Alex Nowbar reviews the latest research from the top medical journals

Non-cancer palliative care

Quinn and colleagues meta-analysed 28 randomised controlled trials of palliative care interventions to see if they reduced acute healthcare use, improved quality of life, and reduced symptom burden compared with usual care. Ten trials were in heart failure, four in dementia, three in chronic obstructive pulmonary disease, and the rest were in mixed disease. Palliative interventions were associated with less acute healthcare use and moderate reductions in symptom burden but no difference in quality of life. However, there was a quality of life benefit when trials at high risk of bias were excluded from the analysis. One would expect that less use of acute healthcare would be associated with better quality of life, but measuring quality of life is tricky. Measures get validated before use in trials, but how can they be validated when there is no gold standard? It is hard for people to assess their quality of life in a tick-box fashion because there are no points of reference.

JAMA doi:10.1001/jama.2020.14205

Clotting and lumbar punctures

It makes sense to check a patient’s clotting before an invasive procedure, especially one where compression would be difficult and a haematoma could have severe consequences (such as a spinal haematoma causing paraparesis). This large Danish cohort study compared rates of spinal haematoma after lumbar puncture in people with and without clotting abnormalities (defined as platelet counts <150×109/L, international normalised ratio >1.4, or activated partial thromboplastin time >39 seconds). The authors found that 0.20% of those without coagulopathy got a spinal haematoma compared with 0.23% of those with coagulopathy. This result is reassuring, but it comes with the caveat that doctors probably avoided performing lumbar punctures in people with more severe coagulopathies, so the risk hasn’t really been fully assessed.

JAMA doi:10.1001/jama.2020.14895

New device approvals in the US

Before a device, or indeed any intervention, is applied to people, we should be assured of its safety and effectiveness. The US Food and Drug Administration has a system called the De Novo Pathway for novel interventions. Johnston and colleagues analysed the clearance of devices via this pathway between 2011 and 2019. Sixty five devices were cleared in this period. None was life sustaining, and only 10 were implantable. Most had been through pivotal clinical trials (most of which were randomised), but a fifth were approved without. And a third were approved despite the trials not meeting their primary effectiveness endpoint. Post-market studies were rarely required. The bar for approval seems unacceptably inconsistent. This is worrying as these approvals go on to support the 510(k) process of approval—510(k) requires the device to be equivalent in safety and effectiveness to a legally marketed device.

JAMA Intern Med doi:10.1001/jamainternmed.2020.3214

Bariatric business

In a large Swedish observational study, bariatric surgery was associated with a three year increase in life expectancy compared with usual obesity care, but life expectancy was still 5.5 years lower than that of the general population. We already knew that bariatric surgery provides prognostic benefit, but this quantification of life expectancy is useful for conceptualising and discussing this benefit. Years of life gained can be quite a persuasive argument for a therapy, but we should also be cautious because these data are from a cohort, not a randomised study. This means that bariatric surgery is not the only factor contributing to the difference in prognosis. There is likely to be some confounding here, as there are systematic differences between people who undergo surgery and those who don’t, such as lack of fitness for surgery. So, nice numbers, but they have to be taken with a pinch of salt.

N Engl J Med doi:10.1056/NEJMoa2002449

Promise in a Janus kinase inhibitor

Newcomer to the field of rheumatoid arthritis is the Janus kinase inhibitor upadacitinib. Rubbert-Roth and colleagues tested it in a six month, double-blind randomised trial looking for non-inferiority to abatacept (a CTLA-4 inhibitor). All patients received conventional disease modifying anti-rheumatic drugs. Upadacitinib was found to be not only non-inferior but also superior in its reduction in disease activity score (the primary endpoint) and in achievement of clinical remission, which is highly impressive. The major caveat is that there were more serious adverse events with upadacitinib, so it is probably not that viable. The authors suggest longer and larger trials, but these will be expensive and challenging given the results of this trial.

N Engl J Med doi:10.1056/NEJMoa2008250

Alex Nowbar is a clinical research fellow at Imperial College London.