Abstract
Objective:
Risperidone is a promising agent for the treatment of schizophrenia, Tourette's disorder, mood disorders, and disruptive behavior disorders in young populations. However, adverse effects of risperidone may take a long time to emerge. The objective of this study was to investigate the changes in the liver function tests (LFTs) associated with more than 6 months of risperidone treatment in children and adolescents.
Method:
A total of 102 youths treated with risperidone for more than 6 months were eligible for the study. For this study, patients' baseline and follow-up weight and hepatobiliary function tests, including alanine aminotransferase (ALT), aspartate aminotransferase (AST), gamma glutamyl transpeptidase (GGT), alkaline phosphatase (ALP), and serum bilirubin levels, were measured at baseline and at 6 months.
Results:
Asymptomatic abnormalities of LFTs, mostly ALP elevation, were found in 38.2% of the cases, and marked elevation of liver enzymes was found in 0.8% of cases treated with risperidone. The mean levels of liver enzymes and bilirubin of the patients were significantly higher than the baseline after first and sixth months of treatment. However, there was no statistically significant change in the liver enzymes and bilirubin levels between the first and sixth months. There was no significant relationship between changes in weight and liver enzymes and biluribin levels after 6 months of risperidone treatment.
Conclusion:
These findings suggest that risperidone treatment in the long term commonly leads to liver function changes, although at therapeutic doses in children and adolescents it may rarely induce a serious hepatic toxicity. Concomitant use of antidepressants and methylphenidate and variations in age and pubertal status are limitations of present study. Further studies are needed to assess the importance and role of other variables over LFT abnormalities in youth population.
Introduction
Atypical antipsychotics commonly have been reported to lead to asymptomatic elevation of liver enzymes in the adult population, and, recently in children and adolescents (Gaertner et al. 2001; Mouradian-Stamatiadis et al. 2002; Erdogan et al. 2008). Results from controlled clinical trials, reports of spontaneous adverse events, and published studies/case reports suggest that severe hepatotoxicity may be rare in the pediatric population (Szigethy et al. 1999; Zuddas et al. 2000; Malone et al. 2002; Turgay et al. 2002; Masi et al. 2003). However, several cases of hepatotoxicity with long-term use of risperidone and other antipsychotic agents have been reported recently in adults and in the pediatric population (Kumra et al. 1997; Krebs et al. 2001; Mouradian-Stamatiadis et al. 2002; Llinares et al. 2005). Furthermore, weight gain during risperidone therapy has already been documented along with fatty infiltration of the liver, which has been reported to occur in obese children and adolescents (Moran et al. 1983; Ratzoni et al. 2002).
Studies of safety and effectiveness of risperidone conducted on large samples have found no clinically significant abnormalities of liver function tests (LFTs) with treatment; however, the rate and type of clinically nonsignificant LFT abnormalities were not been mentioned in those studies (Zuddas et al. 2000; Malone et al. 2002; Turgay et al. 2002; Masi et al. 2003; Shea et al. 2004). Furthermore, until now there has been no consensus on the necessity and timing of testing for liver functions in young patients who are using atypical antipsychotics. Also, adverse effects of atypical antipsychotics on the pediatric population can take a long period of time to emerge. Therefore, it may be argued that the effects of risperidone on liver function of children and adolescents, especially in the long term, are currently unclear. In this study, we aimed to investigate the changes in the LFTs associated with more than 6 months of risperidone treatment in children and adolescents. The relationship between sociodemographic and clinical features and co-morbid drug use and weight gain and abnormalities in LFTs were also evaluated.
Methods
The children and adolescents treated with risperidone for any psychiatric problem at the outpatient clinics of Department of Child and Adolescent Psychiatry, University of Zonguldak Karaelmas Hospital were recruited for the study. The parents of all subjects gave written informed consent; oral assent was procured from the children and adolescents for participation in the study after the procedures were fully explained to them and their questions were answered. The University of Zonguldak Karaelmas, School of Medicine Ethical Committee approved the study protocol. All of the procedures in the study were carried out in accordance with the Declaration of Helsinki and local laws and regulations.
