Psychopharmacology DOI 10.1007/s00213-014-3601-7
ORIGINAL INVESTIGATION
Aminotransferase levels as a prospective predictor for the development of metabolic syndrome in patients with schizophrenia Eun Young Kim & Se Hyun Kim & Nam Young Lee & Jin Sang Yoon & Chul Eung Kim & Yong Sik Kim & Yong Min Ahn
Received: 7 January 2014 / Accepted: 13 April 2014 # Springer-Verlag Berlin Heidelberg 2014
Abstract Rationale Increased levels of alanine aminotransferase (ALT) are a biomarker for metabolic syndrome (MetS), but this relationship remains unproven in patients with schizophrenia. Objective We assessed the relationship between aminotransferase levels and MetS in patients with schizophrenia. Method This pooled analysis from two open-label prospective studies included 342 patients with schizophrenia who did not meet criteria for MetS at baseline. The development of MetS was assessed at weeks 12 and 24. Results MetS developed in 19.1 % of patients during the 24week follow-up period. ALT levels were significantly associated with incident MetS: for each sex-specific standard E. Y. Kim : Y. M. Ahn (*) Department of Psychiatry, Seoul National University Hospital, Seoul, Republic of Korea e-mail: [email protected] S. H. Kim Medical Research Institute, Seoul National University, Seoul, Republic of Korea N. Y. Lee : Y. S. Kim Department of Neuropsychiatry, Dongguk University Medical School, Dongguk University International Hospital, Goyang, Republic of Korea J. S. Yoon Department of Psychiatry, Chonnam National University Medical School, Gwangju, Republic of Korea C. E. Kim Department of Psychiatry, Inha University Hospital, Incheon, Republic of Korea Y. M. Ahn Institute of Human Behavioral Medicine, Seoul National University College of Medicine, 28 Yeongeon-dong, Jongno-gu, Seoul 110-744, Republic of Korea
deviation increase in log ALT, the odds ratio (OR) of MetS was 1.357 (p=.006) after adjusting for age, sex, duration of illness, smoking, and previous use of antipsychotics. This result remained significant after adjusting for interim weight changes. Compared with patients in the lowest quartile, the OR of MetS in those in the highest quartile within the normal range of ALT levels was 4.276 (p=.024). However, this association was significant only in male patients. Using a cutoff value of 23.0 U/L, sensitivity and specificity were 70.6 and 68.3 %, respectively, in male patients whose ALT levels were in the normal range. Conclusions A prospective association between ALT levels and MetS highlights the value of ALT levels, even mild ALT elevations within the normal range, as a predictor of the MetS risk in male patients. Baseline liver function tests and monitoring should be obtained during antipsychotic treatment to identify the risk for MetS. Keywords Schizophrenia . Metabolic syndrome . Aminotransferase levels . Atypical antipsychotics
Introduction Recent studies have indicated that schizophrenia is associated with an increased risk for metabolic syndrome (MetS), which leads to higher somatic comorbidities and premature mortality, particularly from cardiovascular disorders (CVDs) (Capasso et al. 2008; Fleischhacker et al. 2008; McEvoy et al. 2005; Meyer and Stahl 2009; Saha et al. 2007). It has been widely accepted that MetS, a multidimensional risk factor for both CVD and type 2 diabetes (Gami et al. 2007; Lakka et al. 2002), is based on central obesity and insulin resistance (Reaven 1988). A systematic review reported that 32.5 % of patients with schizophrenia have MetS (Mitchell
Psychopharmacology
et al. 2013b). In this regard, routine baseline screening and monitoring for MetS are crucial for reducing excess mortality but are insufficient for patients taking antipsychotic medications (Mitchell et al. 2012). MetS can result not only from an unhealthy lifestyle such as heavy smoking (Bobes et al. 2010), physical inactivity (Vancampfort et al. 2011), or poor diet (Strassnig et al. 2003) but also from the adverse effects of antipsychotic medications, particularly atypical antipsychotics (De Hert et al. 2012). Furthermore, it has been suggested that those with schizophrenia are genetically predisposed to develop MetS (van Winkel et al. 2010). Preliminary evidence indicates an increased prevalence of central obesity and glucose abnormalities, such as impaired fasting glucose levels and insulin resistance, in drug-naive patients with schizophrenia, suggesting that metabolic disturbances may begin early, prior to starting medication (Cohen and De Hert 2011). Given the complex etiology of MetS, which includes environmental