- Email: [email protected]
The prevalence, risk factors, and clinical characteristics of insulin resistance in Chinese patients with schizophrenia
Comprehensive Psychiatry 96 (2020) 152145
Contents lists available at ScienceDirect
Comprehensive Psychiatry journal homepage: www.elsevier.com/locate/comppsych
The prevalence, risk factors, and clinical characteristics of insulin resistance in Chinese patients with schizophrenia Chen Lin b, Ke Chen a, Rongzhen Zhang a, Weihong Fu a, Jianjin Yu a, Lan Gao a, Haiqing Ni c, Jiaqi Song d, Dachun Chen a,⁎ a
Beijing HuiLongGuan Hospital, Beijing, 100096, PR China Department of Psychosomatic Medicine, Beijing HuiLongGuan Hospital, Peking University, Beijing, 100096, PR China Guangzhou Brain Hospital, Guangzhou, 510450, PR China d Peking University Health Science Center, Beijing, 100191, PR China b c
a r t i c l e Available online xxxx
Keywords: Schizophrenia Insulin resistance Prevalence Risk factors
i n f o
a b s t r a c t Background: Studies have shown that patients with schizophrenia are at a high risk of developing insulin resistance (IR). We investigated the prevalence of IR and its clinical correlates in hospitalized Chinese patients with schizophrenia. Methods: A total of 193 patients with schizophrenia (113 males and 80 females) were recruited for the study. We collected their demographic and clinical data, including data on their plasma glucose and lipid levels. All patients were rated using the Repeatable Battery for the Assessment of Neuropsychological Status (RBANS) to assess cognitive function, while Positive and Negative Syndrome Scale (PANSS) was used to assess psychopathology. The cut-off value for the homeostasis model assessment of insulin resistance (HOMA-IR) was set at 1.7. Results: The prevalence of IR was 37.82% (73/193). The IR patients had significantly higher waist-to-hip ratio and body mass index (BMI), and higher fasting plasma glucose (FPG), triglyceride (TG), and low density lipoprotein (LDL) levels compared to non-IR patients (all p b .05). Binary logistic regression analysis showed that smoking, BMI, and TG and LDL levels are significant predictors of IR. In addition, correlation analysis showed that IR was significantly correlated with the waist-to-hip ratio, BMI, and LDL level (Bonferroni corrected p b .05). The multivariable linear regression analysis indicated that the BMI and FPG are associated with the IR index. There was no significant difference in IR index between patients who were taking different antipsychotics. Conclusion: We found a high prevalence of IR and its risk factors in Chinese patients with schizophrenia. Active weight control to reduce the BMI and waist circumference and reducing the number of cigarettes consumed, may be essential to decrease the incidence of IR in patients with schizophrenia. © 2019 Published by Elsevier Inc. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
1. Introduction Many studies have shown that insulin resistance (IR) is more prevalent in patients with schizophrenia than in healthy controls [1–3]. IR, an important cause of metabolic syndrome, significantly affects physical and mental health. As early as 1995, Stern [4] proposed the “common soil theory.” He believed that IR was the link between many diseases and that it was the original cause and pathogenic basis of many metabolic abnormalities and cardiovascular diseases. Recently some studies have shown that type 2 diabetes and IR are risk factors for cognitive dysfunction and dementia [5,6]. IR, one of the main pathogenic factors of type 2 diabetes, is an independent risk factor for coronary heart disease [7]. A study with an average follow-up of ⁎ Corresponding author at: Fifth Clinical Department, Beijing HuiLongGuan Hospital, Peking University, Beijing, 100096, PR China. E-mail address: [email protected] (D. Chen).
22 years showed that the relative risk of cerebral infarction in IR patients was 2.12 times higher than that in non-IR patients [8]. IR greatly reduces the quality of life and average life expectancy of patients. The causes of IR have not been fully elucidated. Possible risk factors for IR in the general population include obesity, poor diet, lack of exercise, age, and genetic predisposition [9]. For patients with schizophrenia, many commonly used antipsychotics (e.g., quetiapine and olanzapine) have been shown to be associated with IR and metabolic syndrome [10–13]. However, there is also a high prevalence of IR in first-episode drug-naïve patients with schizophrenia [1,14,15], suggesting that IR in schizophrenia may be caused by factors other than drugs. Presently, the risk factors for IR and their clinical correlations in Chinese patients with schizophrenia have not been fully investigated. The major objectives of our study were to investigate: (1) the prevalence, risk factors, and clinical correlates of IR in Chinese patients with schizophrenia; (2) the correlation between IR index and demographic and clinical variables, psychopathology, and cognitive function; and
https://doi.org/10.1016/j.comppsych.2019.152145 0010-440X/© 2019 Published by Elsevier Inc. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
2
C. Lin et al. / Comprehensive Psychiatry 96 (2020) 152145
(3) the differences in IR and other metabolic parameters in patients treated with different types of antipsychotics. 1. Subjects and Methods 1.1. Subjects A total of 193 patients with schizophrenia were recruited from June 2015 to December 2018 at the Beijing HuiLongGuan Hospital. Only patients who fulfilled the Diagnostic and Statistical Manual of Mental Disorders, fifth edition (DSM-V) criteria for schizophrenia, were between the ages of 18 and 60 years, and were of the Han nationality were included in the study. These group of patients were excluded from the study: (1) patients with obvious central nervous system diseases, such as stroke, tumor, dementia, Parkinson's disease, and epilepsy; (2) patients with serious physical diseases (except diabetes), such as hypertension and heart disease, among others; (3) patients with drug or alcohol dependence; (4) pregnant or lactating patients; and (5) patients with obvious mental retardation and could not complete the scale and cognitive function tests. This study was approved by the ethics committee of the Beijing HuiLongGuan Hospital. All subjects provided informed consent. 2. Methods 2.1.1. Demographic variables The demographic data (gender, age, and education level) of patients with schizophrenia who met the inclusion criteria were collected in detail using the homemade basic information collection table. Weight, height, and waist and hip circumference were measured, and the body mass index (BMI) (weight (kg) divided by the height (m)2) was calculated. Data on medications (atypical antipsychotics) which patients had been taking for approximately six months, prior to their inclusion in the study, were collected. Smokers were classified as: 1) regular smokers: people who smoke more than one cigarette per day; and 2) occasional smokers: people who smoke more than four times a week, but b1 cigarette per day on average, according to the World Health Organization recommendations. In this study, only regular smokers, and not occasional smokers, were included in the smoking group. All patients were hospitalized with a unified management of diet and exercise. 2.1.2. Psychiatric symptom assessment The Positive and Negative Syndrome Scale (PANSS), which was developed by Kay et al. in 1987, is mainly used to evaluate the psychotic symptoms of patients. These symptoms comprise of seven positive symptoms, seven negative symptoms, 16 general pathological symptoms, and three additional symptoms. The PANSS consists of 33 items. It is scored on a scale of 1–7. The last three additional symptoms were not included in the total score. The higher the total score, the more severe the symptoms of the patients. The PANSS is relatively comprehensive in its assessment of the psychiatric symptoms of patients. The PANSS can reflect the overall clinical phase of patients. 2.1.3. Cognitive function assessment For the cognitive function assessment, the Repeatable Battery for the Assessment of Neuropsychological Status (RBANS) [16] was used. This RBANS consists of 12 items: list learning, story memory, figure copy, line orientation, picture naming, semantic fluency, digit span, coding test, list recall, list recognition, story recall, and figure recall. This cognitive assessment can be summarized into five groups of neuropsychological status (5 factors): immediate memory, visuospatial, language, attention, and delayed memory. The higher the score, the better the cognitive function. Before the study commenced, all scales were tested for consistency. The intraclass correlation coefficient (ICC) was N0.8.
2.1.4. Blood samples Patients avoided greasy foods the day before blood sampling. Five milliliters of venous blood was collected from inpatients at 6:00 a.m., following an overnight fast (avoided food for N10 h) and was stored at −40 °C. An automatic biochemical analyzer (Beckman AU5800) was used to measure the fasting plasma glucose (FPG), insulin, cholesterol (CHO), triglycerides (TG), high-density lipoprotein (HDL), and low density lipoprotein (LDL) levels. 2.1.5. HOMA-IR calculation The insulin resistance index (HOMA-IR) = (fasting plasma glucose (FPG, mmol/L) * fasting insulin (FINS, mIU/L)) /22.5. The standard of HOMA-IR for normal-weight normal subjects was set to 1 [17,18]. Although this index is mainly used in the research field, the HOMA-IR assesses an individual's level of IR. The results correlate with those of standard tests for IR, such as tests involving the use of the insulin clamp [19]. After taking the Metabolic Syndrome components into account, the threshold value of the HOMA-IR was set at 1.7 [20–22]. Patients with HOMA-IR values lower than 1.7 were assigned to the nonIR group, and patients with HOMA-IR values N1.7 were assigned to the IR group. 2.1.6. Statistical analysis Epidata 3.0 was used to create a database for all the data. The SPSS software version 18.0 was used for statistical analysis. The demographic and clinical variables were compared between the different groups using the independent sample t-test or analysis of variance (ANOVA) for continuous variables and chi-squared or Fisher's exact tests for categorical variables. We described the prevalence of IR using percentages. The prevalence was analyzed using the chi-square test, and the risk factors for IR were assessed using a binary logistic regression analysis. The Pearson or Spearman correlation analysis was used to examine the relationship between IR and demographic and clinical variables. Finally, the multivariable linear regression analysis was used to investigate significant predictive variables associated with the IR index. Bonferroni correction was applied to adjust for multiple testing. All p-values were two-tailed and significance was set at p b .05. 3. Results 3.1. Demographic data and clinical characteristics In total, 193 patients (113 males and 80 females) were enrolled in the study. The mean age of patients was 50.37 ± 7.55 years (range: 25–60 years). The average duration of education was 10.89 ± 2.52 years (range: 5–20 years). The mean duration of illness was 297.09 ± 118.38 months (range: 6–516 months). The average age at onset was 25.68 ± 8.24 years (range: 13–55 years). Fifty-two patients (26.94%) were smokers. 3.2. Demographic and clinical information of patients with and without IR As shown in Table 1, the demographic and clinical variables were compared between patients with IR and patients without IR. The IR patients had significantly higher BMI levels, waist-to-hip ratio, FPG, FINS levels (p b .001, Bonferroni corrected p b .01), TG levels (p = .001), and LDL levels (p = .002, Bonferroni corrected p b .05) compared to non-IR patients. After controlling for the effects of age and sex using the ANOVA, significant differences were also found in BMI (F = 63.24, p b .001), and FPG (F = 42.24, p b .001), FINS (F = 106.30, p b .001), TG (F = 13.40, p b .001), and LDL (F = 9.41, p = .003) levels. There were no significant differences in age, sex, education, age at onset, antipsychotic dose (chlorpromazine equivalent), duration of illness, and PANSS score between IR patients and non-IR patients (p N .05). There was a significant difference in attention between the two groups (p b
C. Lin et al. / Comprehensive Psychiatry 96 (2020) 152145
3
Table 1 Demographic and clinical information of patients with or without IR.
