The Frontal Cortex

MRI and Back Pain

Posted by Jonah Lehrer on February 7, 2009
(19)
More »

One of the case studies I use in How We Decide when discussing the dangers of information overload concerns the diagnosis of back pain. Before the introduction of MRI’s in the late 1980s, doctors were forced to rely on X-rays when diagnosing back pain. X-rays provide doctors with a limited amount of information, since they only reveal the bones and spinal column. As a result, back pain remained a mostly mysterious phenomenon, and most patients were prescribed bed rest. Nevertheless, this simple treatment plan was still extremely effective. Even when nothing was done to the lower back, about 90 percent of patients with back pain managed to get better within seven weeks.

However, that all changed with the introduction of MRI. Within a few years, the MRI machine became a crucial medical tool. It allowed doctors to look, for the first time, at stunningly accurate images of the interior body. The medical profession hoped that the use of MRI would revolutionize the treatment of lower back pain. Since doctors could finally image the spine and surrounding soft tissue in detail, they should be able to offer precise diagnoses, locating the aggravated nerves and structural problems that caused the pain in the first place. This, in turn, would lead to better medical care.

Unfortunately, MRI’s haven’t solved the problem of back pain. In fact, the new technology has probably made the problem worse. Here’s the Well blog on the NY Times, summarizing a new study published in The Lancet:

Researchers from Oregon Health and Science University in Portland reviewed six clinical trials comprised of nearly 2,000 patients with lower back pain. They found that back pain patients who underwent scans didn’t get better any faster or have less pain, depression or anxiety than patients who weren’t scanned. More important, the data suggested that patients who get scanned for back pain may end up with more pain than those who are left alone, according to the report published this week in the medical journal Lancet.

Why do MRI’s often lead to worse medical outcomes? The machine simply sees too much. Doctors are overwhelmed with information, and struggle to distinguish the significant from the irrelevant. Take, for example, spinal disc abnormalities. While X-rays can only reveal tumors or problems with the vertebrae, MRI’s can image spinal discs⎯the supple buffers between the vertebrae⎯in meticulous detail. After the imaging machines were first introduced, the diagnosis of various disc abnormalities began to skyrocket. The MRI pictures certainly looked bleak: people with pain seemed to have seriously degenerated discs, which everyone assumed caused inflammation of the local nerves. Doctors began administering epidurals to quiet the pain, and, if the pain still persisted, would surgically remove the necessary disc tissue.

The vivid images, however, were misleading. A 1994 study published in The New England Journal of Medicine imaged the spinal regions of ninety-eight people with no back pain or any back related problems. The pictures were then sent to doctors who didn’t know that the patients weren’t in pain. The end result was shocking: two-thirds of normal patients exhibited “serious problems” like bulging, protruding or herniated discs. In 38 percent of these patients, the MRI revealed multiple damaged discs. Nearly 90 percent of these patients exhibited some form of “disc degeneration”. These structural abnormalities are often used to justify surgery and yet nobody would advocate surgery for people without pain. The study concluded that, in most cases, “The discovery by MRI of bulges or protrusions in people with low back pain may be coincidental.”

In other words, seeing everything made it harder for the doctors to know what they should be looking at. The very advantage of MRI⎯its ability to detect tiny “defects” in tissue⎯turned out to be a liability, since many of the defects were actually a normal part of the aging process. This is the danger of too much information: it can actually interfere with our understanding. We confuse correlation with causation, and make theories out of coincidences. We latch onto medical explanations, even when the explanations don’t make very much sense.

As I note in the book, medical experts are now encouraging doctors not to order MRIs when diagnosing back pain. A recent report in The New England Journal of Medicine concluded that MRIs should only be used to image the back under specific clinical circumstances, when doctors are examining “patients for whom there is a strong clinical suggestion of underlying infection, cancer, or persistent neurologic deficit.” In the latest clinical guidelines issued by the American College of Physicians and the American Pain Society, doctors were “strongly recommended…not to obtain imaging or other diagnostic tests in patients with nonspecific low back pain.” In too many cases, the expensive tests proved worse than useless. All of the extra detail just got in the way. The doctors performed better with less information

(19)
More »

Comments

  1. #1 Fertanish
    February 7, 2009

    I think it ties in nicely to the chapter regarding the AI behind computer chess program that learned how to play. The program started learning by losing, but learned from those losses to become a nearly unbeatable gamer.

