Food allergies (FAs) are defined as immune-mediated adverse reactions to food proteins caused by the lack of development or the breakdown of immunological tolerance to food (1, 2). They are broadly categorized into IgE-mediated reactions (e.g., food-induced anaphylaxis), mixed IgE and non-IgE mediated reactions (e.g., eosinophilic gastrointestinal disorders, EGID), and non-IgE-mediated (cell-mediated) reactions (e.g., food protein-induced enterocolitis syndrome, FPIES) (1).

As a consequence of the increase in the rate of organ transplantation, previously non-allergic transplant recipients have been increasingly recognized as being at higher risk for developing de novo FA in comparison to the general population (3–6). Moreover, children undergoing liver transplantation (LT) have resulted to be far more affected by FA when compared to all the other transplant groups (5, 7–10).

Little is known about the pathogenesis underlying the occurrence of de novo FA after transplantation. Two main mechanisms have been hypothesized so far. The first consists of the passive transfer of allergen-specific IgEs and immune cells from the donor to the recipient at the time of transplant. Although this mechanism has been proven in several anecdotal cases, the inherent transience of this immunological transfer does not explain the development of FAs that are absent in the donor, nor their long-term persistence after transplant (6, 11–13). The second postulated mechanism consists of a loss of immune tolerance to orally ingested food antigens induced by calcineurin inhibitors, responsible for a Th2/Th1 imbalance with predominance of Th2 over Th1 responses (4, 6, 9, 14–16).

Since the first description of a peanut allergy transferred from the donor to the recipient of a combined liver-kidney transplant in 1997 (13), there has been a growing interest regarding de novo FA after LT and a parallel surge in the number of published papers. However, comprehensive recommendations and clinical practice guidelines on the topic are still lacking, mainly due to the scarcity of LTs in children, acknowledged by the European Reference Network for Pediatric Transplantation, and to the absence of high-quality evidence. Thus, we conducted a systematic review on de novo FA in pediatric LT recipients to define epidemiology and risk factors, assess the correlation to specific food groups, describe clinical manifestations, investigate the rate of tolerance acquisition over time and report the available therapeutic strategies.

We performed a systematic review of the literature in order to provide clinicians with evidence-based information regarding six topics relevant to the management of de novo FA after pediatric LT, namely epidemiology, risk factors, culprit food allergens, clinical manifestations, prognosis, and treatment.

As concerns epidemiology, the prevalence of de novo FA in pediatric liver transplant recipients obtained from our analysis of the literature was 15%. This result is not only significantly higher in comparison to the prevalence of FA documented in the general pediatric population (about 1.5–3.5% of objectively confirmed FA) (56), but also far superior than that reported in the recipients of other kinds of solid organs (e.g., about 5% and <1% in heart and kidney transplant recipients, respectively) (5, 10). Little is known about the pathogenetic mechanisms leading to FA in transplant recipients. Several factors may explain the increased susceptibility toward FA observed in pediatric LT patients (Figure 2). Firstly, the liver has a central role in the acquisition of immune tolerance to dietary allergens, as demonstrated by the fact that a severe hepatic dysfunction and the diversion of portal blood into the systemic circulation have been associated with an increased sensitization to food allergens (23). Secondly, intestinal surgery, such as creation of a Roux-en-Y loop at the time of LT, in association with the use of postoperative antibiotics, may alter intestinal permeability and modify the gut microbiota, thereby increasing the risk of post-operative FA (10).

Our data also confirm that the likelihood of developing FA is inversely proportional to the age at LT, with the highest risk in the first 2 years of life. This is consistent with the finding that, after LT, atopic disorders in general and FA in particular develop more frequently in children than in adults (28). Several authors have speculated that the immaturity of the immune system and of the gastrointestinal tract may be responsible for the increased likelihood of sensitization to food allergens in younger subjects (6). In addition, the restrictive diet (predominantly based on artificial formula or parenteral nutrition) that often precedes LT in infants with end stage liver disease may favor the onset of FA by limiting the exposure to dietary allergens at a critical age window for achieving immune tolerance toward foods (6, 31, 36).

Other than age at LT, other elements have been investigated as potential risk factors for the onset of de novo FA. A personal or familiar history of allergic disorders of the recipient and a positive allergic status of the donor were present in only one-third and one-fourth of subjects, respectively. This, coupled with the fact that the passive transfer of FA from the donor to the recipient has been seldom documented in children, supports that, in pediatric LT recipients, the emergence of food sensitization occurs most often regardless of donor and recipient allergic status (23).

