Journal Club Podcast May 2022

Dr Bertha Wu
Dr David McCreary
Prof Peter Cameron

Welcome to the May Journal Club Podcast. We are joined by Professor Peter Cameron, Academic Director for the Alfred Emergency and Trauma Centre, and Emergency Physician Dr David McCreary.

In May’s journal club, we covered papers considering vasopressor choice in post OHCA patients in shock, trainee-supervisor power and trust dynamics in context of WBAs, front of neck access techniques, as well as effectiveness of sotrovimab in treating patients with risk of progressing to severe COVID-19.

Paper 1: Epinephrine versus norepinephrine in cardiac arrest patients with post-resuscitation shock

Read it here
Clinical Question

Does epinephrine cause more harm than norepinephrine in treating cardiac arrest patients with post resuscitation shock?

 

Why is this topic important?
  • Poor survival after out of hospital cardiac arrest (10%)
  • Only 1/3 admitted to hospital alive, of which, half to 2/3 die during ICU admission due to neurological injury or haemodynamic failure (refractory shock or recurrent cardiac arrest)
  • Theoretical risk of harm for adrenaline use: increased cardiac contractility + heart rate à increased myocardial O2 demand à theoretically could exacerbate myocardial damage
  • Few head-to-head studies comparing vasopressor use in OHCA. There had been some studies comparing epinephrine and norepinephrine in shock due to sepsis. A recent RCT looked at mortality between the epinephrine and norepinephrine arms in cardiogenic shock post MI, but the study was underpowered.
  • No study has compared epinephrine and norepinephrine use in patients with post resuscitation shock
Study Design & PICO
Design

Registry-based, multicentre observational study

Population

Inclusion:

  • Data of patients included in study was pulled from the Sudden Death Expertise Centre Registry in the Paris metropolitan area
  • Patients admitted to five university hospitals alive and managed for post-resuscitation shock after OHCA from 2011-2018 were included
  • Post resuscitation shock was defined as a need for vasopressors for more than 6hrs despite adequate fluid loading. The target mean was 65mmHg.

Exclusion:

  • Obvious extra cardiac cause of cardiac arrest eg trauma, drowning, drug overdose, electrocution, asphyxia due to external cause
  • Refractory cardiac arrest without sustainable ROSC
  • Refractory shock requiring extracorporeal membrane oxygenation
  • Absence of continuous intravenous treatment with epinephrine or norepinephrine

Pts who had continuous intravenous treatment with both epinephrine and norepinephrine were initially excluded, but later on included in additional analysis.

Intervention

Epinephrine infusion during ICU stay

Comparison

Norepinephrine infusion during ICU stay

Outcomes

Primary outcome:

  1. All-cause hospital mortality during hospital stay

Secondary outcomes:

  • Cardiovascular hospital mortality (recurrent cardiac arrest or refractory haemodynamic shock)
  • Unfavorable neurological outcome (Cerebral Performance Category 3-5)

 

Findings

766 patients from 5 hospitals were included in study à 285 (37%) received epinephrine, 481 (63%) received norepinephrine

Note, Table 1 in paper reported many difference in baseline features in the two groups. Patients in the epinephrine arm vs those in the norepinephrine arm had:

  • less patients with a shockable rhythm – 124 (44%) vs 276 (57%)
  • a lower median pH at admission was lower – pH 7.17 vs pH 7.23
  • a higher arterial lactate at admission – 7.6 vs 4.8
  • higher inotropic requirements – inotropic equivalent (IE) of 68 vs 49 (IE in ug/kg/min = dopamine + dobutamine + 100 x epinephrine + 100 x norepinephrine + 100 x isoproterenol = 15 x milrinone)
  • more patients with myocardial dysfunction – 160 (72%) vs 229 (57%)
  • less patients who underwent a coronary angiogram – 186 (66%) vs 409 (85%)
  • less patients with targeted temperature management – 127(47%) vs 406 (85%)
  • more patients with occurrence of OHCA at home – 171 (60%) vs 229 (48%) – but this was not statistically significant

The study found that:

