I've just injured my back - Do I need an MRI scan?

Aside from getting asked this question on a daily basis in clinic, recent press and media attention made me feel the need to give my point of view. 

MRI scans have in many ways revolutionised medicine and enhanced patient care. The newest systems provide detailed images of anatomy that even 5 years ago lacked true clarity. They've also become increasingly affordable to many, and even on Harley street it is now possible to get a scan for £200 if booked off peak. Unlike CT scans and XRays their direct action is safe, as they don't expose the patient to a dose of radiation. However, many respected voices in the medical world are warning that inappropriate use of MRI to get a low back pain diagnosis is actually harming the healthy (1). 

To understand the problem it's important to understand that most acute low back pain resolves within 4-8 weeks if good management advice and reassurance is given by a medical professional.

The guidelines are in essence:

1. Use non-steroidal anti-inflammatory medication

2. Refer to physical therapy/physiotherapy

3. Avoid early MRI unless exceptional cases involving red flags

4. Avoid habit forming medication

5. Avoid early referral for injections or surgery

Ok, so what's the problem? Well, point three of the guidelines isn't being followed and this is leading to "overdiagnosis" of simple back sprains (2). These sprains are given labels like "slipped disc" and "degenerative disc disease" when really the injury is as benign as an ankle sprain. Let me explain further. A recent study (1) showed that in a normal pain free group of people given an MRI scan, 90% had degenerative discs and 30-40% had disc bulges, joint degeneration, and bone spurs. Just to be clear this is the pain free, normal group. The group that had pain related to pathology was tiny at 1-2% of the population. 

So if a medical professional advises an MRI scan and it comes back with "degenerative disc disease" you now have a label, in fact you have just been told you have a disease. This feeds anxiety and fear, creating what's called a nocebo effect. This is the opposite of the placebo effect, and describes any case where putting someone in a negative state of mind has an adverse effect on their health.

This asks the wider question: what are the boundaries that define an illness or pathology? And what is in fact normal and not going to cause longstanding incapacity? If the data tells us that 98% of people with acute low back pain alongside structural changes on MRI have just suffered a sprain, then we clearly need a more careful interpretation of these results. Definitively, low back pain should never be diagnosed from a scan alone.

Overdiagnosis as a concept may sound counterintuitive and many, many peoples lives have benefitted from appropriate use of MRI, but this final statistic delivers the strongest message: In America, people referred for MRI for routine back pain are eight times more likely to have surgery. Now that's a statistic worth changing.

Tom Lowes B.Ost MSc (Sports Med)
Registered Osteopath - Sports Injury Rehabilitation

Twitter @SportsMedOsteo       

 

Here's a great link to a podcast on this subject: Ray Moynihan 

Thanks to Karim Khan @BJSM_BMJ

1. McCullough et al. Lumbar MR imaging and reporting epidemiology: do epidemiologic data in reports affect clinical management? Radiology. 2012 Mar;262(3):941-6. doi: 10.1148/radiol.11110618.

2. Mafi et al. Worsening Trends In The Management and Treatment of Back Pain, AMA Intern Med. 2013;173(17):1573-1581