Subclinical Mastitis As A Risk Factor For Motherinfant Hiv Transmission

JF Willumsen,1 SM Filteau,1 A Coutsoudis,2 KE Uebel,2 M-L Newell3 and AM Tomkins1

'Centre for International Child Health, Institute of Child Health, London;2Department of Pediatrics and Child Health, University ofNatal, Durban, 3Department ofEpidemiology and Population Health, Institute of Child Health, London

Key words: mastitis, breast milk, sodium, inflammation, HIV

Abstract: Subclinical mastitis, as diagnosed by an elevated sodium/potassium ratio in milk accompanied by an increased milk concentration of the inflammatory cytokine, interleukin-8 (IL8), was found to be common among breast feeding women in Bangladesh and Tanzania. Subclinical mastitis results in leakage of plasma constituents into milk, active recruitment of leukocytes into milk, and possible infant gut damage from inflammatory cytokines. Therefore, we wished to investigate whether subclinical mastitis was related to known risk factors for postnatal mother-to-child HIV transmission, that is, high milk viral load or increased infant gut permeability. HIV-infected South African women were recruited at the antenatal clinic of McCord's Hospital, Durban. Risks and benefits of different feeding strategies were explained to them and, if they chose to breast feed, they were encouraged to do so exclusively. Women and infants returned to the clinic at 1,6 and 14 weeks postpartum for an interview about infant health and current feeding pattern, a lactulose/mannitol test of infant gut permeability, and milk sample collection from each breast separately for analysis of Na/K ratio, IL8 concentration and viral load in the cell-free aqueous phase. Only preliminary cross-sectional analyses from an incomplete database are available at this point. Moderately (0.6 - 1.0) or greatly (>1 . 0) raised Na/K ratio was common and was often unilateral, although as a group right and left breasts did not differ. Considering both breasts together, normal, moderately raised or greatly raised Na/K was found, respectively, in 51%, 28%, 21% of milk samples at 1 week (n=190); 69%, 20%, 11% at 6 weeks (n=167); and 72%, 16%, 12% at 14 weeks (n=l22). IL8 concentration significantly correlated with both Na/K and viral load at all times. Na/K correlated with viral load at 1 and 14, but not 6 weeks. At 1 and 14 weeks, geometric mean viral loads in samples with Na/K > 1.0 were approximately 4 times those in samples with Na/K < 0.6. At 1 week but not later times, exclusive breast feeding was associated with lower milk viral load than was mixed feeding. Gut permeability was unrelated to milk Na/K ratio or IL8 concentration and was not significantly increased by inclusion of other foods than breast milk in the infant's diet. The results suggest that subclinical mastitis among HIV-infected women may increase the risk of vertical transmission through breast feeding by increasing milk viral load. The importance of various causes of subclinical mastitis, which likely differ at 1 week from at later times and may include local infection or sterile inflammation, systemic infection, micronutrient deficiencies, or poor lactation practices, needs to be further clarified so that appropriate interventions can be implemented.


1.1 Breast feeding and mother to infant HIV transmission

Breast feeding is estimated to account for about a third of mother-to-infant HIV transmission among African populations.1 This has resulted in a major public health dilemma since the advantages of breast feeding for preventing infant morbidity and mortality are well established. HIV-infected women who have reliable access to safe, clean breast milk substitutes are advised not to breast feed. However, in much of Africa, provision of clean breast milk substitutes cannot be guaranteed and women must balance the risks of HIV transmission through breast feeding against the risks of infant mortality associated with not breast feeding. A woman's choice is further complicated by the fact that the protection against diarrhea and respiratory infections which is provided by breast milk may be particularly important for infants already HIV-infected before or during delivery. Furthermore, in cultures where breast feeding is almost universal, not breast feeding may disclose HIV status which could have serious social consequences. Therefore, it is important to try to identify which women or which feeding practices are associated with breast milk transmission of HIV so that the subgroup ofmother-infant pairs at risk can be supported in safe use ofbreast milk substitutes and the remaining pairs can be supported in optimising infant health through breast feeding.

A recent study from South Africa found that exclusive breast feeding was associated with a lower risk of mother-to-infant transmission before 3 months than was mixed feeding.2 There are biologically plausible mechanisms whereby mixed feeding may carry the highest risk of HIV transmission. Breast milk, in addition to virus, contains antibodies and glycosaminoglycans which may inhibit virus binding to cells in the infant gut3, 4 various factors which can actively promote the infant's own immune function,5 and factors which promote development of gut integrity.6 Therefore, with exclusive breast feeding, the net result may be a low level of postnatal transmission. Feeding of other foods can damage the infant's gut and increase permeability. In breast feeding mixed with other foods, this increased permeability, together with decreased total amounts ofbreast milk immune factors, may tip the balance between virus and protective factors in milk and permit virus to enter infant cells more readily. An additional factor which we have recently considered is that supplementation of a breastfed infant's diet with other foods may, by decreasing milk intake, result in milk stasis and subclinical mastitis with consequent adverse effects on the infant.

