Charting a new course in sepsis management

‘Motivated and multidisciplinary’ team critical to improving sepsis care


A drug overdose victim is admitted to a hospital. Providers focus on treating the overdose and blame it for some of the patient’s troubling vital signs, including low blood pressure and increased heart rate. Prior to admission, however, the patient had vomited and aspirated, leading to an infection. In fact, the patient is developing sepsis.

This real-world incident is but one of many ways that sepsis can fool hospitalists and other providers, often with rapidly deteriorating and deadly consequences. A range of quality improvement (QI) projects, however, are demonstrating how earlier identification and treatment may help to set a new course for addressing a condition that has remained stubbornly difficult to manage.

Dr. Devin J. Horton, an academic hospitalist at University Hospital in Salt Lake City

Dr. Devin J. Horton

Every year, more than 1.5 million Americans develop sepsis – arising from the body’s overwhelming and self-destructive response to infection – and roughly 250,000 die from it. According to the Centers for Disease Control and Prevention, about one in three hospital deaths can be at least partially linked to sepsis.

Devin J. Horton, MD, an academic hospitalist at University Hospital in Salt Lake City, sometimes compares sepsis to acute MI to illustrate the difficulty of early detection. A patient complaining of chest pain immediately sets in motion a well-rehearsed chain of events. “But the patient doesn’t look at you and say, ‘You know, I think I’m having SIRS [systemic inflammatory response syndrome] criteria in the setting of infection,’ ” he said. “And yet, the mortality of severe septic shock is at least as bad as acute myocardial infarction.” The trick is generating the same sense of urgency without a clear warning.

The location in a hospital also can present a major obstacle for early identification. Hospitalist Andy Odden, MD, SFHM, patient safety officer in the department of medicine at Washington University in St. Louis, calls hospital wards the “third space” of sepsis care, after the ICU and ED. “A lot of the historical improvement efforts and research has really focused on streamlining care in the ICU and streamlining care in the emergency department,” he said. Often, however, sepsis or septic shock isn’t recognized until a patient is admitted to a medical or surgical ward.

Dr. Andy Odden, patient safety officer in the department of medicine at Washington University in St. Louis

Dr. Andy Odden

Patients on the wards, though, usually begin with a nonsepsis diagnosis, which can produce an anchoring bias. Furthermore, Dr. Odden said, the data needed to identify sepsis may arrive asynchronously, increasingly the difficulty of pulling it all together for a timely diagnosis. As Dr. Horton points out, the trigger for transferring a decompensating sepsis patient from the wards to the ICU is murkier as well. “We don’t know what is too sick for the floor,” he said. “A lot of it is kind of a gestalt.”

Observational studies by the Surviving Sepsis Campaign suggested that patients diagnosed on the floor had mortality rates comparable to and substantially higher than theoretically sicker patients diagnosed in the ICU and ED, respectively.1 “That was kind of a sea change for a lot of people and really articulated what a lot of us on the wards had been feeling,” Dr. Odden said. “We can’t simply apply the lessons that we’ve learned from the emergency department and the ICU to the wards if we’re going to provide the right care for these patients,” he said.

Dueling definitions

Better sepsis care in hospital wards will require a better understanding of shifting management guidelines. Confusing and contradictory definitions haven’t helped. In October 2015, the Centers for Medicare & Medicaid Services instituted its Sepsis Core Measure (SEP-1) for Medicare, requiring every hospital to audit a percentage of patients treated with best-practice 3- and 6-hour bundles for severe sepsis and septic shock. The SEP-1 measure uses the traditional definition of severe sepsis as two or more SIRS criteria, a suspected or proven infection, and organ dysfunction.

A separate set of guidelines issued by the international Sepsis-3 task force in February 2016, by contrast, concluded that the term “severe sepsis” is redundant.2 The update defines sepsis as “life-threatening organ dysfunction caused by a dysregulated host response to infection” and asserts that the condition can be represented by an increase in the SOFA (Sequential Organ Failure Assessment) score of 2 or more points.

For hospital wards, the task force recommended a bedside scoring system called qSOFA (quickSOFA) for adult patients with a suspected infection. The risk stratification tool may help rapidly identify those who are likely to have poorer outcomes typical of sepsis if they meet two of the following three clinical criteria: a “respiratory rate of 22 [breaths]/min or greater, altered mentation, or systolic blood pressure of 100 mm Hg or less.”

CMS doesn’t recognize the Sepsis-3 definition at all and multiple providers have described widespread skepticism and uncertainty over how to reconcile it with the prior definition. Dr. Odden says the dueling definitions have “caused a tremendous amount of confusion” over diagnoses, the necessary sense of urgency, and whether severe sepsis is still a recognized entity. “When people aren’t speaking the same language with the same terminology, there is enormous opportunity for miscommunication to occur,” he said.

Dr. Lisa Shieh, medical director of quality in the department of medicine at Stanford (Calif.) University Medical Center

Dr. Lisa Shieh

Hospitalist Lisa Shieh, MD, PhD, medical director of quality in the department of medicine at Stanford (Calif.) University Medical Center, said Sepsis-3 was never meant to be a screening tool. It can, however, help doctors identify patients at higher risk. Craig A. Umscheid, MD, MSCE, of the department of epidemiology and vice chair for quality and safety in the department of medicine at the University of Pennsylvania, Philadelphia, said many providers agree that, at least theoretically, changes in a patient’s qSOFA score can predict bad outcomes better than SIRS criteria.

Obtaining reliable scores is another matter. The qSOFA blood pressure score generally is measured accurately, he said. On noncritical care units, though, nurses aren’t always trained to consistently and accurately document a patient’s mental status. Likewise, he said, documentation of respiratory rate often is subjective, and an abnormal rate can be easily missed. Changing that dynamic, he stressed, will require coordination with nursing leadership to ensure more consistent and accurate measurements.

Dr. Craig Umscheid, of the department of epidemiology and vice chair for quality and safety in the department of medicine at the University of Pennsylvania, Philadelphia

Dr. Craig Umscheid

Another big issue is that sepsis screening still is based on early recognition, Dr. Shieh said. “The problem with Sepsis-3 is that it is later in the continuum of sepsis.” As such, she recommends sticking with the CMS definition for now. “It catches sepsis earlier, which is the whole strategy for improving sepsis mortality,” she said.


Recommended Reading

VIDEO: Balanced crystalloids protect kidney better than saline
The Hospitalist
Doctors’ and nurses’ predictions of ICU outcomes have variable accuracy
The Hospitalist
ARDS incidence is declining. Is it a preventable syndrome?
The Hospitalist
Improving the risk assessment of patients with acute pulmonary embolism
The Hospitalist
Intense urine output monitoring beneficial in ICU
The Hospitalist
Alarm reductions don’t improve ICU response times
The Hospitalist
Inpatient antiviral treatment reduces ICU admissions among influenza patients
The Hospitalist
When should I transfuse a patient who has anemia?
The Hospitalist
FDA approves angiotensin II for shock patients
The Hospitalist
FVC deterioration signals increasing risk in rib fracture patients
The Hospitalist
   Comments ()