Written by:
Dr Hannah Gardner & Dr Helen Vesey (ICU Fellows)
Peer reviewed by:
Dr Jonny Wilkinsoin & Dr Dave Popple
Intro and background to this…
Every year, there are at least 250,000 cases of sepsis in the UK and 52,000 deaths. According to the ICNARC data from 2010-2013, there were 383,314 admissions to critical care centres across the country due to sepsis. We certainly see our fair share of it in Northampton. Sepsis is responsible for more deaths than breast, bowel and prostate cancer combined and so we are always interested to find innovative ways to try and treat these patients. We tried activated protein C and we are even talking a lot about vitamin C!
Sepsis…
Sepsis, according to the 2016 sepsis-3 study, is defined as life-threatening organ dysfunction caused by a dysregulated host response to infection. Life-threatening organ dysfunction is defined as an increase of two or more points in the SOFA score.
This video explains how the SOFA score is calculated and the associated mortality rates. See our mortality calculators here.
Sepsis, if left untreated, can lead to multi-organ damage and eventually, death. As we know, septic patients often become tachycardic as a physiological mechanism to increase cardiac output and therefore oxygen delivery.
The video below summarises what happens in septic shock.
Morelli…
A trial by Morelli et al in 2013, looking at the effect of esmolol on heart rate and clinical outcomes in septic shock, sparked a wealth of discussion on what seemed like an oxymoron!
Beta-blockers…
A little reminder of how beta-blockers work will follow and it’s worth looking at the pros and cons of using these drugs in a septic patient.
Beta adrenoceptor antagonists were first discovered in 1958 and are used in the treatment of angina, heart failure, hypertension as well as glaucoma, migraine and anxiety.
As you can see from the image below, beta-blockers act on beta 1 and beta 2 receptors, preventing the binding of epinephrine and norepinephrine and thus blocking the excitatory effects these endogenous catecholamines have.
Click the image for the pdf
Click here for the moviegraphic
Check out this great video on the receptors themselves:
Cardioselective beta blockers
Cardio selective beta-blockers such as atenolol, primarily bind to beta 1 receptors within the heart. These are primarily found in cardiac nodal tissue, the conducting system and the contracting myocytes. By blocking these receptors, they compete for the binding site and therefore prevent norepinephrine or epinephrine from binding there. In doing this, they cause a decrease in heart rate, a decrease in contractility, slower AVN conduction, and decreased cardiac workload.
Non-selective beta-blockers such as propranolol, inhibit all of the beta receptors and therefore can cause bronchoconstriction, by their action on the beta 2 receptors in the airways. Some can initiate a modicum of vasoconstriction as well. β2-adrenoceptors have only a small modulatory role on basal vascular tone. But, when blocked, there is a small degree of vasoconstriction in many vascular beds. This occurs because beta-blockers attenuate the β2-adrenoceptor vasodilator influence normally opposing the more dominant alpha-adrenoceptor mediated vasoconstricton.
An interesting point to note is that all cardio selective beta blockers begin with the letters A to M (Act on Myocytes) and all non-cardio selective beta blockers with the exception of carvedilol begin with the letters N to Z.
So, beta-blockers lower heart rate and reduce blood pressure. So why on earth would we consider using them in sepsis, when we seem to be constantly cranking up the vasopressors to combat the associated low MAP that is desperately trying to drain all of our vital organs of any decent blood flow?!?
The Trials!
Let’s start with a summary of the Morelli et al trial. TBL gave a great breakdown here:
In 2013, a group in Italy, led by Morelli, carried out a trial in patients with septic shock who were tachycardic and receiving high doses of noradrenaline. Half of the patients were randomly allocated to receive an infusion of esmolol, a cardioselective beta-blocker. The aim was to reduce, and then maintain the patients’ heart rates at 80-94 bpm. The study showed an increase in stroke volume, maintained MAP and reduced norepinephrine requirements, without increasing the need for inotropic support or causing adverse effects on organ function. This group of patients is known to carry a high mortality rate of over 60%, but in this study, there was an associated improvement in 28-day survival. In summary, they found that not only was beta-blockade associated with improvements in survival, as well as decreased length of atsy on ITU, it also seemed safe in this setting.
As we know, septic shock results in an overproduction of nitric oxide (NO) which leads to a reduction in vasomotor tone and hypo-reactivity to both endogenous and pharmacological catecholamines. In addition, the release of molecules such as cytokines all contributes to the development of sepsis-induced myocardial dysfunction.
In general, the first line of treatment has always been a combination of alpha and beta-agonist, such as noradrenaline and dobutamine. However, following on from the Morelli trial, more interest has been focused on the introduction of a beta-blocker.
