Regarding imaging in msk medicine, I understand that it's hard to know when to do what. In my attempt to be systematic I'm giving you an overview of imaging the msk system from the top down.

As we go through, I will highlight the fundamental need for thorough history and examination, to guide the choice of imaging, as well as the need for clear pathways through which to channel, appropriate referrals.

To give you a bt of a background into who I am and where I come from ...

I have a background in MRCP and my specialty is SEM, which I studied through the London Deanery before CCT in 2012. Since then I have worked at these places as well as initially setting up an msk clinic in Herts. I also have an MA in classical acting which sparked an interest in what I call movement medicine.

My job in any sector is to guide the rehab of msk patients, through diagnostics, interpretating imaging and pathology results, introducing different management options and conveying a thorough understanding of the anatomical and physiological causation of msk pain and dysfn.

Questions I deal with on a daily basis include, how can I continue with rehab when physio makes my pain worse? How can I exercise my knee when it hurts so much? I damaged my shoulder in the gym so how do I know I'm not damaging it by doing further exercises? How can I exercise my knee when it swells so much afterwards? Why would I take pain killers to mask the pain in my ankle so I won't know if I'm damaging it further? Why haven't physio and injections helped my elbow pain - what do I do now?

These are the common scenarios we encounter, so I'm going to discuss guidelines for imaging with these in mind.

These are the common scenarios we encounter, so I'm going to discuss guidelines for imaging with these in mind.

The guidelines that I use on a daily basis are written by the RCR and are intended as a guide for referring clinicians in primary and secondary care.

I'm going to present an overview, of guidelines that you should be using, through clinical situations that you're likely to encounter, and the recommended imaging investigations that will give you the best information, for the best decision-making.

No guidelines should be used to restrict your practice; they should direct the clinician to the imaging test, that is most likely to give the A, to the Q being asked,

But I should point out, that we must always take into consideration, the small but significant risk from ionising radiation.

Starting at the top...

Atraumatic neck pain generally improves or resolves with conservative treatment.

Degenerative changes begin in early middle age and are often unrelated to symptoms. XR is indicated only in specific circumstances. Symptoms frequently arise from soft tissues, which will not show on XR.
-OA changes are common on XR.

Consider MRI and specialist referral when
1. pain affects lifestyle or
2. when there are neuro signs or RF features
(inc. vascular insufficiency, trauma, malignancy, infection, inflammation, myelopathy).

CT myelography may occasionally be required to provide further delineation, or when MRI is unavailable or impossible.

Degenerative changes are invariably present from middle age onwards.

Imaging is rarely useful in the absence of neurological signs or pointers to metastases or infection.

Consider more urgent referral in elderly patients with sudden pain, to show osteoporotic collapse or other forms of bone destruction.

Consider NM for possible metastatic lesions.

MRI may be indicated if local pain persists or is difficult to manage, or if there are long tract signs.

MRI is the preferred investigation for the diagnosis of most spinal diseases and is helpful in identifying those patients who may benefit when planning surgical intervention.
XR is only indicated if presentation suggests osteoporotic collapse in the elderly.
CT is used when MR is contraindicated and when further assessment of spondylolyses is required.
NM is non-specific, and has been largely supplanted by MR and CT in the assessment of chronic back pain. It may show occult osteoid osteomas and spondylolyses.

MRI is the imaging of choice and is indicated immediately in patients with acute neurological features, and urgently in those with suspected malignancy or infection. MRI is the initial investigation of choice for all SC lesions, to assess cord compression and to give an indication of postop prognosis.

Plain radiograph may be required preoperatively. MR is preferable as the firstline investigation in patients with red flag signs, since it has a stronger negative predictive value.

CT is useful to guide soft tissue and bone biopsy and may identify sequestra in infection.

NM is non-specific and should be viewed with plain radiographs. It is useful to show the full extent of disease, especially with metastatic deposits.

