We present the case of a 63 year old man who died of severe septic shock in the setting of agranulocytosis induced by dipyrone (metamizole). The patient had previously developed agranulocytosis after dipyrone exposure 18 months prior to this. The case illustrates the seriousness of dipyrone-induced agranulocytosis, highlights the risks associated with re-exposure and underlines the need for excellent communication between treating physicians and their patients. The possible underlying mechanisms, epidemiology and management of dipyrone-induced agranulocytosis are discussed.

Background:For the diagnosis of giant cell arteritis (GCA) several diagnostic tools do exist such as18F-FDG-PET/CT (PET) with excellent diagnostic accuracy for the larger vessels and ultrasound for the temporal arteries (TA). Recent data propose that PET is able to detect vasculitis in vessels as small as the TA (1). Comparison of PET, ultrasound (US) and histology of the TA on a segment level has not been done.Objectives:To describe diagnostic accuracy of PET of the TA in a vasculitis university clinic and to analyse strength and limitations of PET by comparing18F-FDG uptake to US and histology results on a segment level.Methods:We analysed patients, included in our ethical board approved local prospective GCA cohort having received a PET in between 2015 and 2019 because of suspected GCA.PET of the TA was performed using time-of flight technique and was scored ‘vasculitis’ if tracer uptake was higher than in the surrounding tissue. Standard uptake value (SUV) measurement in the trunk (T), parietal branch (PB) and frontal branch (FB) of the TA was recorded. US was performed for each branch.Results:From 37 consecutively recruited patients, GCA was confirmed in 19 patients and excluded in 18 patients which served as controls (Table 1). PET of the TA showed vasculitis in 12/19 GCA patients and in 1/18 controls. Median SUVmaxof all vasculitic FB (n=18) was 2.91, 2.20 for the T (n=14) and 2.34 for the PB (n=5).Table 1.Patient characteristics. Data are expressed as number (%) or median (interquartile range)GCA (n=19)(51%)Control (n=18)(49%)P-valueFemale11 (57)9 (50)0.64Median age (years) at PET73 (64-78)62.5 (57-71.75)0.04Amaurosis fugax/Loss of vision9 (47)9 (50)>0.99New onset headache13 (68)11 (61)0.74Jaw claudication7 (37)3 (16)0.38Scalp tenderness/pathological TA7 (37)6 (33)0.54Proximal muscle pain11 (58)8 (44)0.29Fever1 (5)5 (28)0.06Median Erythrocyte sedimentation rate (mm/h)73 (48-90)50 (28.5 - 68.5)0.06Median C-reactive protein (mg/L)66 (29-105)46 (13-133)0.5716 of the 19 GCA patients received US of the TA and 9 showed vasculitic findings. From the control group 2 patients showed vasculitic findings. Most often vasculitic findings were localized in the FB (n=16), followed by the T (n=13) and the PB (n=13).In the 16 patients that received US, diagnostic sensitivity and specificity of temporal PET for GCA within the TA was 56% and 94% and of US 56% and 89%, respectively.Whereas US detects vasculitis in comparable frequencies in all TA branches, PET recorded vasculitis less often in the PB (only 4 of the 13 in US vasculitic PB). Indeed, the median diameter of all PET positive TA branches, measured in the US, was higher (2.00mm) compared to PET negative branches (1.50mm). Vasculitis was confirmed histologically in 9 of the 13 biopsied patients. Only 2/9 patients showed vasculitis in the preceding PET in the biopsied branch.Conclusion:High diagnostic accuracy for temporal arteries supports PET as an ‘all-in-one’ exam for GCA. A limitation might be the vessel diameter, as sensitivity of PET for vasculitis of the small parietal branch is low. Thus, in cases with high suspicion of GCA despite a negative PET, US of the TA and or biopsy might enhance diagnostic sensitivity.References:[1]Nielsen et al. Simple dichotomous assessment of cranial artery inflammation by conventional 18F-FDG PET/CT shows high accuracy for the diagnosis of giant cell arteritis: a case-control study.Eur J Nucl Med Mol Imaging. 2019;46(1):184-193.doi:10.1007/s00259-018-4106-0Disclosure of Interests:None declared