Infective endocarditis (IE) is a life-threatening disease with high morbidity and mortality, and its treatment involves prolonged antibiotic therapy or even surgical intervention. A correct and timely diagnosis is crucial, and positron emission tomography/computed tomography (PET/CT) with 18F-fluorodeoxyglucose (FDG) has come to play an increasingly accepted role in this setting in recent years, most notably in the context of cardiac implants such as prosthetic heart valves, pacemakers, and defibrillators. Although FDG PET/CT has proven itself an accurate technique in prosthetic heart valve endocarditis especially,1,2 distinguishing between non-infectious inflammation and infection requires expertise.3 Conversely, vegetations without paravalvular complications may be false negative and require other techniques such as echocardiography or CT angiography to assess.4

As FDG PET/CT allows for imaging of the entire body in a single session, secondary findings related to and supporting the diagnosis of IE may also be encountered. For example, metastatic infectious foci may further support the diagnosis, even in the absence of clear foci in the heart.5

In their paper, Boursier et al. describe another secondary finding in the form of hypermetabolism of the spleen and/or bone marrow (HSBM).6 Both play a role in the immune response,7,8 so it is perhaps no surprise that they found a link between the presence of IE and an activation of these tissues. In their cohort of 129 patients with a definite diagnosis of IE in 88, HSBM was a predictor of definite IE. It was especially powerful when combined with cardiac foci, associated with a definite diagnosis of IE in 29 out of 30 patients and probable IE in the final patient.

However, as Boursier et al. point out, these findings are at least partially influenced by their specific patient group with an exceptionally high prevalence of IE, and their results must therefore be read in this particular context. Even in the group with no cardiac, bone marrow, or spleen abnormalities, more than a third of the patients had IE.

In a less selected population, the number of reasons for HSBM is more varied and it is vital for reading physicians to be aware of patient- or treatment-related factors that may influence these findings even outside the setting of infection. Figure 1 shows four examples of scans with HSBM that were not related to infection, ranging from anemia to chemotherapy. Anemia especially is a common enough finding that must be considered when bone marrow is deemed to be more active than normal. Additionally, as the presenting symptoms of IE are non-specific including fever and malaise, other diseases that show increased bone marrow uptake such as leukemia 9 may certainly present in a similar fashion.

Figure 1
figure 1

Four patients with non-infectious bone marrow (A–D) and spleen (B–D) hypermetabolism on FDG PET/CT. A patient scanned for suspicion of lung malignancy. No malignancy was found, and bone marrow hypermetabolism was explained by anemia. B patient presenting with fever. Cause for the bone marrow and spleen activation was determined to be an auto-immune hemolytic anemia. C patient treated with granulocyte colony-stimulating factor. D patient treated with chemotherapy for B cell lymphoma. Full remission of the lymphoma with chemotherapy-induced hypermetabolism of the bone marrow and spleen. No evidence of endocarditis in any of these patients

Full size image

It remains to be seen if HSBM would remain a significant independent predictor of IE in a more varied setting with different pathologies more commonly encountered in daily routine. Nevertheless, the paper by Boursier et al. shows that in patients with HSBM and cardiac foci, the diagnosis of IE is extremely likely. In the absence of cardiac foci, the nature of HSBM is perhaps, as most things in life, often a matter of context.