Serious (RF) features:

a. Neurological

Sphincter & gait disturbance
Saddle anaesthesia
Sev. or prog. motor loss
Widespread neuro deficit

b. Other

Age 55 years
Previous malignancy
Systemic illness
HIV
Weight loss
IV drug use
Steroid use
Structural deformity
Non-mechanical pain (no relief with bed rest)
Fever
Thoracic pain

Serious (RF) features: TUNAFISH

a. Neurological

Sphincter & gait disturbance
Saddle anaesthesia
Sev. or prog. motor loss
Widespread neuro deficit

b. Other

Age 55 years
Previous malignancy
Systemic illness
HIV
Weight loss
IV drug use
Steroid use

(inc. vascular insufficiency, trauma, malignancy, infection, inflammation, myelopathy)

(inc. vascular insufficiency, trauma, malignancy, infection, inflammation, myelopathy).
Structural deformity
Non-mechanical pain (no relief with bed rest)
Fever
Thoracic pain

MRI is the preferred investigation over CT (wider field of view visualising the conus, postop changes, etc).

Demonstration of disc herniation should be considered after failed conservative management.

Clinico-radiological correlation is important because many disc herniations are asymptomatic.

Acute back pain is usually caused by conditions that cannot be diagnosed on XR (osteoporotic collapse is an exception). Normal XR may be falsely reassuring.

XR is the initial investigation but may be normal in early osteomyelitis.

MRI accurately shows osteomyelitis and any associated soft tissue abnormality. It is the best imaging technique in suspected osteomyelitis.

XR should be used in cases of unresolving bone pain.

If the XR appearances are suggestive of primary bone tumour, referral to a specialist centre should not be delayed. MRI is the investigation of choice for local staging.

CT can improve diagnostic information in some tumours, such as osteoid osteoma, and show intratumoral calcification and ossification.

P.S. For skeletal metastases...
More sensitive and specific than NM, MRI is the investigation of choice to confirm symptomatic metastases, especially in the axial skeleton.

XR gives a dedicated view of the symptomatic area.

MRI is appropriate if pain persists with normal XR or apparently normal NM. MRI may also provide further information when XR and/or NM findings are abnormal.

NM is used if pain persists with normal XR or equivocal and abnormal XR in specific circumstances (eg, suspected osteoid osteoma, osteomyelitis, metastases, metabolic bone disease and inflammatory arthropathy). Bone scan is particularly well suited for rib pain.

CT is used to define bony anatomy in areas of abnormality on XR/MRI/NM, especially if bone biopsy is indicated.

US may be helpful to assess suspected infection, tumour and some fractures particularly in children.

US is useful as the first investigation to evaluate cystic and solid masses, and to distinguish them from pseudotumours. It is also useful to monitor benign masses (eg, haematomas) and to assess for local recurrence of soft tissue sarcomas.

MRI is useful to demonstrate anatomy and can provide a specific diagnosis in some patients.

XR may show bony abnormality associated with masses and can show tumour mineralisation. CT may occasionally help in these areas.

Lateral XR of the thoracic and lumbar spine is the first investigation in suspected osteoporotic collapse and in differential diagnosis. [Collapsed vertebrae are often seen as incidental findings at CT.]
-XR also identifies characteristic signs of other metabolic bone disease, including osteomalacia and hyperparathyroidism.

MR may distinguish acute from chronic osteoporotic collapse, and also distinguishes between osteoporotic and malignant vertebral collapse.

DEXA is used for measurement of bone density. DEXA is reserved for patients with risk factors for osteoporosis and may detect moderate and severe vertebral fractures. In the elderly, fracture on XR is adequate to establish a diagnosis and DEXA is unnecessary unless monitoring of treatment is required.

NM is useful in hypercalcaemia after exclusion of myeloma in the identification of metastases.

Quantitative CT provides objective measurements of bone mineral content in patients where DEXA is difficult to interpret, whether because of deformity of spine or hypertrophic degenerative change. Quantitative CT is more sensitive than DEXA but has a higher dose.

