Appropriateness of magnetic resonance imaging of the lumbar spine orders for low back pain in a general hospital

Abstract

Introduction

Inappropriate magnetic resonance imaging (MRI) for low back pain contributes to increased healthcare costs. The American College of Radiology (ACR) appropriateness criteria for low back pain is a guideline to reduce these unnecessary MRIs.

Methods

Retrospective study of the MRI lumbar spine performed for the investigation of low back pain in a General Hospital from 2016-2017. Divided into six clinical categories. (1) No Red Flags. (2) Low-velocity trauma, osteoporosis, elderly individuals, or chronic steroid use. (3) Suspicion of cancer, infection, or immunosuppression. (4) Failed conservative management. (5) Prior lumbar surgery. (6) Suspected cauda equina syndrome or neurological deficit. MRI findings are divided into 1) Critical findings of spinal cord/cauda equina compression, metastatic cancer, spinal epidural abscess, and vertebral osteomyelitis. 2) Significant findings of vertebral compression fracture or compression of the lumbar spinal nerves. 3) Non-specific findings with no compression of the lumbar spinal nerves and no critical findings.

Results

670 MRI lumbar spine were performed. 25.3% in group (1) with 0% critical findings. 35.3% in group (2) with 4.6% critical findings. 5.8% in group (3) with 39.5% critical findings. 23.4% in group (4) with 0% critical findings. 2.8% in in group (5) with 0% critical findings. 7.3% in group (6) with 91.7% critical findings.

Conclusion

25.3% of MRI lumbar spine performed were inappropriate based on the ACR criteria. Moreover, there were no critical findings in this group. The use of ACR appropriateness criteria can reduce unnecessary MRI lumbar spine for low back pain without compromising patient safety.

Keywords

Magnetic Resonance Imaging Low back pain Appropriateness criteria Lumbar spine

Introduction

Low back pain contributes significantly to the global burden of disease and is the third ranking cause of disability adjusted life years.¹ Magnetic resonance imaging (MRI) is often performed to evaluate the cause of this low back pain. This is costly and can contribute to increased healthcare costs if used inappropriately.^2,3 Moreover, the imaging results on MRI studies often do not lead to clinically meaningful benefits.^4,5 If the symptoms clinically are related to lumbar spondylosis, then conservative treatment which includes patient behaviour modification and physical therapy is the mainstay of treatment, whereas surgery rarely is beneficial to the patient.^6,7 Moreover, it has been shown that the prevalence of lumbar spondylosis in asymptomatic individuals increased from 37% of 20-year-old individuals to 96% of 80-year-old individuals.⁸ The American College of Radiology (ACR) appropriateness criteria for low back pain was created as a guideline to reduce the number of these unnecessary scans.⁹ The guideline highlights Red Flags in the clinical history and examination which would indicate a more complicated aetiology of the low back pain that would require investigation with an MRI lumbar spine. These Red Flags are a clinical history of cancer, unexplained weight loss, immunosuppression, infection, intravenous drug use, prolonged corticosteroid use, significant trauma, patients who are elderly or potentially osteoporotic, patients who have failed conservative management and patients who have suspected cauda equina syndrome or rapidly progressive neurological deficit.

We set out on this local study to assess whether the MRI lumbar spine performed at a local General Hospital were appropriate based on the ACR appropriateness criteria.

Methods

MRI Lumbar spine performed for low back pain from 22 January 2016 to 31 October 2017 in the Hospital were reviewed and divided into the six different categories set out by ACR by looking at their electronic health records. These MRIs were requested by the Orthopedic consultants from the Hospital. The referral pattern was either from the outpatient clinic after referral from the Polyclinic or as an inpatient referral after admission for the investigation of low back pain.

The six categories are patients who have symptoms of acute, subacute, or chronic uncomplicated low back pain or radiculopathy and had,

(1) No Red Flags. No prior management.

(2) Low-velocity trauma, osteoporosis, elderly individuals (65-year-old and above), or chronic steroid use.

(3) Suspicion of cancer, infection, or immunosuppression.

(4) Surgery or intervention candidate with persistent or progressive symptoms during or following 6 weeks of conservative management. Of note, the 6-week duration includes the time that the patient waits to be seen by the Orthopedic department after initial referral from the Polyclinic.

(5) New or progressing symptoms with a history of prior lumbar surgery.

(6) Low back pain with suspected cauda equina syndrome or rapidly progressive neurological deficit.

The MRI lumbar spine were reported by the radiologists from the hospital. The reports were reviewed and divided into categories as stated below,

(1) Critical findings of spinal cord/cauda equina compression, metastatic cancer, spinal epidural abscess, and vertebral osteomyelitis which would require urgent medical assessment and treatment,

(2) Significant findings of a vertebral compression fracture or lumbar spondylosis causing compression of the lumbar spinal nerves,

(3) Non-specific findings demonstrating MRI evidence of lumbar spondylosis, but no compression of the lumbar spinal nerves and no critical findings as detailed above.

