Cholestasis is defined as a reduced flow of bile leading to a decreased concentration of bile acids in the intestinal lumen and accumulation of bile substances in the blood. It affects about 1 of 2500 newborns and is most commonly reported in preterm infants [4]. Tufano et al. [5] showed 27 cases of cholestasis with 92.5% premature infants in a cohort of 1289 infants admitted to NICU, 80% of whom had a birth weight < 1000 g. Our case shows that, despite being a frequent findings in NICU, the diagnostic course can be prolonged and can require several invasive exams. Furthermore, frequent complications are malnutrition and macro-micronutrient deficiency due to increased energy expenditure, maldigestion of lipids and fat-soluble vitamins, diarrhoea, peripheral lipolysis and suboptimal protein synthesis in the liver and muscle [6]. The presence of cholestasis can lead to worsening of the prognosis and outcome in such complex patients. Niccum et al. [1] demonstrated a higher risk of cholestasis in preterm infants having lower Apgar scores and lower birth weight, in correlation with decreased weight percentile at discharge with subsequent impact on neurological outcome. Although cholestasis is typically multifactorial and transient, prognosis can be variable and often poor. Our patient's monochorionic twin brother presented higher birth weight and milder and transient cholestasis, he had hospitalization shorter than five months, better weight catch-up and neurological outcome.

The diagnosis of intrahepatic bile duct paucity (PIBD) is mainly histological: defined as loss of at least 50% of bile ducts in ten portal spaces, with correlation to clinical and biochemical findings [3]. Classically there are two forms: syndromic (associated with Alagille syndrome and arthrogriposis renal dysfunction and cholestasis—ARC syndrome) and non-syndromic secondary to chromosomal, infectious, inborn error of metabolism, endocrinological, myeloproliferative or idiopathic causes [3]. Studies have shown that the bile ducts/portal tracts (BD/PT) ratio progressively increases in the postnatal period during the first weekes of life, making difficult to diagnose PIBD before 38 weeks of postconceptional age [7]. The clinical presentation of our patient was similar to that of the series reported in the literature (Table 1) with jaundice, acholic stools and growth retardation, he showed resolution of cholestasis at 8 months of age.

A second liver biopsy was not done in our patient, but the clinical trend suggests that growth led to an improvement in hepatic and biliary damage. Meena et al. [3] reported that some factors such as intrauterine failure to thrive or sepsis may lead to PIBD in premature infants due to interference with the developing ductal system. It could be that prematurity constitutes a spectrum of this pathology with its characteristics and prognosis, future studies will be needed to analyze this particular subpopulation of patients in more detail. Besides, the prognosis and progression of NSPBD have been widely debated. Past studies reported the non-syndromic forms of NSPBD as the type with the worst prognosis [3]. Chiu et al. [9] reported that one patient of 4 non-syndromic patients survived, during follow-up he presented delayed growth and became jaundice-free after the age of 5 years. In the cohorts analysed (Table 1), only 8 cases reported fatal outcome, but frequent and severe complications such as liver and kidney failure, chronic liver disease and cirrhosis were described and can define the poor prognosis of this condition. Indeed, the belief that the non-syndromic variety progresses more towards cirrhosis than the syndromic variety [14], has been recently contested [13]. Associated familial/metabolic conditions may influence morbidity, mortality and prognosis [11]. Diagnosis of non-syndromic form should not be considered as unfavorable prognosis, as the evolvement should be variable and related to the etiology in this form of presentation [8, 10, 12].

Despite different nutritional strategies and complete coverage of energy requirements, our case, born with severe intrauterine and postnatal growth retardation, continue to present poor weight gain and abdominal symptoms. Strategies adopted included increasing meal frequency and feeding volume, supplementing with medium-chain triglycerides, using specific formulas based on whole milk proteins with BCAAs, intended for hepatopathic patients and supplementing with fat-soluble vitamins. Struggling with hypoglycaemia and poor growth, short-chain polymer (maltodextrin) supplementation has been proposed at 23 weeks of postnatal age (9 weeks of corrected age). Unfortunately, after few weeks our patient presented severe pain crises which required hospitalization, so that different dietary modifications were performed to achieve food tolerance. Maltodextrins were initially discontinued, but abdominal symptoms persisted until the formula was changed with casein based extensively hydrolysed formula, with high percentage of MCT (55% of total lipids). Afterwards maltodextrins were reintroduced, while maintaining a diet free of milk proteins, diarrhoea, bloating and abdominal pain promptly reappeared.

Carbohydrate supplementation such as maltodextrins and starch have been proposed as possible alternatives in patients with cholestasis at high risk of hypoglycaemia, also taking into account that the increase in lipid supplementation is limited due to poor absorption [15]. However, studies in preterm animals have shown that supplementation with maltodextrins leads to alterations in the bacterial flora with production of gas and metabolites such as short-chain acids resulting in an increased risk of necrotizing enterocolitis (NEC), currently, no studies have investigated efficacy and safety of maltodextrins in infants and older children [16] and an adverse reaction had never been described in this population. It should be considered that severe abdominal bloating and diarrhea can occur, because of maldigestion due to the enzymatic immaturity of amylases at an early age [2].

In our patient persistence of adbominal symptoms after maltodextrins discontinuation and symptoms relief with the introduction of extensively hydrolyzed based formula are suggestive of a non-IgE mediated cow milk allergy (CMA), more than maldigestion. However, our case has some limitations: the main one is that an oral food test was not performed to confirm the diagnosis of non-IgE CMA, as the parents refused to do it [17]. The second limitation was that intake of different sources of maltodextrin have not been studied.

Our description reports a rare case of non-syndromic bile duct paucity. To our knowledge, this is the first case reported of maltodextrins intolerance in a preterm infant, complicated by a non-IgE mediated cow-milk allergy, which made the nutritional management even more difficult to deal with and interpret. Our case suggests the importance of a multidisciplinary approach and possible switch to an hypoallergenic formula, based on extensively hydrolyzed proteins, in case of symptoms persistence with the use of whole protein based formula intended for cholestasis. Moreover, we suggest caution in supplementing with carbohydrates infants and careful monitoring adverse reaction during the dose increase for possible complication. There are several gaps and lack of studies on the efficacy of different strategies in paediatric patients with cholestasis, also regarding maltodextrin supplementation. Commercial preparations for hepatopatic patients contain higher concentrations of MCTs (30–50%) and BCAAs, but personalized strategies must be tailored to each patient based on comorbidities, clinical response, and specificrequirements.