XR of the affected joint may be helpful to establish cause, although erosions are a relatively late feature of inflammatory arthropathy. XRs are indicated for symptomatic joints only.

In patients with suspected RA, XR of the feet may show erosions even when symptomatic hand/s look normal.

All of US, MRI, NM can show acute synovitis.
-US and MRI can show early erosions.
-MRI can also show articular cartilage.
-NM is rarely needed but can show distribution.

-MRI BMO is a strong predictor of radiographic progression.
-US may be helpful both for assessment and monitoring of activity.

Impingement is clinically diagnosed.

XR is used as a preoperative assessment. XR is indicated for persistent shoulder pain that is unresponsive to conservative treatment to exclude calcific tendinitis and diagnoses unrelated to the rotator cuff.

US is the investigation of choice in the assessment of RC and surrounding soft tissues. It may be used to guide injection. It is reserved for cases unresponsive to firstline treatment and clinically guided injection. It is indicated preoperatively if the surgeon requires assessment of RC integrity.

MRI is an alternative to US and is useful after major trauma to assess complex injury and bony abnormality. MRI excludes rare conditions obscured by acromial arch and bone abnormalities when other investigations and treatment do not establish a diagnosis.

Plain XRs may show characteristic bony lesions in the humeral head and glenoid. (Bony Bankart, Hill Sachs)

MR may show the labrum without intra-articular contrast but MR arthrography is the investigation of choice for labral and ligamentous lesions.


CT will show the bony glenoid and CT arthrography will show cartilaginous labral tears.

Although XR of the SIJs is the firstline investigation for seronegative arthropathy, it is not sensitive for early disease.

MRI will detect inflammation and structural changes and is the investigation of choice for early disease.

CT will show erosive changes and ossification.

NM is of limited value to confirmed suspected scaroilitis.

US in expert hands has been used to show inflammation in periarticular soft tissues and entheses.

XR of the pelvis is indicated for persistent pain. It may demonstrate focal bony pathology, erosive joint changes, dysplasia and anatomical features associated with FAI.
-XR is abnormal in established AVN but is frequently normal within first 6-9 months.

MRI is widely accepted as the best investigation for further evaluation of XR negative hip pain, including AVN.

MR arthrography may be helpful to diagnose labral tears.

Bone scan is less specific than MRI for AVN and other focal lesions but is an alternative when MRI is not possible.

CT is helpful for diagnosing osteoid osteoma or for identifying subchondral fractures.

MRI is useful in patients with persistent undiagnosed pain, including suspected AVN and sepsis.

US is useful for AKP with suspected tendinopathy or associated bursitis.

Symptoms frequently arise from soft tissues, which will not show on XR.
-OA changes are common on XR.
-XR is needed when considering surgery.

Good indications for imaging:
-Sudden onset or exacerbation of pain
-Pain persisting >6 weeks in children and young adults.

MRI is the investigation of choice to identify meniscal tears and loose bodies.

XR will identify radio-opaque loose bodies, a less frequent cause of locking.

XR is useful to guide surgery.

Is there pain?

Is there pain clinically evident of 1st MTP enlargement caused by OA, bursal inflammation, ganglion formation, or gouty or other inflam arthropathy.

As we remember from the Arthropathy slide, US and MRI can show early erosions, US may guide injection.

Most patients with heel pain should be managed on the basis of clinical findings without imaging.

The cause of pain is rarely detectable on XR. Calcaneal spurs are common incidental findings.

US and MRI will show thickening of plantar fascia and inflam change but should be used selectively.

US enables guided injection therapy.

So, ending at the bottom - last but by no means least, in terms of frequency of presentation.

I have a short Case History in case we had time to look at a common presentation, with high expectation and potential for conflict in the request for imaging.

