# Clinical revenues of selective use of [18F]-FDG-PET/CT scanning in the management of Staphylococcus aureus bacteremia

## TL;DR

This study evaluates how using PET/CT scans selectively in managing Staphylococcus aureus bacteremia impacts clinical outcomes and treatment decisions.

## Contribution

The study quantifies the clinical impact of selective PET/CT use in SAB patients using NNT-scan and identifies factors associated with beneficial outcomes.

## Key findings

- PET/CT detected new infection foci in 73 of 143 patients (NNT-scan ≈ 2).
- PET/CT led to new interventions in 33 patients (NNT-scan ≈ 4) and antimicrobial therapy changes in 44 patients (NNT-scan ≈ 3).
- CRP > 200 mg/L and positive follow-up BCs at 48 h were linked to interventions after PET/CT.

## Abstract

Although [18F]-FDG-PET/CT scanning (PET/CT) is recommended for managing Staphylococcus aureus bacteremia (SAB), its added value remains debated. This study investigated the clinical revenues of selective PET/CT use in SAB by considering three consequence-categories: detection of new infection foci, performance of new interventions, and alterations in antimicrobial therapy.

All adult patients with ≥ 1 blood culture (BC) positive with Staphylococcus aureus admitted in a Dutch academic center between 201711 and 202311 were identified. Standard practice was to order PET/CT for patients with community acquired SAB and/or positive BCs after ≥ 48 h of treatment, or if multiple foci, or persistent fever, or endocarditis were present. Clinical- and laboratory data were obtained from electronic health records. Numbers-needed-to-scan (NNT-scan) were calculated for each consequence-category. Regression analyses were performed to identify variables correlated with consequence-bearing PET/CT.

Of 397 SAB patients, 143 (36%) underwent PET/CT. This led to detection of new foci in 73/143 patients (NNT-scan ≈ 2), new interventions in 33/143 patients (NNT-scan ≈ 4), and a change in antimicrobial therapy in 44/143 patients (NNT-scan ≈ 3). A CRP > 200 mg/L at presentation and positive follow-up BCs at 48 h were independently associated with interventions following PET/CT (adjusted OR and 95%CI 3.2 (1.2–8.3) and 2.6 (1.0-6.7) respectively). PET/CT results instigated changes in antimicrobial therapy predominantly in patients < 65 years and those with a CRP < 100 at presentation.

Selective PET/CT ordering in real-life practice resulted in a relatively low NNT-scan across all consequence-categories. Further research is warranted to optimize patient selection for PET/CT using clinical parameters or profiles.

The online version contains supplementary material available at 10.1007/s10096-025-05052-5.