Hematopoietic Stem Cell Transplantation (HSCT) is a medical procedure involving the collection and re-infusion of multipotent hematopoietic stems cells, aimed at the reconstitution of a normal hematopoietic function (Edman et al., 2001; DeMarinis et al., 2009). Stem cells can be derived from the patient themselves (autologous HSCT) or from a donor (allogenic HSCT). These procedures differ both in indication and in the patient population involved. Autologous HSCT is generally performed to rescue from a dose limiting myelotoxicity patient undergoing chemotherapy for solid tumors or multiple myeloma. Allogenic HSCT is aimed at replacing partially or completely the hematopoietic stem cell compartment with cells derived from a donor, in order to cure hematologic malignancies, non-malignant hematologic disorders, primary immunodeficiency diseases, and inborn errors of metabolism. The donor from allogenic HSCT may be a family member or an unrelated volunteer; stem cells may be derived from bone marrow, umbilical cord blood, or peripheral blood after treatment with a mobilizing agent (Park et al., 2015).

More than 50,000 HSCTs are performed every year worldwide and around 5,000 of them every year in Italy (Gratwohl et al., 2010). The process of care for a patient undergoing HSCT involves different phases: preparatory evaluation, high-doses chemotherapy, stem cell infusion, aplasia, marrow recovery, and a life-long follow-up. Although outcomes have improved over the years, patients are at risk of early and late side effects that range from mild to life-threatening events (Sanders, 2002; Tabbara et al., 2002; Andersson et al., 2008). Early complications include regimen-related toxicity, bleeding, and infections (Whedon and Wujcik, 1997). In the setting of allogenic HSCT, Graft-vs.-Host disease (GvHD), an immune reaction of the donor lymphocytes against the recipient tissue, may lead to significant morbidity and mortality. Late effects of HSCT may impact on the function of different organs resulting in chronic cardiovascular, respiratory, metabolic, endocrine, and reproductive disorders (Sanders, 2002; Tabbara et al., 2002).

About 17% of HSCTs are performed in children under 20 years of age (Center for International Blood and Marrow Transplant Research, 2005; Hollingsworth et al., 2008) and particular considerations should be applied to HSCT at this age (Hollingsworth et al., 2008; Chang et al., 2012). Despite its increasing success rates, pediatric HSCT still carries a significant risk of emotional, cognitive, social, and family functioning (Clarke et al., 2008). The aversive experiences and life disruptions experienced by pediatric patients undergoing transplantation can interfere with adjustment and consequently with social and emotional maturation (Vannatta et al., 1998). Survivors have been described as being more withdrawn and less socially competent with respect to their peers (Phipps and Mulhern, 1995; Vannatta et al., 1998). Moreover, given the longer life expectancy of childhood HSCT survivors as compared with adults, different medical complications, such as GvHD, endocrine dysfunctions and pulmonary complications might arise (Buchsel et al., 1996). Although empirical data about the psychological consequences of HSCT during pediatric age are still limited, there is some evidence that children and adolescents that undergo HSCT, as well as their families, are exposed to intense emotional distress and to multiple traumatic situations (Packman et al., 2010). Experiencing disease during childhood constitutes a risk factor for abnormal personality development: given the close relationship between body and mind, especially during childhood and adolescence, each physical intervention could cause trauma in psychological functioning. Moreover, receiving a cancer diagnosis in adolescence may impact on processes such as physical and sexual maturation, development of identity, and the possibility to experience independence, decision making and social growth (Geenen et al., 2007; Kieran et al., 2010; Kahalley et al., 2013). Even though HSCT offers higher life expectancies, it is associated with intense emotional distress for patients and their families, might cause trauma and could affect the possibility to re-establish normal life routines (Panizon, 2005). More than a third of childhood cancer survivors show physical, neurocognitive, or psychological sequelae, with subsequent long-term psychopathological outcomes (Institute of Medicine (US) and National Research Council (US) National Cancer Policy Board, 2003).

