Isolated inflammatory involvement of the occipital artery in giant cell arteritis and polymyalgia rheumatica: findings from a retrospective analysis and the critical role of MRI in diagnosis

Giant Cell Arteritis (GCA) and Polymyalgia Rheumatica (PMR) are systemic inflammatory disorders predominantly affecting elderly individuals over the age of 50, characterized by granulomatous inflammation of medium and large arteries [1]. Some authors suggest that GCA and PMR may not be distinct conditions but exist on a continuum [2]. Patients may also present with manifestations of both conditions simultaneously (GCA-PMR) [3]. The hallmark of GCA is the inflammatory involvement of branches of the external carotid arteries, especially the temporal arteries, leading to cranial symptoms such as headache, jaw claudication, and visual disturbances [4]. However, diagnosis can be challenging, particularly in cases with non-specific symptoms or inconclusive temporal artery biopsy results [3].

While the temporal artery has been the primary focus in diagnosing GCA, the occipital artery, despite its frequent involvement, has been relatively underexplored. This may be due to its less accessible location and the historical reliance on clinical and biopsy findings, which tend to prioritize temporal artery inflammation [5]. The advent of advanced imaging techniques, such as magnetic resonance imaging (MRI), has provided new insights into the inflammatory involvement of extracranial arteries, including the occipital artery [6]. MRI offers high-resolution visualization of arterial walls without radiation exposure, making it particularly useful in detecting inflammation in arteries that are less accessible to other diagnostic methods. While [¹⁸F]Fluorodeoxyglucose positron emission tomography (FDG-PET) can detect systemic inflammation, its cost, radiation exposure, and limited availability favor MRI as a more practical alternative in many cases. Furthermore, FDG-PET can only be used to detect inflammatory changes in the scalp arteries if the scanner provides image quality with sufficient spatial resolution to reliably assess these delicate vessels.

Current guidelines recommend early imaging to support the clinical diagnosis of GCA. Ultrasound, given its availability and cost-effectiveness, is recommended as the first-line diagnostic modality, with MRI or FDG-PET as alternatives [7].

In this short communication, we analyze the pattern of cranial artery involvement and the prevalence of isolated occipital artery inflammation on MRI in patients diagnosed with GCA, PMR and GCA-PMR.

A total of 90 patients with clinically diagnosed GCA, PMR and GCA-PMR were included in the trial. 51/90 (57%) patients were diagnosed with GCA, 20/90 (22%) were diagnosed with PMR, 10/90 (21%) patients presented with clinical characteristics of both GCA and PMR (GCA-PMR). 53/90 (59%) of all patients with diagnosis of GCA, PMR or GCA-PMR showed inflammatory changes of any of the scalp arteries on contrast-enhanced high-resolution MRI (left superficial temporal artery only: 1/90 (1%); right occipital artery only: 4/90 (4%); left occipital artery only: 3/90 (3%); bilateral occipital artery: 2/90 (2%); superficial temporal artery and occipital artery, each of at least one side: 43/90 (48%)). 41/51 (80%) of patients with GCA showed signs of inflammation of any of the scalp arteries (left superficial temporal artery only: 1/51 (2%); right occipital artery only: 3/51 (6%); left occipital artery only: 3/51 (6%); bilateral occipital artery: 1/51 (2%); superficial temporal artery and occipital artery, each of at least one side: 33/51 (65%)). 2/20 (10%) of patients with PMR showed signs of inflammation of the superficial temporal artery and the occipital artery, each of at least one side. 10/19 (53%) of patients with GCA-PMR showed signs of inflammation of any of the scalp arteries (right occipital artery only: 1/19 (5%); bilateral occipital artery: 1/19 (5%); superficial temporal artery and occipital artery, each of at least one side: 8/19 (42%)). Table 1 summarizes the pattern of involvement of the scalp arteries for each diagnosis (GCA, PMR, GCA-PMR). Figure 1 shows contrast-enhanced MRI of two GCA patients with isolated inflammatory involvement of the occipital artery.

Our study highlights the significant role of MRI in diagnosing GCA and silent GCA in PMR phenotype, particularly in detecting isolated inflammatory involvement of the occipital artery - a phenomenon that has been relatively underexplored in the literature thus far. Previous data have already shown that there is a significant number of patients with occipital artery involvement [11] and atypical manifestations of GCA, characterized by sparing of the temporal arteries and instead inflammatory involvement of other arterial structures of the head, such as the facial artery and the occipital artery [12]. One study demonstrated that in a GCA cohort, 31.2% of patients showed inflammatory involvement of the occipital artery, and that 18.2% of GCA patients with a negative ultrasound of the temporal artery exhibited inflammatory changes in the ultrasound of the facial artery or the occipital artery [12]. However, there is little data in the existing literature regarding the specific involvement patterns of the scalp arteries and regarding the proportion of patients with isolated inflammatory involvement of the occipital artery. Clinical symptoms usually depend on the location of the inflamed vessels. Similar to typical temporal arteritis, patients with inflammatory involvement of the facial artery typically report jaw claudication, while those with inflammatory involvement of the occipital artery generally complain of occipital headaches and nuchal pain, along with swelling and tenderness of the occipital artery [11‐13]. Although it is technically possible to biopsy the occipital artery [14, 15], this is not part of the standard workup. Instead, the biopsy is usually performed on the temporal artery due to its easier accessibility and the more familiar surgical technique [14]. Patients with isolated affection of the occipital artery would be missed by standard temporal artery biopsy but can be identified through ultrasound examination [12]. However, the anatomical course of the occipital artery along the back of the head, particularly considering the usual presence of hair in this area, makes it difficult to access for ultrasound examination. Although studies have shown that the occipital artery is at least partially accessible by ultrasound [11, 12], ultrasound examination of the occipital artery is not routinely included in standard ultrasound protocols for the assessment of large-vessel vasculitis in many centers, including our Vasculitis Center at the University Hospital of Wuerzburg.

