the metablog

Does a GA CS increase PPD risk? Plus LMA studies & COVID vaccine optimism

Thu, 22 Jul 2021 02:04:00 +0000

GA caesarean section & post-partum depression

This large study (Guglielminotti 2020) of 428,204 New York caesarean section records (2006-2013), including 34,356 general anaesthetics (8%), investigated the association between mode of anaesthesia and post-partum depression (PPD). Other studies have shown an association between caesarean section (emergency > elective) and PPD. (Sun 2021, Xu 2017, and others), though this is the first to look specifically at general anaesthesia as a PPD risk factor.

Guglielminotti and Li found that general anaesthesia increased the odds of severe PPD by 54% (aOR 1.54, 1.21-1.95), and suicidal ideation by a massive 91% (aOR 1.91, 1.12-3.25), though not a significant increase in anxiety or PTSD.

The researchers discuss many potential causative factors, particularly known associations between GA CS & poor pain control, and subsequent pain & PPD – while also acknowledging the obvious potential for confounders. Of note patients receiving GA were older, more often non-Caucasian, had more co-morbidities, neonatal complexity, and lower socio-economic levels – also all independently associated with PPD risk.

In order to quantify the potential confounding contribution of emergency vs elective status, the researchers employed the novel E value:

To assess the impact of emergent cesarean delivery on our results, we calculated the E value associated with the aOR for the risk of PPD and suicidality. This relatively new metric takes into consideration 2 associations: (1) that between the confounder (emergent cesarean delivery) and the outcome (PPD); and (2) the association between the confounder (emergent cesarean delivery) and the exposure (general anesthesia).

An E value of 1.7 for the unmeasured confounder emergent cesarean delivery indicates that to explain away the association between general anesthesia and depression, either: (1) emergent cesarean delivery increases the risk of depression by at least 70%; or (2) emergent cesarean delivery is at least 70% more prevalent among general anesthesia than among neuraxial anesthesia. Either association is clinically plausible.

Keep it in perspective...

We already know that general anaesthesia for CS is suboptimal: it compromises both maternal experience and safety, but it should (hopefully) only ever be a chosen mode of anaesthesia when there is a true contraindication to regional anaesthesia – even at the modestly-high 8% GA rate among this New York cohort.

Looking at it from the other end, bear in mind that the modestly-faster time-to-incision for GA over regional is also of questionable neonatal benefit.

The take-home:

Just another reason to avoid GA CS when possible – but you already knew that, right?

"...general anesthesia is a potentially modifiable risk factor for PPD. This finding provides further supporting evidence favoring neuraxial over general anesthesia in cesarean delivery whenever possible."

Supraglottic airway training and manikins

Interesting prospective simulation & equipment study by way of the University of Freiburg. Schmutz et al. investigated how effective five different second generation supraglottic airway devices (SADs) performed in two common airway manikins: the TruCorp AirSim® and the crowd favourite, Laerdal's Resusci Anne® Airway Trainer™.

While ventilation was achieved in all SAD-manikin combinations, the Resusci Anne® Airway Trainer™ was associated with better and more consistent performance for SAD position, participant subjective assessment and ease of gastric tube insertion for most of the SADs. The TruCorp AirSim® did however achieve better leak pressures across most of the SADs (LMA® Supreme™, Ambu® AuraGain™, i-gel®, KOO™-SGA & LTS-D™).

But then, what are the implications for airway simulation training? The researchers correctly note that:

The most important quality of a manikin is the ability to simulate the real-world conditions and thus to give the trainee an authentic feedback.

The bottom line for SAD manikins?

While considering how manikin choice and SAD availability match with your aims for simulation training, the bigger picture is that the primary goal of any manikin-SAD coupling is real-life fidelity – and for that reason, participant subjective assessment is king. And so in this study at least, the Resusci Anne® Airway Trainer™ wins.

Head-rotation to improve LMA leak pressure?

Another interesting little airway study, this time out of Sapporo, Japan (Chaki 2021). A neat little randomised but-not-blinded study of the effect of head rotation on the oropharyngeal leak pressure of both the i-gel and LMA Supreme 2nd generation supraglottic airways.

