Growth Hormone and Insulin-like Growth Factor-1 in Children with Cholestatic Diseases and Pediatric Liver Transplantation

Abstract

The growth hormone/insulin-like growth factor 1 (GH/IGF-1) system is the most important regulator of growth, regeneration, and metabolism in children and adults. Children with congenital cholestatic diseases have elevated GH blood levels, which is combined with growth failure and body mass deficit. Congenital cholestatic diseases lead to end-stage liver disease (ESLD), where GH bioavailability, mediated through IGF-1, is impaired. Blood IGF-1 levels are decreased due to impaired production by the liver. This study included 148 children up to 5 years (60 months) old with congenital cholestatic diseases. The patients underwent liver transplantation (LT) at a leading transplant center in Russia. The clinical significance of the GH/IGF-1 axis in pediatric liver recipients was investigated. Relationship between the patients’ GH/IGF-1 levels and anthropometric parameters was analyzed before and after LT. It was shown that LT leads to renewal/recovery of GH-IGF-1 regulation and improved anthropometric parameters (body height and body mass) in pediatric recipients.

1. Introduction

Liver transplantation (LT) is the only effective treatment for patients with end-stage liver disease (ESLD). The first successful pediatric LT was performed by Thomas Starzl in 1967 on a child with biliary atresia (BA). This marked the beginning of the use of LT in pediatric practice [1, 2]. By 2020, the total number of LTs performed was about 32,000 per year, 7–8% of which were pediatric transplant surgeries [3].

In the Russian Federation, the first pediatric LT was performed in 1997. Today, the total number of transplantations performed has exceeded 1000. The need for this type of medical care is fully satisfied in Russia. The LT program is being implemented at Shumakov National Medical Research Center of Transplantology and Artificial Organs (Shumakov Center), Russia’s leading transplant center; over 100 LTs are performed annually. Surgeries are performed in children with extremely low body mass, starting from the first months of life.

Invasive diagnostic procedures, such as needle biopsy, carry high risks for young children with low body mass. Therefore, it is an urgent task to search for markers for the development of noninvasive methods of assessing liver graft function. Given physical developmental delays in children, which are associated with hepatobiliary diseases and subsequent hormonal dysregulation, GH and IGF-1, the key elements in neurohumoral control of liver function can serve as important indicators reflecting the outcome of LT. The significance of GH and IGF-1 in pediatric LT is related to the role of hormones in the regulation of growth and body mass, their influence on hepatocyte function and immune system activity. A number of studies have shown the relationship between the GH/IGF-1 axis and recipient/graft survival [4, 5, 6].

Despite the large number of studies on the role of the GH/IGF-1 axis in children and adults, none of them are devoted to their relationship with physical development in children after LT [7, 8, 9]. The study of the role of the GH/IGF-1 axis in the neurohumoral regulation of liver graft function in young children is an important task. It may allow to assess the functional ability of the graft by noninvasive methods and will also create a vector for coming up with personalized therapy. This chapter describes the impact of the GH/IGF-1 axis on the outcome of pediatric LT (namely the clinical significance of the GH/IGF-1 axis in pediatric liver recipients) and characterizes its relationship with patients’ anthropometric parameters.

2. Growth hormone and insulin-like growth factor 1

GH and IGF-1 are key links in the neurohumoral regulation of overall metabolism and can act through various intracellular signaling pathways. GH stimulates IGF-1 synthesis, which, in turn, affects GH production by the principle of negative feedback, inhibiting its synthesis (Figure 1).

Figure 1.

Scheme of neurohumoral regulation of growth by the GH/IGF-1 system. GHRF – growth hormone releasing factor, GH – growth hormone, GHR - growth hormone receptor, IGF-1 – insulin-like growth factor 1, IGFBP – IGF binding protein, IGFR – IGF receptor.

The physiological effects of GH and IGF-1 on cells are mediated through transmembrane receptors found on the surface of many cell types, including hepatocytes and lymphocytes [10]. The effect of GH and IGF-1 is largely determined by the level of receptor expression, which depends on cell type and can change under the influence of various factors. The action of the GH/IGF-1 axis depends on GH and IGF-1 production on one hand, and IGF-binding proteins, proteases that degrade the IGF-binding proteins complex, and GH and IGF-1 receptors, on the other hand [11].

GH is synthesized in the anterior pituitary cells and secreted into the blood, with peak blood levels every 3–5 hours. The highest blood content of the hormone occurs during fetal development. With age, the baseline, frequency, and amplitude of hormone secretion peaks decrease. The nature of GH secretion differs in men and women and depends on age. The range of baseline GH levels in children aged 1–3 years is 2–10 ng/m, in adults, 1–5 ng/ml [12, 13].

IGF-1 is a polypeptide hormone produced by many tissues, but more than 90% of circulating IGF-1 is synthesized by hepatocytes [3]. Serum IGF-1 levels, in contrast to GH, practically do not change during a day. The range of reference IGF-1 levels in children aged 1–3 years is 5–300 ng/ml; the maximum IGF-1 level in children is observed during puberty and gradually decreases with the years [14, 15].