1.2 Subclinical mastitis

The biology ofmastitis has been extensively studied by dairy researchers. Inflammatory cytokines such as IL8, IL1 and tumour necrosis factor-a are produced within the mammary gland and mediate both recruitment of leukocytes and opening ofthe tightjunctions between epithelial cells.7-9 IL8 can be readily measured in milk10- 11 but, in our experience, milk is a difficult matrix in which to quantitate many of the other key cytokines involved in mastitis.11 However, sodium is an abundant and easily measurable plasma constituent which crosses into milk when tight junctions become permeable.12

Recently, while investigating the impact of maternal vitamin A supplementation on immune factors in the breast milk of Bangladeshi women, we wished to exclude women with mastitis since this inflammation was likely to greatly alter milk immunology.10 Therefore, we measured milk sodium and controlled for the varying proportions of aqueous phase in spot milk samples by expressing sodium as a ratio to potassium. To our surprise, elevated milk sodium was too common - 25% of women at 2 weeks postpartum and 12% at 3 months - for us to simply exclude these women. Since the study was designed for other purposes, women were not asked about mastitis at the time of sampling, although any overt mastitis would have been noted in the general health check. We refer to high milk Na/K ratio accompanied by high IL8 concentration but in the absence ofsymptoms as subclinical mastitis. Importantly, infants ofwomen with raised milk Na/K ratio at either 2 weeks or 3 months gained significantly less weight between these two time points than did infants of women with normal milk sodium.10 Poor weight gain associated with high breast milk sodium has also been observed in American infants.13 A possible mechanism is that poor lactation practice by the mother or weak suckling by the infant results in milk stasis, mammary gland involution,14-15 and consequently both raised milk sodium and poor growth.

Common features ofclinical and subclinical mastitis often occurs unilaterally high milk sodium high milk IL8 high milk pH associated with systemic inflammation decreasedby dietary anti-oxidants

Since subclinical mastitis was common among Bangladeshi women and was associated with poor infant, and possibly also maternal, health, we extended our investigations to other populations. Of Tanzanian women participating in a trial of food-based micronutrient supplementation, 13% at 1 month postpartum and 11% at 3 months had subclinical mastitis.11 Milk Na/K ratio correlated with maternal plasma acute phase proteins. Biologically it seems more plausible that subclinical mastitis may result from systemic infection or inflammation than that a localised and subclinical inflammation can result in a detectable systemic inflammation. Systemic illness is common in the first few months postpartum16 and can increase permeability of other epithelia such as that of the gastrointestinal tract.17We also found in the Tanzanian cohort that supplementation of women during the last trimester of pregnancy and the first 3 months postpartum with vitamin E-rich sunflower oil, but not with vitamin A-rich red palm oil, could decrease milk Na/K. 11 A possible mechanism is that antioxidant micronutrients such as vitamin E can decrease both production of inflammatory cytokines and tissue damage caused by free radicals produced during inflammation.18 Antioxidant micronutrient supplementation is well recognised in the dairy industry as a means ofdecreasing mastitis.19 The box summarises the similarities between subclinical and clinical mastitis.

Our South African studies are the first for which the primary aim was to investigate subclinical mastitis in a population. We first conducted a cross-

sectional study of women attending vaccination clinics in the Durban area with their infants (Willumsen et al submitted). We specifically inquired about symptoms of mastitis or problems with breast feeding and we collected milk from each breast separately. Problems ofbreast pain, cracked nipples or other breast feeding difficulties were reported by only 3% of the women and could not explain the high prevalence of raised Na/K ratio. Of 269 women with infants 20 weeks old or less, 71 (26%) had Na/Kratio >0.6 in milk from one breast and 77 (29%) had raised Na/K ratio in both breasts. Unilateral inflammation thus seems common for both subclinical and clinical mastitis, indicating that local inflammation or infection may be an important cause. Although the different sampling methods prevent simple comparisons across studies, raised Na/K ratio appeared more common in this population than in those we studied in Bangladesh and Tanzania. One possible reason for this is the high prevalence of HIV infection in Durban although for ethical and logistic reasons we did not test for HIV in this cross-sectional, vaccination clinic-based study. We inquired about infant feeding practices within the previous 24 hours and found that exclusive breast feeding was associated with lower Na/Kratio (0.58, 95% CI 0.53 -0.62, n = 320 breasts) than mixed breast milk and formula feeding (1.05, 95% CI 0.82 - 1.35, n = 104), although mixed feeding with breast milk and other foods resulted in the lowest Na/Kratio (0.43, 95% CI 0.38 - 0.47, n = 166 breasts; P<0.05 in each case).

There is little comparable population-based data on Na/K ratios in milk of well-nourished European or North American women since breast feeding in these populations is not universal and the women recruited to research studies have been generally ofhigh socioeconomic status. Of22 women we have recruited to date from a general practice clinic in a middle class suburb of London, 18 had normal Na/K ratio in both breasts, 2 had a slightly elevated Na/K in one breast and 2 had bilateral subclinical mastitis. We have also analysedNa/Kratios in repeated samples from a small group ofwomen expressing milk regularly to feed very sick infants in hospital. Although many samples in this group have high Na/K since infants were mostly premature and/or too ill to suckle, Na/K ratios were found to be quite consistent for an individual woman across time.

Losing Weight Without Starving

Losing Weight Without Starving

Tired of Trying To Loose Weight And It Never Works or You Have To Starve Yourself Well Here's A Weight Loss Plan That takes Care of Your Weight Problem And You Can Still Eat. In This Book, You’ll Learn How To Lose Weight And Not Feel Hungry! In An Easy Step-By-Step Process That Enables You To Feel Good About Loosing Weight As Well As Feeling Good Because Your Stomach Is Still Full.

Get My Free Ebook

Post a comment