First Pass Techniques

What is the best technique for "first- pass" perfusion imaging?  

Several methods are available for first-pass perfusion imaging but none has yet emerged as the unequivocal best choice. Any sequence selected for myocardial perfusion should ideally have the following characteristics:  

• MR signal linearly related to gadolinium concentration 
• Spatial coverage allowing acquisition of at least 3-4 slices within the allotted time frame
• Temporal resolution sufficient to acquire these slices within 1-2 heart beats
• Short acquisition time per slice (≤ 200 ms) to minimize motion artifact 
• Sufficient in-plane resolution (≤ 2.5mm) to distinguish the myocardial layers

Cine images from 1st pass myocardial perfusion study using a saturation recovery spoiled-GRE method

Sequences for first-pass perfusion consist of a magnetization preparation period followed by an image acquisition module. The purpose of the preparatory module is to reduce the signal of non-enhancing tissue so that gadolinium-enhancing blood and myocardium will be rendered more conspicuous. (This is analogous to background suppression commonly used in MR angiography to accentuate vessels).

Saturation recovery (SR) uses non-slice-selective 90º-pulses that tip the longitudinal magnetization (Mz) of all tissues in the imaging volume into the transverse plane. A crusher gradient (G) is typically applied immediately after the pulse to destroy transverse coherences. After being tipped into the transverse plane tissues begin to recover longitudinal magnetization as a function of their T1 values. Gadolinium-containing blood and myocardium have short T1's and recover quickly, while non-enhancing myocardium and other tissues remain relatively saturated/suppressed. An image acquisition module is then played out using a rapid sequence (spoiled-GRE, balanced-SSFP, or hybrid GRE-EPI). Lines at center of k-space are typically acquired at an "inversion" time (TI) of about 85 ms. 

Saturation recovery (SR) myocardial perfusion sequence allowing for 3 slices every heartbeat (or 6 slices every two heartbeats). Here a 90º-nonselective pulse has been used to suppress signal from the entire volume before the image acquisition/readout by any of several methods. Between 64 and 128 lines of k-space are sampled for each slice with the center line at "inversion" time TI where an appreciable difference in signal is noted between enhancing and non-enhancing tissues.

Even better contrast between enhancing and non-enhancing tissues could be obtained by waiting a little longer between the saturation pulse and image acquisition (i.e., increasing TI). However, this would significantly lengthen imaging time per slice and reduce the maximum number of slices possible in a given R-R interval. 

Slice profile of a "notched" RF-pulse. Slices 1 and 3 in the passband feel the brunt of the RF-pulse and are saturated, while Slice 2 (in the notch) is spared.

A clever method to permit lengthening of TI with little or no penalty in the number of slices is to use slice-selective notched RF-pulses. Notched RF-pulses saturate adjacent slices but have a mid-frequency gap leaving the center slice unaffected. TI is nearly doubled as the saturation pulse for a given slice has occurred when the previous slice was imaged. At present, this option is available only on GE scanners.

MR perfusion using notched saturation pulses (blue). Notched pulses contain a midfrequency gap while saturating adjacent slices. For example, the notched pulse corresponding to Slice 2 saturates Slices 1 and 3. This results in a longer inversion time (TI) with theoretically better image contrast.

Several other strategies and options for first-pass imaging exist but are not in widespread use. Interested readers should check the references and click the "Advanced Discussion" tab below.

Advanced Discussion (show/hide)»

Further Details about RF-pulses

Three broad choices are available for the type of preparation pulses used.

Non-selective RF-pulses are most widely used in first-pass imaging. The term "non-selective" means that these pulses affect the entire volume of interest (e.g., the heart) and are not confined to a single slice. They are also known as "hard" pulses, being recognized by their a rectangular shape. A major advantage of non-selective pulses is their short duration (<3 ms). Crusher gradients are commonly applied after the pulses to minimize off-resonance artifacts and stimulated echoes from T2-coherences.

A second advantage of non-selective pulses is the ability to perform multi-planar imaging. With this saturation method, for example, it is possible in a single acquisition to generate 3 short-axis views and 2 long-axis views simultaneously.

