Alex Nowbar reviews the latest research from the top medical journals

Lancet

A nod for nose surgery?

In this randomised controlled trial, septoplasty for correcting septal deviation was compared with non-surgical management. After one year, the surgical group had higher disease-specific quality of life scores and better airflow. However, both staff and patients knew which treatment the patient had been randomised to, so there is potential for bias. While it is a step in the right direction to be establishing evidence for a procedure, this was not a placebo controlled trial. No matter how well conducted and reported this study was, the design means the investigators have measured the effect of surgery, but not how much of this effect was placebo effect and how much actual effect.

Refining radiotherapy for prostate cancer

A large Scandanavian randomised controlled trial compared ultra-hypofractionated radiotherapy for prostate cancer with conventional fractionation. The ultra-hypofractionated group received the total dose over a much shorter time period, but the overall dose was lower. As expected the early side effects were worse in the ultra-hypofractionated group, while late side effects were similar. The trial was “positive” in that non-inferiority of ultra-hypofractionation was shown. This study is important not only for reducing the burden to patients of repeated hospital visits to receive radiotherapy, but also for resource management as it reduces the workload of the radiotherapy department. However, patients were only followed up for five years, so longer term inferiority is still possible.

Tenosynovial giant cell tumours

The ENLIVEN trial randomised 120 people with a tenosynovial giant cell tumour not amenable to surgery to pexidartinib or placebo for six months. This tumour can impact on physical function and get too big to operate on without the surgery causing functional impairment. Pexidartinib is a tyrosine kinase inhibitor targeting CSF1, which is a protein with a key role in this non-malignant, but locally aggressive and clinically troublesome tumour. There was no treatment response in the placebo group, but a response was seen in 39% of people in the pexidartinib group. While effective, the drug did display alarming adverse effects including hepatotoxicity. Perhaps with further study of optimal timing, duration, and liver monitoring strategies, this could be a clinically useful therapy.

Echo after MI

There are many things one might concern oneself with after a myocardial infarction but should an echo be one of them? Well, as with any test, it depends on whether the test changes management. And further than that, does it change outcomes? Randomising people to have an echo or not after a myocardial infarction seems unlikely. It’s not so expensive or invasive that it needs rationing or does it? Pack et al assessed inpatient mortality, length of stay, inpatient costs and 3-month readmission rates in 98,999 hospital admissions. Over 70% had at least one echo. The study is large, but observational so there are many factors that will have affected echo rates and not all can be accounted for. For example, sicker patients would be more likely to have an echo because there is a need to differentiate between causes and help guide management.

These patients would also have a higher mortality compared to those who did not have an echo, but that does not mean it’s safe to omit an echo. The strongest predictor of getting an echo identified in this study was which hospital the patient got admitted to. Trends between hospitals were assessed, for example, places more likely to perform echo also had longer stays and higher costs, but no difference in mortality or readmissions. However, the conclusion that echo could be used more selectively is not fully justified by these data. Outpatient echo and focussed handheld echo were also not considered.

Hip fracture in older women

Hip fractures go hand in hand with older age and frailty, making therapeutic decisions more challenging. How strong is the mandate for treatment to prevent fractures in women over 80 years old with osteoporosis? Ensrud et al investigated this in a large prospective cohort study. They followed 1528 treatment-naïve, community dwelling women with either osteoporosis or a high fracture risk without osteoporosis. Over five years, 8.8% of women had a hip fracture and 18.8% had died before experiencing a hip fracture. The risk of hip fracture was 18.1% in women with osteoporosis who also had more than three comorbidities and 46.7% in those with a poorer prognosis. Those without osteoporosis had far lower fracture rates in the five year period even though they were deemed to be at high risk of fracture. If the treatment benefits over age 80 are similar to those seen in randomised controlled trials in younger women, there is clearly a group of women (those with more comorbidities or poorer prognosis) who had a lot to gain. These observational data cannot tell us whether treatment would be beneficial in these groups, only that there is potential for gain due to the high risks. This study serves as a reminder that people should not be dismissed as not worth treating due to assumptions about age.

NEJM

Cardiac testing in stable coronary artery disease