It’s a common scenario in the emergency department (ED). Peak cold and flu season brings in patients complaining of fever and chills, with increased heart rates and elevated white blood cell counts. Too many patients, not enough beds. As emergency medicine providers, we have to make quick, challenging decisions at triage, prioritizing the sickest patients based on their vitals and symptoms and getting them on the right care pathway.

As an ED physician for over 15 years, I remind our staff that our job is not so much to get the diagnosis completely right every time, but more importantly “don’t get it wrong.” This often causes us to start treatments for conditions on suspicion without a definitive diagnosis. The initial ED diagnosis can be similar to getting on the right track: Once it’s rolling, particularly if the patient is admitted to the hospital, it can be very difficult to switch tracks.

Sepsis is one of those conditions that is particularly difficult to readjust. Once a patient is labeled as septic it may be less likely for the diagnosis to be questioned, or for an alternative to be explored. Sepsis also is one of the conditions we often have to consider and treat early, even without a definitive diagnosis.

There is a lot of pressure put on not missing sepsis, and rightfully so. Sepsis is a life-threatening condition caused by a dysregulated immune response to an infection that kills 11 million people worldwide each year. It’s the leading cause of death in U.S. hospitals, costing the U.S. health care industry $62 billion annually, largely because accurate, early detection in the ED is lacking.

If you let a septic patient sit in the waiting room without antibiotics, their condition can worsen rapidly, with their risk of death increasing by up to 8 percent for each hour untreated if in shock. Because of this, ED providers tend to treat suspected sepsis with the earliest suspicion and figure out the rest later. This approach, unfortunately, results in patients sometimes not receiving the most appropriate care. Due to diagnostic anchoring and premature closure, other critical diagnoses may be overlooked throughout the patient’s entire hospital stay because the initial impression was sepsis, and this too could be deadly.

Ideally, we need a way to more accurately—and quickly—diagnose sepsis.

The challenges with diagnosing sepsis

By the Centers for Medicare & Medicaid Services (CMS) definition, sepsis is the presumed presence of infection plus at least two of the four possible systemic inflammatory response syndrome (SIRS) criteria, including elevated heart rate, temperature and blood pressure, and high white blood cell count. The SIRS criteria are one of the oldest screening tools used in the early identification of sepsis.

But, it is important to recognize that these are not sensitive or specific criteria. A patient with multiple SIRS criteria may not be septic, and SIRS misses 1 out of 8 people ultimately deemed septic. There are many conditions that mimic sepsis, presenting with similar symptoms but caused by different underlying issues. These include anaphylaxis, cardiogenic shock, diabetic ketoacidosis, pancreatitis, and pulmonary embolism, and all require very specific and different treatment plans.

To provide the best care for patients, we need to be able to detect sepsis early and initiate timely appropriate treatment. Or we need a signal to change direction and go searching for another diagnosis. Adding to these challenges are the pressures hospitals face to meet the CMS Severe Sepsis and Septic Shock Management Bundle (SEP-1), which is time sensitive and leans heavily into antibiotic administration and fluids if sepsis is suspected. Critics of the bundle believe this could result in overtreatment, increased cost, and even potential harm.

How host response technology is transforming emergency medicine

My colleagues previously wrote on a new approach to diagnosing sepsis in the ED with host response technologies, which measure the patient’s immune dysregulation, rather than variable, often subtle symptoms. This new technology leverages a simple blood test that can be administered quickly in triage in the ED that predicts a patient’s likelihood of having or developing sepsis based on their immune response. This test allows us to more accurately assess the likelihood of sepsis in minutes. In a peer-reviewed study published in Academic Emergency Medicine, the host response technology correctly identified patients without sepsis 98 percent of the time, giving me a signal to start searching for another possible diagnosis that’s causing the clinical presentation.

This host response technology has been successfully integrated into ED triage as part of a nurse-driven protocol at Our Lady of the Lake Regional Medical Center in Baton Rouge, La., for any patient who presents with suspicion of sepsis. It has resulted in substantial improvements in the speed of identification and treatment of patients with sepsis, and also given providers a signal to look for alternatives.

In one example, a patient presented to the ED with altered mental status and what was thought to be fever with little background information from emergency medical services or family. The host response technology signaled low risk for sepsis and the team continued searching for alternative causes. Ultimately, the team considered the elevated temperature as hyperthermia rather than fever, and they began treatment for serotonin syndrome with good response.

After nearly two years of using this host response technology, I have found it has significantly enhanced my practice. During those months when we see increases in patients, waiting rooms full of SIRS, and have bed holds in the ED, we now have a tool to quickly prioritize which patient we need to assess immediately, who can wait a little longer, what infections may be able to be treated at home, and who may need a closer look for a diagnosis other than sepsis when they are critically ill.

Claude M. D’Antonio, Jr. is an emergency physician.