The researchers investigated the leak pressure (OPLP) of the i-gel and LMA Supreme in paralysed patients with the head: 1. neutral, 2. rotated 30°, and 3. rotated 60°. They found that rotation of the head through 30° and 60° progressively increased OPLP by a clinically-significantly amount (0° vs 60° 5.5 cmH₂O (3.3-7.8) & 6.5 cmH₂O (5.1-8.0) respectively).

Before you get too excited...

The result however may not be reliably applicable to all populations, notably the study subjects were overwhelmingly small (x̄ ~160cm & 60kg) Japanese women (71%), receiving a TIVA muscle-relaxant anaesthetic (propofol, remifentanil, rocuronium). How well this improvement-with-rotation holds up among, for example, spontaneously ventilating large Caucasian males, is unclear.

Bottom-line

When using an i-gel or LMA Supreme 2nd generation supraglottic airway, careful head rotation to 60° may increased oropharyngeal leak pressure and so assist with ventilation troubleshooting. However head and neck rotation of anaesthetised, paralysed patients should be performed gently and cautiously – you are after all, not a chiropractor!

COVID mRNA vaccines and immune persistence

A great Good News study just published in Nature by researchers from Wash U Med investigated persisting COVID immunity following the Pfizer mRNA vaccination (BNT162b2). Turner et al. looked at the presence of not only circulating antibody-secreting B cells, but also germinal centre B cells found in the axillary lymph nodes of 14 study volunteers.

While the persistence of mRNA-vaccine induced immunity to SARS-CoV-2 has already been demonstrated to last at least 6 months, and likely 12 months, we do not yet know if or when vaccine boosters will be required beyond that.

Turner's study is particularly exciting because they found spike-protein binding B cells in the germinal centre of draining lymph nodes in all 14 post-immunisation participants for the full 15 weeks of the study. The germinal centre response was so vigorous and persistent that the researchers believe this could represent COVID-protection lasting for years.

"Ellebedy said the immune response observed in his team’s study appears so robust and persistent that he thinks that it could last for years. The researcher based his assessment on the fact that germinal centre reactions that persist for several months or longer usually indicate an extremely vigorous immune response that culminates in the production of large numbers of long-lasting immune cells, called memory B cells. Some memory B cells can survive for years or even decades..." – Dr Francis Collins, NIH Directors Blog

This study builds on the same team's earlier work (Turner 2021 May) looking at bone marrow plasma cells in those who have recovered from mild COVID infection, also showing a long-lived immune response.

COVID persistent immunity takeaway:

Although COVID-19 and developed vaccines have been circulating for only 12-18 months, these immune-response studies give some hope that the miracle of mRNA vaccines may not only be in their efficacy, but also in the longevity of protection.

Mentioned studies:

Guglielminotti J Li G. Exposure to General Anesthesia for Cesarean Delivery and Odds of Severe Postpartum Depression Requiring Hospitalization. Anesth. Analg. 2020 Nov 1; 131 (5): 1421-1429.
Schmutz A, Bohn E, Spaeth J et al. Comprehensive evaluation of manikin-based airway training with second generation supraglottic airway devices. Ther Clin Risk Manag. 2019 Jan 1; 15: 367-376.
Chaki T, Tachibana S, Kumita S et al. Head Rotation Reduces Oropharyngeal Leak Pressure of the i-gel and LMA® Supreme™ in Paralyzed, Anesthetized Patients: A Randomized Trial. Anesth. Analg. 2021 Mar 1; 132 (3): 818-826.
Turner JS, O'Halloran JA, Kalaidina E et al. SARS-CoV-2 mRNA vaccines induce persistent human germinal centre responses. Nature. 2021 Jun 28.

PONV, Perioperative Bleeding Aids & Surgery Timing After COVID

Sun, 04 Jul 2021 01:43:00 +0000

A big PONV meta-analysis

Interesting Cochrane meta-analysis looking at PONV prophylaxis from German (Weibel et al. 2021) that included almost 100,000 study participants across 585 trials. Interesting not so much because it confirms much of what we already new (or assumed, based on our common PONV prophylaxis drug choices), but because it reassures us that side-effects from commonly used PONV drugs are low to non-existent.

