Should lactate clearance replace SvO2 in sepsis protocols? (Pro/Con, CHEST)

(This post is originally from March 2012, but a few people requested to see it again, so here it is. -MH)

It started with a friendly pro/con debate in the December 2011 Chest, about whether lactate clearance or mixed venous oxygen saturation is a better “goal” for early goal-directed therapy in severe sepsis and septic shock. It ended with Alan Jones reviving the rumors and innuendo that have swirled for years around Emanuel Rivers’s body of work.

Most intensivists agree that people in septic shock who are hypoperfusing their organs should be considered for inotropic therapy, typically dobutamine. (I’m avoiding the whole transfusion controversy here.) Continuous mixed venous oxygenation saturation monitoring (SvO2), which appeared to be central to the mortality benefit seen in Rivers’s 2001 NEJM landmark trial, is expensive and requires a pulmonary artery catheter. Intermittently-obtained central venous oxygen saturation (ScvO2) correlates well with central venous oxygen saturation, and is considered a proper substitute for continuous mixed venous O2 saturation. In practice, central venous oxygenation is probably what most intensivists use.

The randomized LACTATES trial compared early goal directed therapy using lactate clearance (change in lactate / original lactate / time elapsed) against a strategy using ScvO2, and showed equivalent outcomes (mortality) in the two groups. Alan Jones (PI of LACTATES) squares off against the threesome of Emanuel Rivers, Ronald Elkin and Chad Cannon to make the case that lactate clearance is more useful and should replace SvO2 in EGDT.

Jones’s most compelling point is that either central or mixed venous O2 sats can be normal or high in severe sepsis, even in the presence of severe organ hypoperfusion, because if oxygen extraction varies throughout the body (as it does during sepsis), the mixed blood can “average out” to a misleadingly normal/high value. Therefore, following ScvO2 results in mismanagement. Also, lactate clearance has repeatedly been shown to independently predict survival from sepsis, whereas SvO2 (as a single variable) has not. 

Rivers and friends counter that ScvO2 is more sensitive as an early-warning system in sepsis than lactate (based on their own data), that lactate kinetics are too complex and variable between patients to use it as a standard, that lactate levels often don’t rise even in life threatening sepsis, and that using percent lactate clearance is oversimplified (e.g., it considers a lactate going from 10 to 9 as being the same as one from 4 to 3.6: both are a 10% clearance). They also bemoan the fact that people have interpreted Jones’s LACTATES to mean central lines are not needed at all in severe sepsis (although of course they are under current practice standards, to measure central venous pressures).

But in the rebuttals, things really heated up.

In response to criticism of LACTATES’s methodology, Jones revives the persistent suspicions (by some) surrounding Manny Rivers’s original single center trial, even casting aspersions on the care delivered at Henry Ford, noting Rivers’s original control group mortality was “20% higher than any septic shock mortality reported in the recent literature.” Jones also suggests peculiarity of Rivers’s data collected between 1997 and 2000, calling it “markedly different than any other septic shock population reported in the world’s literature” and wondering if “systematic selection bias was a significant problem in the their study.”

As reported in the Wall Street Journal, prominent academics have wondered publicly about the dramatic benefit seen in Rivers’s original study, particularly about the less-than-perfect clarity initially as to whether he included all patients properly according to intention-to-treat. Rivers has repeatedly assured everyone that no patients were cherry-picked or excluded. But a medical resident at Henry Ford doing a follow-up study raised a red flag as to the integrity of the original data, pointing out there were 25 patients who should have been included and had they been, would have reduced or eliminated the trial’s findings. The hospital performed its own internal analysis of the data, and publicly announced everything was on the up-and-up — but declined to share the primary data with anyone on the outside, the WSJ story says. Since Rivers and his hospital have had a longstanding complex relationship with Edwards Lifesciences, makers of the SvO2 monitoring catheter, the whole thing is redolent with intrigue.

I saw Manny Rivers speak once. He gave a great presentation. The main thing I remember about it, though, was that he announced “nothing to disclose” and then showed several videos of his septic shock patients, during which he kept zooming in for tight shots on the Edwards Lifesciences logo on the SvO2 monitor, holding focus for several suspenseful, brand-building seconds. (Wouldn’t you know it, there was a 0% mortality for those three patients! I felt like I should be getting a free lunch, at least.)

Clinical Takeaway: All the combatants agree that ScvO2 and lactate provide complementary information. I’ll keep checking both, and hope the forthcoming Pittsburgh study lights our way.

Jones AE. Point: Should Lactate Clearance Be Substituted for Central Venous Oxygen Saturation as Goals of Early Severe Sepsis and Septic Shock Therapy? Yes. CHEST 2011;140:1406-1408.

Rivers EP et al. Counterpoint: Should Lactate Clearance Be Substituted for Central Venous Oxygen Saturation as Goals of Early Severe Sepsis and Septic Shock Therapy? No. CHEST 2011;140:1408-1413. 

Rebuttal from Alan Jones

Rebuttal from Manny Rivers