Subjects who had a co-morbid alcohol or drug abuse or requiring treatment with medications other than selective serotonin reuptake inhibitors (SSRIs), tricyclic antidepressants (TCAs), and methylphenidate (MPH) were excluded. Patients using vitamin/mineral supplements and/or herbal remedies, having a history of hepatic disease, having any other active and chronic illness, and lacking follow-up information were also excluded. To be included in this study, all patients had to be drug free for at least 2 weeks before the start of the trial. Baseline liver function tests (ALT, alanine aminotransferase; AST, aspartate aminotransferase; ALP, alkaline phosphatase; GGT, gamma glutamyl transpeptidase; IDB, indirect bilirubin; DB, direct bilirubin) had to be normal before the initiation of risperidone treatment.
After subject recruitment and completion of the baseline assessments, children and adolescents were put on open-label risperidone treatment according to their psychiatric conditions. None of the subjects had been on risperidone before the study. An individualized flexible titration procedure was used in all patients to adjust the dosage for optimal efficacy and minimal untoward effects. Baseline demographics, including age, gender, co-morbid medical conditions, and past medical history, of all the patients were evaluated. Follow-up visits to the child and adolescent psychiatry outpatient department occurred after 2 weeks of risperidone administration and monthly thereafter. Medication use and LFTs were evaluated during these visits.
The patients who needed long-term treatment with risperidone and continued treatment for at least 6 months were included in the study. Laboratory data were obtained for each subject to evaluate changes in LFTs, including ALT, AST, GGT, ALP, and serum bilirubin levels. The tests were obtained at the baseline and the first and sixth months of treatment. Monotherapy was defined as the absence of simultaneously administered medication.
Weighing was always performed on the same type of scale, with the patients lightly clothed. The weight change of participants was reported with respect to two measures: (1) Absolute weight change in kilograms and (2) proportional weight change, i.e., percentage change from baseline.
Laboratory data and statistical tests
The lab analyses of all the subjects were done in the same clinical laboratory. Baseline LFTs were measured (ALT, AST, GGT, ALP, and serum bilirubins) before the initiation of risperidone treatment. Follow-up measurements were done at the first and sixth months of treatment.
For this study, abnormal liver function was defined as the serum ALT, AST, GGT, ALP, or serum bilirubin values falling outside the normal laboratory ranges. The following limits were defined as elevations of liver enzymes: ALT (7–35 U/L for females, 10–40 U/L for males), AST (7–35 U/L for females, 10–40 U/L for males), GGT (5–39 U/L for females, 10–66 U/L for males), and ALP (1–9 years age, >420 U/L for females and males; 10–11 years age, >560 U/L for females and males; 12–13 years age, >495 U/L for males, >420 U/L for females; 14–15 years age, >515 U/L for males, >230 U/L for females; 16–19 years age, >260 U/L for males, >130 U/L for females.
Elevations in liver enzymes and bilirubin were accepted as clinically relevant if ALT, AST, GGT, and ALP levels increased to two-fold of the established reference ranges and if DB and IDB values were greater than normal values (>0.4 mg/dL; >0.8 mg/dL, respectively). In the case of a two-fold elevation in hepatic parameters upon treatment with medication, the agent was withdrawn. Then the patients were examined and followed up with regular laboratory monitoring in collaboration with the Department of Pediatrics. The other possible causes of elevations in liver enzymes, such as chronic viral hepatitis B and C, alcoholic liver disease, hemochromatosis, and nonalcoholic fatty liver disease, were also evaluated in these patients.
A total of 162 child and adolescent patients treated with risperidone were included in the study. Five patients who had liver enzyme abnormalities before the antipsychotic therapy were excluded. Of the 157 subjects, 22 did not come for follow-up visits and 17 stopped taking drugs due to untoward effects (dyskinesia, 6 subjects; sedation, 5 subjects; weight gain, 4 subjects; nocturnal enuresis, 2 subjects). Treatment of 18 subjects was withdrawn or their medication was changed due to untoward effects or ineffectiveness of medications (dyskinesia, 2 subjects; sedation, 3 subjects; increased activity, 2 subjects; diurnal enuresis, 2 subjects; ineffectiveness, 9 subjects). After 6 months, 102 patients (74 male, 28 female, ages between 2 and 18, mean age 9.16 ± 4.17) were eligible for the study.
Statistical analyses
Statistical Package for the Social Sciences (SPSS) for WindowsTM, Version 15.0 (Chicago, IL) was used for statistical comparisons. To determine whether serum ALT, AST, GGT, and bilirubin concentrations were significantly different among groups throughout the period that patients used risperidone, data were analyzed by using analysis of variance (ANOVA) for repeated measures followed by F tests. A general linear model was performed with gender, age, diagnosis, daily dose of risperidone, and combined medication usage to determine the significant predictors for abnormality in the total liver enzymes and bilirubin. A value of p < 0.05 was considered significant. Values are expressed as means ± standard deviation (SD). All of the comparisons were two-tailed.