PONV Takeaway:

Granisetron is probably the best single-agent or in combination with other agents because of it's efficacy (better than ondansetron), low-cost, long duration, and absent side-effects.

A cognitive aid to better manage perioperative bleeding

Although the benefits of cognitive aids to many areas of anaesthesia are well established, our resistance to using decision support tools persists. Whether due to misplaced perceptions of losing autonomy or Dunning Kruger-adjacent inflated belief in our ability to perform under pressure, is unclear.

In Anaesthesia, Kataife et al. (2021) describe a cognitive aid for better managing perioperative haemorrhage, the Haemostasis Traffic Light algorithm. Using a simulation-based RCT across two centres (University Hospital Zurich & The Italian Hospital of Buenos Aires, N=84), they showed that using the HTL improved case solutions (OR 7.23, 3.82-13.68), quickened therapeutic decisions, (HR 1.97, 1.18-3.29), improved therapeutic confidence, (OR 4.31, 1.67-11.11) and reduced workload perception.

The aim of the HTL is to improve both situational awareness and decision making, by integrating clinical judgement and point-of-care testing (ROTEM) within an accessible, structured algorithm.

Haemostasis Traffic Light takeaway:

Kataife's study again shows the benefit of cognitive aids, particularly in critical, time-sensitive situations. The anaesthesia and critical care community's historical resistance to decision-support tools requires challenge.

Surgery timing after COVID infection

Following their important 2020 study of the risk of surgery for COVID patients, the COVIDSurg Collaborative reports on their prospective cohort study aiming to determine the optimal delay for planned surgery after COVID infection.

Once again this was an international (116 countries), multicentre, prospective cohort study including all surgery types, over 140,000 patients, and 3,127 post-COVID. Once again the 30-day postoperative mortality was sobering: even in the 5-6 week post-COVID group, 30-day mortality was dramatically higher (OR 3.6, 2.0-5.2) compared to those without a COVID diagnosis. Worryingly the risk was consistent among both low-risk and high-risk surgical groups.

Although after the 7-week mark postoperative mortality was similar to non-post-COVID patients (OR 1.5, 0.9–2.1), those with persisting COVID symptoms still suffered a 6.0% 30-day mortality (3.2–8.7). (30-day mortality among non-COVID patients was 1.5% (1.4-1.5).

Post-COVID surgical timing takeaway:

Surgery should be delayed for at least 7 weeks after COVID, although those with persistent COVID symptoms will still have more than twice the 30-day mortality than those without.

Mentioned studies:

Weibel S, Schaefer MS, Raj D et al. Drugs for preventing postoperative nausea and vomiting in adults after general anaesthesia: an abridged Cochrane network meta-analysis. Anaesthesia. 2021 Jul 1; 76 (7): 962-973.
Kataife ED, Said S, Braun J et al. The Haemostasis Traffic Light, a user-centred coagulation management tool for acute bleeding situations: a simulation-based randomised dual-centre trial. Anaesthesia. 2021 Jul 1; 76 (7): 902-910.
COVIDSurg Collaborative GlobalSurg Collaborative. Timing of surgery following SARS-CoV-2 infection: an international prospective cohort study. Anaesthesia. 2021 Jun 1; 76 (6): 748-758.

Why were Anaesthetists so early on COVID?

Wed, 01 Jul 2020 01:27:00 +0000

On a Monday morning in March, an anaesthetist stood outside his children's inner Sydney school as a solitary protestor, asking parents to keep their children home if they could.

In the heady early days of the coronavirus pandemic, alarm was raised by a disparate mix of professionals: virologists and epidemiologists, journalists and technologists, and a range of frontline and critical care medical specialists exposed to the first COVID patients.

But as concern spread from the earliest hit countries to those threatened by their own surge, one specialty group was over-represented in public calls for early action: anaesthesia.

In Australia, medical anaesthetists from all states and territories spoke-up, not for health authorities and legislators, but for their communities. For a specialty most comfortable when not spoken of, suddenly anaesthetists were appearing in national newspapers, on radio, television, and even (very small) picket lines.

