Shortened Modified Look-Locker Inversion recovery (ShMOLLI) for clinical myocardial T1-mapping at 1.5 and 3 T within a 9 heartbeat breathhold
© Piechnik et al; licensee BioMed Central Ltd. 2010
Received: 20 July 2010
Accepted: 19 November 2010
Published: 19 November 2010
T1 mapping allows direct in-vivo quantitation of microscopic changes in the myocardium, providing new diagnostic insights into cardiac disease. Existing methods require long breath holds that are demanding for many cardiac patients. In this work we propose and validate a novel, clinically applicable, pulse sequence for myocardial T1-mapping that is compatible with typical limits for end-expiration breath-holding in patients.
Materials and methods
The Shortened MOdified Look-Locker Inversion recovery (ShMOLLI) method uses sequential inversion recovery measurements within a single short breath-hold. Full recovery of the longitudinal magnetisation between sequential inversion pulses is not achieved, but conditional interpretation of samples for reconstruction of T1-maps is used to yield accurate measurements, and this algorithm is implemented directly on the scanner. We performed computer simulations for 100 ms<T1 < 2.7 s and heart rates 40-100 bpm followed by phantom validation at 1.5T and 3T. In-vivo myocardial T1-mapping using this method and the previous gold-standard (MOLLI) was performed in 10 healthy volunteers at 1.5T and 3T, 4 volunteers with contrast injection at 1.5T, and 4 patients with recent myocardial infarction (MI) at 3T.
We found good agreement between the average ShMOLLI and MOLLI estimates for T1 < 1200 ms. In contrast to the original method, ShMOLLI showed no dependence on heart rates for long T1 values, with estimates characterized by a constant 4% underestimation for T1 = 800-2700 ms. In-vivo, ShMOLLI measurements required 9.0 ± 1.1 s (MOLLI = 17.6 ± 2.9 s). Average healthy myocardial T1 s by ShMOLLI at 1.5T were 966 ± 48 ms (mean ± SD) and 1166 ± 60 ms at 3T. In MI patients, the T1 in unaffected myocardium (1216 ± 42 ms) was similar to controls at 3T. Ischemically injured myocardium showed increased T1 = 1432 ± 33 ms (p < 0.001). The difference between MI and remote myocardium was estimated 15% larger by ShMOLLI than MOLLI (p < 0.04) which suffers from heart rate dependencies for long T1. The in-vivo variability within ShMOLLI T1-maps was only 14% (1.5T) or 18% (3T) higher than the MOLLI maps, but the MOLLI acquisitions were twice longer than ShMOLLI acquisitions.
ShMOLLI is an efficient method that generates immediate, high-resolution myocardial T1-maps in a short breath-hold with high precision. This technique provides a valuable clinically applicable tool for myocardial tissue characterisation.
In cardiovascular magnetic resonance (CMR), tissue contrast is generated by a combination of intrinsic tissue properties such as spin-lattice (T1) and spin-spin (T2) relaxation times, and extrinsic properties such as imaging sequence and settings. Signal intensity in conventional CMR images is displayed on an arbitrary scale, and thus is not suited to comparison between subjects. T1-mapping provides a quantitative surrogate marker for the cellular environment of the myocardial water allowing direct comparison between patients and examinations, which can operate without the need for exogenous contrast agents.
T1 relaxation times depend on the composition of tissues, and each tissue type exhibits a characteristic range of normal values at a selected magnetic field strength . Deviation from established ranges can thus be used to quantify the effects of pathological processes. Focal and global T1 changes are reported in a number of myocardial diseases, such as myocardial infarction [2, 3], heart failure , valvular heart disease , and systemic diseases with cardiac involvement such as amyloidosis [6, 7] or systemic lupus erythematosus . T1-mapping may be a sensitive technique for detecting diffuse fibrosis in heart failure and valvular heart disease, which have been described by abnormal post-contrast T1 values but not by conventional late gadolinium enhancement (LGE) imaging [4, 5].
