Epidemiologic considerations for endocarditis are complex as there can be infective and noninfective causes. Once a patient is hospitalized for endocarditis, the mortality rate is estimated to be 10%.1 This underscores the importance of early recognition, workup, and treatment by the primary care provider. However, early diagnosis is complicated by the nonspecific and generalized symptoms of the disease such as fever, chills, weight loss, cough, hemoptysis, and/or malaise. The differential diagnosis for this presentation is sizable.2 In patients with clinical suspicion of endocarditis, prompt workup including 3 sets of blood cultures taken at 30-minute intervals within 48 hours is warranted.2 If the cultures have no growth after 7 days but the clinician still suspects endocarditis, then the search begins for the cause of blood culture-negative endocarditis (BCNE).3
Infective endocarditis affects 15 per 100,000 people in the United States and the number continues to rise.1,4 Blood culture-negative endocarditis represents 2% to 7% of cases of infective endocarditis.5 Patients can present with noninfective BCNE that is related to autoimmune diseases, cancer, and other comorbidities.6 In clinical practice, confirmation of the appropriate diagnosis in cases of BCNE can be quite challenging. Endocarditis, whether culture positive or negative, requires an individualized treatment plan owing to the diversity in epidemiologic causes. Timely diagnosis is critical in the optimal management of this infection.
History and Physical Examination
Evaluation of BCNE is challenging. Patient history is often more important than the physical examination when assessing for BCNE. The key questions to ask during the patient history are listed in Table 1.3
First, it is necessary to investigate if the patient has recently taken antibiotics. Antibiotic use is the primary cause of BCNE.7 Inquiring about exposure to farm animal helps determine if serology for Brucella spp or Coxiella burnetti needs to be obtained.6 To recognize Bartonella hensalae, it is appropriate to question patients about exposure to cats, body lice, and residence in a homeless shelter.6
The Duke criteria is a popular tool to aid in diagnosis of endocarditis. The criteria incorporate aspects of patient history, physical examination, and workup. Physical examination findings that must be evaluated when applying the Duke criteria are presence/absence of Osler nodes, Janeway lesions, splinter hemorrhages, petechiae, new regurgitation murmurs, and neurologic changes.8 A new cardiac murmur is present in 85% of patients with BCNE.8 Although not all patients will present with these findings, their correlation with the disease can expedite workup and management for endocarditis.
Diagnosis of infectious endocarditis is established in patients meeting one of the following Duke criteria:
- 2 major clinical criteria; or
- 1 major and 3 minor clinical criteria; or
- 5 minor clinical criteria
Major and minor criteria were modified by the European Society of Cardiology in 2015 (Table 2).2 A transesophageal echocardiogram is the preferred imaging modality to evaluate for valve insufficiency and/or new vegetation.9
If the Duke criteria have been met to establish a diagnosis of endocarditis but the cultures remain negative for 7 days, more extensive investigation into the pathologic cause is needed. The approach begins by obtaining serology for C burnetti, B hensalae, and Brucella spp.10 If these pathogens are ruled out, then rheumatology and autoimmune workup is warranted to evaluate for a noninfectious cause of endocarditis.3 A study conducted in France reported that 2.5% of 759 cases of BCNE had rheumatic arthritis, Bechet disease, or Libman-Sacks endocarditis.3 The study began by using specific laboratory tests including rheumatoid factor, antinuclear antibody, and anti-DNA antibodies to assist in the diagnosis.3
Histopathology remains the gold standard for pathologic analysis but it is not often feasible. Obtaining samples for a histologic approach is quite invasive and often will only occur if the patient has a planned cardiac surgical procedure.10 Even if tissue samples are obtained, not all tissue is viable for the pathologic stains that would be needed.
In recent years, 16S ribosomal ribonucleic acid polymerase chain reaction (16S rRNA PCR) of excised tissue has played an important role in the diagnostic approach for BCNE.11 If a patient needs surgery for valve repair or vegetation removal, excised tissue should be sent for 16S rRNA PCR. In a prospective study of 819 cases, 16S rRNA PCR identified 109 distinct etiologies making it the second most important diagnostic tool (behind serology) for BCNE.11 If the patient’s history includes prior antibiotic use, PCR may demonstrate the presence of staphylococci or streptococci bacteria (responsible for 45% to 60% of culture-negative endocarditis cases).3
If the workup fails to reveal the underlying pathogen or cause, serology workup on Mycoplasma pneumonia, Legionella spp, and Chlamydia spp should be obtained.10 Serology tests are last-line options because of low prevalence of Chlamydia spp. As of 2017, only approximately 15 cases attributed to Chlamydia spp and 9 cases attributed to Mycoplasma pneumonia have been reported.10 If all diagnostic testing for endocarditis has been exhausted and no pathogen or comorbidity has been identified as the cause of the patient’s illness, it is time to investigate other conditions in the differential diagnosis including atrial myxoma, antiphospholipid syndrome, and acute rheumatic fever.6