- Email: [email protected]
Insulin secretion in patients receiving clozapine, olanzapine, quetiapine and risperidone
Schizophrenia Research 143 (2013) 358–362
Contents lists available at SciVerse ScienceDirect
Schizophrenia Research journal homepage: www.elsevier.com/locate/schres
Insulin secretion in patients receiving clozapine, olanzapine, quetiapine and risperidone Peter Manu a, b, c,⁎, 1, Christoph U. Correll a, b, d, 1, Martien Wampers e, Ruud van Winkel e, f, Weiping Yu e, Daphna Shiffeldrim a, John M. Kane a, b, Marc De Hert e a
Zucker Hillside Hospital, Glen Oaks, NY, United States Hofstra North Shore-LIJ School of Medicine, Hempstead, NY, United States Transilvania University, Brasov, Romania d Feinstein Institute for Medical Research, Manhasset, NY, United States e University Psychiatric Center, Kortenberg, Belgium f School of Mental Health and Neuroscience (EURON), University Medical Center, Maastricht, The Netherlands b c
a r t i c l e
i n f o
Article history: Received 12 July 2012 Received in revised form 12 November 2012 Accepted 14 November 2012 Available online 9 December 2012 Keywords: Antipsychotics Clozapine Prediabetes Insulin Oral glucose tolerance test
a b s t r a c t Background: Second-generation antipsychotics (SGAs) increase the risk of type 2 diabetes. The mechanism is thought to center on drug-induced weight gain, which starts the dysmetabolic cascade of insulin resistance, increased insulin production and pancreatic beta-cell failure. An independent effect of SGAs on insulin secretion has been suggested in animal models, but has not been demonstrated in clinical samples. Objective: To determine the post-challenge insulin secretion in patients treated with SGAs. Method: We identified 520 non-diabetic individuals treated with clozapine (N=73), olanzapine (N=190), quetiapine (N=91) or risperidone (N=166) in a consecutive, single-site cohort of 783 adult psychiatric inpatients who underwent a comprehensive metabolic assessment. Insulin secretion was measured as the area under the curve (AUCinsulin) generated by levels recorded at baseline, 30, 60 and 120 min after the intake of 75 g of glucose. The independent predictors of insulin secretion were determined with regression analysis in the entire sample and separately in patients with normal glucose tolerance (NGT) and prediabetes. Results: The post-challenge AUCinsulin was independently predicted by AUCglucose, waist circumference, triglyceride levels and younger age (pb 0.0001); non-smoking status (p=0.0012); and treatment with clozapine (p=0.021). The model explained 33.5% of the variance in insulin secretion (pb 0.0001). The clozapine effect was present in the NGT group, but not in prediabetics. Conclusions: Clozapine, but not olanzapine, quetiapine and risperidone, is an independent predictor of postchallenge insulin secretion in non-diabetics, particularly in those with normal glucose tolerance. The findings suggest that the diabetogenic risk of clozapine may persist even after weight reduction. © 2012 Elsevier B.V. All rights reserved.
1. Introduction The carbohydrate metabolism is frequently abnormal in patients treated with second-generation antipsychotic drugs (Newcomer, 2005; De Hert et al., 2012; Correll et al., 2011). These abnormalities include increased fasting glycemia, impaired glucose tolerance, prediabetes and diabetes and have been attributed to the insulin-resistant state that follows the weight gain observed in a large proportion of patients receiving long-term treatment with these drugs (De Hert et al., 2012; Manu et al., 2012). A weight-independent diabetogenic
⁎ Corresponding author at: Medical Services, Zucker Hillside Hospital, Glen Oaks, NY 11004, United States. E-mail address: [email protected] (P. Manu). 1 Drs. Manu and Correll have contributed equally to this article. 0920-9964/$ – see front matter © 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.schres.2012.11.016
effect of these drugs has been observed in animal models (Smith et al., 2008a,b), limited case series (Kohen et al., 2008) and one cohort study (Bai et al., 2010). Previous work has established the presence of fasting hyperinsulinemia, a biomarker for insulin resistance, in subjects receiving second-generation antipsychotics who have drug-induced weight gain, metabolic syndrome and/or prediabetes (Manu et al., 2012). However, the effect of second-generation antipsychotics on the insulin secretion after an oral glucose challenge has not been thoroughly evaluated by other investigators. The issue is important, because fasting insulin levels reflect only the basal pancreatic beta-cell activity, but not the meal-related insulin secretion. In fact, there is considerable evidence that concomitant measurements of the plasma glucose and insulin after an oral glucose challenge allow for the most accurate evaluation of the insulin secretory function, at least when compared with the Homeostatic Model Assessment, which relies on fasting insulin and glucose
P. Manu et al. / Schizophrenia Research 143 (2013) 358–362
concentrations, or compared with the insulin level measured shortly after an intravenous glucose challenge (Reaven, 2009). We assessed the insulin secretion after a standard oral glucose challenge in patients chronically treated with clozapine, olanzapine, risperidone and quetiapine. These drugs are prescribed world-wide and have demonstrated effectiveness for the treatment of bipolar mania, schizophrenia spectrum disorders and other psychotic syndromes. Two of these drugs, clozapine and olanzapine, are known to be the most obesogenic of the class, while the weight accrual and dysmetabolic liability of risperidone and quetiapine are usually described as moderate (Newcomer, 2005; De Hert et al. 2012). We hypothesized a) that insulin secretion after glucose challenge will confirm the difference between clozapine and olanzapine, on the one hand, and risperidone and quetiapine, on the other, and b) that some or all of these drugs will be an independent correlate of insulin secretion.
359
blood glucose ≥ 100 mg/dL; serum triglycerides ≥ 150 mg/dL; HDLcholesterol b 40 mg/dL in men and b50 mg/dL in women; and arterial blood pressure ≥ 130/85 mm Hg or current treatment with antihypertensive agents (Grundy et al., 2004). The treating psychiatrists assessed the severity of symptoms and rated them using the Global Assessment of Function (GAF) from 0 (worst) to 100 (best) (Aas, 2011) and the Clinical Global Impressions Severity (CGI-S) Scale from 1 (normal) to 7 (extremely ill) (Rabinowitz et al., 2006). 2.3. Selection of subjects
From November 2003 through July 2007, 798 non-diabetic patients receiving antipsychotic drugs consecutively admitted to the University Psychiatric Center, Kortenberg, Belgium, were asked by their treating psychiatrist to agree to a standardized battery of tests to identify metabolic syndrome and insulin resistance. A great majority of the patients (95.1%) were European-born Whites; others were Black Africans (2.4%), Maghreb-born individuals (1.5%) and Asians (1%). Psychiatric diagnoses were established according to DSM-IV by experienced psychiatrists affiliated with the University Psychiatric Center and responsible for the patient's treatment. All subjects gave written informed consent and the study was approved by the University Psychiatric Center's Ethics Committee.
Of the original 798 patients, we excluded 15 who declined to consent to metabolic screening and 80 with evidence of diabetes mellitus, as indicated by a fasting plasma glucose greater than 125 mg/dL or 2-hour glucose level during OGTT of at least 200 mg/dL or hemoglobin A1c 6.5% or greater. From the remaining cohort of 703 patients, we excluded 183 patients who either were on antipsychotic polypharmacy (N= 87), or who were treated with aripiprazole, amisulpride or various first-generation antipsychotics, but with sample sizes of well below 10% of the cohort. As the final sample, we selected 520 patients who were treated with antipsychotics used as monotherapy and who comprised at least 10% of the sample, i.e., olanzapine (N= 190), risperidone (N= 160), quetiapine (N= 91) and clozapine (N= 73). The proportions of patients treated with the same antipsychotic drug for more than 3 months were 76.0% for clozapine, 46.2% for risperidone, 46.2% for olanzapine, and 38.7% for quetiapine. The study group (N= 520) was divided into subgroups with normal glucose tolerance (fasting plasma glucose less than 100 mg/dL, 2-hour glucose level during OGTT less than 140 mg/dL and hemoglobin A1c 5.6% or less) and with prediabetes (fasting plasma glucose 100–125 mg/dL or 2-hour glucose level during OGTT in the 140–199 mg/dL range or hemoglobin A1c in the 5.7–6.4% range).
2.2. Clinical and laboratory measurements
2.4. Statistical analyses
The procedures included measurements of height, weight, body mass index (BMI), waist circumference, arterial blood pressure, fasting blood glucose, insulin and blood lipids, hemoglobin A1c, and a 2-hour oral glucose tolerance test (OGTT) after the ingestion of 75 g of glucose. Glucose and insulin levels were assessed 30, 60 and 120 min after the glucose load. As described in a previous publication, all tests were performed in the same laboratory and using the same methods throughout the entire study period (Manu et al., 2012). The fasting glucose, 2-hour postprandial glucose during OGTT and hemoglobin A1c data were used to define normal glucose tolerance (fasting glucose less than 100 mg/dL, 2-hour postprandial glucose less than 140 mg/dL and A1c less than 5.7%), prediabetes (fasting glucose 100–125 mg/dL or 2-hour postprandial glucose 140–199 mg/dL or A1c 5.7%–6.4%), and diabetes mellitus (fasting glucose greater than 125 mg/dL or 2-hour postprandial glucose greater than 199 mg/dL or A1c greater than 6.4%). Glucose and insulin measurements at 0, 30, 60 and 120 min after a 75 gram oral glucose load were used to calculate the area under the curve (total AUC) for insulin (AUCinsulin) and glucose (AUCglucose). The data were also used to calculate the Insulin Disposition Index, a reliable indicator of beta-cell function adjusted for insulin sensitivity (Utzschneider et al., 2009). The fasting glucose and insulin data were used for the Homeostatic Model Assessment of Insulin Resistance (HOMA-IR) (Matthews et al., 1985). The weight and height were used to calculate the body mass index (BMI). The waist circumference, arterial blood pressure, and fasting glucose, triglycerides and high-density lipoprotein (HDL) cholesterol levels were used to determine whether the patient met criteria for metabolic syndrome, which was defined by the presence of at least 3 of the following 5 items: waist circumference > 88 cm in women and >102 cm in men; fasting
Analyses of variance and chi-square tests were used to compare continuous and categorical variables, respectively, in patients with normal glucose tolerance and prediabetes. Logistic regression was used to identify independent predictors of AUCinsulin in the entire study group of nondiabetic patients receiving antipsychotic monotherapy with these four drugs and separately in the subgroups with normal glucose tolerance and with prediabetes. All analyses were two-sided, with alpha of pb 0.05, using JMP 5.0.1., 1989–2003, SAS Institute Inc.