We hypothesise that it is possible to determine T1 maps of the heart with high precision in a short breath-hold. We thus present a shortened alternative to MOLLI (ShMOLLI) which can generate rapid and high-resolution myocardial T1-maps in a single short breath-hold of only 9 heartbeats. We investigate the accuracy of T1 measurements by ShMOLLI against MOLLI in simulation and gel phantom experiments over a wide range of heart rates and T1 values. The findings are validated in-vivo for normal human myocardium at 1.5T and 3T. To extend the in-vivo validation range we also present findings in four human subjects with recent myocardial infarction and four normal subjects after gadolinium contrast.
Materials and methods
Simulation and sequence design
Simulations were performed in IDL (Interactive Data Language ver. 6.1, ITT Visual Information Solutions) by implementing equations for the piece-wise calculation of longitudinal magnetisation (Mz(t)) and the signal samples generated by a train of arbitrarily-spaced ideal excitation pulses. Inversion pulses are assumed to be perfect 180° excitations; readouts are simulated as single pulses of 35°. Both sequences had three IR epochs. Simulations were performed for MOLLI based on its optimised variant , which collects 3+3+5 samples in three consecutive IR epochs separated by long recovery periods, which are approximated as being fully recovered in the reconstruction. The shortest effective TI  in each IR are 100, 180 and 260 ms; incremented by heart beat period for subsequent samples (Figure 1A). ShMOLLI uses a similar effective TI principle but collects only 5+1+1 samples and IR epochs are separated by only one TRR (R-R interval), which does not approximate to full recovery of the magnetization (Figure 1B).
Simulations using simplified pulse sequences outlined in Figure 1 were performed for T1 ranging from 50 to 2700 ms (50 ms increments) and for HR between 40-100 bpm (20 bpm increments). For each combination of parameters, we analysed 200 sample trains with noise representative of our phantom measurements.
The MOLLI and ShMOLLI acquisitions used a research sequence provided by Siemens (Siemens Healthcare, Germany) utilising the optimised parameter set for MOLLI  and ShMOLLI. T1 s were fitted using the algorithms described above with an AMOEBA optimisation method . This reconstruction fitting was coded in C++ directly into the scanner's multi-threaded parallel processing image pipeline. As a result, T1-maps are available for viewing in about 10 seconds following image acquisition.
Fifteen 50 ml Agarose and NiCl gel phantoms  with T2~50 ms and T1 ~100-2500 ms were studied at 1.5T (Avanto, Siemens Healthcare, Germany) using a 32 channel body array and at 3T (Trio, Siemens Healthcare) using a 16 channel body array. The artificially generated ECG scanner triggers were used to validate HR range from 40-100 bpm with 10 bpm increments. Imaging parameters were the same for MOLLI and ShMOLLI: FOV = 300 × 300 mm, voxel size = 1.17 × 1.17 × 8 mm, matrix = 256 × 256 (interpolated from 128 × 128 acquisition matrix, 95 phase encoding steps), flip angle = 35°, TR/TE = 1.96/0.98 ms.
Reference T1 relaxation times were calculated offline based on images collected using slice-selective IR with non-segmented spin echo readout. TI = 33, 100, 300, 900, 2700, 5000 ms. TR/TE = 10 s/6.3 ms; 80 phase encoding steps with total image acquisition time of 13 minutes. Sizing and positioning were identical to the studied MOLLI methods. Regions of interest were placed in each tube using an automated method and reference T1 s were fitted per pixel and a mean T1 determined.
Ethics approval was granted for all study procedures and informed consent was obtained from all subjects.