    Medical science has a lot more variables and higher risks for failure (as well as different levels of failure, not a binary “win/loss” the way a chess game has), but in many ways the learning is same. The MRI opened an expansive data set to help identify/isolate the cause of back pain (and, of course, other things). Eventually it was learned (I suspect with some arguments remaining) that the data was mostly irrelevant. So, now, the data is simply not reviewed, lesson learned, and new decision making process implemented.

    However, when a new technology is presented that may produce even more data that can help identify causes of back pain, we will still explore its information and likely make a number of mistakes in the process of translation. There will be lessons to be referenced from what was learned during the evolution of MRI as a diagnostic tool. But not exploring that data, and likely at times making the same mistakes, is not an option if we want to progress. Like the chess game, with a finite (albeit large) number of moves and decisions, we will continue strive for perfection in diagnostics and treatment despite the seemingly infinite amount of data to evaluate.

  2. #2 Comrade PhysioProf
    February 7, 2009

    Fascinating post, holmes!

  3. #3 Steve
    February 7, 2009

    I enjoyed reading the piece in the Sunday Times Magazine. I’m an educator and staff developer who uses cognitive psychology and neuroscience in presentations to help educators understand how and why kids learn, and to make good decisions when it comes to learning. This material you post is very helpful. Thanks.

    Steve in the Adirondacks

  4. #4 Arj
    February 7, 2009

    excellent post on a medical condition common to 10′s of 1000′s each yr., who spend LOTS of time/money on doctors, tests, and treatment often yielding no better results than common sense (and proper exercise!) will lead to.

  5. #5 Alex
    February 7, 2009

    I am a chiropractic student currently taking my advanced imaging coursework. We are very clearly taught that the amount of degeneration visualized on imaging whether it be x-ray or MRI, does not correlate at all with the severity of pain/symptomatology. In fact, our protocol typically will consist of a trial period of conservative care (unless trauma, significant past history, degree of symptom progression, older age..etc) which if the case doesn’t improve after a 2-6 week period depending the condition, then we may elect for further imaging. Put clearly, whether we have imaging or not, the treatment is often the same. MRI is then utilized to determine a possible surgical consultation.

    Most of the pain associated with degeneration is an inflammatory/biochemical event which can be mediated through psychological and dietary measures. In fact the more painful the condition is, the better the prognosis because it shows that the patient’s immune system is physically responding to the lesion. Patients in these circumstances should only elect surgery (according to the research) in cases such as a pro athlete who needs to get back on the field because he’s getting paid millions to be able to perform or the case of a blue collar worker where they can’t afford to take the necessary time off for their condition to heal or lack the home support resources in order to manage a painful condition.

    All jargon aside, it is my expectation that the types of physicians examined would be orthopedic surgeons who upon looking at the MRI found some of these incidental findings that may or may have not been confirmed to be causing the patient’s problem, and because of the tools they have, elected (if you have a hammer, everything’s a nail) to put these patients under surgery…..the better surgeons will work with chiropractors and other types of professionals in order to manage a patient conservatively because the stats behind back surgeries are very poor (albeit slowly getting better), but this is partly because these surgeons are being extra selective for the sake of their success numbers.

    There’s a review article that was conducted in the February 2008 issue of Spine that categorized all treatments as “should strongly be considered, should not be considered, might give benefit”, etc.

    Chiropractic manipulation showed, “some benefit”, while most everything else including surgeries, corticosteroid injections, ibuprofen/NSAIDs, were in a category entitled something on the lines of “should not be considered” based on reviews of the evidence in the literature. My guess was these patients who had MRI’s were more likely to get those other therapies that are under the “should not be considered” category and therefore had poorer outcomes.

    Summary: Chiropractic/conservative care should be a first choice whether MRI confirms or not, we have relatively modest benefits when it comes to pain (this is not to say function, and other outcomes aren’t better), but are better than anything else out there.

    As an aside, if you look at the degree of insurance reimbursement, the therapies with the least evidence get the most reimbursement, and the therapies with the most evidence get the least reimbursement.