Growth regulation is one of the main functions that GH and IGF-1 have in common. In addition, GH in both children and adults plays an important role in the regulation of metabolism. IGF-1, being a major mediator of anabolic and mitogenic effects of GH in peripheral tissues, is an important factor in body mass regulation. However, GH and IGF-1 have different effects on glucose and lipid metabolism: GH increases blood glucose levels and promotes lipolysis, whereas IGF-1 has opposite effects [16, 17, 18].

3. Effect of GH/IGF-1 on liver function

The liver is the largest internal organ in humans; in a child it makes up 4.4% of body mass, which is 1.5 times more than in an adult—2.8%. The importance and diversity of liver functions in young children should be noted: it is not only bile production, general metabolism, detoxification, but also, importantly for pediatric liver recipients, the liver works as an immune organ, maintaining overall immune homeostasis of the body (Figure 2).

Figure 2.

Liver functions in children.

Humoral regulation of liver function is provided by the peculiarity of the hepatic portal system and the diversity of the cellular composition, which are targets for so many hormones and cytokines (Figure 3).

Figure 3.

Peculiarities of liver structure and cells composition. The hepatic blood supply system consists of the portal vein, the hepatic artery, and the sinusoidal capillary system of the liver, which flows into the central vein. Liver cells with antigen-presenting function: dendritic, Kupffer, stellate, and endothelial. GH – growth hormone, GHR – growth hormone receptor, IGF-1 – insulin-like growth factor 1, IGFR – IGF receptor, TGF-beta1 – transforming growth factor beta 1, sFas/sFasL – soluble Fas/soluble Fas ligand.

The liver forms a unique microenvironment due to its anatomy, in particular the circulatory system consisting of the portal vein, hepatic artery, and liver sinusoid. Portal venous blood is rich in dietary antigens and toxins, to which tolerance is developed by a large set of liver cells with antigen-presenting function: dendritic, Kupffer, stellate, and endothelial cells [19, 20, 21, 22, 23]. Liver sinusoidal endothelial cells, which, unlike the vascular endothelium, do not have a basement membrane and tight junctions, are capable of blocking proinflammatory CD8+ T cells [24]. The liver is characterized by the generation of autoantigen-specific regulatory T cells, which, through active immunosuppression of type 1 and 2 T helper cells (Th1/2), form a mechanism of autoimmune tolerance [25, 26, 27]. It is also assumed that natural killers play a significant role in the immunocompetent properties of the liver; their number in the human liver is up to 10% of all lymphocytes [28]. Liver stellate cells are also capable of suppressing immune response, causing apoptosis of activated T cells through expression of costimulatory molecules (CD28/B7) B7-H1 and the production of cytokines—interleukins (IL) 6 and 10, transforming growth factor beta (TGF-β) [29]. Stellate cells induce tolerance through apoptosis of activated T cells via the Fas/Fas ligand system, IL-10 and TGF-β [20].

The GH/IGF-1 axis is the most important regulatory system of the body that controls cell growth and is closely related to liver function [4, 5, 30]. Previously, it was believed that IGF-1 does not directly affect the function of hepatocytes, because in a healthy liver, a small number of IGF-1 receptors are expressed on the surface of hepatocytes. However, further studies have shown that their expression is increased in some liver diseases [30]. In acute viral hepatitis, chronic hepatitis C and B, the expression of IGF-1 receptors on hepatocytes is higher than in a healthy liver. Meanwhile, there are elevated IGF-1 levels, which is believed to accelerate the regeneration of damaged hepatocytes [31].

The antifibrotic effect of IGF-1 is realized both directly through the GH/IGF-1 axis and indirectly through regulation of other profibrogenic factors [32, 33]. Stellate cells play a key role in the development of liver fibrosis: their activation caused by chronic trauma, oxidative stress, and increased levels of inflammatory cytokines and lipopolysaccharides leads to transformation into fibroblasts [34]. It has been shown that IGF-1 is able to inactivate hepatic stellate cells and induce their aging, thus limiting the development of fibrosis [35]. GH and IGF-1 regulate lipid metabolism and influence the development of steatosis, inflammation, and liver fibrosis in a certain way [36]. Thus, reduced IGF-1 production in the liver is not only the result of impaired liver function, but also plays an important role in the development of fibrosis.

4. GH/IGF-1 in liver diseases

IGF-1 synthesis by hepatocytes is impaired in liver diseases. This leads to reduced IGF-1 levels and, according to the principle of negative feedback, leads to increased GH secretion. Despite the high serum GH levels in patients with chronic liver diseases, low IGF-1 levels can also be the result of hepatocyte resistance to GH [37, 38]. It is assumed that metabolic disorders—insulin resistance, malnutrition, osteopenia, etc.—often found in patients with liver diseases are associated with impaired neurohumoral regulation and are caused by IGF-1 deficiency [39, 40, 41].

