Great Falls, Montana

Email Us


Follow us :

Issue 5

Table of Contents

Want to download the PDF?

Prevalence of knee osteoarthritis features on magnetic resonance imaging in asymptomatic uninjured adults: a systematic review and meta-analysis

Dr. Joe Camoratto PT, DPT

Key takeaways:
  • The prevalence of knee osteoarthritis features on MRI in otherwise healthy, asymptomatic, uninjured knees is high— up to 43%
  • Prevalence of knee findings on imaging almost always increases with age
  • It is inappropriate to use imaging findings as a sole determining factor in clinical care


Say cheese!

This month we are going to be focusing on the wide and daunting topic of imaging in medicine! Not one but two articles that cover the nuance, challenges, pitfalls and role that things like x-rays and MRIs play when considering healthcare!

We have so much to cover so let’s get to it!

MRIs have been on the up and up for about 20 years now, currently tipping the scales with a massive 30% increase in use! Due to this large increase in imaging, we should probably figure out if the utilization of this imaging (and spending to the tune of $14 Billion PER YEAR) actually tells us information that is pertinent to managing a patient’s pain and disability.

Now I know what you’re saying:

“Joe, we have more advanced technology now. Of course we are going to be using it more. Why wouldn’t we look deeper and with this better tech our goal is to find the root cause of people’s pain?”

Well I’ll tell you why: because it’s not that simple *insert The Notebook video here*

Follow me on a brief walk down Pain Lane

Pain is complex and we are constantly updating our idea of what it is, where it fits into our experience and how to manage it. The idea that we can just look with an image and see a metaphorical splinter that needs to be pulled out of the tissue for all of someone’s symptoms to go away comes from what is called the Biomedical Model of Pain.

It’s super old. Like 1400s old. This model describes the following: For each and every symptom that someone experiences, there must be a driver for that. Easy peasy. Back hurts? Disc herniation. Headache? Brain is too big. Knee pain? Arthritis. Now, 600ish years later, we would call the Biomedical Model, reductionist. Meaning, it’s too simple of a take on too complex of an experience, in too complex of an organism.

What has taken its place? The BioPsychoSocial Model of Pain. Again, briefly: pain is influenced by many factors. Factors that come from a biological, psychological and social realm. This is also not to be mistaken as a trichotomy. All of these realms ALWAYS bring contributing factors to pain perception (sensation?) all at once. They cannot each exist individually.

So now that we have that out of the way, let’s proceed into the weeds about this systematic review and meta analysis which is cool and applicable to clinical practice on a daily basis. Time to go over the study itself. The title is Prevalence of knee osteoarthritis features on magnetic resonance imaging in asymptomatic uninjured adults: a systematic review and meta-analysis and it took 63 studies with 5397 knees of asymptomatic individuals (individuals without symptoms) and looked at imaging they had done.

Specifically they looked to see how many knees had; articular cartilage defects, meniscus damage, osteophytes and bone marrow lesions.

On a quick category by category basis, they found:



All but one category had an important disclaimer on it:

“The prevalence of _______ significantly increased with age”

The category that was left out was bone marrow lesions. Seems the younger folks sustain more bone marrow lesions due to weight bearing sports participation. I know that it might not need to be pointed out again, considering that if you’re reading this, you’re incredibly intellectual, but these prevalences were all in people with ASYMPTOMATIC knees. So where does that leave us? Scratching our heads about the relationship between imaging findings, symptom presentation and where imaging fits into clinical practice if it can even tell us if someone is hurting or not based on the picture.

It would be bad if we didn’t discuss the dark side of this imaging phenomenon. Radiation aside, there are some huge negatives to taking as many images as the medical system is.

First, it costs an absolute ton. Fourteen billion dollars a year is no joke, especially when we just saw how it’s not even that great of a tool at showing us what the “pain driver” is. Usually when money is spent, ideally it would be on something that is going to really hit the nail on the head regarding things that will alter medical care. I mean, that’s the whole point right? How is this going to change how we are treating the person in front of us? Anecdotally, this is my main point that I drive home when holding the conversation about x-rays or MRIs in the clinic.

“Let’s say we order an MRI for your knee. What happens after we get the results?

