Journal club occurs most Wednesdays after morning handover. The link to the paper will be sent out in the trainee whatsapp group. The paper is critically appraised and presented by one of the trainees and the discussion is facilitated by a consultant.
If you are interested in exploring the ICU literature further it is worth having a look at the Critical Care Reviews website, WICS Bottom Line website and the ICU Trials App.
To aid with critical appraisal of the papers our trainees created a journal club checklist.
Alternatively you may find it useful to to use the Critical Appraisal Skills Programme 2022 checklists or the Centre for Evidence Based Medicine critical appraisal tools.
This Week’s Paper:
Previously Discussed Papers:
What do we need to think about?
- What factors in the care of stroke patients affect functional outcome?
- Could patient centred outcomes have been included? (e.g. comfort)
- What do we think about the two groupings within the modified Rankin scale for the primary outcome?
Did this change our practice?
No. We will continue to assess our patients on an individual basis and favour late tracheostomy insertion.
Further Reading:
What do we need to think about?
Did this change our practice?
Further Reading:
- Impact of obesity in critical illness
- Non-invasive respiratory support for post extubation respiratory failure
- Obesity in the critically ill: a narrative review
Angiography after Out of Hospital Cardiac Arrest without ST Elevation (TOMAHAWK)
What do we need to think about?
- How does the definition of “delayed angiography” compare with local practice?
- What other data would it have been useful to collect?
- what effect would the relatively high rate of cross-over have on the results?
Did this change our practice?
No. Some centres have a policy whereby all cardiac arrests are taken to the cath lab. This does not appear to have a benefit in terms of mortality unless clinically indicated (ST elevation, cardiogenic shock, arrhythmia etc).
Further Reading:
- NICE Acute coronary syndrome guidelines
- Systematic review: diagnosis and treatment of MI in critical care
- Intensive care network echo cases
Oxygen Saturation Targets for Critically Ill Adults Receiving Mechanical Ventilation (PILOT)
What do we need to think about?
- What are the determinants of oxygen delivery?
- What effect would the trial being performed in a single centre have?
- Why did they include an “analytic washout period”?
Did this change our practice?
No. We will continue to avoid hyperoxia and provide individualised oxygen targets for our patients depending on their physiology, co-morbidity and presenting condition.
Further Reading:
Haloperidol for the Treatment of Delirium in ICU Patients (AID-ICU)
What do we need to think about?
- Is a proposed 15% lower risk of death in the haloperidol group realistic?
- Should there be a difference in treatment strategy between the hypoactive and hyperactive delirium groups?
- What effect would the common use of benzodiazepines both pre and post randomisation have on the results?
Did this change our practice?
No. We will continue to routinely use non-pharmacological measures for preventing and treating ICU delirium. We will continue to prescribe pharmacological treatments only where agitation interferes with the ability to care for the patient or represents a risk to patients or staff.
Further Reading:
What do we need to think about?
- What effect would the sample size recalculation have had?
- What about the confounding effect of co-administration with adrenaline?
- Why was the trial stopped?
Did this change our practice?
No. We will continue to only administer calcium in cardiac arrests where it is indicated (e.g. hyperkalaemia).
Further Reading:
What do we need to think about?
- What effect would the open label design have had?
- What proportion of the control group received bicarbonate and what effect would this have had?
- What proportion of patients in the intervention group achieved the target pH? What effect would this have had?
Did this change our practice?
Yes. We would consider administration of intravenous bicarbonate to patients with metabolic acidosis and AKI.
Further Reading:
Balanced multi-electrolyte solution versus saline in critically ill adults (PLUS)
What do we need to think about?
- what effect would the COVID lockdown have had on results?
- what effect would receiving a significant volume of saline (drug diluents) have on the BMES group results?
- Should 90 day mortality have been the primary outcome?
Did this change our practice?
No. We will continue to use balanced electrolyte solutions as our solution of choice unless there are specific indications (e.g. hyponatraemia or traumatic brain injury).
