Nutrition in early childhood affects health and disease risk in later life( Reference Lucas 1 ). In particular, protein intake, both in quantity and in quality, during the first 2 years of life is recognized as having an important role in growth, neurodevelopment and long-term health( Reference Michaelsen and Greer 2 ).
Breast-feeding is an unequalled way of providing optimal food for infants’ healthy growth and development and the WHO recommends that infants should be exclusively breast-fed for the first 6 months of life. For mothers who are unable to breast-feed or who decide not to, infant formulas are a safe alternative( 3 , 4 ). Despite recommendations, it is possible that parents make wrong nutritional choices for their children because of cultural beliefs or misinformation on infant nutrition.
Goat’s milk is not approved for use in infants( 4 ), because its protein and potassium load is about double and the chloride load four times the amount present in most infant formulas( Reference Hendriksz and Walter 5 ). The consumption of undiluted goat’s milk diets by infants has been associated with metabolic acidosis and severe electrolyte imbalances. Moreover, reports have also described associated morbidities such as acute stroke, megaloblastic anaemia, infections, haemolytic uremic syndrome and allergic reactions including life-threatening anaphylactic shock( Reference Hendriksz and Walter 5 – Reference Chapman, Ganesh and Ficicioglu 8 ). Newborn screening (NBS) results close to those observed in patients affected by tyrosinaemia( Reference Hendriksz and Walter 5 , Reference Techakittiroj, Cunningham and Hooper 7 ), a disorder of the catabolic pathway of the amino acid tyrosine, or maple syrup urine disease( Reference Chapman, Ganesh and Ficicioglu 8 ), a disorder affecting the metabolism of branched-chain amino acids, have also been reported in infants fed goat’s milk.
We report the case of a newborn fed with undiluted goat’s milk who was recalled for a positive NBS result, characterized by severe hypertyrosinaemia with high methionine and phenylalanine levels, a pattern indicative of severe liver impairment( Reference Morgan, Marshall and Milsom 9 ).
Case presentation
Born at 40 weeks of gestational age, after an uneventful pregnancy, our patient had a birth weight of 2480 g (3rd percentile). In the first 24 h of life she presented with transient clotting abnormalities treated with plasma transfusion. The NBS performed at 36 h of life was not valid because of the transfusion and, in line with standard procedures, it was repeated at 9 d of life. At that time, the newborn was growing and feeding well with breast milk and was discharged.
NBS was also repeated at 14 d of life as a standard procedure for low birth weight and showed high tyrosine, phenylalanine and methionine levels (Fig. 1). In particular, tyrosine was greater than 1000 μmol/l (normal range: 46·5–250 μmol/l) on dried blood spot. Succinylacetone (SUAC), a specific marker of tyrosinaemia type 1, was normal.
Fig. 1 Tyrosine (Tyr; 
), phenylalanine (Phe; 
) and methionine (
) levels on dried blood spots at the patient’s main checks. Nutritional diets are also reported in relation to dried blood spot levels
After the confirmation of NBS results at 23 d of life, a severe liver disease was suspected and the newborn was recalled for examination and further investigation. At admission, clinical examination was normal. Initial testing results were consistent with moderate metabolic acidosis (
$${\rm HCO}_{3}^{{\minus}} $$
17 mmol/l, base excess 7 mmol/l, normal anion gap). Liver function values and electrolytes were within normal limits. Plasma tyrosine was above 2000 μmol/l (normal plasma value: 55–250 µmol/l), plasma phenylalanine and methionine were increased to 427 µmol/l (normal plasma value: 39–90 µmol/l) and 328 µmol/l (normal plasma value: 11–40 µmol/l), respectively. Tyrosine, phenylalanine and methionine levels on dried blood spot were 2463 μmol/l, 352 μmol/l and 357 μmol/l, respectively.
Organic acids analysis on urine confirmed the absence of SUAC. The patient was started on a tyrosine- and phenylalanine-free formula feed. The acidosis was quickly corrected and did not recur. Blood tyrosine, phenylalanine and methionine normalized in 48 h.
