QUESTION | Guidelines suggest routine imaging for low
back pain is not recommended. Are there any situations where imaging is useful?
ANSWER | Overview: A 2009 systematic review and meta-analysis
of the use of imaging (radiograph, CT or MRI) in non-specific low back pain
found no significant differences in pain or function either short term (<3
months) or long term (6-12 months) (Chou, 2009). This finding should not be
used however, to conclude imaging has no role in the management of back pain. This
review excluded patients with suspected infection or malignancy, both accepted
indications for obtaining imaging in those with back pain. Further, studies
often fail to account for whether diagnostic information from imaging is
associated with subsequent appropriate evidence-based management. If
ineffectual treatment modalities are utilised in the setting of an accurate
diagnosis, then it follows that outcomes will not necessarily be enhanced.
Radiographs have limited utility in the
investigation of non-specific low back pain. They may be a a reasonable option
in patients when concerned for infection or malignancy, but clinical suspicion
is not high (Jarvik 2002). When plain radiographs of the lumbar spine are
indicated, AP and lateral views are usually adequate. Oblique radiographs, once
popular for identifying pars stress fractures, substantially increase radiation
exposure with little benefit, and should not routinely be performed. Dynamic
flexion-extension views are often helpful in the assessment of instability,
such as can occur with bilateral pars defects leading to spondylolisthesis.
Lumbo-sacral nerve root irritation often
refers pain to the hip and groin. Consequently, hip radiographs can be useful
in distinguishing between lumbo-sacral and hip joint pathology. Weight bearing
views should be obtained including a false profile view to assess the posterior
aspect of the joint. If consideration is being given to sacro-iliac joint
instability, a flamingo view x-ray performed in single leg stance can identify
any resulting movement at the pubic symphysis.
Spinal MRI is more informative than radiographs
and CT because it can also identify other pathologies, including inflammatory,
malignant, and vascular disorders. In addition, MRI is not associated with ionising
radiation. One justified criticism of MRI is that findings are often
mis-interpreted. Incidental findings are often seen on imaging which are then
incorrectly attributed as a cause of pain. Disc herniations are seen on MRI in
22 to 67 percent of asymptomatic adults (Jensen M. C. 1994, Boden 1990, Weishaupt
D 1998). Given the high prevalence of these findings, the discovery by MRI of
bulges or protrusions in people with low back pain may be coincidental, and
drawing attention to them may contribute to fear avoidance behaviours
contributing to central sensitisation. Often overlooked in these studies
however, is that severity of herniations does appear to correlate with
symptoms, with disc extrusions rarely seen in asymptomatic populations (Jensen
1994). Consequently, it is reasonable to conclude there is a high probability
that a disc extrusion is contributing to pain if present and should not be
One limitation of MRI is that most scans
are taken with the patient supine, with the spine unloaded. Loading and
movement of spinal structures can often lead to dynamic changes correlating
with postural symptoms. There are now MRI scanners available which can assess
dynamic change in the upright position with flexion and extension. In carefully
selected patients, such as those with significant postural variation in
symptoms and equivocal findings on standard MRI, upright MRI can be very
useful. The magnet strength of the upright MRI scanners however is not comparable
to supine scanners, resulting in reduced image quality and limiting usefulness
as a first line investigation. Many patients with claustrophobia also tolerate
the upright scanner well. MRI enhancement with gadolinium allows the
distinction of scar from disc in patients with prior back surgery.
There are a variety of possible MRI
sequences impacting contrast, resolution and speed of scan acquisition. Not
every possible sequence is performed with every scan, the decision of which
sequences to perform usually the decision of the supervising radiologist. This
can lead to some variation in the scans obtained between different imaging
companies. Radiographers with expertise in musculoskeletal MRI often provide
more informative reports which assists in the clinical interpretation of
findings. Providing a clear clinical history outlining relevant examination
findings when ordering an MRI will assist the radiologist in selecting the most
appropriate sequences and improve the accuracy reporting.
Historically, CT scans have been used to
diagnose pars interarticularis stress fractures in the lumbar spine. Recent
research has demonstrated that 3 Tesla MRI scan using a thin slice VIBE
sequence is 100% accurate in diagnosing complete pars fractures, comparing very
favourably to CT. Additionally, MRI is able to detect bone marrow oedema and
does not employ ionising radiation. MRI should now be the first line
investigation for suspected pars stress fractures (Ang 2016), although this
sequence is not yet routinely performed by every imaging company.
