Eur J Pediatr (2007) 166:189194 DOI 10.

1007/s00431-006-0218-9

ORIGINAL PAPER

Zinc-deficiency dermatitis in breast-fed infants

Antonia Kienast & Bernhard Roth & Christiane Bossier & Christina Hojabri & Peter H. Hoeger

Received: 7 November 2005 / Revised: 8 May 2006 / Accepted: 8 May 2006 / Published online: 8 September 2006 # Springer-Verlag 2006

Abstract We report ten infants (mean gestational age: 30 weeks; range: 25 to 40 weeks) with zinc deficiency dermatitis who developed erosive, impetiginized periorificial dermatitis at 10 weeks of age (corresponding to a mean gestational age of 41.4 weeks, with a range of 36 44 weeks), but who were otherwise well. Cutaneous symptoms were initially misdiagnosed as eczema or impetigo in 8/10 (80%) children who received either topical (4/8) and/or systemic (6/8) antibiotics. Topical corticosteroids were applied in 4/10 infants for a mean time of 4 weeks (range: 2 to 5 weeks) before the correct diagnosis was established by decreased serum zinc levels; skin atrophy (telangiectasia, thinning) as a complication of topical steroid treatment (class II steroids) was observed in two infants. All children responded to oral therapy with zinc sulfate or zinc gluconate (1.54 mg/kg/d). Skin lesions started to clear within 24 h after the initiation of therapy and had completely cleared in all infants after 14 days of therapy (range: 314 days). We conclude that nutritional zinc deficiency is a frequently misdiagnosed problem in thriving, fully breast-fed preterm babies. It is attributable to the decreased zinc content of human milk as compared to
A. Kienast : P. H. Hoeger (*) Catholic Childrens Hospital Wilhelmstift, 22149 Liliencronstr. 130, Hamburg, Germany e-mail: hoeger@kkh-wilhelmstift.de C. Hojabri : P. H. Hoeger Department of Dermatology, Division of Paediatric Dermatology, University of Hamburg, Hamburg, Germany B. Roth : C. Bossier Department of Neonatology, University of Cologne, Childrens Hospital, Cologne, Germany

cows milk, and the increased demand of zinc in rapidly thriving preterm infants. It seems advisable to routinely check serum zinc levels in fully breast-fed preterm infants who do not receive regular oral zinc supplementation once they reach a gestational age of 40 weeks. Keywords Zinc deficiency . Acquired . Breast milk . Preterm infants . Dermatitis

Introduction The trace element zinc serves as a cofactor in a variety of biochemical reactions. It is required for normal keratinisation [19] and serves as a cofactor for many enzymes of the gastrointestinal tract (e.g., pancreatic protease) and bone metabolism (e.g., alkaline phosphatase) [21]. Zinc is also important for immune function, affecting proliferation of T- and B- lymphocytes, baseline and antibody dependent cytotoxic activity of killer lymphocytes, and phagocytotic and bactericidal capacities of neutrophils [2]. Consequently, zinc deficiency may cause multiple dysfunctions in various organ systems. Due to their rapid growth, zinc requirement is particularly high in young, and especially in preterm, infants [11]. Alimentary zinc deficiency dermatitis has been reported in infants receiving long term parenteral nutrition and in preterm infants who were exclusively breast-fed. It is characterized by periorificial and acral dermatitis; unlike acrodermatitis enteropathicaa rare genetic disorder caused by mutations in the SLC39A4 (hZIP4) gene on chromosome 8q24.3 [15]chronic diarrhea, failure to thrive, irritability, alopecia, nail dystrophy or delayed wound healing [1, 9, 14] are rarely observed. Furthermore, acrodermatitis enteropathica develops in bottle-fed infants

190

Table 1 Nutritional and baseline characteristics of our patients Birth weight Enteral nutrition Postnatal age at onset of skin symptoms (weeks) 18 43 20 2,465 307 268 548 10 (weeks) (weeks) (g) Breast-milk zinc levels at diagnosis [normal range: 7842416]* (g/l) Serum zinc levels of the mother [normal range: 600900] (g/l) Complete resolution of skin symptoms after zinc supplement (days) 3 Gestational age at onset of skin symptoms Postnatal age at diagnosis Weight at diagnosis Dose of zinc supplement

Case no.

