COVID-19 Journal Club: The Director’s Cut #2

It’s tough keeping up to date with the ever changing world of Covid19. Whilst we are all working hard at treating patients in the ED and ICU, we also need to try and keep up to date with new evidence coming out all the time. Our aim is to help you hold back the hoardes of papers that threaten your doorstep, and only allowing those worthy of your attention through. If you are in a rush RCEM’s Top 5 is the bitesize snack you need, if you want an interactive sit down with the top Profs the webinar is where you should be, and if you have head space for 10 more impressive papers read on!

Every week we are running the journal club webinar, this week’s cast not ONLY includes four Professors but also a guest appearance from the international FOAMeD community.

You can listen to the podcast on the link below. We only cover 6 papers on the podcast, which leaves plenty for the director’s cut.

The team has been working hard to deliver this and a massive thanks to Dr Govind Oliver, Dr Anisa Jafar, Dr Mina Naguib, Rick and Simon! We are excited to have a guest contributor Dr McCabe from Hereford ED!

As always if you have seen an interesting paper please submit it here!

Highlights

• Gattinoni et al described two phenotypes of pulmonary COVID-19
• Giannakeas et al predicted healthcare resources use with a new model
• Henry et al reminded us to watch out for increased serum ferritin
• Lui et al adds to the emerging case for neutrophil/lymphocyte ratios
• Garcia et al hinted that antibody tests are best suited to late presenters
• Gudbjartsson et al shows off Iceland’s detailed pandemic data
• Zhang et al signal that ACEi/ARB might be protective
• Smith et al gave USS a lead in the running debate for its use
• Tang et al downgraded the hope for HCQ treatment

1. COVID-19 pneumonia: different respiratory treatments for different phenotypes? by Gattinoni et al

• Topic: Pathophysiology
• Rating: Head Turner
• Scout: Professor Simon Carley

This paper is an observational piece that has gained a lot of interest amongst the ICU community. Their observation is that there are two sorts of lung injury in Covid-19 patients. The L-type is described as low elastance, low ventilation to perfusion, low lung weight and low lung rectruitability. The H-type is characterised by high elastance, a high right to left shunt, high lung weight and high lung recruitability. This matters because the supportive therapy that we give to severely affected patients, i.e. ventilatory support clearly will differ between these two phenotypes. We have certainly seen very different clinical presentions with perhaps the ‘Happy Hypoxic’ patient who has little distress and a rapid improvement on oxygen resembling the L-type whereas the patient requiring urgent intubation perhaps resembling the H-type. More work is needed on this, but it is clear that Covid-19 expresses itself in very different ways.

2. Estimating the maximum daily number of incident COVID-19 cases manageable by a healthcare system by Giannakeas et al

• Topic: Epidemiology
• Rating: Worth a peek
• Scout: Dr Anisa Jafar

Mixed feelings about this one. Pre-print from Canadian team who have developed a program for modelling the capacity a given healthcare system has to manage new daily COVID-19 cases.  The model tells you the max. manageable patients “in” based on the turnover of acute/crit care beds and even ventilator requirements. It does all of its clever calculations based on published data.  Some of these assumptions include an average of 11 day in acute care, 20 days in crit. care and average mechanical ventilation of 20 days. So far so good? Population-weighted expected probabilities of COVID-19 acute care hospitalisation/critical care admissions are thrown in as well and an assumption that 50% of crit. care patients require mechanical ventilation.  At this point you may start to wonder how modifiable the model is, assuming local numbers, proportions and lengths of stay are very different.  The authors do tell us that the tool allows for customised age-based case distribution and severity, which is a big tick.  However the tool works on the assumption that the cases are immediately hospitalised and some of the default parameters are based on some specific Canadian, US and Chinese data.  They admit it is a work in progress and will continue to update it.  Take home message: great idea BUT must be used with caution and really by people who understand modelling and numbers well enough to avoid the “rubbish in, rubbish out” phenomenon which will occur with blind usage.

