Ann Robinson reviews the latest research from the top medical journals

Patient centred intensive care

This important US prospective quality improvement study looked at the outcomes of training staff to use time-limited trials (TLTs) as the default communication and care planning approach for critically ill patients in intensive care units (ICU). It found that the use of TLTs was associated with significant reductions in duration of ICU stay (7.4 v 8.7 days), fewer invasive procedures, and no change in mortality (58%) or family satisfaction rating. TLTs involve detailed discussions of patients’ preferences for care and an agreement between clinicians and patients or their families that therapies will be tried for a specified time, then evaluated and stopped if there are no objective signs of improvement. It’s disconcerting to learn that this rational approach isn’t standard care. The study designs may have allowed for some bias in patient selection, and the multifaceted approach makes it hard to separate out which changes produced the positive results. Yet, overall, the quest to minimise suffering in an ICU setting is to be applauded.  

JAMA Intern Med doi:10.1001/jamainternmed.2021.1000

Patient, heal thyself

A randomised trial of 206 adults with a clinical diagnosis of knee osteoarthritis found that a 24 week, web based programme of specific strengthening exercises and guidance about physical activity, bolstered by automated text messages to encourage adherence, improved self reported knee pain and function at 24 weeks compared with a control group who were directed to a website with general information only. Given that knee osteoarthritis is a common problem with no known cure (apart from knee replacement presumably), a digital programme that can dispense education and exercises without the involvement of a human being is an attractive idea. The level of engagement was unusually high in this study—perhaps because of the content and frequency of text messaging (average of 2.5 texts/week.) The study was conducted among a self sufficient, educated, and tech savvy group so may not be generalisable to the wider public, and there was no follow-up beyond 24 weeks so we have no way of knowing whether the benefits were sustained.

JAMA Intern Med doi:10.1001/jamainternmed.2021.0991

Bad news for levetiracetam in epilepsy

Marson and colleagues report the results of SANAD II; two useful pragmatic, open-label, randomised trials of treatment options in epilepsy. In the first, in patients with newly diagnosed focal epilepsy, levetiracetam did not meet non-inferiority criteria in the intention-to-treat analysis of time to 12-month remission from seizures. A major limitation of this study is the low rate of under 18 year olds (<18%). Lamotrigine should remain as first line treatment, rather than levetiracetam or zonisamide (both treatments are licensed for monotherapy), which are less effective and more likely to cause depression and anxiety, say the authors. Further studies are needed now to assess the clinical and cost effectiveness of other anti-seizure medications (such as lacosamide, brivaracetam, and perampanel).

In the second study, of patients with newly diagnosed generalised epilepsy, levetiracetam didn’t fare well either. It didn’t meet criteria for non-inferiority in the intention-to-treat analysis of time to 12-month remission, and in the per-protocol analysis valproate was more effective than levetiracetam for all outcomes. This is a problem for women of childbearing age who need to avoid valproate for its teratogenicity.

Lancet doi:10.1016/S0140-6736(21)00247-6

Lancet doi:10.1016/S0140-6736(21)00246-4

Liver shortage

Liver donors are in short supply, so even though the incidence of non-anastomotic biliary strictures (caused by ischaemia-reperfusion injury) is three times higher in livers obtained from donors after circulatory death than brain death, donated livers from the former are increasingly used. This study found that there was a lower risk of non-anastomotic biliary strictures within six months when donated livers were kept on hypothermic oxygenated machine perfusion rather than conventional cold storage (6% v 18%). Machine perfusion didn’t affect patient or graft survival; outcomes are generally good after liver transplantation, so larger trials would be needed to show an effect. Further work into the cost-benefit of machine perfusion is needed.

N Engl J Med doi:10.1056/NEJMoa2031532

Ivermectin: no evidence of effectiveness against covid-19

Ivermectin, an antiparasitic drug with antiviral activity, is widely used in the treatment of mild covid-19, but this randomised clinical trial found that a five day course didn’t significantly change the duration of symptoms compared with placebo (median time to resolution of symptoms 10 v 12 days). Most people had recovered within 21 days (82% v 79%). Clinical deterioration rates were so low in this young and healthy population (<3%) that the study design had to be changed six weeks after it started. Larger trials may find that ivermectin has other benefits, but there seems to be little justification for its use at the moment.

JAMA doi:10.1001/jama.2021.3071

Ann Robinson is an NHS GP and health writer and broadcaster