Gender

Gestational age

(weeks) 580

(g)

Serum zinc levels at diagnosis [normal range: 7201570] (g/l)

mg/kg/d

25

26

920

14

40

16

2,275

300

10

2.5

27

860

16

43

18

3,360

163

325

770

1.5

27

720

16

43

18

3,560

163

163

416

27

640

36

11

1,800

410

14

6 1,160 1,430 2,560 3,520 4 44 20 5 42 12 8 40 12 14 44 17 3,320 3,660 4,920 7,500

29

1,380

10

39

16

4,800

567 159 354 232 206

10 3 5 5

4 3 4 3 3

30

32

37

10

40

breastmilk + fortifier breastmilk + fortifier breastmilk + fortifier breastmilk + fortifier breastmilk + fortifier breastmilk breastmilk breastmilk breastmilk breastmilk

Eur J Pediatr (2007) 166:189194

*at 20th week of lactation

Eur J Pediatr (2007) 166:189194

191

during or after weaning, and breast milk is considered prophylactic. Here we describe ten cases of zinc deficiency in exclusively breast-fed infants who were initially misdiagnosed (and mistreated) as impetigo or eczema.

Case reports We saw ten infants (six male and four female) who developed zinc deficiency dermatitis. All children had erythematous, erosive and partly impetiginous skin lesions in the face. Seven patients also showed anogenital lesions and four infants had erythematous and erosive skin lesions of the finger pulps. Nine of ten patients were preterm babies (mean gestational age: 30 weeks; range: 25 to 40 weeks; mean birth weight: 1,377 g; range: 5803,520 g). Symptoms occurred at a mean postnatal age of 11.4 weeks (range: 4 to 18 weeks), corresponding to a mean gestational age of 41.4 weeks (range: 36 to 44 weeks). Diagnosis was established at a mean postnatal age of 16 weeks (range: 11 to 20 weeks).

All children were fully breast-fed by the time the symptoms occurred and four of them also received a protein and mineral milk fortifier with the usual amount of zinc (200 g additional zinc/100 ml of milk). All patients showed low serum zinc levels (<720 g/l). Breast milk and serum zinc levels were evaluated in three mothers; in all of them breast milk zinc was lower than the normal range in the 20th week of lactation (<784 g/l). Two mothers had low serum zinc levels (<600 g/l); one mother had a normal serum zinc level, but a low breast milk level. All children responded to oral therapy with zinc sulfate or zinc gluconate (1.54 mg/kg/d). Skin lesions started to clear within 24 h after the initiation of therapy and had completely cleared in all infants after 14 days of therapy (range: 314 days). Cases 110 are summarized in Table 1. Figure 1 illustrates case 8 before (Fig. 1a and b) and after (Fig. 1c and d) zinc supplementation, respectively. Figure 2 correlates individual growth rates and onset of symptoms (Fig. 2a: chronological age; Fig. 2b: gestational age).

Fig. 1 Case no. 8. a, b Before treatment. c, d Four days after initiation of oral zinc supplementation (4 mg/kg BW/d). Note signs of skin atrophy on the cheeks (c) following topical steroid treatment prior to diagnosis

192 Fig. 2 Weight curves. a Patients weight gain (birth weight=1) vs. postpartal ages. b Patients weight gain (birth weight=1) vs. gestational ages. Note that the onset of clinical symptoms correlates obviously much more with gestational than with chronological age

Eur J Pediatr (2007) 166:189194

Discussion We present a series of ten fully breast-fed and thriving infants who developed zinc deficiency dermatitis at the age of 420 postnatal weeks. Zinc deficiency results from an increased demand on the side of the rapidly growing infant and decreased supply by the breast-feeding mother. Although this constellation of alimentary zinc deficiency has been described as early as 1979 [1, 16], virtually all children presented here had been misdiagnosed initially and had received inappropriate topical and oral therapy for periods of up to 5 weeks. Demand for zinc is increased during periods of rapid growth, and it is therefore particularly high in preterm

infants. Reduced intestinal absorption of zinc in preterm infants may also play a role in zinc deficiency, but this is unlikely as patients exhibit rapid weight gain and do not have deficiencies of other trace elements. Furthermore, zinc loss by urine, faeces and skin is much greater than that observed in mature infants. Zinc is essential for the structure and function of a large number of macromolecules. It represents an essential cofactor for more than 300 different enzymes where it has both catalytic and structural roles [25]. Several Zn-metalloenzymes are of particular importance for the regulation of homoiostasis (or the induction of disease) in preterm infants: carboanhydrase (metabolic acidosis), DNA- and RNA-polymerases