2. Materials and methods 2.1. Setting
3. Results 3.1. Demographic and psychiatric characteristics The sample included 340 patients with NGT and 180 with prediabetes. The NGT group had a lower age and a greater proportion of males (Table 1). The NGT and prediabetic subgroups were similar with respect to the distribution of psychiatric diagnoses, functional status, and the severity of psychiatric illness. Olanzapine, risperidone and quetiapine utilization did not differ between the two subgroups, but prediabetic patients were more likely to be treated with clozapine than patients with NGT (Table 1). The proportion of males was lower among subjects treated with quetiapine. Patients treated with the four SGAs were similar in age and prevalence of cigarette smoking. The body mass index was slightly greater in patients receiving clozapine (Table 2). 3.2. Anthropometric features and fasting metabolic parameters The fasting glucose was significantly greater in the prediabetic group, as expected. The other components of the metabolic syndrome
360
P. Manu et al. / Schizophrenia Research 143 (2013) 358–362
Table 1 Demographic and psychiatric characteristics. Characteristic
Total (N = 520)
NGT (N = 340)
Prediabetes (N = 180)
p-Value
Age, years ± S.D. Male gender, N (%) Smoking Psychiatric diagnosis, N (%) Schizophrenia Schizoaffective disorder Bipolar disorder Major depression Personality disorder GAF, score ± S.D. CGI, score ± S.D. Antipsychotic drug N (%)b Olanzapine Risperidone Quetiapine Clozapine
36.5 ± 11.9 319 (61.4%) 298 (57.3%)
33.8 ± 11.6 220 (64.7%) 119 (57.4%)
41.7 ± 12.3a 99 (55.0%) 103 (57.2%)
b0.0001 0.031 0.98 0.090
348 (66.9%) 59 (11.4%) 82 (15.8%) 10 (1.9%) 21 (4.0%) 55.8 ± 12.8 4.4 ± 0.9
241 (70.9%) 33 (9.7%) 47 (13.8%) 5 (1.5%) 14 (4.1%) 55.4 ± 13.1 4.4 ± 0.9
107 (59.4%) 26 (14.4%) 35 (19.4%) 5 (2.8%) 7 (3.9%) 56.5 ± 12.3 4.3 ± 0.9
190 (36.5%) 166 (31.9%) 91 (17.5%) 73 (14.0%)
125 (36.8%) 118 (34.7%) 62 (18.2%) 35 (10.3%)
65 48 29 38
Table 3 Anthropometric features, arterial blood pressure, fasting metabolic parameters, and prevalence of metabolic syndrome. Characteristic
0.34 0.23 0.006
(36.1%) (26.7%) (16.1%) (21.1%)
Total (N = 520)
NGT (N = 340)
Waist circumference, cm ± S.D. Males 95.1 ± 12.0 94.5 ± 11.6 Females 90.9 ± 14.3 88.9 ± 14.1 Systolic blood pressure, 123.5 ± 15.5 122.9 ± 15.9 mm Hg± S.D. Diastolic blood pressure, 77.4 ± 11.5 76.7 ± 11.6 mm Hg± S.D. Triglycerides, mg/dL 148.4 ± 88.3 145.8 ± 88.3 High-density lipoprotein cholesterol, mg/dL Males 47.1 ± 13.0 46.9 ± 13.3 Females 60.5 ± 16.6 62.2 ± 16.5 Glucose, mg/dL 89.9 ± 8.4 85.8 ± 6.2 Insulin, μg/mL 11.0 ± 7.9 10.5 ± 7.6 HOMA-IR 2.5 ± 1.9 2.3 ± 1.7 Metabolic syndrome, N (%) 141 (27.2%) 59 (17.4%)
Prediabetes (N = 180)
p-Value
96.4 ± 12.9 93.8 ± 14.5 124.6 ± 14.8
0.20 0.018 0.21
78.7 ± 11.2
0.053
153.8 ± 88.4
0.31
47.5 ± 12.2 58.0 ± 16.8 95.4 ± 11.5 11.9 ± 8.6 2.9 ± 2.2 82 (45.8%)
0.70 0.084 b0.0001 0.036 0.0003 b0.0001
NGT = Normal Glucose Tolerance; S.D. = standard deviation; GAF = Global Assessment of Functioning Scale; CGI = Clinical Global Impressions Scale. a p b 0.0001. b p b 0.006.
NGT = Normal Glucose Tolerance; S.D. = standard deviation; HOMA-IR = Homeostatic Model Assessment of Insulin Resistance.
were similar in the two subgroups, with the exception of a larger waist circumference in prediabetic females. The proportion of patients who fulfilled criteria for metabolic syndrome was more than double in the prediabetic subgroup. Fasting insulin levels and resistance to insulin were significantly greater in prediabetic patients (Table 3).
second-generation antipsychotic drugs, i.e., clozapine, olanzapine, quetiapine and risperidone, which are known to lead to weight accrual and to increase the risk of the emergence of metabolic syndrome (Newcomer, 2005; De Hert et al., 2012; Correll et al., 2011) and prediabetes (Manu et al., 2012). The study confirmed that fasting insulin levels and degree of insulin resistance are greater in prediabetic patients. The post-challenge insulin secretion correlated with glycemia, abdominal adiposity, triglyceride levels, younger age, non-smoking status, and treatment with clozapine. The effect of clozapine on insulin secretion was stronger in patients with normal glucose tolerance, who can be assumed to have a “healthier” pancreatic betacell mass than those with prediabetes. Despite a significant difference in glycemic levels between the two subgroups, the post-challenge insulin secretion was similar in subjects with normal glucose tolerance as compared with those with prediabetes, an indicator of the decrease in the pancreatic functional reserve present in prediabetics. The study is limited by the racial homogeneity of the cohort, smaller sample of prediabetic patients reducing the power for the multivariate analysis, and by its cross-sectional design, which did not allow for the observation of the relationship between treatment with clozapine and insulin secretion as individual patients transition from normal glucose tolerance to prediabetes.
3.3. Post-challenge glucose and insulin levels The 2-hour post-challenge and the AUCglucose were significantly higher in the prediabetic group, while the 2-hour insulin level, the AUCinsulin, and the magnitude of the Insulin Disposition Index were similar in the 2 subgroups (Table 4). 3.4. Predictors of post-challenge insulin secretion In the entire sample of non-diabetic subjects treated with SGAs (N= 520), the variables independently correlated with AUCinsulin were AUCglucose, waist circumference, triglycerides level, non-smoking status, younger age, and treatment with clozapine (r squared for the entire model 0.335, p b 0.0001). These variables retained their predictive value in the NGT subsample (r squared 0.371, p b 0.0001). The post-challenge insulin secretion in prediabetic patients correlated (r squared 0.313, p b 0.0001) only with AUCglucose, waist circumference, triglycerides and younger age (Table 5). 4. Discussion
Table 4 Endpoint, change and area under the curve values for 2-hour post-challenge glucose and insulin levels. Characteristic
The present study investigated the insulin secretion in a large cohort of non-diabetic patients treated with four widely used
Table 2 Psychotropic drug treatment and clinical characteristics. Characteristic
Olanzapine (N = 190)
Risperidone Quetiapine Clozapine (N = 166) (N = 91) (N = 73)
p-Value
Age, years ± S.D. Male gender, N (%) Smoking, N (%) BMI, kg/m2 ± S.D.
35.19±12.5 115 (69.3%) 107 (56.3%) 25.5 ± 5.0
37.1 ± 13.2 120 (72.3%) 103 (62.1%) 26.3 ± 4.6
0.64 0.041 0.11 0.029
37.4±12.9 38 (41.8%) 43 (47.3%) 25.6 ± 5.0
35.6 ± 9.5 48 (65.8%) 45 (61.6%) 27.3 ± 5.0
Total (N = 520)
NGT (N = 340)
Prediabetes (N = 180)
p-Value
Glucose 2-hour post-challenge, 95.8 ± 31.5 91.8 ± 29.8 103.4 ± 34.5 b0.0001 mg/dL ± S.D. Change compared with 7.6 ± 34.8 6.9 ± 33.8 9.0 ± 36.6 0.53 fasting level, % ± S.D. AUCglucose ± S.D. 14.6 ± 3.4 13.9 ± 3.1 16.1 ± 3.9 b0.0001 Insulin 2-hour post-challenge, 41.6 ± 44.3 39.1 ± 42.3 44.5 ± 47.8 0.075 μg/mL ± S.D. Change compared with 316.6 ± 388.4 306.5 ± 394.6 335.9 ± 376.1 0.42 fasting level, % ± S.D. 7.2 ± 4.4 7.1 ± 4.2 7.4 ± 4.8 0.55 AUCinsulin ± S.D. Insulin Disposition Index 0.13 ± 0.45 0.13 ± 0.53 0.12 ± 0.23 0.66 NGT = Normal Glucose Tolerance; S.D. = standard deviation; AUC: area under the curve.