Age (years) Sex Male (%) Female (%) Education (years) Age of onset (years) Duration of illness (months) Smokers /non-smokers BMI (kg/m2) Waist-to-hip ratio Antipsychotic Clozapine Olanzapine Risperidone Other drugs Antipsychotic dose (mg/day) FPG (mmol/L) FINS(mIU/L) CHO (mmol/L) TG (mmol/L) LDL (mmol/L) HDL (mmol/L) PANSS Positive subscore Negative subscore General psychopathology subscore PANSS total score
All patient (N = 193)
Patient without IR (N = 120)
Patient with IR (N = 73)
t or χ2
P value
50.37 ± 7.55
51.05 ± 6.80
49.26 ± 8.57
1.52 0.14
0.132 0.704
113(58.55%) 80(41.45%) 10.89 ± 2.52 25.68 ± 8.24 297.09 ± 118.38 52/141 25.13 ± 4.27 0.96 ± 0.07
69(57.5%) 51(42.5%) 10.73 ± 2.38 25.88 ± 8.71 301.98 ± 115.69 32/88 23.49 ± 3.61 0.94 ± 0.07
44(60.27%) 29(39.73%) 11.15 ± 2.73 25.37 ± 7.46 289.05 ± 123.05 20/53 27.83 ± 3.88 0.99 ± 0.07
−1.14 0.41 0.73 0.01 −7.87 −4.58 6.95
0.256 0.681 0.464 0.912 b0.001* b0.001 0.074
73(37.82%) 51(26.42%) 34(17.62%) 35(18.14%) 342.82 ± 178.40 5.18 ± 1.21 7.88 ± 5.24 4.17 ± 0.88 1.67 ± 0.95 2.37 ± 0.65 1.06 ± 0.27
48/120(40.00%) 35/120(29.17%) 22/120(18.33%) 15/120(12.50%) 347.88 ± 162.31 4.79 ± 0.84 4.46 ± 1.78 4.08 ± 0.86 1.47 ± 0.68 2.25 ± 0.62 1.08 ± 0.29
25/73(34.25%) 16/73(21.91%) 12/73(16.44%) 20/73(27.40%) 334.52 ± 203.01 5.83 ± 1.44 13.50 ± 9.12 4.32 ± 0.93 1.99 ± 1.23 2.57 ± 0.64 1.04 ± 0.27
0.50 −5.61 −8.38 −1.79 −3.31 −3.22 1.02
0.615 b0.001* b0.001* 0.075 0.001* 0.002* 0.310
16.3 ± 5.76 25.47 ± 6.07 35.22 ± 6.25 77 ± 13.69
16.36 ± 5.77 25.93 ± 6.10 35.57 ± 5.86 77.85 ± 13.01
16.19 ± 5.80 24.71 ± 6.00 34.64 ± 6.85 75.60 ± 14.72
0.19 1.36 0.99 1.11
0.846 0.176 0.321 0.270
Abbreviations: Dosage of antipsychotics refers to the dose converted to chlorpromazine; The other drugs included quetiapine fumarate, aripiprazole, and ziprasidone; insulin resistance (IR); positive and negative syndrome scale (PANSS); body mass index (BMI); fasting plasma glucose (FPG); fasting insulin (FINS); triglyceride (TG); low density lipoprotein (LDL); cholesterol (CHO); high-density lipoprotein (HDL); *: controlling for the age and sex, p value b.005.
.05). There were no significant differences in other cognitive function subscale and total RBANS scores (p N .05), as shown in Table 2.
3.3. Prevalence of IR and risk factors The prevalence of IR in patients was 37.82% (73/193). Using the binary logistic regression analysis, we found that smokers were 1.624 times more likely to suffer from IR compared to non-smokers (Wald x2 = 7.295, OR = 1.624, 95% CI: 1.142–2.309, p = .007) and that patients with a higher BMI level were more likely to develop IR (Wald x2 = 6.93, OR = 1.26, 95% CI: 1.06–1.49, p = .008). The other risk factors were TG (Wald x2 = 3.95, OR = 1.92, 95% CI: 1.01–3.66, p = .047) and LDL (Wald x2 = 5.90, OR = 11.43, 95% CI: 1.60–81.61, p = .015) levels, as shown in Table 3.
3.4. Correlation of IR index with demographic and clinical variables Correlation analysis showed that the IR index was correlated with the age (r = −0.145, p b .05), BMI (r = 0.613, p b .001), and waist-tohip ratio (r = 0.381, p b .001), and CHO (r = 0.185, p b .05), TG (r = 0.343, p b .005), and LDL (r = 0.257, p b .005) levels. However, after the Bonferroni correction, only the significant associations between IR and BMI, waist-to-hip ratio, and LDL level remained (p b .05). No significant correlation was found between the IR index and cognitive function subscale and RBANS total scores. Further, the IR indices were transformed into a normally distributed data by taking the logarithm of the indices. We then used the multivariable linear regression analysis to predict the correlation of the IR index with the demographic and clinical variables. The results indicated that the BMI and FPG level were still associated with the IR index, as shown in Table 4. 3.5. Metabolic information of patients treated with different types of antipsychotics
Table 2 Cognitive function of patients with or without IR.
Immediate memory Visuospatial Language Attention Delayed memory Total RBANS scores
All patient (N = 193)
Patient without IR (N = 120)
Patient with IR (N = 73)
t
66.87 ± 19.50 84.80 ± 19.52 83.96 ± 11.94 85.88 ± 17.33 76.08 ± 21.02 74.34 ± 16.33
66.03 ± 20.12
68.26 ± 18.48 87.19 ± 18.14 84.59 ± 11.82 89.29 ± 13.54 78.99 ± 17.93 76.38 ± 13.65
−0.77 0.441
83.35 ± 20.25 83.58 ± 12.04 83.81 ± 19.04 74.32 ± 22.58 73.09 ± 17.70
P value
−1.33 0.186
We divided all patients into four groups according to the types of drugs they were taking: the clozapine group, olanzapine group, risperidone group, and the other group. The types of drugs patients in the other group were taking mainly included quetiapine fumarate, Table 3 Predictors generated by Binary Logistic Regression with IR as dependent variables.
−0.57 0.568
Coefficients
Wald
P value
EXP(B)
−2.33 0.021 −1.59 0.114 −1.45 0.149
Abbreviations: Repeatable Battery for the Assessment of Neuropsychological Status (RBANS).
(Constant) SMOKE BMI TG LDL
B
Std. Error
−12.046 0.485 0.230 0.653 2.436
5.374 0.180 0.087 0.329 1.003
5.025 7.295 6.931 3.953 5.900
0.025 0.007 0.008 0.047 0.015
1.624 1.258 1.922 11.429
95.0% Confidence Interval for EXP(B) Lower
Upper
1.142 1.061 1.009 1.601
2.309 1.493 3.659 81.611
4
C. Lin et al. / Comprehensive Psychiatry 96 (2020) 152145
Table 4 Multivariable linear regression with IR as dependent variables. Coefficients
(Constant) FPG BMI
T
B
Std. Error
−2.781 0.598 0.157
1.055 0.098 0.028
−2.637 6.112 5.666
P value
0.009 b0.001 b0.001
95.0% Confidence Interval for EXP(B) Lower
Upper
−4.863 0.405 0.102
−0.699 0.791 0.211
Abbreviations: B: Regression coefficient T: The results of t-test for regression coefficients.