In 2015, The Journal of Anaesthesiology Clinical Pharmacology released a systematic review of the use of beta-blockers in sepsis. 9 studies were included, and they revealed some results which helped to confirm what Morelli had found.
They all found the following:
- Beta-blockers led to an improvement in heart rate, with no detrimental effect on the MAP. We know that tachycardia has an increase in cardiac events in critically ill patients and that a beta-blocker can reduce this. But how? It is thought that the decrease in heart rate, allows the diastolic relaxation time to increase and therefore the diastolic function to improve. This results in an increase in coronary perfusion and therefore less cardiac ischaemia.
- There was an increase in stroke volume or stroke volume index but a decrease in cardiac output. This is likely due to the reduction in heart rate, allowing for greater diastolic filling and therefore an increase in stroke volume. The control of the heart rate could also improve the myocardial contractility which would increase stroke volume. We should be aware however, that the cardiac output may drop with a reduction in heart rate and this is because in sepsis the CO is predominantly dependent on HR.
- A reduction in lactate. This could be due to an increase in microvascular flow, improving cellular oxygen delivery and therefore a reduction in anaerobic metabolism and hyperlactatemia.
- Another point which was raised in this review was regarding metabolic dysfunction. Beta-blockers are known to lower gluconeogenesis, hyperglycemia, proteolysis and resting energy expenditure. We know that in burns, beta-blockers reduce muscle catabolism and therefore we could perhaps assume that they have a similar effect in sepsis. However, one study showed that there was no change in muscle catabolism and therefore it is difficult to draw a conclusion on the exact effect of beta-blockers in modulating septic metabolic dysfunction.
In most of the studies, beta-blockers were started after 24 hours on ITU, because it was felt that the initial compensatory response of tachycardia and increase in systemic vascular resistance, was required in the acute phase. However, another study has shown the benefits that beta-blockers can have in those patients who use them on a long-term basis.
A study carried out by Fuchs et al in 2017, showed an independent association between the discontinuation of beta-blockers and 90-day mortality in a cohort of 296 patients with sepsis or septic shock. The study found that if beta-blockers were stopped in the acute phase of the illness, in patients who usually use them, their risk of dying increased substantially.
This data confirmed what the BASEL-II-ICU had already found, which is that there seems to be some kind of protective effect in continuing pre-existing beta-blockers on both short- and long-term mortality. The study found that in patients admitted for acute respiratory failure in whom the pre-existing using of beta-blockers was continued at discharge, the 1-year mortality was 16% vs 46% for patients discharged from the hospital without a beta-blocker.
But how?!
Unfortunately, the studies have not been able to conclude the exact mechanisms behind this protection and can only display an association. Some ideas surrounding this association include:
- The discontinuation of beta-blockers could be a marker of severity of illness and therefore these patients already had a higher mortality risk.
- Another option is that beta-blockers protect against ischaemic heart disease and therefore, you reduce the mortality risk by preventing IHD in septic patients.
- High levels of catecholamines, could lead to sympathetic overstimulation in sepsis, leading to persistent tachycardia, despite adequate fluid resuscitation. This, in turn, can lead to diastolic dysfunction and sepsis-related myocardial dysfunction, which would usually be an indication for a beta-blocker.
- In a high catecholamine state, there is inflammation, oxidative stress and abnormal intracellular calcium trafficking seen within cardiac myocytes. This leads to stunning, apoptosis and even necrosis. By increasing the stroke volume (by reducing the heart rate), we can perhaps prevent the catecholamine-induced toxicity on the cardiac myocytes.
- Finally, the results by Fuchs and colleagues potentially highlight the benefits of preventing beta-blocker withdrawal syndrome. This occurs when there is a mild and transient hypersensitivity of cardiac beta-adrenergic receptors. It presents with tachycardia, sweating, tremors, headaches and angina pectoris. We know that acute interruption in the use of beta-blockers, in conditions such as heart failure, is associated with an attributable risk of death and so perhaps this explains an element of the protection we see in this study.
So, beta-blockers for all then??
Using the evidence from the Morelli trial, and further work that has been carried out since then, it seems that beta-blockers are the way forward. The evidence seems to suggest that in reducing the heart rate, they improve diastolic function, increase stroke volume and reduce cardiac ischaemia. Not to mention the potential protective effect they have on patient’s who use them long term and the possibility of them interrupting the catecholamine-induced cardiomyocyte toxicity.