factors, antipsychotic medication, and genetic predisposition, research regarding the biological markers for MetS in patients with schizophrenia is rare. Although several clinical variables, such as age, duration of illness, and MetS components, have been suggested as risk factors (Mitchell et al. 2013a), the identification of patients at greater risk for metabolic disorders remains a major challenge in clinical practice. Liver enzymes, particularly alanine aminotransferase (ALT), are among the most promising markers for predicting MetS as they are sensitive biomarkers of non-alcoholic fatty liver disease (NAFLD) (Ekstedt et al. 2006). NAFLD is linked with both hepatic and systemic insulin resistance and is often referred to as the hepatic manifestation of MetS (Marchesini et al. 2001). Studies of the general population have associated ALT levels with the development of incident type 2 diabetes, MetS, CVDs, and all-cause mortality (Goessling et al. 2008; Hanley et al. 2005; Schindhelm et al. 2007). Additionally, even higher levels of within-normal-reference-limits of ALT have been associated with the development of MetS and diabetes mellitus in healthy individuals (Goessling et al. 2008; Steinvil et al. 2010). One cross-sectional study of a psychiatric population demonstrated that serum ALT levels are significantly related to MetS in patients with schizophrenia taking clozapine (Lee et al. 2013), suggesting the possibility of serum ALT level as an indicator for MetS. However, the relationship between ALT levels and MetS is complicated by several factors in patients with schizophrenia. Drug-naïve patients with schizophrenia are more likely to show hepatic insulin resistance regardless of their intra-abdominal fat mass (van Nimwegen et al. 2008). Atypical antipsychotics are associated with metabolic dysregulation, and this appears to be mediated primarily via a peripheral mechanism (insulin resistance) and secondarily by a central one (Houseknecht et al. 2007).
This study aimed to use a prospective follow-up design to (1) assess the relationship between aminotransferase levels and the incidence of MetS in patients with schizophrenia who were treated with paliperidone extended-release (ER) controlling for potential confounding factors and (2) determine whether these relationships also occur in patients with aminotransferase values within the normal range.
Methods This analysis used data pooled from two separated 24-week, multicenter, open-label studies that assessed the efficacy and tolerability of paliperidone ER in patients with schizophrenia who were switched from other medications (Kim et al. 2012; Na et al. 2013). The study was conducted in accordance with the guidelines of the International Conference on Harmonization for Good Clinical Practice as stipulated in the Declaration of Helsinki. Individual institutional review boards at each study site approved the final study protocols. All participants provided written informed consent according to local requirements. Subjects We included only patients without MetS at baseline in our analyses. The sample included patients with schizophrenia (DSM-IV criteria) (age, 18–65 years) who had received treatment with any oral antipsychotic agent for at least 2 weeks before screening and required a switch in antipsychotic medication due to lack of efficacy, tolerance, or non-compliance. Patients with a history of neuroleptic malignant syndrome or treatment with a long-acting injectable antipsychotic or clozapine in the 4 weeks before the screening visit, current substance abuse or dependence, or a history of substance abuse or dependence within the past 6 months were excluded. Patients with hypersensitivity to risperidone or paliperidone, any significant medical or neurological condition, a risk for suicide or aggressive behavior, a severe pre-existing gastrointestinal narrowing or inability to swallow the medication whole, pregnancy or breastfeeding, and/or participation in any investigational drug trial within 1 month prior to the screening visit were also excluded. Of the 685 participants who participated in the two studies, 172 were excluded because of major protocol violations or missing data related to liver enzymes, the diagnosis of MetS components, and/or covariates such as duration of illness at baseline. An additional 171 (33 %, 171/513) participants were excluded because they were diagnosed with MetS at baseline. The prevalence of MetS was comparable to that among patients with schizophrenia in Taiwan (Huang et al. 2009). The final sample size was 342.
