An Older Female with Fevers, Myalgias, and Fatigue
A 67-year-old female was admitted for a several day history of fevers, myalgias, and fatigue. She also noted poor appetite and diminished oral intake during this time. She denied cough or production of sputum as well as dysuria or other urinary symptoms.
Significant past medical history reveals long-term hypertension with subsequent end-stage renal failure on hemodialysis for the past year. Her medicines include lisinopril, metroprolol, and amlodipine, in addition to a standard renal failure/hemodialysis regimen, including erythropoietin to maintain hemoglobin levels between 10 g/dL to 11 g/dL.
She denies alcohol use or drug abuse of any kind. Systems review was negative for paroxysmal nocturnal dyspnea, exertional dyspnea, or pedal edema.
Upon admission, physical exam revealed temperature of 101°F, pulse of 108 beats per minute, and blood pressure 108/84 mm Hg. Head, eyes, ears, nose, and throat exams was non-remarkable.
Chest showed clear lungs. There was some redness at the right subclavian hemodialysis access site. The cardiac exam revealed a tachycardia with a grade II/VI systolic murmur at the apex, otherwise more difficult to characterize due to the tachycardia. Abdominal, extremity, and neurologic exam were negative.
Laboratory exam showed creatinine levels of 10 mg/dL, blood urea nitrogen levels of 41 mg/dL, potassium 5.1 mEq/L, sodium 134 mEq/L, and no anion gap. Hemoglobin was 8.7 g/dL and white blood cell count was 11.9 K/mm3. Chest x-ray had no infiltrates.
The patient was admitted with concern regarding an infected access site. Blood cultures were obtained and renally-adjusted dose vancomycin was started. Her temperature initially declined to 99°F and she subjectively improved. However, on the third hospital day, positive blood cultures from both the first 2 days of admission were reported as gram positive cocci in chains.
Vancomycin was continued and the line removed. An echocardiogram was ordered. That night she reported chest tightness and significant shortness of breath with difficulty lying down. Temperature was 99°F but blood pressure was 90/70 mm Hg.
Yet another previously drawn blood culture was positive and all were now being identified as Enterococcus faecalis. An urgent echocardiogram revealed left ventricular ejection fraction of 50%, with vegetations on the aortic and mitral valves. The large vegetation of the anterior leaflet of the mitral valve resulted in severe mitral regurgitation directed posteriorly. There was also moderate aortic regurgitation. Gentamycin was added to her antibiotic regimen.
Which of the following is a correct statement about the presented case and management?
A. Her age of 67 years is uncommon for her illness.
B. Her blood culture findings have become atypical for her condition.
C. Cerebral complications are a common manifestation of disease.
D. Transesophageal echocardiography is preferred and should be the initial imaging study.
(Answer and discussion on next page)
Correct Answer: C
This patient is suffering from bacterial endocarditis. In fact, the findings presented in the case fulfill the modified Duke criteria for diagnosing endocarditis.1 These guidelines continue to be used in practice and establish 2 types of major criteria:
• Microbiological involvement includes typical organisms, such as Enterococcus faecalis, from 2 separate blood cultures without a primary focus or typical microorganisms isolated from 2 persistently positive blood cultures (our patient’s situation) or single blood cultures of Coxiella burnetti.
• Endocardial involvement criteria include new valvular regurgitation or a positive echocardiogram for vegetations. The presence of the 2 major criteria as seen here is diagnostic.
Minor criteria include epidemiologic predisposition to endocarditis—including preexisting valvular disease and injection drug use, vascular phenomena (eg, Janeway’s hemorrhagic lesions), and immunologic phenomena (eg, Osler nodes and glomeronephritis).
Note: One major criterion plus at least 3 minor criteria also define endocarditis. Our patient’s source of infection was the infected vascular access catheter, which resulted in bacteremia that subsequently infected her valves. A “typical microorganism,” E. faecalis, was repeatedly cultured from her blood and echocardiogram demonstrated vegetations of her aortic and mitral valves.
As is typical for medicine in general, population demographics and advances in technologies related to diagnosis, medical, and surgical therapies can and do change the clinical characteristics and findings in infective endocarditis. For example, the more frequent use of prosthetic valves and a variety of other intracardiac devices have resulted in the traditional risk factor of rheumatic disease now accounting for only 10% of endocarditis cases; the previously mentioned iatrogenic hardware accounts for about half of cases.2
Meanwhile, the aging and morbidity of the population has resulted in hemodialysis (with its attendant vascular devices), HIV, and other chronic medical conditions being more commonly involved risk factors; as many as a third of cases are now being categorized as “healthcare associated.” Similarly, there has been an increase in the age of endocarditis cases, such that there is an increased incidence in patients >65 years.2 Thus, Answer A is not correct—in fact, endocarditis is becoming more common in people over 65 years.
As alluded to above, Duke’s criteria for diagnosis, with the microbiology findings listed, still has good sensitivity and specificity of greater than 80%.1 Conversely, when 3 sets of cultures are properly timed and obtained, a pathogen will be identified in about 90% of cases.2 So blood cultures remain the core diagnostic maneuver in the diagnosis of endocarditis and therefore, Answer B as stated is not correct.
Answer D relates to which image is optimal to detect murmurs and vegetations; valvular lesions are the second major Duke criteria for diagnosis. Currently transthoracic echocardiography (TTE) and not transesophageal echocardiography (TEE) is the initially preferred study in an endocarditis evaluation; therefore, Answer D is not correct. TTE can detect a vast majority of vegetations and is superior in assessing valvular dysfunction and hemodynamics.2 Although TEE is more sensitive and specific, it is far more invasive and should be reserved for patients with prosthetic valves and intracardiac devices as an initial study and in secondary situations of poor image quality or nondefinitive findings in high clinical suspicion cases. Further, clinicians should not forget that repeating studies, TTE or TEE, is a good idea when clinical findings are changing. But for now, except in patients with valve prostheses, the initial image should be TTE and not TEE.
Answer C is correct. Cerebral complications are now the most frequent and severe extracardial complications, replacing the old historic listing of splinter hemorrhages, Janeway’s fingertip lesions, and Roth spots that filled textbooks in the past, which are now seen quite uncommonly in about 5% of cases. Meanwhile, cerebral effects from vegetation emboli is seen in up to 20% of cases.3 Further, such cerebral complications have been shown to be strongly associated with increasing S. aureus microbiology and large vegetation sizes (1 cm to 1.5 cm).
In large vegetation, cerebral embolism risk is so common that such a finding has been proposed to be an independent indication for urgent, if not emergent, early valve replacement in endocarditis.3 This represents a significant change in endocarditis management.
Ronald Rubin, MD, is a professor of medicine at Temple University School of Medicine and chief of clinical hematology in the department of medicine at Temple University Hospital, both in Philadelphia.
1.Li JS, Sexton DJ, Mick N, et al. Proposed modifications to the Duke criteria for the diagnosis of infective endocarditis. Clin Infect Dis. 2000;30(4):633-638.
2.Hoen B, Duval X. Clinical practice. Infective endocarditis. N Engl J Med. 2013;
3.Kang DH, Jim YJ, Kim SH, et al. Early surgery versus conventional treatment for infective endocarditis. N Engl J Med. 2012;366:2466-2473.