Investigating the relationship between serum testosterone levels and
negative symptomatology in male patients with schizophrenia
Final Report - Research Project 1 Supervisor: Bodyl Brand
Abstract
Males with schizophrenia tend to show more negative symptoms in comparison to females leading to a reduced quality of life, increased functional impairment and poorer long-term outcomes. Low testosterone levels have been associated with greater negative symptom severity in males. The aim of this study was to investigate the overall level of testosterone and free testosterone in male patients with schizophrenia. Moreover, this study specifically related serum free testosterone levels to negative symptomatology, while exploring the role of age, Body Mass Index (BMI) and antipsychotic medication in this relationship. 52 male patients above the age of 18 with schizophrenia participated in this study. Blood samples were drawn between 8AM and 12PM to measure the level of free testosterone and testosterone. Negative
symptomatology was based on the negative subscale of the Positive and Negative Syndrome Scale (PANSS) and the Brief Negative Symptom Scale (BNSS). Two supplementary clusters of the BNSS were included in the analysis; the Motivation And Pleasure cluster and the Emotional Expressivity cluster. Results showed that serum levels of testosterone and free testosterone were significantly lower than those in the corresponding laboratory reference ranges of adult healthy males above the age of 18. No significant correlation was found between free testosterone and negative symptomatology and this relationship was not significantly influenced by age, BMI or antipsychotic medication. The present study indicates that male patients with schizophrenia have lower levels of free testosterone and testosterone. Free testosterone however was not associated with negative symptomatology nor was there effect of age, BMI or antipsychotic medication.
1. Introduction
Schizophrenia is a chronic, complex brain disorder characterized by symptoms including hallucinations, delusions, and a lack of motivation and concentration (Parekh, 2017; Kasai et al., 2002). Schizophrenia is characterized by three categories of symptoms; positive, negative, and cognitive symptoms (DSM-IV, American Psychiatric Association, 2000). Positive symptoms consist of psychotic behaviours that do not occur in healthy individuals including
hallucinations, delusions, thought disorders and movement disorders (National Institute of Mental Health, 2019). In contrast, negative symptoms are associated with an absence of regular emotions; a flat general affect and a reduction in overall pleasure, feelings or speaking.
Cognitive symptoms include reduced executive functioning, difficulties concentrating and poor overall working memory (National Institute of Mental Health, 2019). Currently, the majority of patients are treated with chronic medical treatment and therapy to help with
disease-management and coping (APA, 2017). Treatment-resistance may occur due to poor adherence, intolerance or inappropriate dosing of antipsychotic drug therapy (Conley & Kelly, 2001). However, even when these factors are positively met, pharmacological treatment and therapy show low treatment efficacy in managing persistent negative and cognitive symptoms in patients with schizophrenia (Cella, Preti, Edwards, Dow & Wykes, 2017; Chue & Lalonde, 2014). Notably, negative symptoms have been designed to capture a class of symptoms that can lead to functional impairment, yet effective treatment appears to remain inexistent (Chue & Lalonde, 2014; Buchanan, 2007).
Typically, in 80 percent of the patient population, disease onset of schizophrenia occurs around adolescence; when the release of luteinizing hormone, estrogens and androgens all increase rapidly (Stevens, 2002). This onset of symptoms around the reproductive period may indicate a possible link between sex steroid hormones and schizophrenic symptomatology (Ko et al., 2007). Sex differences are also consistently reported as one of the main aspects of the disease (Abel, Drake, R & Goldstein, 2010; Aleman, Kahn & Selten, 2003; de Boer, Prikken, Lei, Begemann & Sommer, 2018). These sex differences are evident when observing the
incidence rate ratio of 1.4:1 for males versus females with schizophrenia (McGrath, Saha, Chant & Welham, 2008). Overall, male patients have a younger age at onset with schizophrenic
symptoms occurring 3 to 5 years earlier than in females (Leung & Chue, 2000). As the disease progresses, males tend to show more negative and disorganized symptoms, whereas females are inclined to show affective symptoms with lower psychotic and negative symptoms (Abel et al., 2010; de Boer et al., 2018). This predicts a more favourable course of the disease for females as severe negative symptomatology has been associated with poorer functioning and worse clinical outcomes (Austin et al., 2013). The lack of efficacious treatment for the management of
negative symptoms represents a significant unmet need for patients with predominant negative symptoms (Chue & Lalonde, 2014). This is especially concerning for male patients with schizophrenia as this patient population is more inclined towards negative symptomatology.