Eur J Pediatr (2007) 166:189194

193

(impaired growth), alkaline phosphatase (osteopenia praematurorum), superoxide dismutase (bronchopulmonary dysplasia) and pancreatic protease (impaired digestion) [20]. Decreased levels of alkaline phosphatase are therefore indicative of serum zinc deficiency. Zinc levels of mature human breast milk are lower than those of cows milk (1.18 mg/l vs 3.9 mg/l), and they further decrease during lactation [5, 22, 23]. The relative decrease of zinc levels in breast milk is to a certain extent counterbalanced by the fact that human milk apparently facilitates absorption of zinc by mechanisms still unidentified [8]. Preterm formula milk contains higher zinc levels (59.3 mg/l), but some breast milk fortifiers and infant formulas currently used in Germany do not appear to meet the demands of rapidly growing extremely low birth weight infants for zinc and other trace elements during the first months of life [20]. Low breast milk zinc in mothers of preterm babies, whose milk showed normal zinc levels in the following (term) babies, led to the hypothesis that low zinc levels in breast milk may be due to immaturity of maternal milk produced after pre-term delivery. In other investigations zinc concentrations in human milk are comparable between preterm and term mothers. Others suggested that decreased breast milk zinc often occurs in mothers with normal serum zinc levels and results from a defect in transfer of zinc from maternal serum to breast milk [3, 8, 20]. Since breast milk zinc concentration at the beginning of lactation is 34 times higher than the serum concentration there must be an active transport mechanism [13]. Zinc-transporter proteins were identified in 1997/98 [10, 17]. Iron and folate supplements, as well as calcium substitution, which are often routinely prescribed during pregnancy, seem to decrease the oral bioavailability of zinc in pregnant women [24]. Mothers with small-for-gestational-age babies consumed significantly less zinc than mothers of appropriate-for-gestational-age babies [24]. Mature neonates are apparently equipped with a sufficient zinc reserve accumulated during the third trimester of pregnancy [22]. The risk of zinc deficiency is therefore increased by conditions adversely affecting hepatic storage during fetal development, such as prematurity [7]. Interestingly, gestational age was negatively correlated to the postnatal age at onset of skin symptoms (r=0.86). Thus, the mean gestational age at onset of skin symptoms was 41.4 weeks (range: 3644 weeks). This might be the period of time where zinc demand is highest. This is evidenced by a comparison of chronological and gestational age (Fig. 2). With respect to very low breast milk zinc concentrations, zinc supplementation for fully breast-fed premature infants seems to be mandatory. Zinc substitution of mothers during pregnancy and/or lactation would have no effect on those children whose mothers have normal serum zinc levels in

