QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Dancheck, B. Articles by Semba, R. D. Search for Related Content PubMed PubMed Citation Articles by Dancheck, B. Articles by Semba, R. D. Pubmed/NCBI databases Compound via MeSH Substance via MeSH Medline Plus Health Information Breast Feeding Hazardous Substances DB VITAMIN A

---

Community and International Nutrition

Breast Milk Retinol Concentrations Are Not Associated with Systemic Inflammation among Breast-Feeding Women in Malawi¹

Barbara Dancheck², Veronique Nussenblatt, Michelle O. Ricks, Newton Kumwenda^*, Margaret C. Neville, Dana T. Moncrief, Taha E. Taha^** and Richard D. Semba

Department of Ophthalmology, Johns Hopkins School of Medicine, Baltimore, MD; ^* College of Medicine, University of Malawi, Blantyre, Malawi; Department of Physiology, University of Colorado School of Medicine, Denver, CO; and ^** Department of Epidemiology, Johns Hopkins Bloomberg School of Public Health, Baltimore, MD

²To whom correspondence should be addressed. E-mail: bdanche1{at}jhmi.edu.

	ABSTRACT

TOP ABSTRACT SUBJECTS AND METHODS RESULTS DISCUSSION LITERATURE CITED

 
The acute phase response and inflammation are associated with lower plasma retinol concentrations, but their effect on breast milk retinol concentrations is unclear. We measured plasma retinol concentrations, acute phase proteins, and breast milk retinol concentrations in 237 breast-feeding women at 2 wk postpartum in Blantyre, Malawi; 16.5% of the women had plasma retinol < 0.70 µmol/L and 14.8% had breast milk retinol < 1.05 µmol/L. Among women with and without inflammation [₁-acid glycoprotein (AGP) > 1 g/L and/or C-reactive protein (CRP) > 5 mg/L], geometric mean (95% CI) plasma retinol was 0.89 (0.84, 0.94) and 1.05 (1.01, 1.17) µmol/L, respectively (P < 0.0001). Among women with and without inflammation, geometric mean (95% CI) breast milk retinol was 2.12 (1.89, 2.36) and 2.05 (1.75, 2.39) µmol/L, respectively (P = 0.74). In multiple linear regression models adjusting for age, parity, education, BMI, and days postpartum, plasma retinol concentrations were associated with plasma AGP and CRP concentrations (P < 0.0001 and P = 0.01, respectively), whereas breast milk retinol concentrations were unaffected by plasma AGP and CRP concentrations (P = 0.22 and P = 0.86, respectively). These findings suggest that breast milk retinol concentrations are not affected by systemic inflammation.

KEY WORDS: • acute phase response • inflammation • milk • retinol • vitamin A deficiency

Vitamin A deficiency is a major cause of morbidity and mortality among children and women of childbearing age in developing countries worldwide (1). Vitamin A deficiency is characterized by immune suppression and increased susceptibility to infectious diseases (2). The identification of populations at risk for vitamin A deficiency is important for targeting groups that would benefit from interventions to improve vitamin A status. Vitamin A deficiency is defined to be of public health importance if 15% of a specific population has a plasma retinol concentration < 0.70 µmol/L (3).

Plasma retinol concentrations may be decreased during subclinical infections or inflammation because retinol bound with retinol-binding protein is a negative acute phase reactant (4). Acute phase proteins such as plasma ₁-acid glycoprotein (AGP)³ and C-reactive protein (CRP) are markers for an acute phase response and can indicate the presence of a subclinical infection or inflammation (4,5). It was proposed that individuals with elevated acute phase proteins should be excluded from data analyses on surveys that rely on plasma retinol as an indicator of vitamin A status; thus, the prevalence of those with plasma retinol concentration < 0.70 µmol/L would be based only on those without markers of infection (5). Because subjects with vitamin A deficiency are more prone to infections (2,6,7), the exclusion of all individuals with biochemical indicators of infection could potentially underestimate the prevalence of vitamin A deficiency.

Breast milk retinol is considered a good indicator of vitamin A status in lactating women (8), but it is not well known whether breast milk retinol concentrations are affected by subclinical infection. A breast milk retinol concentration < 1.05 µmol/L is considered to be consistent with vitamin A deficiency in lactating women (8), and prevalence rates of <10%, 10 to <25%, and 25% of breast milk retinol <1.05 µmol/L are considered to indicate mild, moderate, and severe vitamin A deficiency, respectively, as a public health problem (1,3). Because much of the retinol in breast milk comes directly from the diet, we hypothesized that inflammation (9), as indicated by elevated acute phase proteins, would be associated with lower plasma retinol concentrations but not lower breast milk retinol concentrations. To address this hypothesis, we measured breast milk and plasma vitamin A concentrations and plasma acute phase proteins in breast-feeding women in Blantyre, Malawi.

	SUBJECTS AND METHODS

TOP ABSTRACT SUBJECTS AND METHODS RESULTS DISCUSSION LITERATURE CITED

 
The study population consisted of women who gave birth at Queen Elizabeth Central Hospital in Blantyre, Malawi. Women were eligible for the study if they were HIV-negative, chose to breast-feed, resided in the Blantyre District, and were at least 18 y old. Written, informed consent was obtained for HIV screening, and pre- and post-test HIV counseling was provided. Women were enrolled in the study at 2 wk postpartum after giving written, informed consent. At the enrollment visit, demographic information was collected, a medical history, vital signs, physical exam, and anthropometry were conducted, and venous blood and breast milk were collected from the mothers. Maternal weight and height were measured to the nearest 0.1 kg and 1 cm, respectively, using a SECA 700 balance. The study protocol was approved by institutional review boards at Johns Hopkins School of Medicine in Baltimore, MD, and the College of Medicine, University of Malawi, Blantyre, Malawi.

Venous blood was collected from nonfasting women using trace element–free S-Monovette Li-Heparin syringes (Sarstedt Monovette); the blood was stored at 4°C until it was processed (600 x g, 7 min, room temperature). Plasma aliquots were made in sterile cryovials (Wheaton Science Products) and stored at –70°C. Casual milk samples (not controlling for the time since the last breast-feeding episode) were collected from mothers in the morning or early afternoon; 5–7 mL of breast milk was collected from each breast by manual expression after sterilizing the nipple and areola with betadine solution. Whole breast milk aliquots were prepared by gentle swirling of the sample using a vortex mixer before pipetting the whole milk into 2-mL aliquots. Milk aliquots were frozen and stored at –70°C. A creamatocrit was not obtained, and milk vitamin A concentrations were expressed as retinol per volume of milk (µmol/L) rather than per gram of fat. Maternal HIV-1 infection was diagnosed on the basis of a positive rapid test (Determine 1/2 Rapid test, Abbott) and confirmed by a positive ELISA for HIV-1 antibodies (Wellcozyme, Wellcome Diagnostics). Women with positive HIV tests were excluded from this study.

Retinol was measured in plasma using HPLC and a modified method from the Nutrition Laboratory, Inorganic Toxicology and Nutrition Branch, Division of Laboratory Sciences, National Center of Environmental Health, Centers for Disease Control and Prevention (Rosemary Schleicer, personal communication). Briefly, tocol (Hoffmann-La Roche) was added to the samples as an internal standard, along with isopropanol to precipitate the proteins. Samples were extracted with hexane; extracts were dried under nitrogen and redissolved in 50:50 acetonitrile:ethanol. The mobile phase consisted of one pump in acetonitrile with 0.1% triethylamine, and a second pump in ethanol with 0.1% triethylamine. A gradient method was applied by varying the solvent concentrations from 85% acentonitrile:triethylamine to 50% aceonitrile:triethylamine and again to 85% acetonitrile:triethylamine.

Breast milk samples were randomly sampled from either the right or left breast milk aliquots in the repository, and retinol was measured in breast milk using HPLC. The saponification/extraction method was a slight modification of that of Tanumihardjo et al. (10). Briefly, samples were saponified in ethanol and KOH-HOH with gentle rotation for 1 h at 45°C, followed by 15 min on ice. Samples were extracted with hexane containing 0.1% BHT and dried under nitrogen. Once dried, the extracts were resuspended in 50/50 acetonitrile and ethanol. The HPLC method was the same as that used for plasma.

Standard curves were assayed periodically using standard reference material (SRM) 968C (National Institute of Standards and Technology; NIST), and sample retinol concentrations were calculated based on these curves. SRM 968C is a set of 2 plasma-based standards with NIST-certified levels of retinol. Quality control was assessed by repeated analysis of either a pooled human plasma or human breast milk control assayed at the beginning and end of each analysis. Between- and within-run CVs for controls were 18.1 and 10.6% for maternal plasma retinol, and 2.9 and 2.7% for breast milk retinol, respectively.

Plasma AGP was measured using a commercial radial immunodiffusion kit (Bindarid, The Binding Site). Plasma CRP was measured using a commercial ELISA (Virgo CRP 150, Hemagen Diagnostics). Controls provided by the Manufacturer were used as contols in both cases. Between- and within-run CVs for AGP and CRP were 1.4 and 1.4%, and 4.0, and 4.3%, respectively.

Plasma and breast milk retinol, AGP, and CRP were skewed to higher values and were transformed by natural log to achieve a normal distribution. Vitamin A deficiency in mothers was defined as plasma retinol concentrations < 0.70 µmol/L or breast milk retinol < 1.05 µmol/L (1). Elevated AGP was defined as AGP > 1 g/L and elevated CRP was defined as CRP > 5 mg/L (5,7). Systemic inflammation was defined as elevation in either or both AGP and CRP (AGP > 1 g/L and/or CRP > 5 mg/L). Student’s t test and Wilcoxon rank-sum tests were used to compare continuous variables. ² and exact tests were used to compare categorical variables between groups. Spearman correlations were used to examine the relation between plasma and milk retinol concentrations. A 1-sample t test for means was used to compare plasma retinol concentrations between all women and women without inflammation only. A 1-sample test for proportions was used to compare plasma retinol concentrations between all women and women without inflammation only. Multiple linear regression models were used to examine the relation between plasma or breast milk retinol concentrations and CRP and AGP concentrations, adjusting for age, parity, education, BMI, and days postpartum. Multivariate logistic regression models adjusting for age, parity, education, BMI and days postpartum were used to examine the relation between plasma or breast milk retinol concentrations and inflammation. All statistics were analyzed using SAS version 8.1. Differences were considered significant in this study at P 0.05.

	RESULTS

TOP ABSTRACT SUBJECTS AND METHODS RESULTS DISCUSSION LITERATURE CITED

 
From February 2000 to April 2002, 250 women were enrolled in the study; of these, 237 (94.8%) had both plasma and breast milk samples available for laboratory analysis. Geometric mean (95% CI) breast milk retinol concentration was 2.09 (1.92, 2.29) µmol/L and geometric mean (95% CI) plasma retinol concentration was 0.93 (0.89, 0.97) µmol/L in the 237 women. There were 39 women (16.5%) who had plasma retinol < 0.70 µmol/L and 35 women (14.8%) who had breast milk retinol < 1.05 µmol/L.

The characteristics of women by inflammation status (inflammation defined as AGP > 1 g/L and /or CRP > 5 mg/L), are shown in Table 1. Women with inflammation had lower plasma retinol concentrations (P < 0.0001), and the proportion of women with plasma retinol < 0.70 µmol/L was higher in the women with inflammation compared with women without inflammation (P = 0.05). Breast milk retinol concentrations did not differ between women with and without inflammation (P = 0.74), and the proportion of women with breast milk retinol < 1.05 µmol/L also did not differ between the 2 groups (P = 0.98). The 2 groups did not differ by age, parity, education, or BMI. The Spearman correlation between maternal milk and plasma retinol concentrations at 2 wk postpartum was r = 0.24 (P < 0.0002).