10 normal volunteers (7 men; age 35 ± 7 years, normal ECGs without history of cardiac diseases or symptoms) underwent CMR imaging at 1.5T and 3T on the same day. Following standard planning, end-expiration basal, mid-cavity and apical short-axis images using MOLLI and ShMOLLI were collected. Images for specific TI were collected using exactly the same SSFP readouts for both methods to allow direct comparisons, typically: TR/TE = 2.14/1.07 ms, flip angle = 35°, FOV = 340 × 255 mm, matrix = 192 × 144, 107 phase encoding steps, interpolated voxel size = 0.9 × 0.9 × 8 mm, GRAPPA = 2 with 24 reference lines, cardiac delay time TD = 500 ms; 206 ms acquisition time for single image. A single slice, which was judged to have the "best quality" at the time of scanning, was repeated twice at the end of the protocol to assess short-term intra-scan variability of the T1 measurements; this was not performed in the first pilot case. Offline post-processing involved manual tracing of endo- and epi-cardial contours for analysis of the T1 measurements in myocardial segments 1 to 16 of the American Heart Association (AHA) 17-segment model  using in-house software.
Gadolinium contrast application
Matching pairs of ShMOLLI and MOLLI pre-contrast and post-contrast T1 maps were obtained in 4 female subjects (61 ± 3 years old) without pre-existing cardiac disease who underwent a separate research protocol at 1.5T. Subjects underwent adenosine stress perfusion at 140 μg/kg/min for 3 min, followed by a bolus of Gd (Gadodiamide, Omniscan, GE Healthcare, Amersham, UK, 0.03 mmol/kg body weight). After 20 minutes, resting perfusion imaging was performed using 0.03 mmol/kg of Gd followed immediately by a top-up Gd of 0.10 mmol/kg for LGE imaging. Matching T1-maps were obtained at baseline and ~14 minutes after adenosine stress perfusion. Finally, 4 pairs of images were collected before, and one after, the LGE images. The dynamic evolution of T1 recovery after the final Gd bolus was corrected with 3rd order polynomial for the purpose of constructing Bland-Altman plots.
Patients with recent myocardial infarction
4 patients (3 men; age 53 ± 10 years) underwent CMR at 3T following the diagnosis of a first acute ST-elevation type myocardial infarction (STEMI) post primary percutaneous coronary intervention (PCI). LGE images were obtained 24-48 hours post acute infarct . T1-maps using ShMOLLI and MOLLI were obtained 5-17 days after the ischemic event at a single representative slice replicating settings from the normal control volunteer study. Manual contouring of the endo- and epicardium was followed by calculation of the distribution of T1 values within the defined myocardium. The resulting distributions were clearly bi-modal and were fitted using a two-component Gaussian model in order to assess T1 in injured and unaffected myocardium.
Unless stated otherwise, the results are presented as mean ± SD/mixed SD, where the "mixed SD" is estimated as the independent combination of average individual SD and the interindividual SD. For estimation of the relative variability we use the coefficient of variation (CV = 100%*SD/Mean). The term "noise penalty" is used to describe the expected relative increase in CV resulting from predicable factors, i.e. the reduced number of samples. Significance of the differences between population means is calculated using 2-tailed Student T-tests, paired whenever possible, and quoted when p <0.05.
Simulation and phantom studies
The variability of T1 measurements across all simulations or pixels is shown as whiskers in Figure 3. For long T1 s, measurements by ShMOLLI are based on 5 samples whereas MOLLI measurements are based on 11 samples. Therefore ShMOLLI has a predicted noise penalty of 48% versus MOLLI (√(11/5)). For shorter T1 s, 6 or 7 TI samples are used with a predicted noise penalty of 35% and 25%. Inclusive of the heavily nonlinear processing, for the reference T1 range of 300-2600 ms, we find the simulated average CV is 2.7% for ShMOLLI and 2.1% for MOLLI, translating to an overall 28% noise penalty for using ShMOLLI. In phantom studies, the noise penalty for ShMOLLI is 21% at 3T (Figure 3B) and 61% at 1.5T (Figure 3C).
