A Retrospective Study of Physiologic Markers to Assess Radiopharmaceutical Infiltration Compared to External Injection Monitoring

Background: Infiltration is a significant issue in PET/CT imaging because it can lead to uncertainty in the accuracy of values recorded from regions of interest drawn on the data and underestimation of standard uptake values (SUVs). This issue is further complicated because the infiltration site is often not visible in the PET/CT field of view. A commercial device is available that can be used for injection monitoring as well as providing a score related to the quality of the injection (Lara, Lucerno Dynamics, LLC). This study examines physiologic regions drawn in PET data and compared between severely infiltrated injection and good quality injections to determine if any physiologic parameters alone can serve as an indicator of infiltration. Methods: The first phase of this study reviewed 897 SUV max values reported from the right lobe of the liver as this is the most commonly reported SUV for routine PET. Reported SUVs were compared to injection scoring from the injection monitoring device to determine if any significant correlation existed. The second phase of the study examined additional physiological parameters. Five patients (5 infiltrated injections, 5 good quality injections )were identified and regions of interest (ROIs) drawn on the areas listed in the attached table. The five infiltrated studies were the five worst infiltrations recorded at our site over a period of approximately 1.5 years while the 5 good quality injection were five of our best over the same period. Statistical analysis using t-tests were performed to compare means. Results: Phase one liver assessments indicated no significant correlations between injection quality score and right hepatic lobe SUV maximum measurements. Comparison of means in the expanded ROI analysis in phase two indicated that statistically significant differences in mean (p<0.05) were only observed in the cerebellum, vertebrae, and lung. Although these showed statistical significance there is still significant overlap of measured values between scores that indicate significant infiltration and those that indicate good quality injections. Measurements of the cerebellum showed the greatest separation of patient populations.

Conclusions: Physiologic parameters may be helpful in assessing potential infiltration, however, commonly reported areas such as the liver do not appear to have any significant correlation with injection quality making it an unreliable marker of infiltration. Other regions may be more useful, however, the overlap of values for infiltrated and non-infiltrated scans make it difficult to simply assign a threshold for rapid assessment. External injection monitoring seems to be the most reliable method for assessing PET radiopharmaceutical injection quality for individual patients.