Multiple linear regression models were used to examine the association between plasma retinol concentrations (Table 2), breast milk retinol concentrations (Table 3), and AGP and CRP, adjusting for potential confounders. Of the 237 women with plasma and breast milk retinol concentrations, only 221 had all data available for the multiple linear regression analysis. After adjusting for age, education, parity, BMI, and days postpartum, AGP and CRP concentrations were associated with plasma retinol concentrations (P < 0.0001 and P = 0.01, respectively). After adjusting for age, parity, education, BMI, and days postpartum, AGP and CRP concentrations were not associated with breast milk retinol concentrations (P = 0.22 and P = 0.86, respectively). In addition, multivariate logistic regressions models adjusting for age, parity, education, BMI, and days postpartum showed that inflammation (AGP > 1 g/L and/or CRP > 5 mg/L) was associated with an increased risk of plasma retinol < 0.70 µmol/L [odds ratio (OR) 2.68, 95% CI 1.08, 6.64), but was not associated with breast milk retinol < 1.05 µmol/L (OR 1.15, 95% CI, 0.50, 2.63).

	DISCUSSION

TOP ABSTRACT SUBJECTS AND METHODS RESULTS DISCUSSION LITERATURE CITED

The 2 major sources of vitamin A in the circulation that are available for uptake by the mammary gland for secretion in breast milk are retinol carried by retinol-binding protein and retinyl esters carried by chylomicrons. Studies in animals suggest that chylomicrons provide at least one third to one half of vitamin A to mammary gland tissue during lactation for secretion into milk (11), and that in rats, milk retinol can vary with dietary vitamin A intake even when plasma retinol concentrations are relatively unaffected (12). Animal studies showed that there may be a homeostatic mechanism by which the transfer of retinol from the circulation to breast milk is protected within a fairly wide range of liver vitamin A stores and serum retinol concentrations (13). In rats, moderate restrictions of dietary fat or protein had little effect on the transfer of vitamin A from dam to pup, but more severe dietary restrictions had a negative effect on milk volume (13). All of these observations suggest that breast milk retinol is shielded from changes in plasma retinol because the mammary gland utilizes a large proportion of retinyl esters that have entered the circulation after absorption in the gastrointestinal tract.

Although it was proposed that individuals with biochemical markers for infection should be excluded from data analysis in surveys that rely upon plasma retinol concentrations (5), it appears that exclusion of individuals with subclinical infection could underestimate the prevalence of vitamin A deficiency. In the present study, the prevalence of vitamin A deficiency among breast-feeding women was 14.8%, based upon breast milk retinol concentrations, and 16.5%, based upon plasma retinol concentrations. If women with elevated acute phase proteins were excluded from the analysis, the prevalence of vitamin A deficiency as estimated using plasma retinol would be 9% compared with a prevalence of 14.8% based upon the criterion of breast milk retinol. Although there was a trend toward fewer women with low plasma retinol when all women with inflammation were removed from the analysis, this decrease was of marginal significance, probably because there was a small number of women without inflammation who had low plasma retinol. When acute phase reactants were used to "adjust" the plasma retinol concentrations, the overall sample was larger and the "adjusted" prevalence of women with plasma retinol < 0.70 µmol/L was significantly lower. In this study population, the exclusion of individuals with elevated acute phase proteins or the use of "adjusted" retinol concentrations underestimated the prevalence of vitamin A deficiency by nearly 40%. Because vitamin A deficiency is a syndrome characterized by immune suppression, increased susceptibility to infections, and elevated acute phase proteins (6,7,14), the exclusion of individuals with elevated acute phase proteins or "adjustment" of plasma retinol concentrations may also lead to an underestimation of the prevalence of vitamin A deficiency in some populations.

