Non-HACEK Gram-Negative Bacillus Endocarditis

Abstract & Commentary

By Dean L. Winslow, MD, FACP, FIDSA, Chief, Division of AIDS Medicine, Santa Clara Valley Medical Center; Clinical Professor, Stanford University School of Medicine, Section Editor, HIV, is Associate Editor for Infectious Disease Alert.

Synopsis: Of 2761 patients with endocarditis enrolled in a prospective, multinational cohort study, 49 patients with non-HACEK gram-negative endocarditis were identified. E. coli and Pseudomonas aeruginosa were the most common pathogens, and 57% were considered to be health care associated. Fifty-nine percent of cases were associated with prosthetic valves. The mortality rate was high (24%) despite cardiac surgery being performed in 51% of cases.

Source: Morpeth S, et al. Non-HACEK gram-negative bacillus endocarditis. Ann Intern Med. 2007;147:829-835.

Sixty-one hospitals in 28 countries participated in a prospective cohort study of hospitalized patients with definite endocarditis. There were 2761 patients who enrolled between 2000 and 2005. From this cohort, 49 patients with non-HACEK (HACEK stands for: Haemophilus, Actinobacillus, Cardiobacterium, Eikenella, or Kingella) gram-negative endocarditis were identified. Some of the factors differentiating these 49 patients from those with endocarditis due to other organisms included: 59% of the non-HACEK gram-negative cases involved prosthetic valves; 29% had other implanted endovascular devices (pacemakers, implanted cardioverter defibrillators, or aortic stents); 71% had comorbid conditions; 57% were health care associated; only 2 patients (4%) were injection drug users. The following organisms were isolated: E. coli (14), Pseudomonas aeruginos (11), Klebsiella (5), Serratia (4), and other (15). Inpatients with non-HACEK gram-negative bacillus endocarditis, 25% had intracardiac abscess vs 14% of patients with endocarditis due to other organisms; in-hospital mortality was 24% vs 17%, respectively. A large number of different antimicrobial regimens were used but all seemed appropriate. A clear advantage of combination antimicrobial vs monotherapy was not seen.


Dr. Deresinski (Editor of Infectious Disease Alert) and I keep threatening to write an article entitled, "Infectious Disease Truisms Which are not True." One of the truisms we were taught is that non-HACEK gram-negative bacillus endocarditis is largely a disease of injection drug users. This impression was probably due to the publication of several small case series of outbreaks of gram-negative bacillary endocarditis in injection drug users reported from a few large cities during the 1970s and 1980s. This important paper gives a more realistic perspective on this rare disease. Non-HACEK gram-negative bacillus endocarditis should be viewed more properly as predominantly associated with health care-associated acquisition, older age, prosthetic cardiac valves, and other endovascular devices. Paravalvular complications and intracardiac abscess are commonly encountered. Mortality is high despite appropriate medical and surgical therapy.