18 F-FDG-PET/CT assessment of Takayasu arteritis and effect of time-of-flight reconstruction: an observational study

Background To image reconstruction on assessment of Takayasu arteritis (TKA). We reviewed 14 patients (15 cases) who underwent TOF-PET/CT. PET images were reconstructed using ordered-subsets expectation maximisation ± TOF. Uptake in 10 vascular regions was assessed using the four-point visual grading system (3, higher than liver uptake; 0, no uptake). The grades in the TOF-PET/CT and non-TOF-PET/CT groups were compared using the sign test. Relationships between visual grade and arterial FDG uptake pattern in the 10 regions and clinical findings were assessed. Results Except for the highest visual grade, the visual grades in all regions were significantly higher in the TOF-PET/CT group (p=0.003). FDG uptake pattern and region were not significant determinants of the effect of TOF reconstruction. Assessment of disease activity was inconsistent between the groups in 6 (4%) of 150 regions when grade 2/3 was defined as active and in 5 (3.3%) of 150 when grade 3 was defined as active. were higher Use same reconstruction and after recommended when evaluating the response to treatment.


Background
To assess the impact of time-of-flight (TOF) positron emission tomography (PET)/computed tomography (CT) image reconstruction on assessment of Takayasu arteritis (TKA). We reviewed 14 patients (15 cases) who underwent TOF-PET/CT. PET images were reconstructed using orderedsubsets expectation maximisation ± TOF. Uptake in 10 vascular regions was assessed using the fourpoint visual grading system (3, higher than liver uptake; 0, no uptake). The grades in the TOF-PET/CT and non-TOF-PET/CT groups were compared using the sign test. Relationships between visual grade and arterial FDG uptake pattern in the 10 regions and clinical findings were assessed.

Results
Except for the highest visual grade, the visual grades in all regions were significantly higher in the TOF-PET/CT group (p=0.003). FDG uptake pattern and region were not significant determinants of the effect of TOF reconstruction. Assessment of disease activity was inconsistent between the groups in 6 (4%) of 150 regions when grade 2/3 was defined as active and in 5 (3.3%) of 150 when grade 3 was defined as active.

Conclusions
Visual grades were significantly higher in all regions when assessed by TOF-PET/CT. Use of the same reconstruction algorithm before and after treatment is recommended when evaluating the response to treatment.

Background
Takayasu arteritis (TKA) is a chronic granulomatous large-vessel vasculitis that is rare and has severe potentially lethal complications, including aortic dissection/rupture, refractory hypertension, and myocardial infarction. 1,2 Early diagnosis and detection of inflammation in the arterial wall is essential for improvement of the short-term and long-term outcomes in patients with TKA. However, the disease is difficult to evaluate because it has no specific symptoms and there are no laboratory biomarkers that accurately reflect its activity or progression.
3 functional imaging modality that has an important role in the management of patients with cancer.
Moreover, FDG-PET/CT has been demonstrated to be an effective imaging modality for assessment of TKA. 3−8 Time-of-flight (TOF) reconstruction uses the time difference between the arrival of two photons and localises the emission point with much higher precision than non-TOF detection. 9 Previous studies have demonstrated the effects of TOF on PET/CT imaging in the fields of oncology and cardiology. 10−13 However, to the best of our knowledge, the impact of TOF reconstruction on evaluation of TKA has not been reported previously. Therefore, the aim of this study was to investigate how TOF reconstruction affects PET/CT assessment of patients with TKA.

Patients
Consecutive patients who underwent TOF-FDG-PET/CT at Tokyo Medical and Dental University for assessment of large-vessel vasculitis between July 2018 and January 2019 were retrospectively reviewed. Patients who were under the age of 20 years and those who had undergone aortic surgery before FDG-PET/CT were excluded.
We evaluated disease activity using the following four National Institutes of Health (NIH) criteria devised by Kerr et al. 14 : 1, systemic features; 2, features of vascular ischemia or inflammation; 3, laboratory data; and 4, typical angiographic features. Cases with new occurrence or aggravation of at least two of these four features were defined as active. Our assessment of laboratory parameters included the C-reactive protein or matrix metalloproteinase-3 level measured within 3 weeks before or after PET/CT in addition to the erythrocyte sedimentation rate. Angiographic features were assessed on CT or magnetic resonance images acquired in the 2 months before or after PET/CT scanning.

PET/CT protocol
All FDG-PET/CT images were obtained using a Celesteion scanner (Canon Medical Systems, Otawara, Japan). After fasting for at least 4 hours, all patients received an intravenous injection of 3.7 MBq/kg of FDG. Whole-body CT data were obtained for attenuation correction and anatomical co-registration with free breathing. The technical CT scan parameters were as follows: tube voltage, 120 kV; field of view, 550 mm; pitch, 16.0; and slice thickness, 2.0 mm. After CT, whole-body PET scanning was started about 60 min after the injection of FDG. PET emission scans were acquired in 3D mode using the following parameters: 2 min per bed position, a pixel size of 4 mm, and a Gaussian filter size of 6 mm. The PET images were reconstructed using two algorithms, i.e., ordered-subsets expectation maximisation (OSEM) and OSEM + TOF. The reconstruction parameters were two iterations and 10 subsets for both algorithms.

