Low back pain: dangerous to get early MRIs

 Ordering an MRI for patients within the first six weeks of a new episode of uncomplicated nonspecific low back is associated with much more surgery as well as more opiate prescriptions and higher final pain score (see low back pain early MRI dangerous jgim2020 or DOI: 10.1007/s11606-020-06181-7)

 

Details:

-- 405,965 VA patients were analyzed from July 2015-July 2016

-- mean age 54, 6% female, 82% saw their assigned primary care provider, history of opiate use 20%, comorbidities COPD 11%, diabetes 18%, hypertension 45%, mental health diagnoses 40% (mostly bipolar disorder in 26%, PTSD in 22%)

-- baseline pain score 0 in 13% getting an early MRI/32% not having one, and up to a pain score of 9-10 out of 10 in 6% with an early MRI and 3% without

    -- overall, those receiving an early scan were younger, less likely to have an assigned primary care provider, less likely been seen by their primary care provider, had a higher baseline pain score, were less likely to have had a back pain episode in the previous 24 months, had fewer chronic conditions, and were less likely to have a diagnosis of hypertension or ischemic heart disease in the prior year. Though overall there were not dramatic differences between the groups  [ie, the number of patients was so high that even small differences were statistically significant]

-- 9977 people (2.46%) received a lumbosacral MRI within six weeks of their index visit; 395,718 did not

 

Results (adjusted by propensity and covariate outcomes):

-- lumbar surgery: 1.48% in those with early MRI, 0.12% in those without an early MRI, p <0.001

     -- adjusted relative risk of 12.7 (10.3-15.5) (ie, approx 13-fold risk of surgery in the group with early MRI)

-- prescription opioid use: 35% versus 29%, p<0.001

    -- adjusted relative risk 1.23 (1.20-1.27)

-- last recorded pain score: 3.99 versus 3.87, p<0.001 [ie, the last time the patient saw their primary care provider, their reported pain score was actually a bit higher if they had early MRIs done]

-- total cost: $8082 versus $5560, p<0.001

    -- difference of $138 for inpatient care, $1635 for outpatient care

 

Commentary:

-- this study found that those with early MRIs had more surgery, more opioid prescriptions, more costs, and, of note, more pain at their last outpatient visit

-- low back pain is the second most common symptom leading to a physician visit, with huge and increasing healthcare costs

    -- review of the overall VA registry found 1.17 million primary care visits for nonspecific low back pain from July 2015 to July 2016

-- routine use of MRIs is common, with 16-21% of patients with commercial insurance and 12% on Medicare.

-- guidelines from both the American College of Physicians and the American Pain Society recommend not getting an MRI in the first 6 weeks of low back pain unless there are red flag conditions present (e.g., trauma in the past 45 days, lumbar spine surgery in the past 90 days, neurologic impairment, infections, injection drug use in the past year, autoimmune/inflammatory conditions, cancer, radiation therapy, or congenital malformations in the last five years)

    -- however, studies suggest that 26-44% of spine MRIs are not guideline-concordant

    -- and, other studies have found that surgery was 5-20x more likely in those getting an early MRI

-- One large concern with MRIs of the back is their lack of specificity. For example:

    -- a study looking at MRIs one year after surgery for sciatica found no difference in the MRIs in those who responded versus didn’t respond to surgery. see http://gmodestmedblogs.blogspot.com/2013/03/mri-1-year-after-sciatica.html 

    -- one early study found that 52% of 98 asymptomatic patients without a history of low back pain had abnormal disc bulges, 27% with disc protrusion. see lbp mri in pts without pain nejm 1994 in dropbox, or Jensen MC, N Engl J Med. 1994;331(2):69.)

    -- another study found that MRI on 67 asymptomatic people without any history of low back pain/sciatica had dramatically abnormal findings: in those <60yo 20% had a herniated nucleus pulposus and one had spinal stenosis. In those >60yo, the findings were abnormal on 57% of the scans: 36% had a herniated nucleus propulsus and 21% had spinal stenosis. There was degeneration or bulging of a disc at least one lumbar level in 35% of those 20-39yo, and in all but one 60-80yo (see Boden SD. J Bone Joint Surg Am. 1990;72(3):403)

    -- It seems that MRIs just have “too much information….”, and often we don’t really know how to interpret them clinically...