It has been postulated that the observed reduction of symptom severity in female
patients in comparison to male patients is due to differing levels of sex steroids. Several studies have found a negative correlation between estrogen levels and symptom severity of
schizophrenia (Bergemann, Parzer, Runnebaum, Resch & Mundt, 2007; Häfner et al., 1993) More specifically, it is hypothesized that estrogen has a neuroprotective effect in schizophrenic patients, with increased estrogen being more beneficial (de Boer et al., 2018; Ko et al., 2007).
Interestingly, pharmacological interventions using estrogen may exert their effects through modifying serum levels of testosterone in patients with schizophrenia (Owens et al., 2019). Testosterone can either elicit its effects directly; by binding and signaling through the androgen receptor (Ko et al., 2007), or indirectly; by binding and signaling through the estrogen receptor after its conversion to estrogen or estrogen-like steroids (Moore et al., 2013; Owens et al., 2019). Biochemically, steroid hormones can rapidly cross the blood-brain barrier allowing brain levels to mirror blood levels, altering neural functioning through several neurotransmitters and neuropeptides (Bia³ek, Zaremba, Borowicz & Czuczwar, 2004; Banks, 2012). In summary, female patients appear to show a reduction in symptom severity due to higher levels of estrogen, which may exert its effects by modifying serum levels of testosterone.
Generally, males with schizophrenia appear to have lower levels of serum testosterone in comparison to healthy controls (Akhondzadeh et al., 2006; Owens et al., 2019). Adolescent males showing early signs of psychosis, had a significantly lower level of serum testosterone in comparison to non-clinical controls (Van Rijn et al., 2011). In addition to a general reduction of
serum testosterone (Akhondzadeh et al., 2006; Goyal, Sagar, Ammini, Khurana & Alias, 2004). Furthermore, a significant inverse correlation was found between negative symptoms and serum levels of testosterone in patients with predominant negative symptoms (Akhondzadeh et al., 2006). These findings were supported by another study in which moderate negative symptoms showed a negative correlation with serum testosterone levels (Shirayama, Hashimoto Suzuki & Higuchi, 2002). Conversely, other research found no relationship between serum testosterone levels and the severity of positive or negative symptoms in male patients with schizophrenia (Moore et al., 2013). According to a systematic review comparing three separate study trials of testosterone augmentation as adjunctive therapy, results are inconclusive with contradictory or non-significant findings (Elias & Kumar, 2007). The body of research investigating the
influence of testosterone in schizophrenic patients with predominant negative symptoms thus appears to be limited, with inconsistent results.
Most studies investigating the link between testosterone and negative symptomatology in male patients with schizophrenia do not include the possible influence of antipsychotic medication. Research has indicated that antipsychotic treatment leads to reduced serum testosterone levels in male patients in comparison to non-clinical subjects (Baptista, Reyes & Hernández, 1999). More specifically, previous studies have suggested that a decrease in testosterone levels may be due to antipsychotics induced hyperprolactinemia; or reduced prolactin levels (Owens et al., 2019; Ko et al., 2007; Li et al., 2015). This suggests that lower levels of serum testosterone may be the consequence of treatment with antipsychotic medication which may lead to an unclear relationship between serum testosterone and negative
symptomatology. As most patients with schizophrenia are chronically treated with
antipsychotics, it is important to explore the role of antipsychotic medication when investigating serum testosterone and negative symptomatology. In addition to antipsychotics increasing age and weight may also reduce the level of testosterone in patients (Ko et al., 2007; Allan & McLachlan, 2010; Kaufman & Vermeulen 2005; Goncharov et al. 2009). It is therefore important to control for age and weight (i.e. Body Mass Index) in addition to antipsychotic medication when analysing the relationship between serum testosterone levels and negative symptoms.