Psychopharmacology
Dosing of the study medication Pre-study antipsychotic agents were discontinued or gradually tapered off during the first 4 weeks of the study. The recommended dose of paliperidone ER is 6 mg/day after the morning meal or in a fasting state. However, the paliperidone ER dose could be adjusted within the allowed 3–12 mg/day during the 24-week study period. Mean (SD) doses of the study medication were 5.5 (1.5) mg/day and 5.4 (1.6) mg/day at study entry, 8.4 (2.9) mg/day and 7.8 (2.8) mg/day at 12 weeks, and 8.4 (2.8) mg/day and 7.8 (3.0) mg/day at 24 weeks in male and female subjects, respectively. Antipsychotics other than paliperidone ER were not allowed. Mood stabilizers or antidepressants could not be added during the study period. However, if a stable dose of antidepressants and mood stabilizers had been used for more than 1 month at the time of enrollment, patients were allowed to continue with these medications during the study period. Concurrent medications, including anticholinergics, propranolol, and hypnotics including benzodiazepines and zolpidem, were permitted for treating extrapyramidal symptoms (EPS) and sleep disturbances. There were no differences in the use of concomitant medications between male and female patients during the study period (Table 1). Assessment of MetS The presence of MetS was assessed at baseline and at weeks 12 and 24 according to the guidelines of the American Heart Association and the National Heart, Lung, and Blood Institute’s adaptation of the National Cholesterol Education Program Adult Treatment Panel III, with modified waist circumference (WC) criteria provided by the Korean Society for the Study of Obesity. MetS was defined by any three or more of the following criteria: (1) fasting plasma glucose ≥5.6 mmol/L (100 mg/dL) or drug treatment for elevated glucose, (2) serum triglycerides ≥1.7 mmol/L (150 mg/dL) or drug treatment for hypercholesterolemia, (3) high-density lipoprotein (HDL)-cholesterol 85 cm for women. Statistical analysis Baseline characteristics of the participants are summarized and displayed as means (standard deviations [SDs]) for continuous variables and as number of patients (expressed as a percentage) for categorical variables. Differences in demographic and clinical characteristics between men and women were determined using χ2 and t tests. The primary outcome variable was incident MetS. Aspartate aminotransferase (AST) and ALT values were natural
log-transformed to normalize their skewed distributions. All analyses were performed at 24 weeks of follow-up. As an additional analysis, AST and ALT were limited to the normal range (up to 40 U/L each). We used the generalized estimating equation for all samples (n=342) and logistic regression analysis for the per-protocol samples (n=199) to model the change in the odds of the event per one sex-specific SD increase in log AST or log ALT; this was calculated by multiplying the regression coefficient by the SD (OR=e coefficient*SD). Covariates included age, sex, smoking status, illness duration, and pre-study antipsychotic agents. Exponentiated coefficients and corresponding confidence intervals (CI) are presented for each model. Secondary analyses included additional adjustment for interim weight changes from baseline to week 24. We calculated the ORs for having MetS relative to the lowest quartile using logistic regression analysis to better appreciate the magnitude of the differences in the prevalence of MetS between sex-specific quartiles of ALT levels within the normal range. Finally, we conducted receiver-operating-characteristic (ROC) curve analysis of the ALT levels within the normal range, estimating the area under the curve (AUC) with 95 % confidence intervals, to quantify the accuracy of predictions about whether subjects would develop MetS at week 24. The optimal cutoff value was determined using a Youden index value (one minus specificity subtracted from sensitivity) (Fluss et al. 2005). Additionally, 95 % confidence limits were computed using an exact binominal test. A two-tailed p value
ORIGINAL INVESTIGATION
Aminotransferase levels as a prospective predictor for the development of metabolic syndrome in patients with schizophrenia Eun Young Kim & Se Hyun Kim & Nam Young Lee & Jin Sang Yoon & Chul Eung Kim & Yong Sik Kim & Yong Min Ahn
Received: 7 January 2014 / Accepted: 13 April 2014 # Springer-Verlag Berlin Heidelberg 2014
