Comment on: A novel model to assess disease activity in Takayasu arteritis based on 18F-FDG PET/CT: a Chinese cohort study.

Dear Editor, We read with interest the recent article by Ma et al. [1] outlining a novel approach to the assessment of disease activity in Takayasu arteritis (TAK) using 18F-fluorodeoxyglucose positron emission tomography/computed tomography (18F-FDG PET/CT). TAK is an important cause of large vessel vasculitis (LVV), now recognized across the globe [2]. As it is a lifelong condition, the ability to better match treatment intensity to disease activity may result in improvements in patient outcomes, including quality of life. Current tools to quantify disease activity are flawed [3] and, as recognized by the authors, and reinforced by Campochiaro and Misra in their editorial [4], imaging metrics should be incorporated in any TAK disease activity score. In this article the authors used 18F-FDG PET/CT metrics together with several serological markers in their retrospective cohort of 91 patients with TAK. They found that a number of 18F-FDG standardized uptake value (SUV) metrics were higher in those with active disease compared with inactive disease. These were then combined with serological data to form a disease activity assessment model that outperformed the Kerr score in their cohort.

We note with particular interest that Ma et al. [1] utilized SUVmean and found that this outperformed SUVmax in distinguishing active from inactive disease. The SUV provides a numerical scale of PET activity. Each pixel is assigned an SUV, with ‘darker’ pixels (indicating increased tracer uptake) having higher scores. Previously, studies that aimed to quantify arterial PET activity utilized SUVmax or ratios involving SUVmax. Calculation of SUVmax involves identification of the pixel with the highest score within any given arterial segment. This value is then assigned to the whole segment. This method is useful but can overestimate disease activity within an arterial territory, particularly when tracer uptake is patchy, and can lead to misdiagnosis in conditions such as atherosclerosis. The SUVmean provides an average activity score for an arterial segment and thus provides a more accurate representation of the tracer uptake within a given area (Fig.  1). SUVmean may also be more sensitive to subtle changes in disease activity over time. This approach has recently been adopted for the assessment of aortic microcalcification, with promising results [5].

SUVmax  vs SUVmean for the assessment of PET activity in arterial segments

Here, both the normal arterial segment and segment with active LVV are correctly quantified by both SUVmax and SUVmean. However, the atherosclerotic segment is incorrectly quantified (high) by SUVmax but correctly quantified (low) by SUVmean.

While we congratulate the authors for evaluating SUVmean, it is not entirely clear how the authors have calculated this. They state that ‘an oval three-dimensional volume of interest was applied to measure the SUV in the vascular wall of the aorta and its branches’. Further clarification would be helpful to guide future studies. For an SUVmean to be accurate, we believe that the entire arterial segment of interest should be contoured carefully and, for the aorta at least, the lumen should be subtracted so as not to dilute the signal. We recognize that such practice is time consuming, but it may be potentially useful for monitoring disease activity over time in patients with TAK, and LVV more broadly.

As highlighted by Campochiaro and Misra [4], the use of hybrid PET/MRI as opposed to PET/CT may have particular utility in patients with TAK. The reduced radiation dose of PET/MRI permits interval scanning. This is of relevance to TAK, which is often diagnosed at a relatively young age (mean age at recruitment was 37 years in this cohort). The use of MRI also allows complementary assessment of the vessel wall, including mural thickness and increased mural signal on T2-weighted images [6]. The addition of an MR angiographic sequence allows detailed evaluation of luminal abnormalities (e.g. vessel occlusion/stenosis/aneurysm) and post-contrast assessment of mural enhancement. It is unclear whether the authors of this study included an angiographic component as part of their scanning protocol. Such an addition may add further weight to any composite disease activity score, as the development or progression of vascular lesions may suggest active disease.

Finally, we also note with interest that the ratio of female to male patients in this cohort (2.6:1) was much lower than in many historical cohorts (typically ∼6:1). This is an increasingly recognized phenomenon in some parts of the world and reinforces the fact that a diagnosis of TAK should not be discounted on the basis of male gender alone.

The current study adds to the growing body of evidence supporting a role for 18F-FDG PET in the longitudinal evaluation of disease activity in LVV. Utilization of the SUVmean in this setting is likely to be of particular value, allowing detection of subtle changes in inflammatory activity and distinguishing LVV from common mimics. We propose that hybrid PET/MRI, with inclusion of a dedicated angiographic sequence, may have particular utility in young patients with TAK.

Funding: No specific funding was received from any bodies in the public, commercial or not-for-profit sectors to carry out the work described in this article.

Disclosure statement: The authors have declared no conflicts of interest. All authors contributed to data acquisition and analysis, as well as drafting and critically appraising the manuscript.

Data availability statement

All data are included within the text.