All sounds perfect?? Surely it’s a no-brainer. We should just get cracking and prescribe beta-blockers to septic patients? Not quite…
There are some issues with blanket adding beta-blockers into sepsis care bundles:
Asthmatic patients
Patients with asthma or obstructive airways diseases are at risk of beta-blocker induced bronchospasm. Pulmccm discusses the newer data emerging regarding the cardioselective beta-blockers suggests we may be safe to use these under these circumstances, well certainly in COPD.
- A modicum of caution is still needed though, these trials have confounders, for example, the effect of recent steroid courses decreased adverse events.
True hypovolaemia
This is a controversial one in sepsis, as we tend to obsess they are all volume-depleted when they are certainly not. Sepsis is a distributive shock state requiring vasopressors after initial fluid boluses (not in excess of 2000ml). Nonetheless…if they are truly inadequately filled, beta-blockers can exacerbate the associated hypotension. This is common sense!
Alpha agonism
There is also a risk of unopposed alpha agonism when using beta-blockers and catecholamines. This could be catastrophic for anyone with an undiagnosed phaeochromocytoma.
Morelli’s paper also raised concerns!
Most of the concerns fielded online are to do with the original Morelli single-centre trial:
- It was a single centre, non-blinded trial with no placebo control arm.
- The primary outcome is not clinically relevant and a mere target. The important secondary outcome of mortality (the gold standard), was not powered for.
- Many of the secondary outcomes would have been very useful as the primary(s). Thus, further studies should probably be performed before we go using esmolol safely within our routine practice.
- The usual care arm rings EGDT River’s style bells! This was significantly different from standard UK practice (PA catheters, the ubiquitous use of steroids, fluid resuscitation for 96 hours guided by filling pressures, high use of levosimendan 49.4% in esmolol group). Therefore, are these results generalisable?
- The patients in this study were extremely unwell according to the physiological scoring. They had a high prevalence of positive blood cultures and when entered, many were on quite sporting doses of norad.
- 28-day mortality of 80% in the control group is concerning and raises substantial questions of whether similar results would be seen in patients with lower severity of illness.
- SAPS II scoring was done at a different time to fit its purpose, which leads to reproducibility difficulty when applied to most of our ICU’s, where we would calculate the worst scores during the first 24h of care.
Continuation is the key…not starting new beta blockers!
The other issue – many of the papers published discussed the benefit of continuing beta-blockers in septic patients. What they did not focus upon was starting new prescriptions of beta-blockers in patients not already taking them. This is mentioned in Fuchs et al’s paper above and also in Christensen et al’s below:
Was the mortality profile of the test groups in these papers improved because it is quite obviously perhaps safer to continue beta-blockers in those who require them in the first place? After all, stopping them could lead to the recurrence of their pre-blockade pathologies. So, this group were therefore not disadvantaged?
What none of these papers has focussed on is previously young/fit cohorts of patients and the effect beta-blockade would have on their outcomes in sepsis.
The new kid on the block…Landiolol!
This leads us nicely on to the STRESS-L trial.
This study is currently in the recruitment phase over 41 sites. The aim is to complete a Phase IIb trial, in which 340 participants will be randomised to either the usual sepsis care bundle or to usual care and landiolol.
- The primary outcome is to collect data on the SOFA score up to 14 days.
- The secondary outcomes include
- mortality at 28 and 90 days
- length of stay
- reduction in duration of vasopressor treatment
- More detail on MOA of beta blockers in sepsis
The data that will hopefully be collected from this trial will help to make the picture on beta blockers in sepsis clearer – are they beneficial, or not?
You can download the ppt. presentation here:
The other aim is to extend the study performed by Morelli by taking blood samples to measure the effects of the drug on the patient’s immune system, metabolism and heart function. We were left completely in the dark over this the first time around.
The other side to this is to discover if the safety and efficacy Morelli described can be reproduced in a prospective randomised multicentre study, in which mechanisms of action can also be explored.
To summarise…
In summary, there is certainly an increasing amount of evidence to suggest that beta-blockade may be a good adjunct to therapy in the treatment of septic shock, however, it is relatively early days. As many have pointed out, we need further evidence from more targeted, multi-center RCT’s, before using them becomes common practice. We may discover more about their mechanisms of action within the complex immunological realms of sepsis.
A great podcast from John Myburgh is linked here, but the summation is that it would be “naïve and stupid” to think one single drug could be some kind of magic bullet for sepsis treatment. Afterall…look what happened to activated Protein-C!
Other references:
https://www.icnarc.org/Our-Audit/Audits/Cmp/Our-National-Analyses/Sepsis
https://bjanaesthesia.org/article/S0007-0912(17)53789-7/pdf
http://www.emdocs.net/beta-blockers-in-sepsis/
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4676233/
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4068212/