    How many times have you heard a patient with low back pain say “well the chiropractic/physical therapy was working, but my insurance ran out and I couldn’t continue so now i’m back almost where i started….”

  6. #6 Gen
    February 7, 2009

    First-time autopsies are called to my mind while reading this. It is not surprising that a skilled professional in their field would be mislead with such detailed visual representation, the likes of which they have never seen before. Fertanish is exactly right in that new diagnostic technology will always lead to medical insight, after mistakes are made, and perhaps time and time again.

    Not only did the MRI technology bring to light the effects that natural aging can have on the spine, but that the haste in which doctors diagnose and treat can often have drastic consequences. Anyone in the field will be of the “fixer-type” personality, wanting to make right what is wrong with their patients. However, the urgency felt may lead a professional to misdiagnose, over-medicate, and operate unnecessarily. The more often these incidents are brought to attention, the more likely doctors/surgeons/etc will slow themselves in jumping to conclusions, and acting rashly.

    Thank you for this article! I hope it spreads, far and wide.

  7. #7 Donna B.
    February 8, 2009

    Excellent article. My father (age 85) has significant back pain and it occasionally “freezes him up” where he cannot straighten or move. Being otherwise healthy, he wanted this fixed as he thought (correctly, IMHO) that if he couldn’t be active he would die sooner.

    He’s in physical therapy now after bed rest for a week and says he never felt better (he says this a lot, so take it with a grain of salt).

    He has a half-sister, 15 years younger who has had 3 surgeries on her back. Her three children and her husband have all had at least one surgery. They all go to the same doctor. No comment here, just anecdotal data, that sometimes has me wondering. And has convinced me to never go to that doctor. Yes, all these surgeries happened since MRIs came to popular use.

    Since none of her siblings, half or full, nor her parents had or have back problems that do not result from injury, well… much as I love my aunt and cousins, I’ve really got to wonder about their current disabilities – always there, or induced by surgery? My father’s back problems are likely due to a decades ago injury that left him with one leg a bit shorter than the other.

    As a mere science/medical junkie I’m constantly amazed at the things medicine can accomplish, but even more amazed at what some people think it can accomplish.

  8. #8 Rachael
    February 8, 2009

    Well that’s sadly appropriate. A family member of mine was diagnosed with two slipped discs a few weeks ago, and proceeded to take to bed rest as a preventative measure (having never experienced back pain before).

    The interesting part about back pain is that there is no biomechanical correlate to the pain, and there isn’t too much that a doctor can suggest to alter the course of pain once a patient starts experiencing pain. My favorite example is the lumbar support belts that people who work at home improvement stores wear. Insurance companies often require that the stores have their employees wear these devices, since anecdotal reports suggest that wearing one results is a decreased in risk of back injury. But it’s pretty clear from the field of biomechanics that the support systems do absolutely nothing to support muscles or alter posture. The only thing they might do is serve as a reminder (by the constant sensation of cloth) to “lift from the knees” and not bend ones back, etc. In other cases, patients believe the back belt will support them, so they attempt to lift things they shouldn’t and injure themselves anyway. It’s also clear from epidemiological studies that any benefit gained from lumbar support systems is too small to be measured, yet insurance companies continue insisting that their clients institute lumbar belt. The supposed benefit of lumbar support is a wash.

    I saw a great talk by a biochemics professor on this subject – an hour of trying to convince his audience that a career of research in his own field (biomechanics of the lumbar system) offers zero explanation for back pain. Talk about hocus pocus.

  9. #9 Alison
    February 8, 2009

    Another possibility is that a lot of back pain is psychosomatic. If you read Dr. John Sarno’s book on this, he makes a good case for the idea that the brain can induce pain pretty much anywhere (via restricted circulation), and if you get an MRI saying that there are all sorts of physical problems with your discs, etc., then your brain will have a ripe field for ‘making up’ some appropriate pain to go with it. He states in his book that most disc problems that show up in imaging don’t even correlate with the location of the pain. On the other hand, if the patient can be convinced (in the absence of any legitimate injury beyond normal aging) that there is nothing actually wrong with them, then they may be released from their pain, as long as they work through whatever stress is triggering them. Admittedly, this is getting into some fairly uncharted territory, but IIRC, recent research has confirmed that the brain can selectively reduce bloodflow to areas of the body, thereby inducing pain. (This was from a recent issue of Scientific American Mind, though I’m unable to find the reference at the moment. I think it was research related to phantom limb syndrome.)