Abstract Rationale Increased levels of alanine aminotransferase (ALT) are a biomarker for metabolic syndrome (MetS), but this relationship remains unproven in patients with schizophrenia. Objective We assessed the relationship between aminotransferase levels and MetS in patients with schizophrenia. Method This pooled analysis from two open-label prospective studies included 342 patients with schizophrenia who did not meet criteria for MetS at baseline. The development of MetS was assessed at weeks 12 and 24. Results MetS developed in 19.1 % of patients during the 24week follow-up period. ALT levels were significantly associated with incident MetS: for each sex-specific standard E. Y. Kim : Y. M. Ahn (*) Department of Psychiatry, Seoul National University Hospital, Seoul, Republic of Korea e-mail: [email protected] S. H. Kim Medical Research Institute, Seoul National University, Seoul, Republic of Korea N. Y. Lee : Y. S. Kim Department of Neuropsychiatry, Dongguk University Medical School, Dongguk University International Hospital, Goyang, Republic of Korea J. S. Yoon Department of Psychiatry, Chonnam National University Medical School, Gwangju, Republic of Korea C. E. Kim Department of Psychiatry, Inha University Hospital, Incheon, Republic of Korea Y. M. Ahn Institute of Human Behavioral Medicine, Seoul National University College of Medicine, 28 Yeongeon-dong, Jongno-gu, Seoul 110-744, Republic of Korea
deviation increase in log ALT, the odds ratio (OR) of MetS was 1.357 (p=.006) after adjusting for age, sex, duration of illness, smoking, and previous use of antipsychotics. This result remained significant after adjusting for interim weight changes. Compared with patients in the lowest quartile, the OR of MetS in those in the highest quartile within the normal range of ALT levels was 4.276 (p=.024). However, this association was significant only in male patients. Using a cutoff value of 23.0 U/L, sensitivity and specificity were 70.6 and 68.3 %, respectively, in male patients whose ALT levels were in the normal range. Conclusions A prospective association between ALT levels and MetS highlights the value of ALT levels, even mild ALT elevations within the normal range, as a predictor of the MetS risk in male patients. Baseline liver function tests and monitoring should be obtained during antipsychotic treatment to identify the risk for MetS. Keywords Schizophrenia . Metabolic syndrome . Aminotransferase levels . Atypical antipsychotics
Introduction Recent studies have indicated that schizophrenia is associated with an increased risk for metabolic syndrome (MetS), which leads to higher somatic comorbidities and premature mortality, particularly from cardiovascular disorders (CVDs) (Capasso et al. 2008; Fleischhacker et al. 2008; McEvoy et al. 2005; Meyer and Stahl 2009; Saha et al. 2007). It has been widely accepted that MetS, a multidimensional risk factor for both CVD and type 2 diabetes (Gami et al. 2007; Lakka et al. 2002), is based on central obesity and insulin resistance (Reaven 1988). A systematic review reported that 32.5 % of patients with schizophrenia have MetS (Mitchell
Psychopharmacology
et al. 2013b). In this regard, routine baseline screening and monitoring for MetS are crucial for reducing excess mortality but are insufficient for patients taking antipsychotic medications (Mitchell et al. 2012). MetS can result not only from an unhealthy lifestyle such as heavy smoking (Bobes et al. 2010), physical inactivity (Vancampfort et al. 2011), or poor diet (Strassnig et al. 2003) but also from the adverse effects of antipsychotic medications, particularly atypical antipsychotics (De Hert et al. 2012). Furthermore, it has been suggested that those with schizophrenia are genetically predisposed to develop MetS (van Winkel et al. 2010). Preliminary evidence indicates an increased prevalence of central obesity and glucose abnormalities, such as impaired fasting glucose levels and insulin resistance, in drug-naive patients with schizophrenia, suggesting that metabolic disturbances may begin early, prior to starting medication (Cohen and De Hert 2011). Given the complex etiology of MetS, which includes environmental factors, antipsychotic medication, and genetic predisposition, research regarding the biological markers for MetS in patients with schizophrenia is rare. Although several clinical