What's New in Critical Illness and Injury Science? Utility of central venous oxygen saturation to risk stratify septic patients.

Sepsis is defined as a life-threatening organ dysfunction caused by a dysregulated host response to infection. Despite advances in medical therapy, sepsis-associated morbidity and mortality remains high. Globally, sepsis-associated mortality may be as high as 60%,[1] whereas in the United States, sepsis ranks among the leading causes of death by contributing to 33%–50% of in-hospital deaths.[2] The financial strain that sepsis exerts on the medical system topped $24 billion in 2013,[3] a sharp rise from $20 billion just 4 years prior.[4]

Numerous studies have evaluated the use of goal-directed resuscitation in sepsis management, with varying results. In the early 2000s, early goal-directed therapy (EGDT) was promoted to identify high-risk patients, obtain source control, and administer appropriate antibiotics.[5] Therapy was directed at the optimization of oxygen delivery and hemodynamics by using surrogate markers and measurements including surrogates for preload (central venous pressure), afterload (mean arterial pressure), arterial oxygen content, cardiac output, and oxygen consumption (mixed venous oxygen saturation [SvO2]). In 2001, Rivers et al. famously reported that EGDT improved survival in patients with sepsis and septic shock,[5] and subsequently EGDT was widely incorporated into the first 6 h of sepsis management. Following the reports of methodological and data integrity concerns, several validation studies were conducted. Ultimately, the results of the ProCESS, ARISE, and PROMISE trials challenged the EGDT paradigm by reporting that protocol-based resuscitation of sepsis patients did not improve mortality outcomes.[678] While protocol-based resuscitation failed to show consistent outcomes of improvement in mortality, the targeting of individual components of guided treatment has garnered research interest with some evidence to substantiate improved outcomes.

SvO2 is measured in blood taken from the pulmonary artery via a pulmonary artery catheter. It has been a target of resuscitation in previous studies of EGDT but has lost favorability due to invasive means needed to achieve measurements. Central SvO2(ScvO2) has long been studied as a prognostic and target marker of resuscitation in patients with shock.[9] It is a surrogate of SvO2 and can be measured via traditional central venous access. ScvO2 is strongly correlated with SvO2,[510] but differs as it measures the oxygen supply–consumption ratio of only half of the body (upper if sampled from the superior vena cava and lower if sampled from the inferior vena cava). Moreover, ScvO2 overestimates SvO2 by 1.7% ±7.1% in sepsis patients.[1112] It is simple to measure and may have important prognosticative value in patients with sepsis and septic shock. In this issue of the International Journal of Critical Illness and Injury Science, Kumar et al. report the results of a prospective observational cohort study assessing the prognostic significance of ScvO2 among septic patients in the emergency department.

ScvO2 is thought to be a useful indicator of overall oxygen delivery to and consumption of the cells and tissues in the body.[1314] Low ScvO2 indicates decreased oxygen delivery and/or increased oxygen extraction to the tissues in the body. Normal ScvO2 indicates that oxygen delivery is adequate and can be seen in states of cardiac compensation. High ScvO2 indicates that either oxygen delivery is in excess to oxygen demand, or there is a decrease in oxygen extraction due to microvascular or mitochondrial dysfunction.[11]

It has been reported that extreme ScvO2 levels (low or high) have been associated with increased mortality in emergency department and intensive care unit patients with sepsis.[111516] Tissue hypoxia (whether from decreased oxygen delivery, increased consumption, or decreased extraction) may promote lactate overproduction, an independent predictor itself of sepsis mortality. The LACTATES study concluded that lactate clearance is equivalent to ScvO2 in the management of individual patients, but failed to demonstrate a mortality benefit by targeting lactate clearance or ScvO2 normalization.[17] A multicentric trial by Arnold et al. found that 79% of patients with persistently elevated lactate had ScvO2 values >70%.[18] While studies have inconsistently reported that lactate clearance may be associated with improved mortality, studies on general critical care patients,[19] as well as those with sepsis,[17202122] have not shown normalization of ScvO2 to be a better mortality predictor.[22] Moreover, its additional prognostic value may lie in cases where lactate values fail to normalize after resuscitation.[22]

Although the study by Kumar et al. did find the expected differences in baseline illness severity scores and lactate between survivors and nonsurvivors, the lactate did not differ in such a way to be able to distinguish those with high mortality in a clinically meaningful way. Moreover, neither set of variables differed significantly between groups when stratified by ScvO2 into hypoxia, normoxia, or hyperoxia groups. As such, the picture regarding the use of lactate clearance and normalization of ScvO2 as goal-directed resuscitation end points remains unclear, and further prospective randomized assessments are needed.