By Michael Filbin, MD
Results of recent studies lead to more effective approaches to detecting and treating septic shock. These approaches need to now become integrated into clinical practice.
Septic shock is a common and deadly disease that has not enjoyed significant breakthroughs over the past two decades in diagnostics or therapeutics, unlike other common emergency conditions such as heart attack, stroke, and trauma. Until recently, treatment of septic shock in the emergency department had been informed by two relatively small randomized-controlled trials. The first by Dr. Emanuel Rivers in 2001, a single-center trial enrolling only 263 patients, revolutionized how we think about septic shock. Rivers demonstrated that early protocolized therapy with aggressive, goal-oriented fluid resuscitation and restoration of tissue perfusion in the emergency department leads to a significant reduction in mortality.
This heralded the era of early goal-directed therapy (EGDT) for septic shock. EGDT formed the basis of and inspiration for the international Surviving Sepsis Campaign, providing guidelines for physicians around the globe on how best to treat sepsis. Dr. Rivers incorporated continuous central venous pressure (CVP) and oxygen saturation (ScvO2) monitoring as the foundation for goals of resuscitation. However, these measurements require placement of a central venous catheter, an invasive, time-consuming procedure that is often not feasible in a busy, resource-stretched emergency department. Furthermore, there is ongoing debate regarding the usefulness of CVP and ScvO2 as surrogates for intravascular volume status and tissue perfusion, as Dr. Rivers suggested they be used.
A subsequent study demonstrated that optimization of venous lactate as the ultimate resuscitation goal instead of ScvO2 results in equivalent outcomes, with the advantage that lactate measurement does not require central venous catheterization. A larger question regarding EGDT was whether invasive monitoring employed in the Rivers’ trial was responsible for the mortality benefit, or whether it was the mere presence of a study team focused on aggressive fluid resuscitation and optimizing tissue perfusion that made the difference.
The Protocolized Care for Early Septic Shock (ProCESS) trial was designed to answer this question, and to determine whether the findings of Rivers’ single-center study could be generalized to other US emergency departments. ProCESS was funded by the National Institute of Health (NIH) and conducted in over 30 emergency departments in the US. It randomized patients to Rivers’ EGDT executed by a study team versus usual emergency department care. Additionally, there was a third arm that used protocolized care delivered by the study team, but employing traditional, non-invasive goals of therapy (thus not requiring central venous catheterization) such as cuff-measured blood pressure, shock index (ratio of heart rate to systolic blood pressure), urine output, and clinical impression of adequate tissue perfusion.
After 5 years of enrollment and 1,341 patients randomized, the long-anticipated ProCESS findings were published in the New England Journal of Medicine (April 2014). Surprisingly, there was no mortality difference between the three arms (18-21%), with mortalities significantly lower than the 30% mortality seen in the study arm of the Rivers’ trial. Concurrently, the Journal of the American Medical Association (JAMA) published a 10-year review of all patients admitted to Australian and New Zealand ICUs with sepsis (over 100,000 patients studied), and found that the overall baseline mortality for sepsis has declined from 35% to 18% over the past decade. This overall mortality decline has been attributed to the Surviving Sepsis Campaign effort and increased disease awareness associated with it.
So what conclusions can be taken from the results of ProCESS? All three arms of the study had three common elements: 1. Early identification of septic shock by research assistants, reminding clinicians that there was a “septic” patient eligible for study, 2. Early and aggressive intravenous fluid bolus that was required for study inclusion, and 3. Broad-spectrum antibiotics, which are now standard in sepsis care. Therefore, once septic shock has been identified early, and aggressively addressed with basic therapies, the specific method by which protocolized care is delivered does not matter, as long as it adheres to general principals promoted by Dr. Rivers and the Surviving Sepsis Campaign guidelines. Specifically, central venous catheter derived variables such as CVP and ScvO2 are not superior to variables available through traditional non-invasive methods, such as vital sign assessment and clinical judgment.
Yet the onset of septic shock is often insidious, and initial recognition can be difficult and associated with significant delays. Our emergency departments are increasingly overwhelmed with large numbers of patients. Physicians are forced to rely upon an initial brief evaluation of a patient, with little data available, an encounter that often dictates the course of treatment for subsequent hours. The future challenge is to identify early those at risk for developing septic shock, and employ advanced monitoring techniques to provide warning of decompensation. For patients identified as having septic shock, the challenge is to develop easily-obtainable, non-invasive yet relevant resuscitation goals.
Specifically, we need intelligent systems that factor human judgment (e.g., suspicion for infection), static data available from the electronic health record (e.g., medical history components), and dynamic vital sign data trended over time. We need fast and accurate means of monitoring non-invasive goals of resuscitation such as cardiac output and arterial blood pressure. We need to integrate these measured goals with the therapies that are being delivered (i.e., fluid and vasopressors) to evaluate the response to therapy. Many of these technologies exist in some form, but have yet to be introduced into clinical practice in a non-intrusive, effective way.
ProCESS has reinforced the concept that disease awareness and good clinical judgment around the use of basic therapies is as effective as invasive monitoring techniques. However, clinical trials do not necessarily mirror every day practice. The time has come to merge clinical acumen with advanced technologies to diagnose sepsis earlier and avoid preventable deterioration.
For Further Reading
Michael Filbin, MD, is an emergency physician at Massachusetts General Hospital and Assistant Professor at Harvard Medical School. Dr. Filbin attended medical school at Baylor College of Medicine and did his residency training in the Harvard-Affiliated Emergency Medicine Residency (HAEMR) program. Dr. Filbin’s research interest is in human clinical trials of septic shock with a particular focus on early identification and intervention. Read more