HSCT survivors are at higher risk of developing mood disorders and anxiety disorders when compared to the general population and to individuals with chronic disorders (Felder-Puig et al., 2006; Packman et al., 2010).

Several studies reported that a consistent amount of patients frequently experiences depression, peer isolation and behavior problems during the 6 months post-HSCT (Pot-Mees, 1989; Ahomäki et al., 2015). Worse depressive symptoms resulted associated with extended hospital stays (Patenaude, 1990) whereas more severe withdrawal and a higher loss of adaptive skills were associated with younger age at transplantation (Pot-Mees and Zeitlin, 1987).

As far as it concerns anxiety, the ongoing of symptoms seems to assume a different direction: a high rate of pediatric patients experiences intense anxiety before HSCT, whereas symptoms are likely to decrease after transplantation and to remain low during follow-up visits (Meyers et al., 1994). Moreover, some studies reported lower degrees of anxiety in younger children, whereas older children and adolescents reported higher anxiety (Nuss and Wilson, 2007).

Besides mood disorders and anxiety disorders, researches have also detected the presence of social anxiety, post-traumatic stress disorders, low levels of self-esteem, and poor quality of life among this population (Stuber et al., 1990; Bessel, 2001; Kazak et al., 2001; Stam et al., 2006). In line with these results, some authors reported that a high percentage of patients show moderate post-traumatic stress symptoms immediately after HSCT (Stuber et al., 1990) and suggested that some of them experience these symptoms even 12 months after HSCT (Pot-Mees, 1989).

Furthermore, although some experiences report good resilience in cancer survivors, others suggested that at least 10–30% of these children is likely to exhibit difficulties in the adjustment subsequent to treatments.

The main risk factors that were associated with these outcomes were: type of disease, higher age at communication of diagnosis, and the adverse effects of the therapies (Levi, 2006; McDougall and Tsonis, 2009). A recent Finnish study compared a cohort of nearly 14,000 cancer survivors with a sample of healthy siblings, finding a higher incidence of psychiatric disorders such as: mood disorders, anxiety disorders, psychotic disorders, somatic disorders, feeding disorders, and personality disorders (Ahomäki et al., 2015). The same study identified radiotherapy, female gender and younger age at diagnosis as principal risk factors. Moreover, other studies reported the type of transplantation to be another clinical variable with an impact on psychological well-being of children and adolescents exposed to HSCT (Phipps et al., 2002).

Although different studies evidenced different HSCT-related psychological stressors, the existing literature still appears scarce. Moreover, it is plausible that, in the context of health care, treatments may not focus specifically on the psychological aspects involved in transplantation but rather consider physical issues and medical complaints (Feeny et al., 1999; Tanzi, 2011). A deeper understanding of the psychological impact of HSCT could help to implement treatment with subsequent improvements in the general Quality of Life (Tanzi, 2011). Moreover, screening ex-patients' emotional and behavioral functioning, and understanding the potential risk factors associated with long-term psychopathological outcomes, could help to inform interventions during recovery, improving the survivors' experience and preventing unnecessary distress.

Data were analyzed using IBM SPSS statistics version 23 [SPSS(R) 23.0] (IBM Corporation, Armonk NY, USA). Descriptive analyses (frequencies, mean scores, standard deviations) were performed on the CBCL and the YSR scores in order to assess the presence of psychopathology. Subsequently, Cohen's kappa coefficient was applied in order to measure inter-rater agreement between the parents and the survivors' reports on psychopathology; for this purpose, contingency tables with the CBCL and the YSR were made. Finally, ANCOVAs were carried out in order to assess the effect of the single independent variables in predicting the presence of psychopathology.

