The signal and the noise: overemphasis on imaging and MRIs

The “signal” is any meaningful information that can be related to a particular solution or treatment and the “noise” is any random findings that are often times irrelevant. When it comes to healthcare, evidence has shown for a long time that noise is a common theme in imaging. This is important when we think about what the purpose of imaging is as it pertains to healthcare. For most providers, it is to look for a source that may be contributing to an individual’s particular symptoms(usually related to pain). A common theme we see is when people experience persistent pain without any trauma - their healthcare providers are left without an explanation as to why that occurs so they go ahead and order imaging.

It is human nature to try and figure out what is wrong but this leads us to finding a lot of false positives during imaging. The concept of false positive is that “abnormal” findings may show up on the imaging reports but they may not actually be relevant to how the person is presenting. Abnormal is put into quotations because medical research shows that people WITHOUT any symptoms are presenting with these findings which makes me us wonder how can they be abnormal. Let’s explore the prevalence of these findings below.

Research article #1:

Dr. Waleed Brinjikji, MD from the Mayo Clinic and his team conducted a study looking at 33 published research articles. The purpose of the study was to see what kind of findings were prevalent in people that did not exhibit any back pain. The total amount of participants was 3110 people WITHOUT any back pain. They found spinal degeneration in a variety of different age groups (Brinjikji 2014):

20-29 y/o: 37%
30-39 y/o: 52%
40-49 y/o: 68%
50-59 y/o: 80%
60-69 y/o: 88%
70-79 y/o: 93%
80-89 y/o: 96%

These are ALL people without symptoms that had degenerative changes in their spine. They also showed findings including: disc bulge, disc signal loss, disc height loss, disc protrusion, annular fissure, facet degeneration, and spondylolisthesis (Brinjikji 2014).

Research article #2:

Dr. Adam Culvenor, PT and his team conducted a study that looked at over 63 different peer-reviewed research papers that studied 5397 knees (Culvenor 2019). Knee osteoarthritis(OA) was found in different age groups listed below:

<40 y/o: 4-14%
>40 y/o: 19-43%

Once again, these are all participants WITHOUT any knee pain. They also found meniscal tears, cartilage defects, bone marrow lesions, and ostephytes or calcification were common (Culvenor 2019).

Research article #3:

Dr. Laura-Maria Horgan, PhD and her team conducted a similar study looking at MRIs of the knees of 230 individuals without knee pain. It shouldn’t surprise you to know that her research shows that 97% of the knees demonstrated at least one abnormality (Horgan 2020). Here are some of the other findings:

57% of the individuals had cartilage abnormalities
48% of the individuals had bone marrow abnormalities
30% of the individuals had meniscus tears
19-31% of the individuals actually demonstrate swelling and edema
6-21% of the individuals had abnormal changes in their tendon

Let that sink in for a second that almost 97% of the knees had at least one abnormal finding, yet these participants had no knee pain at all (Horgan 2020).

Research article #4:

This was a beautifully designed research study done by Dr. Rodrigo Barreto, PT, PhD which looked at individuals with ONE sided shoulder pain. Dr. Barreto and his team looked at 123 MRIs comparing left and right; here is what they found:

"Results: Abnormal MRI findings were highly prevalent in both shoulders. Only the frequencies of fullthickness tears in the supraspinatus tendon and glenohumeral osteoarthritis were higher (approximately 10%) in the symptomatic shoulder according to the surgeon’s findings.

Conclusion: Most abnormal MRI findings were not different in frequency between symptomatic and asymptomatic shoulders. Clinicians should be aware of the common anatomic findings on MRI when considering diagnostic and treatment planning.” (Barreto 2019)

Rotator cuff tendinopathies, rotator cuff tears, and AC joint abnormalities were found on BOTH shoulders despite having pain on only one side (Barreto 2019).

Research article #5:

One of the common questions we receive is do these findings actually hinder performance - good thing there are studies done on elite level athletes within this context. Dr. Christopher Sy Lee, MD who is the head physician for the USA National Indoor Volleyball Team conducted research with his team on the prevalence of “abnormal” findings on elite level volleyball players (Lee 2019). These participants did NOT have any pain and this is what they found (Lee 2019):

88% had rotator cuff tendon abnormalities
65% had partial rotator cuff tears
23% had labral tears

These are “abnormal” findings in elite level athletes without any symptoms which are often attributed to pain, yet these individuals are not only pain-free but are also performing at a very high level.

