- Top of page
- Materials and Methods
- Clinical and research implications
The lower rate of extrapyramidal side effects and hyperprolactinemia has led to an increase in preference for second-generation antipsychotics (SGAs) in children and adolescents (Patel, Sanchez, Johnsrud, & Crismon, 2002). Risperidone is the first SGA approved by the US Food and Drug Administration (FDA) for patients aged <18 years. As with other SGAs, risperidone is used off label to treat self-injurious behaviors, impulsivity, irritability, and aggression associated with disruptive behavior disorders (DBD), tic disorders, anxiety disorders, and mood disorders (Croonenberghs, Fegert, Findling, De Smedt, & Dongen, 2005). At the time, the present study was conducted risperidone was the only SGA approved by the Turkish Ministry of Health for children aged <18 years and was commonly prescribed to children with autism spectrum disorders (ASD) (Ulay & Cengel Kultur, 2006).
Although SGAs are reported to be well tolerated, children and adolescents are more vulnerable than adults to their hyperprolactinemia, weight gain, and cardiometabolic side effects (Cohen, Bonnot, & Laurent, 2012; Correll, 2008; Correll et al., 2009; Hellings, Zarcone, Crandall, Wallace, & Schroeder, 2001; Safer, 2004; Sikich, Hamer, Basford, Sheitman, & Lieberman, 2004). When compared with other SGAs, the potential for weight gain and an increase in glucose and lipid levels associated with risperidone is considered moderate (Baptista, Kin, Beaulieu, & de Baptista, 2002; Cohen et al., 2012; Sikich et al., 2004). The FDA issued a warning about SGA-induced hyperglycemia and diabetes mellitus (DM), suggesting that every patient given SGAs must be evaluated for metabolic syndrome (MS), however, MS is a complex disorder and its very definition remains a controversial topic especially in children (Kassi, Pervanidou, Kaltsas, & Chrousos, 2011; Zimmet et al., 2007).
Although MS is associated with overweight, and insulin resistance (IR) plays a central role in the pathophysiology of MS, a number of other conditions are involved in the pathogenesis (Kassi et al., 2011). Numerous interactions between genetic (Farooqi & O'Rahilly, 2006) and nongenetic factors cause overweight and obesity; in many patients the etiology of obesity is too complicated to be definitively determined except for iatrogenic risk factors.
SGA-related weight gain has been explained via a series of hypotheses (Coccurello & Moles, 2010). Although the underlying mechanism of weight gain due to SGAs is associated with H1 and H3 receptors (Huang, Weston-Green, & Deng, 2010; Kroeze et al., 2003) and 5-HT2c receptor polymorphism (Reynolds, Templeman, & Zhang, 2005), the pharmacological basis of other metabolic changes is not fully understood. The impact of peptide hormones (leptin, adiponectin) on food intake and other metabolic parameters has gained importance as adipose tissue has become considered an endocrine organ (Jin, Meyer, Mudaliar, & Jeste, 2008; Zhang et al., 1994). Studies on the effects of SGAs on these hormones have shown that leptin increases in association with weight gain during SGA treatment, but the reported findings have been inconsistent, especially in terms of adiponectin (Calarge, Xie, Fiedorowicz, Burns, & Haynes, 2012; Jin et al., 2008).
Although there are many studies on the metabolic effects of SGAs in adult and adolescent patients, to the best of our knowledge only a few have investigated metabolic changes associated with SGA in prepubertal children (Calarge, Acion, Kuperman, Tansey, & Schlechte, 2009; Calarge et al., 2012). When children aged 3–6 years were prescribed SGA, risperidone was preferred more often than all other antipsychotics (Governale & Mehta, 2009). Therefore, it is crucial that weight gain and metabolic changes related to risperidone are investigated in prepubertal patients because, as mentioned earlier, the risk of metabolic changes is higher in children (Cohen et al., 2012; Correll, 2008; Correll et al., 2009; Hellings et al., 2001; Safer, 2004; Sikich et al., 2004) and because of the disparity between prepubertal, pubertal, and adult physiology. Additionally, as shown in adults, the risk of metabolic side effects of SGA increases as the age at which such treatment begins decreases (Bai et al., 2009). As such, this study aimed to investigate changes in weight, metabolic parameters (glucose and lipid profiles), and adipocytokines (leptin, adiponectin) after 14–16 weeks of open-label prospective risperidone treatment in patients aged ≤10 years. The study hypothesis was that risperidone treatment in prepubertal children would result in significant change in weight and other metabolic parameters similar to those reported in adults and adolescents.
- Top of page
- Materials and Methods
- Clinical and research implications
In this study, risperidone was used for a mean 120.20 ± 17.08 days at doses of 0.25–1 mg/day (0.01–0.07 mg/kg/day) in prepubertal children. Twenty-nine of the 31 patients (93.55%) gained varying degrees of weight (mean: 2.51 ± 1.94 kg) and weight, height, BMI, BMI-p, BMI z-score, triglyceride, VLDL, insulin, and leptin levels were significantly higher after treatment. There was no relationship between baseline BMI-p and weight gain; although overweight prepubertal children may have a high risk of developing IR soon after commencing risperidone treatment. The present study illustrates the emergence of metabolic abnormalities even in prepubertal children treated with low doses of risperidone in the absence of the confounding effect of puberty, an important developmental stage due to changes in body fat, blood pressure, lipids, and insulin sensitivity. Also to the best of our knowledge the present study is the first to investigate leptin and adiponectin levels in prepubertal patients treated with risperidone. Although metabolic parameters – especially weight gain – are thought to be mediated by many factors, such as age, gender, initial BMI, and parents' BMI (Baptista et al., 2002; Bobes et al., 2003; Hellings et al., 2001; Ratzoni et al., 2002), as described below the changes observed in this study are mostly congruent with the literature on postpubertal youth.