P. Manu et al. / Schizophrenia Research 143 (2013) 358–362 Table 5 Independent correlates of post-challenge AUCinsulin.
AUCglucose Waist circumference Triglycerides Non-smoking status Younger age HDL cholesterol Female gender Treatment with clozapine
Total (N = 520) p-Value
NGT (N = 340) p-Value
Prediabetes (N = 180) p-Value
b0.0001 b0.0001 b0.0001 0.0012 b0.0001 ns ns 0.021
b0.0001 b0.0001 b0.0001 b0.0001 b0.0001 0.029 0.041 0.0506
b0.0001 b0.0001 0.0037 ns 0.035 ns ns ns
ns = not significant.
The independent association between glucose levels, abdominal adiposity, triglyceride levels, smoking status and age, and post-challenge insulin levels found in our cohort is not surprising. The plasma glucose level is the primary physiologic stimulus for insulin secretion. A larger waist circumference is a near-universal corollary of obesity, which is associated with insulin resistance and hyperinsulinemia (Polonsky et al., 1988; Adiels et al., 2008; Bacha et al., 2009; Jill et al., 2009). The obesity-related hyperinsulinemia reflects an increased insulin secretion as well as decreased insulin clearance through hepatic extraction (Seino et al., 2011). The latter is, however, a minor determinant of hyperinsulinemia in obese individuals. The insulin response in obesity is inversely correlated with the degree of glucose tolerance (Bacha et al., 2009), but the regulatory mechanisms are normal and the increase in insulin production is due to a larger beta-cell mass and not explained by hyperresponsiveness to extrinsic stimuli (Polonsky et al., 1988; Seino et al., 2011). The correlation between hypertriglyceridemia and insulin resistance leading to increased insulin secretion is also very wellestablished (Vossen et al., 2011). Insulin resistance influences triglyceride levels, particularly the very low-density lipoprotein concentration, by increasing the free fatty acid output from the adipose tissues (Adiels et al., 2008) and by decreasing the catabolism of chylomicrons and remnant particles (Lopez-Miranda et al., 2007). The excess of plasma triglycerides is partly distributed in nonadipose tissues, particularly in the skeletal muscle, where it becomes the substrate for the generation of harmful metabolites, such as ceramide and diacylglycerol, which in turn contribute to insulin resistance (Zhang et al., 2010; Eckardt et al., 2011). The correlation between non-smoker status and insulin secretion remains to be clarified, but smoking decreases the insulin secretion and increases the risk of pancreatic beta-cell failure and type 2 diabetes (Nakanishi et al., 2000; Cho et al., 2009). The effect appears to be related to the cytotoxic effect of the nitrosamines contained on the pancreatic islet cells, which may end in their apoptosis (Liu et al., 2011). Finally, age is universally accepted as a risk factor for decreased insulin secretion and type 2 diabetes in humans (American Diabetes Association, 2010) and a drop in the number of insulin-positive islet cells with aging was observed in experimental models using genetically lean and obese animals (Howarth et al., 2011). In clinical samples, the lower postchallenge insulin secretion in older subjects has been attributed not only to a decrease in the beta-cell mass, but also to the blunting of the age-related increase in adiponectin levels (Jill et al., 2009). Of the four second-generation antipsychotics studied by us, only clozapine was a predictor of the post-challenge insulin secretion. The correlation was independent of glucose levels, waist circumference, age and smoking status. This correlation has not been previously described, but in an 8-year cohort study of patient s receiving clozapine, the glucose dysregulation was associated with treatment duration rather than with weight gain (Bai et al., 2010). It is clear that the diabetogenic effect of clozapine is explained, at least in part, by its effect on insulin resistance, a fact amply demonstrated in clinical studies (Newcomer, 2005; De Hert et al., 2006, 2012) and in animal
361
models in which clozapine induced peripheral insulin resistance via increases in hepatic glucose output (Houseknecht et al., 2007; Smith et al., 2008a,b, 2009). However, these experiments have pointed out that olanzapine is no different than clozapine with respect to inducing an acute decrease in whole-body insulin sensitivity in a dose dependent manner, an effect not observed with risperidone (Houseknecht et al., 2007). The metabolic difference between olanzapine and clozapine cannot be attributed to differences in the half life of these drugs, which were administered the evening prior to our biochemical evaluation, as for olazapine, the half life of the parent compound approaches 30 h (Correll, 2010), while for clozapine and norclozapine is approximately 22 h (Renwick et al., 2000). The difference between clozapine and olanzapine, despite much larger power in the olanzapine sample compared to the clozapine sample (N = 190 vs. N = 73), observed in our study may be explained by the fact that the administration of clozapine may induce an increase in glucagon levels even when glucose levels are high (Smith et al., 2008a,b). In a convincing animal experiment performed after a period of chronic drug exposure (42 days) to mice fed either regular or high fat/high sugar, clozapine stimulated the 1 h postchallenge insulin and glucagon release (Smith et al., 2009). Notable is the fact that the glucagon secretion was increased even in the face of increased blood glucose levels as well as elevated insulin levels, both of which would normally act to suppress glucagon release. The clozapine effect on glucagon secretion appears mediated by a decrease in the plasma levels of GLP-1 (Smith et al., 2009). This clozapine-induced inappropriate secretion of glucagon has been confirmed in a recent experiment (Jassim et al., 2012), in which the expression levels of hepatic insulin receptor substrate 2 (Irs2), an essential component of the signaling pathway, which is typically down-regulated in the presence of hepatic insulin resistance, were measured. Contrary to what is expected, a significant upregulation of Irs2 was seen after 30 min (pb 0.01) in clozapine-exposed rats, which supports the hypothesis that the observed changes in glucose metabolism are, at least in part, also related to increased glucagon and not only to decreased insulin sensitivity. A limitation of our study is the fact that we did not assess glucagon levels or c-peptide levels which may have provided information regarding pre-hepatic insulin response (Faber et al., 1981; Mittelman et al. 2000). Recent data have also indicated increased glucagon secretion in healthy volunteers during subacute administration of olanzapine (Vidarsdotttir et al., 2010). It is also worth mentioning that the insulin response after oral glucose challenge is influenced by gastric-emptying rates, speed of enteral absorption and incretin responses. Despite these limitations, our findings can be safely interpreted to suggest the possibility that patients treated with some of the commonly used second-generation antipsychotics, particularly clozapine and possibly olanzapine, develop not only insulin resistance, but also hyperglucagonemic states, which may hasten pancreatic beta-cell failure and the emergence of diabetes. Treatment with clozapine appears to stimulate insulin secretion in patients with normal glucose tolerance independent of adiposity, a finding with two important implications. First, the insulin overproduction may lead to an earlier decrease in the beta-cell mass and secretory function, and a more rapid and prevalent transition to prediabetes and then diabetes compared with treatment with other obesogenic second-generation antipsychotics. Second, the progression to clinically overt dysglycemic states may occur in the absence of significant weight gain or of marked insulin resistance. From a practical standpoint, patients treated with clozapine must be considered a high risk group for the emergence of type 2 diabetes and periodic monitoring of fasting glucose and hemoglobin A1c should become the standard of care. Therapies combining drugs that improve insulin sensitivity (metformin) and increase the GLP-1 activity (exenatide) may decrease the risk of diabetes in clozapine-treated patients and should be tested in prospective clinical trials.
362
P. Manu et al. / Schizophrenia Research 143 (2013) 358–362
Role of the funding source Support for clinical and laboratory data collection. Contributors Dr. Manu designed the study and wrote the protocol and the first draft of the manuscript. Dr. Correll contributed to study design, performed the data extraction and analyses. Drs. Wampers, van Winkel, Yu and De Hert performed the main data collection and contributed to the interpretation of the results. Dr. Schiffeldrim and Kane managed the literature searches and contributed to the interpretation of the results. All authors contributed to and approved the final manuscript. Conflict of interest Dr. Correll has been a consultant and/or advisor to or has received honoraria from Actelion, Alexza, AstraZeneca, Biotis, Boehringer-Ingelheim, Bristol-Myers Squibb, Cephalon, Desitin, Eli Lilly, GSK, IntraCellular Therapies, MedAvante, Merck, Novartis, Ortho-McNeill/Janssen/J&J, Otsuka, Pfizer, ProPhase and Sunovion. Dr. van Winkel has received unrestricted grant support from Eli Lilly and Astra-Zeneca. Dr. Kane has been a consultant to or has received honoraria from Astra-Zeneca, Bristol-Myers Squibb, Cephalon, Eli Lilly, Janssen Pharmaceutica, Johnson and Johnson, Lundbeck, Otsuka, Pfizer Inc., PgXHealth, Proteus, Vanda and Wyeth, has served on the speaker's bureau of AstraZeneca, Bristol-Myers Squibb/Otsuka and Eli Lilly, and is a share holder of MedAvante. Dr. De Hert has served as a consultant to, received grant/research support and honoraria from, and served on the speakers or advisory boards of AstraZeneca, Bristol-Myers Squibb, Eli Lilly, Janssen-Cilag, Sanofi-Adventis and Lundbeck. Drs. De Hert, Wampers van Winkel and Yu's work on this study was supported by a European Foundation for the Study of Diabetes/Lilly grant. Drs. Manu and Shiffeldrim report no financial or other relationships relevant to the subject of this article. Acknowledgments None.