Conclusively, multiple studies have postulated the existence of an inverse relationship between serum testosterone levels and the severity of negative symptoms, yet without exploring the role of antipsychotic medication (Akhondzadeh et al., 2006; Owens et al., 2019; Goyal et al., 2004; Ko et al., 2007). Results of studies investigating the relationship between serum testosterone and negative symptomatology appear to be inconsistent and contradictory. In order to better understand potential pharmacological treatment options such as testosterone for treating negative symptomatology, this relationship thus needs to be clarified.
The aim of this study is to investigate the relationship between serum testosterone levels and negative symptomatology in male patients with schizophrenia. Both the level of total testosterone and free testosterone will be assessed; free testosterone indicates the amount of unattached and thus available testosterone in the blood (Bia³ek, Zaremba, Borowicz & Czuczwar, 2004). Within this study negative symptom severity is assessed using the Positive and Negative Syndrome Scale (PANSS) (Kay, Fiszbein & Opfer, 1987) and the Brief Negative Symptom Scale (BNSS) (Kirkpatrick et al., 2011). In addition to the PANSS the BNSS will be included with good reason, namely due to its high internal consistency for measuring negative symptoms in clinical practice. Two supplementary clusters of the BNSS will be included in the analysis; the Motivation And Pleasure cluster and the Emotional Expressivity cluster (San Ang, Rekhi & Lee, 2019). By using two negative symptom scales, the PANSS and BNSS, and two BNSS clusters, Motivation And Pleasure and Emotional Expressivity, this study attempts to obtain a thorough and complete assessment of negative symptomatology in male patients with schizophrenia (San Ang, Rekhi & Lee, 2019).
With the use of these measurements, this study investigated the following three aims. The first aim was to compare the level of serum testosterone of schizophrenic male patients to laboratory reference ranges in order to investigate whether male patients have lower levels of serum testosterone than healthy controls. The second aim was to investigate the relationship between serum testosterone levels and the severity of negative symptoms in male patients with schizophrenia using the negative subscale of the PANSS, the total BNSS score, and two BNSS clusters Motivation And Pleasure and Emotional Expressivity. The third aim was to explore the role of antipsychotic medication, age and BMI in this relationship. Based on prior research, it is
serum testosterone levels would be inversely correlated with negative symptoms, and that this relationship is influenced by the dosage of antipsychotic medication, age, and BMI.
2. Methods 2.1 Participants
For this study, baseline data of the RAPSODI study conducted in the University Medical Center Utrecht (UMCU) was used (Raloxifene Augmentation in Patients with a Schizophrenia
Spectrum Disorder, UMCU, 2016); approved by the Ethics Committee in the Netherlands in August 2016. The RAPSODI study is an ongoing double-blind placebo-controlled clinical trial investigating the effectiveness of Raloxifene augmentation as adjunctive therapy in patients with schizophrenia. Participants are either recruited via referral of their psychiatrist, via an advertisement, or directly after having consented to be approachable for research. All subjects are capable of understanding the purpose and details of the study in order to provide written informed consent. For a full protocol description of this study, see ClinicalTrials.gov. (Identifier: NCT03043820).
The data of 52 male patients over the age of 18 was used. All subjects had a DSM-IV-R diagnosis of schizophrenia, schizophreniform disorder, schizoaffective disorder, or psychotic disorder NOS (code 295.x or 298.9), affirmed by the Mini-International Neuropsychiatric Interview (M.I.N.I., Sheehan et al., 1998). Subjects were required to be on a stable dosage of antipsychotic medication for at least two weeks prior to the day of data collection in order to participate in the study.
2.2 Measurements 2.2.1. Hormonal Assays
All measurements took place on the same day. Blood samples were drawn between 8 am and 12 pm to determine the hormonal levels of participants. Patients were instructed to fast six hours prior to the blood draw. One serum separator tube (3.5ml) was collected to measure the
to the Laboratory of Clinical Chemistry and Haematology (LKCH) in the UMCU for further analysis. The reference ranges used for males above the age of 18 were total testosterone, 11 - 40 nmol/L (LKHC, UMC Utrecht, 2020-a); and free testosterone, 230 - 600 pmol/L (LKHC, UMC Utrecht, 2020-b). Values within these reference ranges are those within the normal distribution of a reference group taken from the general population of healthy males above the age of 18.