Unless we are referring you out for surgery, you’re probably just going to continue with what we are doing here in rehab. It’s unlikely that there is going to be something on that MRI that is going to vastly change your care, and if there was something like that going on, I would probably be able to pick up on it just by talking to you and about your symptoms and presentation” – Me

Nine times out of ten this is met with “yeah, but I just want to see.”

What a perfect transition into our next point: the nocebo effect. If you are unfamiliar with the nocebo effect, it’s similar to the placebo effect in that an expectation influences an outcome. More specifically, a negative expectation influences an outcome negatively. In this context, just the knowledge of something that isn’t an absolute perfect picture of “health” existing on the imaging that was obtained can and does influence how people perceive their symptoms, the resiliency of their body and their ability to tolerate the world around them. To relate this back to the current article, the 43% of 40 year olds that have an articular cartilage defect but no symptoms are more likely to develop symptoms after having seen that imaging. The power of suggestion!

This isn’t the first or the last time a patient will unintentionally shoot themselves in the foot out of curiosity. It’s not their fault either. This stuff is difficult and it very quickly gets into the “woo” realm when we start talking about perceptions and beliefs and drift away from the mechanical world.

How are they shooting themselves in the foot? It’s called a domino effect.

> Patient has a symptom and asks for imaging

> Imaging shows age related changes

> Provider sees age related changes on radiology report

> Patient is referred to orthopedics/pain management “to get it fixed”

> Patient undergoes surgery on a thing that might not have been contributing to their symptom

At that point, consider Pandora’s box wide open.

So if this stuff isn’t what is driving the symptoms, then what is it? There is a great movement going on in medicine right now asking clinicians to start calling stuff like the changes seen on the imaging in this study, age related changes. Considering that the prevalence increases with age, it makes total sense.

“This increase of approximately 10%-15% per decade for osteophytes and cartilage defects, and 3% per decade for meniscus tears, suggests that these features reflect normal aging related changes”

Moving away from degenerative changes or abnormal findings can help to lighten the blow when the radiology tech calls the patient and tells them about the radiologist’s impression. Using a comparison to gray hair or wrinkles is a favorite of mine. Then you get to field the comment, “I guess I’m just getting older”.

“What else would you be doing?” – Me in my head

The truth is, you are going to get patients who get imaging ordered for them. If not by you, then by other providers. It is best to, when appropriate, downplay the severity of the findings to avoid the nocebo effect as it has a few loose ends of its own in the form of decreased physical activity and quality of life. I certainly don’t want this suggestion to be misrepresented as “lie to your patient”, but there is a benefit in presenting the findings with less certainty in their ability to drive symptoms. You can sleep soundly doing this because, well, we are truly uncertain of this topic.

Let’s put a bow on this.

If the goal of patient care is to improve function and manage pain, and there isn’t a well defined relationship between pain and imaging, why get the imaging knowing that there are harms and costs that don’t need to occur? Commission bias can really take hold when sitting in front of a patient. It can be tough not to give in to the pressure of someone who is just trying to advocate for their health.

Always consider the unintended consequences of what happens when they see the very likely present normal aging changes. Are they the type of patient who isn’t going to internalize that radiology impression? Or are they going to hit up Web MD and see what the internet tells them to do.

  1. Culvenor AG, Øiestad BE, Hart HF, et al. Prevalence of knee osteoarthritis features on magnetic resonance imaging in asymptomatic uninjured adults: a systematic review and meta-analysisBritish Journal of Sports Medicine 2019;53:1268-1278.


Systematic Literature Review of Imaging Features of Spinal Degeneration in Asymptomatic Populations

Dr. Joe Camoratto PT, DPT

Key takeaways:
  • Changing from “degenerative changes” to “normal aging changes” is a more accurate descriptor of imaging findings
  • Up to 37% of asymptomatic individuals in their 20s have disc degeneration
  • Prevalence of findings on imaging in asymptomatic population increases with age


Time to take this party to the lumbar spine, especially considering that low back pain is the leading cause of disability on the globe! What else should we assume the leading disabler on the globe is associated with? Exorbitant health care costs! The tune remains the same here, unfortunately.

As we witnessed in the previous article, things between what is seen on imaging and what is experienced is fuzzy. But that is a good thing because that means that the human is more complex than just an *insert mechanical object here*. As a side note, please refrain from any vehicular, robotic or mechanical descriptions of the human body. So where do we pick up with the lumbar spine where we left off with the knees?

