Ann Robinson reviews the latest research from the top medical journals

NEJM

GABA modulation in depression⁠—will it work?

The inhibitory neurotransmitter γ-aminobutyric acid (GABA) may play a role in clinical depression. This small, double blind, phase 2 trial asked whether a new drug, SAGE-217, that modulates GABA receptors is effective and safe in treating moderate to severe depression compared with placebo. Short term results were encouraging. Depression, as measured by the Hamilton Depression Rating Scale, improved more in the SAGE-217 group than the placebo group at day 15. There were no serious adverse effects, although side effects of headache, dizziness, nausea, and drowsiness were more common in the SAGE-217 group. Bigger, longer studies including head to head comparison with existing antidepressants are needed. It’s too early to say whether SAGE-217 will live up to its promise.

JAMA

Warfarin after arthroplasty⁠—a Gordian knot 

How intense does anticoagulation before elective hip or knee replacement (arthroplasty) need to be to prevent venous thromboembolism (VTE) within two months, or death within a month of surgery? In this randomized clinical trial, the rate of VTE or death did not meet the criterion for noninferiority in the low-intensity group (INR 1.8, rate of VTE or death 5.1%) compared with the standard-intensity group (INR 2.5, rate of VTE or death 3.8%). Interestingly, low-intensity warfarin didn’t lower the rate of bleeding. The rates of VTE or death strike me as high, but the authors say the rate was actually lower than expected. They comment that the study may have been underpowered to establish noninferiority. This trial used a web application to guide warfarin dosing algorithms and randomized half of the patients to genotype-guided warfarin dosing—both may have enhanced safety. The tools may be new, but the problem is age-old—how to minimise the risk of thrombosis without increasing the risk of bleeding.

JAMA Internal medicine

Time to ditch soft drinks?

A well publicised, population based, cohort study of nearly half a million people from 10 European countries found that greater consumption of total, sugar sweetened, and artificially sweetened soft drinks was associated with a higher risk of all-cause mortality. Consumption of artificially sweetened soft drinks was positively associated with deaths from circulatory diseases, and sugar sweetened soft drinks were associated with deaths from digestive diseases. There was no increased cancer mortality. The authors say that: “Results of this study appear to support ongoing public health measures to reduce the consumption of soft drinks.” This type of study can’t demonstrate causality; there may be residual confounding factors, and there was only a single assessment of soft drink consumption at baseline. Artificial sweeteners may induce glucose intolerance, but the reasons for the association with circulatory diseases is unclear.

Lancet

Inebilizumab for treating neuromyelitis optica spectrum disorder

Neuromyelitis optica spectrum disorder (NMOSD) causes episodes of optic neuritis, and transverse myelitis. Once considered a variant of multiple sclerosis, it is now viewed as a distinct autoimmune, inflammatory CNS disorder. We don’t have much to offer patients with NMOSD. Eculizimab has recently received FDA approval for reducing relapse in NMOSD, but only for a subset of patients. This randomised, placebo controlled, phase 2/3 study found that inebilizumab (a drug that depletes CD19 B lymphocytes) increased the expected attack-free interval compared with placebo (12% v 39% had an attack during the study), with a number needed to treat of 3.73. Disappointingly, the severity and recovery from attacks of optic neuritis appeared unchanged. “A high unmet medical need for proven therapies exists in NMOSD,” say the authors. Inebilizumab is of limited effectiveness, but, together with other emerging developmental drugs such as satralizumab, it offers a possible way forward.

Annals of Internal Medicine

Microplastics in human stool

The ubiquity of microplastics is of huge environmental concern and we know that they’ve entered the food chain. This tiny study⁠—the first of its kind—is published now, but was publicised in The Guardian a year ago. Eight people in Europe and Asia kept a food diary and gave one stool sample to researchers who looked for particles of microplastics which are widely found in food and the gastrointestinal tract of marine animals. All eight samples tested positive for nine types of plastic. Adults excrete around 100g of stool a day and in this study researchers found 20 microplastic particles/10g of faecal matter. There were no data on the source or impact of the microplastics, but the food diaries showed that six of the eight ate seafood, seven drank from plastic bottles and three used cosmetic polymers. The key questions are: Why would anyone do a study with only eight people in it? Are the particles of plastics absorbed in the gastrointestinal tract (especially in those with increased gut permeability due to inflammatory bowel disease) and, most importantly,  what, if any, damage do they cause?

Antipsychotics don’t treat delirium…

Delirium is an acute attention and cognition disorder caused by an underlying medical problem, and it affects 1 in 5 hospital inpatients. It’s associated with bad outcomes such as longer hospital stays, long term cognitive decline, and increased mortality. There is no single drug approved for treating delirium, and this systematic review found that current evidence does not support routine use of haloperidol or second generation antipsychotics. The drugs cause more cardiac side effects, but no excess neurological harm. There’s a distinct lack of evidence for older people and palliative care patients. Despite some limitations⁠—heterogeneity of severity and cause of delirium, drug type and regimen, and definition of outcomes⁠—this review supports previous published studies and guidelines.

…And antipsychotics don’t prevent delirium

Prevention of delirium is key, especially given that treatment options are so limited and the symptoms so distressing for patients, families, staff, and other hospital inpatients. Algorithms or old fashioned clinical nouse can help to identify those at risk of delirium, but then what? Antipsychotics are used to prevent delirium, but this systematic review of 14 randomised controlled trials found that they don’t work; there were no differences in incidence of delirium, length of hospital stay, and mortality between haloperidol and placebo. There was limited evidence that second-generation antipsychotics may lower the risk of delirium postoperatively. Further research, including standardisation of outcome measures, is sorely needed into both prevention and treatment of delirium. 

Ann Robinson is an NHS GP and health writer and broadcaster

Correction: This article was corrected on 24 September 2019 to correct a factual error.