2.2.2 Negative Symptomatology
Negative symptomatology was assessed using the Positive and Negative Syndrome Scale (PANSS) (Kay, Fiszbein & Opfer, 1987) and the Brief Negative Symptom Scale (BNSS) (Kirkpatrick et al., 2011), respectively.
The PANSS is a medical scale used to assess symptom severity consisting of 30 items rated on a 7-point Likert scale ranging from 1 (absent) to 7 (extremely severe), subdivided into three domains: positive symptoms (7 items), negative symptoms (7 items) and general
psychopathology (16 items). This study used data corresponding to the negative symptom scale consisting of the following 7 items: blunted affect, emotional withdrawal, poor rapport,
passive/apathetic social withdrawal, difficulty in abstract thinking, lack of spontaneity and flow of conversation, and lastly stereotyped thinking. The occurrence of symptoms is rated for the week preceding the PANSS interview. Scores on the positive and negative Scales ranges from 7 to 49, while scores on the general psychopathology scale ranges from 16 to 113. A total score can be derived from summing these 3 subscale scores, ranging from a minimum total score of 30 to a maximum score of 210. Generally, the PANSS takes 30 to 40 minutes to complete and is conducted by a validated first and a second rater who will rate the symptom severity in
conjunction.
The BNSS is used to assess negative symptom severity and consists of 13 items divided into 6 subscales of negative symptoms: anhedonia, distress, asociality, avolition, blunted affect, and alogia. The subscales, corresponding items, and score ranges are depicted in Table 1. Some subscales have a higher exclusive score than others due to an increased number of items or higher score ranges; items can be scored on a 6-point Likert scale ranging from 0 (absent) to 6
(severe) or a 9-point likert scale ranging from 0 (absent) to 9 (extremely severe). The total BNSS score can be derived from summing each subscale score and ranges from 0 to 87. The BNSS requires approximately 15 minutes to be completed. Additional questions were asked regarding the following subscales: anhedonia, distress, asociality and avolition. Symptom severity was based on the internal experience of the subject and their behaviour in the preceding and near future.
In addition to the total score of the BNSS this study included two BNSS clusters: the Motivation And Pleasure cluster and the Emotional Expressivity cluster (San Ang, Rekhi & Lee, 2019). The BNSS subscales and score ranges corresponding to each cluster are depicted in Table 2.
Table 1
Subscales, corresponding items, and score ranges of the BNSS
BNSS Subscales Items (score range) Subscale
score range 1. Anhedonia 1. Intensity of pleasure during activities
(0 - 6)
2. Frequency of pleasure during activities (0 - 6)
3. Intensity of expected pleasure from future activities (0 - 6)
0 - 18
2. Distress 4. The absence of regular distress (0 - 6)
0 - 6
3. Asociality
Lack of motivation to participate in social activities:
5. Behaviour (0 - 6)
6. Internal experience (0 - 6)
0 - 12
4. Avolition
Lack of motivation to initiate and perform self-directed activities:
7. Behaviour (0 - 6)
8. Internal experience (0 - 6)
0 - 12
5. Blunted Affect 9. Facial expression (0 - 6) 10. Vocal expression (0 - 9) 11. Expressive gestures (0 - 6)
0 - 21
6. Alogia 12. Quantity of speech (0 - 9) 13. Spontaneous elaboration (0 - 9)
0 - 18
Total range: 0 - 82
Table 2
This table depicts the BNSS clusters MAP and EE and the corresponding BNSS subscales, score ranges and cluster score ranges
BNSS Clusters BNSS Subscales (score range) Cluster score range 1. MAP 1. Anhedonia (0 - 18) 3. Asociality (0 - 12) 4. Avolition (0 - 12) 0 - 42 2. EE 5. Blunted Affect (0 - 21) 6. Alogia (0 - 18) 0 - 39
Note. BNSS = Brief Negative Symptom Scale; MAP = Motivation And Pleasure; EE =
Emotional Expressivity.