In Victoria, Dr Pieter Peach was a prominent early voice pushing for cancellation of Melbourne's Grand Prix. The Australian Society of Anaesthetist's fearless president Dr Suzi Nou guided the society's careful campaigns to prepare for the pandemic, pause elective surgery and then cautiously restart. In NSW, Dr Tanya Selak's advocacy was celebrated on Telstra's #saythanks billboard, while Rob picketed outside his children's school.

Like a warning telegram from 1940s London, the message scrawled on Rob's chest captured the zeitgeist of our specialty at the time:

"Lives depend on it. Government too slow to act."

More anaesthetists added their voices, like Dr Cameron Graydon in Queensland, and WA AMA President Dr Andrew Miller, who cut straight through the noise to call a spade a f%$king shovel to the appreciation of the Western Australian public. Plus many, many more of us working publicly or behind the scenes.

What was it about anaesthetists and anaesthesiologists¹ that made for such vocal Cassandras? Was it simply panic built upon our fear of personal risk? A plugged-in critical care community watching stories from Italian anaesthetists in horror? And then the same from London… and New York…?

While these played a part, more importanty there are unique cultural and cognitive features of anaesthesia and anaesthetists that motivated many to give voice to the risk. Some of these also appear in sibling critical care specialties, among emergency medicine and intensive care specialists – but not in the same ways, and diluted by their own tribe's characteristics.

Anaesthesia culture

Culturally anaesthesia is typified by strong group cohesion, teamwork, bias toward collaboration and in-group trust – and yet mistrust of institutional authority. Ironically Australia's outgoing Chief Health Officer, Mr Covid himself, learnt this the hard way in his earlier life as CEO of Austin Health.²

It is probably the simmering mistrust of authority mixed with our quiet subversive streak that motivated anaesthetists to speak out once the pandemic risk was internalised. Although many anaesthetists actually have additional Master's degrees in Public Health, Biostatistics and even Epidemiology (we're an over-educated bunch; there are quite a few PhD's out there too...) as pandemic-expert-adjacent we do not carry the same reputational risk as fully-fledged epidemiologists, public health or infectious disease specialists that might have otherwise tempered our loud advocacy.

Broadly, we are always up for epistemic trespass.

Anaesthesia cognition

But why did anaesthetists so quickly identify the growing risk even before the WHO declared a pandemic?

Anaesthetists have an intuitive understanding of Bayesian probability, asymmetric risk³, systems thinking, and contrary to stereotypes, profound risk aversion. Anaesthetists know risk.

I don't mean "hey bro, we understand risk", I mean we really know risk. We feel risk. The way an experienced pilot feels the plane as she flies. The way Neo sees the matrix.

Each day anaesthetists balance the competing risks of hundreds of decisions, big and small. Factoring in patient wishes, surgical needs, priors and specific situations. We are obsessively proactive, understanding the lag between intervention and effect, carefully adjusting the now to avoid small diversions that may multiply later into bigger, badder outcomes.

We allow for black box decision making, knowing that our understanding of human physiology is only a simplified model that helps navigate how someone behaves before, during and after anaesthesia. Many steps of our decision making are guided by necessarily-imperfect models.

Anaesthetists understand that responses are time critical – we are specialists of dynamic decision making. We know that acting early with incomplete information is often better than acting too late with perfect information. We are biased to act and intervene, perhaps to a failing, rather than to watch and wait.

These traits are common to many anaesthetists, whether the specialty selects the type A personalities that typify our craft, or we become this way as we train, encultured into the behaviour.

Asymmetric risk and the pandemic

Whether nature or nurture, anaesthetists immediately understood the asymmetric risk of the SARS-CoV-2 pandemic with it's almost uncapped downside, but manageable upside. We were wise to the reality of exponential growth, sigmoid curves and the lag time for exposure, illness, hospitalisation and death. Anaesthetists were very comfortable with the uncertainty of incomplete case information, but could see the rough probability tree of outcomes. Many of these were very bad, enough were apocalyptical.

