Elastography is a method of imaging that detects the compressibility or stiffness of tissues in the imaging field and then overlays a false-color map upon the greyscale image to indicate which tissues are hard/stiff versus soft/compressible. The science behind the technique is beyond the scope of a blog post, particularly as there are several methods by which elastography can be performed.
In practical terms, elastography is useful in identifying lesions that are sonographically iso-dense compared to their surroundings. Such lesions, while they are therefore visually “iso-grey” (if you will tolerate a neologism), may not be iso-compressible despite their iso-density, and thus when their differential compressibility is identified by elastography it becomes possible to characterize a lesion whose greyscale appearance is not instructive. Among the most common current uses of elastography are the characterization of breast and liver lesions, and indeed the well-known Fibroscan device is, in essence, liver elastography.
There are several instances in the field of musculoskeletal (MSK)/rheumatologic ultrasound in which this technology is appealing, but more work is needed before widespread use will be advisable. I will mention only two of the most obvious examples here.
Example One
The first example is in the interrogation of a symptomatic tendon or ligament. Such a structure, whose normal function involves incredible amounts of linear tension, when disrupted by trauma or disease, would be expected to lose integrity in the region of the insult and become softer/more compressible than normal in that area.
Traditionally, elastography is not used to measure tendons and ligaments despite the validity of the above statement. The reason for this is that the stiffness of tissue, when measured by elastography, can be expressed in terms of the speed at which a deformation (compression wave) in the tissue propagates, usually in meters per second (there are other units by which stiffness can be measured, but for simplicity’s sake, I will leave it at that).
In the classical case of breast and liver lesions, this is not an issue since the surrounding normal tissue is relatively soft and compressible, so the speed of the propagation of a compression wave is relatively slow. Thus, most elastography measurements top out at a propagation speed of about 10 meters per second, and most normal and abnormal breast/liver tissue will have stiffness values somewhat slower than this. Tendons and ligaments, on the other hand, are by nature very hard/noncompressible. Even in their “relaxed” state, these tissues are so bowstring-tight (relatively) that measuring a normal Achilles’ tendon, for example, will yield only a maxed-out value of “offscale hard” throughout the entire structure.
It is tempting to say that one could simply recalibrate the machine to measure faster propagation speeds, but, unfortunately, we run into limitations of our current technology. It is simply not possible currently to measure velocities much faster than 10 m/s.
While we await advancements in technology, the current workaround is to trust that a damaged region of tendon or ligament will be significantly softer, and thus transmit compression waves much more slowly. Therefore, we simply consider any propagation speed that falls out of “offscale” and into the measurable range to be an indicator of pathology.
Example Two
The second example of the potential rheumatologic utility of elastography is in the assessment of systemic sclerosis, commonly known as scleroderma. As the Greek name would suggest, this disease usually includes a characteristic hardening of the skin. The problem is that there is currently no reliable way to quantify skin stiffness. The existing gold standard is a semi-quantitative scoring of skin thickening performed by simple physical examination in which each of several predefined regions of the skin is palpated and assigned a value from 0 to 3. This results in an overall score known as the Modified Rodnan Skin Score (MRSS). Performing Rodnan scoring requires an experienced clinician, and since scleroderma is a rare disease, very few physicians have a large enough cohort in their practice to be able to consider themselves expert Rodnan scorers.
This leads to a host of problems, and one of the worst is that clinical trials in scleroderma (a devastating and potentially fatal disease for which no good treatment exists) are very difficult to conduct because one of the primary endpoints of any trial will be the degree of improvement found in this semi-quantitative and hard-to-perform examination, which is subject to severe inter-rater reliability problems.
When I first started as a rheumatology fellow, I agreed to help with a scleroderma clinical trial in the role of a blinded efficacy assessor. The sponsor brought a dozen or so of us to a hotel for training, and all morning long we cycled through a series of hotel meeting rooms, each containing a volunteer patient for us to score.
It was a disaster.
After lunch, the representative from the sponsor got up to the podium and told us to rip up our afternoon agendas—we were going back to the meeting rooms to examine the volunteers again in an effort to improve the scoring consensus.
Clearly, this situation screams for elastography. The objective measurement of skin stiffness is precisely the datum that is sorely needed. Sadly, our current technology again fails us, as present-day elastography has limitations in resolution and the skin by its anatomic location, will always be very nearly directly applied to the probe face, in a region outside the focal zone of the beam where the measurement physics work best. Further, one of the techniques for performing elastography is highly operator-dependent, because the compression waves being measured are generated by manually varying the pressure of the probe against the skin—definitely a skill that must be learned over time and one that opens the door once more to inter-rater variability.
Overall, elastography holds great promise for MSK/rheumatologic applications in the future, as described in the two examples above. For now, however, it’s currently a technology that is “not ready for prime time” in this field.
This post is intended as a companion to “What Rheumatologists Really Need for Ultrasound Is…”, which discusses advances in ultrasound technology that are sorely needed in the field of MSK ultrasound, and specifically in rheumatology.
Dr. Mandelin is an academic rheumatologist, registered in MSK ultrasound (RhMSUS) by the American College of Rheumatology and certified in MSK ultrasound (RMSK) by the Alliance for Physician Certification & Advancement. He currently serves the AIUM as secretary of the High-Frequency Clinical and Preclinical Imaging Community. Connect with him on Twitter @NU_Rheum_MSK_US.
at medical students are graduating and insisting on using ultrasound in their residency training. It would seem that many of our medical students are learning ultrasound at a rate that will outpace attending physician knowledge, exposure, and experience. Indeed, when teaching ultrasound to many of the medical students at West Virginia University as part of their medical education, I was astounded to see how proficient they were at using the machine, the transducer, and correctly identifying both normal and pathologic anatomy. It’s my understanding that many universities have included medical ultrasound into the academic curricula as a bridge to their respective gross anatomy courses and in their general clinical medical education.