Imaging analysis
All PET/CT images were analysed and scored by two nuclear medicine radiologists working independently. The four-point grading system was used for visual assessment of vascular FDG uptake (3, vascular uptake higher than liver uptake; 2, vascular uptake similar to liver uptake; 1, vascular uptake lower than liver uptake; 0, no vascular uptake) [3][4][5]. Ten vascular regions, i.e., the ascending aorta, aortic arch, innominate artery, right and left common carotid arteries, right and left subclavian arteries, descending aorta, abdominal aorta, and pulmonary artery, were visually assessed in each patient. When FDG uptake was detected in a specific vascular region, we evaluated whether it was diffuse or patchy. When there was a discrepancy, the final decisions on visual grade and uptake pattern were reached by consensus. In each patient, the highest visual grade in the 10 vascular regions was defined as the final visual grade. In all regions, a difference in visual grade was identified by subtraction of the grade for non-TOF (OSEM)-PET/CT from that for TOF (OSEM + TOF)-PET/CT.

Statistical analysis
The sign test was used to compare the visual grades between the TOF and non-TOF groups. Fisher's exact test was used to compare the diffuse pattern with the patchy pattern and to compare the 10 vascular regions. All statistical analyses were performed using R 3.5.1 software (R Foundation for Statistical Computing, Vienna, Austria). A p-value < 0.05 was considered statistically significant.

Results
Eighteen patients (19 cases) met the study inclusion criteria. Four patients (4 cases) were excluded (3 because of previous aortic surgery and 1 who was below the age limit). The characteristics of the 14 patients (15 cases) included in the analysis are shown in Table 1. The mean blood glucose level was 102.2 ± 10.8 mg/dl. The mean fasting duration was 391 ± 85 min (data for 4 patients were excluded from the calculation because their fasting duration was over 4 hours but their last mealtime was uncertain). We did not assess disease activity in 5 cases because of lack of laboratory and angiographic data; in 7 of 10 cases, we were able to assess disease activity but not necessarily accurately because of lack of angiographic data. The final visual grades were not significantly different between the non-TOF and TOF-PET/CT groups (p = 1.0). The visual grades for the 10 vascular regions in all patients are shown in Table 2. The visual grades were significantly higher in the TOF-PET/CT group than in the non TOF-PET/CT (p = 0.003). Table 2 Visual grade values in 150 regions The relationship between change in visual grade and symptoms is shown in Table 3. The visual grade was elevated from 0 or 1 (non-TOF-PET/CT) to 2 (TOF-PET/CT) for 6 regions in 2 cases and from 2 to 3 for 4 regions in 3 cases. There was no significant difference in the distribution of differences in visual grade between the patchy and diffuse uptake patterns (p = 0.66 in the non-TOF group and p = 0.83 in the TOF group; Fig. 1).
There were significant between-group differences in the distribution of visual grades in the 10 vascular regions (p = 0.002 in the non-TOF group and p = 0.011 in the TOF group). The presence of FDG uptake was markedly more common at the aortic arch and innominate artery in both the TOF and non-TOF groups (Fig. 2). There was no significant difference in the distribution of differences in the visual grade between the 10 vascular regions (p = 0.37, Fig. 3). A representative case (case 2) is shown in Fig. 4.