Anaesthesia has a long history of driving patient safety improvements⁴, both because of our daily risk consumption, and also because culturally we see ourselves as patient advocates. For many anaesthetists it was only a small extra step to go from advocating for patient safety in hospitals, to the shared safety of the wider community.

Most importantly we saw the need to act early – the disproportionately greater benefit but lower cost of early intervention versus waiting. When WHO pandemic experts extolled the need to "go hard and go early" it felt immediately familiar.

Finally, threaten anaesthetists with conscription to intensive care units and we will move heaven and earth to ensure it will not be neccessary.

In Australia and the UK all anaesthetists are doctors (ie. medical anaesthetists), so I use this term throughout interchangeably with 'anaesthesiologists' – although we are considering a name change to align with our North American peers. ↩
Brendan Murphy had a rather long and fraught battle with a Melbourne Department of Anaesthesia, in which he belatedly discovered the heft of the specialty and it's one-for-all-and-all-for-one advocacy for work conditions, patient safety and anaesthesia services. The dominant theme revealed by Dr Murphy's public statements at the time was simply disbelief, betraying his lack of understanding of the considerable pride that anaesthetists take in the care they provide patients. ↩
Also related and releveant is Daniel Kahneman's prospect theory, the foundation of modern behavioural economics, describing the asymmetry in how individuals assess risk from a loss and gain perspective – both our tendency to be 'loss-averse' but also how this may flip depending on the level of risk and the expected outcome of a decision. This has a lot of relevance to how the general public has responded to the COVID pandemic, however is different to 'asymmetric risk' which describes situations with disproportionate difference between magnitude of downside risk vs upside gain. ↩
For bragging-rights, anaesthesia is the only specialty to have achieved six-sigma safety levels when considering anaesthesia-related deaths. Of course death is a very imperfect measure of anaesthesia quality – no anaesthetist would consider its avoidance even a minimum quality standard, quality is after all a multi-domain concept – but the incredible and purposeful improvement in anaesthesia safety since the 1950s is an impressive example of the prioritisation of safety by our entire specialty. Over 70 years we have brought about a 100-fold decrease in anaesthesia mortality through the work of many anaesthesia organisations and societies, bringing anaesthesia-related mortality down to a "6 sigma defect rate". In the manufacturing industry this describes a 99.99966% defect-free process, 3.4 'defects' per million or a 1 in 294,000 chance of dying due to an anaesthesia-related cause. Roughly comparable to the mortality data we have today. ↩

The Paradox of Avoidance

Wed, 24 Jun 2020 02:04:00 +0000

The cries of 'over-reaction' are as predictable as they are simplistic. Epidemiologists and pubic health experts knew they were coming, because avoiding a disaster brings little thanks.

As Australia emerges unevenly from its soft lockdown, anxious and still responding to COVID flares, it is the envy of much of the world. Alongside success achieved in places as varied as New Zealand, Vietnam, Taiwan, South Korea, Iceland and even austerity-inflicted Greece – the contrast with those that have suffered disastrous outcomes is obvious.

And yet the naysayers still question the painful necessity of the lockdowns, even as study after study demonstrates how the measures have avoided or delayed hundreds of millions of infections.

The United Kingdom, uniquely and even justifiably proud of its National Health System, first chose instead to ignore expert advice and offer up its venerated NHS as a funeral-pyre sacrifice to COVID and the gods of conservatism. Today more than 43,000 British are dead. (You know what would help NHS healthcare workers more than clapping? Earlier border closure and adequate access to PPE!)

Sweden pursued a Claytons lockdown founded on an ideological mix of misplaced intellectual-exceptionalism and responsibility-abdicating libertarianism (my eyes are rolling...). It has not gone well.

Despite making up less than 40% of the Nordic population, Sweden now accounts for five out of of six Nordic COVID deaths. Sweden has twice the population of neighbouring Norway but twenty-times the COVID mortality.

And then there is the Land of the Free, the United States. The wealthiest country on the planet, now the epicentre of the pandemic with well over 2 million infected and 120,000 deaths. A disaster due to a failure to act either early or decisively, a dash of magical thinking and an embarrassing absence of leadership.