Adiabatic pulses. The 90º-crusher combination is sensitive to B₁ field variations, so composite adiabatic RF-pulses (like BIR-4) have been used to create more uniform saturation effects. However, these pulses take longer to perform and deposit more energy in the tissues (higher SAR).

Slice-selective pulses, like the notch pulse described above, have some advantages of increased number of slices for a given TI. The major disadvantage is that multiplanar acquisition is not possible as it is with non-selective pulses. Selective pulses will cause cross-talk saturation interference in other imaging slices. So usually only a set of parallel short-axis views are obtained.

Further Details about Image Acquisition Modules

The principal techniques available all require tradeoffs between signal linearity, signal-to-noise, spatial resolution, temporal resolution, and artifacts.

Spoiled-GRE methods are probably the most popular and widely used at present, especially at 3T. Typical imaging parameters might be TR/TE/FA = 2/1/10° at 150 ms per section. Due to the low flip angle, spoiled-GRE sequences have the lowest signal-to-noise The major advantage of this method is reduced motion and blood flow artifacts.

Balanced-SSFP methods (e.g. FIESTA, TrueFISP) produce the highest signal-to-noise of the major techniques. Typical imaging parameters would be similar to the spoiled-GRE technique but with a much higher flip angle (e.g. 50° instead of 10°). Major disadvantages include off-resonance artifacts ("moiré pattern") and frequency shifts induced by the gadolinium bolus. These features currently restrict the use of b-SSFP sequences to 1.5T and below.

Hybrid echo-planar/gradient echo techniques permit the fastest image acquisition times per slice (100-120 ms) and slightly greater spatial coverage especially at higher heart rates. Faster imaging translates to fewer motion-induced artifacts especially near the endocardium. Typical imaging parameters might be TR/TE/FA = 6/1/25° with an echo train length (ETL) = 4-6. The major disadvantages are the presence of susceptibility artifacts and T2-blurring, both of which make these sequences more challenging to implement at 3T where reduction in ETL is usually required.

---

References
     Coelho-Filho OR, Rickers C, Kwong RY, Jerosch-Herold M. MR myocardial perfusion imaging. Radiology 2013; 266:701-725.  
     Ding S, Wolff SD, Epstein FH. Improved coverage in dynamic contrast-enhanced cardiac MRI using interleaved gradient-echo EPI. Magn Reson Med 1998; 39:514-519. (original description of hybrid EPI technique).
     Fair MJ, Gatehouse PD, DiBella EVR, Firmin DN. A review of 3D first-pass, whole-heart, myocardial perfusion cardiovascular magnetic resonance. J Cardiovasc Magn Reson 2015; 17:68e
     Gerber BL, Raman SV, Nayak K, et al. Myocardial first-pass perfusion cardiovascular magnetic resonance: history, theory and current state of the art. J Cardiovasc Magn Reson 2008; 10:18.
     Hunold P, Maderwald S, Eggebrecht, et al. Steady-state free precession sequences in myocardial first-pass perfusion MR imaging: comparison with TurboFLASH imaging. Eur Radiol 2004; 14:409-416
      Jung B, Honal M, Hennig J, Markl M. k-t-Space accelerated myocardial perfusion. J Magn Reson Imaging 2008; 28:1080-1085.
     Kellman P, Arai AE. Imaging sequences for first pass perfusion − a review. J Cardiovasc Magn Reson 2007; 9:525-537.  
     Shehata ML, Basha T, Hayeri MR, et al. MR myocardial perfusion imaging: insights on techniques, analysis, interpretation, and findings. RadioGraphics 2014; 32:1636-1657.
     Slavin GS, Wolff SD, Gupta SN, Foo TKF. First-pass myocardial perfusion MR imaging with interleaved notched saturation: feasibility study. Radiology 2001; 219:258-163. (The notched RF method).
     Wang Y, Moin K, Akinboboye O, Reichek N. Myocardial first pass perfusion: steady-state free precession versus spoiled gradient echo and segmented echo planar imaging. Magn Reson Med 2005; 54:1123-1129.

---

Related Questions
     How does a myocardial perfusion study MRI work? Why would you want to do one? 

---

←  Previous Question

Next Question  →

↑ Complete List of Questions ↑