3. Hematologic, biochemical and immune biomarker abnormalities associated with severe illness and mortality in coronavirus disease 2019 (COVID-19): a meta-analysis by Henry et al

• Topic: Prognosis
• Rating: Worth a peek
• Scout: Dr Govind Oliver

This meta-analysis of 3,377 patients and 33 laboratory parameters disappointingly has few hard take homes for Emergency Physicians to adjust their decision-making. Henry et al. found that patients with severe and fatal COVID-19 had significantly increased white blood cell, lymphocyte and platelet counts whilst IL-6, IL-10 and serum ferritin were markers for potential progression to critical illness. There are methodological weaknesses in this study. In particular, the differing strategies reported in treating this novel infection, causes the study specific clinical definition of “severe” (a composite of critical care/ ventilatory / organ support) to have inherent heterogeneity and ambiguity. All but one of the 21 included studies comes from China so there are also questions in the transferability of findings to suspected COVID-19 population being seen in EDs. These reported associations are worth being aware of. 

4. Neutrophil-to-lymphocyte ratio as an independent risk factor for mortality in hospitalized patients with COVID-19 by Lui et al

• Topic: Prognosis
• Rating: Worth a peek
• Scout: Dr McCabe Hereford County Hospital 

One of the first blood tests returned to Emergency Physicians is the FBC. The potential of the neutrophil to lymphocyte ratio (NLR) as a prognostic feature has been considered before. This single centre retrospective cohort study examined 245 COVID-19 patients from Wuhan, China. Through an adjusted analysis they demonstrated that NLR is an independent risk factor for in-hospital mortality for COVID-19. This meets face validity matching what we have seen in the ED, and also the published reports of lymphopenia. Good things; they used an adjusted analysis and kept the NLR continuous! Bad things; the adjusted model used a supplanted method for variable selection, and it is a small retrospective sample. NLR should be treated as a potential prognostic marker, and will require external prospective validation prior to clinical use.

5. Rapid diagnosis of SARS-CoV-2 infection by detecting IgG and IgM antibodies with an immunochromatographic device: a prospective single-center study by Garcia et al

• Topic: Diagnosis
• Rating: Worth a peek
• Scout: Professor Richard Body

We hear a lot about antibody (IgM/IgG) tests for SARS-CoV-2 but we don’t yet really know how they might be used in practice. In this study, the authors evaluated IgM/IgG tests manufactured by AllTest. The tests are immunochromatographic, meaning that you read a positive or negative result from a strip (a qualitative test). They evaluated this in three groups: (1) healthy controls; (2) PCR-positive COVID-19; and (3) pneumonia of unknown aetiology but PCR-negative for SARS-CoV-2. All of the healthy controls tested negative (implying 100% specificity). In group 2, sensitivity for COVID-19 was only 47%. However, in patients presenting at least 14 days after symptom onset, sensitivity was 74%, implying that this may be helpful for diagnosis in late presenters. In group 3, 89% of patients tested positive (including 91% of patients presenting at least 14 days after symptom onset). The results suggest that antibody testing could be useful for diagnosis of COVID-19 in late presenters (when PCR tests from nose/throat swabs may be negative). However, these are early results – more definitive evidence is still needed.

6. Spread of SARS-CoV-2 in the Icelandic Population by Gudbjartsson et al

• Topic: Epidemiology
• Rating: Worth a peek
• Scout: Dr Govind Oliver

Iceland; whilst geographically small, it shines as a beacon of excellence in pandemic surveillance. They have remarkably robust, detailed, and inclusive pandemic data. Gudbjartsson et al. found that 13.3% of people undergoing targeted testing were positive for COVID-19 compared to only 0.8% and 0.6% from open-invitation and random population screening in Iceland. In the early phase of spread (prior to government implementation of restrictions and social distancing), 65% of people testing positive had travelled outside Iceland (86.1% to designated high risk areas) and 40.1% reported contact with a known infected person. This compared to only 15.5% of people testing positive in the late phase having travelled outside Iceland and 60.2% having contact with a known infected person. Interestingly, 29% of those testing negative in the population screening had reported some symptoms, whilst 43% testing positive had no symptoms. Overall children and women had lower incidence of COVID-19 and contact tracing showed spread through family, social, work, travel and unknown sources. Slow spread during the late phase supports the role of government interventions in slowing spread.

