E1578. Multicenter Investigation Into Frequency and Significance of Infiltration/Extravasation Events in PET Imaging
  1. Twyla Bartel; Global Advanced Imaging, PLLC
  2. Dusty York; Chattanooga State Communicty College
  3. Christine Mundt; University of Iowa Hospitals and Clinics
  4. John Sunderland; University of Iowa Hospitals and Clinics
There have been recent concerns raised regarding potential clinical risks from potential infiltration during the injection of diagnostic radiopharmaceuticals. The purpose of this study is to objectively investigate the frequency, severity, and clinical impact of injected dose (ID) infiltrations in a multicenter PET/CT 1000 patient cohort. Negative potential clinical outcomes include tissue damage, uninterpretable images, and quantitative SUV bias.

Materials and Methods:
1000 PET/CT patient studies were collected from ten different imaging institutions. To avoid bias a variety of institutions were chosen including, an academic medical center (n = 1), private radiology groups (n = 2), private oncology groups (n = 2), community hospital (n = 1), multispecialty groups (n = 2), and a research facility (n = 1). Approximately 50 patients from smaller clinics were combined. At each institution, consecutive patients that had the injection site in the field-of-view were chosen. The radiopharmaceutical, ID (Bq), time of injection, time of imaging, injection site, injection type (e.g., IV catheter, port…) were recorded. PET/CT images were analyzed by one of two expert readers. Volumes of interest (VOIs) were drawn around the injection site to include all injection site activity potentially associated with the injection. A similar VOI was drawn on the contralateral side. Mean and maximum radioactivity concentration (Bq/mL) and VOI was measured. Net activity at the injection site was calculated as the difference between activity at the injection site and that on the contralateral side.

Of the 1000 injections measured, 952 were 18F-FDG, 17 68Ga-DOTATATE or -DOTATOC, 18F-Fluciclovine, and 13 18F-PSMA. Interestingly, 929 used IV catheters, 68 used port access, and 3 used a PICC line. No direct sticks or butterflies were noted which may help explain why the infiltration rate was significantly lower than reported in the literature. In the oncologic patient, vein access can be challenging; 645 patients were injected in the antecubital vein, 135 in the hand, 111 in the forearm, and 34 in the wrist. There were zero infiltrations with more than 1.0% ID, net, at the injection site. In six cases, activity in excess of 10 µCi (370 kBq) was measured at 11, 11, 23, ,34, 41, and 44 µCi representing 0.12, 0.20, 0.27, 0.47 , 0.39, and 0.92% ID, respectively. Absorbed dose estimates using recent literature infiltration dose calculation paradigms suggest all of these were below 0.1 Gray (100 mSv), significantly below any threshold for tissue damage, compromised image quality, or SUV quantitation >4%.

In this multicenter study, attempts were made to fairly sample the clinical PET imaging space to inform concerns about infiltration. This study showed no evidence of a significant problem and demonstrated a substantially lower infiltration rate than previously reported. This may be due to the uniform adherence to the recommended practice of using IV catheterization for injection of PET radiopharmaceuticals which may be more universally practiced than previously thought.