In this weekly round-up, Richard Lehman looks at a personal selection of articles of relevance to clinicians dealing with covid-19
Covid and thinking
Medicine is made up of human encounters based on mutual recognition. This goes all the way from simple pattern recognition of disease to the deepest kind of human exchange. Medicine and covid both show us that we don’t all think alike. When I started writing reviews about covid-19 seven weeks ago, I thought it might be useful to pick out a few key themes for clinicians and write a few readable comments about them, as I did for many years each week about the main medical journals. In fact I’ve found it more difficult, because it’s a massive exercise in cognitive adaptation (aka re-cognition): a cruel test for the ageing brain. In this review, I revisit some themes from my previous ones to see where I might have got it right, or wrong.
Moral: re-cognition is hard work, but you can’t do medicine or life properly without it.
Unclean
The oldest form of protection from infectious disease is to keep infected people away. Old Testament lepers are the best-known example. Distancing rules for covid-19 struck me as bizarre from the start, and lack of quarantine as even more so. How can an arbitrary distance of 2m apply across the board, from people sitting inside a restaurant to those in the open air following behind a panting cyclist? [Bahl, P. et al. Airborne or droplet precautions for health workers treating COVID-19? J Infect Dis]. How come that, until now, people could get off planes from epidemic areas and be left to roam the country at will? Why, when it has been known since February that infected people can shed virus for up to 3 weeks following onset of symptoms, can they go back to work 7 days from the onset of illness?
Moral: silly rules strongly stated give a clear message to the public: never trust a word we say.
The test of my words
On tests, I was wrong in my first review. I said on 23 March that “Standardised IgG based testing became simple and scalable weeks ago”. That was based on reports from China, plus the fact that dependable assays for immunoglobulins have been around for about 60 years. It’s a paradox that although the SARS-CoA-2 virus was completely characterised within a month of discovery, we still don’t have a reliable way of testing for active infection, and we can’t even be sure of those for the presence of neutralising antibodies following recovery. The excessive claims of accuracy made by some leading test manufacturers have repeatedly exposed the inadequacy of regulation in this area, and the lack of open data about validation. They’ve also exposed the problems of defining “accuracy” in diagnostic tests, according to how we intend to use them. I feel some new undergraduate teaching materials coming on.
Moral: EBM needs to recover its scope and get back into the business of questioning both tests and diagnoses.
Serum slickness
Serum is the rich sticky liquid that’s left if you allow blood to clot and separate. Plasma is the similar liquid you get if you prevent blood from clotting and spin off the cells. Both are full of assorted proteins and antibodies, and you can use them to confer passive immunity. This magic property was discovered early and from then on serums became the last-resort option for pre-antibiotic era doctors, not to mention charlatans and mad scientists in horror movies. In my first reviews, written before UK schools were closed down, I speculated that government advisers might be planning to use schools as serum farms. Clean, kerchiefed children would sing their way to donation booths, where their blood would be harvested for the good of the nation. Any spare serum not required for covid treatment would be used to support the vital British facial cosmetics industry. This could yet happen, but in the meantime there has been a very good response to a call for adult blood donors who have had confirmed covid-19 infection. Convalescent plasma is undergoing several randomised trials, including REMAP-CAP and an arm of RECOVERY.
Moral: Other people’s antibodies may save your life, but randomised evidence is always needed.
In dreams begin responsibilities
I’m quoting from WB Yeats here, and Yeats spent a good deal of his life in a state of dreaminess, though he also managed to be quite a good and practical Senator for the Irish Free State. Yeats’ youthful dreams began in Celtic fairyland but took an increasingly dark turn in old age. As I’ve mentioned in more than one of these reviews, the dreams of covid-19 patients being ventilated are extremely vivid and often terrifying: fully as bad as great beasts slouching toward Bethlehem or Crazy Jane ranting to the Bishop about copulation. Yeats was an oneirophant, a person who deliberately sought and expounded his visions, whereas people on ICUs crave the opposite. Most people forget ordinary dreams within 4 minutes (this is known as oneirolysis), but ICU dreams characteristically imprint their terror on patients for life. It is time we discovered drugs to avert this. Going with the Greek, we could call these oneirolytics or, if you prefer, ephialtediaskorpizics.
Moral: The horrible dreams of ICU patients need effective oneirolytic drugs, which may already exist undiscovered.