aripiprazole, and ziprasidone. After testing for homogeneity of variance using the univariate ANOVA Bonferroni method, we found that there were no significant differences in IR index, BMI, and waist-hip ratio, and the FPG, HDL, and TG levels (p N .05) among these groups. However, CHO and LDL levels (p b .05) were significantly different among the groups. Further pairwise analysis showed that the difference in LDL level between the clozapine and olanzapine groups was statistically significant (p = .007). The difference in CHO level between the clozapine and olanzapine groups was statistically significant (p = .029), and the differences in CHO and LDL levels between the other groups were not statistically significant (p N .05), as shown in Table 5. 4. Discussion In this study, a high prevalence of IR (37.82%) was found in patients with schizophrenia, which is similar to the results of majority of the previous studies [1–3]. We also found some demographic and clinical variables that were risk factors for IR in patients, including regular smoking, high BMI, and high TG and LDL levels. The sex, age, antipsychotics, severity of psychotic symptoms, and cognitive function had no effect on the risk of IR in patients with schizophrenia. This study found that the BMI had an effect on IR in patients with schizophrenia. In obese patients, the number of insulin receptors is reduced. In these patients, the insulin receptors are also defective. Therefore, IR and elevated fasting insulin level occur easily, which can affect glucose transport and utilization and protein synthesis. Obesity, excessive waistline, and lack of physical activity are important factors that promote the progress of insulin resistance [23]. When visceral fat increases, the adipose tissue secretes a higher amount of serum-free fatty acid (FFA) and fat-derived factors, including tumor necrosis factor alpha (TNF-α), interleukin-6 (IL-6), C-reactive protein (CRP), and angiotensinogen. The FFA secreted into liver cells inhibits insulin receptor signaling, leading to insulin resistance [24]. TNF-α promotes fat decomposition, increases FFA, and indirectly causes insulin resistance. It also affects glucose uptake by reducing glucose transporter protein-4 (GLUT-4) translocation [25]. Liu et al. compared fat metabolism in obese mice and lean mice, and their results showed that fat tissues in obese mice can produce large amounts of small fat cells, resulting in impaired cell differentiation and the activation of inflammatory factors, leading to IR. IR also increases the likelihood of the occurrence of obesity, and IR is associated with the risk factors for coronary heart disease [26]. A study has found that the BMI and waist circumference of patients with schizophrenia treated with clozapine are significantly correlated
with IR, but not olanzapine, which is partly consistent with our results [27]. In the study of Henderson DC, only 22 subjects were treated with olanzapine. A significant correlation was not found probably due to the small sample size in the aforementioned study. The stepwise multiple linear regression models used in the study of Henderson DC showed that regardless of whether the patient was treated with clozapine or olanzapine, a greater waist circumference predicted increased IR. Therefore, active weight control to reduce the BMI and waist circumference may be essential to reduce IR in patients with schizophrenia. In the present study, we found that the risk of IR among smoking patients with schizophrenia was 1.624 times higher than that among nonsmoking patients. A Chinese study showed that long-term hospitalized patients with schizophrenia who were smokers were more likely to develop metabolic syndrome than nonsmokers [28]. Smoking produces a large amount of free radicals and reactive oxygen species (ROS), and when the free radical scavenging capacity of the antioxidant system is exceeded, it can lead to an imbalance in the oxidation - antioxidant system in liver tissues and to oxidative stress, which can cause liver damage, reduce insulin clearance, and cause IR and hyperinsulinemia. The nicotine in tobacco can directly damage the islet beta cells and reduce the sensitivity of insulin receptors, leading to the deterioration of the IR and the occurrence of type 2 diabetes [29]. Therefore, medical workers and family members should actively control and manage the smoking behavior of patients with schizophrenia, strengthen health education, and reduce the number of cigarettes smoked by patients as much as possible. We found that there was no gender difference in the risk of IR in patients with schizophrenia. A study on metabolic syndrome in 503 inpatients with schizophrenia found that the prevalence of metabolic syndrome in female patients was higher than that in male patients, and the prevalence in patients over 40 years old was higher than that in patients under 39 years old. Dyslipidemia, obesity, and hyperglycemia are the most important risk factors and prediction indices for metabolic syndrome in schizophrenia [30]. Women are at a higher risk for weight gain after taking antipsychotic medications than men; they are about five times more likely to be overweight after treatment. Additionally, women meet the criteria for abdominal obesity for metabolic syndrome twice as often as men [31]. However, other studies have found that among long-term hospitalized patients with schizophrenia, men are more likely to develop metabolic syndrome than women [28], which is inconsistent with the two aforementioned studies. There is currently no consensus on whether men or women are at a higher metabolic risk for schizophrenia. We did not find an effect of age on the development of IR in patients with schizophrenia, which is inconsistent with our expectations. This may be due to the small sample size, which resulted in little age difference, or interference of confounding factors. It may also be because IR already existed before the onset of the disease. Metabolic abnormality is very common in patients with schizophrenia, and when 1–2 indicators are abnormal, the treatment plan should be adjusted quickly or early preventive intervention should be performed. Abnormal blood lipid levels (increased TG and LDL levels, and decreased HDL cholesterol levels) and obesity (increased waistline and BMI) are important predictors of metabolic syndrome (MS).
Table 5 Metabolic information of patients treated with different types of antipsychotics.
IR BMI Waist-hip ratio FPG CHO LDL TG HDL
Clozapine (n = 73)
Olanzapine (n = 51)
Risperidone (n = 34)
Other (n = 35)
F
P value
1.63 ± 1.44 24.65 ± 3.75 0.97 ± 0.08 5.14 ± 1.05 3.96 ± 0.92 2.20 ± 0.67 1.70 ± 0.61 1.00 ± 0.21
1.69 ± 1.46 25.27 ± 3.80 0.95 ± 0.07 4.96 ± 1.08 4.42 ± 0.97 2.60 ± 0.71 1.69 ± 0.97 1.06 ± 0.29
2.18 ± 2.82 24.64 ± 4.49 0.95 ± 0.07 5.24 ± 1.38 4.26 ± 0.63 2.34 ± 0.48 1.49 ± 0.70 1.14 ± 0.32
2.51 ± 1.89 26.41 ± 5.45 0.97 ± 0.07 5.57 ± 1.45 4.16 ± 0.85 2.41 ± 0.56 1.75 ± 1.57 1.13 ± 0.30
2.33 1.55 2.03 1.91 2.85 3.69 0.51 2.56
0.076 0.203 0.111 0.130 0.039 0.013 0.674 0.057
C. Lin et al. / Comprehensive Psychiatry 96 (2020) 152145
Increased levels of triglycerides have a greater effect than decreased levels of HDL cholesterol, and increased waist circumference has a greater effect than BMI [32]. Studies have shown that elevated blood lipid levels is associated with IR [33], and the ratio of TG to HDL levels is better than other blood lipid indicators (LDL, HDL, and TG levels) at predicting IR in patients with schizophrenia [34,35]. However, Anindya Dasgupta [36] conducted a study involving 30 newly diagnosed drugnaive patients with schizophrenia and found that IR was more prevalent in these patients than in the normal control group. The IR was not significantly correlated with dyslipidemia, suggesting that there may be other risk factors, including genetics, which contribute to the occurrence of IR in patients with schizophrenia. In this study, we found no correlation between IR and the dose of antipsychotics in patients with schizophrenia. There was also no significant difference in the IR index between the olanzapine, clozapine, and risperidone groups and the other drug groups, which is consistent with a Chinese study [28]. Many studies have found that the use of antipsychotic medications is associated with impaired glucose tolerance in patients with schizophrenia, including IR [37,38]. However, IR is also prevalent in patients with first-episode untreated schizophrenia [39]. The results of a meta-analysis suggested that patients with firstepisode drug-naive schizophrenia had elevated fasting plasma glucose levels, decreased glucose tolerance, elevated fasting plasma insulin levels and increased IR at the onset of the disease. The results of this meta-analysis indicated that there were changes in glucose homeostasis in patients with chronic schizophrenia from the onset of the disease [1]. A survey in Japan found that after three months of treating 100 patients with schizophrenia with either olanzapine or risperidone, the IR index of these patients was not significantly different from that of the normal control group. Additionally, these authors found no direct correlation between antipsychotic drugs and the risk of impaired glucose tolerance in patients with schizophrenia [40]. Therefore, the risk of IR, caused by antipsychotics in schizophrenia, may be lower than initially thought. In the present study, the severity of psychotic symptoms or cognitive function did not affect the risk of IR in patients with schizophrenia. Studies have shown that low cognitive function or severe negative symptoms may affect the ability of patients to manage their own blood glucose levels, and psychiatric symptoms and cognitive functions may affect the glucose metabolism of patients [41]. The severity of positive symptoms in first-episode untreated schizophrenic patients was negatively correlated with IR [3]. A study involving 224,743 people from 52 countries showed that psychotic symptoms are associated with increased prevalence of diabetes in nonclinical samples [42]. There is a common genetic predisposition for both schizophrenia and diabetes [43], which suggests that there may be a correlation between the severity of psychotic symptoms and glucose metabolism. However, we did not find evidence of this correlation in our study. This may be due to the incomplete evaluation of psychiatric symptoms and cognitive function. Although HOMA-IR is widely used in clinical and epidemiological studies, it is only a surrogate measure for insulin resistance levels, and it is less accurate compared to using a glucose clamp. More advanced measurements can be used when they become available in the future. Since our study is a cross-sectional study, we cannot demonstrate the causality of risk factors for IR in patients with schizophrenia. To reduce the influence of confounding factors and to obtain more evidence to confirm our results, longitudinal studies should be conducted using a large sample size and should involve first-episode and drug-naive patients. 5. Conclusions In summary, our results showed a high prevalence of IR in patients with schizophrenia. Smoking increases the risk of IR. The IR patients also had high BMI and FPG, TG, and LDL levels. Active lifestyle changes in patients with schizophrenia, such as weight and blood lipid control