The same general idea persists, even with some of the most renowned spine peoples in the world: “we just need to look harder for the pain generator.” Some are under the impression that our technology just isn’t good enough to see what is really causing the low back pain. This above description is far from what you think though, good reader, especially after having gone through the previous article with a fine toothed comb. If you still need convincing, this is the article for you.

This is one of the greatest hits articles in the current #NewWaveRehab world. As it should be! This article does just about the same exact thing as the last one. It is titled Systematic Literature Review of Imaging Features of Spinal Degeneration in Asymptomatic Populations. It accumulated 33 articles reporting imaging findings for 3110 asymptomatic individuals and bunched all of that data together in one neat super article.

Also just like the last article, it has a list of things it looked for; Disc degeneration, disc signal loss, disc height loss, disc bulge, disc protrusion, annular fissures, facet degeneration and spondylolisthesis. It looked for these changes over a very wide range of ages as well. Starting with 20 years old all the way up into the 80s. The reason, in my opinion, that this is a greatest hit for our younger rehab world is that we can use data like this to educate and reassure other people in their younger decades that even though they have changes on a lumbar image that doesn’t mean that they are destined to just continue downhill from there.

The authors were nice enough to put together a neat table for our eyeballs:

This is the Age specific prevalence estimates of degenerative spine imaging findings in asymptomatic patients.


I’ll have to email them about the whole degenerative vs age related snafu there. What a great table though. Just like the knee changes, almost everything increases with age, asymptomatically.

“Our study suggests that imaging findings of degenerative changes…are generally part of the normal aging process rather than pathologic processes requiring intervention.”

This is a very important line in our career field, as we can point to articles like this (at least figuratively point) and say look, we know that getting older is this seemingly black pit of sickness and dying, but it doesn’t have to mean that we continuously feel worse and worse as the years go by. It gives us as practitioners an opportunity to look at our patients and say “yeah, you are getting older, but we know that these changes happen without you even knowing them.”

The source escapes me right now, but I once heard someone compare the degenerative changes that are seen on the spine (and other body parts) like gray hair or wrinkles. Most people get them at some point secondary to aging. Some earlier than others and with varying levels of severity, but gray hair and wrinkles don’t come with symptoms. Armed with the knowledge that this is just a part of being alive and existing as a biological organism, we can at the very least help to reassure the humans in front of us that they are totally normal to have these changes occurring, if they have them occurring. Even those in their early 20s! “…disk degeneration was found in one third of asymptomatic 21 year olds.”

We also have the opportunity to educate them on the divorce that needs to happen between how we think about symptom presentation and the things that are going on within our bodies or what is seen on the imaging (we won’t even get into the large amount of variation on what is seen on an image based on who the radiologist is and where the radiologist is located).

Medicine and rehab are all about small wins. Planting seeds of thought and doubt about preconceived notions while also showing patients how to continue to move forward given the new context of their lives. Highlighting (in a paraphrased manner) the post hoc fallacy can be a great foot in the door to help them generate their own conclusion with your help in finessing their thinking.

Now for my favorite part of this whole conversation with the patient: the spin we can put on these sorts of chats is that even if there are normal aging changes on imaging, and even if you are having symptoms right now, there is something that can help! We can build a fantastic physical activity plan together that will not only be fun, but can aid in alleviating symptoms and wildly improve your health.

The moral of the story here is that there is an opportunity for not only the patient to become nocebo’d from the findings that they see on imaging, but also you as a clinician. This applies to any imaging in any body part as well. Keep in mind a couple big takeaways as you’re being confronted with abnormal changes on imaging: how is this going to change patient care and how likely is it that this finding was there before the symptoms presented themselves?

Answering these two questions can begin to point you and your patient in a direction forward, whether that be rehabilitation or referrals to someone that might be more appropriate. Either way, don’t let imaging in and of itself be a sole determining factor in your interventions and recommendations. The way forward is usually much fuzzier than that.

  1. Brinkikji et al Systematic Literature Review of Imaging Features of Spinal Degeneration in Asymptomatic Populations American Journal of Neuroradiology Apr 2015, 36 (4) 811-816; DOI: 10.3174/ajnr.A4173