Author + information
- Jean-François Sarrazin, MD⁎ (, )
- François Philippon, MD,
- Jean Guimond, MD,
- Michel Tessier, MD and
- Mikaël Trottier, MD
- ↵⁎Institut Universitaire de Cardiologie et, Pneumologie de Québec, 2725, chemin Sainte-Foy, Québec, QC, G1V 4G5, Canada
We thank Dr. Erba and colleagues as well as Dr. Sharma for their interest in our paper (1). In the letter from Dr. Erba and colleagues, they suggested different rates of false-positive and false-negative results with 18F-fluorodeoxyglucose (FDG) positron emission tomography (PET)/computed tomography (CT). We agree that diagnosis of device infection should not be limited to a single test. Conventional testing, clinical judgment, and practice according to guidelines remain critical. However, our published data and additional personal experience using this technology showed that the addition of PET/CT could be of great value in difficult cases. It is common to use a group of patients with higher disease prevalence to demonstrate a proof of concept. Tests were also performed in 2 lower-risk control groups with recent device surgery or no active infection. We are aware that Ploux et al. (2) had FDG uptake on leads in 7.5% of controls; however, no information is available on recent antibiotic use, whether attenuation-corrected or noncorrected images were used, and follow-up duration because this observation could be related to early signs of infection. The negative predictive value of PET/CT is useful, as demonstrated in our study of 10 patients having an excellent evolution with no late recurrence following conservative treatment at 15 months. Further studies are required to better understand causes for false-positive and false-negative results. Our study was not powered to examine the sensitivity/specificity of PET/CT with different micro-organisms. Head-to-head comparison between 99mTc–hexamethylpropyleneamine oxime–leukocyte single-photon emission CT/CT and 18F-FDG PET/CT could be useful. There was no bias in patient recruitment because all patients were recruited consecutively. All patients were followed up until the end of the study. In conclusion, we do not believe that PET/CT is a “magic” test, but it should now be considered as an additional tool for decision making.
In response to the second letter by Dr. Sharma, our study was not intended to evaluate the impact of hyperglycemia on PET/CT. Patients with blood glucose >8 mmol/l (>144 mg/dl) received intravenous rapid-acting insulin 1 h prior to FDG injection. All patients underwent PET/CT within 48 h of initiation of antibiotic therapy. However, we agree that prior antibiotic therapy could have an impact on test results. False-positive results can occur with Dacron pouches, inflammation, or hematoma, but FDG uptake is usually higher with infection. We agree that further studies with longer follow-up are required. We do not believe that PET/CT should be used in all cases, but it appears useful in patients with cardiac device and fever of unknown origin, as well as in patients for which knowledge of the extent of infection could change the decision to perform device and lead extraction.
- American College of Cardiology Foundation