Ann Robinson’s research reviews—4 December 2020

Ann Robinson reviews the latest research from the top medical journals

Weekly insulin on the horizon for type 2 diabetes?

A weekly dose of a synthetic basal insulin analogue (insulin icodec) was as effective and safe as once daily insulin (glargine) for patients whose type 2 diabetes was poorly controlled with oral hypoglycaemic drugs and who had not previously taken long term insulin. This small, randomised, 26 week, phase II trial found that weekly insulin icodec, which is still in development, lowered glycated haemoglobin levels to a similar extent as a once daily dose of 100 U glargine (decline of 1.33% v 1.15%). Safety end points, including episodes of hypoglycaemia, were also similar and rare. The results overall were encouragingif modestbut cannot be overstated: the trial wasn’t powered to detect significant differences between treatments for any given endpoint, and the study population was small and non-diverse. Because of the trial design and the need to treat to a target, the insulin dose adjustment had to be identical in both groups so wasn’t tailored for each treatment. All of the participants were taking metformin, with or without a dipeptidyl peptidase-4 inhibitor, but patients taking a sulphonylurea were excluded from the study. Further study is needed before the weekly insulin dose can be offered to patients, but the prospect is likely to be popular and improve concordance.

N Engl J Med doi:10.1056/NEJMoa2022474

Ray of hope for the cloud around febuxostat?

The urate-lowering drug febuxostat has been under a cloud since the European Medicines Agency recommended a post-licensing study to assess cardiovascular safety compared with allopurinol. This prospective, randomised, multicentre trial of patients with gout who were over 60 years old, had at least one cardiovascular risk factor (but no myocardial infarctions or strokes yet), and were already taking allopurinol, found that febuxostat was non-inferior to allopurinol in terms of cardiovascular or all-cause death rates. In fact, overall, there were fewer deaths in the febuxostat group. These reassuring findings are in contrast to a previous study (CARES), but the participants in CARES were generally at greater cardiovascular risk and less likely to complete the trial. 

Lancet doi:10.1016/S0140-6736(20)32234-0

Covid-19: no end in sight yet

Is the end in sight for the global pandemic? One key piece of information is how many of us have already been infected. Bajema and colleagues asked what proportion of people in 52 US jurisdictions had detectable antibodies against SARS-CoV-2 between July and September of this year. Based on a repeated, cross sectional study of 177 919 residual clinical specimens, between <1% and 23% of people had antibodies, and in most areas (42/49 jurisdictions) the figure was <10%. Antibody prevalence clearly varies a great deal, but extrapolation from this study suggests that the vast majority (>90%) of people do not show evidence of previous infection. Admittedly participants who volunteered may not represent the population at large; for instance, they may be more health conscious so more likely to stick to infection avoidance measures. And seroprevalence in the US may have increased since September. But even so, it’s clear that most people in the US do not have antibodies and that measures to contain infection need to continue. Roll on the vaccines.

JAMA Intern Med doi:10.1001/jamainternmed.2020.7976

Fighting fat to combat covid-19

Two studies shed light on the link between obesity and outcomes in covid-19. A retrospective cohort study in New York City found that, of 2466 people who were hospitalised with severe symptoms, 22% were intubated, 25% died, and 2% remained in hospital. The severely obese patients (BMI >40) unsurprisingly fared worse than those who were merely overweight. More surprisingly, the association was primarily in those younger than 65 years. Also notable was that the severely obese didn’t have higher admission levels of biomarkers of inflammation, cardiac injury, or fibrinolysis. One important (possibly critical) caveat is that data on BMI was missing in 28% of patients, so that really limits the validity of any conclusions.

Ann Intern Med doi:10.7326/M20-3214

Obesity and mortality in patients with covid-19

Another retrospective cohort study, this time from a large integrated healthcare organisation in California, also found an association between BMI and the risk of death from covid-19, this time in a J-shaped curve, even after adjustment for obesity related comorbidities. Those with a BMI of 40-44 had a relative risk of death of 2.68, increasing to 4.18 for BMI >45, compared with people in the normal weight range (BMI 18.5-24). Again, the risk was most striking in people under 60 years old and in men. There was no increased risk of death associated with Black or Latino ethnicity or other demographic factors. Some data on deaths may have been missed, but the two studies taken together reinforce the association (causality can only be inferred, but it seems plausible) between severe obesity and risk of death from covid-19, especially in people under 60 years old.

Ann Intern Med doi:10.7326/M20-3742

Ann Robinson is an NHS GP and health writer and broadcaster