Further Reading:
- Intravenous fluid in the perioperative and critical care setting
- Glycocalyx in critical illness
- NICE guidance: intravenous fluid therapy in adults in hospital
What do we need to think about?
- What effect could cluster randomisation have? Why was an inflation factor used?
- what effect would altering the threshold for prolonged grief on the PG13 score have?
- what effect would only using segments of questionnaires that have been validated for use as a whole?
Did this change our practice?
Yes. There will be cultural and communication differences between practice in the UK and France. However, this is a thought provoking paper. We can take away that families appreciate regular communication and information before, during and after their loved one has died. Could we include a psychological support plan in our end of life care plan?
Further Reading/ Watching:
- FICM end of life care guidance
- Bereavement in intensive care: a narrative review
- Complicated grief after death of a relative on ICU
- Worth a watch – Extremis on Netflix
What do we need to think about?
- What effect would the choice of induction agent have on the results?
- What is a predictive enrichment strategy?
- Would the volume of fluid given prior to RSI have an effect? Should this data have been collected?
Did this change our practice?
Yes. The majority of those involved in the discussion did not routinely give a fluid bolus at RSI unless clinically indicated (hypovolaemia etc). This has prompted us to perform a snapshot of current practice and review our RSI standard operating procedure which mandates a fluid bolus.
Further Reading:
- Life in the Fast Lane – RSI
- Prognostic and predictive enrichment in sepsis
- DAS ICU intubation guidelines
Effect of Vasopressin and Methylprednisolone vs Placebo on Return of Spontaneous Circulation in Patients With In-Hospital Cardiac Arrest. A Randomized Clinical Trial (VAM IHCA)
What do we need to think about?
- What are the logistical challenges of running this trial?
- What impact could the fact that the trial was only performed in Denmark have?
- How could post cardiac arrest care have affected the results?
Did this change our practice?
No. Although vasopressin and methylprednisolone improved rates of ROSC there was no difference in terms of 30 day mortality or neurological outcomes. It will be interesting to see if there is a difference in terms of long term outcomes.
Further Reading:
- Life in the fast lane: Vasopressin
- National Cardiac Arrest Audit Reports
- Outcomes after in hospital cardiac arrest
- Resuscitation Council UK Guidelines
What do we need to think about?
- How is iron utilisation influenced by critical illness?
- How might the dosing regime affect the outcome?
- What impact would the power calculation have on the conclusion of the study?
Did this change our practice?
No. This trial was underpowered and mainly included cardiac surgical patients. We will continue to try to prevent anaemia on ICU and, in severe anaemia with accompanying iron deficiency, supplement according to the manufacturer’s dosing regime.
Further Reading:
- Iron metabolism in critical illness
- BSH guideline anaemia and transfusion in critically ill
- Type I and II errors
What do we need to think about?
- How do differences in healthcare systems around the world affect the translation of research findings to local practice?
- How can the characteristics of patients included affect the generalisability of a study?
- Why has delta sofa been used as the primary endpoint?
Did this change our practice?
No. We will continue to not administer vitamin c and thiamine to patients presenting with septic shock.
Further Reading:
Trial of Decompressive Craniectomy for Traumatic Intracranial Hypertension (RESCUE ICP)
What do we need to think about?
- Is management of intracranial hypertension the most important factor in outcome for these patients?
- What effect would crossover among patients in the medical group have on results?
- How should we decide what is an acceptable benefit?
Did this change our practice?
Yes. This paper, combined with information from DECRA, informs discussions surrounding best interests decisions about surgical management and what would be acceptable to that individual.
Further Reading:
- What does favourable outcome mean?
- Brain Trauma Foundation TBI Guidelines
- Overview of Long Term Outcomes Post Severe TBI
- LITFL: Traumatic Brain Injury Management
What do we need to think about?
- Peripheral vs central capillary refill time
- Confidence intervals
- How can we measure the effect of complex interventions?
Did this change our practice?