Dietary history revealed that, after the first discharge, the patient was fed undiluted goat’s milk plus small volumes of breast milk as breast-feeding was considered not sufficient. The parents did not verify the lack of breast milk and did not search support of a paediatric service because they did not consider infant feeding a complex task. The parents, both of Romanian origin, had lived in Italy for a few months in a restricted community of Romanian people. The mother was a 20-year-old housewife and she was not able to speak Italian or English. The father was a 22-year-old farmer and he seemed to understand Italian. The father stated that they chose goat’s milk in a superficial manner because they considered it digestible and suitable for infants.
We estimate that the patient has received three to four times more protein than is recommended (Table 1). Her parents were really surprised to know that goat’s milk is potentially harmful for infants. They did not know that it has a very different composition compared with human milk and infant formula( 10 ).
Table 1 Comparison of nutritional intake between our patient and a breast-fed or standard formula-fed newborn of same weight (3·2 kg) and fluid intake (550 ml/d). The source of nutritional information is reported in parentheses
USDA, US Department of Agriculture. *Professional nutrition facts of Infant formula Mellin 1. Data about the phenylalanine and tyrosine content are not available online.
After our evaluation, the infant was discharged and specific and precise feeding advice was given to the family. The patient is now 4 months old and she is fed with infant formula. She is gaining weight within a normal range and her development is normal.
Discussion
Health outcomes in developed countries differ substantially for infants who formula-feed compared with those who breast-feed. In fact, formula-fed infants face higher risks of infectious morbidity, childhood obesity and chronic disease( Reference Stuebe 11 ).
Our experience demonstrates that, when breast-feeding is not possible, infants may be also at risk of dangerous nutritional practices, such as inappropriate high-protein diets.
High-protein diets in infancy have an important effect on plasma amino acid levels, particularly tyrosine. This is demonstrated by the evidence that infant tyrosine levels have fallen dramatically in the last four decades because of concomitant increase in breast-feeding rates and reduction in the protein content of commercial infant formulas.
High tyrosine levels have been associated in the long term to intellectual deficits with variable degrees of developmental retardation, eye symptoms and skin lesions. Severe neurological manifestations, including microcephaly, seizures, tremors and behavioural difficulties, have been also described( Reference Hendriksz and Walter 5 – Reference Chapman, Ganesh and Ficicioglu 8 ).
At the time of evaluation our patient was clinically asymptomatic and only a mild metabolic acidosis was detected. It is difficult to predict what would have happened if the baby had continued to take goat’s milk. If NBS had not been repeated during the high protein administration, it is possible that the inappropriate diet of our patient would not have been precociously detected.
Since NBS is usually performed between 36 and 48 h of life when the baby is still in hospital, NBS programmes may not identify high-protein diets unless specific follow-ups are requested. Moreover, NBS is performed in only a few countries.
Our case highlights that, when breast-feeding is not possible or is not exclusive, infants may be at risk of potentially harmful diets. To reduce the risks of not breast-feeding, it is important to promote breast-feeding support groups for referral of mothers to them on discharge from the hospital. Infant feeding should be assessed by hospital and community paediatric services at every clinical evaluation, not only for promoting breast-feeding, but also for educating and counselling families about the risks associated with alternative diets. Particular attention may be required to families belonging to a small foreign community. When breast-feeding is not possible or is not exclusive, families may need appropriate advice on safe alternatives for infant nutrition to avoid dangerous nutritional practices.
Acknowledgements
Acknowledgements: The authors are grateful to Mr Giorgio Mazzoleni and Mrs Helena Andersson for editing the English version of the manuscript. Financial support: This research received no specific grant from any funding agency in the public, commercial or not-for-profit sectors. Conflict of interest: The authors declare that they have no competing interests. Authorship: E.M. conceived the work and wrote the draft; G.G. performed dietetic evaluation; E.T. and A. Pietrobelli revised the manuscript critically for important intellectual content; N.C., A. Pasini, F.I.-P., M.V., F.T. and M.C. performed expanded NBS and amino acids analysis; A.B. ensured the accuracy of the data and revised the manuscript critically. Ethics of human subject participation: All procedures were performed in compliance with the Declaration of Helsinki.
