It took all of one day for emergency physicians to organize opposition after an update to the Surviving Sepsis Campaign Bundle was released online.
The day after the new version was released, Scott D. Weingart, MD, the editor-in-chief of EMCrit (https://emcrit.org), and Josh Farkas, MD, the editor-in-chief of PulmCrit (http://bit.ly/PulmCrit), posted a petition calling for its retraction. Their petition was signed by more than 4,000 people as of press time. (emcrit.org/sscpetition.)
The new bundle, written by Mitchell M. Levy, MD; Laura E. Evans, MD, MSc; and Andrew Rhodes, MBBS, all members of either the Surviving Sepsis Campaign executive or steering committees, contains an hour-one bundle that combines the three-hour and six-hour bundles from previous iterations of the Surviving Sepsis Campaign guidelines. The guidelines are described as evidence-based recommendations of how sepsis should be treated, practical advice for physicians on the front line, and was published simultaneously by the journals Critical Care Medicine and Intensive Care Medicine. (http://bit.ly/2x7Ud5d.)
The bundle concept, developed by the Institute for Healthcare Improvement, is designed to short-circuit the average time (17 years) it can take for new clinical findings to be put into practice. The 2018 hour-one bundle says resuscitation and management should begin immediately, defining sepsis as a medical emergency on par with serious trauma, heart attack, and stroke. It advises physicians to:
- Measure lactate level and measure it again if the initial finding was greater than 2 mmol/L. (Weak recommendation, low quality of evidence.)
- Obtain blood cultures before administering antibiotics. (Best practice statement.)
- Administer broad-spectrum antibiotics. (Strong recommendation, moderate quality of evidence.)
- Begin rapid administration of 30 ml/kg of crystalloid for low blood pressure or lactate above or equal to 4 mmol/L. (Strong recommendation, low quality of evidence.)
- Begin vasopressors if the patient has low blood pressure during or after fluid resuscitation to main maximum arterial pressure (MAP) greater than or equal to 6 mm Hg. (Strong recommendation, moderate quality of evidence.)
Sepsis in a Vacuum
Dr. Weingart, who opposes the new bundle, said, “It sounds wonderful. Let's treat these patients as quickly as possible. [But the authors] have fallen into the trap of many guideline writers who pretend their topic exists in a vacuum.”
The emergency department, however, is not a vacuum, he said, and there is no leeway time. Determining who has sepsis or septic shock and who does not can be complicated, Dr. Weingart noted. “They actually said in some of their materials that they see patients being treated quickly for trauma and heart attack. Why not the same for septic patients? But they are not showing us a septic patient with an extremely low blood pressure, breathing 80 times a minute. That would be comparable with a patient who fell out a four-story window. Then the one-hour bundle would be reasonable,” he said.
Instead, Dr. Weingart said, emergency physicians could be faced with one really sick patient, nine not-so-sick patients, and 40 others who need care. They would have to start treating them all within an hour if they adhered to the bundle, including exposing patients who might not have sepsis to powerful antibiotics just to comply with the bundles. “We won't be able to wait for cultures to come back,” he said.
The petition outlines these concerns, leveling criticism that the recommendations are based on low to moderate quality of evidence but are made at a strong level. The bundle also ignores clinician judgment about whether the recommendations will help or harm an individual patient and does not consider the effect the guidelines will have on nonseptic patients also receiving care in the emergency department.
Tiffany Osborn, MD, an emergency physician who was on the panel for the 2016 guidelines, said she was unaware of the 2018 bundle update until colleagues pointed it out to her. She said she remains unclear, even after talking to one of the writers of that version, about the rationale for a bundle that varies so dramatically from the carefully crafted 2016 guideline that had an hour-three goal for initial treatment.
Dr. Osborn, a professor of surgery and emergency medicine at Washington University School of Medicine in St. Louis, said she didn't understand why the shift between the three- and the one-hour goal occurred, noting that it's a change that affects emergency physicians most. Perhaps, she said, they should have focused on their own service line and encouraged getting patients into the intensive care unit as soon as possible and ensuring that the patients received resuscitative fluids, blood draw for lactate, and cultures and antibiotics as soon as possible in the place where the patient will receive ongoing care.
Alan Jones, MD, a professor and the chair of emergency medicine at the University of Mississippi Medical Center, said he is concerned that the bundles could have a negative effect on emergency physicians. “The leaders of the guidelines got the bundles [for previous guidelines] endorsed by CMS [the federal Centers for Medicare and Medicaid] as a quality metric for treatment of patients with sepsis,” he said. “If you didn't treat by the bundle, payment could be withheld because you didn't provide high quality care.
“You can have bundles without the guidelines, but the two things are often confused. The bundles are of some concern because some recommendations are not as evidence-based as one would like. It was a bigger concern when CMS picked up the bundle as a potential quality metric. There is not sufficient evidence to support that that's what we should be doing in all cases,” he said. “Sepsis is complex, and can be difficult to detect. Treatment can be dictated by underlying comorbidities of patient acuity and age.”
Dr. Jones said some physicians may end up doing the wrong thing for some patients. “Just because you have a process in place doesn't mean you have provided the best care for the patient,” he said, noting that evidence shows that patients receiving early fluid resuscitation have improved outcome. “But the recommendation of 30 ml/kg for all patients is not evidence-based.” Some patients, including those with heart failure, kidney and liver problems, and adult respiratory distress syndrome, may be harmed by that volume of fluid, he said.
Mitchell Levy, MD, an author of the 2018 Surviving Sepsis Bundle update, said it was not unusual for the revision to be written by members of the executive and steering committees as other interim guides have been in the past. He called the hour-one bundle “aspirational,” and acknowledged that it will not always be possible to accomplish the bundle's recommendations in an hour. “It's always been an ongoing debate [and] discussion between emergency clinicians, nurses, and intensive care clinicians,” Dr. Levy said, admitting that getting a patient checked in, evaluated, and started on fluid resuscitation and antibiotics after drawing blood for cultures will be difficult to do within an hour. He maintained, however, that that should be the goal.
The alternative would be to start the clock when the clinician writes a sepsis diagnosis in the chart, he said, but added that “would miss the window of opportunity to improve care.”
The New York Department of Health uses the 2016 bundle as a metric, but not as a punishment, Dr. Levy said. Instead, hospitals are measured on how close they come to the bundle's standards in comparison with other hospitals in the state. “If you are 60 percent compliant with the measure, you could end up being in the 95th percentile in comparison with the other hospitals in the state of New York,” he said. CMS will ultimately use a similar kind of metric nationwide, he added.
Dr. Levy said he thinks the new tweak of the guidelines makes sense. “When an emergency physician sees a patient with sepsis, he is not going to wait three hours to start treatment if he thinks the patient needs fluids. If there is a mean arterial blood pressure of 50, he will start the patient on vasopressors. That reflects clinical reality.”
“No matter what you publish, some people just want to argue with it,” Dr. Levy said. “CMS has picked up our measures, and they are mandated. Physicians don't like to be told what to do.”