The degree of decrease in IGF-1 levels in patients with chronic liver diseases correlates with the severity of hepatocyte dysfunction [36, 42]. Introduction of recombinant IGF-1 stops liver fibrous degeneration [43, 44]. In an animal experiment with nonalcoholic steatohepatitis (NASH), recombinant IGF-1 was shown to improve liver function in cirrhosis. Also in the experiment, it was found that the degree of hepatic ischemia-reperfusion injury is less at higher IGF-1 levels [45, 46]. Experimental studies suggest the possibility of using IGF-1 in clinical practice. Administration of recombinant IGF-1 in patients with liver cirrhosis has been shown to increase serum albumin levels and improve energy metabolism [47].

In adult patients with liver cirrhosis, GH levels are significantly higher than in healthy ones, which is associated with impaired IGF-1 synthesis by the liver in ESLD [17]. Already on day 7 following LT in adults, GH levels decrease to normal values, and those of IGF-1 increases [39].

In recent decades, there has been much focus on the study of the GH/IGF-1 axis as factors associated with longevity and quality of life [48]. It is known that life expectancy is closely related to the GH/IGF-1 axis, which may be of some importance for the development of techniques for predicting recipient and graft survival [6, 49]. It has been shown that IGF-1 can improve survival rates in rats with lipopolysaccharide/D-galactosamine-induced acute liver failure. Prophylactic administration of IGF-1 in the animals prevented an increase in bilirubin levels and transaminase activity [44].

5. GH/IGF-1 in pediatric LT

In pediatric LT, GH and IGF-1 play a significant role in the regulation of growth and body mass, hepatocyte function, and immune system activity [4, 5, 6].

Indications for LT in children include ESLD with a life expectancy of less than 1 year, acute liver failure, unresectable liver tumors, and metabolic disease. In 80% of cases, ESLD is caused by congenital anomalies of the biliary tract, such as biliary atresia (BA) and biliary hypoplasia (BH), Alagille syndrome and disease, etc. The incidence of such diseases is about 1 in 10,000–20,000 newborns, atresia is slightly more common in newborn girls than boys. Biliary disorders lead to cholestasis, fibrosis, and cirrhosis, which in turn result in esophageal varices, splenomegaly, hemorrhagic syndrome, esophageal or gastrointestinal bleeding, etc. BA with the inevitable formation of fibrosis and cirrhosis leads to decreased detoxification of metabolites, chronic malnutrition, coagulopathy, growth factor deficiencies, and an excess of proinflammatory cytokines, which together can contribute to delayed physical and psychomotor development. In the absence of early surgical intervention, BA children usually do not live up to one year, dying from liver failure, bleeding, associated pneumonia, cardiovascular disease, and intercurrent infections. With incomplete biliary atresia, some children may live up to 10 years. Therefore, congenital anomalies affecting the biliary tract are a strong indication for LT. LTs performed at Shumakov Center fully cover the needs for this type of care in Russia [50].

The shortage of donor organs for children is compensated by living related donation of liver fragments. As a rule, young children are transplanted with the left lateral lobe of the liver of a living or deceased donor (Figure 4).

Figure 4.

Transplantation of the left lateral liver section in children from an adult living donor.

Transplantation of a liver fragment from a living related donor to children makes it possible to fully satisfy the need for this type of therapy. Improved pre-transplantation preparation, patient selection, organ preservation techniques, better immunosuppressive therapy, and post-transplant follow-up techniques have led to outstanding results in this area.

Unlike other solid organs, the liver is an immune-privileged organ, which is due to the peculiarities of development, blood supply, structure, and function of the liver [22, 51, 52]. After LT, incidence of graft rejection is significantly lower than after transplantation of other solid organs, such as the kidney or heart. The best recipient and graft survival rates are observed in pediatric LT from a living related donor. This is due to the peculiarities of immune response development at an early age, the possibility of ensuring greater tissue compatibility, and a shorter graft ischemia time. Pediatric LT from a living related donor is a model of the relationship between the graft and the recipient’s body in conditions of minimal risk of rejection and low immunosuppression.

LT has become a standard therapy for children with ESLD, acute or chronic, as well as liver neoplasms and metabolic disorders. According to world statistics, the 5- and 10-year graft survival in children is 85–90% and 77–80%, respectively [53]. The results we achieved at Shumakov Center are fully consistent with global ones. For example, the average 1-year survival rate was 93%, and the 3-year survival rate was up to 90% [50, 54].

5.1 Patients

The present study included 148 children ≤5 years (60 months) old, selected by random sampling, who underwent liver fragment transplantation at Shumakov Center. The study was approved by the Local Ethics Committee of Shumakov Center (Minutes No.28/1d dated April 15, 2016). All the patients or their guardians gave written informed consent to participate in the study.

The mean age of the recipients was 11 ± 9.8 months (median, 8 months). LT was indicated due to ESLD resulting from various pathologies: biliary atresia (BA, n = 86), Byler disease (n = 15), biliary hypoplasia (BH, n = 14), Alagille syndrome (n = 12), Caroli disease and syndrome (n = 5), and others (n = 16). In the “other” group, we included patients with diseases, which did not exceed three cases, such as autoimmune and fulminant hepatitis, Von Gierke disease, portal vein developmental anomaly, Crigler-Najjar syndrome, Budd-Chiari syndrome, hepatoblastoma, alpha-1 antitrypsin deficiency, and previous liver transplant dysfunction (Table 1).