The subsequent management of patients were divided into the follow groups:

(1) Conservative management.

(2) Active intervention which includes spinal surgery, spinal injection, intravenous antibiotics, and palliative chemoradiotherapy.

MRI lumbar spine studies were excluded if there is insufficient clinical history provided, if they were performed for other indications other than back pain or were follow-up MRI studies.

There were 670 MRI Lumbar spine scans done from 22 January 2016 to 31 October 2017. The male to female ratio was 1.5 to 1. The average age was 53.2 years old, and the age range was 13 to 99 years old.

170 (25.3%) cases with no Red Flags and no prior management were identified. These inappropriate orders account for the 2nd largest group among all categories with zero critical findings in this category. Five patients (2.9%) required active intervention for symptomatic relief of pain related to intervertebral disc degeneration. All five patients had spinal operations which included one extreme lateral interbody fusion, two microdiscectomies and two microdiscectomies with laminectomies.

237 (35.3%) cases with low-velocity trauma, osteoporosis, involving elderly individuals (65-year-old and above), or chronic steroid use were identified. This formed the largest category with 11 critical findings (4.6%). Six patients (2.5%) required active intervention which included four patients receiving spinal operations and two patients receiving chemotherapy.

38 cases (5.8%) with suspicion of cancer, infection or immunosuppression were identified. There were 15 critical findings (39.5%). 9 cases (23.7%) required active intervention which includes four patients receiving intravenous antibiotics, two patients receiving spinal operations and three patients receiving chemotherapy.

157 cases (23.4%) with a history of progressive symptoms during or following 6 weeks of conservative management were identified. There were no critical findings. 7 cases (4.5%) required active intervention which includes one patient receiving intravenous antibiotics, one patient receiving facet joint injection, one patient receiving epidural steroid injection and four patients receiving spinal operations.

19 (2.8%) cases with previous lumbar surgery were identified. There were no critical findings with none requiring active intervention.

24 (7.3%) cases with suspected cauda equina syndrome or rapidly progressive neurological deficits were identified. This formed the smallest category but accounts for one of the greatest proportions of critical findings requiring intervention. There were 22 (91.7%) critical findings with 9 cases (37.5%) requiring active intervention which includes five patients receiving palliative chemotherapy, three patients receiving intravenous antibiotics and one patient receiving spinal operation Table 1.

Table 1.

Summary of active interventions by category.

No Red Flags

Lumbar extreme lateral interbody fusion

Lumbar posterior decompression laminectomy and discectomy

L5-S1 microdiscectomy

Left L5-S1 microdiscectomy

L5-S1 laminotomy, microdiscectomy and right sided decompression

Low-velocity trauma, osteoporosis, involving elderly individuals (65-year-old and above) or chronic steroid use

Palliative chemotherapy (2 cases)

Posterior decompression laminectomy T9, instrumentation and posterolateral fusion from T6 to L2

T11 To S1 posterior instrumentation, decompression laminectomy L3-L4, L4-L5 & L5-S1 and interbody fusion

Anterior cervical decompression & fusion C4-C5, C5-C6 & C6-C7 with plating

Cancer, infection, immunosuppression

Palliative chemotherapy (3 cases)

Intravenous antibiotics (4 cases)

T10 to L3 posterior instrumentation and decompression laminectomy

Failed conservative

Facet joint injection

Epidural steroid injection

Intravenous antibiotics

Transforaminal lumbar interbody fusion

Decompression laminectomy L4-L5 and interspinous spacer insertion

Left posterior decompression laminotomy and discectomy

L4-L5, L5-S1 discectomy

Prior lumbar spine surgery

Nil

Suspected cauda equina syndrome or rapidly progressive neurologic deficits

Palliative chemotherapy (5 cases)

Intravenous antibiotics (3 cases)

Right hemilaminotomy and facetectomy L4 and L5

Discussion

The ACR appropriateness criteria advocates that many patients who have symptoms of acute, subacute, or chronic low back pain do not require MRI imaging. The physician should perform careful clinical history and examination to identify the Red Flags that would indicate a more complicated aetiology of the low back pain that would require investigation with an MRI lumbar spine.

Our results show 170 out of 672 cases (25.3%) cases were inappropriate based on the ACR criteria. Moreover, 0 out of 170 cases (0%) had any critical findings which would require urgent medical assessment and treatment. This suggests that delaying or even withholding an MRI study in this patient group in our local population is unlikely to result in adverse clinical outcomes. The five patients who have had active intervention in this group were for symptomatic relief of pain related to intervertebral disc degeneration. Hence, even if an MRI Lumbar spine was not performed, it would unlikely lead to adverse clinical outcomes.

In contrast, in patients with suspicion of cancer, infection or immunosuppression, 39.5% of cases had critical findings with 23.7% receiving active intervention. Also, in patients with suspected cauda equina syndrome or rapidly progressive neurological deficits identified, 91.7% of cases had critical findings with 37.5% receiving active intervention. Hence, in these two groups of patients urgent MRI lumbar spine should be performed to help in prognosis and management.

