B, Hausenloy DJ, Cook SA, Chin CWL, Tan RS. Cardiac magnetic resonance T1 and extracellular volume mapping with motion correction and co-registration based on fast elastic image registration. MAGMA 2018;31:115-29. [Crossref] [PubMed]
22. Zhang S, Okuaki T, Kabus S, Leong B, Han Y, Hung YH, Wang P, Tan RS. Free-breathing native cardiac T1 and T2 mapping with flexible elastic image registration: Preliminary Clinical Result. Proc Intl Soc Magn Reson Med 2017;2754: