Determinants of diagnostic yield and scan quality of [18F]FDG PET/CT in critically ill patients suspected of infection or inflammation of unknown origin
Authors' conclusion
[18F]FDG PET/CT has a high diagnostic yield in critically ill patients. Diagnostic yield, scan quality and myocardial suppression are influenced by patient characteristics and preparation.
Study context
- Population
- Adult ICU patients suspected of infection or inflammation of unknown origin undergoing [18F]FDG PET/CT, four Dutch hospitals (2013–2023)162 indications
- Participants
- 151 patients · 169 scans
- Outcome
- Diagnostic yield: new conclusive diagnoses and/or therapy changes
- Follow-up
- Within 72 h
Methodological detailsmulticentre
- Multicentre
Main finding
As reported in the abstract
Diagnostic yield (per indication)
Adjusted odds ratio (95% CI)
Diabetesfewer new diagnoses
0.35(0.15–0.80)
Partial-body imaging vs Total-body imagingfewer new diagnoses
0.48(0.24–0.99)
Longer pre-scan ICU staypredicted therapy change
1.03(1.01–1.05)
- Conclusive diagnosis
- 109 indications (67%)
- New diagnoses
- 79 (49%)
- New diagnoses involving musculoskeletal system
- n = 39; 24%
- New diagnoses involving respiratory tract
- n = 31; 19%
- Therapy change
- 70 indications (43%)
Observational data: associations, not effects.
Secondary outcomes
Scan quality
| aOR (CI) | ||
|---|---|---|
| Higher BMIImproved scan quality | 1.08 (1.00–1.17) | |
| Lower serum creatinineLog-transformed; improved scan quality | 0.36 (0.20–0.68) | |
| 0.20.5125Adjusted odds ratio (log scale) |
Adequacy of myocardial suppression
| aOR (CI) | ||
|---|---|---|
| Male sexSuboptimal myocardial suppression | 0.45 (0.21–0.96) | |
| Pre-scan unfractionated heparinSuboptimal myocardial suppression | 0.43 (0.19–0.98) | |
| 0.20.5125Adjusted odds ratio (log scale) |
Original publication
Determinants of diagnostic yield and scan quality of [18F]FDG PET/CT in critically ill patients suspected of infection or inflammation of unknown origin. Crit Care. Published 10 April 2026. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →