Ann Robinson reviews the latest research from the top medical journals
The challenge of change
Residents of nursing homes who are living with advanced dementia are often prescribed antibiotics for possible urinary or chest infections despite a lack of evidence of infection or a clear rationale. This pragmatic cluster trial of 426 nursing home residents with advanced dementia found that a multicomponent intervention reduced the use of chest x rays compared with the control arm, but otherwise had no significant impact on management. Healthcare professionals were given pocket cards with communication tips, management flowcharts, feedback on their prescribing, in-person seminars, and online courses. Relatives of patients got a booklet. All this effort was associated with a non-significant reduction in antibiotic prescribing and no change in unwarranted initiation of antibiotics, catheterisations, blood tests, and hospital admissions. “Comfort” is the most common goal of treatment for this group of people, and intrusive testing and treatment can cause discomfort and harm. But this study shows just how hard it is to change entrenched medical practice.
JAMA Intern Med doi:10.1001/jamainternmed.2021.3098
mRNA covid-19 vaccines work well in patients with cirrhosis
Do covid-19 vaccines work among patients with cirrhosis, who often have impaired immune responses to vaccines and who were excluded from phase III clinical trials? This retrospective cohort study of over 20 000 US veterans with cirrhosis compared with a propensity-matched control group found that the vaccines worked well: a single dose of the Pfizer or Moderna (mRNA) vaccine was associated with a 64.8% reduction in covid infections and a 100% reduction in covid related hospitalisation or death after 28 days in both compensated and, to a lesser extent, decompensated cirrhosis. The second jab was associated with a 78.6% reduction in infections and 100% reduction in covid hospitalisation or death after seven days. The vaccinated and unvaccinated controls may not have been as well matched as the researchers had hoped, especially in terms of their covid exposure risk and frequency of PCR testing. But the timing of the study was good; it was done when there were still comparable numbers of vaccinated people and controls, whereas now it would be a struggle to find enough unvaccinated controls.
JAMA Intern Med doi:10.1001/jamainternmed.2021.4325
Room for optimism in chronic GVHD?
Chronic graft versus host disease (GVHD) is a major and distressing complication after allogeneic stem cell transplantation. Steroids (glucocorticoids) are usually prescribed, but either stop working or require constant use (with the attendant side effects) in around half of all cases. Good evidence about alternatives is lacking. This phase III open-label randomised trial into ruxolitinib—a protein kinase inhibitor targeting Janus kinase subtypes 1 and 2 that is used in myeloproliferative disorders and has been found to have some effect against GVHD—is welcome. Overall response at week 24 was greater in the ruxolitinib group than in the control group. The ruxolitinib group had longer median failure-free survival (>18.6 months v 5.7 months). The most common adverse effect was thrombocytopenia (15.2% in the ruxolitinib group, 10.1% in the control group) and anaemia (12.7% and 7.6% respectively). The incidence of cytomegalovirus infections and reactivations was similar in the two groups. More work is needed, and ruxolitinib is unlikely to be a panacea (hard-to-treat GVHD in liver and lungs showed little response), but it may offer some respite for patients who have already been through so much.
N Engl J Med doi:10.1056/NEJMoa2033122
Safe delivery of small babies
Is iatrogenic delivery (early induction of labour or caesarean before labour) warranted in fetal growth restriction? It’s tricky because the aim is to minimise the increased risk of stillbirth associated with fetal growth restriction against the known harms of early delivery such as neurodevelopmental and educational delay. In addition, detection of fetal growth restriction is far from perfect, with over half of suspected cases turning out to have normal growth. This interesting, large, exploratory Australian retrospective cohort study found that infants who were severely small for gestational age, with a birth weight below the third centile, who were appropriately delivered early for suspected fetal growth restriction had poorer developmental outcomes when they started primary school. They had poorer educational outcomes compared with children with the same degree of growth restriction at birth who had not been suspected of having fetal growth restriction. This study raises lots of questions: data on maternal factors that may affect a child’s development (such as smoking, alcohol use, drug use, and obesity) were lacking, as were details about the causes of being small for gestational age. But the finding that such infants who were delivered early had poorer school outcomes suggests that iatrogenic prematurity may be harmful in itself. Delaying delivery to 38 weeks and early childhood interventions to improve neurocognitive function in this high risk group could help.
Does anyone still think that remdesivir or hydroxychloroquine is particularly helpful in the treatment of covid-19? The WHO Solidarity trial showed no effect on mortality, and this Norwegian add-on (NOR-Solidarity), which included biobank data and three months of clinical follow-up, found no significant differences in mortality, degree of respiratory failure, or inflammatory markers between those given remdesivir, hydroxychloroquine, or standard care. There was no placebo arm, and numbers were small (a total of 185 participants), but it surely represents yet another nail in this particular coffin.
Ann Intern Med doi:10.7326/M21-0653
Ann Robinson is an NHS GP and health writer and broadcaster