Ann Robinson’s research reviews—22 January 2021

Ann Robinson reviews the latest research from the top medical journals

Appendicitis: staying out of hospital

Antibiotics are an effective and safe alternative to appendicectomy in cases of CT-confirmed acute appendicitis, but the best regimen isn’t yet clear. This Finnish study found that either one week of oral antibiotic (moxifloxacin) or two days of intravenous antibiotic (ertapenem) followed by five days of levofloxacin and metronidazole met the prespecified threshold for treatment success—discharge from hospital without needing surgery and no recurrence within one year. The oral treatment is more practical to deliver, and moxifloxacin is broad spectrum, but runs the risk of antibiotic resistance. Unfortunately, the oral option failed to demonstrate non-inferiority in this trial. The World Society of Emergency 2020 guideline suggests offering antibiotics as a safe alternative to surgery in uncomplicated acute appendicitis without appendicolith, and this has, for obvious reasons, been more widely adopted during the pandemic.

JAMA doi:10.1001/jama.2020.23525

Saliva as good as a swab up the nose?

Nasopharyngeal swabbing, done properly, is unpleasant; spitting into a tube is much easier and more acceptable to patients. So this study comparing saliva with nasopharyngeal nucleic acid amplification testing (NAAT)—a generic term covering techniques such as PCR—in testing for covid-19 is timely, and the results are welcome. The systematic review and latent class meta-analysis found that saliva NAAT had a similar sensitivity (83.2% v 84.8%) and specificity (99.2% v 98.9%) to that of nasopharyngeal NAAT. The findings seem robust despite significant heterogeneity among the included studies. However, an important caveat is that there were little data about patients’ symptoms or how ill they were, so it remains unclear whether the two sampling methods are equally good in all circumstances.

JAMA Intern Med doi:10.1001/jamainternmed.2020.8876

The quest for heart failure drugs that work

There’s a gap in the market for treatments that boost systolic function and improve outcomes in patients with heart failure and reduced ejection fraction. Omecamtiv mercabil, a novel selective cardiac myosin activator, is already known to improve cardiac function. This study claimed a lower incidence in a composite of heart failure event or death from a cardiovascular cause among the patients receiving omecamtiv mercabil than among those taking placebo. But the results were actually quite disappointing; there was no significant difference in the death rate from cardiovascular causes (19.6% v 19.4%) or the frequency of ischaemic or ventricular arrhythmia events. There was only a modest fall in primary outcome events—a composite of first heart failure event requiring hospitalisation or an urgent visit or death from cardiovascular cause (37.0% v 39.1%). The study was limited to patients under 85 years old in a stable clinical condition; so not fully representative of the patient population.   

N Engl J Med doi:10.1056/NEJMoa2025797

Machine learning for predicting adverse events

Devising an individualised management strategy after an acute coronary syndrome relies on accurate prediction of ischaemic and bleeding events. But current prediction tools are often not accurate enough at an individual level. This study found that a risk stratification model based on machine learning (the PRAISE score) performed well in predicting all-cause death, recurrent acute myocardial infarction, and major bleeding after acute coronary syndrome. The days of one treatment size fits all are receding. 

Lancet doi:10.1016/S0140-6736(20)32519-8

If the shoe fits

Walking is the only activity left for many of us in these pandemic days. So pity those whose enjoyment of walking is hampered by osteoarthritis of the knee. This study asked whether wearing flat, flexible shoes is preferable to stable, supportive shoes as some previous evidence has suggested. This was clearly a highly motivated group of participants; an impressive 98% completed the six month intervention of wearing flat flexible or stable supportive shoes for at least six hours a day. No evidence was found that one type of shoe is better than the other. Stable, supportive ones were associated with some modest reduction in pain, knee-related quality of life, and hip pain on the same side, but there was no difference in physical function. There was no control group who wore “usual shoes,” and the results aren’t necessarily generalisable to a wider population. The type of shoes probably matters less than finding a comfy pair.

Ann Intern Med doi:10.7326/M20-6321

Ann Robinson is an NHS GP and health writer and broadcaster