5
and reduction in smoking, can help to reduce the incidence of IR in patients. These changes can improve the quality of life and decrease the prevalence of metabolic diseases such as diabetes and cardiovascular diseases. Conflict of Interest Disclosures No conflicts of interest was disclosed for the authors. Funding/Support This study was supported by Beijing Municipal Administration of Hospitals Incubating Program, code: PX2018066. Author contributions Chen Lin, Rongzhen Zhang, Weihong Fu, and Dachun Chen conceived and designed the study. Chen Lin, Ke Chen, Jianjin Yu, Lan Gao, Haiqing Ni, and Jiaqi Song recruited the patients, performed the clinical assessments and collected the clinical data. Chen Lin and Ke Chen were responsible for the statistical analysis and manuscript preparation. Chen Lin, Ke Chen, and Dachun Chen were responsible for writing the protocol, editing the paper and revising the manuscript. All authors contributed to the study and approved the final manuscript. References [1] Pillinger T, Beck K, Gobjila C, Donocik JG, Jauhar S, Howes OD. Impaired glucose homeostasis in first-episode schizophrenia: a systematic review and meta-analysis. JAMA Psychiat 2017;74(3):261–9. [2] Wu X, Huang Z, Wu R, et al. The comparison of glycometabolism parameters and lipid profiles between drug-naïve, first-episode schizophrenia patients and healthy controls. Schizophr Res 2013;150(1):157–62. [3] Chen S, Broqueres-You D, Yang G, et al. Relationship between insulin resistance, dyslipidaemia and positive symptom in Chinese antipsychotic-naive first-episode patients with schizophrenia. Psychiatry Res 2013;210(3):825–9. [4] Stern MP. Diabetes and cardiovascular disease. The “common soil” hypothesis. Diabetes 1995;44(4):369–74. [5] Luchsinger JA. Adiposity, hyperinsulinemia, diabetes and Alzheimer’s disease: an epidemiological perspective. Eur J Pharmacol 2008;585(1):119–29. [6] Gudala K, Bansal D, Schifano F, Bhansali A. Diabetes mellitus and risk of dementia: a meta-analysis of prospective observational studies. J Diabetes Investig 2013;4(6): 640–50. [7] Konig M, Lamos EM, Stein SA, Davis SN. An insight into the recent diabetes trials: what is the best approach to prevent macrovascular and microvascular complications. Curr Diabetes Rev 2013;9(5):371–81. [8] Pyörälä M, Miettinen H, Halonen P, Laakso M, Pyörälä K. Insulin resistance syndrome predicts the risk of coronary heart disease and stroke in healthy middle-aged men: the 22-year follow-up results of the Helsinki policemen study. Arterioscler Thromb Vasc Biol 2000;20(2):538–44. [9] Kahn SE, Cooper ME, Del PS. Pathophysiology and treatment of type 2 diabetes: perspectives on the past, present. And Future Lancet 2014;383(9922):1068–83. [10] Balf G, Stewart TD, Whitehead R, Baker RA. Metabolic adverse events in patients with mental illness treated with antipsychotics: a primary care perspective. Prim Care Companion J Clin Psychiatry 2008;10(1):15–24. [11] Perez-Iglesias R, Mata I, Pelayo-Teran JM, et al. Glucose and lipid disturbances after 1 year of antipsychotic treatment in a drug-naïve population. Schizophr Res 2009;107 (2–3):115–21. [12] Foley DL, Morley KI, Carroll KE, Moran J, McGorry PD, Murphy BP. Successful implementation of cardiometabolic monitoring of patients treated with antipsychotics. Med J Aust 2009;191(9):518–9 [author reply 519]. [13] Correll CU, Robinson DG, Schooler NR, et al. Cardiometabolic risk in patients with first-episode schizophrenia spectrum disorders: baseline results from the RAISEETP study. JAMA Psychiat 2014;71(12):1350–63. [14] Petrikis P, Tigas S, Tzallas AT, Papadopoulos I, Skapinakis P, Mavreas V. Parameters of glucose and lipid metabolism at the fasted state in drug-naïve first-episode patients with psychosis: evidence for insulin resistance. Psychiatry Res 2015;229(3):901–4. [15] Petruzzelli MG, Margari M, Peschechera A, et al. Hyperprolactinemia and insulin resistance in drug naive patients with early onset first episode psychosis. BMC Psychiatry 2018;18(1):246. [16] Zhang BH, Tan YL, Zhang WF, et al. Repeatable battery for the assessment of neuropsychological status as a screening test in Chinese:reliability and validity. Chinese Mental Health Journal 2008;22(12):865–9. [17] Kubota N, Terauchi Y, Yamauchi T, et al. Disruption of adiponectin causes insulin resistance and neointimal formation. J Biol Chem 2002;277(29):25863–6. [18] Zhang JQ. New HOMA-IR—insulin resistance measured from fasting glucose and fasting insulin. Chinese Journal of Diabetes Mellitus 2005;13(4):245–6.
6
C. Lin et al. / Comprehensive Psychiatry 96 (2020) 152145
[19] Wallace TM, Levy JC, Matthews DR. Use and abuse of HOMA modeling. Diabetes Care 2004;27(6):1487–95. [20] Tang Q, Li X, Song P, Xu L. Optimal cut-off values for the homeostasis model assessment of insulin resistance (HOMA-IR) and pre-diabetes screening: developments in research and prospects for the future. Drug Discov Ther 2015;9(6):380–5. [21] Gayoso-Diz P, Otero-González A, Rodriguez-Alvarez MX, et al. Insulin resistance (HOMA-IR) cut-off values and the metabolic syndrome in a general adult population: effect of gender and age: EPIRCE cross-sectional study. BMC Endocr Disord 2013;13(47. [22] Esteghamati A, Ashraf H, Esteghamati AR, et al. Optimal threshold of homeostasis model assessment for insulin resistance in an Iranian population: the implication of metabolic syndrome to detect insulin resistance. Diabetes Res Clin Pract 2009; 84(3):279–87. [23] Alberti KG, Eckel RH, Grundy SM, et al. Harmonizing the metabolic syndrome: a joint interim statement of the international diabetes federation task force on epidemiology and prevention; National Heart, Lung, and Blood Institute; American Heart Association; world heart federation; international atherosclerosis society; and International Association for the Study of obesity. Circulation 2009;120(16):1640–5. [24] Pereira S, Park E, Mori Y, et al. FFA-induced hepatic insulin resistance in vivo is mediated by PKCδ, NADPH oxidase, and oxidative stress. Am J Physiol Endocrinol Metab 2014;307(1):E34–46. [25] Stephens JM, Lee J, Pilch PF. Tumor necrosis factor-alpha-induced insulin resistance in 3T3-L1 adipocytes is accompanied by a loss of insulin receptor substrate-1 and GLUT4 expression without a loss of insulin receptor-mediated signal transduction. J Biol Chem 1997;272(2):971–6. [26] Liu A, Sonmez A, Yee G, et al. Differential adipogenic and inflammatory properties of small adipocytes in Zucker obese and lean rats. Diab Vasc Dis Res 2010;7(4):311–8. [27] Henderson DC, Fan X, Sharma B, et al. Waist circumference is the best anthropometric predictor for insulin resistance in nondiabetic patients with schizophrenia treated with clozapine but not olanzapine. J Psychiatr Pract 2009;15(4):251–61. [28] Qiao Y, Wen H, Chen HY, et al. The study of risk for metabolic syndromes in longterm hospitalized patients with schizophrenia. Shanghai Arch Psychiatry 2008;20 (2):65–8. [29] Institute of Medicine (US) Committee on Secondhand Smoke Exposure and Acute Coronary Events Secondhand Smoke Exposure and Cardiovascular Effects: Making Sense of the Evidence. Washington (DC): National Academies Press (US); 2010. [30] Cheng ZH, Ling WM, Zhou XQ, Ji Q. Prevalence of metabolic syndrome in patients with schizophrenia and its related factors. Chinese journal of behavioral medicine and brain. Science 2011;20(11):1012–5. [31] Smith S. Gender differences in antipsychotic prescribing. Int Rev Psychiatry 2010;22 (5):472–84.
[32] Koponen HJ, Hakko HH, Saari KM, et al. The prevalence and predictive value of individual criteria for metabolic syndrome in schizophrenia: a northern Finland 1966 birth cohort study. World J Biol Psychiatry 2010;11(2 Pt 2):262–7. [33] Laws A, Reaven GM. Evidence for an independent relationship between insulin resistance and fasting plasma HDL-cholesterol, triglyceride and insulin concentrations. J Intern Med 1992;231(1):25–30. [34] McLaughlin T, Reaven G, Abbasi F, et al. Is there a simple way to identify insulinresistant individuals at increased risk of cardiovascular disease. Am J Cardiol 2005; 96(3):399–404. [35] Fan X, Liu EY, Hoffman VP, Potts AJ, Sharma B, Henderson DC. Triglyceride/highdensity lipoprotein cholesterol ratio: a surrogate to predict insulin resistance and low-density lipoprotein cholesterol particle size in nondiabetic patients with schizophrenia. J Clin Psychiatry 2011;72(6):806–12. [36] Dasgupta A, Singh OP, Rout JK, Saha T, Mandal S. Insulin resistance and metabolic profile in antipsychotic naïve schizophrenia patients. Prog Neuropsychopharmacol Biol Psychiatry 2010;34(7):1202–7. [37] Smith RC, Lindenmayer JP, Davis JM, et al. Effects of olanzapine and risperidone on glucose metabolism and insulin sensitivity in chronic schizophrenic patients with long-term antipsychotic treatment: a randomized 5-month study. J Clin Psychiatry 2009;70(11):1501–13. [38] Chen FB, Sun J, Yan TJ, Xu LP, Wang HL. The liver fat content and its relationships with insulin resistance in schizophrenia patients receiving olanzapine or aripiprazol:a randomized comparative study. Chinese Journal of Nervous and Mental Diseases 2015;41(11):646–50. [39] Perry BI, McIntosh G, Weich S, Singh S, Rees K. The association between first-episode psychosis and abnormal glycaemic control: systematic review and meta-analysis. Lancet Psychiatry 2016;3(11):1049–58. [40] Yasui-Furukori N, Sato Y, Furukori H, Saito M, Nakagami T, Kaneko S. Glucose metabolism in Japanese schizophrenia patients treated with risperidone or olanzapine. J Clin Psychiatry 2009;70(1):95–100. [41] Wykes TL, Lee AA, McKibbin CL, Laurent SM. Self-efficacy and hemoglobin A1C among adults with serious mental illness and type 2 diabetes: the roles of cognitive functioning and psychiatric symptom severity. Psychosom Med 2016;78(3):263–70. [42] Nuevo R, Chatterji S, Fraguas D, et al. Increased risk of diabetes mellitus among persons with psychotic symptoms: results from the WHO world health survey. J Clin Psychiatry 2011;72(12):1592–9. [43] Lin PI, Shuldiner AR. Rethinking the genetic basis for comorbidity of schizophrenia and type 2 diabetes. Schizophr Res 2010;123(2–3):234–43.
