What’s new in the other journals?

Screening mammography is controversial. In populations where there are screening programmes, it provides early detection of cancers and is quoted as reducing mortality by 25%. These gains must be weighed against the losses of expense, radiation and false-positive diagnoses with their accompanying anxiety and investigations. Screening programmes are by no means universal, with only developed countries being able to afford them. These are countries where women live longer, radiology facilities are available and certain lifestyles increase breast cancer risks—such as obesity, sedentary habits and combined hormonal therapy. One would have thought that, given the profile of breast cancer, the radiological techniques used and the quality of reporting would be improving because of upgraded technology and image interpretation. However, there seem doubts about both these aspects of mammography. From the radiologists’ point of view, mammography is unattractive. The equipment is expensive and only justified where large volumes are dealt with. It is not a lucrative part of radiological practice, often requiring cross-subsidisation in the private and public sectors. It is heavily regulated, with mandated procedural requirements that are stressful to staff who are more and more called upon to perform image-guided needle biopsies and interpret results. In the USA, screening mammography is the most common basis for lawsuits in radiology. For these reasons, it is the weakest of the imaging subspecialities with a shortfall of trained mammographers, ‘many of whom are older and overworked’ (Hall, N Engl J Med 2007;356:1464–6). It was hoped that two recent advances would enhance breast imaging; digital mammography and computer-aided detection. Digital mammograms are expensive, and only digitised images can be used by computer-aided detection software, whereas standard analogue films have to be digitised first. But the basic question still has to be answered—does computer-aided detection improve results? In an article destined to disappoint the mammography world, it appears not—according to Fenton et al. (N Engl J Med 2007;356:1399–409) from the USA. They looked at over 400 000 mammograms across 43 states with and without computer-aided detection. It was found that the specificity of diagnoses decreased when computer-aided detection was introduced. The positive predictive value deceased while biopsy rates increased by 20%. There was no change in cancer detection rates before and after computer-aided detection implementation. The Food and Drug Administration approved computer-aided detection technology nearly 10 years ago, and expensive hardware and software are selling well (digital mammography comprising 40% of new units sold a few years ago). No randomised trials have proved the new technique superior to existing methods, and the profession is left wondering whether computer-aided detection is not simply a failed experiment. Human papillomavirus types 16 and 18 are responsible for most cervical cancers and types 6 and 11 are responsible for most genital warts. Gardasil®, Merck’s vaccine against all four types was licensed last year, and Cervarix® from GlaxoSmithKline is due to be approved soon. In the USA, it is up to each state to decide the terms of introducing the vaccine, and at present, 20 states are debating legislation that would make vaccination a requirement for school attendance. The prevailing mood is for 11–12-year-old girls to receive inoculation unless their parents object on social or religious grounds (Coombes, BMJ 2007;334:721–3). Right-wing views suggesting vaccination will somehow endorse sexual experimentation are countered by remarks that a vaccine is unlikely to change a woman’s decision to have sex or not. Australia, Canada, Mexico, New Zealand and European countries have all moved to adopt mass vaccination, but the UK seems stalled at the cost implications—the 3-injection course costing £180. Let us hope that the global community rallies to assist developing countries obtain and administer the vaccines as they bear the greatest burden of the disease. Mother-to-child HIV transmission occurs in 30–40% of deliveries when there is no intervention. The most common time of transmission is intrapartum and various strategies of antiretroviral drug administration antenatally and peripartum reduce MTCT rates dramatically. The most effective regimens are highly active antiretroviral therapy (HAART) started pre-pregnancy, followed by various timings of HAART introduced during pregnancy—generally the earlier the better. Next come combination antiretrovirals in the third trimester, then triple, dual and single treatments at delivery. In developing countries, there are often late bookings, ignorance about HIV status, a lack of healthcare staff and a scarcity of antiretroviral drugs. In these circumstances, advice about breastfeeding is crucial. Formula feeding would be ideal, but it is impractical for a variety of reasons—social and financial. In any event, the risk of diarrhoeal diseases in these settings is so high as to negate any benefits of formula feeding. Therefore, breastfeeding has to be an option for HIV-positive mothers in these circumstances, but the next question is whether exclusive breastfeeding