Normal human subjects
Gadolinium contrast application
None of the subjects demonstrated LGE lesions. The baseline T1 values were 982 ± 28 ms, comparable with other controls at 1.5T. Depending on the dose of gadolinium and time point the T1 showed expected shortening with the lowest average myocardial T1 of 135 ± 33 ms and subsequent recovery towards normal values (Figure 4). The differences between the ShMOLLI and MOLLI measurements are shown together with the remaining in-vivo data in Figure 6 and confirm good agreement in-vivo in the low T1 regime predicted in simulation and phantom measurements.
Patients with recent myocardial infarction
Assessment of myocardium by T1-mapping and LGE at 3T in acute STEMI treated by 1° PCI.
LGE (days prior)
1200 ± 46
1392 ± 33
1189 ± 39
1369 ± 34
1269 ± 69
1425 ± 21
1244 ± 50
1418 ± 30
1248 ± 47
1391 ± 21
1244 ± 51
1445 ± 45
1230 ± 33
1389 ± 46
1171 ± 47
1433 ± 48
1165 ± 55
1469 ± 59
53 ± 10
1217 ± 37
1424 ± 23
48 ± 9%
44 ± 4%
Excluding any such corrections (as no corrected values are presented in this work), the measurements of myocardial T1 by ShMOLLI at 1.5T are in good agreement with in-vivo data published in the literature , particularly with previous measurements using the MOLLI sequence . The observed minor differences between the two methods conform closely to simulation and phantom validation.
The purpose of the limited in-vivo material was to directly compare ShMOLLI performance with MOLLI over a wide range of clinically relevant scenarios. This included measurements in normal myocardium, following shortening of T1's as a result of contrast administration and T1 prolongation related to myocardial infarction. T1 measurements agreed in healthy controls within ± 17 ms noise boundaries for normal myocardium and after Gd administration. Our MI data is consistent with previously published reports [2, 9, 13, 20] showing that the area of infarct demonstrates increased T1 values on non-contrast T1 maps. LGE images were acquired 4-16 days prior to T1-maps, which precludes direct comparisons of infarct size assessment by LGE imaging and T1-mapping. Nonetheless, it is clear that the distribution of T1 values within the myocardium shows two distinct peaks of normal and long T1 s, the latter co-localising to the area of injured myocardium as seen on LGE. ShMOLLI showed a 15% larger T1 difference between injured and unaffected myocardium, which can be directly attributed to its superior quantitation of long T1 s.
Typically, the penalty for reducing the number of images collected for T1 reconstruction is increased variability within the resulting T1 maps. Our simulation and phantom measurements predicted the noise penalty for ShMOLLI to be as much as 61%. However, in-vivo T1 measurements by ShMOLLI showed noise penalty of only 14% (1.5T) and 18% (3T) as compared to MOLLI T1 s. The repetition error was 1.6% for both methods at 1.5T and only excluding the worst measurements brought out the expected noise advantage of MOLLI at 3T. We found that while MOLLI benefited from more samples for the majority of measurements, the higher SNR was bought at the cost of more prominent outliers, especially at 3T. This may be due to the ShMOLLI conditional processing acting as an error checker, but detailed analysis of such effects exceeds the scope of this work. Overall, for most of the T1 values tested, the in-vivo noise penalty for ShMOLLI due to a shorter imaging time is only 10-20%. We attribute this favourable observation mostly to the beneficial effect of the short breath-hold, which limits the incidence of breathing motion, estimated as the cause for 31% of image artefacts . The small reduction in precision of ShMOLLI T1 maps in-vivo is an excellent trade-off for halving the imaging time and breath-hold, rendering the ShMOLLI both accurate and clinically acceptable.
The ShMOLLI technique holds promise for wider application of non-contrast T1-mapping towards cardiac conditions in which there are focal changes, such as myocardial infarction and myocarditis. It may be pivotal in studying conditions in which there are diffuse myocardial changes, such as chronic heart failure, hypertrophic cardiomyopathy and infiltrative diseases such as cardiac amyloidosis and sarcoidosis. The presented comparisons warrant the ongoing ShMOLLI deployment on its own merit in several more comprehensive patient validation studies.
