Esmolol infusion reduced septic shock mortality by 40% in RCT

Do I.V. Beta-Blockers Save Lives in Septic Shock?

Catecholamines can be toxic -- just ask anyone experiencing the heartbreak of tako-tsubo syndrome. Blocking the heart-flogging effects of the hormones epinephrine (adrenaline) and norepinephrine have long been known to improve long-term survival in congestive heart failure. People with septic shock might be the last group you'd consider giving beta-blockers, but a pilot study showed oral metoprolol improved cardiac output and markers of perfusion among 40 septic shock patients, without apparent adverse effects. A new phase 2 randomized trial in JAMA takes the concept to the next step, and floats the provocative idea that aggressive catecholamine blockade with esmolol saves lives in people with septic shock.

What They Did

Andrea Morelli et al enrolled patients with septic shock while receiving norepinephrine infusions, who had not received beta-blocker therapy previously, at a single ICU at the University of Rome (Italy). The vast majority had either pneumonia or peritonitis, with median SAPS II scores of ~55, which would predict about a 60% chance of death.

All patients were fluid resuscitated, sedated with sufentanil and propofol and mechanically ventilated with low tidal volumes, and had pulmonary artery catheters. All also got continuous hydrocortisone infusions (300 mg / day). Levosimendan (a inotropic calcium sensitizer used in decompensated CHF) was given to patients with low mixed venous oxygen saturations, who would have received dobutamine in the U.S.

After 24 hours of optimization, 154 patients still in septic shock with heart rates > 94 / minute were randomized to receive open-label esmolol infusions to achieve a goal heart rate of 80-94 / minute, or usual care (i.e., there was no blinding of treatments; no placebo was given to the control group). Esmolol was continued until ICU discharge or death, with continuing fluid boluses and norepinephrine titration to achieve wedge pressure and mean arterial pressure goals.

The primary endpoint was esmolol's efficacy at lowering heart rate to the goal range; secondary endpoints included mortality, organ failure, etc.

What They Found

All patients in the esmolol group successfully achieved a goal heart rate of 80-94 / minute. Rather than needing more norepinephrine and I.V. fluids, they needed less: about 7-8 μg / minute less norepinephrine and about 500 mL less crystalloid in the average patient receiving esmolol. Hemodynamic variables (stroke volume, systemic vascular resistance, left ventricular stroke work indices) were all higher in those receiving esmolol. Kidney function, lactate and pH were all slightly improved in the esmolol group as well, compared to controls.

Most strikingly, the esmolol group had a 40% relative reduction in mortality at 28 days: 49% vs 81% in the control group (P < .001).

What It Means

I expect most experienced intensivists will be skeptical of the observed 40% mortality reduction with esmolol for septic shock. That's justified, since single-center trials are more likely to show large effects that are often reduced or eliminated entirely in subsequent larger multicenter randomized trials. The very high mortality in the control group -- even more than expected given the initial SAPS II scores -- raise the possibility that the large mortality benefit could be a chance finding. The open-label design (lack of blinding to treatment group by the care teams) increase the chance of unconscious bias in care delivery, with unmeasured differences in treatments between groups. It would be easy to unknowingly give 500 mL less saline over several days, or unconsciously encourage a reduction in vasopressor dose in patients receiving the intervention, if one were unblinded. (Hemodynamic variables, which also improved, would seem to be less influenceable by these potential biases.)

With those caveats, this is an exciting study showing benefits besides mortality (reduced need for vasopressors and I.V. fluids, and improved hemodynamic parameters) with a biologically plausible mechanism (reduced toxic effects of catecholamines and improved stroke volumes with longer diastolic filling times). There doesn't seem to be an adverse safety signal, and even if the mortality benefit seen is mostly chance, it makes harm from esmolol even less likely. Further, many will argue the very high mortality of patients with severe septic shock justifies the use of apparently safe, conceivably-but-not-definitively-proven-beneficial therapies like this.

As the authors rightly say, "Appropriately powered, randomized, controlled multicenter trials are required to confirm our findings." Baxter markets esmolol in the U.S. under the brand name Brevibloc, but the drug has been off-patent and available as a generic for years. The financial incentive may be lacking for industry sponsorship of a definitive trial, which would have to enroll hundreds of patients and cost millions. The lead author (Dr. Morelli) reported receiving honoraria in the past from Baxter, which did not directly fund this study.

Andrea Morelli et al. Effect of Heart Rate Control With Esmolol on Hemodynamic and Clinical Outcomes in Patients With Septic Shock: A Randomized Clinical Trial. JAMA. 2013;310(16):1683-1691. doi:10.1001/jama.2013.278477.