Introduction

Crigler—Najjar syndrome (CNS) is a rare autosomal recessive genetic disease that occurs in less than one in a million births. The syndrome is caused by the inactivity of an enzyme called UDP glycuronyltransferase A1 (UGT1A1) [1]. In the liver UGT1A1 binds unconjugated bilirubin to two glucuronic acid molecules and convert it to conjugated bilirubin. Unconjugated bilirubin is insoluble in water, so, it cannot be excreted in bile and urine, but is fat-soluble and thus crosses the brain and placental blood barriers. The conjugated bilirubin is soluble in water, so it enters the bile and is excreted through the small intestine [2]. CNS is associated with the unconjugated hyperbilirubinemia. The disease is divided into two categories based on severity. No enzymatic activity is seen in patients with type 1 of the disease and unconjugated bilirubin levels are as high as 20% to 45% of total bilirubin. This condition can lead to unconjugated bilirubin deposition in the basal nuclei of the brain, which causes kernicterus, long-term and even fatal neurotic and brain symptoms [3]. Before the introduction of phototherapy and blood transfusion as treatments of high bilirubin levels, patients used to die from kernicterus in their first 18 months of life. One of the treatments suggested for patients with type 1 of Crigler—Najjar syndrome (CNS1) is phototherapy up to 20 hours a day for infants and 12 hours a day for children [4]. Over time, the effectiveness of phototherapy decreased due to factors such as lack of patient cooperation, decreased surface-to-volume ratio, increased skin thickness, and skin lesions [5]. Treatment with ursodiol is often prescribed, and phenobarbital administration can be more effective in patients with less enzymatic defects. Other treatments include blood transfusion, albumin injection, and avoidance of drugs that cause bilirubin displacement [6, 7]. Definitive treatment of CNS1 is liver transplantation, although there is disagreement about the appropriate time for transplantation due to its associated complications. There is also controversy about the efficacy of transplantation for improvement of neurologic symptoms [7]. Although there are some disagreements over the level of improvement after transplantation, the goal of treatment is to perform the transplant before the neurological symptoms appear [8]. Theoretically, samples taken from patients with CNS1 should not provide specific histological findings via electron or light microscope, but a recent study confirmed the presence of silent fibrosis in the liver of CNS1 patients [9]. The main objective of the present study is to examine the liver histology of children with CNS1 who underwent organ transplantation in the Organ Transplantation Center of Shiraz University of Medical Sciences in the period from 2008 to 2018.

Methods

This retrospective cross-sectional study was performed to evaluate the liver pathologies of patients with CNS1. The research population included all children under 18 who had CNS1 and underwent liver transplantation in the period from 2008 to 2018 at the Namazi Liver Transplant Center affiliated with Shiraz University of Medical Sciences. Most of the patients were less than 2 years old, with an average age of 1.7 years. We studied the tissue of the explanted liver, which was replaced by transplants. The patients who had passed out during or after liver transplantation or those with insufficient information in their medical records were excluded from the study. Therefore, a questionnaire was developed to collect data from the patients’ pre- and post-transplantation medical records and follow-up sheets. The questionnaire contained all the patients’ pre- and post-transplantation data including demographic data, family history of CNS1, age of disease onset, age of the patient at the time of transplantation, history of blood transfusion, total bilirubin level at the time of transplantation, and neurological complications and disorders. Histological slides were recovered and assessed for cholestasis levels in H&E, fibrosis staining in Masson Trichrome, staining in case of the availability of ductular reaction by cytokeratin 7 immuno-histochemical staining. The intrahepatic severity of cholestasis was assessed for the pericentral regions of the portal area, and the bile plugs removed from the patient’s liver were graded. The presence of the pericentral cholestasis in zone 3 alone was rated as +1, cholestasis in the zones 2 and 3 was rated +2, and the panlobular distribution was considered as +3. In addition, the bile plugs were rated from 0 to +3, respectively, depending on the severity of cholestasis in these zones. Fibrosis was rated for both the traditional lobular and the portal areas. If the pericentral fibrosis had no evidence of fibrin-expanding filaments extending to the central venous collagen, it was rated as 0, if extended to sinusoids it was rated as +1, if it extended to areas 2 and 3 was rated as +2, and any evidence of central fibrous bridging was rated as +3. Portal fibrosis was rated between 0 and 6 degrees using the Ishak system. CK7 was used to estimate the intensity of the bile ductular reaction so that 0 showed the absence of ductular reaction, +1 indicated a mild ductular reaction, +2 was considered as a moderate ductular reaction, and +3 denoted extensive ductular reaction. Furthermore, the severity of cholestasis was assessed using a score of 0 to +3 depending on the amount, quality, and expansion of CK7 staining. The collected data were analyzed using SPSS 22 software. In addition, the data that were normally distributed were analyzed using Kruskal—Wallis and Mann—Whitney tests. All analyses were performed at the significance level of 0.05 (p=0.05).