Results
Patients included in our sample had a wide variety of mental disorders, with the most common being disruptive behavior disorders (including conduct disorder, oppositional defiant disorder, mental retardation or subaverage intelligence quotient [IQ], and ADHD, n = 66, 64.7%). Other diagnoses were listed as follows: Anxiety disorders (n = 13, 12.7%), PDDs (n = 11, 10.8%), tic disorders (n = 9, 8.8%), and psychotic disorders (n = 3, 2.9%).
The mean dose of risperidone ranged from 0.25 to 6 mg/ day (mean, 0.73 ± 0.77) or 0.01 to 0.32 mg/kg per day (mean, 0.023 ± 0.032). Thirty-three participants (32.4%) took 0.25 mg/day, 36 participants (35.3%) took 0.5 mg/day, 22 participants (21.6%) took 1 mg/day, and 11 (10.7%) took 1.5 mg/day or more. Twenty-five percent of the patients (n = 25) on risperidone were taking antidepressant medications (all of them using SSRIs, 24.5%) in addition to risperidone. Thirty-two patients (31.4%) were taking MPH in addition to risperidone.
Liver function tests and weight changes
No cases of jaundice or hepatic failure were seen in the treatment period of 6 months. Only in 1 male patient, aged 8 years, were ALT levels increased up to three-fold and AST levels increased up to two-fold from the baseline levels (ALT, 225 U/L; AST, 83 U/L; GGT, 27 U/L; ALP, 641 U/L; DB, 0.11 mg/dL and IDB, 0.40 mg/dL, at a risperidone dose of 0.25 mg/day) at the first month of the treatment. When risperidone was discontinued, the liver function tests of that patient returned to normal after 1 month. The liver enzymes levels did not increase up to two-fold from the baseline levels in any of the other cases in the treatment period of 6 months.
Mean levels of liver enzymes and bilirubin levels of the patients at the first and sixth months were significantly higher than baseline. Elevation of liver enzymes and bilirubin from the defined normal levels was observed in 38.2% of the cases at the sixth month. However, there was no significant change in the liver enzymes and biluribin levels between the first and sixth months. Comparison of LFTs at the baseline, first, and sixth months are presented in Table 1. The increase of ALT above the normal range was seen in 1 (1.0%), increase of AST in 9 (8.8%), increase of ALP in 22 (21.6%), increase of DB in 3 (2.9%), and increase of IDB in 3 (2.9%) patients.
Abbreviations: ALT, alanine aminotransferase; AST, aspartate aminotransferase; GGT, gamma glutamyl transpeptidase; ALP, alkaline phosphatase; DB, direct bilirubin; IDB, indirect bilirubin.
A general linear model was performed with gender, age, diagnosis, daily dose of risperidone, and combined medication use to determine the significant predictors for abnormality in the total liver enzymes and bilirubin. Change in the levels of ALT, ALP, GGT, and DB were not significantly related to any variables (p > 0.05). A significant relationship was found between age and change in IDB (p = 0.048), with IDB change increasing with age. A significant relationship was found between risperidone dose and IDB change (p < 0.0005). The mean levels of liver enzymes and bilirubin according to combined medication use at the baseline and after sixth months of treatment are shown in Table 2.
Abbreviations: SD, Standard deviation; ALT, alanine aminotransferase; AST, aspartate aminotransferase; GGT, gamma glutamyl transpeptidase; ALP, alkaline phosphatase; DB, direct bilirubin; IDB, indirect bilirubin; SSRI, selective serotonin reuptake inhibitor; MPH, methylphenidate.
Mean body weight increased significantly for most of the participants (p = 0.002). Weight gain was found in 67 patients out of 102 patients (65.7%). At the end of the sixth month of the study, mean weight had increased to 37.5 kg from a baseline mean weight of 31.98 kg (mean increase from baseline, 5.5 kg). However, there was no significant association between changes in weight gain and liver enzymes and bilirubin levels.
