High false positive rates for red flags in spinal malignancy
GPs should not rely on ‘uncritical use’ of red flags in identifying spinal malignancy in patients with low back pain, as they are prone to false positives, says a gold-standard review.
Cochrane Collaboration researchers reviewed eight studies looking at red flags for spinal malignancy in 6,622 patients presenting in primary or secondary care with symptoms of low back pain. The studies compared the red flags found from taking a clinical history or a physical examination, with CT and MRI scans or a follow-up of patients of over six months.
In primary care, the post-test probability for diagnosis of spinal malignancy following a positive red flag remained below 1% - the generally considered prevalence in primary care of serious pathologies affecting the spine. Only a previous history of cancer, at 4.6%, was found to have a sufficiently high likelihood ratio in primary care to meaningfully increase the probability of malignancy. Unexplained weight loss had the next highest likelihood ratio in primary care at 1.2%. The remaining red flags investigated all fell below a likelihood ratio of 1% in primary care.
What does it mean for GPs?
The authors concluded that there is ‘insufficient evidence’ to support or refute the clinical usefulness of most red flags to screen for spinal malignancy in patients with low back pain.’ They added that ‘red ﬂags such as insidious onset, age over 50, and failure to improve after one month have high false positive rates, suggesting that uncritical use of these red ﬂags as a trigger to order further investigations will lead to unnecessary investigations that are themselves harmful and the consequences of these investigations themselves producing false positive results.’
Dr Majid Artus, GP clinical lecturer with an interest in back pain at Keele University: ‘None of the included studies were designed to achieve the specific objectives of the review. So, evidence was not sufficient to pass recommendations on the diagnostic accuracy of the identified red flags. This, of course doesn’t mean there is no evidence, we just have to wait for evidence to be found for or against.’