CMPA is an adverse reaction to cow milk proteins mediated by immune mechanisms, and the incidence of CMPA is estimated to be 0.5–3% in developed countries within one year of age [7–10]; the incidence of CMPA is approximately 2.69% in China [11]. Neonates constitute a special group due to their immature immune mechanisms and other factors. The incidence of milk protein allergy is relatively low, its clinical manifestation lacks specificity, and it is often misdiagnosed as necrotizing small bowel colitis and sepsis in neonates—leading to unnecessary surgery or antibiotic use[12]. Although reports of milk protein allergy in neonates have been increasing in recent years, there is a lack of definitive epidemiologic data. Morita et al. [13] depicted an incidence of CMPA of about 1.98% among 2,116 neonates in neonatal intensive care units (NICUs), with their multicenter analysis revealing that 69,796 were hospitalized. A multicenter study also showed that the incidence of CMPA was 0.21% in 69,796 hospitalized neonates [14]. Lin et al.[15] reported that CMPA allergy was diagnosed in 160 (53.3%) of 300 hospitalized children with bloody stools in the neonatal unit of their hospital, of which 112 (51.1%) were full-term infants and 48 (59.3%) were preterm infants. Pssariello et al. [16] analyzed 39 cases of diarrhea in their NICU, of whom 20.5% were caused by food allergy, and Miyazawa et al. [17] observed that among 263 NICU infants, 53 cases showed CMPA (with a mean onset age of 6 d) and 41% of them were low-birth-weight infants. In the present study, five cases were full-term infants, and we showed that the onset of disease in full-term infants was 1–2 weeks; the onset of disease in ultra-low-birthweight infants was also significantly later than in low-birthweight infants (LBWI) or normal birthweight infants [14, 17]. There are also reports of allergic colitis in the first few hours and one day after birth[18]. In the present study, the children showed disease onset at three to 28 days, and it was evident that the age at onset of disease could be in either the early or late neonatal period. It is hypothesized that neonates with milk protein allergy have a history of eating dairy products one week before the onset of the disease. In our analysis, all neonates had eaten dairy products, including three cases with artificial feeding and two cases with mixed feeding. We suggest that infants administered artificial feeding may be more prone to severe milk protein allergy.
The clinical manifestations of neonatal CMPA are primarily gastrointestinal symptoms such as diarrhea, bloody stools, vomiting, abdominal distension, and diminished milk intake—some of which are accompanied by skin symptoms such as eczema and atopic dermatitis; respiratory symptoms are less common. Depending on the severity, neonatal food allergy can be categorized as mild-moderate and severe [19]: mild-moderate digestive symptoms may be manifested as diarrhea, bloody stools, abdominal distension, vomiting, colic, and, in a few cases, constipation, and these symptoms may occur alone or in combination. In addition to gastrointestinal symptoms, chronic diarrhea and vomiting can cause growth retardation, feeding difficulties, nutritional iron-deficiency anemia, protein-losing enteropathy, and severe ulcerative colitis in some cases. Siu et al.[20] reported a case of chronic diarrhea in a Chinese newborn due to CMPA, and this was followed by severe developmental disorders, metabolic acidosis, hypoalbuminemia, severe atopic dermatitis combined with hypoalbuminemia, and severe diarrhea. Atopic dermatitis combined with hypoalbuminemia or combined with growth restriction or iron-deficiency anemia, circulatory failure, and shock may also be present. In our study, of five cases, four children presented with diarrhea and one with bloody stools showing vomiting, fever, malnutrition, edema, and skin ulceration. Laboratory tests showed anemia, hypoproteinemia, elevated leukocytes, and CRP. Gastroenteroscopy suggested various degrees of mucosal inflammation, including esophagitis, pyloric stenosis, lymphatic dilatation of the small intestine, erosive gastritis, erosive colitis, and colonic ulceration with polypoid hyperplastic changes. It is acknowledged that children with severe disease often have anemia, hypoproteinemia, elevated inflammatory indices, differing degrees of inflammation upon gastroenteroscopy, hyperplastic stenosis, and even erosive ulceration that is similar to the manifestation of Crohn’s disease—which is prone to misdiagnosis.
