Sepsis a Factor in Many Hospital Deaths, Investigators Say

Norra MacReady

May 21, 2014

Sepsis figures in 1 of every 2 to 3 hospital deaths, according to findings from a study of more than 7 million adult hospitalizations across the United States.

More than half of the deaths from sepsis occurred in people who initially appeared less seriously ill, Vincent Liu, MD, and colleagues report in a research letter published online May 18 issue in JAMA and presented at the American Thoracic Society 2014 International Conference.

"Given the prominent role it plays in hospital mortality, improved treatment of sepsis (potentially a final hospital pathway for multiple other underlying conditions) could offer meaningful improvements in population mortality," the authors write.

Dr. Liu, from the Kaiser Permanente Division of Research, Oakland, California, and colleagues studied 482,828 adult hospitalizations between 2010 and 2012 at 21 Kaiser Permanente Northern California (KPNC) facilities. They also studied 6,555,621 hospitalizations reported in 2010 as part of the Healthcare Cost and Utilization Project Nationwide Inpatient Sample (NIS), a nationally representative sample of 1051 hospitals.

They identified patients with explicit and implicit sepsis. Explicit sepsis included cases diagnosed with septicemia, sepsis, severe sepsis, or septic shock, according to International Statistical Classification of Diseases, Ninth Revision, Clinical Modification, codes. Patients with evidence of infection and organ failure were assumed to have implicit sepsis.

In the KPNC cohort, explicit sepsis was diagnosed in 55,008 KPNC cases (11.4%; 95% CI, 11.3% - 11.5%), including 50,520 cases (10.5%; 95% CI, 10.54% - 10.5%) that were present on admission. Another 80,678 KPNC patients were diagnosed with implicit sepsis (16.7%; 95% CI, 16.6% - 16.8%); of those, 73,933 were present on admission (15.3%; 95% CI, 15.2% - 15.4%).

In the NIS cohort, explicit sepsis was diagnosed in 280,663 patients (4.3%; 95% CI, 4.3% - 4.3%), of whom 49,664 died (17.7%; 95% CI, 17.6% - 17.8%). Implicit sepsis was diagnosed in 717,718 (10.9%; 95% CI, 10.9%-11.0%) of the NIS patients, of whom 74,451 died (10.4%; 95% CI, 10.3%-10.4%).

There were 14,206 deaths in the KPNC cohort and 143,312 deaths in the NIS cohort. In the KPNC cohort, explicit sepsis contributed to 44.2% (95% CI, 43.3% - 45%) of the hospital deaths and implicit sepsis contributed to 55.9% (95% confidence interval [CI], 55.1% - 56.7%). In the NIS cohort, explicit sepsis contributed to 34.7% (95% CI, 34.4% - 34.9%) of deaths and implicit sepsis contributed to 52.0% (95% CI, 51.7% - 52.2%) of hospital deaths.

“[W]e found that sepsis contributed to 1 in every 2 to 3 deaths, and most of these patients had sepsis at admission,” the authors write.

In the KPNC cohort, there were 6272 deaths among the patients with explicit sepsis (11.4%; 95% CI, 11.1% - 11.7%), including 5238 deaths among patients with sepsis on admission (10.4%; 95% CI, 10.1% - 10.6%). Among patients with implicit sepsis, there were 7,941 deaths (9.8%; 95% CI, 9.6% - 10.0%), including 7,391 deaths among patients with implicit sepsis at admission (10.0%; 9.8% - 10.2%).

In a subset of the KPNC cohort, those patients treated in 2012, 2536 patients with sepsis on admission met the criteria for early goal-directed therapy (EGDT), but only 1200 actually received it. EGDT-eligible patients accounted for 32.6% (95% CI, 30.4% - 34.7%) of sepsis-related deaths overall, but mortality among those who received EGDT was 11.7% (95% CI, 10.2% - 13.1%) compared with 20.9% (95% CI, 19.0% - 22.7%) among the patients who did not receive EGDT. Another 55.9% (95% CI, 53.6% - 58.1%) of those deaths occurred in septic patients who had normal blood pressure and with measured lactate levels lower than 4 mmol/L.

"Patients with initially less severe sepsis made up the majority of sepsis deaths.... [I]mproving standardized care for patients with less severe sepsis could drive future reductions in hospital mortality," the authors write.

These findings are unsurprising to several hospital medicine experts. "We have had good, high-quality data for over a decade demonstrating that severe sepsis and septic shock are incredibly common, highly mortal, and grossly underrecognized," Andrew Odden, MD, assistant professor, University of Michigan Medical School and the Veterans Administration, Ann Arbor Healthcare System, told Medscape Medical News.

"Clinicians need to always keep sepsis in their differential diagnosis and to have patients seek medical help early even if they seem well," added Lisa Shieh, MD, PhD, clinical associate professor and hospitalist, and director of quality for the Department of Medicine at Stanford University School of Medicine, Palo Alto, California, told Medscape Medical News.

Commenting on the finding that most of the deaths occurred in patients who seemed less severely ill, Dr. Shieh said, "these healthier patients sometimes get triaged to medical-surgical units, where their sepsis may progress unrecognized to severe sepsis or septic shock. This is one of the reasons why the Society of Hospital Medicine is collaborating with the Society of Critical Care Medicine on the Surviving Sepsis Campaign: to emphasize and improve screening and early treatment on the medical-surgical units."

"These findings highlight the need for vigilance in teaching the principles of sepsis evaluation and management," Aroop Pal, MD, associate professor and hospitalist at the University of Kansas, Kansas City, told Medscape Medical News. "Sepsis is a reflection of severe infection and is [an indication] that trouble is present."

One study author has reported receiving nonfinancial support from Edwards Inc; personal fees from Pfizer, Med-Immune LLC, Eli Lilly, Ferring Pharmaceuticals, and Roche Diagnostics International Ltd; and a grant from Eisai Inc. The other study authors, Dr. Odden, Dr. Shieh, and Dr. Aroop have disclosed no relevant financial relationships. Dr. Odden, Dr. Shieh, and Dr. Aroop are all members of the Society of Hospital Medicine and the Surviving Sepsis Campaign.

JAMA. Published online May 18, 2014. Full text


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.