  10. #10 Phil
    February 9, 2009

    2 things occur to me.
    1. the comments and maybe the article are not really from a “world medicine” point of view.
    Insurance seems to have as much a role as evidence.
    2.Do we medics have a problem saying “don’t worry, keep quietly active, the pain will get better”. Also do the patients have a problem hearing that?

  11. #11 cube
    February 9, 2009

    Too much information from an MRI. Who knew? Certainly speaks volumes for adopting a conservative approach to medical treatment.

  12. #12 Amanda
    February 10, 2009

    ALso to be noted is a problem with how doctors learn anatomy, most cadavers they study aren’t the healthiest of examples, and most textbooks show bodies of similarly imperfect nature. There is no real normal out there so any minor differences from the specific cadaver or textbook studied is going to seem strange to the doctor in question.

  13. #13 richard scott
    February 13, 2009

    As an Osteopathic orthopedic surgeon and semi retired spine subspecialist I have, like many of us, spent the last forty years attempting to figure out back pain. Every new imaging modality from myelograpjy, discography, ct scans and mri’s we hoped would show us the anatomic cause of back pain. Most of us know that imaging rarely or not always accurately shows us where the pain generator is. The more sophisticated the imaging modality, the more likely it is to show too much meaningless data. From the onset of ct scans through the present high tech ct sscans and mri’s I see daily patients who either bring an mri ordered by the primary care physician with the idea that it will induce us to operate, or they will demand an mri to “prove” what is wrong.
    There was an interesting editorial on mri’s a few months ago which noted the obverse. It implied that many mri’s are read by poorly trained radiologists and that mri of spine should be interpreted by sub speciaiist. The case I believe indicated that the re-view of the imagings led to urgent surgery to prevent paralysis. As a cynic I would possibly look at it the other way; sometimes surgeons see an image and imagine the worst. The physical examination is of equal importance to the imaging and I doubt that even a first year radiology resident would “miss” significant cord compression or myelomalacia.
    We now know the same over call with the 64 bit cts cans. Imaging artifacts often look like pathology.

  14. #14 charlotte
    February 13, 2009

    I think most doctors (especially neuros and orthopods) are aware of the shortcomings in using MRIs as a diagnostic tool for back pain issues. Almost any medical spine specialist will tell you they’ve seen plenty of mild disc herniation in a patient who presents with debilitating sciatica and plenty of folks who feel only slight discomfort but who nonetheless show severe degeneration of the discs. Conclusion? No one really knows why some people feel the back pain they do. They just don’t. Medically, we’re not there yet; there are too many neurological variables at play. Perhaps the parts of the brain that process pain are overactive; perhaps there are other things going on. It’s a bit of a mystery, that’s for sure.

    I also believe that the medical system is very screwed up when it comes to dealing with pain, especially when it involves prescribing opioids. Virtually no doctor is going to fill a pain meds prescription more than maybe twice without some sort of “definitive’ proof that some kind of pathology exists, hence the MRI (which they already know is probably not going to tell them much). The boards that regulate physicians would be all over them if they failed to provide some kind of medically-approved documentation for the continued use of controlled substances.

    Also, our fear and loathing of narcotics pervades all through the medical system (as well the public) in this country, even when those drugs may be the one thing that actually helps a patient lead a productive life. The fear that the patient may become “addicted” blocks the important fact that supervised narcotic intervention may actually be the best option for chronic pain patients, especially when all other treatment methods fail. There is a difference between dependency and addiction, and again, I think most back pain specialists are aware of this. But what kind of enormous flack would a doctor come under for writing all these prescriptions? The answer: a lot. What sort of flack do they come under for suggesting surgery after other methods have failed? Answer: not much, especially if they’re surgeons. Seriously, who the hell cares if a patient becomes dependent on a reasonable dose of Vicodin to control pain? If they can function in their lives, feel some measure of relief, be present for their families, and generally join the world of the living again, why is that such a terrible thing? What, surgery is better? I think not. Just look at the stats on that for reducing chronic pain. Let’s just say they’re not that great. The cultural mindset that tells us narcotics are “bad” has made us reluctant to use the one thing that actually works for many people. Physicians are gun-shy of narcotic review boards, many doctors and their patients are afraid of “addiction,” and just about everybody’s worried about drug-seekers.