One potential limitation of this study is that the fat content in breast milk samples was not determined, as a recent study suggested that the most accurate indicator of the response of breast milk retinol to vitamin A supplementation was the measurement of breast milk retinol per gram of fat in casual breast milk samples (15). Another study of breast-feeding women in Indonesia showed that breast milk vitamin A was most sensitive to changes in vitamin A status when expressed as retinol per volume of breast milk (16). In the present study, breast milk samples were collected at 2 wk postpartum; during this period, milk retinol concentrations tend to decline since time of delivery (17). Thus, it is possible that the prevalence of vitamin A deficiency may be underestimated when based on breast milk samples that were collected 2 wk postpartum. Because sample collection was not controlled for time since the last breast-feeding episode, the results are less precise than they might have been otherwise. It would be beneficial for future studies to examine breast milk retinol in relation to markers of inflammation at 1 mo postpartum or later when breast milk retinol concentrations have stabilized. It is not known whether the influence of the acute phase response differs during breast-feeding, and later analysis of breast milk retinol and inflammation could provide additional insight. The geometric mean breast milk retinol concentrations of 2 µmol/L among lactating women in Malawi at 2 wk postpartum is similar to values reported in other populations in which vitamin A deficiency is prevalent. At 2 wk postpartum, breast milk retinol concentrations of 2.69 µmol/L were reported in Indonesia (18) and 1.41 µmol/L were reported in Bangladesh (15). A low-to-moderate correlation between plasma retinol and breast milk retinol concentrations was shown in previous studies, with correlations of 0.30 reported among Thai women who had been breast-feeding for >6 mo (19) and 0.32 among Indonesian women at 6 mo postpartum (8).

In summary, this study suggests that breast milk retinol concentrations are relatively unaffected by modest levels of systemic inflammation in this population of breast-feeding women in Malawi. In contrast, plasma retinol concentrations were lower in women who had elevated acute phase proteins. Exclusion or inclusion of women with elevated acute phase proteins may potentially underestimate or overestimate, respectively, the prevalence of vitamin A deficiency in a population in studies in which plasma retinol is used as the sole indicator of vitamin A status. If further studies in other populations corroborate these findings, this would suggest that breast milk retinol concentrations are a relatively good indicator of vitamin A status in breast-feeding women in whom subclinical infections may be common.

	ACKNOWLEDGMENTS

 
We thank Saskia de Pee for helpful comments on the manuscript.

	FOOTNOTES

 
¹ Supported by National Institutes of Health grants R01 HD30042, and R01 HD32247.

³ Abbreviations used: AGP, ₁-acid glycoprotein; CRP, C-reactive protein; NIST, National Institute of Standards and Technology; OR, odds ratio; SRM, standard reference material.

Manuscript received 20 July 2004. Initial review completed 27 August 2004. Revision accepted 2 November 2004.

	LITERATURE CITED

TOP ABSTRACT SUBJECTS AND METHODS RESULTS DISCUSSION LITERATURE CITED

 

1. McLaren, D. S. & Frigg, M. (2001) Sight and Life Manual on Vitamin A Deficiency Disorders (VADD) 2nd ed. 2001 Task Force Sight and Life Basel, Switzerland.

2. Semba, R. D. (2002) Vitamin A, infection, and immune function. Calder, P. C. Field, C. J. Gill, H. S. eds. Nutrition and Immune Function 2002:151-169 CABI Publishing New York, NY. .

3. World Health Organization (1996) Indicators for Assessing Vitamin A Deficiency and Their Application in Monitoring and Evaluating Intervention Programmes 1996 Document WHO/NUT/96.10. WHO Geneva, Switzerland.

4. Filteau, S. M., Morris, S. S., Abbott, R. A., Tomkins, A. M., Kirkwood, B. R., Arthur, P., Ross, D. A., Gyapong, J. O. & Raynes, J. G. (1993) Influence of morbidity on serum retinol of children in a community-based study in northern Ghana. Am. J. Clin. Nutr. 58:192-197.[Abstract/Free Full Text]

5. Thurnham, D. I., McCabe, G. P., Northrop-Clewes, C. A. & Nestel, P. (2003) Effects of subclinical infection on plasma retinol concentrations and assessment of the prevalence of vitamin A deficiency: meta-analysis. Lancet 362:2052-2058.[Medline]

6. Christian, P., Schulze, K., Stoltzfus, R. J. & West, K. P., Jr (1998) Hyporetinolemia, illness symptoms, and acute phase protein response in pregnant women with and without night blindness. Am. J. Clin. Nutr. 67:1237-1243.[Abstract]

