Morelli

Effect of Heart Rate Control With Esmolol on Hemodynamic and Clinical Outcomes in Patients With Septic Shock – A Randomized Clinical Trial

Morelli et al. JAMA 2013; 310(16):1683-1691

Clinical Question

  • Does the use of the short-acting β-blocker esmolol effectively control heart rate (HR) in patients with septic shock?

Design

  • Phase 2 randomised controlled trial
  • Computer based randomisation
  • Non-placebo
  • Non-blinded
  • Sample size calculation
    • 128 patients required to to detect a 20% change in HR with:
      • estimated standard deviation 40%
      • probability of false:
        • positive trial 5%
        • negative trial 20%

Setting

  • Single centre, mixed-ICU in Rome, Italy
  • Conducted between November 2010 – July 2012

Population

  • Inclusion:
    • Adults ≥18 years old with diagnosis of septic shock
    • Requiring noradrenaline to maintain mean arterial pressure ≥65 mmHg after 24 hours of haemodynamic optimisation in ICU
    • Heart rate ≥95/min
  • Exclusion:
    • Prior treatment with β-blockers; Pronounced cardiac dysfunction (Cardiac index ≤2.2 + PAOP >18mmHg); significant valvular heart disease; pregnancy
  • 154 randomised out of 336 patients assessed for eligibility

Intervention

  • Esmolol infusion
    • 25-2000mg/hour
    • targeting HR 80-94
    • continued for entire length of ICU stay

Control

  • Standard care
All patients in both intervention + control groups managed with standardised practice:
    • Intubated & ventilated, with restrictive tidal volume strategies.
    • Propofol & sufentanil used as sedative agents.
    • Pulmonary artery catheters inserted
    • In the 1st 96 hours, received fluid resuscitation, targeting CVP >8mmHg, PAOP >12mmHg, mixed venous oxygen saturations >65%.
    • Noradrenaline titrated to maintain MAP >65 mmHg.
    • IV hydrocortisone (continuous infusion 300mg/d).
    • Transfused if Hb <7 or clinical signs of inadequate systemic oxygen supply
    • Evidence of inadequate systemic oxygen delivery, despite Hb >8, treated with levosimendan.

Outcome

  • Primary outcome: HR lowered to <95 + maintained at 80-94 for ICU stay
    • significantly more likely in esmolol group
  • Secondary outcomes:
    • Noradrenaline and fluid requirements significantly reduced in esmolol group
    • Stroke volume + systemic vascular resistance index significantly increased in esmolol group
    • Mortality significantly reduced in esmolol group

Authors’ Conclusions

  • For patients in septic shock, the use of esmolol reduced heart rate to target levels. Compared with standard treatment esmolol also increased stroke volume, reduced noradrenaline requirements + was associated with an improved 28 day mortality.

Strengths

  • Randomised controlled trial
  • Powered for primary outcome
  • Standardised resuscitation strategies

Weaknesses

  • Single centre, non-blinded, no placebo (potential for overestimate of effect size, and for unconscious bias in care delivery)
  • Primary outcome is not clinically relevant
  • Not powered for secondary outcomes e.g. mortality
  • Usual care was significantly different to standard UK practice (PA catheters, ubiquitous use of steroids, fluid resuscitation for 96 hours guided by filling pressures, high use of levosimendan 49.4% in esmolol group). Therefore, significant concerns about generalisability of results.
  • The patients in this study appear to have been very unwell at baseline (median dose of noradrenaline approaching 0.4ug/kg/min at enrolment), with a high prevalence of multi-resistant organisms cultured. Despite this, 28 day mortality of 80% in the control group is concerning, and raises substantial questions of whether a similar results would be seen in patients with a lower severity of illness. As Pinksy states “80% mortality in the control group can hide a lot of sin.”
  • A non-standard method of calculating SAPS II scores was used (at study enrolment after resuscitation, rather than taking the worst values in the first 24 hours of ICU admission) for reasons that aren’t well explained in the paper. This makes it difficult to compare this cohort with those in other intensive care units.

The Bottom Line

  • The use of esmolol for tachycardic patients in septic shock resulted in good control of heart rate, and a reduction in noradrenaline dose, without any significant side effects.  A large mortality benefit was reported with the use of esmolol; but with a number of  study limitations, further evidence is required before a change in practice is justified.

External Links

Metadata

Summary author: @MikeW_J (from Portsmouth journal Club)
Summary date: 10 November 2014
Peer-review editor: @davidslessor

Leave a Reply