Discussion
In this study, we evaluated the changes of liver enzymes and serum bilirubin levels with 6 months of risperidone treatment in a child and adolescent patient population. Elevation of liver enzymes and bilirubin from the defined normal levels was observed in 38.2% of the cases at the sixth month of treatment. Our results are consistent with previous studies that reported clinically insignificant changes of liver enzymes and bilirubin levels with risperidone treatment in adults and youths (Whitworth et al. 1999; Gaertner et al. 2001; Dumotier et al. 2002; Garcia-Unzueta et al. 2003; Pae et al. 2005; Atasoy et al. 2007; Erdogan et al. 2008). Concordant with those studies, the majority of transaminase elevations seen in the present study population were clinically insignificant. In the present study, three-fold elevation in liver enzymes was observed only in 1 case (0.8%) in the first month of treatment, but the levels returned to the normal levels in 4 weeks after discontinuation of risperidone. Similar to our findings, the majority of significant transaminase elevations that were seen during the risperidone treatment were reported to be mostly reversible (Dumotier et al. 2002; Atasoy et al. 2007). In contrast to our findings, some studies did not find liver enzyme abnormalities with risperidone (Szigethy et al. 1999). Szigethy et al. (1999) reviewed the charts of 38 youths treated with risperidone and found that 37 of the 38 youths treated with risperidone had no liver enzyme abnormalities. The mean length of risperidone treatment was 15.2 months and mean dose was 2.5 mg/day in their study population.
To our knowledge, there have been safety and effectiveness studies with large samples that found no clinically significant LFT abnormalities with risperidone treatment in developing children and adolescents, although the rate and type of clinically nonsignificant LFT abnormalities were not mentioned in these studies (Zuddas et al. 2000; Malone et al. 2002; Turgay et al. 2002; Shea et al. 2004). It may also be important to point out that the relative frequency of ALP elevation in the present study differs clearly from the previous study results with risperidone in adult and child populations (Kumra et al. 1997; Gaertner et al. 2001; Sikich et al. 2004; Atasoy et al. 2007). Although elevation of ALP did not cause any treatment discontinuation, asymptomatic enzyme elevations, observed mostly in ALP and AST after 6 months of risperidone treatment in the present study, are consistent with our previous study that reported the effects of 1 month of risperidone treatment on LFTs (Erdogan et al. 2008). This result contrasts with the results of studies in adult patients, where most authors reported that ALT is the enzyme most often and most strongly affected during antipsychotic treatment (Hummer et al. 1997; Gaertner et al. 2001). In a child and adolescent population, Sikich et al. (2004) found increases in AST and ALT whereas Kumra et al. (1997) found increases in AST, ALT, GGT, and ALP with risperidone.
Similar results of increased levels of ALP were observed only with other typical antipsychotics, namely phenotiazines, haloperidol, and chlorpromazine, in an adult population (Dincsoy and Saelinger 1982; Obata 1983; Cepelak et al. 1986; Hütteroth 1989; Garcia-Unzueta et al. 2003). Accordingly, those antipsychotics were reported to induce a cholestatic form of injury or primary liver damage and injury to liver cells caused by cholestasis (Obata 1983; Dölle and Martini 1984; Cepelak et al. 1986; Thomas 1995; Garcia-Unzueta et al. 2003). This elevation of ALP, which is not seen in a risperidone-treated adult population, needs to be carefully evaluated and interpreted (Atasoy et al. 2007). It may be due to the risperidone side-effect profile being similar to typical antipsychotic agents compared to other atypical antipsychotic agents, namely olanzapine, clozapine, and quetiapine (Serretti et al. 2004). ALP elevation was reportedly observed during treatment with psychoactive drugs (phenytoin, primidone, and phenobarbital) that have been shown to interfere with metabolism of calcium, phosphorus, and vitamin D (Kafali et al. 1999). Therefore, and taking into consideration the case report of Kumra et al. (1997), an alternative explanation may be that risperidone may display an age-dependent effect on the metabolism of calcium, phosphorus, and vitamin D. It can be said that the relationship of risperidone treatment and elevation in ALP needs to be further investigated.
One of the limitations of the present study may be that approximately 50% patients were using SSRIs and psychostimulants in addition to risperidone. Abnormalities in liver enzymes were reported with antidepressant agents, but liver enzyme elevations have been rarely reported with MPH in the literature (Wigal et al. 2006; Schatzberg 2007). However, in the present study, no difference was found in enzyme and bilirubin levels between subjects using combined antidepressant agents or MPH and subjects receiving monotherapy with risperidone. Additionally, patients who had been using mood-stabilizing or hypnotic drugs were excluded because prescription of those agents was reported to more commonly increase the susceptibility to hepatic trouble compared to the antidepressant medicines (Sherlock and Dooley 1993; Chitturi and George 2002; Gonzalez-Heydrich et al. 2003).