-- Even in those with sciatica, early recommendations were for imaging only in those with red-flag conditions or when disc surgery is considered, but overall initial treatment should be conservative for 6-8 weeks (see https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1895638/ ).

 

Limitations:

--this was not a randomized controlled trial, so there might well be undocumented differences between those getting vs not getting an early MRI

--for example, this study raises the issue of the "ecological fallacy" whereby the results of group data may be inappropriately assumed to apply to individuals. in this case, the outcome regarding more surgery was actually related to whether the population had a higher rate of early MRIs, and not to whether an individual did. it may be assumed that it was getting the surgery that was related to the increased opiate use, or the higher pain scores. but:

    --was the subgroup getting early MRIs different in unmeasured ways to those not? was the provider's assessment of the overall condition of the patient (their unwritten gestalt) different? did they think the patient was in much more pain or psychological discomfort? was the patient in fact more inconsolable either because of their condition (perhaps augmented by stress, their psychosocial issues, etc), their pain threshold, etc. 

        -- for example, we do know from some studies that provider prescriptions for opiates are not necessarily related to a patient's pain level: eg see http://gmodestmedblogs.blogspot.com/2021/01/opioids-very-high-rate-of-prescriptions.html

    --ie, were the clinical outcomes measured in this study actually related to the early MRI, or was the early MRI a marker for other issues. and the early MRI was just one of the provider orders that also included more opiates. and the increased rates of surgery were actually related to some other issue?? perhaps to the provider's assessment that surgery was going to be needed in the near future led to the early MRI??

--the VA is not a generalizable institution to the overall population:

    --almost all men

    --VA providers are salaried. In the rest of the US system, there are financial incentives to doing surgery (perhaps this explains the fact that in other, non-VA studies, the rate of surgery in those with early scans was 14-22%, v s 1-3% in those not getting the early scans, about 10x the rate of the current VA study)

    --VA clinicians are less likely to practice “defensive medicine”, since the Feds cover malpractice costs

--the database for this study was several years ago, so there may be differences with current practices

--the cost of care was pretty meager compared to what we might expect. Part is likely from the VA system assessment of costs. Part because the costs represented those only for the 6-week period after the MRI, and perhaps there were significant costs later?

so, though this study had significant limitations, overall it does reinforce the importance of NOT doing an early MRI for nonspecific low back pain without red-flag indications. 

--pretty much all of the important outcomes were worse in those getting early MRI exams. 

--conservative management continues to be the cornerstone of therapy: non-opioid pain control, early activity (avoiding prolonged bedrest), exercise/physical therapy, perhaps muscle relaxants (probably best at night, since daytime somnolence might lead to more bedrest/less activity, and they do not confer huge benefit), manipulation therapy/yoga, not prescribing braces unless specific indication, etc

for a blog showing no benefit of opiates vs non-opioid pain meds on chronic back pain: http://gmodestmedblogs.blogspot.com/2018/03/opioids-not-better-for-chronic.html

for a blog finding no benefit for opiates or muscle relaxants for acute low back pain: http://gmodestmedblogs.blogspot.com/2015/10/opiates-for-acute-low-back-pain.html

geoff

 

If you would like to be on the regular email list for upcoming blogs, please contact me at gmodest@uphams.org

 

to get access to all of the blogs (2 options):

1. go to http://gmodestmedblogs.blogspot.com/ to see them in reverse chronological order

2. click on 3 parallel lines top left, if you want to see blogs by category, then click on "labels" and choose a category​

3. or you can just click on the magnifying glass on top right, then  type in a name in the search box and get all the blogs with that name in them

 

or: go to https://www.bucommunitymedicine.org/ , a website from the Community Medicine section at Boston Medical Center.  This site does have a very searchable and accessible list of my blogs (though there have been a few that did not upload over the last year or two). but overall it is much easier to view blogs and displays more at a time.

 

 

please feel free to circulate this to others. also, if you send me their emails, i can add them to the list