Clinical prevalence and population incidence of serious pathologies among patients undergoing magnetic resonance imaging for low back pain

Abstract

BACKGROUNDIn rare cases low back pain may be caused by underlying serious pathology such as fracture, malignancy, cauda equina syndrome, or spinal infection. The lack of evidence regarding either the clinical prevalence or population incidence of serious pathologies in the lumbar spine makes it difficult for clinicians to adequately assess a patient's risk of serious pathology. PURPOSETo determine the prevalence of serious pathologies in patients with low back pain who have been referred for a lumbar magnetic resonance imaging (MRI) by a specialist in a private secondary care or public tertiary care setting. The incidence of these serious pathologies in the geographic region of South Auckland, New Zealand was also investigated. STUDY DESIGNRetrospective, observational cohort study. PATIENT SAMPLEConsecutive patients referred for lumbar MRI over a 10-month period (1st of October 2013–31st of July 2014). METHODData from all eligible MRI reports was analyzed and any serious pathologies were identified and recorded. Prevalence (along with 95% confidence intervals) was calculated as a percentage of the study population. Prevalence specific to private secondary care and public tertiary care settings was also calculated and prevalence rate ratios were determined to allow comparison between settings. Incidence in the geographic region of South Auckland, New Zealand, was determined using data collected from participants recruited from the regional public hospital. Population incidence with respect to age, gender, and ethnicity for each target condition was calculated and incidence rate ratios were computed to compare groups. RESULTSA total of 2,383 participants referred for lumbar MRI scans were included in this study. Prevalence was significantly higher in the public tertiary care setting than in the private secondary care setting for all pathologies investigated in this study. Pathology specific prevalence in secondary care vs tertiary care settings was: malignancy, 0.3%, 4.4% (p<.001); fracture 2.2%, 6.7% (p<.001); cauda equina compression 0.6%, 2.3% (p=.001); infection 0.1%, 3.4% (p<.001). The combined prevalence in secondary care was 3.2% and in tertiary care 14.8% (p<.001). Pathology specific total incidence was: fracture, 13 per 100,000 person-years (p-y); malignancy 8.5 per 100,000 p-y; cauda equina compression 4.4 per 100,000 p-y; spinal infection 6.6 per 100,000 p-y. CONCLUSIONSThe prevalence of serious pathologies was significantly higher in tertiary care (public health) than in private secondary care settings. One in every 6.5 patients referred for MRI in tertiary care demonstrated structural abnormalities associated with serious pathology, which raises the question of whether access to MRI should be re-evaluated.