Ann Robinson’s research reviews—22 October 2021

Carbon dioxide laser on postmenopausal vaginal symptoms

This small but interesting Australian study found that treatment with fractional carbon dioxide laser did not reduce the severity of self rated postmenopausal vulvovaginal symptoms such as vaginal dryness or improve quality of life or vaginal histology compared with sham treatment after 12 months (change in score from baseline -17.2 v -26.6 on a visual analogue scale and -3.1 v -1.6 on the Vulvovaginal Symptom Questionnaire score). These results contrasted markedly with previous observational studies that suggested a 75-100% reduction in symptoms for women treated with vaginal laser. The device was widely adopted across the world on fairly weak evidence, prompting a warning by the US Food and Drug Administration in 2018 that robust evidence of efficacy and safety were still lacking. This study was small and imperfect (for example, minor vaginal trauma caused by the probe could have prompted healing which helped symptoms even when sham treatment was used). Bigger and better studies are much needed, as it’s a common and distressing problem with few effective options.

JAMA doi:10.1001/jama.2021.14892

Covid immunity and infection risk for household members

What is the impact of individuals who are immune to covid-19 on the risk of infection among non-immune family members? This large Swedish cohort study (nearly 1.8 million people from 0.8 million households) found that the risk of non-immune family members contracting covid-19 reduced as the number of immune family members increased (by 45-61% with one immune family member, 75-86% with two, and 91-94% with three). There may have been some other variables to account for the results, but it’s reassuring for pandemic control that having a family member with immunity (from a previous infection or from full vaccination) has a protective impact on non-immune family members. The authors rightly caution that single doses of vaccine may be less effective against emerging highly transmissible variants.

JAMA Intern Med doi:10.1001/jamainternmed.2021.5814

Statins: good for the heart, bad for diabetes?

Are statins good for diabetic people? There’s a dichotomy: US guidelines recommend statins for all diabetics aged 40-75 years with an LDL cholesterol level >70 mg/dL (1.8 mmol/L) to reduce cardiovascular risk, but statins themselves have been associated with insulin resistance and higher blood glucose levels, which increase cardiovascular risk. This 12 year US retrospective cohort study of 83 022 matched pairs of statin users and non-users found that, among those diagnosed with diabetes during the study period, initiation of statins was associated with a worsening of diabetes (for example, greater likelihood of needing to start taking insulin or significant hyperglycaemia and complications). The odds of diabetes progression among statin users versus non-users were 1.83, 1.55, and 1.45 for high, moderate, and low intensity cholesterol lowering, respectively. Tight lipid control with statins needs to be weighed against the risk of diabetes progression; one size fits all is too crude.

JAMA Intern Med doi:10.1001/jamainternmed.2021.5714

Empagliflozin for heart failure

Sodium-glucose cotransporter 2 inhibitors such as empagliflozin are known to reduce the risk of hospitalisation in patients with heart failure and a reduced ejection fraction, but do they have a similar effect if ejection fraction is preserved? This randomised trial conducted over 26 months found that, in addition to usual care, treatment with empagliflozin reduced hospitalisation for heart failure compared with placebo (8.6% v 11.8%) but had little impact on cardiovascular deaths (7.3% v 8.2%). Thirty one patients needed to be treated with empagliflozin to prevent one primary outcome event (hospitalisation or death). There was a high drop-out rate (23%) across both groups, which may have skewed results to an extent.

N Engl J Med doi:10.1056/NEJMoa2107038

Most children with chest infections don’t benefit from antibiotics

A well conducted randomised controlled trial across 56 general practices in England found that children aged 6 months to 12 years with acute uncomplicated lower respiratory tract infection (LRTI) fared no better after a seven day course of amoxicillin than with placebo. There was no difference in primary outcome (duration of symptoms rated moderately bad or worse (measured with a validated diary) for up to 28 days). This held true irrespective of whether the clinician was using chest signs, fever, assessment of how unwell the child was, or the presence of sputum, chest rattle, or breathlessness as their reason to prescribe antibiotics. The study’s authors are clear that “unless pneumonia is suspected, clinicians should provide safety-netting advice but not prescribe antibiotics for most children presenting with chest infections.” If anything, antibiotics are even less likely to be effective in real life practice, as the children recruited to the trial were more unwell than is usual in a low risk population.

Lancet doi:10.1016/S0140-6736(21)01431-8

Ann Robinson is an NHS GP and health writer and broadcaster