Lovely models available
I have cultivated a flippancy about models of covid-19, languidly waving them aside like Oscar Wilde with his leg over the arm of a chair. “Ah, you bring me another model! Did you make it yourself? How thoughtful. Leave it on the table by the gardenia, dear fellow, and help yourself to a hock and selzer.” Unfortunately this is a pose which I’ve been forced to give up by academic obligation. Public policy about covid-19 has all along depended on the mathematical modelling of spread and of the likely effect of interventions, taken singly or together. Those of us who are bad at sums need guidance from critically aware mathematicians about their assumptions and the scale of their uncertainty. Unfortunately, almost all the evidence we have about interventions comes as bundles of elements adopted at varying times by different countries. Evidence about single interventions is only available in the rare cases where they have been omitted from natural experiments in some populations. That way we can assess their contribution to the main effect in other populations.
Moral: All models are wrong, but modellers should be able to tell us the bounds of their wrongness.
No place like home?
Gather round children, while I tell you about the days when there were very few ICUs and even fewer hospices, and young GPs like myself would care for all sorts of things at home, day and night. Myocardial infarction before thrombolytics, terminal cancer before syringe drivers, heart failure with dig and diuretics: it was just you and the district nurse. If we were still in the 1970s, no doubt most of covid-19 would be looked after in the community, with the odd cylinder of oxygen and bottle of morphine. I spent several fraught weeks in March and April trying to find out what was going on in 2020 amongst patients with covid who decided to remain at home. How did they access care and what did it consist of? Was anyone in charge nationally? How big was the unmet need? I still don’t know the answers. The problem may not be of the scale originally feared (except in nursing homes) but it must be there, largely hidden.
Moral: We are weeks into the UK epidemic, but our primary care based NHS seems to have no coordinated plan for home care.
ACE in spades
One of the running controversies about covid-19 is the effect of taking angiotensin converting enzyme (ACE) inhibitors on the severity of disease. I alluded to this last month. It’s a neat illustration of how biomechanical logic can always be run backwards or forwards according to what you are trying to explain. Initially there were fears that people who took ACEIs were more susceptible to covid-19, and some doctors even advised patients to discontinue them if they had covid symptoms. The logic was that taking blockers of the ACE receptor led the body to produce more ACE-2 receptors and these were the sites which allow SARS-CoV-2 to enter human cells. But a systematic review of the observational evidence now firmly concludes that “prior use ACEIs /ARBs is associated with a decreased risk of death or critical outcome among SASR-CoV-2 infected patients.” So reverse logic is now needed: something to do with functional downregulation; or upregulation in ACE2 receptors? I can’t remember which. I’ve even heard suggestions that ACEIs might now be candidates for interventional trials in covid-19, though they wouldn’t be my automatic choice for an illness characterised by an incessant painful cough.
Moral: when the facts change, you can always change the physiological logic.
Silver lining
I began by mentioning re-cognition. It’s a cognitive process that GPs in particular need to perform all day long. People can sense right away whether you are trying to align yourself cognitively with them in order to see things from their point of view, or if you are brushing aside their point of view in order to assert your own. The first builds trust, and the second builds distance.
Covid-19 shows that whole of medicine needs a process of re-cognition to bring it into alignment with what matters to people. Evidence-based medicine started off with that intention, but over 25 years it has drifted towards becoming a set of technical rules for the conduct and analysis of randomised controlled trials. The pandemic forces us to look at the actual kinds of evidence around us when we have to make important decisions in a hurry. Nobody has got it entirely right, but the UK has come closer than most countries to getting it entirely wrong. We need to learn from the decision-making processes of successful nations. We need to look at the heuristics of dealing with irreducible uncertainty. We need to revisit and revise the basic principles of diagnostic testing applied to different problems and different populations.
This is only part of an agenda of re-cognition that is needed: let’s get on with listening properly to the concerns of people, and sharing the work out
Confucius: If you think in terms of a year, plant a seed; if in terms of ten years, plant trees; if in terms of 100 years, teach the people.
Richard Lehman is professor of the Shared Understanding of Medicine at the University of Birmingham.
Competing interests: None declared
Articles from MedRxiv get special prominence in these reviews. MedRxiv is a completely free site for the rapid exchange of knowledge which was launched last year as a joint initiative by The BMJ, Cold Spring Harbor Laboratory and the Yale University Open Data Access project. All MedRxiv articles carry the caveat: This article is a preprint and has not been certified by peer review. It reports new medical research that has yet to be evaluated and so should not be used to guide clinical practice. This review is intended as a quick source of information in a readable form. While every effort has been made to be accurate, the opinions are those of the author and should not be relied on without reading the full articles cited in the context of current NHS guidance.