The novel ShMOLLI sequence for myocardial T1-mapping generates robust, high resolution quantitative T1 maps in agreement with published data in the literature in just 9 heart-beats across a wide range of heart rates and T1 values. Single short breath holds are typical for routine examinations to make them easily achievable for patients and permit wider clinical application of quantitative mapping. Implementation of nonlinear T1 fitting directly in the scanner image reconstruction pipeline yields immediate access to T1-maps for viewing, allowing for re-acquisition if necessary. In patients presenting with recent myocardial infarction, preliminary data demonstrates ShMOLLI superiority in distinguishing injured from normal myocardium, with areas of long T1 co-localizing with injured myocardium as assessed by LGE. ShMOLLI provides a valuable clinically applicable tool for myocardial tissue characterisation with or without the use of contrast agents.
This research was funded by the NIHR Oxford Biomedical Research Centre Programme. Dr. Vanessa Ferreira is funded by the Alberta Heritage Foundation for Medical Research (AHFMR) and the University of Oxford Clarendon Fund Scholarship.
- Bottomley PA, Foster TH, Argersinger RE, Pfeifer LM: A review of normal tissue hydrogen NMR relaxation times and relaxation mechanisms from 1-100 MHz: dependence on tissue type, NMR frequency, temperature, species, excision, and age. Med Phys. 1984, 11 (4): 425-448. 10.1118/1.595535.View ArticlePubMedGoogle Scholar
- Messroghli DR, Niendorf T, Schulz-Menger J, Dietz R, Friedrich MG: T1 mapping in patients with acute myocardial infarction. J Cardiovasc Magn Reson. 2003, 5 (2): 353-359. 10.1081/JCMR-120019418.View ArticlePubMedGoogle Scholar
- Messroghli DR, Walters K, Plein S, Sparrow P, Friedrich MG, Ridgway JP, Sivananthan MU: Myocardial T1 mapping: application to patients with acute and chronic myocardial infarction. Magn Reson Med. 2007, 58 (1): 34-40. 10.1002/mrm.21272.View ArticlePubMedGoogle Scholar
- Iles L, Pfluger H, Phrommintikul A, Cherayath J, Aksit P, Gupta SN, Kaye DM, Taylor AJ: Evaluation of diffuse myocardial fibrosis in heart failure with cardiac magnetic resonance contrast-enhanced T1 mapping. J Am Coll Cardiol. 2008, 52 (19): 1574-1580. 10.1016/j.jacc.2008.06.049.View ArticlePubMedGoogle Scholar
- Sparrow P, Messroghli DR, Reid S, Ridgway JP, Bainbridge G, Sivananthan MU: Myocardial T1 mapping for detection of left ventricular myocardial fibrosis in chronic aortic regurgitation: pilot study. AJR Am J Roentgenol. 2006, 187 (6): W630-635. 10.2214/AJR.05.1264.View ArticlePubMedGoogle Scholar
- Maceira AM, Joshi J, Prasad SK, Moon JC, Perugini E, Harding I, Sheppard MN, Poole-Wilson PA, Hawkins PN, Pennell DJ: Cardiovascular magnetic resonance in cardiac amyloidosis. Circulation. 2005, 111 (2): 186-193. 10.1161/01.CIR.0000152819.97857.9D.View ArticlePubMedGoogle Scholar
- Maceira AM, Prasad SK, Hawkins PN, Roughton M, Pennell DJ: Cardiovascular magnetic resonance and prognosis in cardiac amyloidosis. Journal of Cardiovascular Magnetic Resonance. 2008, 10 (1): 10.1186/1532-429X-10-54.