CT scans can assess osseous structures
better than either plain radiography or MRI and is therefore helpful in
assessing for bony disease. While CT scans can identify disc prolapses with
similar sensitivity to MRI, they are unable to visualise nerve roots limiting
utility in radicular presentations. MRI is thus the preferred modality. In some
circumstances, CT, especially fine cut CT, can be useful to look for specific bony issues
difficult to visualise on an MRI. This may include suspicion of non-union
following a surgical fusion, or cases where susceptibility artefact from
implants impacts MRI quality.
guided local anaesthetic and corticosteroid injections:
Not all pain originating from nerve root
compression presents in a classical dermatomal distribution. Irritation of the
small nerve fibres innervating the dura around the nerve root itself leads to
somatic referred pain, typically experienced as an intense ache. Recognising
the common somatic referral patterns from L5 (postero-lateral hip) and S1 (buttock
region) can assist the interpretation of MRIs, however there may still be some
doubt. CT guided local anaesthetic injections are very useful to confirm nerve
root irritation as a source of symptoms. These injections are usually directed
at nerve roots through the intervertebral foramen (perineural) or more
centrally between the lamina into the spinal canal (epidural) based on
examination and imaging findings. Significant improvement in symptoms while the
local anaesthetic is active is diagnostically useful.
Facet joints are another common target of
CT guided injection. Isolated facet joint pathology as a cause of symptoms is
uncommon however, and a strong clinical suspicion should exist before
injections are performed (see SPECT/CT below).
Hypertrophic facet joints often lead to descending nerve root
irritation, or even posterior annulus irritation, and a positive response to
injection does not guarantee a facet joint as a source of pain, though it is
The type and level of any spinal injection
should be carefully determined to ensure the highest chance of success.
Clinician experience and skill can significantly impact the utility of CT
guided injections, and the use of experience radiographers is recommended.
Cortico-steroid is usually injected concurrently, although benefit can be
This is a combined test in which the
nuclear medicine test Single Photon Emission Computed Tomography is combined
with a CT. This is a very useful test to identify symptomatic facet joints. Follow-up
with a diagnostic injection using local anaesthetic can assist in the diagnosis.
The concurrent injection of cortico-steroid into facet joints with anaesthetic
occasionally leads to sustained benefit.
If a facet joint is confirmed as a source
of symptoms and does not respond to cortico-steroid injection, ablation of the
innervating medial branch nerves has been shown to reliably provide sustained
relief and can also be performed under image guidance.
Occasionally despite a strong index of
suspicion of nerve related pathology, examination and imaging findings remain
equivocal. The combination of nerve conduction studies (NCS) and electromyography
(EMG) studies can provide further insight. These tests are most useful in patients with
in whom imaging findings are inconsistent with the clinical symptoms of
radiculopathy, especially weakness. In cases of radiculopathy, NCS and EMG can
localise the specific spinal nerve root involved, though this is more reliable
at lower levels (L5 and S1). Electrophysiology testing also can identify
conditions that mimic radiculopathy.
Chou R, Fu R, Carrino JA, Deyo RA. Imaging
strategies for low-back pain: systematic review and meta-analysis. Lancet 2009;
Jensen MC, Brant-Zawadzki MN, Obuchowski N,
et al. Magnetic resonance imaging of the lumbar spine in people without back
pain. N Engl J Med 1994; 331:69.
Boden SD, Davis DO, Dina TS, et al.
Abnormal magnetic-resonance scans of the lumbar spine in asymptomatic subjects.
A prospective investigation. J Bone Joint Surg Am 1990; 72:403.
Weishaupt D, Zanetti M, Hodler J, Boos N.
MR imaging of the lumbar spine: prevalence of intervertebral disk extrusion and
sequestration, nerve root compression, end plate abnormalities, and
osteoarthritis of the facet joints in asymptomatic volunteers. Radiology 1998;
Jarvik JG, Deyo RA. Diagnostic evaluation
of low back pain with emphasis on imaging. Ann Intern Med 2002; 137:586.
Ang, E. C., Robertson, A. F., Malara, F.
A., O’Shea, T., Roebert, J. K., Schneider, M. E., & Rotstein, A. H. (2016).
Diagnostic accuracy of 3-T magnetic resonance imaging with 3D T1 VIBE versus computer
tomography in pars stress fracture of the lumbar spine. Skeletal Radiology,
Lehman, V. T., Murphy, R. C., Schenck, L.
A., Carter, R. E., Johnson, G. B., Kotsenas, A. L., … Maus, T. P. (2016).
Comparison of facet joint activity on 99mTc-MDP SPECT/CT with facet joint
signal change on MRI with fat suppression. Diagnostic and Interventional
Radiology, 22(3), 277–283.
Dr Paul Mason