Etiology of Sepsis in Patients Admitted from the ED

Abstract and 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, is Associate Editor for Infectious Disease Alert.

Dr. Winslow serves as a consultant for Siemens Diagnostics and is on the speaker's bureau for GSK and Cubist.

Synopsis: In a prospective, observational study, > 50% of patients identified and treated for severe sepsis in the emergency department (ED) had negative cultures; 18% of patients had a noninfectious diagnosis that mimicked sepsis.

Source: Heffner AC, et al. Etiology of illness in patients with severe sepsis admitted to the hospital from the emergency department. Clin Infect Dis. 2010;50:814-820.

A prospective, observational study of patients 18 years of age and older treated with goal-directed therapy of sepsis in the ED was conducted at a large county hospital in North Carolina. Inclusion criteria included two or more criteria for systemic inflammation and evidence of hypoperfusion. Clinical data were prospectively collected for two years. Blinded observers used standardized criteria to determine the final cause of hospitalization.

A total of 211 patients were enrolled in the study. Of those, 95 (45%) had positive culture results and 116 (55%) had negative cultures. Overall mortality was 19%. Patients with positive cultures were more likely to have active malignancy, have a vascular line, be a resident of a nursing home, have UTI as a primary source, and were less likely to have a pulmonary source.

Of the patients negative by culture, 51 (44%) had clinical evidence of infections, with pneumonia being the most common in 38 patients. Nine patients had atypical infections, including C. difficile disease (5), cryptococcosis (2), TB (1), and viral encephalitis (1). Thirty-seven patients (32%) had noninfectious mimics. The most common diagnoses included inflammatory colitis, hypovolemia, medication effect, adrenal insufficiency, acute MI, pulmonary edema, pancreatitis, diabetic ketoacidosis, and small bowel obstruction. In 19 cases (16%), the cause of the sepsis picture on presentation was indeterminate.


While this was a relatively small single-center study, I felt it was an interesting and important paper. I frequently lead the morning report with the medicine house staff at our own county hospital in San Jose, CA, and enjoy being challenged by the diagnostic possibilities present in patients who present to our ED acutely ill and require admission to the hospital. The finding of noninfectious causes of a surprisingly large number of cases of patients admitted to the hospital for "sepsis" is an important reminder to keep a broad differential diagnosis in mind in caring for such critically ill patients. This further reinforces the importance of infectious diseases clinicians remaining skilled as internists (or pediatricians) as they approach the management of these complicated patients.

As a side note, the second author on this paper, Dr. Jim Horton, is Chief of the ID Division at Carolinas Medical Center. Jim and I trained together in New Orleans many years ago. He is a great clinician and one of the finest individuals I've ever known. He invited me to give medicine grand rounds at his hospital in 2008, where I had an opportunity to meet his colleagues and tour the wards. It was apparent that CMC is a wonderful hospital that provides outstanding care to the people of Charlotte.