References Aas, I.H., 2011. Guidelines for rating Global Assessment of Functioning (GAF). Arch. Gen. Psychiatry 20 (10), 2. Adiels, M., Olofsson, S.O., Taskinen, M.R., Borén, J., 2008. Overproduction of very low-density lipoproteins is the hallmark of the dyslipidemia in the metabolic syndrome. Arterioscler. Thromb. Vasc. Biol. 28 (7), 1225–1236. American Diabetes Association, 2010. Stand. Med. Care Diabetes 33 (Suppl. 1), S11–S61. Bacha, F., Gungor, N., Lee, S., Arslanian, S.A., 2009. In vivo insulin sensitivity and secretion in obese youth; what are the differences between normal glucose tolerance, impaired glucose tolerance, and type 2 diabetes? Diabetes Care 32 (1), 100–105. Bai, Y., Lin, C., Chen, J., Chen, T., Su, T., Chou, P., 2010. Association of weight gain and metabolic syndrome in patients taking clozapine: an 8-year cohort study. J. Clin. Psychiatry e1–e6. Cho, N.D., Chan, J.C., Jang, H.C., Lim, S., Kim, H.L., Choi, S.H., 2009. Cigarette smoking is an independent risk factor for type 2 diabetes: a four-year community based prospective study. Clin. Endocrinol. (Oxf) 71 (5), 679–685. Correll, C.U., 2010. From receptor pharmacology to improved outcomes: individualizing the selection, dosing, and switching of antipsychotics. Eur. Psychiatry Suppl. 2, S12–S21. Correll, C.U., Lencz, T., Malhotra, A.K., 2011. Antipsychotic drug and obesity. Trends Mol. Med. 17, 97–107. De Hert, M., Van Winkel, R., Van Eyck, D., et al., 2006. Prevalence of diabetes, metabolic syndrome and metabolic abnormalities in schizophrenia over the course of the illness: a cross-sectional study. Clin. Pract. Epidemiol. Ment. Health 2, 14. De Hert, M., Detraux, J., van Winkel, R., Yu, W., Correll, CU, 2012. Metabolic and cardiovascular adverse effects associated with antipsychotic drugs. Nat. Rev. Endocrinol. 8 (2), 114–126. Eckardt, K., Taube, A., Eckel, J., 2011. Obesity associated insulin resistance in skeletal muscle: role of lipid accumulation and physical inactivity. Rev. Endocr. Metab. Disord. 12, 163–172. Faber, O.K., Christensen, K., Kehlet, H., Madsbad, S., Binder, C., 1981. Decreased insulin removal contributes to hyperinsulinemia in obesity. J. Clin. Endocrinol. Metab. 53, 618–621. Grundy, S.M., Brewer, H.B., Cleeman, J.I., et al., 2004. American Heart Association/National heart, Lung and Blood Institute. Definition of metabolic syndrome. Circulation 109, 433–438.
Houseknecht, K.L., Robertson, A.S., Zavadoski, W., Gibbes, E.M., Johnson, D.E., Rollema, H., 2007. Acute effects of atypical antipsychotics on whole-body insulin resistance in rats: implications for adverse metabolic effects. Neuropsychopharmacology 32 (2), 289–297. Howarth, F.C., Al Kitbi, M.D., Hameed, R.S., Adeghate, E., 2011. Pancreatic peptides in young and elderly Zucker type 2 diabetic fatty rats. JOP 12 (6), 567–573. Jassim, G., Skrede, S., Vázquez, M.J., Wergedal, H., Vik-Mo, A.O., Lunder, N., Diéguez, C., Vidal-Puig, A., Berge, R.K., López, M., Steen, V.M., Fernø, J., 2012. Acute effects of orexigenic antipsychotic drugs on lipid and carbohydrate metabolism in rat. Psychopharmacology (Berl) 219 (3), 783–794. Jill, C.P., Shamoon, H., Cohen, W.H., Reid, M., Gajavelli, S., Trandafirescu, G., Tabatabaie, V., Barzilai, N., 2009. Post-challenge hyperglycemia in older adults is associated with increased cardiovascular risk profile. J. Clin. Endocrinol. Metab. 94, 1595–1601. Kohen, I., Gampel, M., Reddy, L., Manu, P., 2008. Rapidly developing hyperglycemia during treatment with olanzapine. Ann. Pharmacother. 42, 588–591. Liu, T., Chen, W., David, S., Tyndale, R., Wang, H., Chen, Y., Yu, X., Chen, W., Zhou, Q., Ling, W., 2011. Eur. J. Endocrinol. 165, 961–967. Lopez-Miranda, J., Williams, C., Lairon, D., 2007. Dietary, physiological, genetic and pathological influences on postprandial lipid metabolism. Br. J. Nutr. 98 (3), 458–473. Manu, P., Correll, C.U., Van Winkel, R., Wampers, M., DeHert, M., 2012. Prediabetes in patients treated with antipsychotic drugs. J. Clin. Psychiatry 73, 40–466. Matthews, D.R., Hosker, J.P., Rudenski, A.S., Naylor, B.A., Treacher, D.F., Turner, R.C., 1985. Homeostasis model assessment: insulin resistance and beta-cell function from fasting plasma glucose and insulin concentration in man. Diabetologia 28, 412–419. Mittelman, S.D., Van Citters, G.W., Kim, S.P., Davis, D.A., Dea, M.K., Hamilton-Wessler, M., et al., 2000. Longitudinal compensation for fat-induced insulin resistance includes reduced insulin clearance and enhanced beta-cell response. Diabetes 49, 2116–2125. Nakanishi, N., Nakamura, K., Matsuo, Y., Suzuki, K., Tatara, K., 2000. Cigarette smoking and risk for impaired fasting glucose and type 2 diabetes in middle-aged Japanese men. Ann. Intern. Med. 133 (3), 183–191. Newcomer, J.W., 2005. Second-generation (atypical) antipsychotics and metabolic effects: a comprehensive literature review. CNS Drugs 19 (Suppl. 1), 1–93. Polonsky, K.S., Given, B.D., Hirsch, L., Shapiro, E.T., Tillil, H., Beebe, C., Galloway, J.A., Frank, B.H., Karrison, T., Van Cauter, E., 1988. Quantitative study of insulin secretion and clearance in normal and obese subjects. J. Clin. Invest. 81 (2), 435–441. Rabinowitz, J., Mehnert, A., Eerdekens, M., 2006. To what extent do the PANSS and CGI-S overlap? J. Clin. Psychopharmacol. 26, 303–307. Reaven, G., 2009. Insulin secretory function in Type 2 diabetes: does it matter how you measure it? J. Diabetes 1, 142–150. Renwick, A.C., Renwick, A.G., Flanagan, R.J., Ferner, R.E., 2000. Monitoring of clozapine and norclozapine plasma concentration–time curves in acute overdose. J. Toxicol. Clin. Toxicol. 38, 325–328. Seino, S., Shibasaki, T., Minami, K., 2011. Dynamics of insulin secretion and the clinical implications for obesity and diabetes. J. Clin. Invest. 121 (6), 2118–2125. http:// dx.doi.org/10.1172/JC145680. Smith, G.C., Chaussade, C., Vickers, M., Jensen, J., Shepherd, P.R., 2008a. Atypical antipsychotic drugs induce derangements in glucose homeostasis by acutely increasing glucagon secretion and hepatic glucose output in the rat. Diabetologia 51 (12), 2309–2317. Smith, M., Hopkins, D., Peveler, R.C., Holt, R.I., Woodward, M., Ismail, K., 2008b. First- v. second-generation antipsychotics and risk for diabetes in schizophrenia: systematic review and meta-analysis. Br. J. Psychiatry 192, 406–411. Smith, G.C., Vickers, M.H., Cognard, E., Shepherd, P.R., 2009. Clozapine and quetiapine acutely reduce glucagon-like peptide-1 production and increase glucagon release in obese rats: implications for glucose metabolism and food choice behaviour. Schizophr. Res. 115, 30–40. Utzschneider, K.M., Prigeon, R.L., Faulenbach, M.V., Tong, J., Carr, D.B., Boyko, E.J., et al., 2009. Oral disposition index predicts the development of future diabetes above and beyond fasting and 2-h glucose levels. Diabetes Care 32, 335–341. Vidarsdotttir, S., de leeuw van Weenen, J.E., Frolich, M., Romijn, J.A., Pijl, H., 2010. Effects of olanzapine and haloperidol on the metabolic status of healthy men. J. Clin. Endocrinol. Met. 95, 118–125. Vossen, M., Tödter, K., Altenburg, C., Beisiegel, U., Scheja, L., 2011. Atherosclerosis 217, 214–219. Zhang, L., Keung, W., Samokhvalov, V., Wang, W., Lopaschuk, G.D., 2010. Role of fatty acid uptake and fatty acid beta-oxidation in mediating insulin resistance in heart and skeletal muscle. Biochim. Biophys. Acta 1801 (1), 1–22.