No. Capillary refill time may be a useful indicator of shock resolution in resource poor countries. We will continue to use a multimodal approach including clinical assessment, serum lactate, cardiac output monitoring and POCUS.
Further Reading:
- Surviving Sepsis Guidelines
- Lactate and Lactic Acidosis
- The 10 Pitfalls of Lactate Clearance in Sepsis
What do we need to think about?
- What effect would the timing of randomisation had on the results?
- What effect would the drug being open label have had?
- Should serum levels of drugs have been measured?
Did this change our practice?
No. We will continue not to administer the combination therapy of vitamin c, hydrocortisone and thiamine to patients with septic shock.
Further Reading:
- Physiology of vitamin c in sepsis
- Corticosteroid therapy for sepsis: BMJ Rapid Recommendations
- Time to go back to fundamental principles?
What do we need to think about?
- Definition of sepsis
- Should shock resolution have been considered as an end point/ measured?
- How does the method of HDF differ to our practice?
Did this change our practice?
No. We will not be using high volume continuous haemodiafiltration for sepsis associated AKI.
Further Reading
- Renal replacement therapy in ICU
- Acute kidney injury in sepsis
- Nutrients and micronutrients at risk during renal replacement therapy
Pantoprazole in Patients at Risk for Gastrointestinal Bleeding in the ICU – SUP ICU
What do we need to think about?
- What are the harms of administering PPI’s in ICU patients?
- Is mortality a useful primary end point here?
- How should we interpret and apply secondary endpoints to clinical practice? Especially when no p values are provided.
Did this change our practice?
No. As a unit we continue PPIs for those taking them in the community. For those not taking PPIs in the community those at high risk of stress ulceration would receive a PPI until feeding is established.
Further Reading:
Treating Rhythmic and Periodic EEG Patterns in Comatose Survivors of Cardiac Arrest (TELSTAR)
What do we need to think about?
- Do these EEG patterns represent seizure activity or evidence of cerebral damage?
- What effect would the intervention being unblinded have on results?
- What effect would the variability in interpretation of CPC scores have between assessors?
Did this change our practice?
No. This trial was discussed as a matter of interest as we will be introducing cEEG for selected patients on the unit. The anticonvulsant regime used is not in line with our current practice. We will continue to treat confirmed seizure activity with anticonvulsants in line with the ERC guidelines.
Further Reading:
- Continuous EEG monitoring in ICU
- European Resuscitation Council & European Society of Intensive Care Medicine Guidelines: Post Resuscitation Care
- Reliability of Cerebral Performance Category
What do we need to think about?
- What is clinical equipoise?
- What effect could changing the primary outcome have?
- Types of bias
Did this change our practice?
Yes. We will not use tranexamic acid empirically in GI bleeding. We will also use point of care thromboelastography (ROTEM in our centre) to guide haemostasis.
Further Reading:
- Sengstaken-Blakemore, Minnesota and Linton-Nachlas tubes
- BMJ State of the Art Review: Management of GI bleeding
- Uses of Tranexamic Acid
Lower or Higher Oxygenation Targets for Acute Hypoxaemic Respiratory Failure (HOT-ICU)
What do we need to think about?
- What is heterogeneity of treatment effect?
- What are the potential harms of a lower oxygenation strategy vs a higher oxygenation strategy?
- What have the other trials preceding this concluded?
Did this change our practice?
No. We will continue to aim for a PaO2 of more than or equal to 8Kpa with the lowest possible Fio2.
Further Reading:
Early vs Late Tracheostomy for Prevention of Pneumonia in Mechanically Ventilated Adult ICU Patients
What do we need to think about?
- How do we define early vs late?
- How do we balance between benefits and the risks of a procedure?
- How did practice change during COVID when patients remained intubated for long periods than would usually be accepted?
Did this change our practice?
No. We remained in favour of a later insertion of tracheostomy preferably after a trial of extubation where safe.
Further Reading:
- Cardiff Tracheostomy LOCSSIP
- National Tracheostomy Safety Project Resources
- Percutaneous vs Surgical Tracheostomy