Contents lists available at ScienceDirect
Comprehensive Psychiatry journal homepage: www.elsevier.com/locate/comppsych
The prevalence, risk factors, and clinical characteristics of insulin resistance in Chinese patients with schizophrenia Chen Lin b, Ke Chen a, Rongzhen Zhang a, Weihong Fu a, Jianjin Yu a, Lan Gao a, Haiqing Ni c, Jiaqi Song d, Dachun Chen a,⁎ a
Beijing HuiLongGuan Hospital, Beijing, 100096, PR China Department of Psychosomatic Medicine, Beijing HuiLongGuan Hospital, Peking University, Beijing, 100096, PR China Guangzhou Brain Hospital, Guangzhou, 510450, PR China d Peking University Health Science Center, Beijing, 100191, PR China b c
a r t i c l e Available online xxxx
Keywords: Schizophrenia Insulin resistance Prevalence Risk factors
i n f o
a b s t r a c t Background: Studies have shown that patients with schizophrenia are at a high risk of developing insulin resistance (IR). We investigated the prevalence of IR and its clinical correlates in hospitalized Chinese patients with schizophrenia. Methods: A total of 193 patients with schizophrenia (113 males and 80 females) were recruited for the study. We collected their demographic and clinical data, including data on their plasma glucose and lipid levels. All patients were rated using the Repeatable Battery for the Assessment of Neuropsychological Status (RBANS) to assess cognitive function, while Positive and Negative Syndrome Scale (PANSS) was used to assess psychopathology. The cut-off value for the homeostasis model assessment of insulin resistance (HOMA-IR) was set at 1.7. Results: The prevalence of IR was 37.82% (73/193). The IR patients had significantly higher waist-to-hip ratio and body mass index (BMI), and higher fasting plasma glucose (FPG), triglyceride (TG), and low density lipoprotein (LDL) levels compared to non-IR patients (all p b .05). Binary logistic regression analysis showed that smoking, BMI, and TG and LDL levels are significant predictors of IR. In addition, correlation analysis showed that IR was significantly correlated with the waist-to-hip ratio, BMI, and LDL level (Bonferroni corrected p b .05). The multivariable linear regression analysis indicated that the BMI and FPG are associated with the IR index. There was no significant difference in IR index between patients who were taking different antipsychotics. Conclusion: We found a high prevalence of IR and its risk factors in Chinese patients with schizophrenia. Active weight control to reduce the BMI and waist circumference and reducing the number of cigarettes consumed, may be essential to decrease the incidence of IR in patients with schizophrenia. © 2019 Published by Elsevier Inc. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
1. Introduction Many studies have shown that insulin resistance (IR) is more prevalent in patients with schizophrenia than in healthy controls [1–3]. IR, an important cause of metabolic syndrome, significantly affects physical and mental health. As early as 1995, Stern [4] proposed the “common soil theory.” He believed that IR was the link between many diseases and that it was the original cause and pathogenic basis of many metabolic abnormalities and cardiovascular diseases. Recently some studies have shown that type 2 diabetes and IR are risk factors for cognitive dysfunction and dementia [5,6]. IR, one of the main pathogenic factors of type 2 diabetes, is an independent risk factor for coronary heart disease [7]. A study with an average follow-up of ⁎ Corresponding author at: Fifth Clinical Department, Beijing HuiLongGuan Hospital, Peking University, Beijing, 100096, PR China. E-mail address: [email protected] (D. Chen).
22 years showed that the relative risk of cerebral infarction in IR patients was 2.12 times higher than that in non-IR patients [8]. IR greatly reduces the quality of life and average life expectancy of patients. The causes of IR have not been fully elucidated. Possible risk factors for IR in the general population include obesity, poor diet, lack of exercise, age, and genetic predisposition [9]. For patients with schizophrenia, many commonly used antipsychotics (e.g., quetiapine and olanzapine) have been shown to be associated with IR and metabolic syndrome [10–13]. However, there is also a high prevalence of IR in first-episode drug-naïve patients with schizophrenia [1,14,15], suggesting that IR in schizophrenia may be caused by factors other than drugs. Presently, the risk factors for IR and their clinical correlations in Chinese patients with schizophrenia have not been fully investigated. The major objectives of our study were to investigate: (1) the prevalence, risk factors, and clinical correlates of IR in Chinese patients with schizophrenia; (2) the correlation between IR index and demographic and clinical variables, psychopathology, and cognitive function; and
https://doi.org/10.1016/j.comppsych.2019.152145 0010-440X/© 2019 Published by Elsevier Inc. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
2
C. Lin et al. / Comprehensive Psychiatry 96 (2020) 152145
(3) the differences in IR and other metabolic parameters in patients treated with different types of antipsychotics. 1. Subjects and Methods 1.1. Subjects A total of 193 patients with schizophrenia were recruited from June 2015 to December 2018 at the Beijing HuiLongGuan Hospital. Only patients who fulfilled the Diagnostic and Statistical Manual of Mental Disorders, fifth edition (DSM-V) criteria for schizophrenia, were between the ages of 18 and 60 years, and were of the Han nationality were included in the study. These group of patients were excluded from the study: (1) patients with obvious central nervous system diseases, such as stroke, tumor, dementia, Parkinson's disease, and epilepsy; (2) patients with serious physical diseases (except diabetes), such as hypertension and heart disease, among others; (3) patients with drug or alcohol dependence; (4) pregnant or lactating patients; and (5) patients with obvious mental retardation and could not complete the scale and cognitive function tests. This study was approved by the ethics committee of the Beijing HuiLongGuan Hospital. All subjects provided informed consent. 2. Methods 2.1.1. Demographic variables The demographic data (gender, age, and education level) of patients with schizophrenia who met the inclusion criteria were collected in detail using the homemade basic information collection table. Weight, height, and waist and hip circumference were measured, and the body mass index (BMI) (weight (kg) divided by the height (m)2) was calculated. Data on medications (atypical antipsychotics) which patients had been taking for approximately six months, prior to their inclusion in the study, were collected. Smokers were classified as: 1) regular smokers: people who smoke more than one cigarette per day; and 2) occasional smokers: people who smoke more than four times a week, but b1 cigarette per day on average, according to the World Health Organization recommendations. In this study, only regular smokers, and not occasional smokers, were included in the smoking group. All patients were hospitalized with a unified management of diet and exercise. 2.1.2. Psychiatric symptom assessment The Positive and Negative Syndrome Scale (PANSS), which was developed by Kay et al. in 1987, is mainly used to evaluate the psychotic symptoms of patients. These symptoms comprise of seven positive symptoms, seven negative symptoms, 16 general pathological symptoms, and three additional symptoms. The PANSS consists of 33 items. It is scored on a scale of 1–7. The last three additional symptoms were not included in the total score. The higher the total score, the more severe the symptoms of the patients. The PANSS is relatively comprehensive in its assessment of the psychiatric symptoms of patients. The PANSS can reflect the overall clinical phase of patients. 2.1.3. Cognitive function assessment For the cognitive function assessment, the Repeatable Battery for the Assessment of Neuropsychological Status (RBANS) [16] was used. This RBANS consists of 12 items: list learning, story memory, figure copy, line orientation, picture naming, semantic fluency, digit span, coding test, list recall, list recognition, story recall, and figure recall. This cognitive assessment can be summarized into five groups of neuropsychological status (5 factors): immediate memory, visuospatial, language, attention, and delayed memory. The higher the score, the better the cognitive function. Before the study commenced, all scales were tested for consistency. The intraclass correlation coefficient (ICC) was N0.8.
2.1.4. Blood samples Patients avoided greasy foods the day before blood sampling. Five milliliters of venous blood was collected from inpatients at 6:00 a.m., following an overnight fast (avoided food for N10 h) and was stored at −40 °C. An automatic biochemical analyzer (Beckman AU5800) was used to measure the fasting plasma glucose (FPG), insulin, cholesterol (CHO), triglycerides (TG), high-density lipoprotein (HDL), and low density lipoprotein (LDL) levels. 2.1.5. HOMA-IR calculation The insulin resistance index (HOMA-IR) = (fasting plasma glucose (FPG, mmol/L) * fasting insulin (FINS, mIU/L)) /22.5. The standard of HOMA-IR for normal-weight normal subjects was set to 1 [17,18]. Although this index is mainly used in the research field, the HOMA-IR assesses an individual's level of IR. The results correlate with those of standard tests for IR, such as tests involving the use of the insulin clamp [19]. After taking the Metabolic Syndrome components into account, the threshold value of the HOMA-IR was set at 1.7 [20–22]. Patients with HOMA-IR values lower than 1.7 were assigned to the nonIR group, and patients with HOMA-IR values N1.7 were assigned to the IR group. 2.1.6. Statistical analysis Epidata 3.0 was used to create a database for all the data. The SPSS software version 18.0 was used for statistical analysis. The demographic and clinical variables were compared between the different groups using the independent sample t-test or analysis of variance (ANOVA) for continuous variables and chi-squared or Fisher's exact tests for categorical variables. We described the prevalence of IR using percentages. The prevalence was analyzed using the chi-square test, and the risk factors for IR were assessed using a binary logistic regression analysis. The Pearson or Spearman correlation analysis was used to examine the relationship between IR and demographic and clinical variables. Finally, the multivariable linear regression analysis was used to investigate significant predictive variables associated with the IR index. Bonferroni correction was applied to adjust for multiple testing. All p-values were two-tailed and significance was set at p b .05. 3. Results 3.1. Demographic data and clinical characteristics In total, 193 patients (113 males and 80 females) were enrolled in the study. The mean age of patients was 50.37 ± 7.55 years (range: 25–60 years). The average duration of education was 10.89 ± 2.52 years (range: 5–20 years). The mean duration of illness was 297.09 ± 118.38 months (range: 6–516 months). The average age at onset was 25.68 ± 8.24 years (range: 13–55 years). Fifty-two patients (26.94%) were smokers. 3.2. Demographic and clinical information of patients with and without IR As shown in Table 1, the demographic and clinical variables were compared between patients with IR and patients without IR. The IR patients had significantly higher BMI levels, waist-to-hip ratio, FPG, FINS levels (p b .001, Bonferroni corrected p b .01), TG levels (p = .001), and LDL levels (p = .002, Bonferroni corrected p b .05) compared to non-IR patients. After controlling for the effects of age and sex using the ANOVA, significant differences were also found in BMI (F = 63.24, p b .001), and FPG (F = 42.24, p b .001), FINS (F = 106.30, p b .001), TG (F = 13.40, p b .001), and LDL (F = 9.41, p = .003) levels. There were no significant differences in age, sex, education, age at onset, antipsychotic dose (chlorpromazine equivalent), duration of illness, and PANSS score between IR patients and non-IR patients (p N .05). There was a significant difference in attention between the two groups (p b
C. Lin et al. / Comprehensive Psychiatry 96 (2020) 152145
3
Table 1 Demographic and clinical information of patients with or without IR.