Contents lists available at SciVerse ScienceDirect
Schizophrenia Research journal homepage: www.elsevier.com/locate/schres
Insulin secretion in patients receiving clozapine, olanzapine, quetiapine and risperidone Peter Manu a, b, c,⁎, 1, Christoph U. Correll a, b, d, 1, Martien Wampers e, Ruud van Winkel e, f, Weiping Yu e, Daphna Shiffeldrim a, John M. Kane a, b, Marc De Hert e a
Zucker Hillside Hospital, Glen Oaks, NY, United States Hofstra North Shore-LIJ School of Medicine, Hempstead, NY, United States Transilvania University, Brasov, Romania d Feinstein Institute for Medical Research, Manhasset, NY, United States e University Psychiatric Center, Kortenberg, Belgium f School of Mental Health and Neuroscience (EURON), University Medical Center, Maastricht, The Netherlands b c
a r t i c l e
i n f o
Article history: Received 12 July 2012 Received in revised form 12 November 2012 Accepted 14 November 2012 Available online 9 December 2012 Keywords: Antipsychotics Clozapine Prediabetes Insulin Oral glucose tolerance test
a b s t r a c t Background: Second-generation antipsychotics (SGAs) increase the risk of type 2 diabetes. The mechanism is thought to center on drug-induced weight gain, which starts the dysmetabolic cascade of insulin resistance, increased insulin production and pancreatic beta-cell failure. An independent effect of SGAs on insulin secretion has been suggested in animal models, but has not been demonstrated in clinical samples. Objective: To determine the post-challenge insulin secretion in patients treated with SGAs. Method: We identified 520 non-diabetic individuals treated with clozapine (N=73), olanzapine (N=190), quetiapine (N=91) or risperidone (N=166) in a consecutive, single-site cohort of 783 adult psychiatric inpatients who underwent a comprehensive metabolic assessment. Insulin secretion was measured as the area under the curve (AUCinsulin) generated by levels recorded at baseline, 30, 60 and 120 min after the intake of 75 g of glucose. The independent predictors of insulin secretion were determined with regression analysis in the entire sample and separately in patients with normal glucose tolerance (NGT) and prediabetes. Results: The post-challenge AUCinsulin was independently predicted by AUCglucose, waist circumference, triglyceride levels and younger age (pb 0.0001); non-smoking status (p=0.0012); and treatment with clozapine (p=0.021). The model explained 33.5% of the variance in insulin secretion (pb 0.0001). The clozapine effect was present in the NGT group, but not in prediabetics. Conclusions: Clozapine, but not olanzapine, quetiapine and risperidone, is an independent predictor of postchallenge insulin secretion in non-diabetics, particularly in those with normal glucose tolerance. The findings suggest that the diabetogenic risk of clozapine may persist even after weight reduction. © 2012 Elsevier B.V. All rights reserved.
1. Introduction The carbohydrate metabolism is frequently abnormal in patients treated with second-generation antipsychotic drugs (Newcomer, 2005; De Hert et al., 2012; Correll et al., 2011). These abnormalities include increased fasting glycemia, impaired glucose tolerance, prediabetes and diabetes and have been attributed to the insulin-resistant state that follows the weight gain observed in a large proportion of patients receiving long-term treatment with these drugs (De Hert et al., 2012; Manu et al., 2012). A weight-independent diabetogenic
⁎ Corresponding author at: Medical Services, Zucker Hillside Hospital, Glen Oaks, NY 11004, United States. E-mail address: [email protected] (P. Manu). 1 Drs. Manu and Correll have contributed equally to this article. 0920-9964/$ – see front matter © 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.schres.2012.11.016
effect of these drugs has been observed in animal models (Smith et al., 2008a,b), limited case series (Kohen et al., 2008) and one cohort study (Bai et al., 2010). Previous work has established the presence of fasting hyperinsulinemia, a biomarker for insulin resistance, in subjects receiving second-generation antipsychotics who have drug-induced weight gain, metabolic syndrome and/or prediabetes (Manu et al., 2012). However, the effect of second-generation antipsychotics on the insulin secretion after an oral glucose challenge has not been thoroughly evaluated by other investigators. The issue is important, because fasting insulin levels reflect only the basal pancreatic beta-cell activity, but not the meal-related insulin secretion. In fact, there is considerable evidence that concomitant measurements of the plasma glucose and insulin after an oral glucose challenge allow for the most accurate evaluation of the insulin secretory function, at least when compared with the Homeostatic Model Assessment, which relies on fasting insulin and glucose
P. Manu et al. / Schizophrenia Research 143 (2013) 358–362
concentrations, or compared with the insulin level measured shortly after an intravenous glucose challenge (Reaven, 2009). We assessed the insulin secretion after a standard oral glucose challenge in patients chronically treated with clozapine, olanzapine, risperidone and quetiapine. These drugs are prescribed world-wide and have demonstrated effectiveness for the treatment of bipolar mania, schizophrenia spectrum disorders and other psychotic syndromes. Two of these drugs, clozapine and olanzapine, are known to be the most obesogenic of the class, while the weight accrual and dysmetabolic liability of risperidone and quetiapine are usually described as moderate (Newcomer, 2005; De Hert et al. 2012). We hypothesized a) that insulin secretion after glucose challenge will confirm the difference between clozapine and olanzapine, on the one hand, and risperidone and quetiapine, on the other, and b) that some or all of these drugs will be an independent correlate of insulin secretion.
359
blood glucose ≥ 100 mg/dL; serum triglycerides ≥ 150 mg/dL; HDLcholesterol b 40 mg/dL in men and b50 mg/dL in women; and arterial blood pressure ≥ 130/85 mm Hg or current treatment with antihypertensive agents (Grundy et al., 2004). The treating psychiatrists assessed the severity of symptoms and rated them using the Global Assessment of Function (GAF) from 0 (worst) to 100 (best) (Aas, 2011) and the Clinical Global Impressions Severity (CGI-S) Scale from 1 (normal) to 7 (extremely ill) (Rabinowitz et al., 2006). 2.3. Selection of subjects
From November 2003 through July 2007, 798 non-diabetic patients receiving antipsychotic drugs consecutively admitted to the University Psychiatric Center, Kortenberg, Belgium, were asked by their treating psychiatrist to agree to a standardized battery of tests to identify metabolic syndrome and insulin resistance. A great majority of the patients (95.1%) were European-born Whites; others were Black Africans (2.4%), Maghreb-born individuals (1.5%) and Asians (1%). Psychiatric diagnoses were established according to DSM-IV by experienced psychiatrists affiliated with the University Psychiatric Center and responsible for the patient's treatment. All subjects gave written informed consent and the study was approved by the University Psychiatric Center's Ethics Committee.
Of the original 798 patients, we excluded 15 who declined to consent to metabolic screening and 80 with evidence of diabetes mellitus, as indicated by a fasting plasma glucose greater than 125 mg/dL or 2-hour glucose level during OGTT of at least 200 mg/dL or hemoglobin A1c 6.5% or greater. From the remaining cohort of 703 patients, we excluded 183 patients who either were on antipsychotic polypharmacy (N= 87), or who were treated with aripiprazole, amisulpride or various first-generation antipsychotics, but with sample sizes of well below 10% of the cohort. As the final sample, we selected 520 patients who were treated with antipsychotics used as monotherapy and who comprised at least 10% of the sample, i.e., olanzapine (N= 190), risperidone (N= 160), quetiapine (N= 91) and clozapine (N= 73). The proportions of patients treated with the same antipsychotic drug for more than 3 months were 76.0% for clozapine, 46.2% for risperidone, 46.2% for olanzapine, and 38.7% for quetiapine. The study group (N= 520) was divided into subgroups with normal glucose tolerance (fasting plasma glucose less than 100 mg/dL, 2-hour glucose level during OGTT less than 140 mg/dL and hemoglobin A1c 5.6% or less) and with prediabetes (fasting plasma glucose 100–125 mg/dL or 2-hour glucose level during OGTT in the 140–199 mg/dL range or hemoglobin A1c in the 5.7–6.4% range).
2.2. Clinical and laboratory measurements
2.4. Statistical analyses
The procedures included measurements of height, weight, body mass index (BMI), waist circumference, arterial blood pressure, fasting blood glucose, insulin and blood lipids, hemoglobin A1c, and a 2-hour oral glucose tolerance test (OGTT) after the ingestion of 75 g of glucose. Glucose and insulin levels were assessed 30, 60 and 120 min after the glucose load. As described in a previous publication, all tests were performed in the same laboratory and using the same methods throughout the entire study period (Manu et al., 2012). The fasting glucose, 2-hour postprandial glucose during OGTT and hemoglobin A1c data were used to define normal glucose tolerance (fasting glucose less than 100 mg/dL, 2-hour postprandial glucose less than 140 mg/dL and A1c less than 5.7%), prediabetes (fasting glucose 100–125 mg/dL or 2-hour postprandial glucose 140–199 mg/dL or A1c 5.7%–6.4%), and diabetes mellitus (fasting glucose greater than 125 mg/dL or 2-hour postprandial glucose greater than 199 mg/dL or A1c greater than 6.4%). Glucose and insulin measurements at 0, 30, 60 and 120 min after a 75 gram oral glucose load were used to calculate the area under the curve (total AUC) for insulin (AUCinsulin) and glucose (AUCglucose). The data were also used to calculate the Insulin Disposition Index, a reliable indicator of beta-cell function adjusted for insulin sensitivity (Utzschneider et al., 2009). The fasting glucose and insulin data were used for the Homeostatic Model Assessment of Insulin Resistance (HOMA-IR) (Matthews et al., 1985). The weight and height were used to calculate the body mass index (BMI). The waist circumference, arterial blood pressure, and fasting glucose, triglycerides and high-density lipoprotein (HDL) cholesterol levels were used to determine whether the patient met criteria for metabolic syndrome, which was defined by the presence of at least 3 of the following 5 items: waist circumference > 88 cm in women and >102 cm in men; fasting
Analyses of variance and chi-square tests were used to compare continuous and categorical variables, respectively, in patients with normal glucose tolerance and prediabetes. Logistic regression was used to identify independent predictors of AUCinsulin in the entire study group of nondiabetic patients receiving antipsychotic monotherapy with these four drugs and separately in the subgroups with normal glucose tolerance and with prediabetes. All analyses were two-sided, with alpha of pb 0.05, using JMP 5.0.1., 1989–2003, SAS Institute Inc.