Characteristics	Value	Healthy children	p value
Number of patients, n	148	16	-
Age, median, (range), month	8 (2-60)	12 (3-25)	0,78
Gender (M/F) (n; %)	61/87; 41/59	9/7; 56/44	0,84
Diagnosis (n; %)			-
BA	86; 58
Byler’s disease	15; 10
BH	14; 10
Allagille syndrome	12; 8
Caroli disease	5; 3
Other	16; 11

Table 1.

Baseline patients’ characteristics, BA, biliary atresia; BH, biliary hypoplasia.

The comparison group consisted of 16 children, 9 boys and 7 girls, practically healthy, examined after treatment of intestinal dysbacteriosis, without other diseases, aged 3–25 (median, 12) months, anthropometric parameters (body height, body mass) were in the range of average population values (25th–75th percentile). The age and sex of the children included in the study and those of healthy children did not differ (p = 0.78 and p = 0.84, respectively).

All recipients underwent hepatectomy with preservation of the inferior vena cava, followed by implantation of a liver fragment (left lateral sector) in an orthotopic position. Blood flow characteristics were monitored intraoperatively using ultrasound. Bile duct reconstruction was performed after graft revascularization on the defunctionalized jejunal loop.

5.2 Methods

Patients were examined before and after LT in accordance with the protocol approved at Shumakov Center. Routine examination and treatment of patients before and after LT were carried out in accordance with the clinical guidelines of the Russian Transplant Society and international consensus guidelines.

The required methods were:

1. Physical examination, including observation, palpation, percussion, and auscultation;

2. Laboratory examination: biochemical blood test, coagulogram, general blood test, hemodynamic parameters.

3. Instrumental methods: ultrasound, echocardiography and the study of the electrical activity of the heart; multislice CT scan of the brain, chest, and abdominal cavity with intravenous contrast.

Double- and triple-combination immunosuppressive therapies were used at Shumakov Center.

During preparation for donor organ implantation, all recipients received 10 mg of basiliximab. Reintroduction of basiliximab was performed on day 4 after surgery. Immediately before graft reperfusion, methylprednisolone was administered at a dose of 10 mg/kg.

Maintenance immunosuppressive therapy included tacrolimus (at 6–8 ng/mL target concentration), glucocorticoids, and mycophenolic acid preparations. Cyclosporin A was administered in a few cases with tacrolimus neurotoxicity. Gradual reduction of methylprednisolone doses to 4–2 mg per day was continued.

Mycophenolic acid preparations were administered as the third component of the therapy in the absence of haplotype compatibility with donors and when a rejection reaction occurred.

In the case of rejection, intravenous pulse therapy was performed with high doses of glucocorticoids (20 mg/kg).

The concentration of biomarkers was studied in plasma or serum obtained from peripheral blood after centrifugation at 1500–2500 g at room temperature. Blood was collected in plastic tubes (BD Vacutainer, Becton Dickinson, USA) without anticoagulant or with sodium citrate as an anticoagulant. Serum and plasma samples were stored at −50°C until analysis.

GH levels were measured by enzyme immunoassay using a reagent kit (DBC, Canada) according to the manufacturer’s instructions in blood serum. The optical density of the reaction mixture was measured at 450 nm wavelength using a spectrophotometer. Analytical sensitivity of the method was 0.2 ng/ml. The range of determined concentrations is 0.2–50 ng/ml.

IGF-1 levels were measured by the sandwich immunoassay method using polyclonal sheep antibodies and monoclonal antibodies against IGF-1 in plasma (IDS Ltd. 77 OCTEIA® IGF-1, UK). The analytical sensitivity of the method was 3.1 ng/ml. The range of determined concentrations is 12–627 ng/ml.

5.3 Statistical analysis

Data are presented as mean and standard deviation (M ± SD), upper and lower limits of the 95% confidence interval for parametric variables, and median and interquartile range from 25th to 75th percentile for nonparametric variables. The data were statistically processed using nonparametric statistics methods: paired Wilcoxon test was calculated to compare dependent samples, and the Mann-Whitney or Kruskal-Wallis U test was used to compare independent variables. Spearman rank correlation coefficient was calculated to evaluate the relationship between quantitative and qualitative ordinal signs. In the case of error probability (p < 0.05), the differences were considered statistically significant. The obtained data were processed using Microsoft Office Excel with the IBM SPSS STATISTICS 20 software package for scientific and technical calculations (IBM SPSS Inc., USA).

6. GH/IGF-1 and anthropometric characteristics of children with cholestatic diseases

Children suffering from congenital hepatobiliary diseases have multiple and various homeostasis disorders, largely due to a violation of the key functions of the liver—synthesizing, detoxifying, regulatory, etc.—which ultimately lead to growth retardation and underweight.