  • All cause mortality was significantly higher in the epinephrine group (OR 2,6, 95% CI 1.4-4.7, P = 0.002)
  • Cardiovascular hospital mortality was also higher with epinephrine (aOR 5.5, 95% CI 3.0-10.3, P < 0.001), as was the proportion of patients with CPC of 3-5 at hospital discharge
  • Sensitivity analyses produced consistent results
  • The analysis involving adjustment on a propensity score to control for confounder showed similar findings (aOR 2.1, 95% CI 1.1-4.0, P = 0.02)
Authors’ conclusions

Among patients with post resuscitation shock after out of hospital cardiac arrest, use of epinephrine was associated with higher all cause and cardiovascular specific mortality compared with norepinephrine infusion

Journal Club thoughts

The study question is an important one, but not an easy one to perform prospectively. Not unexpectedly, this retrospective registry-based study had some major issues with the study design. Importantly, there were many baseline differences between the two study arms. Essentially, patients in the adrenaline group sound much sicker to being with. The authors tried hard to compensate for confounders and differences between the groups by using various statistical maneuvers. However, you just can’t compare apples with oranges!

In clinical practice, clinicians’ choice of vasopressors in patients with post-resuscitation shock is largely based on clinical judgement. Whether to use adrenaline vs noradrenaline in post-resuscitation shock is generally based on the cause of the patient’s post-ROSC hypotension: cardiogenic vs inflammatory (distributive). In general, for the former, adrenaline tends to be used, and for the latter, noradrenaline would be the agent of choice.

There were also potential flaws in the study’s statistical analysis. Whist performing multivariate logistic regression to adjust for confounders, independent variables should be used. Variables looked at in study (eg initial rhythm, time from collapse to CPR, time from CPR to ROSC, arterial pH, myocardial dysfunction) were dependent on each other (multicollinearity). So to account for codependence, conditional logistic regression has to be used. When the authors performed conditional logistic regression, no statistical difference was found between the epinephrine and norepinephrine arm. However, authors went on to perform propensity analysis on the conditional logistic regression – essentially performing statical analysis to adjust for confounders twice, in effect canceling each other out.

Bottom Line

This is not a practice changing study, although it may be taken into account during out decision making when deciding whether to use adrenaline vs noradrenaline in patients with persistent shock post arrest (Ed. – unless you are of the same opinion as Dr McCreary, who feels this paper in no way can alter our clinical practice). It’s a hypothesis generating study – an RCT is needed to answer this question.

Paper 2: Trust, power and learning in workplace-based assessment: The trainee perspective

Read it here
Clinical Question:

How do trust and power dynamics between trainee and supervisor influence WBAs?

Summary

This is a qualitative research paper with a clinical question of – how do trust and power dynamics between trainee and supervisor influence WBAs?

Semi-structured interviews were performed on 17 anaesthetics trainees from Australia and New Zealand. The interviewers were the first author who was consultant anaesthetists (also Supervisor of Training and also supervises WBS), and a nonclinician clinical research assistant.

It used a qualitative research method called the constructivist grounded theory (CGT) – a method that uses an inductive approach to generate a new theory from data gathered through participant interviews or focus groups. The aim of CGT is to understand and explore a social process where no adequate prior theory exists.

As a results of the interviews, the authors theorized that WBA supervisor-trainee power dynamic is underpinned by an interaction of supervisor power over and trainee power to.

Supervisor power over is the result of an expected natural order of a hierarchical environment. There is a reciprocal dominance and subservience in the supervisor/trainee relationship due to the power gradient.

  • An example of this is trainees calling their supervisors “bosses”
  • How supervisors use their power and how trainees respond to this can produce a different outcome for trainees
  • Supervisor power over can provide trainees with multiple benefits:
    • Provide norms for trainee and supervisor as to how to behave and relate to each other
    • Clarified responsibility for patient care
    • Interpersonal conflict avoided and facilitates a cordial and efficient working relationship
    • Facilitate trainee access to specialist expertise and support in their learning
  • Supervisors can influence trainee’s reputation, their progression in training and their access to future employment, for better or for worse

Trainee power to reflects the trainees agency and how they choose to exercise their trust on supervisors.