The aim of the present study was to investigate the psychological aspects associated with HSCT in pediatric age. Although HSCT increases the possibility to survive and to recover from disease it affects all aspects of the child's and the adolescent's life, including social, cognitive, emotional and behavioral functioning. More specifically, the objectives of the study were to: (a) assess the presence of psychopathology in HSCT survivors at distance from transplantation, (b) compare the psychological aspects of pediatric HSCT survivors with those of healthy peers, (c) detect whether parents and children differed in the perception of mental health status, and (d) identify medical and socio-demographic factors associated with psychopathology. For these purposes, the CBCL and the YSR were given to a group of HSCT survivors and their families and to a comparison group matched according to age, gender and geographic origin.

With respect to the first objective (a), both the parents and the survivors reported scores above normative cut offs for the Total competence scale, highlighting the presence of difficulties in the domains concerning activities and social competencies. These results seem to be in line with the hypothesis made by some authors that the experiences associated with transplantation could somehow interfere with adjustment and social competence (Phipps and Mulhern, 1995; Vannatta et al., 1998). Regarding more specifically the presence of psychopathology, on the other hand, higher scores were reported for somatic symptoms, suggesting both the presence of actual long-term somatic sequelae after treatment or a possible decrease in pain tolerance, as already highlighted by the literature (Pederson et al., 2000).

It is interesting to note that, in contrast to what other studies reported (Kazak et al., 2001, 2004), the group of subject involved in our study did not report significant scores in the Post-traumatic stress problems scale. It is possible to speculate as to whether HSCT survivors could experience anticipatory fear linked to their previous traumatic experience and express it through worries about physical well-being, a hypothesis that could in part explain the parallel presence of high scores on somatic symptoms and low scores on post-traumatic stress symptoms (Bingen et al., 2012; Ahomäki et al., 2015).

With reference to the second objective (b), the comparison between the HSCT Group and the HEALTH Group highlighted differences only regarding somatic symptoms. No differences were found with respect to the other variables. This is not surprising, given the high probability that the majority of survivors has to experience physical late effects after transplantation (Skinner and Leiper, 2004). Moreover, this data should also be interpreted considering the fundamental role played by the body both during childhood and adolescence. As previously highlighted childhood represents a developmental period characterized by a close relationship between body and mind, and where emotional difficulties are likely to be expressed through somatic complaints. On the other hand, during adolescence bodily issues are of particular importance especially given the processes of physical and sexual growth, which are typical of this period.

With respect to the third objective (c), the comparison between the CBCL and the YRS reported low inter-rater agreement only in a few scales (i.e., Total competence, somatic problems and attention problems), suggesting a different perception of psychopathology between the survivors and their parents. More specifically, the results showed higher scores on the CBCL, suggesting the perception of higher behavioral/emotional difficulties on the parents' side. The low agreement between child and parent reports has already been described by other studies (Perrin et al., 1991; Briggs-Gowan et al., 1996). This data could be read both as a higher denial of psychopathology in the survivors but also as their possibility to reach a more functional adjustment, an achievement that could be somehow favored by their previous experience. Moreover, some authors have suggested that the parents' emotional state might influence their perception regarding the children/adolescent's emotional state in a way that, for example, a higher degree of anxiety experienced by caregivers might lead to a higher parental perception of anxiety in the adolescent as well (Parsons et al., 2006, 2009; Jobe-Shields et al., 2009; Chang et al., 2012). In this sense, the collection of multiple perspectives would be preferable during the assessment of HSCT survivors' well-being, in order to better identify subjects that are potentially at higher risk of psychological problems and to promptly offer supportive interventions (Chang et al., 2012).

