What We Can Learn From Treating Sepsis in Low-Resource Settings

Gregory S. Martin, MD, MSc


December 08, 2017

Effect of an Early Resuscitation Protocol on In-hospital Mortality Among Adults With Sepsis and Hypotension: A Randomized Clinical Trial

Andrews B, Semler MW, Muchemwa L, et al
JAMA. 2017;318:1233-1240

Study Summary

Early and targeted sepsis resuscitation[1] has become the cornerstone of effective sepsis therapy,[2,3] yet questions still exist about implementation and efficacy in low-resource settings.[4] The authors of this study, which took place in Zambia, sought to determine whether early resuscitation with intravenous fluids, vasopressors, and blood transfusion would decrease mortality as compared with usual care. They randomly assigned 212 patients, of whom 90% had HIV and 209 completed the study, and observed that in-hospital mortality was higher in the early resuscitation group (48.1% vs 33.0%; relative risk, 1.46; P=.03). During the 6 hours of the resuscitation protocol, patients in the sepsis protocol group received, on average, 1.2 L more intravenous crystalloid fluid compared with the usual care group (P<.001), and 14.2% versus 1.9% received vasopressors (P<.001). The authors concluded that among adults with sepsis, hypotension, and HIV in a resource-limited setting, an early resuscitation sepsis protocol increases in-hospital mortality.


This study challenges the conventional wisdom that: (1) early targeted resuscitation of sepsis patients with fluids, vasopressors, and blood products reduces mortality; and (2) that bundles of care are effective at saving lives. Is it possible that the recent ProMISe, ProCESS, and ARISE trials,[5] which each failed to show benefit for protocolized early sepsis resuscitation bundles, are best interpreted to mean that such an approach to care is inappropriate and even harmful? No; that would be an incorrect conclusion. But this new study from Andrews and colleagues does challenge what we know about fluid resuscitation in sepsis, timing, fluid and blood product composition, and monitoring.

An earlier study offering fluid resuscitation to children with sepsis in a low-resource setting in Africa similarly found an increase in mortality for those receiving fluids,[6] with the adverse effects seemingly related to cardiovascular complications rather than fluid overload,[7] which may have been expected in a setting with limited ability to monitor patients during and after fluid administration. The Andrews study raises a concern about use of blood products in sepsis resuscitation, as that is a notable difference from the Maitland study, yet there is no direct comparison to the Andrews study, particularly using the Maitland study of African children. Furthermore, sepsis resuscitation bundles have demonstrated conflicting efficacy in other lower-middle-income countries,[8,9] including the suggestion that full-bundle compliance is less important than early recognition.[10] Taken together, these studies suggest that early fluid resuscitation may have adverse consequences that are mitigated by rigorous monitoring and intensive care systems available in high-resource settings. Another interpretation is that these adverse consequences are unique to the low-resource environment of the patient population (such as the prevalence of HIV, which may be different from in the United States, where it varies greatly).[11] In the end, we need to further explore the physiologic and clinical outcome effects of fluid resuscitation in sepsis, disentangled from other components of bundled sepsis care; but until then, we should continue following guidelines such as the Surviving Sepsis Campaign because of their broadly demonstrated efficacy.[3]


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.