Alex Nowbar reviews the latest research from the top medical journals

Improving quality is harder than it looks
If only people understood their condition and its treatment better, they would be more likely to adhere to treatment and thus have better outcomes—right? No. An education intervention in hospital and post-discharge for people with heart failure with reduced ejection fraction was tested in a cluster randomised trial called CONNECT-HF. It did not improve clinical outcomes (defined as rehospitalisation for heart failure and all-cause mortality) or care (defined as the percentage of recommendations followed). On one hand, this negative trial result could be seen as disappointing. On the other hand, resources devoted to such endeavours could potentially be allocated to strategies focused on altering prognosis. Or we could quibble about whether an education intervention would be more effective if delivered differently—for example, digitally or in a more personalised way, or targeted to certain patients, such as those with less social support or those with many other comorbidities.

JAMA doi:10.1001/jama.2021.8844

Smoking cessation after lung cancer diagnosis
Yes to smoking cessation. Obviously, always. But is there a point when it is too late? After the diagnosis of lung cancer, for example, does quitting smoking alter prognosis? Sheikh and colleagues performed a prospective cohort study to address this question in 517 smokers with early stage non-small cell lung cancer in Russia. The difference in survival between those who quit and those who continued smoking was stark: it was 21.6 months longer in those who quit. In an adjusted analysis there was a clear reduction in all-cause mortality, cancer-specific mortality, and disease progression. Of course, this is observational research, so the relationship could be confounded. For example, people who are more likely to quit may also be less likely to have other comorbidities, and the mechanism of improved prognosis might not be smoking cessation alone. Still, it’s hard not to be convinced by these data.

Ann Intern Med doi:10.7326/M21-0252

Covid-19, strokes, and heart attacks
Katsoularis and colleagues’ Swedish study confirmed covid-19 as a risk factor for myocardial infarction and stroke. A cynic might say that any critical or inflammatory illness is a cardiovascular risk factor. It seems that there is more to it than just that, although this study only compared the risk to that of the background population, so we don’t know. This study (the first to use self-controlled case series methodology in this area) supports the theory that covid-19 predisposes a person to thromboembolic events, but it cannot determine how specific that risk is to covid-19 compared with, say, other serious viral infections. However, it is generally accepted to carry a higher risk than influenza. The authors suggest that these acute cardiovascular complications of covid-19 warrant prioritisation of prevention strategies. I’m not sure this study makes prevention any more important than it was already, but for people and organisations who aren’t yet convinced perhaps this is important. I think the take-home message here is more that therapies could be targeted at preventing or reducing the impact of the cardiovascular complications from covid-19, and perhaps to be vigilant for the onset of myocardial infarctions and strokes.

Lancet doi:10.1016/S0140-6736(21)00896-5

Disruption to routine childhood vaccination
The pandemic has destroyed many good things for so long that it is easy to be apathetic about things we once considered essential and routine. One of the greatest (but not that much talked about) successes of the modern era is widespread childhood vaccination, such as for measles and diphtheria-tetanus-pertussis. Globally, millions of children have missed doses for these vaccines, leaving them under-vaccinated or unvaccinated against preventable diseases at the end of 2020. Children in north Africa and the Middle East were particularly affected. Causey and colleagues’ modelling study indicates that these gaps are likely to extend throughout 2021. Targeting resources to make up the missed doses is going to be key.

Lancet doi:10.1016/S0140-6736(21)01337-4

Covid-19 in vaccinated healthcare workers
If you wanted to research how protective vaccines are against infections, you’d do a randomised trial, wouldn’t you? You would randomise people to vaccine or placebo and see how many people got infected and how many severely so. That’s the most unbiased assessment. But there’s another piece of the puzzle—what makes people less likely to be protected by the vaccine? Bergwerk and colleagues’ case-control analysis of over 1000 fully vaccinated healthcare workers found that the infected workers (cases) had lower neutralising antibody titres than the controls (uninfected vaccinated workers). Reassuringly, most of the cases were mild or asymptomatic. Most cases (85%) involved the B.1.1.7 variant. I suppose this is also reassuring because at least the vaccine protected against what it was initially designed to protect against, and we just have to wait for the booster research.

N Engl J Med doi:10.1056/NEJMoa2109072

Alex Nowbar is a clinician at Royal Brompton Hospital.