or mixed feeding is better—and for how long. To answer this, Coovadia et al. from South Africa encouraged exclusive breastfeeding in a large group of HIV-infected mothers who chose not to exclusively formula feed (Lancet 2007;369:107–16). The overall MTCT rates were 15–20% at the end of 6 months, but the breakdown of these figures was interesting. Those acquiring infection from breastfeeding were more likely to be those born below a birthweight of 2500 g or whose mothers had low CD4 counts. Infants who were born HIV negative and were exclusively breastfed had only a 4% conversion rate at 6 months of age. Infants given a mixture of breast and formula feeding had nearly double this conversion rate, and those given breast and solid feeds had a ten times greater conversion rate. It seems the infantile gut is more prone to virus penetration if breast milk plus foreign proteins are ingested, or perhaps mothers with higher viral loads are unable to exclusively breastfeed because of poorer health or breast infections. The statistics for mortality were equally stark. At 3 months, 6% of the exclusively breastfed infants had died compared with 15% of those receiving mixed feeding. It seems clear that HIV-positive mothers should avoid breastfeeding where formula feeding is ‘acceptable, feasible, affordable, sustainable and safe’; otherwise exclusive breastfeeding is recommended for the first few months of life. Most regions with high HIV prevalence rates are resource-poor. They are also short of education about primary HIV prevention, testing and open discussion on HIV/AIDS. Homes and Savage (Lancet 2007;369:1065–6) could well be right in suggesting that promoting exclusive breastfeeding may save infant lives and allow more dissemination of information about all forms of HIV transmission. Recently, the American College of Obstetricians and Gynecologists held its annual meeting, and presentations from it were published in a supplement to Obstet Gynecol (2007;109:4). Summaries of some of the papers are as follows: Natural pregnancy rates are about 20% in a single normal cycle and 80% after a year in average population studies. Cronin et al. (p3S) followed more than 2000 women stopping oral contraceptives (OCs) to become pregnant and found that their conception rates were not different from those who has not been taking OCs. The data were similar for those who had been taking OCs for 2 years or longer. Serious adverse effects of OCs are rare, and therefore, it takes large studies to quantify their risk. Dinger et al. (p5S) followed up European women taking OCs for the equivalent of 140 000 women-years and found that serious adverse effects had an incidence similar to nonusers. Three months after stopping OCs, the risk ratio was 1.5 times greater because of the adverse effects of pregnancy, so OCs can be considered to have a beneficial influence on the public health of women. Modern oral contraceptive use is associated with a low absolute risk of venous thromboembolism—just over 1 event per 1000 women-years. Seeger et al. (p59S) observed a large cohort of women starting routine low-dose OC use and compared them with women starting an OC containing 30 micrograms ethinyl estradiol plus 3 mg drospironone (a novel progestin). The venous thromboembolic event rates were 1.4 for regular OCs and 1.2 for the newer preparation substantiating the claims for low incidence. Smoking is bad for mothers and babies. Living or working where others smoke is bad for your health, but will it kill you? It appears so, according to Hill et al. from New Zealand who looked at disease and mortality in those who did not smoke but lived with someone who did (Am J Epidemiol 2007;165:530–40). Passive smokers had higher rates of cardiovascular and respiratory disease than people not living with smokers, and mortality rates were raised, irrespective of social status, income or where they lived. The authors feel that they have sufficient evidence to conclude the increase in mortality was caused by other’s smoke. You have been warned. In most countries, the law says you do not have to declare your HIV status to anyone. This is to rightfully protect people living with HIV, but it does not apply if you intend having sex with someone. An HIV-infected person in Scotland found this out to his cost when he did not divulge his HIV-positive status to his lover and did not practice safe sex, thus infecting her (News, BMJ 2007;334:766). He also passed on hepatitis C to her and now faces 9 years in jail. Fetal sex testing is being offered commercially in Germany during the first trimester. According to Tuffs (BMJ 2007;334:712), a woman can get her doctor to send 2 ml of blood to a laboratory, which will determine the fetus’s sex with 99% accuracy by searching the maternal plasma for fetal DNA, then looking for Y chromosome material. Clinicians are concerned that the service, which costs £100 could be used to precipitate abortions if the sex of the fetus is not that desired by the parents. Early sex determination has very limited value in diagnosing inherited disorders. Making such tests available outside of academic units sounds suspiciously like making money out of people who may have nefarious reasons for wanting to know their baby’s sex.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.