There is a relatively small proportion of 4.6% of critical findings in patients with a history of low-velocity trauma, osteoporosis, involving elderly individuals (65-year-old and above), or chronic steroid use when compared to the other groups. However, it is considered appropriate based on the ACR criteria, and an MRI lumbar spine should be performed to help in prognosis and management.

Limitations

One of the limitations of this study is that we only reviewed the MRI lumbar spine requested in a tertiary hospital setting which may not be reflective of how low back pain is managed in primary care, both in the private primary care and in the Polyclinic setting. It is uncertain in primary care what proportion of patients with low back pain get managed conservatively without MRI imaging. In addition, we are uncertain if these current findings are also seen in the rest of the national healthcare system.

Another limitation is that clinical history is based on electronic records. We are reliant on good history taking and recording by the clinicians to determine whether there was prior management and treatment. Patients seeking their own physiotherapy or even alternative treatment methods may not be recorded.

Conclusion

This study demonstrates that 25.3% of MRI lumbar spine performed for patients with low back pain in a General Hospital were inappropriate based on the ACR criteria. Moreover, there was no critical finding on MRI detected in this group which would require urgent medical assessment or treatment. Conversely, Red Flags on clinical history or examination which would require investigation with an MRI lumbar spine should be able to detect all the critical findings that would require urgent medical assessment or treatment Figures 1 and 2.

Figure 1.

The 6 categories are patients who have symptoms of acute, subacute, or chronic uncomplicated low back pain or radiculopathy and had 1) No Red Flags. No prior management. 2) Low-velocity trauma, osteoporosis, elderly individuals, or chronic steroid use. 3) Suspicion of cancer, infection, or immunosuppression. 4) Surgery or intervention candidate with persistent or progressive symptoms during or following 6 weeks of conservative management. 5) New or progressing symptoms with a history of prior lumbar surgery. 6) Low back pain with suspected cauda equina syndrome or rapidly progressive neurological deficit. MRI results were divided into 1) Critical findings of spinal cord/cauda equina compression, metastatic cancer, spinal epidural abscess, or vertebral osteomyelitis. 2) Significant findings of vertebral compression fracture or compression of the lumbar spinal nerves. 3) Non-specific findings with no compression of the lumbar spinal.

Figure 2.

Percentage of cases by category requiring active medical intervention. 1) 2.9% of patients with no Red Flags. 2) 2.5% of patients who have low-velocity trauma, osteoporosis, elderly individuals, or chronic steroid use. 3) 23.7% with suspicion of cancer, infection, or immunosuppression. 4) 4.5% who have failed conservative treatment. 5) 0% of those who have prior lumbar surgery and 6) 37.5% of those with clinical suspicion of neurological compromise.

Primary care and tertiary hospitals should consider adopting the ACR appropriateness criteria for low back pain which is an effective tool at reducing the number of unnecessary MRIs, leading to reduced health care cost without compromising patient safety.

Footnotes

Author contributions

LWY collected and processed the data. LWY wrote the paper. TPH provided guidance for this paper and provided the direction for this paper. RSP was key in providing the data from the institute.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Ethical statement

Research ethics

Singhealth institutional review board 2017/2078

ORCID iD

Weiyong Lee

Data Availability Statement

The datasets generated and/or analysed during the current study are available from LWY (the corresponding author)*.

References

Murray

Lopez

. Measuring the global burden of disease. N Engl J Med 2013; 369(5): 448–457.

Chou

Qaseem

Owens

, et al. Diagnostic imaging for low back pain: advice for high-value health care from the American College of Physicians. Ann Intern Med 2011; 154(3): 181–189.

Jarvik

Hollingworth

Martin

, et al. Rapid magnetic resonance imaging vs radiographs for patients with low back pain: a randomized controlled trial. JAMA 2003; 289(21): 2810–2818.

Chou

Qaseem

Snow

, et al. Diagnosis and treatment of low back pain: a joint clinical practice guideline from the American College of Physicians and the American Pain Society. Ann Intern Med 2007; 147(7): 478–491.

Jarvik

Gold

Comstock

, et al. Association of early imaging for back pain with clinical outcomes in older adults. JAMA 2015; 313(11): 1143–1153.

Bigos

Bowyer

Braen

, et al. Acute low back Problems in adults. Clinical Practice guideline No. 14. AHCPR Publication No. 95-0642. Rockville, MD: Agency for health care Policy and Research, Public Health Service, U.S. Department of Health and Human Services. December 1994.

Last

Hulbert

. Chronic low back pain: evaluation and management. Am Fam Physician 2009; 79(12): 1067–1074.

Brinjikji

Luetmer

Comstock

, et al. Systematic literature review of imaging features of spinal degeneration in asymptomatic populations. AJNR Am J Neuroradiol 2015; 36(4): 811–816.

Patel

Broderick

Burns

, et al. ACR appropriateness criteria low back pain. J Am Coll Radiol 2016; 13(9): 1069–1078. DOI: 10.1016/j.jacr.2016.06.008.