- Been M, Thomson BJ, Smith MA, Ridgway JP, Douglas RHB, Been M, Best JJK, Muir AL: Myocardial involvement in systemic lupus erythematosus detected by magnetic resonance imaging. European Heart Journal. 1988, 9 (11): 1250-1256.PubMedGoogle Scholar
- Messroghli DR, Radjenovic A, Kozerke S, Higgins DM, Sivananthan MU, Ridgway JP: Modified Look-Locker inversion recovery (MOLLI) for high-resolution T1 mapping of the heart. Magn Reson Med. 2004, 52 (1): 141-146. 10.1002/mrm.20110.View ArticlePubMedGoogle Scholar
- Messroghli DR, Plein S, Higgins DM, Walters K, Jones TR, Ridgway JP, Sivananthan MU: Human myocardium: single-breath-hold MR T1 mapping with high spatial resolution--reproducibility study. Radiology. 2006, 238 (3): 1004-1012. 10.1148/radiol.2382041903.View ArticlePubMedGoogle Scholar
- Marks B, Mitchell DG, Simelaro JP: Breath-holding in healthy and pulmonary-compromised populations: effects of hyperventilation and oxygen inspiration. J Magn Reson Imaging. 1997, 7 (3): 595-597. 10.1002/jmri.1880070323.View ArticlePubMedGoogle Scholar
- Solano JP, Gomes B, Higginson IJ: A comparison of symptom prevalence in far advanced cancer, AIDS, heart disease, chronic obstructive pulmonary disease and renal disease. J Pain Symptom Manage. 2006, 31 (1): 58-69. 10.1016/j.jpainsymman.2005.06.007.View ArticlePubMedGoogle Scholar
- Messroghli DR, Greiser A, Frohlich M, Dietz R, Schulz-Menger J: Optimization and validation of a fully-integrated pulse sequence for modified look-locker inversion-recovery (MOLLI) T1 mapping of the heart. J Magn Reson Imaging. 2007, 26 (4): 1081-1086. 10.1002/jmri.21119.View ArticlePubMedGoogle Scholar
- Nelder JA, Mead R: A Simplex-Method for Function Minimization. Comput J. 1965, 7 (4): 308-313.View ArticleGoogle Scholar
- Cochlin L, Blamire A, Styles P: Dependence of T1 and T2 on high field strengths in doped agarose gels; facilitating selection of composition for specific T1/T2 at relevant field. Proceedings of 11th Annual ISMRM Meeting: 2003; Toronto, Canada. 2003, 885-Google Scholar
- Cerqueira MD, Weissman NJ, Dilsizian V, Jacobs AK, Kaul S, Laskey WK, Pennell DJ, Rumberger JA, Ryan T, Verani MS: Standardized myocardial sementation 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. Circulation. 2002, 105 (4): 539-542. 10.1161/hc0402.102975.View ArticlePubMedGoogle Scholar
- Kim RJ, Wu E, Rafael A, Chen EL, Parker MA, Simonetti O, Klocke FJ, Bonow RO, Judd RM: The use of contrast-enhanced magnetic resonance imaging to identify reversible myocardial dysfunction. N Engl J Med. 2000, 343 (20): 1445-1453. 10.1056/NEJM200011163432003.View ArticlePubMedGoogle Scholar
- Piechnik SK, Ferreira VM, Dall'Armellina E, Cochlin LE, Robson MD: Shortened Modified Look Locker Inversion recovery (Sh-MOLLI) cardiac gated mapping of T1 - Theory and phantom verification. 26th Annual Scientific Meeting, ESMRMB Oct 1-3, 2009 2009; Antalya, Turkey. 2009, 485-Google Scholar
- Sharma P, Socolow J, Patel S, Pettigrew RI, Oshinski JN: Effect of Gd-DTPA-BMA on blood and myocardial T1 at 1.5T and 3T in humans. J Magn Reson Imaging. 2006, 23 (3): 323-330. 10.1002/jmri.20504.View ArticlePubMedGoogle Scholar
- Been M, Ridgeway JP, Douglas RHB: Characterisation of acute myocardial infarction by gated magnetic resonance imaging. Lancet. 1985, 2 (8451): 348-350. 10.1016/S0140-6736(85)92494-8.View ArticlePubMedGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (<url>http://creativecommons.org/licenses/by/2.0</url>), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.