    We’re all in a tizzy about this when what we really need to do is just plain get over it. People in pain want some sort of relief. That’s what sent them to the back specialist in the first place. Prescribing some NSAIDS and rest is a fine first start, but when that doesn’t work, and the epidural doesn’t work, and the stim treatment doesn’t work, and the steroid treatment doesn’t work, and the physical therapy doesn’t work, and the chiropractic treatments don’t work, and the acupuncture doesn’t work and finally the surgery doesn’t work, well, then what? Sorry, we’ve done all we can do for you? It must be all in your head? Nice. Helpful, too.

    After years of dealing with sciatica so bad he couldn’t walk, or sit (literally), my husband went through every treatment in the book, including two failed lamenectomies and ultimately culminating with a spinal fusion with instrumentation performed by one of the doctors cited in your book, Dr. Eugene Carrageee. None of it worked. You want to know what did? The lowly, much maligned Vicodin; that’s what worked. He uses what he needs when he’s in pain, he uses nothing when he isn’t. He’s okay now, not perfect, but he’s once again a functioning person in the world. Just simply knowing that the drug is available to him if he needs it helps stave off the fear and anxiety of dealing with recurrent pain. The answer was so simple, and it was the whole time. Everything else was just a painful, useless, expensive detour. His case is not everyone’s case, and I’m not slamming what works for other people, but their are plenty of folks out there who simply just need some pain meds to help them cope, and then they’ll do the rest of the work themselves. Once we gave up on the medical establishment (especially specialists) and found a sympathetic doctor who wasn’t afraid of his prescription pad, life has gotten a whole helluva easier for us, lemme tell you.

    Sorry for the rant, but I really do think part of the reason physicians order MRIs is not because they think it will reveal something something particularly diagnostic, but more so they can cover their medical asses when they reach for their triplicate forms and also so that their patients will think they’re “doing something.” Whatever. I just think that many times, the answer is sitting right in front of these doctors, they’re just too cowed by their cultural biases and professional constraints to do the one thing they know will work.

    Just my two cents.

  15. #15 Flymises
    February 14, 2009

    “…the brain can induce pain pretty much anywhere (via restricted circulation), and if you get an MRI saying that there are all sorts of physical problems with your discs, etc., then your brain will have a ripe field for ‘making up’ some appropriate pain to go with it. He states in his book that most disc problems that show up in imaging don’t even correlate with the location of the pain.”

    Interesting. That may mean that, after seeing the scan results, and after the ‘faulty’ spinal plates have been shown to patients, the patient’s pain may move to the area where the apparent problem is, or it may spread to that area. This could be tested!
    I doubt there would be such an effect – it would require that the brain makes an unconscious activation of vasal constriction in a specific area of the body based upon visual, or perhaps other external, information. Nevertheless, it would still be worth testing to at least eliminate the possibility!

  16. #16 e-okul
    February 18, 2009

    I enjoyed reading the piece in the Sunday Times Magazine. I’m an educator and staff developer who uses cognitive psychology and neuroscience in presentations to help educators understand how and why kids learn, and to make good decisions when it comes to learning. This material you post is very helpful. Thanks.

  17. #17 ısı yalıtım
    February 19, 2009

    thanks.

  18. #18 Bhetti B
    February 21, 2009

    According to the latest in UK (where an MRI is only ordered if there are red flags e.g. trauma), apparently better outcomes occur if mobility is encouraged within the limits of pain: more people return to work that way instead of taking time permanently off.

    Not sure where the data to back this up might be found, though, but it has been what I was taught by more than one health professional!

  19. #19 Gary Brazzell
    April 19, 2010

    I just wanted to make a quick note about the glaring factual error in the lead paragraph. Bed rest is not “extremely effective” for back pain. Research has proven beyond a shadow of a doubt that bed rest is bad for back pain. Appropriate activity beats bed rest for overcoming back pain. Get to a physical therapist or doctor to learn the activities that will beat your back pain.