It is disappointing that the main contribution the United States is currently making to the global COVID response is to serve as a warning to other nations.

The common theme among these failures is the inability to develop a rational response guided by expert advice. A virus does not care for your wishful thinking. A virus does not care for your political posturing, your dog-whistling or your belief in national uniqueness. A virus is the honey-badger of microbiology.

Which leads us back to the braying calls of 'over-reaction'. Whether former politicians, conservative economists, or performance artists playing opinion writers, they are all making the same mistake: misattributing causality and misunderstanding the purpose of modelling.

As my friend @jamesdgriffiths suggested, it's like angrily claiming "the boy cried wolf" while around us our neighbours are decimated by packs of hungry wolves.

"We've had so few wolves here – why did we even bother with these fences!?"¹

To label Australia's successful COVID response an 'over-reaction' is an irrational, reality-defying argument that willfully ignores the terrible suffering happening globally. The same accusers denied the threat early in the pandemic, opposed the need for significant action, and then threw out the false choice of the economy over health.² It is indefensible that they now attack these very measures because they have actually worked.

Models by their very nature are dynamic, each unique and covering a spectrum of outcomes. For policy makers they are not about accurate prediction, they exist to offer an abstracted mathematical view of the real world that can be used to answer 'what if' questions.

What if we do nothing? What if we quarantine arrivals? Close businesses? Schools? Socially distance? How much hospital capacity could we need? How close can people safely approach Peter Dutton? Should anyone ever approach Peter Dutton? What if public behaviour does not change at all?

For the greater public, model projections are even more critical. They warn and motivate us to act, to change our behaviour so as to avoid disaster. At their best, the most dire models are self-defeating prophecies.

It is ridiculous to decry the success of Australia's COVID response by pointing at the avoided worse-case scenarios of the initial projections.

The models incorporate a complex world of assumptions and necessary simplifications, producing a metaverse of outcomes – almost all of which were bad. Some were very, very bad.

Their purpose was to show an uncertain population 'this is the suffering that could happen', so that it might not.

So instead we are happily condemned to a paradox of avoidance.³

Or Homer Simpson's unintentionally insightful comment: "You know, we're always buying Maggie vaccinations for diseases she doesn't even have!" ↩
This false dichotomy pushed early in the pandemic, pitting the economy against the health-needs of the population was both philosophically immoral and is now demonstrably wrong. Economic prosperity is tightly intwined with the community's health – sick people tend not to be particularly economically productive, and the dead even less so. Several studies now show that the public moved to alter their own behaviour and reduce their economic activity before lockdown orders came into effect. Similarly, countries that have opened up have seen marked reluctance of their communities to return to pre-pandemic levels of cosumption. It turns out it's not so much the government enforcing lockdowns that tanks the economy, but the risk of a contagious fatal disease when buying groceries. The economy will recover by controlling the pandemic as quickly as possible. Never forget that the economy serves the people, not the converse. ↩
I searched for an existing term for this phenomenon, but stayed with 'the paradox of avoidance'. Surely disaster experts and public health specialists must have something they already use to describe this? The paradox of prevention has pleasing alliteration, but is an existing public health term that describes a different prevention dynamic. @DrLKVaughan suggested 'Leavitt's Law' after Michael Leavitt – but not only was Leavitt talking about resistance to action before a pandemic hits rather than the perception after success, more importantly there is already a Leavitt's Law in astronomy – and it seems doubly wrong to erase the historical legacy of another woman of science. My friend @jgolshevsky suggested 'victory blindness' which does have a nice ring to it… ↩

Flattening the curve of pandemic research

Tue, 09 Jun 2020 05:45:00 +0000

The wave of COVID research continues, much of it low-quality and hurriedly published. This is apparently the norm for academic publishing during a pandemic: fast, furious and haphazard.

However, two very significant reviews appeared in The Lancet this week that impact and inform anaesthetists and other critical care specialists.

Post-operative mortality, COVID & surgery

Early pandemic data suggested that COVID-19 patients having even minor elective surgery suffered worse post-operative outcomes, particularly high post-operative mortality. The COVIDSurg Collaborative has confirmed this after a multicenter trial across 24 countries.