  • Trainees are more likely to initiate WBAs when
    • they perceive that the encounter could be in their benefit
    • supervisor demonstrate benevolent use of their power over eg had shown committment to trainees’ learning
  • Trainees are reluctant to expose their own practise to supervisor scrutiny when they felt supervisors demand close control over patient care. Trainees minimise their vulnerability by:
    • Withdrawing trust
    • Complying closely to the supervisor’s wishes
    • Matching their practise to the supervisor’s, ie “gaming” or “staging” their performance
  • Trainees choose when to initiate formal assessments eg they avoid formal assessments with supervisors who they know to be unhelpfully critical
Authors’ conclusions

The authors concluded that trainees’ trust in their supervisors was expressed in their use of their power to depending on how they perceived supervisors used their power over in WBAs:

  • when trainees choose to trust in response to benevolent use of supervisor power, they are more likely to practise authentically and expose their weakness for the purpose of learning
  • when trainees choose to trust less, eg with more “controlling” supervisors, they forgo their learning opportunities, and are likely to “game” or “stage their performance to match their practise to how they think they are expected to perform,
  • with unsupportive and overly critical supervisors, trainees choose to avoid doing WBAs with them altogether

The authors suggested a few ways to refine assessment systems to. rebalance the power dynamic between trainees and supervisors, with the aim to enhance trainee trust and mitigate trainee vulnerability

  • transparency in what performance information is considered in decision making
  • Prospectively designating WBAs as either evidence for decision making or learning events to separate assessment of learning and assessment for learning
  • Granting trainees some control over the selection of performance information presented to inform progress decisions
Journal Club thoughts

It’s a good to see clinical educators doing research on an area that is relatively new – WBAs have only been introduced quite recently. However, the study participants are anaesthetic trainees – supervision levels in anaesthetic training is very different to emergency training. Emergency trainees have to be independently practising at a much earlier stage of training. The interviewer in the study is also an anaesthetic consultant who is a supervisor of WBAs for anaesthetic trainees – it’s hard to imagine the conflict of interest did not affect the participants’ answers in some way.

WBAs are somewhat tick boxes that ED trainees have to fulfil in order to complete training, but the stakes get higher as they approach the end of their training. One suggestion in our Journal Club is to focus on doing WBAs earlier in Advanced Training so the stakes are not so high – this way there is a better chance of trainee performing authentically for the purpose of learning. Another suggestion is to prospectively allocate WBAs so the trainee won’t get to “game” or stage” the performance as much.

Bottom Line

The process of assessment via WBAs is not perfect. But at the end of the day, trainees can always get supervisors to informally assess their performance, independent to WBAs, in order to learn.

Paper 3: Success and Time to Oxygen Delivery for Scalpel-Finger-Cannula and Scalpel-Finger-Bougie Front-of-Neck Access: A Randomized Crossover Study With a Simulated “Can’t Intubate, Can’t Oxygenate” Scenario in a Manikin Model With Impalpable Neck Anatomy

Read it here

Clinical Question

In “can’t intubate, can’t oxygenate” scenarios, is the scalpel-finger-cannula technique superior to the scalpel-finger-bougie technique in providing timely oxygenation?

Study Design & PICO
Design

Randomized prospective observational study conducted in the Department of Anaesthesiology, Intensive Care and Pain Medicine at Tan Tock Sen Hospital Singapore

  • 65 Attending Anaesthetists or senior residents were invited to participate in study
  • They attended scheduled basic FONA training, followed by advanced training on SFC and SFB cricothyrotomy
  • On the same day, they performed both cricothyrotomy techniques under simulated conditions
  • The order in which they perform either of the cricothyrotomy techniques were randomized
Population
  • Interventions were performed on manikins in simulation scenarios
  • Manikins simulated an obese patient with non palpable anatomy – trachea was 8cm deep, with lots of fake blood involved to simulate bleeding
Intervention

Using the scalpel-finger-cannula (SFC) technique for front of neck access (FONA) followed by performing a Melker cricothyrotomy

Comparison

Using the scalpel-finger-bougie (SFB) technique for FONA

Outcomes

Primary:

  • Time to in seconds from CICO declaration to oxygen delivery (defined as observation of manikin lung inflation), and
  • First attempt success
    • Each attempt was defined as the removal and reinsertion of the cannula or bougie
    • Successful attempt was defined as oxygen delivery within 180 seconds of CICO declaration and within 3 or fewer attempts
  • For SFC, the time for insertion of a Melker cricothyrotomy tube was recorded from the participant picking up the guidewire to successful oxygen delivery
    • Deemed successful if the combined time for successful cannula and Melker cricothyrotomy was within 300 sec and within 3 or fewer attempts