In the patients before LT, the average body height was 71.2 ± 8.2 cm, and body mass was 7.9 ± 2.3 kg. Comparison of average height and weight in the study group with reference values for healthy children of the same age (mean age, 11 months) showed that the average height of the recipients was significantly lower than the average reference value for healthy children according to World Health Organization (WHO) [55]—75 ± 6 cm, p = 0.00, and the average body mass of the recipients was significantly lower than reference values - 9.5 ± 2 kg, p = 0.00.

It should be noted that a correct comparison of anthropometric indicators of the study group with healthy patients is a difficult task due to the absence of a group of healthy children of the same ethnic group that is similar in age and sex. Therefore, the authors consider it possible to compare the histogram of the distribution of a trait with a normal distribution curve superimposed on a similar range of an anthropometric indicator (Figure 5).

Figure 5.

Histogram of the distribution of body height (A) and body mass (B) of children with ESLD in comparison with the normal distribution curve.

The figure shows that the histograms of body height and body mass distribution of in children with ESLD differ significantly from the normal distribution curves and are shifted to lower values.

Thus, in the examined group of children with ESLD, anthropometric indicators are lower than the values typical for healthy children of the same age. Obviously, the low body height and body mass of patients are associated with ESLD.

The Pediatric End-stage Liver Disease (PELD) scale was used to assess the relative severity of liver disease. This scale is used to evaluate the probable 90-day survival of patients awaiting LT and was developed to objectively prioritize LT candidates in children. Like the Model End-stage Liver Disease (MELD) score, the PELD score was derived from biological markers of liver function (albumin, bilirubin, and international normalized prothrombin time ratio) and growth retardation [56, 57].

The PELD score, which reflects the severity of liver disease, averaged 19.0 ± 8.2 (median, 18.0; range, 0 to 51) in the study group (Figure 6).

Figure 6.

Histogram of the distribution of PELD score in children with ESLD.

The distribution histogram of the PELD score in the examined children shows that in most patients, the score ranges from 10 to 25 and indicates that they have severe liver failure.

To assess the relationship between the severity of liver failure and the anthropometric parameters of the liver recipients, a correlation analysis was carried out between the PELD scores and the body height or body mass of the patients (Figure 7).

Figure 7.

Correlation of PELD score with body height (A) and body mass (B) of pediatric liver recipients.

The dependencies presented in the graphs indicate that patients’ height and weight, determined before transplantation, significantly correlate with the PELD score. The higher the PELD score in patients before radical treatment of the underlying disease—liver fragment transplantation—the lower the level of anthropometric parameters (body mass and height).

The obtained result suggests that, along with other changes, there may be impairment of hormonal growth regulation in this group of children. The most important growth regulators are GH produced by the pituitary gland and IGF-1 synthesized in the liver. Comparative analysis of GH levels in patients before LT and healthy children of the same age indicates a significant difference in the indicator (Figure 8).

Figure 8.

The level of GH in the blood of healthy children and patients with ESLD, * - p < 0.05.

The median plasma GH level of children with ESLD was 4.3 (1.6–7.2) ng/ml and was significantly higher than in healthy children of the same age—1.2 (0.3–2.4) ng/ml, p = 0.001. Thus, there is a paradoxical situation when high GH levels in children with ESLD are combined with growth retardation and weight loss.

A comparative analysis of IGF-1 levels in patients before LT and healthy children of the same age also showed significant differences (Figure 9).

Figure 9.

The level of IGF-1 in the blood of healthy children and patients with ESLD, * - p < 0.05.

The median IGF-1 level in children with liver failure was 7.9 (0.0–25.1) ng/ml and was significantly lower than in healthy children: 38.0 (26.9–56.5) ng/ml, p = 0.001; this is consistent with the literature data and is associated with the inability of the damaged liver to produce this growth regulator [58, 59]. Significantly low level of IGF-1 in liver recipients, obviously, reflects impaired synthesis and production of IGF-1 in the liver and explains the high level of GH in the blood.

Correlation analysis between GH and IGF-1 levels in children before transplantation revealed no significant relationship between hormone levels (Figure 10).

Figure 10.

Correlation between IGF-1 and GH blood plasma levels in children with ESLD.

The analysis confirms that in children with ESLD, there is a violation of the relationship between GH and IGF-1 concentrations in the blood.

To characterize the hormonal regulation of anthropometric parameters and liver function, we analyzed the correlations between the GH and IGF-1 levels in children with height, body mass, and PELD scores (Table 2).

Biomarker	Correlation coefficient
	Body mass	Body height	PELD
GH	−0.17	−0.14	0.02
IGF-1	0.29*	0.28*	0.15

Table 2.

Correlation of levels of GH (GH) and IGF-1 in the blood plasma of recipients with anthropometric parameters and the value of the PELD index before LT.

*

−p < 0.05.

The analysis revealed no statistically significant associations between GH levels and body height, body mass, or PELD score. Whereas IGF-1 concentrations in the blood were significantly associated with body height and body mass (p = 0.001), but not with PELD score.

The result obtained shows that in children with ESLD, their body height and body mass are not associated with plasma GH levels, while IGF-1 levels correlate with these anthropometric characteristics. In the study group, no correlation between hormone levels and ESLD was found, which is inconsistent with an earlier study that showed a relationship between GH levels and PELD score [60, 61]. This can be explained by many factors, in particular, differences in the structure of the incidence and degree of liver damage in patients in the studied samples.

