User login
SAN DIEGO – Although hospitals in recent years have done a good job catching severe sepsis, less severe cases are falling through the cracks and ultimately proving fatal, according to a retrospective database study presented at an international conference of the American Thoracic Society.
That’s probably the main reason investigators at Kaiser Permanente Northern California (KPNC) found that up to half of hospital deaths are sepsis related. Of 14,206 adult inpatient deaths at KPNC hospitals between 2010 and 2012, 36.9% had sepsis-related codes. When the team included patients without sepsis codes but with evidence of both infection and acute organ failure – implying sepsis – the number rose to 55.9%.
Most patients had indications of sepsis at admission, and patients with initially less severe sepsis made up the majority of sepsis deaths. About 56% of sepsis-related deaths were in people with normal blood pressure and normal or intermediate serum lactate levels (less than 4 mmol/L) on admission. Patients who met the criteria for early goal-directed therapy at admission, but for whatever reason did not get it, accounted for 21% of the deaths (JAMA 18 May 2014 [doi:10.1001/jama.2014.5804]).
The story was similar when the team looked at 143,312 deaths in the 2010 Healthcare Cost and Utilization Project Nationwide Inpatient Sample, which captures about 20% of U.S. community hospitals: 34.7% had sepsis codes, and 52% had codes or evidence of both infection and acute organ failure.
Sepsis could have been the final common pathway in already-ill patients, but the numbers hint that at least in some cases, sepsis that could have been extinguished early got out of hand before it was recognized.
The Surviving Sepsis Campaign and other efforts "have had a huge impact on how we treat the most severely ill sepsis patients. We’ve seen about a halving of mortality in the past 15 years. Now we need to broaden our perspective to focus intervention on the less severely ill, who tend to be less severe up front and underidentified," said lead investigator Dr. Liu, with the KPNC division of research, Oakland, Calif.
Based on the results, KPNC has started applying its sepsis bundle to patients with intermediate-lactate levels, "but there is very limited data about the benefit of bundle care in less severe sepsis patients, so we are still tracking our outcomes," he said. There’s also a culture shift involved, which includes heightening clinician awareness, updating communication protocols, and other measures, Dr. Liu said in presenting the results.
There’s a role for more research dollars as well, and education efforts to make the public aware of sepsis and the need for early intervention, similar to what’s been done for stroke, he said.
 |
|
Dr. Steven Q. Simpson, FCCP, comments: "These national data corroborate data from nearly every individual hospital that has looked at the phenomenon in the Kansas Sepsis Project and other quality improvement efforts with which I work personally.
It is clear that early identification of severe sepsis and aggressive intervention remain suboptimal in U.S. hospitals, in spite of a decade of attempts by numerous organizations, including CHEST, the Surviving Sepsis Campaign, the New York Hospital Association, and others to make providers aware of the need. One hopes that the press engendered by these findings helps to focus more attention in that direction and helps to prevent unnecessary deaths."
|
|
Dr. Steven Q. Simpson, FCCP, comments: "These national data corroborate data from nearly every individual hospital that has looked at the phenomenon in the Kansas Sepsis Project and other quality improvement efforts with which I work personally.
It is clear that early identification of severe sepsis and aggressive intervention remain suboptimal in U.S. hospitals, in spite of a decade of attempts by numerous organizations, including CHEST, the Surviving Sepsis Campaign, the New York Hospital Association, and others to make providers aware of the need. One hopes that the press engendered by these findings helps to focus more attention in that direction and helps to prevent unnecessary deaths."
|
|
Dr. Steven Q. Simpson, FCCP, comments: "These national data corroborate data from nearly every individual hospital that has looked at the phenomenon in the Kansas Sepsis Project and other quality improvement efforts with which I work personally.
It is clear that early identification of severe sepsis and aggressive intervention remain suboptimal in U.S. hospitals, in spite of a decade of attempts by numerous organizations, including CHEST, the Surviving Sepsis Campaign, the New York Hospital Association, and others to make providers aware of the need. One hopes that the press engendered by these findings helps to focus more attention in that direction and helps to prevent unnecessary deaths."
SAN DIEGO – Although hospitals in recent years have done a good job catching severe sepsis, less severe cases are falling through the cracks and ultimately proving fatal, according to a retrospective database study presented at an international conference of the American Thoracic Society.
That’s probably the main reason investigators at Kaiser Permanente Northern California (KPNC) found that up to half of hospital deaths are sepsis related. Of 14,206 adult inpatient deaths at KPNC hospitals between 2010 and 2012, 36.9% had sepsis-related codes. When the team included patients without sepsis codes but with evidence of both infection and acute organ failure – implying sepsis – the number rose to 55.9%.
