Ann Robinson reviews the latest research from the top medical journals

JAMA

Effect of dexamethasone in pediatric cardiac surgery

Dexamethasone is in the spotlight, with positive results for its use in critically ill adults with covid-19. Could it also be useful in reducing complications and mortality in infants undergoing cardiac surgery with cardiopulmonary bypass? This randomised trial of 394 infants with congenital heart defects found that intraoperative dexamethasone didn’t reduce a primary composite endpoint of death and a range of major cardiovascular, renal, or neurological complications compared to placebo. Dexamethasone is widely used in paediatric cardiac surgery, with the rationale that steroids reduce systemic inflammation. The study may have been underpowered to detect a significant clinical benefit, but more work is needed to support or refute the practice. 

JAMA Internal Medicine

Covid-19 death tallies: unreliable and unhelpful?

I think that we’ve all cottoned on to the unreliability of numbers of deaths attributed specifically to covid-19 in the current pandemic. In this US cohort study, the number of deaths due to any cause increased by 122,000 over a 3 month period (March to May 2020) compared to the same period in previous years (an increase of around 15% above normal rates.) This figure was 28% higher than the reported number of covid-19 deaths, reflecting differences in testing and reporting strategies in different states and meaning that it has been hard to estimate the true prevalence of the virus in the US. Excess all-cause mortality is a tried and tested way of tracking flu mortality and this study suggests that it’s the most reliable way of keeping tabs on covid-19 mortality, at least until specific covid-19 testing becomes more widespread and uniform. Those grim daily tallies of covid cases and deaths shed little light and much fear. 

Lancet

Surgery and covid-19

Many patients were fearful of having surgery at the height of the covid-19 crisis. Was their fear justifiable? This international multicentre cohort study looked at all the patients in 235 hospitals who underwent surgery while testing positive for SARS-CoV-2 in the period of one week before to one month after their operation. Three quarters of the operations were emergencies and the 30 day mortality rate was very high—23.8%. Just over half of all those operated on developed lung complications such as pneumonia and this caused most of the deaths. Unsurprisingly, older, male, and frailer patients and those undergoing major, emergency, and cancer surgery did worse. The rates of lung compilations and deaths are higher than those normally quoted for even the highest risk patients in non covid times. The authors recommend postponing non critical surgery during pandemics, which I would have thought is obvious.

NEJM

Covid-19 and children

Children have been relatively unaffected by covid-19. But reports of a multisystem inflammatory syndrome in children (MIS-C) associated with the virus and with similarities to Kawasaki disease and toxic shock syndrome, has caused widespread anxiety among parents (and some urgent revision of the presentation by doctors). This study pulled together descriptive analyses of 191 potential cases in New York State, finding high prevalence of myocarditis, coagulopathy, and gastrointestinal symptoms, though only mild respiratory symptoms. 80% of the children were admitted to intensive care and two died. It was outside the scope of this study, but it seems likely that a post covid-19 inflammatory syndrome may exist in adults too. The extent of multisystem inflammatory syndrome in children may have been underestimated due to gaps in serological testing, medical recognition, and reporting. Increased awareness and active surveillance are key to learning more about this potentially dangerous condition. As we were taught about Kawasaki disease; it may be rare, but always consider it or you’ll miss it.

Annals of Internal Medicine

Metformin alone for low risk diabetes?

According to this study, in patients with diabetes who were at low risk of cardiovascular disease (CVD), all the commonly used add on drugs to metformin based therapies were no better than placebo in improving vascular and mortality outcomes. The systematic review and network meta-analysis of 453 recent clinical trials and databases, assessed 21 drugs given as monotherapy or added on to metformin. 23 trials were arguably the most useful because they compared the drug alone to adding it to metformin. But I was astonished that in diabetic patients at low risk of cardiovascular disease (CVD), adding any of the drugs to metformin resulted in no clinically meaningful improvement in mortality or vascular outcomes. Hba1c levels may have fallen, especially when insulin or glucagon-like peptide-1 receptor agonists (GLP-1 RAs) were added to metformin, but vascular outcomes were no better than adding placebo. Yet another reminder that we’re treating the patient not the blood test result. In diabetic patients at high risk of CVD, adding GLP-1 RAs and SGLT-2 inhibitors to metformin was more justifiable, with a reduction in all cause mortality and stroke risk.

Ann Robinson is an NHS GP and health writer and broadcaster