The facet joint has been increasingly implicated as a significant source of pain which can arise from any structure within the facet joint complex including the fibrous capsule, synovial membrane, hyaline cartilage and bone. Facet arthropathy is prevalent in patients with low back pain and has been studied for an association with pain and for a potential impact on treatment indications and outcome. Facet osteoarthritis is the most frequent form of facet pathology [21, 22].
The aim of our study was to investigate the potential role of 18F-PET/CT imaging to precisely pinpoint the sites of active disease in the facet joints.
Our study demonstrates a strong correlation between FDG activity and the location of the painful facet joints even in the absence of underlying CT degenerative changes (Fig. 2). Increased FDG uptake in inflammatory cells, such as macrophages, has been documented in the literature. The pathophysiological of FDG activity in infection and inflammation can be explained by hyperemia which facilitates FDG delivery to the site of inflammation. In addition, enhanced glycolytic pathway and upregulation of glucose transporters increase the utilization of FDG [23, 24]. Irmler and co-workers showed a significant correlation between inflammatory cell bulk and the degree of FDG activity [25]. Arthropathy is a common disorder and is associated with physical and disability encumbrance of involved patients [26, 27]. 18F-FDG PET/CT can efficiently assess the disease extent and response to therapy in many musculoskeletal inflammatory disorders. Based on the glucose metabolism, 18F-FDG PET/CT can point to diseased joints by demonstrating inflammatory peripheral cells and fibroblasts [28]. Rosen et al. [29] evaluated the level of FDG uptake in the spine and correlated it with the findings from CT. The authors found a good correlation between the severity of FDG uptake and the severity of degeneration on CT.
This study shows the potential efficacy of 18F-FDG PET/CT in diagnosing abnormalities of cervical and lumbar facets in 129 adult patients who presented with neck or back pain. According to our results, CT diagnosed 72 patients with facet degenerative changes. From 59 patients with confirmed specific facet joints as the source of pain, 38 cases were false positive on CT images (Fig. 3). In addition, the FDG abnormality was not associated with clear degenerative changes on CT images in 18.6% of our cases. Several studies have reported that the CT facet joint degenerative changes are non-specific and not correlated with the actual site of pain generation [30,31,32].
Further, in the present study 18F-FDG PET/CT not only showed an additional asset in precisely identifying pain location, where 89 (69%) had bilateral distribution compared to 40 (31%) unilateral pain distribution, but also demonstrated significantly higher sensitivity and specificity compared to CT (88.1% vs. 57.6% and 97.1% vs. 45.7%, p < 0.01). Similarly, the drawbacks of CT were previously mentioned by Gorbach et al. [33] who reported that the degree of facet joint arthropathy as outlined by CT was not a significant predictor for the outcome in 42 patients who performed facet joint blocks (p = 0.57–0.95). Carrino et al. [34] measured inter-observer agreement, among 4 radiologists, in MR lumbar spine for facet arthropathy in 111 studies and found a variability of 0.54 (CI 95%: 0.50–0.57). The inconsistent variability indicated that MRI does not provide a clear assessment of facet arthropathy inflammatory status, and the report can be inconclusive. In the same study, the authors reported that conventional MRI and CT imaging convey facet arthropathy at multiple levels with no clear findings for those joints causing pain and no definite guide to the level of injection therapy.
There is scarce literature examining the role of 18F-FDG PET/CT imaging in facet joint arthropathy. Yet, our data corroborate that of Gamie et al. [35] where 67 patients with suspected facetogenic or discogenic pain were studied. Imaging studies were performed with 18F-FDG PET/CT without contrast. Abnormal uptake of the tracer was found in 56 patients (83.6%) with 45 of these patients having abnormal activity at the facet joint. Moreover, the authors reported a sensitivity of 84% overall, and a sensitivity of 88% for patients without a prior history of lumbar surgery. However, the study was not solely focused on facet joints, which suggests a different set of sensitivity values in reality.
This study has some limitations. Selection of patients based on back pain referral may cause selection bias. Nevertheless, all patients have been examined neurologically and with the aim of this study to explore the potential role of 18F-FDG PET/CT in patients with facet joint disorder, we think that the present data are informative. In the same context, an assessment of the effect of management was assigned to neurology physicians and may potentially result in some variability among them. However, as the assessment happened in a single center with a close experience level, we assume that the variability was insignificant.