Age (years) Sex Male (%) Female (%) Education (years) Age of onset (years) Duration of illness (months) Smokers /non-smokers BMI (kg/m2) Waist-to-hip ratio Antipsychotic Clozapine Olanzapine Risperidone Other drugs Antipsychotic dose (mg/day) FPG (mmol/L) FINS(mIU/L) CHO (mmol/L) TG (mmol/L) LDL (mmol/L) HDL (mmol/L) PANSS Positive subscore Negative subscore General psychopathology subscore PANSS total score
All patient (N = 193)
Patient without IR (N = 120)
Patient with IR (N = 73)
t or χ2
P value
50.37 ± 7.55
51.05 ± 6.80
49.26 ± 8.57
1.52 0.14
0.132 0.704
113(58.55%) 80(41.45%) 10.89 ± 2.52 25.68 ± 8.24 297.09 ± 118.38 52/141 25.13 ± 4.27 0.96 ± 0.07
69(57.5%) 51(42.5%) 10.73 ± 2.38 25.88 ± 8.71 301.98 ± 115.69 32/88 23.49 ± 3.61 0.94 ± 0.07
44(60.27%) 29(39.73%) 11.15 ± 2.73 25.37 ± 7.46 289.05 ± 123.05 20/53 27.83 ± 3.88 0.99 ± 0.07
−1.14 0.41 0.73 0.01 −7.87 −4.58 6.95
0.256 0.681 0.464 0.912 b0.001* b0.001 0.074
73(37.82%) 51(26.42%) 34(17.62%) 35(18.14%) 342.82 ± 178.40 5.18 ± 1.21 7.88 ± 5.24 4.17 ± 0.88 1.67 ± 0.95 2.37 ± 0.65 1.06 ± 0.27
48/120(40.00%) 35/120(29.17%) 22/120(18.33%) 15/120(12.50%) 347.88 ± 162.31 4.79 ± 0.84 4.46 ± 1.78 4.08 ± 0.86 1.47 ± 0.68 2.25 ± 0.62 1.08 ± 0.29
25/73(34.25%) 16/73(21.91%) 12/73(16.44%) 20/73(27.40%) 334.52 ± 203.01 5.83 ± 1.44 13.50 ± 9.12 4.32 ± 0.93 1.99 ± 1.23 2.57 ± 0.64 1.04 ± 0.27
0.50 −5.61 −8.38 −1.79 −3.31 −3.22 1.02
0.615 b0.001* b0.001* 0.075 0.001* 0.002* 0.310
16.3 ± 5.76 25.47 ± 6.07 35.22 ± 6.25 77 ± 13.69
16.36 ± 5.77 25.93 ± 6.10 35.57 ± 5.86 77.85 ± 13.01
16.19 ± 5.80 24.71 ± 6.00 34.64 ± 6.85 75.60 ± 14.72
0.19 1.36 0.99 1.11
0.846 0.176 0.321 0.270
Abbreviations: Dosage of antipsychotics refers to the dose converted to chlorpromazine; The other drugs included quetiapine fumarate, aripiprazole, and ziprasidone; insulin resistance (IR); positive and negative syndrome scale (PANSS); body mass index (BMI); fasting plasma glucose (FPG); fasting insulin (FINS); triglyceride (TG); low density lipoprotein (LDL); cholesterol (CHO); high-density lipoprotein (HDL); *: controlling for the age and sex, p value b.005.
.05). There were no significant differences in other cognitive function subscale and total RBANS scores (p N .05), as shown in Table 2.
3.3. Prevalence of IR and risk factors The prevalence of IR in patients was 37.82% (73/193). Using the binary logistic regression analysis, we found that smokers were 1.624 times more likely to suffer from IR compared to non-smokers (Wald x2 = 7.295, OR = 1.624, 95% CI: 1.142–2.309, p = .007) and that patients with a higher BMI level were more likely to develop IR (Wald x2 = 6.93, OR = 1.26, 95% CI: 1.06–1.49, p = .008). The other risk factors were TG (Wald x2 = 3.95, OR = 1.92, 95% CI: 1.01–3.66, p = .047) and LDL (Wald x2 = 5.90, OR = 11.43, 95% CI: 1.60–81.61, p = .015) levels, as shown in Table 3.
3.4. Correlation of IR index with demographic and clinical variables Correlation analysis showed that the IR index was correlated with the age (r = −0.145, p b .05), BMI (r = 0.613, p b .001), and waist-tohip ratio (r = 0.381, p b .001), and CHO (r = 0.185, p b .05), TG (r = 0.343, p b .005), and LDL (r = 0.257, p b .005) levels. However, after the Bonferroni correction, only the significant associations between IR and BMI, waist-to-hip ratio, and LDL level remained (p b .05). No significant correlation was found between the IR index and cognitive function subscale and RBANS total scores. Further, the IR indices were transformed into a normally distributed data by taking the logarithm of the indices. We then used the multivariable linear regression analysis to predict the correlation of the IR index with the demographic and clinical variables. The results indicated that the BMI and FPG level were still associated with the IR index, as shown in Table 4. 3.5. Metabolic information of patients treated with different types of antipsychotics
Table 2 Cognitive function of patients with or without IR.
Immediate memory Visuospatial Language Attention Delayed memory Total RBANS scores
All patient (N = 193)
Patient without IR (N = 120)
Patient with IR (N = 73)
t
66.87 ± 19.50 84.80 ± 19.52 83.96 ± 11.94 85.88 ± 17.33 76.08 ± 21.02 74.34 ± 16.33
66.03 ± 20.12
68.26 ± 18.48 87.19 ± 18.14 84.59 ± 11.82 89.29 ± 13.54 78.99 ± 17.93 76.38 ± 13.65
−0.77 0.441
83.35 ± 20.25 83.58 ± 12.04 83.81 ± 19.04 74.32 ± 22.58 73.09 ± 17.70
P value
−1.33 0.186
We divided all patients into four groups according to the types of drugs they were taking: the clozapine group, olanzapine group, risperidone group, and the other group. The types of drugs patients in the other group were taking mainly included quetiapine fumarate, Table 3 Predictors generated by Binary Logistic Regression with IR as dependent variables.
−0.57 0.568
Coefficients
Wald
P value
EXP(B)
−2.33 0.021 −1.59 0.114 −1.45 0.149
Abbreviations: Repeatable Battery for the Assessment of Neuropsychological Status (RBANS).
(Constant) SMOKE BMI TG LDL
B
Std. Error
−12.046 0.485 0.230 0.653 2.436
5.374 0.180 0.087 0.329 1.003
5.025 7.295 6.931 3.953 5.900
0.025 0.007 0.008 0.047 0.015
1.624 1.258 1.922 11.429
95.0% Confidence Interval for EXP(B) Lower
Upper
1.142 1.061 1.009 1.601
2.309 1.493 3.659 81.611
4
C. Lin et al. / Comprehensive Psychiatry 96 (2020) 152145
Table 4 Multivariable linear regression with IR as dependent variables. Coefficients
(Constant) FPG BMI
T
B
Std. Error
−2.781 0.598 0.157
1.055 0.098 0.028
−2.637 6.112 5.666
P value
0.009 b0.001 b0.001
95.0% Confidence Interval for EXP(B) Lower
Upper
−4.863 0.405 0.102
−0.699 0.791 0.211
Abbreviations: B: Regression coefficient T: The results of t-test for regression coefficients.