Discussion
This study investigated the effect of PET/CT reconstruction using TOF on assessment of patients with TKA when using the established four-point visual grading system. Although TOF reconstruction had no significant impact on the final visual grade, we found that the visual grade for all regions assessed was significantly higher in the TOF-PET/CT group than in the non-TOF-PET/CT group. An earlier study demonstrated the effect of TOF reconstruction on uptake in the normal aorta 15 ; however, subjects with suspected large-vessel vasculitis were excluded in that study. Therefore, the present study is the first to investigate the effect of TOF reconstruction on assessment of TKA.
An earlier review showed that a visual grade of 2 was possibly indicative and that a grade of 3 was positive when assessing disease activity in large-vessel vasculitis. 3 In our study, 18/150 regions were regarded as active on non-TOF-PET/CT and 24/150 were regarded as active on TOF-PET/CT when grade 2 or 3 uptake was defined as active ( Table 2). Assessment of disease activity by non-TOF-PET/CT and TOF-PET/CT was inconsistent for 6 (4%) of 150 regions when grade 2 or 3 was considered as active. These regions were shown in case 4 (4 regions) and case 14 (2 regions). In these cases, TOF reconstruction had no effect on the final visual grades; in 2 of 10 cases there was a discrepancy between the final visual grade and disease activity according to the NIH criteria (cases 3 and 14).   Fig. 4). These findings indicate that use of different reconstruction algorithms before and after treatment may lead to incorrect evaluation of the response to therapy in patients with TKA. Therefore, we recommend use of the same reconstruction algorithm when evaluating their response to treatment. Table 3, the visual grade increased from 0/1 to 2 in 2 cases (4 and 14) when TOF-PET/CT was used. In case 4, new-onset bruit and vascular obstruction in the right subclavian region were detected and the visual grade was elevated near the right subclavian artery (innominate and right common carotid arteries), even though the visual grade for the right subclavian artery in this case was assessed as 0 by both reconstruction algorithms. In case 14, aortic surgery was performed 37 days after PET/CT, at which time pathology specimens were obtained from the ascending aorta, left subclavian artery, and left common carotid artery. Infiltration of focal lymphocytes and neutrophils with disruption of elastic fibres and broad adventitial fibrosis were observed in the left common carotid artery, indicating coexistence of acute and chronic inflammation. 16 Elevation of the visual grade may reflect mild chronic inflammation in these regions. Moreover, the visual grade was elevated from 2 to 3 on TOF-PET/CT in 4 regions (Table 3). Vascular pain was observed in 3 of these 4 regions (the right subclavian artery in case 1, left common carotid artery in case 2, and right common carotid artery in case 13). The symptomatology suggested acute inflammation in these regions, and TOF-PET/CT may reflect acute inflammation more accurately. TOF reconstruction has been reported to increase the standard uptake value for lymph node metastases, 11,17 and this finding is compatible with our present results. In previous studies that used semi-quantitative analysis, there was no significant difference in FDG uptake in the normal aorta or liver between non-TOF-PET/CT and TOF-PET/CT. 15,18 Furthermore, Daube-Witherspoon et al. demonstrated that TOF reconstruction improved the accuracy and precision of measurement of FDG uptake using clinical whole-body patient data with embedded artificial lesions that had known activity uptake. 19 Therefore, on the basis of the above findings, elevation of the visual grade from non-TOF-PET/CT to TOF-PET/CT may reflect a more accurate assessment of FDG uptake in the arterial wall.

As shown in
There was no significant difference in the effect of TOF reconstruction on visual grade between the patchy and diffuse uptake patterns. As shown in an earlier phantom study, TOF reconstruction improved the contrast of hot lesions, particularly if they were small. 10 Therefore, we expected that the difference in visual grade for the patchy uptake pattern would be higher than that for the diffuse uptake pattern because the patchy uptake pattern was smaller. We speculate that the arterial wall lesions were so small that there was no significant difference between the patchy and diffuse uptake pattern.
In most cases, FDG uptake was detected at the aortic arch and innominate artery. This finding is consistent with previous studies showing that TKA involves predominantly the thoracic and cervical arteries in the Japanese population. 20,21 Finally, as shown in Fig. 3, the effect of TOF reconstruction was not significantly different between the 10 vascular regions assessed. This is in contrast with an earlier study that found a difference in the gain in the signal-to-noise ratio as a result of TOF reconstruction in three body regions, i.e., the head and neck, abdomen, and lung. 22 In that study, the gain in the abdominal lesions showed a clear increase as a function of body mass index (BMI) in the range of 1.1-1.8. In contrast, the gain in the head and neck lesions showed a slight increase as a function of BMI in the range of 1. This study has several limitations. First, the sample size was small, which was inevitable given the rarity of TKA. Second, patients with types of vasculitis other than TKA were included. Third, our investigation of the association between visual grade and clinical findings was inadequate because of the small sample size and our lack of ability to assess disease activity fully using the NIH criteria. A further prospective study would be required to investigate the relationship between the findings of TOF-FDG-PET/CT and the activity of TKA. Fourth, this study was performed at a single centre in a Japanese population. Therefore, studies in other populations may be needed. Fifth, we used the fourpoint visual grading system without semi-quantitative analysis. However, this grading system is recommended for assessment of TKA in the clinical setting. 3 The present study was the first to evaluate the effect of TOF reconstruction on assessment of TKA using this system.

Conclusions
In this study we found that the visual grades observed by TOF-PET/CT were significantly higher in all vascular regions than those observed by non-TOF-PET/CT in patients with TKA. Although evaluation of the activity of TKA does not change markedly when TOF-PET/CT is used, we recommend that the same reconstruction algorithm be used before and after treatment when possible to assess the response to therapy in patients with TKA.

Consent for publication
Not applicable

Availability of data and material
The data that support the findings of this study are available from the corresponding author on request.

Competing interests
The authors declare that they have no competing interests.