In the Treatment of Early-Onset Schizophrenia Spectrum Disorders Study, the efficacy and safety of SGAs and molindone were compared, and it was found that weight gain was significant in the SGAs group—the mean change in BMI-p in the patients treated with risperidone was in the 6.8 percentile (Sikich et al., 2008). In a recently published meta-analysis, mean weight gain in 25 short-term (3–12 weeks) studies that included 1040 patients treated with risperidone was 2.02 ± 0.32 kg (Cohen et al., 2012). Nicolson, Awad and Sloman (1998) reported a weight gain of 3.5 kg after 12 weeks of risperidone treatment, whereas Aman, De Smedt, Derivan, Lyons and Findling (2002) observed a weight gain of 2.2 kg after 6 weeks and Turgay, Binder, Snyder and Fisman (2002) noted 8.5 kg of weight gain after 1 year. Weight gain during risperidone treatment was reported as 7.2 ± 5.3 kg during 45 weeks of treatment (Fleischhaker et al., 2008). In this study, there was no correlation between risperidone dosage and a change in BMI; however, it is important to note that many metabolic changes were seen even at this low dose.
In the present study significant increases were observed in the BMI z-scores and BMI-p. A prospective (8 week) double-blind study compared risperidone, olanzapine, and haloperidol in 8- to 19-year-olds with psychosis (Sikich et al., 2004), and reported that a BMI change exceeding the developmentally expected change in each patient and the increase in BMI during SGA treatment was significant. Although BMI-p and weight gain were not correlated, overweight children had the highest degree of weight gain in this study, which is in accordance with findings in adults (Baptista et al., 2002), whereas the results should be interpreted with caution due to the small number of patients in each BMI-p category.
Metabolic changes (especially HOMA-IR) observed in the present study tended to be higher in overweight children, which is similar to the findings reported by Calarge, Acion et al. (2009). They reported that increases in weight and BMI were significant during a mean 2.9 years of risperidone treatment, and that hypertriglyceridemia, hyperinsulinemia, and a high HOMA-IR were observed, and were especially higher in overweight patients. We think that the risk of weight gain and IR during risperidone treatment might be mediated by a baseline BMI, because the weight gain and metabolic effects associated with SGAs are mediated by environmental and individual factors (Baptista et al., 2002).
In the present study, there was a significant increase in insulin and HOMA-IR. Another study reported that insulin increased by 35% and HOMA-IR increased by 27% in patients treated with risperidone within 2.9 years (Calarge, Ellingrod et al., 2009). It was reported that an elevated insulin level might be the result of compensation to avoid an increase in glucose (Eyzaguirre & Mericq, 2009). The underlying mechanism of the development of IR during SGA treatment remains unknown (Anil & Meltzer, 2001; Yurtsever, Esen-Danaci, & Deveci, 2007). In this study an increase in insulin was weakly associated with an increase in leptin and was not associated with weight gain, although it has been reported that insulin increases with weight gain; however, the relevant published findings are inconsistent (He, Votruba, Venti, & Krakoff, 2011; Potretzke, Warren, Schmitz, Gottsacker, & Jensen, 2011). One study reported a significant weight change of 4.16 kg after 8 weeks of risperidone treatment, in the absence of significant changes in leptin, glucose, and fasting insulin levels (Maayan & Vakhrusheva, 2010). Other effects (significant increases in TG and VLDL) of risperidone were also noted in this study. Correll et al. (2009) reported a mean weight gain of 5.3 kg and an increase in TG in 135 children and adolescents treated with risperidone for 12 weeks.
The relationship between adipocytokines and SGAs has also been reported in the literature (Bai et al., 2009; Brömel et al., 1998; Maayan & Vakhrusheva, 2010). Leptin was reported to be associated with weight gain during risperidone treatment (Calarge, Acion et al., 2009, 2012; Yurtsever et al., 2007). It is difficult to improve our understanding of the relationship between adipocytokines and weight gain during risperidone treatment because of the many confounding factors associated with appetite and weight change observed in healthy individuals. Despite the potential association between a leptin increase and weight gain, a weak correlation between leptin, weight gain and an increase in BMI was observed in this study. As such, the observed weight gain might not be the only cause of the observed increase in leptin. Similarly, Martin et al. (2004) reported that leptin and weight gain did not predict each other. In addition, it should be noted that we did not measure subcutaneous adipose tissue, which has a stronger association with leptin than visceral adipose tissue. Adiponectin is also associated with obesity, and its level is negatively correlated with weight and insulin level (Weyer et al., 2001). In this study, a nonsignificant decrease in adiponectin was noted. A recent study that compared the effects of risperidone and olanzapine on adiponectin for 3 months reported that there was a direct antipsychotic effect on adipose tissue (Wampers et al., 2012); however, Jin et al.'s (2008) meta-analysis reported that adiponectin did not differ between patients and normal controls, and did not change post-SGA treatment, in most studies on the effects of SGAs on leptin and adiponectin.
The present study has some limitations that should be mentioned. Firstly, the small sample limits the generalization of the findings and the statistical power to test the possible relationships between parameters. Also we could not assess the impact of gender. The short duration of follow-up is another limitation; although many changes in weight and metabolic parameters were observed, we could not forecast the progression. Additional limitations were an open-label study design, the lack of a comparison group, and lack of measurements of diet and other lifestyle factors. In order to obtain more generalizable results we believe it is necessary to conduct studies with larger samples, control groups, and longer follow-up periods.