How many people with lower back pain have a serious underlying problem such as a tumour?
While lower back pain is a common condition around the world, in only a very small proportion of people – less than 5% - will this be caused by a serious problem such as fracture, infection, or tumour.
Of these serious problems, spinal tumours are the second most common (after fractures) and are present in about 1% of all patients with lower back pain.
Who is spinal cancer most likely to affect?
The spine is much more frequently affected by metastatic disease than it is the site of primary tumours.
This means that spinal cancers occur mostly in people already affected by cancer in other parts of their body (such as prostate, lung, or breast cancer).
What do current guidelines recommend for assessing patients with lower back pain?
Clinical guidelines suggest that when assessing patients with lower back pain, clinicians should be aware of “red flags”, which are features thought to be associated with a higher risk of serious disease.
These include features such as age greater than 50, unexplained weight loss, or a previous history of cancer.
When these “red flags” are present, clinicians should consider sending the patient on for further diagnostic testing.
What further diagnostic tests may be performed and when are these recommended?
Clinical guidelines recommend against the routine use of diagnostic imaging (such as x-rays, MRI, or CT scans) for patients with lower back pain. In some cases, certain blood tests can help to identify serious disease.
However, both diagnostic testing and blood tests are only recommended when a clinician thinks that a patient has a more serious problem causing their lower back pain.
How did your research into assessing the accuracy of using red flags to diagnose spinal malignancies originate?
While the clinical guidelines often suggest using these “red flags” to identify patients with serious underlying problems, there is very little data provided to support their usefulness. This makes it hard for clinicians to know how best to use these “red flags” in clinical practice.
To address this, we aimed to summarise all of the available evidence around commonly recommended “red flags” for spinal cancer.
What did your research involve?
We performed a systematic literature review to identify all studies that had reported on the diagnostic accuracy of “red flags” (clinical characteristics from the medical history or physical examination) to screen for spinal cancer in patients with lower back pain.
We searched electronic databases and screened over 2000 articles for inclusion. We included six of these studies which were performed in a primary care setting – evaluating over 6,600 patients with lower back pain – from which we extracted data on 20 different “red flags”.
How accurate did your research find using red flags to be when diagnosing spinal malignancies?
In general, we found that the diagnostic performance of most "red flags" is poor, especially when used in isolation. The exception was a previous history of cancer which, when present in a patient with lower back pain, seemed to meaningfully increase the probability of spinal cancer.
The problem lies in the fact that spinal cancer is very rare, and many of the “red flags” had high false-positive rates which would result in a lot of unnecessary diagnostic testing.
Can the further tests that these red flags trigger be harmful to the patient?
Usually these further tests involve diagnostic imaging which in some cases (such as x-ray and CT) is associated with exposure to radiation.
Added to this, these tests can be quite expensive and result in a lot of unnecessary worry for patients.
Your review focussed on the diagnostic accuracy of individual red flags. In practice, several red flags taken into consideration. How can the accuracy of these combined red flags be tested and are there plans to carry out this research?
In practice, a clinician will usually consider all aspects of the presentation of a patient and then decide whether further testing is warranted. However, in the studies we included in our review data was usually only available on the diagnostic accuracy of isolated “red flags”.
Future studies should provide data on a number of “red flags” tested on the same patients – in this way the most useful combinations of these can be identified.
We are currently looking into further research, however given the low prevalence of spinal cancer any study will need to recruit a very large sample size to provide good data on these red flags.
What impact do you think your research will have?
Our research provides some insight into the difficulty clinicians face when attempting to screen for serious problems in patients with lower back pain.
Clinicians should be aware that spinal cancer is rare in patients with lower back pain and that many common red flags should be used with caution as they have high false-positive rates.
The lack of data to support many of these red flags highlights a need for improved performance and reporting of diagnostic studies in this area.
How do you think the diagnosis of spinal cancer can be improved?
Our review highlights a need for high quality diagnostic studies in this area and further evaluation of the implications of clinical findings in patients with lower back pain.
Studies which evaluate combinations of “red flags” may lead to an improved ability of clinicians to diagnose spinal cancer in patients with lower back pain.
Where can readers find more information?
Reader can find more information on the review at the Cochrane Library (www.thecochranelibrary.com).
About Dr Nicholas Henschke
Dr Nicholas Henschke is an epidemiologist at the Institute of Public Health in the University of Heidelberg, Germany.
He has a background in physiotherapy and completed his PhD in 2008 at the University of Sydney in Australia.
He has previously worked as a research fellow at the EMGO+ Institute for Health and Care Research in Amsterdam, the Netherlands, and The George Institute for Global Health in Sydney, Australia.
His research focuses on the diagnosis and management of musculoskeletal conditions, rehabilitation, and research methodology.