2.2.3 General measurements
Following the measurement of negative symptomatology, participants were asked to report their age, their length in centimeters and their weight was measured using a weighing scale in
kilograms. Body Mass Index (BMI) was assessed by dividing the weight over the squared height.
2.3 Statistical Analyses
The collected data was analyzed in RStudio using R version 3.6.3 (RStudio Team, 2020). The assumption of linearity was checked by inspecting the plotted residuals versus fitted values. The assumption of multivariate normality was tested using the Shapiro-Wilk test. The assumption of homoscedasticity was checked using the Breusch-Pagan Test and multicollinearity was checked by looking at the Variance Inflation Factor (VIF). In case of violation, data was transformed with the use of a suitable transformation (Field, Miles & Field, 2012). If outliers were present in
the dataset, the analysis would be conducted on both the complete dataset and the dataset following outlier removal.
In order to attain the first aim, two one-sample t-tests were conducted comparing the mean of testosterone and the mean of free testosterone to the mean of the corresponding laboratory reference range.
Prior to achieving the second aim, a correlation analysis was conducted comparing the independent variables testosterone and free testosterone. Subsequently both testosterone and free testosterone were correlated to the dependent variables; PANSS Negative, BNSS,
Motivation And Pleasure, and Emotional Expressivity. All subsequent analyses were conducted on the independent variable with the strongest correlation to negative symptomatology.
To achieve the second aim, four multiple linear regression (MLR) analyses were
conducted using the independent variable as predictor variable. Any of the three covariates; age, Body Mass Index (BMI), or the chlorpromazine equivalent, with a significant correlation to the predictor variable were included in the analysis. The four linear regression models predicted scores on the i) negative subscale of the PANSS, ii) the total BNSS score, iii) the Motivation And Pleasure cluster score, and iv) Emotional Expressivity cluster score.
The third aim was attained by conducting a Pearson’s correlation analysis comparing free testosterone to age, Body Mass Index (BMI), and the chlorpromazine equivalent (Gardner, Murphy, O'Donnell, Centorrino & Baldessarini, 2010). For a significant correlation, the variable would be added to all four MLR analyses as a covariate.
3. Results
Following the exclusion of six participants due to missing data values, the data of 52 male patients was analysed. Outliers were not omitted from the analysis; the complete dataset was analysed. The descriptive statistics of this sample are summarized in Table 3.
Table 3
Descriptive statistics including the M, SD, SE, and range of each continuous variable measured in a sample of male patients with schizophrenia
Variable N M SD SE Range PANSS Total 52 55.90 13.63 1.89 [31 , 94] PANSS Negative 52 12.83 4.97 0.69 [7 , 29] PANSS Positive 52 14.77 4.86 0.67 [7, 25] PANSS General 52 28.31 7.33 1.02 [16 , 47] BNSS Total 52 14.42 12.05 1.67 [0 , 50] BNSS cluster MAP 52 8.83 7.48 1.04 [0 , 35] BNSS cluster EE 52 5.31 6.44 0.89 [0 , 25] Free Testosterone 52 297.04 91.75 12.72 [120 , 520] Testosterone 52 16.19 6.86 0.95 [4.6 , 34] Age (years) 52 39.25 10.53 1.46 [20 , 73] BMI 52 26.02 4.29 0.59 [17 , 40] Chlorpromazine equivalent 52 400.14 200.18 30.19 [0 , 1150]
Note. SD = standard deviation; SE = standard error; M = mean; PANSS = Positive And
Negative Syndrome Scale; BNSS = Brief Negative Symptom Scale; MAP = Motivation And Pleasure; EE = Emotional Expressivity; BMI = Body Mass Index
3.1 Comparing testosterone and free testosterone to the laboratory reference range The analysis showed a significant difference between the sample mean of testosterone (M=16.19, SD = 6.86) and the mean of the corresponding reference range (M=25.5, SD =7.25); (t (51) = -2.96, p = <.01), indicating that testosterone levels are significantly lower in male patients with schizophrenia in comparison to the laboratory reference range of testosterone (Figure 1a).