aripiprazole, and ziprasidone. After testing for homogeneity of variance using the univariate ANOVA Bonferroni method, we found that there were no significant differences in IR index, BMI, and waist-hip ratio, and the FPG, HDL, and TG levels (p N .05) among these groups. However, CHO and LDL levels (p b .05) were significantly different among the groups. Further pairwise analysis showed that the difference in LDL level between the clozapine and olanzapine groups was statistically significant (p = .007). The difference in CHO level between the clozapine and olanzapine groups was statistically significant (p = .029), and the differences in CHO and LDL levels between the other groups were not statistically significant (p N .05), as shown in Table 5. 4. Discussion In this study, a high prevalence of IR (37.82%) was found in patients with schizophrenia, which is similar to the results of majority of the previous studies [1–3]. We also found some demographic and clinical variables that were risk factors for IR in patients, including regular smoking, high BMI, and high TG and LDL levels. The sex, age, antipsychotics, severity of psychotic symptoms, and cognitive function had no effect on the risk of IR in patients with schizophrenia. This study found that the BMI had an effect on IR in patients with schizophrenia. In obese patients, the number of insulin receptors is reduced. In these patients, the insulin receptors are also defective. Therefore, IR and elevated fasting insulin level occur easily, which can affect glucose transport and utilization and protein synthesis. Obesity, excessive waistline, and lack of physical activity are important factors that promote the progress of insulin resistance [23]. When visceral fat increases, the adipose tissue secretes a higher amount of serum-free fatty acid (FFA) and fat-derived factors, including tumor necrosis factor alpha (TNF-α), interleukin-6 (IL-6), C-reactive protein (CRP), and angiotensinogen. The FFA secreted into liver cells inhibits insulin receptor signaling, leading to insulin resistance [24]. TNF-α promotes fat decomposition, increases FFA, and indirectly causes insulin resistance. It also affects glucose uptake by reducing glucose transporter protein-4 (GLUT-4) translocation [25]. Liu et al. compared fat metabolism in obese mice and lean mice, and their results showed that fat tissues in obese mice can produce large amounts of small fat cells, resulting in impaired cell differentiation and the activation of inflammatory factors, leading to IR. IR also increases the likelihood of the occurrence of obesity, and IR is associated with the risk factors for coronary heart disease [26]. A study has found that the BMI and waist circumference of patients with schizophrenia treated with clozapine are significantly correlated
with IR, but not olanzapine, which is partly consistent with our results [27]. In the study of Henderson DC, only 22 subjects were treated with olanzapine. A significant correlation was not found probably due to the small sample size in the aforementioned study. The stepwise multiple linear regression models used in the study of Henderson DC showed that regardless of whether the patient was treated with clozapine or olanzapine, a greater waist circumference predicted increased IR. Therefore, active weight control to reduce the BMI and waist circumference may be essential to reduce IR in patients with schizophrenia. In the present study, we found that the risk of IR among smoking patients with schizophrenia was 1.624 times higher than that among nonsmoking patients. A Chinese study showed that long-term hospitalized patients with schizophrenia who were smokers were more likely to develop metabolic syndrome than nonsmokers [28]. Smoking produces a large amount of free radicals and reactive oxygen species (ROS), and when the free radical scavenging capacity of the antioxidant system is exceeded, it can lead to an imbalance in the oxidation - antioxidant system in liver tissues and to oxidative stress, which can cause liver damage, reduce insulin clearance, and cause IR and hyperinsulinemia. The nicotine in tobacco can directly damage the islet beta cells and reduce the sensitivity of insulin receptors, leading to the deterioration of the IR and the occurrence of type 2 diabetes [29]. Therefore, medical workers and family members should actively control and manage the smoking behavior of patients with schizophrenia, strengthen health education, and reduce the number of cigarettes smoked by patients as much as possible. We found that there was no gender difference in the risk of IR in patients with schizophrenia. A study on metabolic syndrome in 503 inpatients with schizophrenia found that the prevalence of metabolic syndrome in female patients was higher than that in male patients, and the prevalence in patients over 40 years old was higher than that in patients under 39 years old. Dyslipidemia, obesity, and hyperglycemia are the most important risk factors and prediction indices for metabolic syndrome in schizophrenia [30]. Women are at a higher risk for weight gain after taking antipsychotic medications than men; they are about five times more likely to be overweight after treatment. Additionally, women meet the criteria for abdominal obesity for metabolic syndrome twice as often as men [31]. However, other studies have found that among long-term hospitalized patients with schizophrenia, men are more likely to develop metabolic syndrome than women [28], which is inconsistent with the two aforementioned studies. There is currently no consensus on whether men or women are at a higher metabolic risk for schizophrenia. We did not find an effect of age on the development of IR in patients with schizophrenia, which is inconsistent with our expectations. This may be due to the small sample size, which resulted in little age difference, or interference of confounding factors. It may also be because IR already existed before the onset of the disease. Metabolic abnormality is very common in patients with schizophrenia, and when 1–2 indicators are abnormal, the treatment plan should be adjusted quickly or early preventive intervention should be performed. Abnormal blood lipid levels (increased TG and LDL levels, and decreased HDL cholesterol levels) and obesity (increased waistline and BMI) are important predictors of metabolic syndrome (MS).
Table 5 Metabolic information of patients treated with different types of antipsychotics.
IR BMI Waist-hip ratio FPG CHO LDL TG HDL
Clozapine (n = 73)
Olanzapine (n = 51)
Risperidone (n = 34)
Other (n = 35)
F
P value
1.63 ± 1.44 24.65 ± 3.75 0.97 ± 0.08 5.14 ± 1.05 3.96 ± 0.92 2.20 ± 0.67 1.70 ± 0.61 1.00 ± 0.21
1.69 ± 1.46 25.27 ± 3.80 0.95 ± 0.07 4.96 ± 1.08 4.42 ± 0.97 2.60 ± 0.71 1.69 ± 0.97 1.06 ± 0.29
2.18 ± 2.82 24.64 ± 4.49 0.95 ± 0.07 5.24 ± 1.38 4.26 ± 0.63 2.34 ± 0.48 1.49 ± 0.70 1.14 ± 0.32
2.51 ± 1.89 26.41 ± 5.45 0.97 ± 0.07 5.57 ± 1.45 4.16 ± 0.85 2.41 ± 0.56 1.75 ± 1.57 1.13 ± 0.30
2.33 1.55 2.03 1.91 2.85 3.69 0.51 2.56
0.076 0.203 0.111 0.130 0.039 0.013 0.674 0.057
C. Lin et al. / Comprehensive Psychiatry 96 (2020) 152145
Increased levels of triglycerides have a greater effect than decreased levels of HDL cholesterol, and increased waist circumference has a greater effect than BMI [32]. Studies have shown that elevated blood lipid levels is associated with IR [33], and the ratio of TG to HDL levels is better than other blood lipid indicators (LDL, HDL, and TG levels) at predicting IR in patients with schizophrenia [34,35]. However, Anindya Dasgupta [36] conducted a study involving 30 newly diagnosed drugnaive patients with schizophrenia and found that IR was more prevalent in these patients than in the normal control group. The IR was not significantly correlated with dyslipidemia, suggesting that there may be other risk factors, including genetics, which contribute to the occurrence of IR in patients with schizophrenia. In this study, we found no correlation between IR and the dose of antipsychotics in patients with schizophrenia. There was also no significant difference in the IR index between the olanzapine, clozapine, and risperidone groups and the other drug groups, which is consistent with a Chinese study [28]. Many studies have found that the use of antipsychotic medications is associated with impaired glucose tolerance in patients with schizophrenia, including IR [37,38]. However, IR is also prevalent in patients with first-episode untreated schizophrenia [39]. The results of a meta-analysis suggested that patients with firstepisode drug-naive schizophrenia had elevated fasting plasma glucose levels, decreased glucose tolerance, elevated fasting plasma insulin levels and increased IR at the onset of the disease. The results of this meta-analysis indicated that there were changes in glucose homeostasis in patients with chronic schizophrenia from the onset of the disease [1]. A survey in Japan found that after three months of treating 100 patients with schizophrenia with either olanzapine or risperidone, the IR index of these patients was not significantly different from that of the normal control group. Additionally, these authors found no direct correlation between antipsychotic drugs and the risk of impaired glucose tolerance in patients with schizophrenia [40]. Therefore, the risk of IR, caused by antipsychotics in schizophrenia, may be lower than initially thought. In the present study, the severity of psychotic symptoms or cognitive function did not affect the risk of IR in patients with schizophrenia. Studies have shown that low cognitive function or severe negative symptoms may affect the ability of patients to manage their own blood glucose levels, and psychiatric symptoms and cognitive functions may affect the glucose metabolism of patients [41]. The severity of positive symptoms in first-episode untreated schizophrenic patients was negatively correlated with IR [3]. A study involving 224,743 people from 52 countries showed that psychotic symptoms are associated with increased prevalence of diabetes in nonclinical samples [42]. There is a common genetic predisposition for both schizophrenia and diabetes [43], which suggests that there may be a correlation between the severity of psychotic symptoms and glucose metabolism. However, we did not find evidence of this correlation in our study. This may be due to the incomplete evaluation of psychiatric symptoms and cognitive function. Although HOMA-IR is widely used in clinical and epidemiological studies, it is only a surrogate measure for insulin resistance levels, and it is less accurate compared to using a glucose clamp. More advanced measurements can be used when they become available in the future. Since our study is a cross-sectional study, we cannot demonstrate the causality of risk factors for IR in patients with schizophrenia. To reduce the influence of confounding factors and to obtain more evidence to confirm our results, longitudinal studies should be conducted using a large sample size and should involve first-episode and drug-naive patients. 5. Conclusions In summary, our results showed a high prevalence of IR in patients with schizophrenia. Smoking increases the risk of IR. The IR patients also had high BMI and FPG, TG, and LDL levels. Active lifestyle changes in patients with schizophrenia, such as weight and blood lipid control