7. Association of Inpatient Use of Angiotensin Converting Enzyme Inhibitors and Angiotensin II Receptor Blockers with Mortality Among Patients With Hypertension Hospitalized With COVID-19 by Zhang et al

• Topic: Prognosis
• Rating: Worth a peek
• Scout: Dr Mina Naguib

Should we continue a patient’s routine ACEi / ARBs in COVID-19? As hypertension is one of the most common comorbidities; this is a big question that requires a big answer. The controversy is that ACEi/ARB are typically stopped in acute illness; the negative effect on blood pressure and potentiation of kidney injury being the (very good) reasons for this. This was a retrospective cohort study, and comes with the usual caveats of such a design; risk of both recall and misclassification bias. 1128 hypertensive patients were included of whom 188 were on ACEi/ARB. Overall mortality of those continuing ACEi/ARB in hospital was 3.7% vs 9.8%. Both raw and propensity matched analysis found a consistent association between ACE-i/ARB use and reduced mortality; this held up with sensitivity analysis. At first glance an in-hospital population suggests a sicker cohort of patients, and thus a more important study, however they excluded patients with acute organ dysfunction. Zhang et al attempt to answer this big question but the study does not give us the big answer that we need. Prospective and/or larger sample sizes with local populations are required before this can be practice changing – ideally via RCT for a drug intervention. Carry on as usual for now.

8. Point‐of‐care lung ultrasound in patients with COVID‐19 – a narrative review by Smith et al

• Topic: Diagnostics
• Rating: Worth a peek
• Scout: Dr Charlie Reynard

Ultrasound as divisive, as it is user dependent. as This is a deep dive into COVID-19 lung ultrasound by way of a narrative review. Smith et al note that ultrasound’s superior superficial resolution is well suited to detecting the early stages of COVID-19 given the initial distal pulmonary changes. When you get the probe on the chest you are looking for pleural line irregularities and B-line artefacts. The machines are inherently more portable than the classical x-ray and easier to decontaminate. The authors recommend serial USS may be able to track disease progression, and that a 12 zone lung score could provide objectivity to the scan. I can’t see serial scanning catching on in ED, but an objective score that is monitored through the patient journey has face validity.

9. Guidance for the prevention of COVID-19 infections among high-risk individuals in camps and camp-like settings by Favas et al

• Topic: Global Health 
• Rating: Worth a peek
• Scout: Dr Anisa Jafar

A practical guidance document for a spectrum of professionals and lay-persons managing those at highest risk of suffering most from COVID-19 whilst in camp-like settings such as refugee and IDP camps.  It covers the general principles of pragmatic shielding; specific groups who should be considered for shielding within such settings; development of “green zones” within various contexts; the role of community engagement; and managing commodity distribution and healthcare provision.  Its horizontal emphasis on a community-led approach hold potential lessons for more stable, higher-resource settings.  Given the very top-down approach the UK has embraced, both societally and notably within secondary care, there may be lessons within this and similar guidance from the humanitarian sector which could be applied for greater efficacy. 

10. Hydroxychloroquine in patients with COVID-19: an open-label, randomized, controlled trial by Tang et al

• Topic: Treatment
• Rating: Worth a peek
• Scout: Professor Simon Carley

Hydroxychloroquine is controversial despite what Donald Trump appears to believe. In this open label RCT HCW failed to alter viral clearance in subjects given the drug. However, there are caveats. It’s not really a patient centred outcome, though a reasonable one some might say. The drug was given late in the disease and only to mild/moderate cases so not realy the group we are seeing in Virchester ED. What is interesting is that there were significantly more adverse events in the HCQ group. What this means is that we still don’t know whether HCQ benefits patients or whether it might in fact be harmful. The bottom line is that I think it should only be given as part of a much larger randomised controlled trial with patient orientated outcomes, and I’m pleased that there are now several of these currently recruiting. There is a great review of this on the EMCRIT site by Josh Farkas if you want to read more.

vb

Charlie

NB Don’t forget to check out the last ‘director’s cut’ post if you missed it

RCEM COVID-19 Team

• Dr Charles Reynard, NIHR Clinical Doctoral Research Fellow, University of Manchester
• Dr Tom Roberts, RCEM Trainee Emergency Research Network Fellow, Bristol
• Professor Pamela Vallely, Medical Virologist, University of Manchester
• Professor Richard Body, Consultant in EM, University of Manchester
• Dr Patricia van den Berg, NIHR Academic Clinical Fellow, Manchester
• Dr Anisa Jafar, NIHR Clinical Lecturer, University of Manchester
• Dr Govind Oliver, EM trainee, TERN & RCEM learning editor
• Dr Mina Naguib, NIHR Academic Clinical Fellow, Manchester
• Professor Simon Carley, Consultant in EM, RCEM CPD Director
• Guest Contributor: Dr McCabe, Emergency Department, Hereford County Hospital