Alex Nowbar reviews the latest research from the top medical journals.

Lancet

China’s Fangcang shelter hospitals

Only a month ago, pop-ups were hipster cafes. Now the only pop-ups anyone is interested in is the hospital variety. As a field hospital, Nightingale Hospital in London serves a very different set of functions to the Fangcang shelter hospitals that were rapidly erected in China in response to covid-19. Chen et al’s letter presents the key features of these shelter hospitals developed in Wuhan: (1) one to two days to build, (2) thousands of beds, (3) low cost achieved through low staff to patient ratios. They were intended for isolation of people with mild to moderate covid-19 from their families and communities, while providing medical care, disease monitoring, food, shelter, and social activities. When pre-existing hospitals filled up in Wuhan, these people were the ones who were not severe enough to warrant a hospital bed but also needed not to return to their communities yet. Can this model be widely implemented in the UK? I hope so, because, until then, people with symptoms have to isolate at home, and woe betide their family members.

JAMA

Lots in Los Angeles

Mild flu-like symptoms are the talk of the town these days (read: globe). Does everyone with mild flu-like symptoms have the SARS-CoV-2 virus? Without gigantic upscaling of testing, we cannot know and, therefore, cannot optimise transmission reduction. Various approaches have been taken to answer this question. A large US hospital, LA County and USC Medical Center, attempted to estimate community prevalence by testing everyone who presented to them with a mild flu-like illness between 12-13 March and 15-16 March as a sample. This 131 person dataset did not include people with severe respiratory presentations or those with risk factors for the virus because the idea was to represent the people who are quietly wondering if they could be part of the spread. Seven people (5%) were positive for SARS-CoV-2. The authors describe this as “concerning.” I agree, but I’m worried it is an underestimate. It would be useful to do sampling across more environments and regions to try and estimate the true prevalence.

Annals of Internal Medicine

SARS-CoV2 testing of different body sites

I’m impressed that data relevant to the pandemic is being published quickly and clearly but it is a shame so much of it is small and observational. This Chinese dataset looks at 22 people admitted to hospital with a diagnosis of COVID-19 who had gone on to test negative (on real-time PCR) for the virus on a pharyngeal sample and had had sputum or faecal samples taken at any point that were positive. The researchers then extensively tested these people by taking samples from different body sites. The aim was to investigate positivity of the test in samples from different body sites once the pharyngeal sample had converted from positive to negative because this could have implications for subsequent isolation. They found that sputum remained positive up to 39 days after the negative pharyngeal samples while faeces stayed positive up to 13 days later. This suggests that pharyngeal sample negativity in insufficient to determine whether isolation needs to continue. This is critical to containment planning, if indeed there are any, because some parts of the world appear to just be winging it. However, this is not a test of how infectious the individuals were in practice so we can’t jump and say we have to be isolating people for a lot longer than we are now. However, it is not a pleasant thought that people still exude virus weeks after they recover.

NEJM

A case series from Seattle

Oh to be pre-outbreak. Those were the days. Seattle was on the brink of its outbreak in March 2020. Bhatraju et al report the course of 24 patients treated across 9 hospitals in the Seattle area. They all had hypoxaemic respiratory failure and none had had known exposure to a returning traveller (something we don’t talk about anymore since the disease flooded into every corner of the globe). As previously reported for COVID-19, cough and shortness of breath were common. Lymphopenia was present in 75% of them. Interestingly, only 50% had fever on admission. Does this represent that temperature goes up and down over the course of the illness, that fever is a feature that develops earlier or later in the course or simply that fever is not always a feature? 12 of them died. These statistics are gloomy sounding but it is important to remember that most disease is much milder. The gloomiest thing is that we don’t have evidence for any prognosis-altering therapy yet.

Vaginal delivery in mother with COVID-19

Covid-19 case reports are receiving a lot more attention that case reports usually would. While this is a natural human tendency, these elegant anecdotes have huge power to mislead. In this letter, a woman was admitted with Covid-19 symptoms 39 weeks into her pregnancy. A day later she went into labour, then the test back back confirming her Covid-19 diagnosis, then she had an uncomplicated vaginal delivery. She breast fed and she and her baby went home well 6 days after delivery. Telephone follow-up did not identify any signs of neonatal infection.

We can’t draw conclusions about the risks of vertical Covid-19 transmission with vaginal delivery nor safety of breast feedings from this one case. The key element of interest in this report isn’t so much what happened to mother and baby (because this was happily uneventful) but more the approach taken to prevent staff infection i.e. minimising contact and using PPE. This is fundamental as infected staff could go on to put many other patients at risk.

Alex Nowbar is a clinical research fellow at Imperial College London

Competing interests: None declared