Reducing SAR

What practical steps can a technologist use to reduce SAR?  

From a prior Q&A we developed a simplified equation to express specific absorption rate (SAR) in terms of several other tissue related and technical factors, including: σ (tissue conductivity), A (patient cross sectional area), ω (RF frequency, which is proportional to Bo main field strength by the Larmor equation), B1 (the intensity of the RF field, which is proportional to flip angle α), and D (the duty cycle, or percent time RF is "on" during a given pulse sequence. Considering these factors leads to a better understanding of what sequences produce the highest levels of SAR and provide a wide range of potential strategies for reducing it. 

---

Since SAR is proportional to the square of the flip angle, sequences with the highest SAR levels tend to be those containing a large number of 180º refocusing-pulses applied over a short period of time. Fast (Turbo) Spin Echo is the archetype, followed by the Inversion Recovery family. Saturation pulses of all types (e.g., Fat-Sat, Spatial Sat, and especially Magnetization Transfer) may cause SAR limits to be exceeded. As a rule, Gradient Echo sequences produce much lower SAR effects (as they use gradient reversals rather than refocusing pulses to rephase the MR signal). Exceptions to this rule are GRE sequences performed at extremely short TR's, such as bSSFP/FIESTA and TOF MR, the latter of which also include traveling saturation bands that add to the SAR load. 

---

Methods for Reducing SAR

1. Avoid or minimize use of high SAR sequences (in the box above). If they must be used, interleave them in the protocol with lower SAR sequences to allow tissue cooling in between.
2. Use Gradient Echo instead of Spin Echo sequences where possible. Note: increased susceptibility artifacts
3. Reduce initial RF flip angle (α). Especially useful for super-short TR sequences like fully rewound GRE (e.g. bSSFP). Image contrast and SNR may be affected.
4. Increase TR (without increasing # of slices). Note: incurs a time penalty
5. Decrease # of slices (without reducing TR). Note: incurs a time penalty
6. Decrease # of k-space views. Can be accomplished by decreasing # phase encode steps (accompanied by loss of resolution), using rectangular FOV (not applicable to all anatomy), using parallel imaging (SNR loss and potential artifacts). 
7. For FSE/TSE reduce # echoes by decreasing ETL/turbo factor. Note: incurs a time or slice number penalty
8. For FSE/TSE reduce angle of refocusing pulses (from 180º to 135º-150º). Surprisingly there is relatively little effect on image contrast but SNR will decrease. Some special 3D techniques like CUBE/SPACE do this automatically.
9. Use Low SAR or Variable-Rate Selective Excitation (VERSE) pulses. Note: incurs a time or slice number penalty; not available on all scanners. See advanced discussion for details.​​

Advanced Discussion (show/hide)»

Two modifications of RF-pulses to reduce SAR are pictured below:

Low SAR Pulses reduce the maximum amplitude of the RF envelope and associated gradient and extend the time of the pulse and gradient by an equivalent amount. In Siemens scanners, for example, the "Low SAR" pulse is simply the standard RF-pulse scaled by a factor of 1.5 (i.e., 1.5 times as long and 1/1.5 = 2/3 as high). The penalty here is that the the relatively longer RF-pulses affect the minimum allowable TE and potentially number of slices.

Variable-Rate Selective Excitation (VERSE) Pulses for low SAR applications leave the pulse length unchanged, but modify the envelope of the RF pulse with a corresponding "dip" in gradient amplitude where at times where the RF pulse is highest. Minimum SAR VERSE pulses are typically somewhat flat in their central region and amplified toward their periphery. The main disadvantage is some smearing of the slice profile that can lead to spatial blurring in the image.

---

References
     Allison J, Yanasak N. What MRI sequences produce the highest specific absorption rate (SAR), and is there something we should be doing to reduce the SAR during standard examinations? Am J Roentgenol 2015; 205:W140. [DOI Link]
     Carluccio G, Collins CM. Optimization of the order and spacing of sequences in an MRI exam to reduce the maximum temperature and thermal dose. Magn Reson Med 2019; 81:2161-2166.  [DOI Link]
     Conolly S, Nishimura D, Macovski A, Glover G. Variable-rate selective excitation. J Magn Reson 1988; 78:440-458. (original description of VERSE pulses) [DOI Link]
     Hargreaves BA, Cunningham CH, Nishimura DG, Conolly SM. Variable-rate selective excitation for rapid MRI sequences. Magn Reson Med 2004; 52:590-597. [DOI Link]
     Prost JEH, Wehrli FW, Drayer B, et al. SAR reduced pulse sequences. Magn Reson Imaging 1988; 6:125-130. [DOI Link]
     Sarkar SN, Alsop DC, Madhuranthakam AJ, et al. Brain MR imaging at ultra-low radiofrequency power. Radiology 2011; 259:550-557. [DOI Link]

---

Related Questions
     How do radiofrequency fields affect biological tissues?
     What is SAR? ​

---

←  Previous Question

Next Question  →

↑ Complete List of Questions ↑