7. Semba, R. D., Muhilal, , West, K. P., Jr, Natadisastra, G., Eisinger, W., Lan, Y. & Sommer, A. (2000) Hyporetinolemia and acute phase proteins in children with and without xerophthalmia. Am. J. Clin. Nutr. 72:146-153.[Abstract/Free Full Text]

8. Stoltzfus, R. J. & Underwood, B. A. (1995) Breast-milk vitamin A as an indicator of the vitamin A status of women and infants. Bull. WHO 73:703-711.[Medline]

9. Haskell, M. J. & Brown, K. H. (1999) Maternal vitamin A nutriture and the vitamin A content of human milk. J. Mammary Gland Biol. Neoplasia 4:243-257.[Medline]

10. Tanumihardjo, S. A. & Penniston, K. L. (2002) Simplified methodology to determine breast milk retinol concentrations. J. Lipid Res. 43:350-355.[Abstract/Free Full Text]

11. Green, M. H., Green, J. B., Akohoue, S. A. & Kelley, S. K. (2001) Vitamin A intake affects the contribution of chylomicrons vs. retinol-binding protein to milk vitamin A in lactating rats. J. Nutr. 131:1279-1282.[Abstract/Free Full Text]

12. Ross, A. C., Pasatiempo, A. M. & Green, M. H. (2004) Chylomicron margination, lipolysis, and vitamin A update in the lactating rat mammary gland: implications for milk retinoid content. Exp. Biol. Med. 229:46-55.[Abstract/Free Full Text]

13. Pasatiempo, A.M.G. & Ross, A. C. (1990) Effects of food or nutrient restriction on milk vitamin A transfer to neonatal vitamin A stores in the rat. Br. J. Nutr. 63:351-362.[Medline]

14. Semba, R. D., Kumwenda, N., Taha, T. E., Mtimavalye, L., Broadhead, R., Miotti, P. G., Eisinger, W., Hoover, D. & Chiphangwi, J. D. (2000) Plasma and breast milk vitamin A as indicators of vitamin A status in pregnant women. Int. J. Vitam. Nutr. Res. 70:271-277.[Medline]

15. Rice, A. L., Stoltzfus, R. J., de Francisco, A. & Kjolhede, C. L. (2000) Evaluation of serum retinol, the modified-relative-dose-response ratio, and breast-milk vitamin A as indicators of response to postpartum maternal vitamin A supplementation. Am. J. Clin. Nutr. 71:799-806.[Abstract/Free Full Text]

16. de Pee, S., Yuniar, Y., West, C. E. & Muhilal, (1997) Evaluation of biochemical indicators of vitamin A status in breast-feeding and non-breast-feeding Indonesian women. Am. J. Clin. Nutr. 66:160-167.[Abstract/Free Full Text]

17. Chappell, J. E., Francis, T. & Clandinin, M. T. (1985) Vitamin A and E content of human milk at early stages of lactation. Early Hum. Dev. 11:157-167.[Medline]

18. Stoltzfus, R. J., Hakimi, M., Miller, K. W., Rasmussen, K. M., Dawiesah, S., Habicht, J. P. & Dibley, M. J. (1993) High dose vitamin A supplementation of breast-feeding Indonesian mothers: effects on the vitamin A status of mother and infant. J. Nutr. 123:666-675.

19. Panpanich, R., Vitsupakorn, K., Harper, G. & Brabin, B. (2002) Serum and breast-milk vitamin A in women during lactation in rural Chiang Mai, Thailand. Ann. Trop. Paediatr. 22:321-324.[Medline]

This article has been cited by other articles:

				S. H. Gieng and F. J. Rosales Plasma {alpha}1-Acid Glycoprotein Can Be Used to Adjust Inflammation-Induced Hyporetinolemia in Vitamin A-Sufficient, but Not Vitamin A-Deficient or -Supplemented Rats J. Nutr., July 1, 2006; 136(7): 1904 - 1909. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Dancheck, B. Articles by Semba, R. D. Search for Related Content PubMed PubMed Citation Articles by Dancheck, B. Articles by Semba, R. D. Pubmed/NCBI databases Compound via MeSH Substance via MeSH Medline Plus Health Information Breast Feeding Hazardous Substances DB VITAMIN A