Why should this concern you? A lot of times, healthcare providers rely on imaging in order to dictate their treatment plan. The problem arises when the treatment plan is built around a finding that may not even be relevant to your symptoms. I mean, take a look at the numbers and how many people are presenting with these “abnormalities” but experiencing no symptoms.

While there are certainly cases when imaging is warranted, often times they are only picking up things that may have actually been there before you even started experiencing any symptoms. One of the thoughts to take into consideration is that we usually don’t have an image of how our particular body part looked BEFORE.

There are multiple other studies that provide similar conclusions but we are keeping them to a minimum to decrease the length of this blog. If you are interested in looking at more, simply do a Google search “asymptomatic mri research x” and replace x with any body part such as hips, elbows, shoulders, etc.

Dr. Gilbert H. Welch who wrote the book ‘Overdiagnosed: Making People Sick in the Pursuit of Health’ practices in Vermont, is a leader in cancer research and a professor at Dartmouth University, said it best during an interview on NPR when speaking about imaging:

”The problem is, when you deploy them on a normal population, the well population, you also find a lot of abnormalities. Almost all of us harbor meniscal tears in our knee, whether or not we have knee pain. We have discs popping out of our back even - whether or not we have a back pain. So the problem is they identified so many abnormalities in normal people, you don't know what to do. But all of the pressures on us are to go ahead and to treat those people.” (Talk of the Nation, NPR)

Well, if my MRI and the findings aren’t the problem then why am I suffering? The truth is that no one can say for sure without an assessment. An assessment that looks at your lifestyle factors (social, psychological), functional capacity (movement, strength, endurance), training (proper programming and progression), and more. There’s a lot that we have to consider and it’s not something that can be derived from imaging.

The main takeaway is that you can have improvements in pain and function without changes in structure. None of the things that show up on your imaging actually have to change in order for you to see differences in those aspects of your life. There are literally people walking around that have the same findings on their MRI as you with none of the symptoms - so are these findings actually abnormal?

These people go on to live normal lives and they return to the activities they love without any problem. They improve their function, and they improve their pain yet never see a change in their “abnormal” imaging.

References:

1.) Brinjikji W, Luetmer P, Comstock B, et al. Systematic Literature Review of Imaging Features of Spinal Degeneration in Asymptomatic Populations. American Journal of Neuroradiology. 2014;36(4):811-816. doi:10.3174/ajnr.a4173.

2.) Culvenor AG, Øiestad BE, Hart HF, Stefanik JJ, Guermazi A, Crossley KM. Prevalence of knee osteoarthritis features on magnetic resonance imaging in asymptomatic uninjured adults: a systematic review and meta-analysis. British Journal of Sports Medicine. 2018;53(20):1268-1278. doi:10.1136/bjsports-2018-099257.

3.) Horga LM, Hirschmann AC, Henckel J, et al. Prevalence of abnormal findings in 230 knees of asymptomatic adults using 3.0 T MRI. Skeletal Radiology. 2020;49(7):1099-1107. doi:10.1007/s00256-020-03394-z.

4.) Barreto RPG, Braman JP, Ludewig PM, Ribeiro LP, Camargo PR. Bilateral magnetic resonance imaging findings in individuals with unilateral shoulder pain. Journal of Shoulder and Elbow Surgery. 2019;28(9):1699-1706. doi:10.1016/j.jse.2019.04.001.

5.) Lee CS, Goldhaber NH, Davis SM, et al. Shoulder MRI in asymptomatic elite volleyball athletes shows extensive pathology. Journal of ISAKOS: Joint Disorders & Orthopaedic Sports Medicine. 2019;5(1):10-14. doi:10.1136/jisakos-2019-000304.

6.) Is Preventive Medicine Actually Overtreatment? NPR. https://www.npr.org/2011/02/11/133686016/Is-Preventive-Medicine-Actually-Overtreatment. Published February 11, 2011. Accessed June 2, 2020.