Most patients had indications of sepsis at admission, and patients with initially less severe sepsis made up the majority of sepsis deaths. About 56% of sepsis-related deaths were in people with normal blood pressure and normal or intermediate serum lactate levels (less than 4 mmol/L) on admission. Patients who met the criteria for early goal-directed therapy at admission, but for whatever reason did not get it, accounted for 21% of the deaths (JAMA 18 May 2014 [doi:10.1001/jama.2014.5804]).
The story was similar when the team looked at 143,312 deaths in the 2010 Healthcare Cost and Utilization Project Nationwide Inpatient Sample, which captures about 20% of U.S. community hospitals: 34.7% had sepsis codes, and 52% had codes or evidence of both infection and acute organ failure.
Sepsis could have been the final common pathway in already-ill patients, but the numbers hint that at least in some cases, sepsis that could have been extinguished early got out of hand before it was recognized.
The Surviving Sepsis Campaign and other efforts "have had a huge impact on how we treat the most severely ill sepsis patients. We’ve seen about a halving of mortality in the past 15 years. Now we need to broaden our perspective to focus intervention on the less severely ill, who tend to be less severe up front and underidentified," said lead investigator Dr. Liu, with the KPNC division of research, Oakland, Calif.
Based on the results, KPNC has started applying its sepsis bundle to patients with intermediate-lactate levels, "but there is very limited data about the benefit of bundle care in less severe sepsis patients, so we are still tracking our outcomes," he said. There’s also a culture shift involved, which includes heightening clinician awareness, updating communication protocols, and other measures, Dr. Liu said in presenting the results.
There’s a role for more research dollars as well, and education efforts to make the public aware of sepsis and the need for early intervention, similar to what’s been done for stroke, he said.
SAN DIEGO – Although hospitals in recent years have done a good job catching severe sepsis, less severe cases are falling through the cracks and ultimately proving fatal, according to a retrospective database study presented at an international conference of the American Thoracic Society.
That’s probably the main reason investigators at Kaiser Permanente Northern California (KPNC) found that up to half of hospital deaths are sepsis related. Of 14,206 adult inpatient deaths at KPNC hospitals between 2010 and 2012, 36.9% had sepsis-related codes. When the team included patients without sepsis codes but with evidence of both infection and acute organ failure – implying sepsis – the number rose to 55.9%.
Most patients had indications of sepsis at admission, and patients with initially less severe sepsis made up the majority of sepsis deaths. About 56% of sepsis-related deaths were in people with normal blood pressure and normal or intermediate serum lactate levels (less than 4 mmol/L) on admission. Patients who met the criteria for early goal-directed therapy at admission, but for whatever reason did not get it, accounted for 21% of the deaths (JAMA 18 May 2014 [doi:10.1001/jama.2014.5804]).
The story was similar when the team looked at 143,312 deaths in the 2010 Healthcare Cost and Utilization Project Nationwide Inpatient Sample, which captures about 20% of U.S. community hospitals: 34.7% had sepsis codes, and 52% had codes or evidence of both infection and acute organ failure.
Sepsis could have been the final common pathway in already-ill patients, but the numbers hint that at least in some cases, sepsis that could have been extinguished early got out of hand before it was recognized.
The Surviving Sepsis Campaign and other efforts "have had a huge impact on how we treat the most severely ill sepsis patients. We’ve seen about a halving of mortality in the past 15 years. Now we need to broaden our perspective to focus intervention on the less severely ill, who tend to be less severe up front and underidentified," said lead investigator Dr. Liu, with the KPNC division of research, Oakland, Calif.
Based on the results, KPNC has started applying its sepsis bundle to patients with intermediate-lactate levels, "but there is very limited data about the benefit of bundle care in less severe sepsis patients, so we are still tracking our outcomes," he said. There’s also a culture shift involved, which includes heightening clinician awareness, updating communication protocols, and other measures, Dr. Liu said in presenting the results.
There’s a role for more research dollars as well, and education efforts to make the public aware of sepsis and the need for early intervention, similar to what’s been done for stroke, he said.
Key clinical point: Have a high index of suspicion for less severe sepsis at hospital admission.
Major finding: About 56% of sepsis-related deaths are in people with normal blood pressure and normal or intermediate serum lactate levels (less than 4 mmol/L) on admission.
Data source: A retrospective database study of about 7 million adult hospitalizations.
Disclosures: The work was funded by the Kaiser Foundation, the Department of Veterans Affairs, and others. One author disclosed personal fees from industry sources. Dr. Liu and the other authors said they had no financial disclosures.