2. Materials and methods 2.1. Setting
3. Results 3.1. Demographic and psychiatric characteristics The sample included 340 patients with NGT and 180 with prediabetes. The NGT group had a lower age and a greater proportion of males (Table 1). The NGT and prediabetic subgroups were similar with respect to the distribution of psychiatric diagnoses, functional status, and the severity of psychiatric illness. Olanzapine, risperidone and quetiapine utilization did not differ between the two subgroups, but prediabetic patients were more likely to be treated with clozapine than patients with NGT (Table 1). The proportion of males was lower among subjects treated with quetiapine. Patients treated with the four SGAs were similar in age and prevalence of cigarette smoking. The body mass index was slightly greater in patients receiving clozapine (Table 2). 3.2. Anthropometric features and fasting metabolic parameters The fasting glucose was significantly greater in the prediabetic group, as expected. The other components of the metabolic syndrome
360
P. Manu et al. / Schizophrenia Research 143 (2013) 358–362
Table 1 Demographic and psychiatric characteristics. Characteristic
Total (N = 520)
NGT (N = 340)
Prediabetes (N = 180)
p-Value
Age, years ± S.D. Male gender, N (%) Smoking Psychiatric diagnosis, N (%) Schizophrenia Schizoaffective disorder Bipolar disorder Major depression Personality disorder GAF, score ± S.D. CGI, score ± S.D. Antipsychotic drug N (%)b Olanzapine Risperidone Quetiapine Clozapine
36.5 ± 11.9 319 (61.4%) 298 (57.3%)
33.8 ± 11.6 220 (64.7%) 119 (57.4%)
41.7 ± 12.3a 99 (55.0%) 103 (57.2%)
b0.0001 0.031 0.98 0.090
348 (66.9%) 59 (11.4%) 82 (15.8%) 10 (1.9%) 21 (4.0%) 55.8 ± 12.8 4.4 ± 0.9
241 (70.9%) 33 (9.7%) 47 (13.8%) 5 (1.5%) 14 (4.1%) 55.4 ± 13.1 4.4 ± 0.9
107 (59.4%) 26 (14.4%) 35 (19.4%) 5 (2.8%) 7 (3.9%) 56.5 ± 12.3 4.3 ± 0.9
190 (36.5%) 166 (31.9%) 91 (17.5%) 73 (14.0%)
125 (36.8%) 118 (34.7%) 62 (18.2%) 35 (10.3%)
65 48 29 38
Table 3 Anthropometric features, arterial blood pressure, fasting metabolic parameters, and prevalence of metabolic syndrome. Characteristic
0.34 0.23 0.006
(36.1%) (26.7%) (16.1%) (21.1%)
Total (N = 520)
NGT (N = 340)
Waist circumference, cm ± S.D. Males 95.1 ± 12.0 94.5 ± 11.6 Females 90.9 ± 14.3 88.9 ± 14.1 Systolic blood pressure, 123.5 ± 15.5 122.9 ± 15.9 mm Hg± S.D. Diastolic blood pressure, 77.4 ± 11.5 76.7 ± 11.6 mm Hg± S.D. Triglycerides, mg/dL 148.4 ± 88.3 145.8 ± 88.3 High-density lipoprotein cholesterol, mg/dL Males 47.1 ± 13.0 46.9 ± 13.3 Females 60.5 ± 16.6 62.2 ± 16.5 Glucose, mg/dL 89.9 ± 8.4 85.8 ± 6.2 Insulin, μg/mL 11.0 ± 7.9 10.5 ± 7.6 HOMA-IR 2.5 ± 1.9 2.3 ± 1.7 Metabolic syndrome, N (%) 141 (27.2%) 59 (17.4%)
Prediabetes (N = 180)
p-Value
96.4 ± 12.9 93.8 ± 14.5 124.6 ± 14.8
0.20 0.018 0.21
78.7 ± 11.2
0.053
153.8 ± 88.4
0.31
47.5 ± 12.2 58.0 ± 16.8 95.4 ± 11.5 11.9 ± 8.6 2.9 ± 2.2 82 (45.8%)
0.70 0.084 b0.0001 0.036 0.0003 b0.0001
NGT = Normal Glucose Tolerance; S.D. = standard deviation; GAF = Global Assessment of Functioning Scale; CGI = Clinical Global Impressions Scale. a p b 0.0001. b p b 0.006.
NGT = Normal Glucose Tolerance; S.D. = standard deviation; HOMA-IR = Homeostatic Model Assessment of Insulin Resistance.
were similar in the two subgroups, with the exception of a larger waist circumference in prediabetic females. The proportion of patients who fulfilled criteria for metabolic syndrome was more than double in the prediabetic subgroup. Fasting insulin levels and resistance to insulin were significantly greater in prediabetic patients (Table 3).
second-generation antipsychotic drugs, i.e., clozapine, olanzapine, quetiapine and risperidone, which are known to lead to weight accrual and to increase the risk of the emergence of metabolic syndrome (Newcomer, 2005; De Hert et al., 2012; Correll et al., 2011) and prediabetes (Manu et al., 2012). The study confirmed that fasting insulin levels and degree of insulin resistance are greater in prediabetic patients. The post-challenge insulin secretion correlated with glycemia, abdominal adiposity, triglyceride levels, younger age, non-smoking status, and treatment with clozapine. The effect of clozapine on insulin secretion was stronger in patients with normal glucose tolerance, who can be assumed to have a “healthier” pancreatic betacell mass than those with prediabetes. Despite a significant difference in glycemic levels between the two subgroups, the post-challenge insulin secretion was similar in subjects with normal glucose tolerance as compared with those with prediabetes, an indicator of the decrease in the pancreatic functional reserve present in prediabetics. The study is limited by the racial homogeneity of the cohort, smaller sample of prediabetic patients reducing the power for the multivariate analysis, and by its cross-sectional design, which did not allow for the observation of the relationship between treatment with clozapine and insulin secretion as individual patients transition from normal glucose tolerance to prediabetes.
3.3. Post-challenge glucose and insulin levels The 2-hour post-challenge and the AUCglucose were significantly higher in the prediabetic group, while the 2-hour insulin level, the AUCinsulin, and the magnitude of the Insulin Disposition Index were similar in the 2 subgroups (Table 4). 3.4. Predictors of post-challenge insulin secretion In the entire sample of non-diabetic subjects treated with SGAs (N= 520), the variables independently correlated with AUCinsulin were AUCglucose, waist circumference, triglycerides level, non-smoking status, younger age, and treatment with clozapine (r squared for the entire model 0.335, p b 0.0001). These variables retained their predictive value in the NGT subsample (r squared 0.371, p b 0.0001). The post-challenge insulin secretion in prediabetic patients correlated (r squared 0.313, p b 0.0001) only with AUCglucose, waist circumference, triglycerides and younger age (Table 5). 4. Discussion
Table 4 Endpoint, change and area under the curve values for 2-hour post-challenge glucose and insulin levels. Characteristic
The present study investigated the insulin secretion in a large cohort of non-diabetic patients treated with four widely used
Table 2 Psychotropic drug treatment and clinical characteristics. Characteristic
Olanzapine (N = 190)
Risperidone Quetiapine Clozapine (N = 166) (N = 91) (N = 73)
p-Value
Age, years ± S.D. Male gender, N (%) Smoking, N (%) BMI, kg/m2 ± S.D.
35.19±12.5 115 (69.3%) 107 (56.3%) 25.5 ± 5.0
37.1 ± 13.2 120 (72.3%) 103 (62.1%) 26.3 ± 4.6
0.64 0.041 0.11 0.029
37.4±12.9 38 (41.8%) 43 (47.3%) 25.6 ± 5.0
35.6 ± 9.5 48 (65.8%) 45 (61.6%) 27.3 ± 5.0
Total (N = 520)
NGT (N = 340)
Prediabetes (N = 180)
p-Value
Glucose 2-hour post-challenge, 95.8 ± 31.5 91.8 ± 29.8 103.4 ± 34.5 b0.0001 mg/dL ± S.D. Change compared with 7.6 ± 34.8 6.9 ± 33.8 9.0 ± 36.6 0.53 fasting level, % ± S.D. AUCglucose ± S.D. 14.6 ± 3.4 13.9 ± 3.1 16.1 ± 3.9 b0.0001 Insulin 2-hour post-challenge, 41.6 ± 44.3 39.1 ± 42.3 44.5 ± 47.8 0.075 μg/mL ± S.D. Change compared with 316.6 ± 388.4 306.5 ± 394.6 335.9 ± 376.1 0.42 fasting level, % ± S.D. 7.2 ± 4.4 7.1 ± 4.2 7.4 ± 4.8 0.55 AUCinsulin ± S.D. Insulin Disposition Index 0.13 ± 0.45 0.13 ± 0.53 0.12 ± 0.23 0.66 NGT = Normal Glucose Tolerance; S.D. = standard deviation; AUC: area under the curve.