7. GH/IGF-1 and clinical parameters of children with cholestatic diseases before LT

The patients included in the study had various etiologies of ESLD. Analysis of the structure of liver diseases showed that the largest group, about 60%, consisted of recipients with BA, a congenital malformation of the biliary tract (Figure 11).

Figure 11.

Structure of morbidity in the studied liver transplant recipients, BA, biliary atresia; BH, biliary hypoplasia.

Plasma GH levels in the patients differed depending on the etiology of liver disease: 5.5 (4.8–13.7) ng/ml in Alagille syndrome, 4.9 (2.0–7.3) ng/ml in BA, 1.6 (1.2–2.3) ng/ml in others, 3.9 (1.3–4.8) ng/ml in Byler disease, 3.6 (1.5–5.2) ng/ml in BH, 3.9 (0.0–7.7) ng/ml in Caroli disease (Figure 12).

Figure 12.

GH levels in the blood of children with various etiologies of liver disease, * - p = 0.012 compared with the level in children with BA, BA, biliary atresia; BH, biliary hypoplasia.

Comparative analysis of GH levels in children with different etiologies of liver disease using the Kruskal-Wallis test for several independent variables revealed no statistically significant differences in GH concentrations (p = 0.16). Pairwise comparison of GH concentrations in the groups with different diagnoses using the Mann-Whitney U test showed that the plasma GH level in children with ESLD resulting from BA was significantly higher than in patients grouped under “other” diseases (autoimmune and fulminant hepatitis, Von Gierke disease, portal vein developmental anomaly, Crigler-Najjar syndrome, Budd-Chiari syndrome, hepatoblastoma, alpha-1 antitrypsin deficiency, and previous liver transplant dysfunction), p = 0.012. No significant differences were found between the other groups.

Analysis of plasma IGF-1 concentrations depending on the etiology of liver disease showed that the hormone content in the blood of patients with Alagille syndrome was 12.6 (0.0–28.1) ng/ml, with BA 1.2 (0.0–20.0) ng/ml, others 42.3(3.3–72.3) ng/ml, Byler disease 25.1 (14.7–51.7) ng/ml, BH 1.1 (0.0–12.0) ng/ml, Caroli disease 3.6(0.0–8.7) ng/ml (Figure 13).

Figure 13.

The level of IGF-1 in the blood of children with various etiologies of liver disease, * - p < 0.05 in comparison with the level in children with BA; BA, biliary atresia; BH, biliary hypoplasia.

Multiple comparative analysis showed significant differences in IGF-1 levels in children with different etiologies of liver disease, p = 0.01 according to Kruskal-Wallis test. Pairwise comparison of IGF-1 levels in the groups with different diagnoses using the Mann-Whitney U test revealed that IGF-1 levels significantly differed in the group BA and “others” (p = 0.004), BA, and Byler disease groups (0.012). Differences were also found in the hormone level between the BH and “other” groups (p = 0.013).

It is possible that IGF-1 levels in the children depend on the etiology of liver disease; however, since the “other” group included various pathologies and each of them is small, further studies are needed for an unambiguous conclusion.

It is known that the nature of GH secretion differs in males and females and depends on age [62]. In our study, we could not identify a significant correlation between GH level and the age of patients before transplantation (r = −0.18, p = 0.10). This could be probably due to both impaired hormone secretion associated with the disease and by the age structure of the sample used in our work. In the control group, there was also no significant correlation with age, which can be explained by the narrow age range of the sample and/or a small number of observations.

A comparative analysis of plasma GH levels in girls (4.7 (1.7–7.1) ng/ml) and boys (3.4 (1.6–7.2) ng/ml) with ESLD revealed no significant differences, p = 0.41. In the control group, we could not also identify significant differences in GH levels in healthy boys and girls.

Serum IGF-1 levels, like GH, change with age, while there are practically no sex differences in protein levels. In the studied children, no significant correlation was found between IGF-1 levels and age (r = −0.001, p = 0.96). In the control group, no significant differences were also found, which is obviously due to the narrow age range in the groups included in the study.

The IGF-1 content in girls and boys with ESLD was 3.9 (0.0–23.5) ng/ml and 11.1 (0.0–33.6) ng/ml, respectively, and did not differ significantly, p = 0.38. The result obtained showed that IGF-1 concentration in children with ESLD is not associated with the sex of the child. No significant differences were found in the control group either.

The absence of age and sex differences in GH and IGF-1 levels in the group of children aged 3 months to 5 years with ESLD can be explained by a number of reasons—the absence of such differences at this early age, and impaired hormone secretion associated with the disease. Our data do not provide an unambiguous answer to this question due to the age composition of the study group. Of course, studies with a large number of observations and in more homogeneous age groups will be informative.

8. Changes in the content of GH and IGF-1 in the blood plasma of children after LT

To assess the effect of LT on the levels of the studied hormones in the early and late post-transplant period, the levels in the recipients were measured 1 month and 1 year after the transplantation.