Secondary:

  • Overall success
  • Incidence of puncture through cricothyroid membrane
  • Posterior tracheal wall damage
  • Preferred FONA technique
  • Pre-simulation and post-simulation confidence in SFC and SFC cricothyrotomies
  • Satisfaction with the simulation model

 

Findings
  • SFC was associated with a short time to oxygen delivery, with mean time difference -62.1s in multivariable analysis.
  • Higher first attempt success was reported with SFC than SFB – 72.3% vs 27.7%
  • Participants had higher odds at achieving first-attempt success with SFC bs SFB (OR 10.7, 95% CI 3.3-35.0, P <0.001)
  • Successful delivery of oxygen after the CICO declaration within 3 attempts and 180sec was higher (84.6% vs 63.1%) and more likely with SFC (OR 5.59, 95% CI 1.7-18.9, P = 0.006)
  • Analysing successful cases only, SFC achieved a shorter time to oxygen delivery (mean time difference -24.9s, 95% CI -37.8 to -12s, P < 0.001), but a longer time to cuffed tube I nsertion (mean time difference +56.0s, 95% CI 39.0-73.0, P < 0.001)
  • After simulation training, most participants preferred SFC in patients with impalpable neck anatomy (75.3% vs 24.6%)
Authors’ conclusion

In manikin simulation of impalpable neck anatomy and bleeding, the SFC approach demonstrated superior performance in oxygen delivery and was also the preferred technique of the majority of study participants. The study findings support the use of a cannula-based front of neck access technique for achieving oxygenation in a CICO situation, with the prerequisite that appropriate training and equipment are available.

Journal Club thoughts

It’s a good attempt in looking at an area that is very difficult to study. CICO situations don’t happen often (and I’d rather they never happen in my clinical practice!). However, this study is limited by being a manikin study. It’s also performed by experienced anaesthetists and anaesthetic registrars who have a different airway skillset to emergency physicians and trainees. The low pressure, rapid O2 delivery system is not readily available. In Australia, a high-pressure jet ventilation device like Manujet is available in most hospitals, but has a high incidence of complications eg device failure and barotrauma. With the scenario used in the study – an obese patient with a bleeding airway – how well can one oxygenate the patient with a small gauge cannula? Establishing a definitive airway for adequate oxygenation will be the priority in these patients. With the SFC technique, having steps to insert the Melker cricothyrotomy tube means extra time before a definitive airway is established. It also means extra cognitive load in a high-pressure high-stake situation.

Bottom Line

In our patients (ED patients requiring emergency RSI), time to definitive airway using easy to access equipment is our priority. This study won’t change the most common practice in ED, which is to use the scalpel-finger-bougie technique.

 

Paper 4: Effect of Sotrovimab on Hospitalization or Death Among High-risk Patients With Mild to Moderate COVID-19

Read it here

Clinical Question

Among patients at risk of disease progression, does early treatment of mild to moderate COVID-19 with sotrovimab prevent progression to severe disease?

Study Design & PICO
Design

Phase 3, double-blind, multicentre RCT conducted at 57 sites in Brazil, Canada, Peru, Spain and the US from 2020-2021

A part of a larger trial to evaluate the efficacy and tolerability of – COVID-19 Monoclonal Antibody Efficacy Trial – Intent to Care Early (COMET ICE)

Population

Non-hospitalized patients with symptomatic, mild to moderate COVID-19 and at least 1 risk factor for progression

Risk factors:

  • Age >/= 55yo
  • Diabetes requiring medication
  • Obesity (BMI >30)
  • CKD (eGFR <60)
  • CCF (>/= NYHA class II)
  • COPD
  • Moderate to severe asthma

Excluded:

  • Hospitalized patients
  • Signs or symptoms of severe COVID-19 (SOB at rest, SpO2 <94%, required supplemental O2)
Intervention

Single intravenous infusion with 500mg of sotrovimab over 1 hour on day 1 (n=528)

Comparison

IV infusion with equal volume of placebo over 1 hour on day 1 (n=529)

Outcomes

Primary:

Proportion of patients with COVID-19 progression through day 29 (all cause hospitalization lasting >24hrs for acute illness management, or death)

 

Secondary:

  • Composite of all-cause emergency department visit
  • Hospitalization of any duration for acute illness management
  • Death through day 29
  • Progression to severe or critical respiratory COVID-19 requiring supplemental O2 or mechanical ventilation

 

Findings

Enrolment was stopped early for efficacy at the prespecified interim analysis.