John, 38, comes to see you with a letter from a private physiotherapist asking that you send him for an MRI scan. Previously he had back pain and sciatica for three weeks, and had not settled very well with NSAIDs and codeine-based analgesics. His referred pain was in the L5 distribution of the buttock and lateral thigh and lateral shin. He has had three sessions of physiotherapy with minimal improvement in his symptoms. He is an accountant who is office based and plays golf when his work allows. He is slightly overweight, but otherwise fit and well. Symptoms started after he lifted a heavy suitcase out of the car boot. There were no red flags in the history.


Why has this MRI request come about?

I'm sure that you recognise this type of case. A private practitioner has raised the patient's expectations. There is potential for conflict, particularly if you not agree with the request, and care is needed.

OE: lumbar flexion is reduced and SLR is 45° on the affected side and 80° on the other side. Neuro exam NAD, the ankle reflexes are equal and DF of the big toe is normal.

There are no RFs of cauda equina syndrome [where a central disc protrusion usually at L4 or L5 level is an orthopaedic emergency (see box below).]


How should you refuse the request?

I would explain that in my opinion, although the pain is bad, there are no signs of pressure on the nerve root. I discuss my usual referral criteria for MRI scan or ortho/neurosurg referral, which are:

• persistent symptoms after 6 weeks

• progressive neuro signs

• lack of specific diagnosis - imaging is often unhelpful

• unrelenting pain.

I would discuss his views on surgery, and explain that an MRI scan would look for pathology that can be treated surgically. (People are often not keen on surgery)

I state that MRI scans can find minor features such as prolapsed discs in at least 20% of cases, which do not need treating, but can cause unnecessary anxiety.

The patient will often agree that he would rather continue with physiotherapy and stronger analgesics, and return for after 3-4 weeks. It can help to write a short note back to the physiotherapist.

In this way you can avert confrontation in consultation. It is down to careful history taking and examination, and explaining your referral criteria. This is not new to you, but by showing the patient you understand the condition, and know when to imag, and can compare them with your usual practice, you gain their confidence.


How do you gauge outcome in the longer term?

After 3-6 weeks, what are you looking at in review?
Is the pain had improved even slightly?
Have exam measures improved slightly as well – if SLR is now 60° instead of 45° on the affected side, this is your objective measure. The patient can then also see that this is better. Have they also been able to reduce their analgesia. Can they feel more positive?

Encourage continuity of the exercise programme, which may include Pilates-type ex that they can access themselves, and encourage them to keep in touch. One of the most difficult things with such patients is to encourage them to do exercises when they are symptom free.

Learning Points
When faced with a request for a referral from another health care professional:
• Take a careful history and do a good exam
• Do not show irritation with the request even if you are
• Discuss your normal referral thresholds
• Outline and agree a management plan with timescale
• Involve the patient in the whole process
• Encourage preventive medicine, which is what you do better than anybody.

red flags for back pain

* PMH ca, particularly breast, prostate, thyroid, kidney
* Neuro signs from different nerve roots, suggesting widespread pathology
* Weight loss or night sweats
* Abnormal blood tests, esp. raised ESR, calcium/ PSA

cauda equina symptoms
* Back or perianal pain
* Faecal or urinary incontinence
* Sudden onset of impotence
* Numbness in legs and difficulty walking
* Reduced perianal sensation, anal tone, lower limb strength and reflexes

So I hope that I've helped you to know when to do what in terms of imaging of the msk system from top to bottom.

I've highlighted the fundamental need for thorough history and examination to guide the choice of imaging as well as the need for clear pathways through which to channel appropriate referrals.

Guidelines will never replace good communication and discussion between a radiologist and a clinician. I am in constant contact with my musculoskeletal radiologist colleagues, whom I speak to regarding patients on an almost daily basis.

The guidelines ought not to be used to restrict your practice in these specific clinical circumstances I've highlighted; instead they should direct the clinician to the imaging test that is most likely to give the A to the Q being asked, while taking into consideration the small but significant risk from ionising radiation.