Although Garcia-Unzueta et al. (2003) found significant relationships between psychiatric diagnoses and increase of liver enzymes and bilirubin levels in adult patients, no significant relationship was found between psychiatric diagnoses and increase of liver enzymes and bilirubin levels in the present study. Their finding may be due to the factors related to life habits, feeding, etc., of adult patients with chronic psychiatric disorders that could negatively affect the liver enzymes and biluribin levels. Contrary to previous reports, there was no significant difference between males and females regarding the increase of liver enzymes and bilirubin levels in the present study (Kumra et al. 1997). This may be due to sampling bias and should be clarified with further studies.
Most of the risperidone-treated children and adolescents experienced significant weight gain between baseline and end point, despite the low dose of risperidone used in this study. Our findings of a 5.5-kg weight gain in 6 months are similar to previous study findings (Kelly et al. 1998; Martin et al. 2000; Fleischhaker et al. 2007). Recently, Correll et al. (2009) studied cardiometabolic effects of second-generation antipsychotics, including risperidone, in 272 pediatric patients (aged 4–19 years). They found that after a median of 10.8 weeks of treatment, weight increased by 5.3 kg (95% confidence interval [CI], 4.8–5.9 kg) with risperidone. Consistent with other study results, no gender differences were found in weight gained by subjects in the present study (Kelly et al. 1998; Martin et al. 2000).
Although there was no significant association between changes in weight and liver enzymes and bilirubin levels in the present study, the association of marked weight gain and hepatotoxicity with risperidone treatment has been reported in youths. Therefore, it can be said that further studies are needed to reveal the relation of long-term weight gain and LFT changes associated with risperidone (Moran et al. 1983; Holtmann et al. 2003; Aman et al. 2005).
This pilot study has a number of limitations, including differences in the diagnoses of participants in the different treatment groups, use of concomitant medications, and variations in age and pubertal status. Prescription of antidepressant agents may increase the susceptibility to hepatic trouble (Wigal et al. 2006; Schatzberg 2007; Findling et al. 2004). Another limitation of this study is that we did not collect information on whether patients have prior exposure to other medications prior to the washout period. However, to ameliorate this limitation, we included only patients who had normal baseline LFTs in the study. Also, there were more male than female patients; however, our study population reflects the preponderance of boys among patients with disruptive behavior disorders (Steiner 1997). In addition, the baseline height and body mass index (BMI) were not measured, because these dimensions were not the primary focus of our research. However, given that a height increase is expected in this age group, BMI rather than body weight might have been a more meaningful measure of weight gain in this population. Another limitation is that no information concerning our patients' dietary behavior before treatment was available, whereas history of dieting and concern of weight gain may have effect on weight gain in risperidone-treated patients (Safer 2004). Although data obtained from the analysis of 102 patients were used in this study, the small number of subgroups limited the opportunity to investigate the stratification of hepatic enzymes with adequate statistical power. Further studies are needed with larger samples to allow more reliable conclusions. Finally, other factors that might affect the level of hepatic enzymes, such as changes of other metabolic parameters like glucose and serum lipids, were not evaluated in the present study by the random timing of samples throughout the day without clear relationship to meals. The potential correlations need to be explored.
Conclusion
We found asymptomatic liver function test abnormalities, mostly ALP elevation in about 38.2% of subjects and marked liver enzymes elevation in 0.8% of subjects treated with risperidone. Although risperidone treatment may rarely cause serious LFT abnormalities, it commonly leads to asymptomatic LFT abnormalities at therapeutic doses in a pediatric population. Because the effects of risperidone on liver function of children and adolescents, especially in the long term, are currently unclear, we suggest that clinicians who prescribe risperdone to a pediatric patient should monitor liver enzymes at baseline and routinely in the long term. Further studies are needed to assess the importance of LFT abnormalities and role of other variables over these abnormalities in a population of youths.
Footnotes
Disclosures
The authors declare that they have no conflicts of interest.
The statistical consultant for this study was Mehmet Ali Kurcer, Associate Professor, Department of Public Health, Zonguldak Karaelmas University, Faculty of Medicine, Zonguldak, Turkey.
This study has not been published elsewhere, either in part or in its entirety. No support was received for this study. All analyses and writing were done by the authors themselves.