variables, such as age, duration of illness, and MetS components, have been suggested as risk factors (Mitchell et al. 2013a), the identification of patients at greater risk for metabolic disorders remains a major challenge in clinical practice. Liver enzymes, particularly alanine aminotransferase (ALT), are among the most promising markers for predicting MetS as they are sensitive biomarkers of non-alcoholic fatty liver disease (NAFLD) (Ekstedt et al. 2006). NAFLD is linked with both hepatic and systemic insulin resistance and is often referred to as the hepatic manifestation of MetS (Marchesini et al. 2001). Studies of the general population have associated ALT levels with the development of incident type 2 diabetes, MetS, CVDs, and all-cause mortality (Goessling et al. 2008; Hanley et al. 2005; Schindhelm et al. 2007). Additionally, even higher levels of within-normal-reference-limits of ALT have been associated with the development of MetS and diabetes mellitus in healthy individuals (Goessling et al. 2008; Steinvil et al. 2010). One cross-sectional study of a psychiatric population demonstrated that serum ALT levels are significantly related to MetS in patients with schizophrenia taking clozapine (Lee et al. 2013), suggesting the possibility of serum ALT level as an indicator for MetS. However, the relationship between ALT levels and MetS is complicated by several factors in patients with schizophrenia. Drug-naïve patients with schizophrenia are more likely to show hepatic insulin resistance regardless of their intra-abdominal fat mass (van Nimwegen et al. 2008). Atypical antipsychotics are associated with metabolic dysregulation, and this appears to be mediated primarily via a peripheral mechanism (insulin resistance) and secondarily by a central one (Houseknecht et al. 2007).
This study aimed to use a prospective follow-up design to (1) assess the relationship between aminotransferase levels and the incidence of MetS in patients with schizophrenia who were treated with paliperidone extended-release (ER) controlling for potential confounding factors and (2) determine whether these relationships also occur in patients with aminotransferase values within the normal range.
Methods This analysis used data pooled from two separated 24-week, multicenter, open-label studies that assessed the efficacy and tolerability of paliperidone ER in patients with schizophrenia who were switched from other medications (Kim et al. 2012; Na et al. 2013). The study was conducted in accordance with the guidelines of the International Conference on Harmonization for Good Clinical Practice as stipulated in the Declaration of Helsinki. Individual institutional review boards at each study site approved the final study protocols. All participants provided written informed consent according to local requirements. Subjects We included only patients without MetS at baseline in our analyses. The sample included patients with schizophrenia (DSM-IV criteria) (age, 18–65 years) who had received treatment with any oral antipsychotic agent for at least 2 weeks before screening and required a switch in antipsychotic medication due to lack of efficacy, tolerance, or non-compliance. Patients with a history of neuroleptic malignant syndrome or treatment with a long-acting injectable antipsychotic or clozapine in the 4 weeks before the screening visit, current substance abuse or dependence, or a history of substance abuse or dependence within the past 6 months were excluded. Patients with hypersensitivity to risperidone or paliperidone, any significant medical or neurological condition, a risk for suicide or aggressive behavior, a severe pre-existing gastrointestinal narrowing or inability to swallow the medication whole, pregnancy or breastfeeding, and/or participation in any investigational drug trial within 1 month prior to the screening visit were also excluded. Of the 685 participants who participated in the two studies, 172 were excluded because of major protocol violations or missing data related to liver enzymes, the diagnosis of MetS components, and/or covariates such as duration of illness at baseline. An additional 171 (33 %, 171/513) participants were excluded because they were diagnosed with MetS at baseline. The prevalence of MetS was comparable to that among patients with schizophrenia in Taiwan (Huang et al. 2009). The final sample size was 342.
