Breastfeeding and CRP in adulthood

None of us need further convincing of the benefits of breastfeeding to children's health. But can this extend long into adult life? The adult literature is full of studies demonstrating convincingly that chronic minor elevations of blood C-reactive protein (CRP) levels are an excellent predictor of major health problems such as atheroma, ischaemic heart disease and stroke. A massive longitudinal cohort study from the US related breastfeeding history to CRP levels at age 24–32 years (McDade T. Proc R Soc B 2014; doi:10.1098/rspb.2013.3116). The National Longitudinal Study of Adolescent Health had data on nearly 10 500 participants. Lower birthweight was associated with higher CRP levels as predicted by the Barker hypothesis. But the findings on breast-feeding were also striking: those breastfed for over 12 months had CRP levels 31% lower than those never breastfed, with a ‘dose-response’ trend seen for those breastfed for shorter periods. Of course there were many confounding factors, but they were able to allow for most of these in their complex analyses. They also compared siblings discordant for breastfeeding duration, where ethnic, socioeconomic and genetic factors are automatically controlled for: there was still a small but significant independent effect.

Although this is a proxy study with no real health outcomes, it represents the best evidence yet about the long-term benefits of breastfeeding. CRP is a biomarker for harmful low-grade inflammation, and the authors hypothesise that something in breast milk down-regulates this permanently.

What happens when bed-wetters grow up?

When we gratefully discharge enuretic children who have become dry at night, should we warn them that their problem might return in adulthood? Researchers from Japan used a questionnaire to look retrospectively at a group of adults aged 30–89, asking them about past and present nocturnal urinary problems (Akashi S, Tomita K. Acta Paed 2014. doi:10.1111/apa.12694). The group they selected, deliberately, were the parents and grandparents of children in their enuresis clinic, because of the known strong genetic influence on enuresis. Remarkably they achieved over 3600 responses, and after exclusions for other significant diseases, were able to analyse over 2500. Over half of these adults reported bed-wetting as children, more in men. Those who did were significantly more likely to have continuing urinary symptoms as adults, for example, for nocturia, in men OR 3.8 (95% CI 3.0–4.8), in women OR 4.1 (95% CI 3.3–5.1). They compared these risks with data from whole population studies of the same age groups, and found that childhood enuresis independently increased risk of nocturia and urgency to a level comparable to diabetes, prostatic disease or just aging. This study is obviously subject to huge selection and recall bias, but it reminds us that the genetic predisposition to enuresis (ie, circadian urine production, functional bladder capacity), doesn't go away.

Catheter or surgical closure for PDA?

Non-surgical catheter techniques for closing a patent ductus arteriosus (PDA) have been in use for two decades, but some centres still prefer to do open surgery. Avoiding a thoracotomy seems to have an obvious advantage, but catheterisation also has it drawbacks. Researchers from China undertook a meta-analysis, and found 7 methodologically adequate trials, which together included 810 children under 17 years (Wang K, et al. Clin Cardiol 2014;37:188–94). Preterm infants and those with complex heart disease were excluded. There were no significant differences in rates of primary procedural failure, nor of complications or the need for blood transfusion postoperatively. Catheterisation left more children with a residual shunt (RR 5.2, 95%CI 1.4–19.2: p=0.01). However, as expected, catheter closure significantly reduced length of hospital stay (mean 1.6 days less, 95% CI 2.6–0.7: p=0.001). There was huge heterogeneity between the studies: for example, mean lengths of stay for surgical treatment varied between 1.1 and 18.7 days. On the face of it, it appears to be a straight choice between higher risk of non-closure, and longer hospital stay, but each centre will have its own outcome data which may be different.

Helicobacter and probiotics

Gastrointestinal infection with Helicobacter pylori remains a significant health problem worldwide, and although eradication regimes with triple therapy (a proton pump inhibitor and two antibiotics) are well-established, success rates are disappointing. Adding probiotics to this regime has shown some promise in adults. A meta-analysis from China looked at the literature in children: they found 7 adequate studies, from diverse European, Asian and South American countries, published between 2005 and 2013 (Shan Li, et al. Eur J Pediatr 2014;173:153–61). Over 500 children, all with proven symptomatic H pylori infection, were involved. A variety of probiotic regimes were used, mostly using strains of Lactobacillus. All were given in addition to standard triple therapy. Eradication rates were roughly doubled (OR for eradication by intention to treat 1.96; 95% CI 1.3–3.2). Perhaps more significantly, adverse effects of the triple therapy, particularly diarrhoea, were reduced: (OR 0.32; 95% CI 0.1–0.8). In practice, eradication failure may be related to poor compliance with treatment, and probiotics may help to improve this.

Helicobacter and allergy

But should we be trying to eradicate Helicobacter anyway? The ‘hygiene hypothesis’ suggests that infections such as this, less common in developed countries, may protect against allergy. A cohort study from Ethiopia looked at over 1000 children (Amberbir A, et al. Clin Exp Allergy 2014: 44; 563–71). At ages 1, 3 and 5 years they were assessed both for allergic/atopic symptoms through standard ISAAC questionnaire and skin prick tests (SPT), and for H pylori infection through stool samples. Infection was common (25%). At age 3 years, infection was inversely associated with symptomatic eczema (OR 0.31; 95% CI 0.1–0.9), and at 5 years with positive SPT (OR 0.26; 95% CI 0.1–0.9). They allowed for possible confounding factors in their multivariate analysis, such as social deprivation, and treatment with antibiotics, and the negative association was still seen. This doesn't prove that H pylori is protective, as it may be a proxy for other factors, including many other types of infections. Screening for and treating asymptomatic H pylori, even with probiotics, is probably not such a good idea.