Results

Out of 53 included patients 29 were female and 24 were female. 20 of them had a family relationship with the donor and 15 had a positive family history regarding CNS1in their close family. 31 patients out of 53 received a whole liver, while the rest of them received a part of liver. The most prevalent pathology found in the liver of these patients was periportal fibrosis (96.2%) which was mostly observed as grade 2 (37.7%). Cholestasis was the second common finding (94.3%) in the liver histology of CNS1 patients, mostly seen as grade 1(63.3%). Other pathologies observed in liver histology of these patients were pericentral fibrosis (86.7%) and ductal reaction (22.6%) (fig. 1—4). The complete distribution of liver pathologic findings are provided in table 1. We assessed the presence of possible correlations between our pathologic findings and some variables including Phenobarbital usage and phototherapy time period, age and liver function tests. Out of all correlation measures that we evaluated, a significant correlation was only present between phototherapy time and ductal reaction grade meaning that the more phototherapy the patient has received the higher his ductal reaction grade was. However, the correlation between direct bilirubin levels and cholestasis grade was close to significant. The P values for correlations among different variables can be seen in table 2.

Fig. 1. Liver cholestasis in a CNS1 patient.

H&E, ×400.

Fig. 2. Ductular reaction in a CNS1 patient.

H&E, ×400.

Fig. 3. Periportal fibrosis in a CNS1 patient.

Masson’s trichrome, ×100.

Fig. 4. Pericentral fibrosis in a CNS1 patient.

Masson’s trichrome, ×100.

Table 1. The prevalence of various grades of liver findings

Table 2. Correlation between phenobarbital usage, time of phototherapy, laboratory measures and different pathologic measurements

Discussion

Type 1 of Crigler—Najjar syndrome is a liver-based rare genetic disorder that affects less than one in a million people [1]. However, the haptic structure and synthetic function were not assumed to be changed in this disease and the major cause of morbidity was considered to bilirubin deposition in brain leading to severe neurological complications. Meanwhile, Mitchell et al. in their cohort study reported a 40% presence of liver histologic fibrosis in CNS1 patients [9]. In their cases the fibrosis was significantly more in the liver of older children and was not associated with altered biochemical measurements. Our study revealed a great prevalence of fibrosis in these patients accounting for 96.2% periportal fibrosis and 86.7% pericentral fibrosis. Although most of these patients have low grades of fibrosis, it must not be considered that 43.3% of patients presented periportal fibrosis in higher grades (>2). Our findings were consistent with Mitchell et al regarding the silent potential of fibrosis development in these patients, as we did not notice any significant correlation between the levels of different liver function tests with fibrosis grade. However, we could not approve what they mentioned about the relation between the patient’s age and the presence of fibrosis. No significant relationship was obtained between the age and fibrosis grade in our study. A case series conducted by Baris et al. on 7 pediatric patients reported mild hepatocanalicular cholestasis in 5 of them [10]. Our research found cholestatic changes in 94% of patients, 62% of which being in grade 1. We also found a significant correlation between the direct bilirubin level and the cholestasis level. In another study Baris et al introduced a case of CNS 2 who developed cirrhosis and concluded that CNS might be able to cause clear liver fibrosis and even liver cirrhosis that requires liver transplantation [11]. Fata et al reported a case of CNS2 as well. They illustrated marked canalicular cholestasis accompanied with periportal and portal fibrosis [12]. The final treatment of CNS1 is liver transplantation. However, before the transplantation, treatments such as phenobarbital, phototherapy and blood transfusion are used to help the patient eliminate the complications of high non-conjugated bilirubin [7]. In the present study we also evaluated the possible effects of these treatments on the level of liver pathologies. The only statistically significant correlation that we found was between ductal reaction grade and the period of time the patient has been receiving phototherapy. Some other encouraging therapies have been used recently such as gene therapy and hepatocyte transplantation. In the former method normal UGT1A1 genes will be transduced in the cultured hepatocytes in an ex vivo environment and will be delivered to the patient’s body using a vector. In addition, in the latter the physician infuses normal hepatocytes into the peritoneal space or the portal vein [13, 14]. Considering our results along with the few previous studies in this field, it can be understood that histologic examination in these patients specially while planning for gene therapy or hepatocyte transplantation is of great importance. We also evaluated the post-transplant complications in our patients which indicated infection (11.3%) as the most prevalent complication followed by ascites (9.4%), and chronic rejection (7.5%). Meanwhile, Baris et al in their study on 7 patients reported hepatic artery thrombosis, acute cellular rejection and bile leak each in 2 patients, while hepatic vein thrombosis, cholestasis and pneumothorax each were seen in one patient.

Conclusion

As a conclusion, our results indicated a high prevalence of fibrosis in different grades among CNS 1 patients which bolds the necessity of histologic examination before considering treatments such as gene therapy or hepatocyte transplantation. We found a significant correlation between phototherapy time and ductal reaction grade. High direct bilirubin levels also showed a positive relation with cholestasis grade.

The authors declare no conflicts of interest.