The diagnosis of neonatal CMPA necessitates being conducted in combination with clinical manifestations, laboratory tests, and the exclusion of infections and surgical diseases. Once neonatal CMPA is considered, a milk avoidance test, oral provocation test, specific IgE assay, skin prick test (SPT), patch test (atopy patch test [APT]), and eosinophil count are performed to confirm the diagnosis [21]. But a multicenter survey study in the UK surfaces that 84% of the current diagnoses of neonatal CMPA are clinical diagnoses only[22].Milk avoidance reflects both diagnostic and therapeutic value, and the foods consumed during the avoidance period should be extensively hydrolyzed formula (eHF), AAF, or soy-based formula [21]. AAF is preferred for those babies presenting with severe allergic or life-threatening manifestations, and soy-based formulas are preferred for those babies and families who cannot tolerate the bitter taste or cost of eHF. If symptoms do not improve within two weeks, only AAF should be chosen. Further diagnosis requires oral provocation testing that is potentially risky and requires appropriate resuscitative preparations. Soy-based formulas are also not implemented in neonates. In our study, three of the five children showed positivity for milk IgE, one had an elevated eosinophil count in the colonic mucosa, and one exhibited a positive oral provocation test. The remaining four children presented with severe gastrointestinal lesions that were difficult to treat with recurrent illnesses, and the provocation test was not performed after avoidance of the diet for 2–4 weeks for safety reasons; thus, all were diagnosed therapeutically. Three children were relieved by switching to amino acid formula feeding and one by switching to lactose-free deeply hydrolyzed formula, and one improved after the mother avoided the diet.
For the treatment of CMPA, the WAO 2010 Food Allergy Guidelines reflected a comparison of the dietary treatment guidelines for CMPA in different countries and organizations, and the common points were that breastfeeding was encouraged, but mothers needed to avoid high-risk foods, and dietary substitutions were recommended vis-à-vis amino acid formula or eHF [23]. According to China’s infant milk protein allergy guidelines, formula-fed children with mild-moderate degrees of milk protein allergy should be given eHF, and if milk protein allergy still occurs after consumption of eHF, then the amino acid formula should be replaced; in severe cases, amino acid formula is preferred [23], and avoidance time is at least six months, or until nine to 12 months of age. In the present study, four of five children with severe CMPA were fed with amino acid formula, and lactose-free eHF was administered in one case due to a diagnosis of lymphatic dilatation of the small intestine and recurrent hypoproteinemia. The avoidance time for four of the children was over six months and as long as two years; this was mainly due to obvious gastroenteroscopic inflammation and manifested by hyperplasia and stenosis, erosions, or ulcers. It also took a much longer period of time for the intestinal mucosa to return to normal. One child with recurrent hypoproteinemia exhibited dietary avoidance for two months, but could be transitioned to partially hydrolyzed milk powder feeding and thus recovered more quickly. This suggests that children with severe CMPA often require a longer avoidance time and that gastroenteroscopy is a reliable examination modality that can be used for evaluating the condition and also to guide feeding.
Milk protein allergy in newborns is not uncommon, often reflecting gastrointestinal symptoms as the chief manifestations, including diarrhea, bloody stools, and vomiting. Some of the children with severe milk protein allergy also showed anemia, hypoproteinemia, malnutrition, and other complications; the condition was recurring and the treatment was complicated. Therefore, early gastroenteroscopy is considered an excellent assessment of the scope of the lesions in gastrointestinal inflammation and their severity within the gastrointestinal tract; these symptoms can accumulate in many parts of the gastrointestinal tract, including the esophagus, stomach, duodenum, and colon. The jejunum and ileum, however, are not visualized due to endoscopic limitations. The endoscopic presentation is variable and may consist of hyperplastic strictures, erosions, or even ulcers. Severely affected children are additionally prone to misdiagnosis of sepsis, NEC, and very-early-onset inflammatory bowel disease. Early recognition of CMPA during the neonatal period can thereby avoid misdiagnosis and unnecessary antibiotic or surgical treatment, shorten treatment time, and improve prognosis. Whether early intervention can avoid the progression of gastrointestinal inflammation, however, requires further study.