Moreover, there was a significant difference between the sample mean of free
testosterone (M=297.04, SD=91.75) and the mean of the corresponding reference range (M = 415, SD = 92.50); (t (51) = -9.27, p = <.001), indicating that free testosterone levels are significantly lower in male patients with schizophrenia in comparison to the laboratory reference range of free testosterone (Figure 1b).
Figure 1
Comparing the (a) range of testosterone and (b) free testosterone in a sample of male patients with schizophrenia to the corresponding laboratory reference range in healthy adult males
3.2 Comparing testosterone and free testosterone as independent variables
A correlation analysis showed that testosterone was significantly correlated to free testosterone (r = .70, p < .001). In addition, free testosterone showed a stronger correlation with the
dependent variables in comparison to testosterone (Table 4). For this reason, the subsequent analyses were conducted using free testosterone as the independent variable.
Table 4
Comparing the correlations of total testosterone and free testosterone with the PANSS negative subscale, BNSS total score, MAP cluster score, EE cluster score, age, BMI, and the
chlorpromazine equivalent.
Variables Testosterone (r) Free testosterone (r)
PANSS Negative -0.03 -0.07 BNSS total -0.09 -0.13 MAP cluster (BNSS) -0.08 -0.16 EE cluster (BNSS) -0.11 -0.08 Age -0.13 -0.38 BMI -0.59 -0.43 Chlorpromazine equivalent -0.02 -0.16
Note. PANSS = Positive And Negative Symptom Scale; BNSS = Brief Negative Symptom
Scale; MAP = Motivation And Pleasure; EE = Emotional Expressivity; BMI = Body Mass Index
3.3 Free testosterone levels and negative symptomatology
Free testosterone showed a significant correlation with the covariate BMI (r = -0.43, p = .001) and with the covariate age (r = -0.38, p = .006), and no significant correlation with the chlorpromazine equivalent (r = -0.16, p = .27). For this reason, age and BMI were included in the linear regression model as covariates in addition to free testosterone. As the chlorpromazine equivalent did not show a significant correlation with free testosterone this variable was
excluded from the linear model.
The summary statistics of each linear model predicting scores on the negative subscale of Positive And Negative Syndrome Scale (PANSS), the total score of the Brief Negative Symptom Scale (BNSS) and the two BNSS clusters Motivation And Pleasure and Emotional Expressivity are depicted in Table 5. In total, the level of free testosterone, BMI and age accounted for 6% of the variance in the PANSS negative score (R2 = 0.06, F (3, 48) = 1 , p = .40), 5% of the variance in the total BNSS score (R2 = 0.05, F (3, 48) = 0.76 , p = .52), 7% of the variance in the Motivation And Pleasure score (R2 = 0.07, F (3, 48) = 1.18 , p = .33) and 3% of the variance in the Emotional Expressivity score (R2 = 0.03, F (3, 48) = 0.45, p = .72). None of the linear models showed a significant result, nor did the covariates. This result therefore does not indicate a relationship between free testosterone and negative
symptomatology in male patients with schizophrenia, Each linear model was plotted in a scatterplot with the addition of a regression line (Figure 2).
Table 5
Summary of four multiple linear regression analyses examining the influence of free
testosterone, BMI, and age on negative symptomatology (dependent variables) in male patients with schizophrenia
Linear Model Beta t p-value 95% CI
a. PANSS Negative Free testosterone -0.00 -0.77 .44 [-.00 , .00] Age 0.00 0.81 .42 [-.01 , .02] BMI -0.02 -1.49 .14 [-.05 , .01] b. BNSS Free testosterone -0.03 -1.44 .17 [-.07 , .01] Age -0.12 -0.68 .49 [-.05 , .23] BMI -0.42 -0.96 .34 [-1.31 , .46] c. MAP cluster (BNSS) Free testosterone -0.02 -1.60 .12 [-.05 , .01] Age -0.00 0.05 .96 [-.22 , .21] BMI -0.40 -1.48 .15 [-.94 , .14] d. EE cluster (BNSS) Free testosterone -0.00 -0.84 .41 [.01 , .00] Age -0.02 -1.02 .31 [-.07 , .02] BMI -0.02 -0.28 .78 [-.13 , .10]
Note. CI= confidence interval; BMI = Body Mass Index; PANSS = Positive And Negative
Syndrome Scale; BNSS = Brief Negative Symptom Scale; MAP = Motivation And Pleasure; EE = Emotional Expressivity
Figure 2
Scatterplots visualizing the relationship between free testosterone and the a) PANSS negative subscale score; b) BNSS total score; c) MAP cluster score; d) EE cluster score in male patients with schizophrenia
a. b.
c. d.