Across the entire 1,128 patient cohort, 30-day mortality was a jaw-dropping 24%. Yes, 1 in 4 died within 30 days of surgery.

Pulmonary complications (pneumonia, ARDS or unexpected post-op ventilation) were very common (51%) and were associated with an even higher mortality (38%; and 83% of all deaths). Mortality was unsurprisingly associated with older age ≥ 70 years, male sex, ASA ≥ 3, emergency surgery, major surgery, and malignancy.

But even among low-risk groups, post-operative mortality was shockingly high: 30-49 year olds (6%), women (18%), ASA 1-2 patients (12%), and even those without comorbidities (7%). Being asymptomatic at admission did not have a significant protective effect (22% vs 27% mortality).

Elective surgery still carried a 19% mortality rate, and even for minor surgery mortality was 16%! Anaesthesia modality (local, regional or GA) did not have a significant impact.

Click through to read the summary or full-text, though the obvious take-away is that non-essential surgery should be avoided as much as is possible in those with confirmed or suspected COVID.

This will have huge implications for recommencement of elective surgery in many pandemic-hit countries. (Cook & Harrop-Griffiths explore this very topic in an NHS-context in their recent editorial.)

Physical distancing, face-masks & eye protection

On the back of another Lancet study the WHO has changed their advice on wearing face-masks. The WHO now recommends that the general public wear face-masks when unable to physically distance.

In this review and meta-analysis, the Canadian COVID-19 SURGE group (Systematic Urgent Review Group Effort) looked at the effect of three non-pharmacological interventions on coronavirus transmission:

Physical distancing
Face masks
Eye protection

They analysed 172 observational studies with over 25,000 patients in both community and healthcare settings, covering not just SARS-CoV-2, but also SARS and MERS.

Reassuringly they showed that physical distancing > 1 meter was associated with lower transmission risk (risk difference 95% CI -11.5 to -7.5%) and also increasing protection as distance extended beyond 2 meters.

Face-masks were also associated with reduced transmission (risk difference 95% CI -14.3% to -15.9%, though with low certainty), as was eye protection (risk difference 95% CI -12.5% to -7.7%).

Interestingly N95 masks were even more strongly associated with risk reduction, as was mask use in a health-care setting vs non-health-care. Both N95 and multi-layer surgical masks were more protective than single-layer masks.

The take-home is that simple protective behavioural changes, namely physical distancing, face-mask use and eye protection ARE associated with a significant risk reduction of coronavirus transmission.

Quick takes...

Bonus speed-round: Boulware et al. reported in the NEJM on their RCT of hydroxychloroquine for COVID post-exposure prophylaxis. Bottom-line, HCQ probably gives no post-exposure benefit when taken within 4 days of COVID exposure.

And keeping everyone on their toes on the back of the Surgiphere COVID data scandal, two notable COVID retractions in both The Lancet and NEJM:

Mehra et al.'s look at cardiovascular disease and COVID, notable for the lack of association they concluded between bad COVID outcome and ARB/ACEIs. [retracted]
Mehra et al's investigation of hydroxychloroquine for COVID that failed to show any benefit, with or without a macrolide. [retracted]

Take-home? Signal or noise, it's hard to say – but a sci-fi writer, an adult-content model and cryptocurrency are a unique mix in the medical data industry...

Relevant papers

COVIDSurg Collaborative. Mortality and pulmonary complications in patients undergoing surgery with perioperative SARS-CoV-2 infection: an international cohort study. Lancet. 2020 May 29.
Cook TM Harrop-Griffiths W. Kicking on while it's still kicking off - getting surgery and anaesthesia restarted after COVID-19. Anaesthesia. 2020 May 19.
Chu DK, Akl EA, Duda S et al. Physical distancing, face masks, and eye protection to prevent person-to-person transmission of SARS-CoV-2 and COVID-19: a systematic review and meta-analysis. Lancet. 2020 Jun 1.
Boulware DR, Pullen MF, Bangdiwala AS et al. A Randomized Trial of Hydroxychloroquine as Postexposure Prophylaxis for Covid-19. N. Engl. J. Med. 2020 Jun 3.