On the Pulse - 12th June 2009
Preventing DVT after stroke
Compression stockings are effective at preventing deep vein thrombosis after surgery, and they are also recommended after stroke by national guidelines, despite a lack of direct evidence. Results published in the Lancet suggest that this advice is quite wrong. In an outcome-blinded multicentre trial in patients admitted with acute stroke, thigh-length compression stockings made no significant difference to the incidence of DVT, and increased the frequency of skin breaks, ulcers, blisters, and skin necrosis. An accompanying Comment suggests it may be time to re-evaluate their use in other medical contexts too.
Waiting times for carotid endarterectomy
According to recently published NICE guidelines, carotid endarterectomy should be performed within 2 weeks of symptoms in appropriate patients. Unfortunately, a study in the BMJ shows this is far from the case in the UK. Of 5513 patients undergoing carotid endarterectomy from December 2005 to December 2007, only 20% had their operation within 2 weeks of symptom onset, and 30% waited more than 12 weeks; some of course may never have made it to the operating table. The authors reckon improvements in these figures will require such surgery to take precedence over elective cases, and improved availability of imaging and vascular anaesthesia.
The immunology of granulomas
Our immune system is usually good at keeping us free of illness, even though we are constantly exposed to infectious agents. But sometimes it over-reacts and causes disease instead of preventing it: asthma, anaphylaxis and auto-immune conditions are obvious examples. Sometimes too, the immune response is maladaptive and gets outwitted by micro-organisms. An article in the NEJM explains recent experiments (in zebra fish!) that cast doubt on the long-held view that the cellular aggregates known as granulomas limit the growth of mycobacteria. On the contrary, they seem to provide a niche for further growth of the pathogen, allowing egress of infected cells to produce new granulomas.
Early infant growth
Back in March, an analysis of data from the Avon Longitudinal Study of Parents and Children showed that rapid weight gain in the first 9 months of life was associated with greater BMI and adiposity when the child reached the age of 10 years. A longitudinal study in JAMA now suggests that it’s the first 3 months of post-natal life that really count. Levels of risk factors for cardiovascular disease and diabetes, including insulin sensitivity, high-density lipoprotein cholesterol and triglyceride concentrations, and waist circumference, measured in early adulthood, tended to be worse in those who had gained the most weight in the first months of infancy.
Extreme premature birth
A study from Sweden published in JAMA examines outcomes in children born before 27 weeks of gestation from 2004 to 2007. Overall perinatal mortality was 45%, varying from 15% at 26 weeks to 90% at 22 weeks. Of children born alive, a remarkable 70% survived to 1 year, and of these, 45% had no major neonatal morbidity. Perinatal interventions such as tocolytic treatment, antenatal corticosteroids, intubation at birth, and surfactant use were associated with survival, suggesting that the substantial improvement in these figures since the 1990s is mostly due to improved perinatal medical care.
HPV vaccination after age 24
Earlier studies proved that HPV vaccination was effective in women aged 16–26 years. A trial in the Lancet now demonstrates its efficacy in those aged 24–45 years. At 26 months, analysed per-protocol, disease or infection related to HPV 6, 11, 16, and 18 lasting 6 months or more was seen in 41/1607 in the placebo group compared with only 4/1615 in the vaccine group. However, as an accompanying Comment explains, whether mass vaccination is cost effective will depend greatly on how long protection lasts.
Reducing antibiotic prescriptions
Les antibiotiques c'est pas automatique (Antibiotics are not automatic) is the name of a campaign launched in France in 2002 to persuade doctors to reduce the prescription of antibiotics and explain to the general public that these drugs are often unnecessary. According to an evaluation in PLoS Medicine, it worked. Between 2002 and 2007, there was a fall of more than 25% in winter antibiotic prescriptions, which was consistent across all age groups for all the main classes of antibiotic. What amazed Dr Lydgate (who hasn’t needed an antibiotic since a dose of clap in his student days) was the staggeringly high level of prescribing before the campaign: an average of 72 prescriptions per 100 people each winter. He wonders what the figures are for the UK.
Plain English for statistical findings of trials
A comment in the Lancet tries to find a way of translating the statistical paraphernalia that characterise reporting of randomised clinical trials into succinct everyday English. It identifies six scenarios for the possible primary outcomes of such trials, according to the size of the treatment effect, the width of the associated confidence interval and the accompanying p value, and proposes that these can be summed up in simple phrases such as “...is superior to” or “... seems not superior to”. The authors admit that these phrases won’t capture every nuance of the data, and that they fail to cover the clinical importance of the findings. Even so, it’s a welcome step towards clarity.