FDG PET quality in hyperglycemic patients: Is routine rescheduling necessary?

Introduction: High quality F-18 FDG PET/CT imaging is essential in the management of patients with cancer and other conditions. PET/CT image quality is known to decrease in patients with elevated blood glucose levels,and subsequently PET/CT studies are often canceled or rescheduled in patients with blood glucose levels over 200 mmol/L.^1-2 This delay may inconvenience patients and negatively impact patient management. In this study, we evaluate the quality of PET/CTs performed in patients with elevated blood glucose levels and evaluate factors associated with non-diagnostic studies.

Methods: We performed an IRB-exempt single-institution retrospective review of patients with glucose levels of 200 mmol/L and above who underwent PET imaging between 10/01/2019 and 9/30/2021. Additional clinical characteristics including BMI, sex, age, medications (insulin, metformin, and steroids), indication for exam, and diabetic status (Type 1 vs. Type 2 and hemoglobin) were obtained from the medical record. A qualitative consensus evaluation of PET/CT image quality was performed by 2 board certified Nuclear Radiologists blinded to blood glucose level. A numeric scale from 1-3 was used with 1 being a diagnostic scan without visible limitations, 2 being diagnostic but not ideal, and 3 being of inadequate quality requiring additional/repeat imaging. Statistics including mean and standard deviation were performed on Microsoft Excel.

Results: 108 PET/CT exams met the inclusion criteria for this study. Average patient blood glucose level was 215 mmol/L (SD 17), average hemoglobin A1c was 8.3% (SD 1.6), average BMI was 32.6 kg/m (SD 6.83), and average age was 62 years old (SD 12.8). At the time of their PET/CT scan, 17 patients were taking corticosteroids (15.7%), 43 were taking insulin (43.5%), and 47 were taking metformin (43.5%). For diabetic status, 9 had a type 1 diabetes diagnosis (~8%), 93 had a type 2 diabetes diagnosis (86%) and the rest were non-diabetic (~6%). Of the 108 studies, 60 (55.6%) were scored as 1 (normal quality), 42 (38.9%) were scored as 2 (suboptimal but diagnostic), and only 6 (5.5%) were scored as 3 (nondiagnostic). Of the 6 patients with nondiagnostic imaging, only 3 of them (50%) had blood glucose levels of less than 210 mmol/L. Amongst outpatients, the rate of non-diagnostic scans was 2.2% (2/93), versus 26.7% (4/15) for inpatients (p<0.001). Furthermore, only 13 exams (12%) included a comment about quality such as "diffuse muscular uptake, limitations of the exam, altered biodistribution, and poor technical quality," in the text of the didactic report.

Conclusions: Most PET/CTs performed in patients with elevated blood glucose were of diagnostic quality, especially in the outpatient setting. These results suggest that further research re-evaluating the optimal upper blood glucose threshold in FDG PET is warranted. In the future, it may be possible to identify a subset of hyperglycemic patients in whom FDG PET can be performed without the need for rescheduling.

• Download figure
• Open in new tab
• Download powerpoint