Significant changes in the plasma GH levels were found in the children before, a month and a year after liver transplantation (Figure 14).

Figure 14.

The level of GH in the blood of children after LT. * – p < 0.05 in comparison with the level before LT.

GH level 1 month after LT was 1.4 (1.1–2.4) ng/ml. It was significantly lower before the operation (p = 0.001), but did not differ from the hormone level in healthy children (p = 0.74).

A year after transplantation, the GH level was 2.5 (1.5–5.7) ng/ml, which was also significantly lower than before transplantation (p = 0.049) and did not differ from the healthy children’s level (p = 0.67).

The change in IGF-1 concentrations in the children a month and a year after liver fragment transplantation is shown in Figure 15.

Figure 15.

The level of IGF-1 in the blood of children after LT, * - p < 0.05 in comparison with the level before LT.

In the blood plasma of the pediatric recipients, there was a significant increase in IGF-1 levels compared with the values before transplantation. One month later, the IGF-1 concentration was 73.5 (52.1–128.5) ng/ml (p = 0.000 compared with the concentration before transplantation). One year after transplantation, plasma IGF-1 level was 76.1 (42.9–111.7) ng/ml and was also significantly higher than before liver transplantation (p = 0.000 compared with the level before transplantation) and did not differ from the level in healthy children (p = 0.48).

A comparative analysis of changes in levels of the hormones in the pediatric recipients before and after LT depending on the gender of the patient was carried out. It was revealed that GH concentrations significantly differ among girls and boys after LT (Figure 16).

Figure 16.

The content of GH in the blood of children after LT, depending on gender, * - p < 0.05.

The analysis showed that plasma GH concentrations in girls and boys did not differ before LT - 4.7 (1.7–7.1) ng/ml and 3.4 (1.6–7.2) ng/ml, respectively, p = 0.41, and a month after - 1.4 (1.1–2.5) ng/ml and 1.6 (1.2–2.4) ng/ml, respectively, p = 0.8. Whereas a year after transplantation, the levels in girls were significantly higher than those in boys—4.6 (2.1–5.9) ng/ml and 1.8 (1.2–2.8) ng/ml, respectively, p = 0.002. It is difficult to assess the reasons for the differences in GH levels after LT in boys and girls in the study sample, because for healthy children under 6 years of age, according to the reference values, differences in GH levels between the sexes are insignificant. However, given the sex differences in GH hormonal regulation in later puberty, it cannot be ruled out that periods of differences between boys and girls can be observed in young children. Prescription of immunosuppressants after LT can also affect the result.

After LT, IGF-1 levels increased in both girls and boys equally (Figure 17).

Figure 17.

The content of IGF-1 in the blood of children after LT, depending on gender, * - p < 0.05.

There were no differences in IGF-1 levels in the children depending on gender, as before transplantation: IGF-1 levels in girls and boys were 3.9 (0.0–23.5) ng/ml and 11.1 (0.0–33.6) ng/ml, respectively, p = 0.38, and a month after it - 74.2 (54.0–150.3) ng/ml and 70.4 (50.6–119.5) ng/ml, respectively, p = 0.37. A year after the operation, the IGF-1 level in girls did not significantly differ from that of boys—83.1 (43.0–118.9) ng/ml and 70.1 (42.9–108.4) ng/ml, respectively, p = 0.68.

9. Relationship between GH/IGF-1 levels and anthropometric characteristics of recipients after LT

Analysis of changes in the anthropometric parameters of the pediatric recipients was carried out 1 year after LT, because 1 month after the operation, these indicators practically do not differ from preoperative ones.

One year after transplantation, the average height (82.1 ± 7.6 cm) and body mass (11.5 ± 2.2 kg) in the examined group were significantly higher than before LT (p = 0.00 for both parameters). Figure 18 shows the change in the histogram of the distribution of height and body mass of the examined children one year after LT.

Figure 18.

Change in the histogram of the distribution of height (A) and body mass (B) of recipient children 1 year after LT.

The results presented in the histogram show that the children’s height and weight change a year after LT, close to the normal distribution curve.

Comparison of the average height and body mass in the study group with the reference values for healthy children of the same age (mean age, 23 months) showed that the body mass of the recipients did not significantly differ from the reference values (12 ± 2 kg), p = 0.06.

The average height of the recipients was still significantly below the average reference value for healthy children according to WHO standards (87 ± 7 cm), p = 0.00 [55].

Currently, there are no published studies of long-term follow-up of the anthropometric parameters of liver transplant pediatric recipients, and therefore, it is difficult to assess how fully their growth is restored in the long-term post-transplant period. For example, it has been found that for child kidney recipients, there is a problem of growth retardation both before and after transplantation, which is associated with many reasons, including taking immunosuppressants. For such patients, recommendations have been developed on how to correct growth retardation by taking recombinant GH [63].

As already noted, we compared the anthropometric indicators of the study group with healthy children by comparing the histogram of the distribution of height and body mass with a normal distribution curve superimposed on a similar range of the anthropometric indicator (Figure 19).

Figure 19.

Histograms of the distribution of height (A) and body mass (B) of recipient children one year after LT in comparison with the normal distribution curve.