Integrating MRI into the diagnostic workup for GCA and PMR presents both opportunities and challenges. While MRI has demonstrated considerable promise in identifying inflammatory changes of the cranial arteries with comparable diagnostic accuracy to ultrasound [16], its accessibility and cost may limit widespread adoption compared to ultrasound as a more readily available and cost-effective modality. Furthermore, it has been shown that ultrasound detects vascular changes more frequently than MRI when not only the scalp arteries but all supra-aortic arteries are taken into account, supporting its role as a crucial part of a comprehensive diagnostic workup [17].

Our findings suggest that MRI should be considered an additional tool in the diagnostic workup for suspected GCA or silent GCA within the PMR phenotype, particularly in or reserved for cases with inconclusive temporal artery biopsy results or discrepancies between biopsy, ultrasound findings, and clinical symptoms. This targeted application not only optimizes the use of MRI but also enhances diagnostic accuracy in complex cases. To better integrate MRI into routine practice, it is essential to establish clinical protocols that clearly define when MRI should be employed, especially in scenarios where ultrasound may be less effective. Furthermore, increasing clinician awareness of the benefits of MRI and exploring cost-effective solutions - such as a streamlined, focused examination protocol - can improve accessibility, ultimately leading to more accurate and timely diagnosis of GCA.

Our observation of isolated occipital artery inflammation in a notable proportion of patients challenges the prevailing view that GCA primarily affects the temporal arteries. Although temporal artery involvement remains a hallmark of GCA, our findings suggest a significant number of patients exhibit atypical manifestations that spare the temporal arteries, potentially missed by standard diagnostic approaches. The utility of MRI in detecting isolated occipital artery inflammation carries significant clinical implications. Given the limitations of temporal artery biopsy, such as sampling error and false-negative results [18], MRI emerges as a valuable complementary tool for identifying extracranial arterial involvement in GCA and PMR. By facilitating the prompt identification of patients with active inflammation, MRI can help initiate appropriate therapy and reduce the likelihood of missed diagnoses. This aligns with recent literature advocating for the incorporation of an independent reference diagnosis into the diagnostic algorithm for GCA, particularly in cases where traditional modalities are inconclusive [18]. Notably, FDG-PET serves as a limited alternative to MRI due to its restricted resolution and availability, relatively high costs, and associated radiation exposure.

Additionally, our findings underscore the importance of considering the heterogeneous presentations of patients with GCA and/or PMR and the necessity of a multidisciplinary approach to diagnosis and management [1]. Clinicians should maintain a high index of suspicion for GCA and/or PMR in patients presenting with non-specific clinical symptoms, even in the absence of temporal artery abnormalities [19].

Despite the strengths of our study, including a relatively large sample size and systematic MRI evaluation, several limitations warrant consideration. The retrospective design inherently carries the risk of selection bias and incomplete data capture. Additionally, the lack of a control group limits our ability to assess the specificity of isolated occipital artery inflammation as a diagnostic marker for GCA and PMR. Future prospective studies that incorporate longitudinal follow-up and comparisons with healthy controls are warranted to further elucidate the clinical significance of isolated occipital artery inflammation in GCA and PMR. Another limitation is that we did not conduct targeted ultrasound of the occipital artery in the presented cohort, which means we cannot determine how many patients we might have identified as positive for vasculitis without the use of MRI, but with a more comprehensive ultrasound examination. Moreover, while we focused on the isolated inflammatory involvement of the occipital artery in this study, it remains to be determined whether other vessels within the GCA spectrum could also be affected by isolated inflammatory changes, potentially resulting in patients being overlooked by conventional diagnostic algorithms.

In conclusion, our study provides novel insights into the role of MRI in diagnosing GCA and PMR, particularly in detecting isolated inflammatory involvement of the occipital artery. By highlighting the prevalence, variability and clinical significance of extracranial arterial involvement patterns, our findings underscore the importance of a comprehensive diagnostic approach that integrates clinical and laboratory assessments, and, when inconclusive, various imaging modalities. Moving forward, further research is needed to validate the diagnostic utility of MRI and optimize the diagnostic algorithm for GCA and PMR.