Alex Nowbar reviews the latest research from the top medical journals

Long covid in China

It’s important for us to know about long covid. I’d like to know what it is (not just anecdotes) and if we can do anything about it. Other common queries are, who is affected and why? Huang and colleagues’ cohort study of almost 2000 people with covid-19 who had been discharged from a hospital in China six months before sheds a little light, but the study design (single centre, only hospitalised patients, no baseline data, no control group) limits what we can learn. For example, the study excluded people who were unable to move freely due to osteoarthropathy or who were immobile (such as due to stroke). Presumably this was because these patients would have been unable to do the exercise testing part of the protocol, but surely their symptoms would still have been worth collecting? The headline findings were that nearly two thirds of the subjects had fatigue or muscle weakness six months after discharge, a quarter had sleep difficulties, and almost a quarter reported anxiety or depression. These are high rates, but many questions about long covid remain. And more questions are raised. For example, how much more prevalent are these features than in people who have been severely unwell with other conditions?

Lancet doi:10.1016/S0140-6736(20)32656-8

Respiratory ruminations

Couturaud and colleagues ask whether we should screen people with chronic obstructive pulmonary disease (COPD) who have been hospitalised with acutely worsening respiratory symptoms for pulmonary embolism? Pulmonary embolism is in the differential diagnosis in such acute scenarios and can be easily missed and labelled as an exacerbation of COPD. The authors assessed the prevalence of pulmonary embolism in a cohort of 740 such patients in seven French hospitals. The prevalence (counted within 48 hours of admission) was 5.9%. In the study, patients were first classified as suspected or not suspected pulmonary embolism. Then a standarised algorithm was used for their diagnostic work-up based on Geneva score, D-dimer, spiral CT pulmonary angiogram, and leg compression ultrasound. In those pre-classfied as suspected, the prevalence of pulmonary embolism was 10%. In those not suspected, the prevalence was still 3%. One study limitation is that people with COPD but milder symptoms or in whom pulmonary embolism wasn’t suspected may not have been admitted and therefore would not have been included. However, this large study lays the groundwork for further studies to assess the utility of systematic screening for people with COPD presenting with acutely worsening respiratory symptoms.

JAMA doi:10.1001/jama.2020.23567

New drug effective for covid-19

Does anyone else find it strange how little fanfare there has been about tocilizumab? I mean, sure, there is a lot to worry about in the world right now, but all the more reason to focus on some good news. In this study 389 people hospitalised with covid-19 pneumonia but who weren’t receiving mechanical ventilation were randomised to intravenous tocilizumab or placebo in a double-blind fashion. There was a 44% reduction in the primary outcome of either mechanical ventilation or death at 28 days. Perhaps people’s reservations about this result are that the drug didn’t seem to reduce mortality alone. But I think we should be pleased that it reduced progression to ventilation; this finding is resistant to bias given that the trial was blinded.

N Engl J Med doi:10.1056/NEJMoa2030340

Convalescent plasma in early covid-19

I’m going to share a secret with you. I never thought convalescent plasma would be of any use in any condition, least of all covid-19. I was still ready to donate my own antibody-laden plasma but, to paraphrase for this low-key humble boast, the NHS said that, as a puny woman, I didn’t have enough circulating volume to help. In Argentina, 160 people aged over 75 years (or over 65 with comorbidities) with covid-19 symptoms that had started within the past three days and who hadn’t already developed severe respiratory disease were randomised to either convalescent plasma or placebo in a double-blind fashion. The risk of the progression to severe respiratory disease was reduced by 48%—again, not much fanfare about this, perhaps because the improvement was in reducing disease progression rather than mortality. The higher the donor IgG titre, the greater the reduction in risk. One key point in this study was that early intervention in the disease process was important.

N Engl J Med doi:10.1056/NEJMoa2033700

Small non-randomised study of repurposed sitagliptin

Farag and colleagues studied sitagliptin as a therapy for preventing acute graft versus host disease (GVHD) after allogeneic stem cell transplantation (such as for acute myeloid leukaemia) along with other immunosuppressive therapy. They found a low incidence of acute GVHD in their single-arm non-randomised study. What does this mean? Well, it’s certainly good news for 34 of the 36 participants who had not experienced acute GVHD by day 100 and suggests this treatment is worth investigating further. But, without a control group for comparison, we don’t know what the incidence would have been without the drug. If you’re interested, sitagliptin inhibits dipeptidyl peptidase 4. This enzyme degrades things that stimulate insulin secretion, so blocking it is useful for diabetes; but the enzyme also enhances the T cell immune response, so blocking it could help in preventing graft versus host disease. Cool.

N Engl J Med doi:10.1056/NEJMoa2027372

Alex Nowbar is a clinical research fellow at Imperial College London