Severe sepsis, a consequence of the body’s response to infection, is a major cause of death in hospitals. The earlier that doctors recognize that a patient has sepsis, the earlier the patient can be treated with antibiotics, fluids and other measures, and the better the chance of survival.

That’s why critical care and emergency medicine researchers have been looking for ways to spot whether someone coming to the hospital might have sepsis, even before arrival.

At Emory, Carmen Polito, Jonathan Sevransky and colleagues recently published a paper in the American Journal of Emergency Medicine on an emergency medical services screening tool for severe sepsis. Polito and Sevransky are in the division of pulmonary, allergy, critical care and sleep medicine in the Department of Medicine. The tool was evaluated based on Grady emergency medical services data from 2011 and 2012.

“Sepsis is largely a face without a name in the EMS setting,” Polito says. “The goal of our study was to create a tool to assist EMS providers in naming this deadly condition at the point of first medical contact. Similar to other life-threatening, time-sensitive conditions like stroke and heart attack, naming sepsis is the first step in developing coordinated care pathways that focus on delivering rapid, life-saving treatment once the patient arrives at the hospital.”

Diagnosing sepsis in a medical clinic or hospital setting has traditionally involved obtaining cultures from blood or other body fluids and takes 24 to 48 hours. More recently, the focus has moved to lactate, white blood cell levels and possibly other biomarkers, which may be difficult to measure in a moving ambulance.

Polito, Sevransky and colleagues developed a model that works in two steps. Patients were defined as “at risk” based on their heart rate, breathing rate and blood pressure. Once someone was defined as “at risk,” a risk score (PRESS, for prehospital recognition of severe sepsis) was calculated using six criteria: age, transport from nursing home, EMS flag for “sick person,” elevated temperature by touch, low systolic blood pressure, and low oxygen saturation. Note that all of these can be measured quickly and non-invasively.

The PRESS score demonstrated a sensitivity of 86 percent and a specificity of 47 percent. This means that by itself, the PRESS score is good at detecting the presence of severe sepsis but may give a “false positive” about half the time.

However, the PRESS score could be combined with lab tests to further increase specificity, with the trade-off of needing more time to obtain the information.

The authors conclude by noting that the PRESS score would need to be validated at other hospitals, and that the “at-risk” screen and exclusion criteria are both necessary; extrapolating to all EMS patients would yield lower sensitivity and specificity.