Psychopharmacology
Dosing of the study medication Pre-study antipsychotic agents were discontinued or gradually tapered off during the first 4 weeks of the study. The recommended dose of paliperidone ER is 6 mg/day after the morning meal or in a fasting state. However, the paliperidone ER dose could be adjusted within the allowed 3–12 mg/day during the 24-week study period. Mean (SD) doses of the study medication were 5.5 (1.5) mg/day and 5.4 (1.6) mg/day at study entry, 8.4 (2.9) mg/day and 7.8 (2.8) mg/day at 12 weeks, and 8.4 (2.8) mg/day and 7.8 (3.0) mg/day at 24 weeks in male and female subjects, respectively. Antipsychotics other than paliperidone ER were not allowed. Mood stabilizers or antidepressants could not be added during the study period. However, if a stable dose of antidepressants and mood stabilizers had been used for more than 1 month at the time of enrollment, patients were allowed to continue with these medications during the study period. Concurrent medications, including anticholinergics, propranolol, and hypnotics including benzodiazepines and zolpidem, were permitted for treating extrapyramidal symptoms (EPS) and sleep disturbances. There were no differences in the use of concomitant medications between male and female patients during the study period (Table 1). Assessment of MetS The presence of MetS was assessed at baseline and at weeks 12 and 24 according to the guidelines of the American Heart Association and the National Heart, Lung, and Blood Institute’s adaptation of the National Cholesterol Education Program Adult Treatment Panel III, with modified waist circumference (WC) criteria provided by the Korean Society for the Study of Obesity. MetS was defined by any three or more of the following criteria: (1) fasting plasma glucose ≥5.6 mmol/L (100 mg/dL) or drug treatment for elevated glucose, (2) serum triglycerides ≥1.7 mmol/L (150 mg/dL) or drug treatment for hypercholesterolemia, (3) high-density lipoprotein (HDL)-cholesterol 85 cm for women. Statistical analysis Baseline characteristics of the participants are summarized and displayed as means (standard deviations [SDs]) for continuous variables and as number of patients (expressed as a percentage) for categorical variables. Differences in demographic and clinical characteristics between men and women were determined using χ2 and t tests. The primary outcome variable was incident MetS. Aspartate aminotransferase (AST) and ALT values were natural
log-transformed to normalize their skewed distributions. All analyses were performed at 24 weeks of follow-up. As an additional analysis, AST and ALT were limited to the normal range (up to 40 U/L each). We used the generalized estimating equation for all samples (n=342) and logistic regression analysis for the per-protocol samples (n=199) to model the change in the odds of the event per one sex-specific SD increase in log AST or log ALT; this was calculated by multiplying the regression coefficient by the SD (OR=e coefficient*SD). Covariates included age, sex, smoking status, illness duration, and pre-study antipsychotic agents. Exponentiated coefficients and corresponding confidence intervals (CI) are presented for each model. Secondary analyses included additional adjustment for interim weight changes from baseline to week 24. We calculated the ORs for having MetS relative to the lowest quartile using logistic regression analysis to better appreciate the magnitude of the differences in the prevalence of MetS between sex-specific quartiles of ALT levels within the normal range. Finally, we conducted receiver-operating-characteristic (ROC) curve analysis of the ALT levels within the normal range, estimating the area under the curve (AUC) with 95 % confidence intervals, to quantify the accuracy of predictions about whether subjects would develop MetS at week 24. The optimal cutoff value was determined using a Youden index value (one minus specificity subtracted from sensitivity) (Fluss et al. 2005). Additionally, 95 % confidence limits were computed using an exact binominal test. A two-tailed p value
