Better Sepsis Care in Resource-Limited Settings
November 23, 2017
Happy Thanksgiving!
Early sepsis treatment with protocolized IV fluid, pressors, and transfusions vs usual care was undertaken in Zambia and resulted in much higher mortality. Most patients had HIV. This is yet another strike against a form of early goal-directed therapy (EGDT) vs. usual sepsis care.
Why does this matter?
EGDT came on the scene in 2001. It was helpful in focusing attention on identifying sepsis and early source control with antibiotics, but it used invasive monitoring of volume resuscitation, aggressive transfusions, dobutamine, and other atypical aspects of sepsis care. Subsequent studies over a decade later showed that usual care and a non-invasive monitoring approach were just as effective. See prior landmark articles for EGDT and PRoCESS. However, many argue that the usual care arm in ProCESS (as well as ARISE and ProMISe) adopted the key elements of EGDT, namely earlier antibiotics and more aggressive fluid administration, leading to a null effect. This study asks a very different question. How does protocolized sepsis care compare to usual care in Zambia where usual care is very different? Two prior African RCTs suggested that early aggressive IV fluid increased mortality in children and those with tachypnea and hypoxemia. Therefore, this study focused on septic adults without pre-existing respiratory distress.
Back to the basics for sepsis care in resource-limited settings
This was a non-blinded RCT with 209 septic and hypotensive patients (SBP<90 or MAP<65) divided between an early sepsis protocol specifying early IV fluid (2L + up to an additional 2L based on clinical re-assessment), peripheral dopamine, visual inspection of the jugular venous pressure, and transfusion if hemoglobin <7g/dL vs usual sepsis care with hemodynamic management at the discretion of the treating physician. Most patients were young, mid 30s, malnourished, at risk for TB and malaria, and 90% had HIV. Despite having poorer lactate clearance, the usual care group did markedly better, with 15% lower mortality (33% vs 48%). NNH for this early sepsis protocol was 7. Patients in the early intervention group received 3.5L of IV fluid in the first 6 hours vs only 2L in the usual care group. Additionally, patients in the early intervention group received more dopamine (14.2% vs 1.9%). Of note, patients in the intervention arm developed tachypnea and hypoxemia more frequently than those in the usual care arm, and there were no mechanical ventilators available for almost all patients.
Source
Effect of an Early Resuscitation Protocol on In-hospital Mortality Among Adults With Sepsis and Hypotension: A Randomized Clinical Trial. JAMA. 2017 Oct 3;318(13):1233-1240. doi: 10.1001/jama.2017.10913.
Peer reviewed by Thomas Davis, MD.
