Abstract Background Vagus nerve activation impacts inflammation. Therefore, we hypothesized that vagal nerve stimulation (VNS) influenced arterial wall inflammation as measured by ¹⁸F-FDG uptake. Results Ten patients with left-sided VNS for refractory epilepsy were studied during stimulation (VNS-on) and in the hours after stimulation was switched off (VNS-off). In nine patients, ¹⁸F-FDG uptake was measured in the right carotid artery, aorta, bone marrow, spleen, and adipose tissue. Target-to-background ratios (TBRs) were calculated to normalize the respective standardized uptake values (SUVs) for venous blood pool activity. Median values are shown with interquartile range and compared using the Wilcoxon signed-rank test. Arterial SUVs tended to be higher during VNS-off than VNS-on [SUV_max all vessels 1.8 (1.5–2.2) vs. 1.7 (1.2–2.0), p = 0.051]. However, a larger difference was found for the venous blood pool at this time point, reaching statistical significance in the vena cava superior [_meanSUV_mean 1.3 (1.1–1.4) vs. 1.0 (0.8–1.1); p = 0.011], resulting in non-significant lower arterial TBRs during VNS-off than VNS-on. Differences in the remaining tissues were not significant. Insulin levels increased after VNS was switched off [55.0 pmol/L (45.9–96.8) vs. 48.1 pmol/L (36.9–61.8); p = 0.047]. The concurrent increase in glucose levels was not statistically significant [4.8 mmol/L (4.7–5.3) vs. 4.6 mmol/L (4.5–5.2); p = 0.075]. Conclusions Short-term discontinuation of VNS did not show a consistent change in arterial wall ¹⁸F-FDG-uptake. However, VNS did alter insulin and ¹⁸F-FDG blood levels, possibly as a result of sympathetic activation.