Finally, with respect to the fourth objective (d), the results evidenced an association between specific medical and socio-demographic variables and the risk of experiencing higher psychological distress. More specifically, with respect to the Total problem scale and to the summary scales, the variables female gender, related allogeneic HSCT, acute and long-term sequelae, psychophysical problems, and time elapsed between the communication of the need to transplant and the effective transplantation were shown to be predictors of a higher risk of experiencing psychopathology. These results are in line with the literature that identifies females at a higher risk of developing psychopathological problems subsequent to oncologic disease and transplantation (Ahomäki et al., 2015). Furthermore, allogeneic related transplantation was shown to be a risk factor for experiencing a higher degree of emotional and behavioral difficulties, in accordance with other studies that already identified children undergoing autologous transplant as less likely to experience significant somatic distress and impairments when compared with those exposed to allogeneic intervention, both related or unrelated (Phipps et al., 2002). On the other hand, it is possible to speculate that although the presence of a related donor could represent a precious resource in terms of survival, it may increase the possibility of experiencing intense psychological distress in the patient (Weaver et al., 2015). Moreover, the physical suffering associated with medical toxicities and waiting, before the actual transplantation after the communication of the need to transplant, that could be either too long or too short, could constitute additional psychological stressors.

It is interesting to observe that the presence of acute consequences predicted higher externalizing symptoms, which could thus be interpreted as a possible reaction to the “internal,” physical and psychological sufferings linked to disease and transplantation.

On the other hand, regarding the syndrome scales, again female survivors were identified at higher risk of experiencing anxiety. Anxiety symptoms were also predicted by a higher amount of time passed in the Transplantation Unit, by the presence of acute consequences and psychophysical problems, and by a shorter time elapsed between the communication of the need to transplant and the actual transplantation. On the contrary, a longer amount of time passed before the intervention predicted depressive symptoms, suggesting the need to schedule the “right” timing between the different phases that follow diagnosis, in order to favor a more functional regulation of the emotional aspects involved in this delicate process.

Moreover, somatic complaints were predicted by a younger age at the moment of diagnosis. This result, that seems somehow to contrast with other studies that suggest older age to be a significant variable associated with poorer psychological functioning (Parsons et al., 2006), could be interpreted with specific reference to the developmental phase that each patient is experiencing; while children could be more likely to complain about somatic disease and to ask their caregivers for help, it is plausible that in adolescence, a period specifically characterized by the search for autonomy, the individuals are more inclined to deal with the difficulties by themselves. Social problems were predicted by a longer time elapsed between the communication of the need to transplant and the actual transplantation, by the presence of longer hospitalizations, and by the presence of intensification with double HSCT, suggesting that the concrete limitations in activities subsequent to medical procedures that patients experience could significantly impact on their relationships with peers. Finally, the variables considered also predicted attention symptoms, in line with the studies that pointed out the possible impact that hospitalizations and medical procedures may have on the cognitive functioning of psychiatric patients (Bingen et al., 2012).

One limitation of the present study concerns the small number of participants, which prevents the generalization of the data obtained; in any case, these results seem to be in line with the existing literature in confirming the presence of an increased risk of experiencing psychological distress and psychopathology in ex-patients exposed to HSCT during childhood. This risk seems to be enhanced by the presence of specific demographic and medical variables, which should thus be taken into account during the different phases that precede, accompany, and follow HSCT. Taken together, these results could help to inform intervention and prevention programs, since attention to the aforementioned demographic, medical and psychological variables could help identify which subjects could still require services and support once the process of reintegration to normal routines following HSCT has begun.

SZ: wrote the central part of the manuscript, partecipated in collecting data. AT: wrote the discussions section of the manuscript; partecipated in critical interpretation of data. MTr: prepared data set, performed statistical analyses. FS: prepared data set, performed statistical analyses. AP: wrote the introduction section of the manuscript and participated to the organization of collected data. GP: performed analyses and supervised statistical analyses. ND: prepared tables and figures; recruited the sample and wrote the references. FC: recruited the sample and followed patients in every steps of the project. AM: recruited the sample and collected medical data; carried out follow up visits. MTu: recruited the sample and collected medical data. CC: prepared the study design and supervised the mental health research team. GB: prepared the study design and supervised the research team. CM: prepared the study design and supervised the research team. All authors reviewed the manuscript.