β-BLOCKER IN SEPSIS – A TRIAL

iPhoneIcon_Big (3)A small study with patients randomised to standard ICU sepsis therapy, or ICU sepsis therapy with beta blockage. We’ve written about the potential for β-blockage in sepsis earlier. This is the first clinical trial. It shows much better survival for the β-blocker group, but there are a few bumps in the road…

Sympathetic storm
The theory is that critical illness like severe sepsis induces a sympathetic storm that overshoots what is needed to keep the body alive. This cathecholamine storm might be harmful for the patient. β-blocking them could countereffect this. We wrote on the theory after a retrospective study here. There is also a good overview article from Shock 2009 here.

The Morelli group did a pilot study first, documenting the effects of beta-blocking in sepsis, and it all looked favourable. They concluded: “This pilot study demonstrated that heart rate control by a titrated esmolol infusion in septic shock patients was associated with maintenance of stroke volume, preserved microvascular blood flow, and a reduction in norepinephrine requirements”. So they did a small randomised trial.

The study
77 patients were randomised to β-blockers or standard treatment. To control the beta-blockade, they used the ultra short-acting β-blocker esmolol on an infusion pump. Clever choice, as they could titrate themselves to the predefined goal of a heart rate between 80-94/min. They mention heart rate might not be the optimal target for the β-blocker group, but it was the easiest and most practical to monitor. Seems fair.

Results
The esmolol group almost halved the mortality of the control group! So this is all good? Hang on: The esmolol group had a mortality of 49%, which seems high. Having nearly halved the mortality from the control group, the control group had a mortality of 80%.

80% mortality in the control group?!? Either these were the sickest sepsis patients the world has ever seen, or the control group treatment must’ve consisted of locking the control group in a room and opening it on day three to see who’s still alive. As a cohort, even a severly sick sepsis cohort, the mortality rate shouldn’t be higher than 40-50% at the most. The latest numbers are showing an overall mortality from severe sepsis of 29%, as we have reported here. So something is not right in this 2012 beta-blocker study. Even with a high incidence of multiresistant bacteria as reported in the Morelli study, it’s hard to explain those mortality rates.

That being said, this study was first and foremost a safety study, and esmolol was shown to give no adverse outcomes. Of course, with the extreme mortality in the control group, adverse outcomes in the esmolol group might be hard to catch. Still, future, larger scale trials are surely in the works, and the results will be very interesting to see.

Conclusion
Impressive mortality reduction by esmolol, but needs additional studies – or a very good explanation on the extremely high overall mortality rates seen in this trial. β-blocker in sepsis was used here without adverse outcomes and is still an interesting concept. But for the time being, nothing more than that.

Effect of Heart Rate Control With Esmolol on Hemodynamic and Clinical Outcomes in Patients With Septic Shock, JAMA, 2013.

Microvascular effects of heart rate control with esmolol in patients with septic shock: a pilot study, Crit Care Med, 2013.

Review article on β-blockers in sepsis:
beta-Blockers in sepsis: reexamining the evidence, Shock, 2009.

This entry was posted in Intensive Care. Bookmark the permalink.

9 Responses to β-BLOCKER IN SEPSIS – A TRIAL

  1. nfkb says:

    hi ! thank you for this post

    the least we can say is that this study makes people talk ;)

    I’m ok with the sympathetic overshoot theory but i’m not ok with the way doctors always try to counteract things in critical care : too much inflammation, let’s give anti-TNF, too much oxidative stress, let’s give anti-oxydants, elevated heart rate ? let’s give beta-blocker !

    I think we have to look for new paradigms to significantly improve our care. I don’t wait for a magic bullet. We already have so many things to setup to improve our care, beginning with less iatrogenic disease ! i understand that may sound a little bit pessimistic but i’m a little bit angry against this run to publish exciting datas loosing common sense… I you want to publish the editor may ask you some really tricky stuff non relevant to common practice… But here you’ve got an exciting article so let’s go for it event if the mortality rate is 80%, no problem !

    About interfering with sympathetic overshoot, i’m a big fan of alpha2-agonists sedation :)

  2. Thomas D says:

    I hear you! Some of this turns out a little crazy, but I still think there’s reason to counteracting the body’s response when it gets out of proportion. The problem is to tell when the body’s response is appropriate and helpful and when it is over-reacting. We all agree an anaphylactic reaction is overshoot and needs counteracting. In other settings, it’s not so clear. The classic example is fever. Good or bad?

    However, it’s not just the reaction from the body, but also the hole setting. In critical care, the patient might be out of reserves, and can’t handle otherwise healthy reactions. Like shivering in hypothermia uses a lot of energy. If you already have metabolic challenges i.e. from impaired ventilation, you don’t need a shivering body using even more energy. So in this setting, shivering might be inappropriate, and needs counteracting – while warming the patient artificially.

    But I agree, we need to think hard about counteracting the body’s responses. ‘Normal’ values might not be normal or what we want at all in a critically ill patient!

  3. nfkb says:

    yep, i’ve been a little bit too quick in my answer ;) my main message is that “publish or perish” pushes us into a weird zone !

    • Thomas D says:

      Publish or perish doesn’t just push us into a weird zone, it also produces a lot of uninteresting papers that create a lot of noise and makes it hard to find the important stuff.

  4. Kristian says:

    The mortality is staggering. However this is subgroup of septic shock with poor prognosis and even the 6S trial had a mortality above 40%. These patient are selected due to their lack of ability to improve despite initial 24-resuscitation. They have persistent shock with tachycardia which certainly has a higher mortality than 50%. Throw in some resistant bugs and you may approach an 80% mortality when considering the consistent poorer performance of Southern European ICUs compared to those in Scandinavia.
    I do agree that the study results seem to good to be true. They are not studying the invention of penicillin. It does however put another nail in the coffin of dobutamine in septic shock. At this point I believe that we should consider dobutamine toxic when used for prolonged resuscitation. Although I do not find this study to be sufficient evidence to protocolize esmolol therapy just yet, the paper has influenced my practice. I have a harder time accepting tachycardia and easily turn to low-dose metoprolol or amiodarone having no current access to esmolol.

    • nfkb says:

      hi !

      I would be very careful with amiodarone and severely injured lungs as we may see in septic shock.

      In our ICU we are also strong believers that easy/routine dobutamine is harmful. But what is the alternative for patients with a severely impared contractile function ?

      • Kristian says:

        I agree with you that there are probably much better options than amiodarone for rate control.
        We tend to use levosimendan for refractory shock when contractility continues to be poor after the first 24 hours. Evidence is almost completely lacking, but this is also done in the esmolol trial. As a bonus it has a nice effect on the pulmonary circulation.

        • nfkb says:

          thank you for your answer

          i thought levosimendan as some kind of long effect which makes it difficult to use in the context of septic shock, can you tell me a little bit more about your experience with levosimendan and sepsis ?

          thanks

          • Kristian says:

            Using levosimendan is a bit tricky due to its prolonged effect and vasodilation. Never use bolus or you will probably kill the patient. Start out with 0,05 mcgr/kg/min and expect increased need for pressor. If the patient tolerates it we increase to 0,1, but rarely go higher in septic shock. The need for a cardiac output monitor in this setting is obvious. We tend to let the infusion run for a day or two.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>