5
and reduction in smoking, can help to reduce the incidence of IR in patients. These changes can improve the quality of life and decrease the prevalence of metabolic diseases such as diabetes and cardiovascular diseases. Conflict of Interest Disclosures No conflicts of interest was disclosed for the authors. Funding/Support This study was supported by Beijing Municipal Administration of Hospitals Incubating Program, code: PX2018066. Author contributions Chen Lin, Rongzhen Zhang, Weihong Fu, and Dachun Chen conceived and designed the study. Chen Lin, Ke Chen, Jianjin Yu, Lan Gao, Haiqing Ni, and Jiaqi Song recruited the patients, performed the clinical assessments and collected the clinical data. Chen Lin and Ke Chen were responsible for the statistical analysis and manuscript preparation. Chen Lin, Ke Chen, and Dachun Chen were responsible for writing the protocol, editing the paper and revising the manuscript. All authors contributed to the study and approved the final manuscript. References [1] Pillinger T, Beck K, Gobjila C, Donocik JG, Jauhar S, Howes OD. Impaired glucose homeostasis in first-episode schizophrenia: a systematic review and meta-analysis. JAMA Psychiat 2017;74(3):261–9. [2] Wu X, Huang Z, Wu R, et al. The comparison of glycometabolism parameters and lipid profiles between drug-naïve, first-episode schizophrenia patients and healthy controls. Schizophr Res 2013;150(1):157–62. [3] Chen S, Broqueres-You D, Yang G, et al. Relationship between insulin resistance, dyslipidaemia and positive symptom in Chinese antipsychotic-naive first-episode patients with schizophrenia. Psychiatry Res 2013;210(3):825–9. [4] Stern MP. Diabetes and cardiovascular disease. The “common soil” hypothesis. Diabetes 1995;44(4):369–74. [5] Luchsinger JA. Adiposity, hyperinsulinemia, diabetes and Alzheimer’s disease: an epidemiological perspective. Eur J Pharmacol 2008;585(1):119–29. [6] Gudala K, Bansal D, Schifano F, Bhansali A. Diabetes mellitus and risk of dementia: a meta-analysis of prospective observational studies. J Diabetes Investig 2013;4(6): 640–50. [7] Konig M, Lamos EM, Stein SA, Davis SN. An insight into the recent diabetes trials: what is the best approach to prevent macrovascular and microvascular complications. Curr Diabetes Rev 2013;9(5):371–81. [8] Pyörälä M, Miettinen H, Halonen P, Laakso M, Pyörälä K. Insulin resistance syndrome predicts the risk of coronary heart disease and stroke in healthy middle-aged men: the 22-year follow-up results of the Helsinki policemen study. Arterioscler Thromb Vasc Biol 2000;20(2):538–44. [9] Kahn SE, Cooper ME, Del PS. Pathophysiology and treatment of type 2 diabetes: perspectives on the past, present. And Future Lancet 2014;383(9922):1068–83. [10] Balf G, Stewart TD, Whitehead R, Baker RA. Metabolic adverse events in patients with mental illness treated with antipsychotics: a primary care perspective. Prim Care Companion J Clin Psychiatry 2008;10(1):15–24. [11] Perez-Iglesias R, Mata I, Pelayo-Teran JM, et al. Glucose and lipid disturbances after 1 year of antipsychotic treatment in a drug-naïve population. Schizophr Res 2009;107 (2–3):115–21. [12] Foley DL, Morley KI, Carroll KE, Moran J, McGorry PD, Murphy BP. Successful implementation of cardiometabolic monitoring of patients treated with antipsychotics. Med J Aust 2009;191(9):518–9 [author reply 519]. [13] Correll CU, Robinson DG, Schooler NR, et al. Cardiometabolic risk in patients with first-episode schizophrenia spectrum disorders: baseline results from the RAISEETP study. JAMA Psychiat 2014;71(12):1350–63. [14] Petrikis P, Tigas S, Tzallas AT, Papadopoulos I, Skapinakis P, Mavreas V. Parameters of glucose and lipid metabolism at the fasted state in drug-naïve first-episode patients with psychosis: evidence for insulin resistance. Psychiatry Res 2015;229(3):901–4. [15] Petruzzelli MG, Margari M, Peschechera A, et al. Hyperprolactinemia and insulin resistance in drug naive patients with early onset first episode psychosis. BMC Psychiatry 2018;18(1):246. [16] Zhang BH, Tan YL, Zhang WF, et al. Repeatable battery for the assessment of neuropsychological status as a screening test in Chinese:reliability and validity. Chinese Mental Health Journal 2008;22(12):865–9. [17] Kubota N, Terauchi Y, Yamauchi T, et al. Disruption of adiponectin causes insulin resistance and neointimal formation. J Biol Chem 2002;277(29):25863–6. [18] Zhang JQ. New HOMA-IR—insulin resistance measured from fasting glucose and fasting insulin. Chinese Journal of Diabetes Mellitus 2005;13(4):245–6.
6
C. Lin et al. / Comprehensive Psychiatry 96 (2020) 152145
[19] Wallace TM, Levy JC, Matthews DR. Use and abuse of HOMA modeling. Diabetes Care 2004;27(6):1487–95. [20] Tang Q, Li X, Song P, Xu L. Optimal cut-off values for the homeostasis model assessment of insulin resistance (HOMA-IR) and pre-diabetes screening: developments in research and prospects for the future. Drug Discov Ther 2015;9(6):380–5. [21] Gayoso-Diz P, Otero-González A, Rodriguez-Alvarez MX, et al. Insulin resistance (HOMA-IR) cut-off values and the metabolic syndrome in a general adult population: effect of gender and age: EPIRCE cross-sectional study. BMC Endocr Disord 2013;13(47. [22] Esteghamati A, Ashraf H, Esteghamati AR, et al. Optimal threshold of homeostasis model assessment for insulin resistance in an Iranian population: the implication of metabolic syndrome to detect insulin resistance. Diabetes Res Clin Pract 2009; 84(3):279–87. [23] Alberti KG, Eckel RH, Grundy SM, et al. Harmonizing the metabolic syndrome: a joint interim statement of the international diabetes federation task force on epidemiology and prevention; National Heart, Lung, and Blood Institute; American Heart Association; world heart federation; international atherosclerosis society; and International Association for the Study of obesity. Circulation 2009;120(16):1640–5. [24] Pereira S, Park E, Mori Y, et al. FFA-induced hepatic insulin resistance in vivo is mediated by PKCδ, NADPH oxidase, and oxidative stress. Am J Physiol Endocrinol Metab 2014;307(1):E34–46. [25] Stephens JM, Lee J, Pilch PF. Tumor necrosis factor-alpha-induced insulin resistance in 3T3-L1 adipocytes is accompanied by a loss of insulin receptor substrate-1 and GLUT4 expression without a loss of insulin receptor-mediated signal transduction. J Biol Chem 1997;272(2):971–6. [26] Liu A, Sonmez A, Yee G, et al. Differential adipogenic and inflammatory properties of small adipocytes in Zucker obese and lean rats. Diab Vasc Dis Res 2010;7(4):311–8. [27] Henderson DC, Fan X, Sharma B, et al. Waist circumference is the best anthropometric predictor for insulin resistance in nondiabetic patients with schizophrenia treated with clozapine but not olanzapine. J Psychiatr Pract 2009;15(4):251–61. [28] Qiao Y, Wen H, Chen HY, et al. The study of risk for metabolic syndromes in longterm hospitalized patients with schizophrenia. Shanghai Arch Psychiatry 2008;20 (2):65–8. [29] Institute of Medicine (US) Committee on Secondhand Smoke Exposure and Acute Coronary Events Secondhand Smoke Exposure and Cardiovascular Effects: Making Sense of the Evidence. Washington (DC): National Academies Press (US); 2010. [30] Cheng ZH, Ling WM, Zhou XQ, Ji Q. Prevalence of metabolic syndrome in patients with schizophrenia and its related factors. Chinese journal of behavioral medicine and brain. Science 2011;20(11):1012–5. [31] Smith S. Gender differences in antipsychotic prescribing. Int Rev Psychiatry 2010;22 (5):472–84.
[32] Koponen HJ, Hakko HH, Saari KM, et al. The prevalence and predictive value of individual criteria for metabolic syndrome in schizophrenia: a northern Finland 1966 birth cohort study. World J Biol Psychiatry 2010;11(2 Pt 2):262–7. [33] Laws A, Reaven GM. Evidence for an independent relationship between insulin resistance and fasting plasma HDL-cholesterol, triglyceride and insulin concentrations. J Intern Med 1992;231(1):25–30. [34] McLaughlin T, Reaven G, Abbasi F, et al. Is there a simple way to identify insulinresistant individuals at increased risk of cardiovascular disease. Am J Cardiol 2005; 96(3):399–404. [35] Fan X, Liu EY, Hoffman VP, Potts AJ, Sharma B, Henderson DC. Triglyceride/highdensity lipoprotein cholesterol ratio: a surrogate to predict insulin resistance and low-density lipoprotein cholesterol particle size in nondiabetic patients with schizophrenia. J Clin Psychiatry 2011;72(6):806–12. [36] Dasgupta A, Singh OP, Rout JK, Saha T, Mandal S. Insulin resistance and metabolic profile in antipsychotic naïve schizophrenia patients. Prog Neuropsychopharmacol Biol Psychiatry 2010;34(7):1202–7. [37] Smith RC, Lindenmayer JP, Davis JM, et al. Effects of olanzapine and risperidone on glucose metabolism and insulin sensitivity in chronic schizophrenic patients with long-term antipsychotic treatment: a randomized 5-month study. J Clin Psychiatry 2009;70(11):1501–13. [38] Chen FB, Sun J, Yan TJ, Xu LP, Wang HL. The liver fat content and its relationships with insulin resistance in schizophrenia patients receiving olanzapine or aripiprazol:a randomized comparative study. Chinese Journal of Nervous and Mental Diseases 2015;41(11):646–50. [39] Perry BI, McIntosh G, Weich S, Singh S, Rees K. The association between first-episode psychosis and abnormal glycaemic control: systematic review and meta-analysis. Lancet Psychiatry 2016;3(11):1049–58. [40] Yasui-Furukori N, Sato Y, Furukori H, Saito M, Nakagami T, Kaneko S. Glucose metabolism in Japanese schizophrenia patients treated with risperidone or olanzapine. J Clin Psychiatry 2009;70(1):95–100. [41] Wykes TL, Lee AA, McKibbin CL, Laurent SM. Self-efficacy and hemoglobin A1C among adults with serious mental illness and type 2 diabetes: the roles of cognitive functioning and psychiatric symptom severity. Psychosom Med 2016;78(3):263–70. [42] Nuevo R, Chatterji S, Fraguas D, et al. Increased risk of diabetes mellitus among persons with psychotic symptoms: results from the WHO world health survey. J Clin Psychiatry 2011;72(12):1592–9. [43] Lin PI, Shuldiner AR. Rethinking the genetic basis for comorbidity of schizophrenia and type 2 diabetes. Schizophr Res 2010;123(2–3):234–43.