P. Manu et al. / Schizophrenia Research 143 (2013) 358–362 Table 5 Independent correlates of post-challenge AUCinsulin.
AUCglucose Waist circumference Triglycerides Non-smoking status Younger age HDL cholesterol Female gender Treatment with clozapine
Total (N = 520) p-Value
NGT (N = 340) p-Value
Prediabetes (N = 180) p-Value
b0.0001 b0.0001 b0.0001 0.0012 b0.0001 ns ns 0.021
b0.0001 b0.0001 b0.0001 b0.0001 b0.0001 0.029 0.041 0.0506
b0.0001 b0.0001 0.0037 ns 0.035 ns ns ns
ns = not significant.
The independent association between glucose levels, abdominal adiposity, triglyceride levels, smoking status and age, and post-challenge insulin levels found in our cohort is not surprising. The plasma glucose level is the primary physiologic stimulus for insulin secretion. A larger waist circumference is a near-universal corollary of obesity, which is associated with insulin resistance and hyperinsulinemia (Polonsky et al., 1988; Adiels et al., 2008; Bacha et al., 2009; Jill et al., 2009). The obesity-related hyperinsulinemia reflects an increased insulin secretion as well as decreased insulin clearance through hepatic extraction (Seino et al., 2011). The latter is, however, a minor determinant of hyperinsulinemia in obese individuals. The insulin response in obesity is inversely correlated with the degree of glucose tolerance (Bacha et al., 2009), but the regulatory mechanisms are normal and the increase in insulin production is due to a larger beta-cell mass and not explained by hyperresponsiveness to extrinsic stimuli (Polonsky et al., 1988; Seino et al., 2011). The correlation between hypertriglyceridemia and insulin resistance leading to increased insulin secretion is also very wellestablished (Vossen et al., 2011). Insulin resistance influences triglyceride levels, particularly the very low-density lipoprotein concentration, by increasing the free fatty acid output from the adipose tissues (Adiels et al., 2008) and by decreasing the catabolism of chylomicrons and remnant particles (Lopez-Miranda et al., 2007). The excess of plasma triglycerides is partly distributed in nonadipose tissues, particularly in the skeletal muscle, where it becomes the substrate for the generation of harmful metabolites, such as ceramide and diacylglycerol, which in turn contribute to insulin resistance (Zhang et al., 2010; Eckardt et al., 2011). The correlation between non-smoker status and insulin secretion remains to be clarified, but smoking decreases the insulin secretion and increases the risk of pancreatic beta-cell failure and type 2 diabetes (Nakanishi et al., 2000; Cho et al., 2009). The effect appears to be related to the cytotoxic effect of the nitrosamines contained on the pancreatic islet cells, which may end in their apoptosis (Liu et al., 2011). Finally, age is universally accepted as a risk factor for decreased insulin secretion and type 2 diabetes in humans (American Diabetes Association, 2010) and a drop in the number of insulin-positive islet cells with aging was observed in experimental models using genetically lean and obese animals (Howarth et al., 2011). In clinical samples, the lower postchallenge insulin secretion in older subjects has been attributed not only to a decrease in the beta-cell mass, but also to the blunting of the age-related increase in adiponectin levels (Jill et al., 2009). Of the four second-generation antipsychotics studied by us, only clozapine was a predictor of the post-challenge insulin secretion. The correlation was independent of glucose levels, waist circumference, age and smoking status. This correlation has not been previously described, but in an 8-year cohort study of patient s receiving clozapine, the glucose dysregulation was associated with treatment duration rather than with weight gain (Bai et al., 2010). It is clear that the diabetogenic effect of clozapine is explained, at least in part, by its effect on insulin resistance, a fact amply demonstrated in clinical studies (Newcomer, 2005; De Hert et al., 2006, 2012) and in animal
361
models in which clozapine induced peripheral insulin resistance via increases in hepatic glucose output (Houseknecht et al., 2007; Smith et al., 2008a,b, 2009). However, these experiments have pointed out that olanzapine is no different than clozapine with respect to inducing an acute decrease in whole-body insulin sensitivity in a dose dependent manner, an effect not observed with risperidone (Houseknecht et al., 2007). The metabolic difference between olanzapine and clozapine cannot be attributed to differences in the half life of these drugs, which were administered the evening prior to our biochemical evaluation, as for olazapine, the half life of the parent compound approaches 30 h (Correll, 2010), while for clozapine and norclozapine is approximately 22 h (Renwick et al., 2000). The difference between clozapine and olanzapine, despite much larger power in the olanzapine sample compared to the clozapine sample (N = 190 vs. N = 73), observed in our study may be explained by the fact that the administration of clozapine may induce an increase in glucagon levels even when glucose levels are high (Smith et al., 2008a,b). In a convincing animal experiment performed after a period of chronic drug exposure (42 days) to mice fed either regular or high fat/high sugar, clozapine stimulated the 1 h postchallenge insulin and glucagon release (Smith et al., 2009). Notable is the fact that the glucagon secretion was increased even in the face of increased blood glucose levels as well as elevated insulin levels, both of which would normally act to suppress glucagon release. The clozapine effect on glucagon secretion appears mediated by a decrease in the plasma levels of GLP-1 (Smith et al., 2009). This clozapine-induced inappropriate secretion of glucagon has been confirmed in a recent experiment (Jassim et al., 2012), in which the expression levels of hepatic insulin receptor substrate 2 (Irs2), an essential component of the signaling pathway, which is typically down-regulated in the presence of hepatic insulin resistance, were measured. Contrary to what is expected, a significant upregulation of Irs2 was seen after 30 min (pb 0.01) in clozapine-exposed rats, which supports the hypothesis that the observed changes in glucose metabolism are, at least in part, also related to increased glucagon and not only to decreased insulin sensitivity. A limitation of our study is the fact that we did not assess glucagon levels or c-peptide levels which may have provided information regarding pre-hepatic insulin response (Faber et al., 1981; Mittelman et al. 2000). Recent data have also indicated increased glucagon secretion in healthy volunteers during subacute administration of olanzapine (Vidarsdotttir et al., 2010). It is also worth mentioning that the insulin response after oral glucose challenge is influenced by gastric-emptying rates, speed of enteral absorption and incretin responses. Despite these limitations, our findings can be safely interpreted to suggest the possibility that patients treated with some of the commonly used second-generation antipsychotics, particularly clozapine and possibly olanzapine, develop not only insulin resistance, but also hyperglucagonemic states, which may hasten pancreatic beta-cell failure and the emergence of diabetes. Treatment with clozapine appears to stimulate insulin secretion in patients with normal glucose tolerance independent of adiposity, a finding with two important implications. First, the insulin overproduction may lead to an earlier decrease in the beta-cell mass and secretory function, and a more rapid and prevalent transition to prediabetes and then diabetes compared with treatment with other obesogenic second-generation antipsychotics. Second, the progression to clinically overt dysglycemic states may occur in the absence of significant weight gain or of marked insulin resistance. From a practical standpoint, patients treated with clozapine must be considered a high risk group for the emergence of type 2 diabetes and periodic monitoring of fasting glucose and hemoglobin A1c should become the standard of care. Therapies combining drugs that improve insulin sensitivity (metformin) and increase the GLP-1 activity (exenatide) may decrease the risk of diabetes in clozapine-treated patients and should be tested in prospective clinical trials.
362
P. Manu et al. / Schizophrenia Research 143 (2013) 358–362
Role of the funding source Support for clinical and laboratory data collection. Contributors Dr. Manu designed the study and wrote the protocol and the first draft of the manuscript. Dr. Correll contributed to study design, performed the data extraction and analyses. Drs. Wampers, van Winkel, Yu and De Hert performed the main data collection and contributed to the interpretation of the results. Dr. Schiffeldrim and Kane managed the literature searches and contributed to the interpretation of the results. All authors contributed to and approved the final manuscript. Conflict of interest Dr. Correll has been a consultant and/or advisor to or has received honoraria from Actelion, Alexza, AstraZeneca, Biotis, Boehringer-Ingelheim, Bristol-Myers Squibb, Cephalon, Desitin, Eli Lilly, GSK, IntraCellular Therapies, MedAvante, Merck, Novartis, Ortho-McNeill/Janssen/J&J, Otsuka, Pfizer, ProPhase and Sunovion. Dr. van Winkel has received unrestricted grant support from Eli Lilly and Astra-Zeneca. Dr. Kane has been a consultant to or has received honoraria from Astra-Zeneca, Bristol-Myers Squibb, Cephalon, Eli Lilly, Janssen Pharmaceutica, Johnson and Johnson, Lundbeck, Otsuka, Pfizer Inc., PgXHealth, Proteus, Vanda and Wyeth, has served on the speaker's bureau of AstraZeneca, Bristol-Myers Squibb/Otsuka and Eli Lilly, and is a share holder of MedAvante. Dr. De Hert has served as a consultant to, received grant/research support and honoraria from, and served on the speakers or advisory boards of AstraZeneca, Bristol-Myers Squibb, Eli Lilly, Janssen-Cilag, Sanofi-Adventis and Lundbeck. Drs. De Hert, Wampers van Winkel and Yu's work on this study was supported by a European Foundation for the Study of Diabetes/Lilly grant. Drs. Manu and Shiffeldrim report no financial or other relationships relevant to the subject of this article. Acknowledgments None.