The pathogenetic role of calcineurin inhibitors in the onset of post-transplant FA has been broadly investigated. Due to the extensive use of tacrolimus compared to cyclosporine and the small number of patients enrolled in individual studies, few authors were able to discriminate whether the onset of FA was associated with the use of the pharmacologic class of calcineurin inhibitors or with tacrolimus per se, favoring the latter hypothesis. The data obtained from our systematic review demonstrate that tacrolimus is significantly more associated than cyclosporine with the onset of de novo FA after pediatric LT. Although calcineurin inhibitors share the same mechanism of action, tacrolimus is 10–100 times more active than cyclosporine in suppressing T- and B-cell responses and in reducing IL-2 production, and this may favor a stronger immunological imbalance toward Th2 responses (27, 54). Furthermore, tacrolimus, but not cyclosporine, increases intestinal permeability and is thus responsible for an amplified immunological exposure to food allergens (16).

Culprit foods of de novo FA do not differ significantly from those usually implicated in otherwise healthy children with FA (1). Similarly, a significant proportion of patients had allergies to multiple foods (57).

More than 15% of children with de novo IgE-mediated FA suffered from anaphylaxis. The prevalence of anaphylaxis in our whole study population was 1.4% (23/1,647 patients considering available data), slightly higher than that of food-induced anaphylaxis in the general pediatric population, ranging from 0.3 to 1.2% (58).

No study has identified markers able to predict the onset of de novo food allergy in LT recipients. Skin prick tests and serum food-specific IgE (sIgE) indicate the sensitization toward food allergens, but do not necessarily correspond to the presence of clinical FA (1). Moreover, no strategies able to prevent the development of FA after LT have been identified to date and the avoidance of exposure could further increase the risk of sensitization to food allergens (1).

Once arisen, FA tend to persist overtime, with only a minority of patients being able to resume an unrestricted diet. To date, the only effective treatment of de novo FA consists, as for allergies arising in non-transplanted subjects, in the avoidance of allergen ingestion and in the prompt pharmacologic treatment in case of allergic reaction (1). As demonstrated by case reports and small case series, the discontinuation of tacrolimus in favor of other immunosuppressive drugs, such as cyclosporine and mycophenolate, or the therapeutic association of tacrolimus with mycophenolate might reduce the immunological imbalance toward Th2 responses, thus leading to the resolution or the improvement of FA. However, before this strategy can be routinely recommended, prospective, controlled studies are needed to evaluate the efficacy and the safety of this therapeutic approach, which, at present, should be reserved for patients with life-threatening allergic reactions or with multiple FA severely limiting their quality of life.

The main limitation of our study relies in the low level of evidence of the vast majority of available articles. Indeed, the retrospective nature of most of the studies, together with their small number of patients, account for the marked heterogeneity of the results and limit the possibility of drawing definitive conclusions. Particularly, very few studies have specifically investigated the characteristics of the donors, possibly hampering our ability to identify donor or graft-related risk factors. However, our work has several strengths. To the best of our knowledge, this is the only systematic review in literature on the subject of de novo FA after pediatric LT. Moreover, by combining data from all relevant studies, we obtained the largest available population of transplanted food-allergic children, so that we could answer to clinically relevant questions with the best possible evidence.

In conclusion, de novo FA develops in 15% of children after LT, especially in the first 2 years after transplant, with higher risk related to younger age at transplant (especially under 2 years of age) and tacrolimus immunosuppression. FAs cause anaphylaxis in 15% of subjects with IgE-mediated reactions and are often multiple. The majority of patients do not spontaneously outgrow their symptoms, and, to date, no treatment has been clearly proven to resolve the immunological imbalance responsible for the onset of de novo FA after LT. Prospective multicenter studies are needed to confirm these findings, guide the risk-based stratification of pediatric LT recipients, and provide therapeutic strategies for children with de novo FA.

The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding author/s.

CB conceptualized and designed the study, provided support with the search strategy, screened databases for eligible studies, and reviewed the manuscript. EA, SG, MCu, and FC undertook data collection and analysis. LB and MCa drafted the manuscript. GF and PG critically reviewed the manuscript. All authors read and approved the final manuscript as submitted.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.