spite of reduced milk zinc levels (one out of three in our patients). During rapid physical growth, full-blown zinc deficiency syndrome can develop even when fortified milk is added (case I, II). Milk fortifiers with higher amounts of zinc are available (e.g. a modified formulation of FM 85, Nestl, Munich, Germany, 5 g per 100 ml breast milk=500 g additional zinc per 100 ml milk). We conclude that alimentary zinc deficiency has an excellent prognosis and, in contrast to acrodermatitis enteropathica, oral zinc supplementation is only temporarily needed. Nutritional zinc deficiency is a frequently misdiagnosed problem in thriving, fully breast-fed preterm, and also occasionally in full term infants. The majority of children are initially misdiagnosed and treated for eczema or impetigo with local steroids and/or antibiotics, which may lead to severe complications [4, 18]. Oral zinc supplementation of fully breast-fed preterm babies should therefore be considered [6, 12]. Alternatively, it seems advisable to routinely check serum zinc levels in fully breast-fed preterm infants who do not receive regular oral zinc supplementation once they reach a gestational age of 40 weeks. References
1. Aggett PJ, Atherton DJ, More J, Davey J, Delves HT, Harries JT (1980) Symptomatic zinc deficiency in a breast-fed preterm infant. Arch Dis Child 55:547550 2. Antila PH, von Willebrand E, Simell O (1986) Abnormal immune responses during hypozincemia in acrodermatitis enteropathica. Acta Pediatr Scand 75:988992 3. Atkinson SA, Whelan D, Whyte RK, Lnnerdal B (1989) Abnormal zinc content in human milk. Am J Dis Child 143:608611 4. Brar BK, Pall A, Gupta RP (2003) Acrodermatitis enteropathicalike rash in exclusively breast fed infant with zinc deficiency. J Dermatol 30:259260 5. Casey CE, Neville MC, Hambidge KM (1989) Studies in human lactation: secretion of zinc, copper, and manganese in human milk. Am J Clin Nutr 49:773785 6. Drea JG (1993) Is zinc a first limiting nutrient in human milk? Nutr Res 13:659666 7. Drea JG (2003) Zinc deficiency in nursing infants. J Am Coll Nutr 21:8487 8. Duncan JR, Hurley LS (1978) Intestinal absorption of zinc: a role for a zinc binding ligand in human milk. Am J Physiol 235E: 556559 9. Heinen F, Matern D, Pringsheim W, Leititis JU, Brandis M (1995) Zinc deficiency in an exclusively breast-fed preterm infant. Eur J Pediatr 154:7175 10. Huang L, Kirschke CP, Gitschier J (2002) Functional characterization of a novel mammalian zinc transporter, ZnT6. J Biol Chem 227:2638926395 11. Krebs NF, Hambidge KM (1986) Zinc requirements and zinc intakes of breastfed infants. Am J Clin Nutr 43:288292 12. Krebs NF, Westcott J (2002) Zinc and breastfed infants: if and when is there a risk of deficiency? Adv Exp Med Biol 503:6975 13. Krebs NF (1999) Zinc transfer to the breastfed infant. J Mammary Gland Biol Neopl 4:259268

194 14. Kuramoto Y, Igarashi Y, Tagami H (1991) Acquired zinc deficiency in breast-fed infants. Semin Dermatolog 10:309312 15. Kury S, Dreno B, Bezieau S, Giraudet S, Kharfi M, Kamoun R, Moisan JP (2002) Identification of SLC39A4, a gene involved in acrodermatitis enteropthica. Nature Genet 31:239240 16. Leigh IM, Sanderson KV, Atherton DJ, Wells RS (1979) Hypozincemia in infancy. Br J Dermatol 101:7375 17. MacMahon RJ, Cousins RJ (1998) Mammalian zinc transporters. J Nutr 128:667670 18. Munro CS, Lazaro C, Lawrence LM (1989) Symptomatic zinc deficiency in breast-fed premature infants. Zinc deficiency in rapidly growing preterm infants. Br J Dermatol 121:773778 19. Niemi KM, Anttila PH, Kanerva L, Johansson E (1989) Histopathological study of transient acrodermatitis enteropathica due to decreased zinc in breast milk. J Cutan Pathol 16:382387

Eur J Pediatr (2007) 166:189194 20. Obladen M, Loui A, Kampmann W, Renz H (1998) Zinc deficiency in rapidly growing preterm infants. Acta Pediatr 87:685691 21. Prasad AS (1985) Clinical, endocrinological, and biochemical effects of zinc deficiency. Spec Top Endocrinol Metab 7:4576 22. Shaw JC (1979) Trace elements in the fetus and young infant. I. Zinc. Am J Dis Child 133: 12601268 23. Sievers E, Oldings H-D, Drner K, Schaub J (1992) Longitudinal zinc balances in breast-fed premature infants. Brit J Dermatol 121:773778 24. Simmer K, Thompson RB (1985) Zinc in fetus and newborn. Acta Paediatr Scand Suppl 319:158163 25. Tapiero H, Tew KD (2003) Trace elements in human physiology and pathology: zinc and metallothioneins. Biomed Pharmacother 57:399411