An analysis of the relationship between GH/IGF-1 levels and anthropometric parameters 1 year after LT showed that GH levels significantly correlate with body mass (r = −0.27, p = 0.046)—the lower the plasma GH level, the greater the body mass. The correlation of GH levels with patients’ height was also negative, but did not reach statistical significance (r = −0.20, p = 0.143). The correlation of IGF-1 content in the blood with the height (r = 0.19, p = 0.055) and body mass of patients (r = 0.18, p = 0.074) was positive, but did not reach statistical significance. The described relationships reflect the restoration of the physiological regulation of anthropometric parameters in children after LT due to the restoration of IGF-1 production by the graft and provision of GH bioavailability.

A correlation analysis was carried out to assess the relationship between GH and IGF-1 levels in the blood of children after LT. It revealed the restoration of the relationship between the levels of hormones in pediatric recipients already a month after LT (r = −0.28, p = 0.011); the correlation strength increased a year after LT: r = −0.47, p = 0.01 (Figure 20).

Figure 20.

Correlation of IGF-1 and GH levels in children 1 year after LT.

The presented results show that in pediatric recipients a year after LT, a statistically significant negative correlation is observed between the plasma GH and IGF-1 levels, which indicates normalization of the relationship between the hormones and normalization of the regulatory mechanisms of anthropometric characteristics in the children. The IGF-1-to-GH ratio can serve as a kind of integral indicator of such normalization.

Analysis before and after LT showed that this ratio increased significantly after LT (Figure 21).

Figure 21.

The ratio of IGF-1/GH levels in pediatric recipients before, 1 month and 1 year after LT.

Before LT, the IGF-1-to-GH ratio was 5.5 ± 4.2 and was significantly lower than that in healthy children - 37.1 ± 18.4 (p = 0.001); a month later, this parameter had increased by almost an order of magnitude, reaching 62.2 ± 41.0 (p = 0.001), and a year later, it had decreased to 32.6 ± 21.8 and did not differ from that in healthy children (p = 0.98), but was higher than before transplantation (p = 0.002). The data obtained suggest that mutual regulation of the GR/IGF-1 axis is restored 1 year after LT.

Obviously, normalization of the feedback between the GH and IGF-1 levels a year after LT is an indicator of the restoration of the normal ratio of the components of the central and peripheral growth regulation system and is a consequence of the restoration of liver function.

Plasma GH and IGF-1 levels not only depend on liver function, but also largely determine its condition and can be considered as indicators of liver function in patients with hepatobiliary diseases. Changes in these levels after LT can serve as an objective indicator of the degree of normalization of the synthetic function of the graft and restoration of neurohumoral regulation of the body of pediatric liver recipients.

Separate long-term post-transplant follow-up of pediatric liver recipients shows that with good graft function, the anthropometric characteristics of children can be fully restored and do not differ from those of healthy children of the same age.

Figure 22 shows a photo of patient P. (in the center outlined by a dotted line) with her classmates 7 years after LT. It can be seen that her height is practically the same as that of her peers.

Figure 22.

Patient P., 7 years after LT.

This study has revealed that the GH/IGF-1 axis plays a critical role in the pathogenesis of ESLD in pediatric patients. Children with ESLD have high PELD score, growth retardation, and decreased body mass. Their serum GH levels are much higher than in healthy children, while IGF-1 levels are lower than in healthy children. After LT, the GH/IGF-1 axis recovers, improving anthropometric parameters in pediatric recipients. Result of the study suggests that the GH/IGF-1 axis is of clinical significance in pediatric liver recipients. The GH/IGF-1 levels and the IGF-1-to-GH ratio can reflect liver graft function.

10. Conclusion

ESLD results in a decline in liver production and deficiency of IGF-1, which impair GH bioavailability for peripheral tissues. This hormonal disorder leads to delayed growth in children with congenital cholestatic diseases.

Our studies have shown that children with ESLD have high GH and low IGF-1 levels, which is combined with growth failure and body mass deficit.

After pediatric LT, GH concentrations fall, while those of IGF-1 increase. As early as a month after LT, the plasma concentrations of both hormones do not differ from those in healthy children, and a year after the transplantation, they remain at normal levels.

After LT, the relationship between plasma GH and IGF-1 levels is restored, which was disrupted in children with ESLD.

Normalization of the ratio between serum levels of IGF-1 and GH a year after LT can serve as an integral indicator of the restoration of hormonal regulation of the physical development of pediatric recipients. A year after LT, GH and IGF-1 levels in pediatric recipients are comparable to those in healthy children.

GH bioavailability, mediated through IGF-1, is restored due to normal production of IGF-1 by the graft. The body height and body mass of pediatric recipients are also comparable to those of healthy children. This reflects the restoration of central and peripheral hormonal regulation of growth, which is a consequence of graft functioning.

Acknowledgments

The authors would like to thank all the transplant teams from Shumakov Center, who provided patients’ data and samples for this study.

The study was supported by research grants from the Ministry of Health of the Russian Federation.

Conflict of interest

The authors declare no conflict of interest.