1057 non-hospitalized patients with symptomatic mild to moderate COVID 19 and at least 1 risk factor for progression were enrolled.

  • All cause hospitalization lasting longer than 24hrs or death was significantly reduced with sotrovimab – 6/258 (1%) vs placebo 30/529 (6%); adjusted RR 0.21 (95% CI 0.09 – 0.50); absolute difference -4.53% (95% CI, -6.00% to -2.37%), P < 0.001
  • Four of the 5 secondary outcomes were statistically significant in favor of sotrovimab, including
    • reduced ED visit, hospitalization or death – 13/528 [2%] for sotrovimab vs 39/529 [7%] for placebo; adjusted RR, 0.34 [95%CI, 0.19 to 0.63]; absolute difference, -4.91%[95%CI, –7.50% to –2.32%]; P < .001)
    • progression to severe or critical respiratory COVID-19 – 7/528 [1%] for sotrovimab vs 28/529 [5%] for placebo; adjusted RR, 0.26 [95%CI, 0.12 to 0.59]; absolute difference, –3.97%[95%CI, –6.11% to –1.82%]; P = .002)
      • no patients treated with sotrovimab required HFNP, O2 via NBM or mechanical ventilation through day 29
    • Adverse events were infrequent and similar between treatment groups (22%for
    • sotrovimab vs 23%for placebo)
Authors’ conclusion

Among non-hospitalized patients with mild to moderate COVID-19 and at risk of disease progression, a single intravenous dose of sotrovimab, compared with placebo, significantly reduced the risk of a composite end point of all-cause hospitalization or death through day 29.

Journal Club thoughts

This is somewhat a historical study now. The study enrolled patients over a finite period of the pandemic before vaccines were available and when the delta COVID strain was still at large. Now, a majority of people are vaccinated and omicron variants are the most common strains of COVID in the community. We know now that sotrovimab is less effective against omicron and variants so this study is somewhat not as relevant now. Additionally, there is a shortage of supply for sotrovimab. Current guidelines recommends using other antiviral agents like paxlovid and molnupiravir first.

Bottom Line

Though not as relevant now, it’s still a good study to do as what we learn from sotrovimab can inform future efforts looking into therapies for other strains of COVID.

References
  1. Bougouin W, Silmanu K, Renaudier M et al. Epinephrine versus norepinephrine in cardiac arrest patients with post-resuscitation shock. Intensive Care Med. 2022 Mar;48(3):300-310.
  2. Castanelli DJ, Weller JM, Moloy E, Bearman M. Trust, power and learning in workplace-based assessment: The trainee perspective. Med Educ 2022 Mar;56(3):280-291.
  3. Zhang J, Ong S, Toh H, et al. Success and Time to Oxygen Delivery for Scalpel-Finger-Cannula and Scalpel-Finger-Bougie Front-of-Neck Access: A Randomized Crossover Study With a Simulated “Can’t Intubate, Can’t Oxygenate” Scenario in a Manikin Model With Impalpable Neck Anatomy. Anesth Analg 2022 Mar 3.
  4. Gupta A, Gonzalez-Rojas Y, Juarez Erick, et al. Effect of Sotrovimab on Hospitalization or Death Among High-risk Patients With Mild to Moderate COVID-19A Randomized Clinical Trial.  JAMA.2022;327(13):1236-1246.
Dr Bertha Wu

Dr Bertha Wu

Emergency Registrar

MBBS, CCPU (eFAST, AAA, BELS). Emergency Medicine Advanced Trainee and Intensive Care Medicine Trainee in Melbourne, Australia. Particular interests in POCUS, medical education and health care in resource-poor settings. Twitter: @berthawu29