HOLLI W. HAYNIE

Sepsis kills more people every year than heart attack or stroke, yet it is a commonly misunderstood and under-recognized condition. The reasons for that are numerous and complex, but the most basic reason is there still is not an industry-wide accepted definition of sepsis or the standard of care for it. There is no simple blood test that identifies sepsis and physicians often have a challenging time recognizing it since the initial symptoms – fever, rapid pulse and respiratory difficulty – can be attributed to many conditions, leading to a late or missed diagnosis. Moreover, with this lack of awareness about sepsis, many stereotypes persist.

"It can happen to anyone, even a young, healthy person," said Paula Jacobs, director of quality performance and improvement at Methodist Hospital North. "Physicians have had this stereotype of an 80-year-old in poor health and frail as the typical sepsis patient and that's not the case."

In fact, young black men are the highest risk group for developing sepsis. Currently, sepsis is the leading cause of hospital death, and is on the rise by 1.5 percent a year. According to the Centers for Disease Control and Prevention (CDC), 29 percent of people who develop sepsis die; that's 215,000 per year. Sepsis has a higher mortality rate than heart attack (19 percent) and stroke (22.5 percent). Much of that mortality rate is due to the fact that by the time sepsis is recognized in a patient, they are already in severe sepsis or septic shock, when very little can be done to stop the cascade.

At Methodist Hospital North, a sepsis alert tool has been used with their EMR system for more than 18 months. This alert system is tied into the hospital's OneChart program that keeps detailed history and lab information on all patients in the hospital. Monitoring patients in all levels, from ED to ICU to med-surg floors, this alert system looks at the entire clinical picture – a combination of vital signs, lab values and organ function – to indicate if the sepsis cascade is starting.

While there is still much cooperation needed in the clinical community to identify sepsis sooner, there are general definitions accepted by the Society of Critical Care Medicine. The sepsis cascade begins with SIRS (Systemic Inflammatory Response Syndrome), an early warning sign introduced just a few years ago. This can be indicated by a combination of two factors out of four: elevated temperature, heart rate, respiratory rate or white blood cell count. The next stage in the cascade is sepsis, which includes SIRS and evidence of infection. After that stage comes severe sepsis, which is sepsis plus organ dysfunction. Organ dysfunction can be indicated by lactic acid, which is one of the first things to show up in organ dysfunction, Jacobs explained. However the system compares lab values with patient history and only uses organ dysfunction as a red flag if it is new to the patient. The final stage is septic shock – severe sepsis plus severe low blood pressure. The mortality rate is significantly higher in septic shock, at 49 percent.

"A lot of times patients can have (the signs of SIRS) and not be considered sick yet," said Karen Hopper, MD, clinical director of Methodist Hospital North. She said the majority of patients come into the hospital already in severe sepsis, when it's harder to halt the cascade. "If things aren't caught at that point and they continue, then organs fail."

Added Jacobs, "You can be very sick with sepsis and may not look that sick."

Even with nurses checking vitals and labs every two hours, it takes a lot of critical thinking, explained Jacobs, to put together the pieces within a certain timeframe to provide early intervention.

The program, Jacobs explained, "actually looks for those triggers in the vital signs and then will go back and scan that record and look for lab values."

If sepsis is indicated, whoever is caring for that patient is immediately paged with a sepsis alert and given the reason why.

When the first alerts go off for sepsis the protocol is to stop the infection, Hopper said, which early on the location of the infection cannot be identified and patients receive a bombardment of antibiotics to cover all possible types of infection. After lab cultures come back and indicate the bacterial culprit, the barrage of antibiotics stops except for the specified medication for the source of infection.

Overall the alert system has worked well, with 100 percent of alerts firing per met criteria. Since the alert system began in late 2006, the mortality rate for sepsis at Methodist North has been reduced by 14 percent, however Jacobs is quick to note that not all sepsis cases are being consistently documented and that percentage is likely much higher. Jacobs said the more compelling statistic is their actual-to-expected mortality rate. The baseline in 2006 was .88 when the alert system began and by year's end 2007, the ratio dropped to .68 and is holding. Another positive is year-to-date, their mortality rate for septic shock dropped 13 percent to 40.9 (down from 47 percent), which is below the national average of 45 to 50 percent.

"Looking at the records now and how quickly it's addressed and how it's gotten better (shows) we've made so much progress in the last year and a half," said Jacobs, adding she hopes to see this sepsis alert system become a best practice in all Methodist hospitals.

Jacobs admits there has been resistance to the alert, as sepsis myths and stereotypes are hard to dispel, but the paradigm is shifting. To address the sepsis documentation issue, the hospital has been designing a data collection tool over the past two years to identify a fully accurate sepsis mortality rate. Soon they will embark on a quality project with the University of Tennessee to study chart reviews.

"We're still in the early stages of defining what sepsis is and making sure all doctors agree on the definition and agree on early treatments that will work," concluded Hopper. "It needs to get out to everybody."