References Aas, I.H., 2011. Guidelines for rating Global Assessment of Functioning (GAF). Arch. Gen. Psychiatry 20 (10), 2. Adiels, M., Olofsson, S.O., Taskinen, M.R., Borén, J., 2008. Overproduction of very low-density lipoproteins is the hallmark of the dyslipidemia in the metabolic syndrome. Arterioscler. Thromb. Vasc. Biol. 28 (7), 1225–1236. American Diabetes Association, 2010. Stand. Med. Care Diabetes 33 (Suppl. 1), S11–S61. Bacha, F., Gungor, N., Lee, S., Arslanian, S.A., 2009. In vivo insulin sensitivity and secretion in obese youth; what are the differences between normal glucose tolerance, impaired glucose tolerance, and type 2 diabetes? Diabetes Care 32 (1), 100–105. Bai, Y., Lin, C., Chen, J., Chen, T., Su, T., Chou, P., 2010. Association of weight gain and metabolic syndrome in patients taking clozapine: an 8-year cohort study. J. Clin. Psychiatry e1–e6. Cho, N.D., Chan, J.C., Jang, H.C., Lim, S., Kim, H.L., Choi, S.H., 2009. Cigarette smoking is an independent risk factor for type 2 diabetes: a four-year community based prospective study. Clin. Endocrinol. (Oxf) 71 (5), 679–685. Correll, C.U., 2010. From receptor pharmacology to improved outcomes: individualizing the selection, dosing, and switching of antipsychotics. Eur. Psychiatry Suppl. 2, S12–S21. Correll, C.U., Lencz, T., Malhotra, A.K., 2011. Antipsychotic drug and obesity. Trends Mol. Med. 17, 97–107. De Hert, M., Van Winkel, R., Van Eyck, D., et al., 2006. Prevalence of diabetes, metabolic syndrome and metabolic abnormalities in schizophrenia over the course of the illness: a cross-sectional study. Clin. Pract. Epidemiol. Ment. Health 2, 14. De Hert, M., Detraux, J., van Winkel, R., Yu, W., Correll, CU, 2012. Metabolic and cardiovascular adverse effects associated with antipsychotic drugs. Nat. Rev. Endocrinol. 8 (2), 114–126. Eckardt, K., Taube, A., Eckel, J., 2011. Obesity associated insulin resistance in skeletal muscle: role of lipid accumulation and physical inactivity. Rev. Endocr. Metab. Disord. 12, 163–172. Faber, O.K., Christensen, K., Kehlet, H., Madsbad, S., Binder, C., 1981. Decreased insulin removal contributes to hyperinsulinemia in obesity. J. Clin. Endocrinol. Metab. 53, 618–621. Grundy, S.M., Brewer, H.B., Cleeman, J.I., et al., 2004. American Heart Association/National heart, Lung and Blood Institute. Definition of metabolic syndrome. Circulation 109, 433–438.
Houseknecht, K.L., Robertson, A.S., Zavadoski, W., Gibbes, E.M., Johnson, D.E., Rollema, H., 2007. Acute effects of atypical antipsychotics on whole-body insulin resistance in rats: implications for adverse metabolic effects. Neuropsychopharmacology 32 (2), 289–297. Howarth, F.C., Al Kitbi, M.D., Hameed, R.S., Adeghate, E., 2011. Pancreatic peptides in young and elderly Zucker type 2 diabetic fatty rats. JOP 12 (6), 567–573. Jassim, G., Skrede, S., Vázquez, M.J., Wergedal, H., Vik-Mo, A.O., Lunder, N., Diéguez, C., Vidal-Puig, A., Berge, R.K., López, M., Steen, V.M., Fernø, J., 2012. Acute effects of orexigenic antipsychotic drugs on lipid and carbohydrate metabolism in rat. Psychopharmacology (Berl) 219 (3), 783–794. Jill, C.P., Shamoon, H., Cohen, W.H., Reid, M., Gajavelli, S., Trandafirescu, G., Tabatabaie, V., Barzilai, N., 2009. Post-challenge hyperglycemia in older adults is associated with increased cardiovascular risk profile. J. Clin. Endocrinol. Metab. 94, 1595–1601. Kohen, I., Gampel, M., Reddy, L., Manu, P., 2008. Rapidly developing hyperglycemia during treatment with olanzapine. Ann. Pharmacother. 42, 588–591. Liu, T., Chen, W., David, S., Tyndale, R., Wang, H., Chen, Y., Yu, X., Chen, W., Zhou, Q., Ling, W., 2011. Eur. J. Endocrinol. 165, 961–967. Lopez-Miranda, J., Williams, C., Lairon, D., 2007. Dietary, physiological, genetic and pathological influences on postprandial lipid metabolism. Br. J. Nutr. 98 (3), 458–473. Manu, P., Correll, C.U., Van Winkel, R., Wampers, M., DeHert, M., 2012. Prediabetes in patients treated with antipsychotic drugs. J. Clin. Psychiatry 73, 40–466. Matthews, D.R., Hosker, J.P., Rudenski, A.S., Naylor, B.A., Treacher, D.F., Turner, R.C., 1985. Homeostasis model assessment: insulin resistance and beta-cell function from fasting plasma glucose and insulin concentration in man. Diabetologia 28, 412–419. Mittelman, S.D., Van Citters, G.W., Kim, S.P., Davis, D.A., Dea, M.K., Hamilton-Wessler, M., et al., 2000. Longitudinal compensation for fat-induced insulin resistance includes reduced insulin clearance and enhanced beta-cell response. Diabetes 49, 2116–2125. Nakanishi, N., Nakamura, K., Matsuo, Y., Suzuki, K., Tatara, K., 2000. Cigarette smoking and risk for impaired fasting glucose and type 2 diabetes in middle-aged Japanese men. Ann. Intern. Med. 133 (3), 183–191. Newcomer, J.W., 2005. Second-generation (atypical) antipsychotics and metabolic effects: a comprehensive literature review. CNS Drugs 19 (Suppl. 1), 1–93. Polonsky, K.S., Given, B.D., Hirsch, L., Shapiro, E.T., Tillil, H., Beebe, C., Galloway, J.A., Frank, B.H., Karrison, T., Van Cauter, E., 1988. Quantitative study of insulin secretion and clearance in normal and obese subjects. J. Clin. Invest. 81 (2), 435–441. Rabinowitz, J., Mehnert, A., Eerdekens, M., 2006. To what extent do the PANSS and CGI-S overlap? J. Clin. Psychopharmacol. 26, 303–307. Reaven, G., 2009. Insulin secretory function in Type 2 diabetes: does it matter how you measure it? J. Diabetes 1, 142–150. Renwick, A.C., Renwick, A.G., Flanagan, R.J., Ferner, R.E., 2000. Monitoring of clozapine and norclozapine plasma concentration–time curves in acute overdose. J. Toxicol. Clin. Toxicol. 38, 325–328. Seino, S., Shibasaki, T., Minami, K., 2011. Dynamics of insulin secretion and the clinical implications for obesity and diabetes. J. Clin. Invest. 121 (6), 2118–2125. http:// dx.doi.org/10.1172/JC145680. Smith, G.C., Chaussade, C., Vickers, M., Jensen, J., Shepherd, P.R., 2008a. Atypical antipsychotic drugs induce derangements in glucose homeostasis by acutely increasing glucagon secretion and hepatic glucose output in the rat. Diabetologia 51 (12), 2309–2317. Smith, M., Hopkins, D., Peveler, R.C., Holt, R.I., Woodward, M., Ismail, K., 2008b. First- v. second-generation antipsychotics and risk for diabetes in schizophrenia: systematic review and meta-analysis. Br. J. Psychiatry 192, 406–411. Smith, G.C., Vickers, M.H., Cognard, E., Shepherd, P.R., 2009. Clozapine and quetiapine acutely reduce glucagon-like peptide-1 production and increase glucagon release in obese rats: implications for glucose metabolism and food choice behaviour. Schizophr. Res. 115, 30–40. Utzschneider, K.M., Prigeon, R.L., Faulenbach, M.V., Tong, J., Carr, D.B., Boyko, E.J., et al., 2009. Oral disposition index predicts the development of future diabetes above and beyond fasting and 2-h glucose levels. Diabetes Care 32, 335–341. Vidarsdotttir, S., de leeuw van Weenen, J.E., Frolich, M., Romijn, J.A., Pijl, H., 2010. Effects of olanzapine and haloperidol on the metabolic status of healthy men. J. Clin. Endocrinol. Met. 95, 118–125. Vossen, M., Tödter, K., Altenburg, C., Beisiegel, U., Scheja, L., 2011. Atherosclerosis 217, 214–219. Zhang, L., Keung, W., Samokhvalov, V., Wang, W., Lopaschuk, G.D., 2010. Role of fatty acid uptake and fatty acid beta-oxidation in mediating insulin resistance in heart and skeletal muscle. Biochim. Biophys. Acta 1801 (1), 1–22.