Be Careful What You Wish For: Demand an MRI from Your Doc, and Your Problem Might Get Worse


MRI ScanFirst off, nobody is telling you not to get an MRI (Magnetic Resonance Imaging), CT Scan, or anything else, but it is probably time somebody -and it might as well be me- lets you know some of the downsides of getting unnecessary imaging. You may not like this, but then again, you don’t like brussels sprouts, either.

Regarding the term “Musculoskeletal:” For the purposes of this article, it refers to non-disease aches and pains. Those pains in your knees, cricks in your neck, back pain, etc, as long as it is not caused by an illness of some sort.

Whenever something is “wrong” with us we all want to know what it is. When it comes to musculoskeletal issues the exact source of a pain and dysfunction can sometimes be unclear, mainly due to the fact that there are many contributing factors. The main point of this article is to take a look at the problems that can arise when we (that is, we at the confluence of the patient/provider interaction) cause when pushing for pictures of our bodies that we don’t need.

Let’s start with a small thought experiment:

Imagine you are standing at a 45 degree angle to your largest mirror and you are bent over just far enough that your belly skin is no longer as taut as it can be. Turn on those fluorescent lights. Have we set the mood? Excellent. Now shoot a naked selfie. If you are a man, imagine it is very cold. Now that we all have that image in our heads, attempt to forget it. Good luck, because you suffer from an incurable dose “some things you can’t un-see.”

An MRI is similar. But it comes with a written commentary from a radiologist. Imagine if your naked selfie had a professionally written, dry sounding explanation along with it: “This image is contrasted with a naked selfie performed 5 years ago. The subject’s skin is comparatively looser, and the abdomen is slightly more pronounced. Tan lines visible in earlier imaging have faded, and the subject is now uniformly pale. The buttocks are not fully visualized, but there are obvious gravitational effects to the left side not seen in earlier imaging.” There is just no way to feel good about this, so who the hell would ask for this type of thing unless it were absolutely necessary?

You, my friends, that’s who. Sometimes we are driven to know what we have no need knowing, and in the case of musculoskeletal imaging, we frequently end up with an image that is of little more use that that naked selfie, but even worse, we could end up making our symptoms worse in the process.

What?! An MRI can make my back pain worse, you might ask?

That is not to say that the MRI itself (i.e. the process of shooting the image) is harmful. Going in the tube is generally safe, but there is evidence that the knowledge that comes along with getting an unneeded image can in itself be harmful. One study looked at two groups of people with lower back pain. One group got MRIs early in the process, while the other did not. The results were troubling, especially considering how cavalier providers and patients alike can be about demanding MRIs. On average, medical costs were somewhere around $13,000 more expensive in the early imaging group, and were much higher even in those who were lucky enough to have their symptoms resolve quickly. Far worse than the cost was the effect that early imaging had on the patients themselves. They had longer disability than those who did not get early imaging.¹ The bottom line is this: The simple act of getting an MRI made patients’ symptoms worse.

There are many thoughts on this, but I will share a little clinical experience with you.

MRIs have gotten much more detailed over the last few years, and can now see an incredibly vivid picture of the human body. Even small deviations from an idealized picture of structural integrity can be seen. But we also have a metric shitload of other methods of determining what a patient’s problem is. For example, when a patient complains that a certain body part hurts, say, the bottom of the kneecap, I will perform a number of physical exam tests I use to try and pinpoint which treatment is best to help the patient. Frequently the primary care provider has already ordered an MRI. When I try and explain to the patient that his knee pain is from problem A, he will frequently state that his MRI clearly shows problem B, and that he is frustrated that he keeps getting different opinions. This takes medical care and turns it into a confusing, irritating experience for patients. Here is the kicker, though. Even if there is a perfect, damn-near-infallible gold standard test for condition B (an MRI is a picture, not proof that the structure is painful) and it is clearly negative, the erroneous perception that the MRI is infallible makes it difficult to treat condition A. Are you following? I lost my focus for a second, but I am back. Here is the deal: A small tear in your meniscus is not always painful, but if something else in your knee is painful you need to treat it.

Pathology found on MRI might not be the main issue

Let’s get back to the idea that pathology (the thing a medical type considers “bad”) found on MRI might not be the main issue. Remember when you scraped your knee or elbow as a kid? If there is still a scar there, we could say that the skin is not in its ideal state. But if it doesn’t hurt, who cares? It is not quite as strong as healthy skin, but good enough to get you by. You forget it’s there until you see it, and the next time your forearm hurts, you wouldn’t think for a second to attribute your symptoms to the scar on your elbow. The back is a bit more complicated, as it includes some structures that can refer pain to other areas. Notice that I said structures, not structure. For those whose focus is not in musculoskeletal medicine, the tendency is to call all referred pain nerve–related. This is, by the way, dead wrong.² This type of pain freaks people out, and makes docs and patients ramp up the pressure to get an MRI when they suspect this.¹

This is where the real problems start. Influenced by a desire to reassure the patient, and even likely by pressure from the patient, the doc orders an MRI. Once the results come back we are very likely to see things that are not perfect-looking, but are unrelated our symptoms. A recent study looked at over 1200 people who had NO SYMPTOMS and performed MRIs of their necks. 87.6% of these subjects showed disc bulging -what some people will refer to as a bulging disc. Even the subject in their 20s, over 70%, as a matter of fact, showed bulging discs. These findings became more extreme with age, but remember, NOBODY HAD ANY SYMPTOMS!³ That adds up to a lot of naked selfies people can never forget!

So what is the solution?

A good physical examination, and a large dose of this (Fear and overreaction is a huge component in lower back pain, but I’ll save that for another article). As for the physical exam, one hint is that if the doctor doesn’t watch you move, he is not likely performing a sufficient musculoskeletal exam. If he doesn’t touch you for more than a short time, he is not likely performing a sufficient musculoskeletal exam. Also, if the whole exam took 5 minutes, well, you can finish this sentence, I bet…

The main point is this: If you have a problem that is musculoskeletal in nature, it can usually be identified without any advanced imaging (e.g. MRI). On the other hand, advanced imaging cannot treat your condition, but it can most certainly make it worse.

Here is what I would suggest for musculoskeletal conditions: Find a good Physical Therapist (if you are not in a state with direct access to a PT, pressure your doc to refer you to one, but don’t push for an MRI) and have that person examine you. We look at the whats, whys and hows of painful conditions of the musculoskeletal system. A PT will do a thorough examination and create a treatment plan for your problem, and doesn’t need an MRI to do it. If that doesn’t work, then it might be time to look elsewhere. There are, of course, other providers who perform examinations of this sort, and a good examination is far superior to imaging in many cases.

Sources (To my colleagues who are screaming about format. I am working on easily importing AMA style into this blog. Each superscript links directly to each source, and if that does not ease your troubled minds, please click here:

  1. Webster, et. al: Iatrogenic Consequences of Early Magnetic Resonance Imaging in Acute, Work-Related, Disabling Low Back Pain.
  2. Nakashima, et. al: Abnormal findings on magnetic resonance images of the cervical spines in 1211 asymptomatic subjects.

Leave a comment

Your email address will not be published. Required fields are marked *