Highlights from the literature
In many countries, especially in sub-Saharan Africa, Millennium Development Goal 4 is unlikely to be achieved. In 2010 there were 7.6 million deaths of children under 5 years of age. The problems are such that a multifaceted approach is needed. The Millennium Villages project, in nine countries in sub- Saharan Africa, provides investments in agriculture, the environment, business development, education, infrastructure, and health (Lancet 2012;379:2179–88; see also Comment, ibid: 2131–3). Village sites selected in East and West Africa included, on average, around 35 000 people with high levels of poverty and undernutrition. The projected cost, starting in 2006, was US $120 per person per year. The actual average spending was $25 at baseline and $116 by year 3 ($25 on health). There were reductions in poverty, food insecurity, stunting, and malaria prevalence in the Millennium Village sites after 3 years. Access to improved water and sanitation was increased and more provision of many maternal-child health interventions was achieved. Under-5s mortality fell by 22% in Millennium Village sites and by 33% relative to matched comparison sites. The integrated multisector approach was associated with rapid reductions in child mortality.
Optimal treatment for the infants of mothers who are diagnosed with HIV-1 infection late in pregnancy, and therefore do not receive antiretroviral therapy during pregnancy, is uncertain. One drug, two drug, and three drug regimens have been compared in a trial in Brazil, South Africa, Argentina, and the USA (N Engl J Med 2012;366:2368–79). The trial included 1684 bottle-fed infants of mothers with HIV-1 infection who had not received antiretroviral therapy. The infants were randomised within 48 h of birth to one drug (zidovudine for 6 weeks), two drugs (zidovudine for 6 weeks plus three doses of nevirapine within the first 8 days), or three drugs (zidovudine for 6 weeks plus nelfinavir and lamivudine for 2 weeks). Overall, the estimated rate of in utero HIV-1 transmission was 5.7% with no significant difference between groups. Intrapartum transmission occurred in 4.8% (one drug), 2.2% (two drugs), and 2.4% (three drugs), a significant difference between one drug and two or three drugs. By the age of 3 months 8.5% of the infants were infected with HIV-1, 11% in the zidovudine alone group, 7.1% in the two drug group, and 7.4% in the three drug group, again significantly better with two or three drugs rather than one. Factors significantly associated with a higher risk of HIV-1 transmission were zidovudine monotherapy, a higher maternal HIV-1 load, and maternal use of illegal substances. Neutropenia was significantly more common with three drugs (16.4% vs 14.9% vs 27.5%). The two-drug regimen might be the best option.
Deciding on the best treatment for HIV-infected young children is problematic. For children who have previously been exposed to nevirapine, a ritonavir-boosted lopinavir-based regimen is recommended but there are difficulties with such a regimen in developing countries. Now a study in six countries of sub-Saharan Africa and India (N Engl J Med 2012;366:2380–9) has shown better results with a ritonavir-boosted lopinavir-based regimen for young children who have not previously been exposed to nevirapine. The study included 288 HIV-infected children aged 2–36 months with no previous exposure to nevirapine. Randomisation was to zidovudine and lamivudine with either nevirapine or ritonavir-boosted lopinavir. The rates of virological failure or treatment discontinuation by week 24 were significantly greater in the nevirapine group (40.8% vs 19.3%). At the time of virological failure 21 of the 32 patients tested in the nevirapine group and 11 of the 20 in the ritonavir-boosted lopinavir group had developed drug resistance. Drug toxicity was greater in the nevirapine group. Although these data demonstrate the superiority of the ritonavir-boosted lopinavir regimen there are problems: the liquid preparation of ritonavir-boosted lopinavir is unpleasant to taste, unstable at high ambient temperature, and the regimen is twice as costly as the nevirapine-based regimen. The decision facing policy makers is difficult and new drug formulations are needed urgently.
Babies with duct-dependent congenital heart disease may appear normal at birth and be discharged from hospital only to become suddenly very ill, collapsed and cyanosed when the ductus arteriosus closes after a few weeks. It is to avoid this happening that screening with pulse oximetry has been suggested and has been, or is being, taken up by at least 16 states of the USA. A new systematic review and meta-analysis (Lancet 2012;379:2459–64; see also Comment, ibid: 2407–8 and Editorial, ibid: 2401) has confirmed the suitability of pulse oximetry for screening of new born babies. The 13 studies analysed included 229 421 infants. Pulse oximetry had a sensitivity of 76.5% and a specificity of 99.9% for the detection of critical congenital heart defects. The false positive rate was 0.5% in the first 24 h after birth and 0.05% when screening was done after 24 h. The conclusion is that pulse oximetry screening should be widely introduced.
The four pillars of the Safe Motherhood Initiative are family planning, antenatal care, safe delivery, and postnatal care. Family planning improves child survival in resource poor countries by reducing sibling competition for family and maternal resources. Data from three international databases have been used to estimate the effects of contraceptive use on maternal mortality in 172 countries in 2008 (Lancet 2012;389:111–25; see also Comment, ibid: 87–8). Using a counterfactual modelling approach it was estimated that there were 342 303 deaths from maternal causes in 2008 and that contraceptive use averted a further 272 040 maternal deaths (a 44% reduction of the potential total). Data from 167 countries suggested that expansion of contraceptive use to satisfy unmet needs could prevent another 104 000 maternal deaths per year (20% reduction). In countries with high (>65%) rates of contraceptive use, such use averted almost 60% of potential maternal deaths whereas in sub-Saharan Africa, where only 22% of married or sexually active women used contraceptives, the proportion of potential maternal deaths averted was 32%. Greater use of contraception could avert many maternal deaths in developing countries. Maternal death is associated with reduced child survival.