Note. PANSS = Positive And Negative Syndrome Scale; BNSS = Brief Negative Symptom
4. Discussion
This study investigated the level of testosterone and free testosterone in male patients with schizophrenia. In addition, the relationship between free testosterone and negative
symptomatology was analysed while examining the influence of age, Body Mass Index (BMI) and the chlorpromazine equivalent. The three main findings of this study were as follows: (i) the level of testosterone and the level of free testosterone in male patients with schizophrenia was significantly lower than the corresponding laboratory reference range; (ii) free testosterone levels did not show a correlation with negative symptomatology as measured by two different rating scales; (iii) and this relationship was not influenced by age, BMI or antipsychotic medication.
This experiment confirms the hypothesis that testosterone and free testosterone levels in male patients with schizophrenia are lower in comparison to corresponding laboratory reference ranges. This finding corresponds with results of prior studies which have also reported lower levels of testosterone and free testosterone in male patients (Owens et al., 2019; Akhondzadeh et al., 2006). In contrast, other studies have reported no difference in testosterone levels between males with schizophrenia and healthy controls (Moore et al., 2013). As testosterone levels are known to decline with increasing age and obesity, these factors may have contributed to these differences (Li et al., 2015). Previous research has implicated that the mean testosterone level of males remains unchanged up to the age of 70, while free testosterone levels start to decline after the age of 50 (Stearns et al., 1974). On average patients within this study were aged
39.25±10.53 years with 99.7% of males below the age of 70 and 68% below the age of 50. Age therefore may have influenced the level of testosterone in 3% of the sample and the level of free testosterone in 32% of the sample. Indeed the correlation between free testosterone and age (-0.38) was higher in this sample than the correlation between testosterone and age (-0.13). Prior studies reporting lower levels of testosterone and free testosterone in male patients with schizophrenia omitted the influence of age by using healthy age-matched controls (Owens et al., 2019; Akhondzadeh et al., 2006).
In addition to age, BMI may also influence testosterone levels. Previous studies have shown a negative relationship between total testosterone, free testosterone and increased BMI
(Kaufman & Vermeulen 2005; Goncharov, Katsya, Chagina & Gooren, 2009). A BMI of 18.5 to 24.9 is an indicator of a healthy weight, whereas a BMI between 25 and 29.9 is considered overweight. The average BMI of patients within this sample was 26.02±4.29, meaning that patients within this sample were marginally overweight which may have influenced the
measured level of testosterone and free testosterone. Indeed BMI showed a moderate correlation of -0.59 with testosterone and -0.43 with free testosterone in this sample.
Decreased levels of testosterone may be damaging in males which is supported by findings indicating that low testosterone levels are associated with greater negative symptom severity (Akhondzadeh et al., 2006; Goyal et al., 2004; Owens et al., 2019). Women show a lower incidence of schizophrenia, a later age at onset, less negative symptoms and a better response to treatment (McGrath, Saha, Chant & Welham, 2008; Abel et al., 2010; de Boer et al., 2018). The hormone estrogen plays a neuroprotective role in female patients with
schizophrenia, it is possible that increased testosterone in females also plays a neuroprotective role by buffering women against severe symptomatology (Owens et al., 2019).
This study did not find a correlation between free testosterone and negative symptomatology in males contradicting earlier findings showing an inverse relationship between free testosterone and negative symptomatology (Akhondzadeh et al., 2006; Goyal et al., 2004). For this reason it is important to evaluate factors which may have influenced the level of free testosterone in our sample. First, free testosterone is known to decline with
increasing age (Ko et al., 2007), by controlling for age this study omitted the confounding effect of this variable. Moreover, free testosterone levels are subject to diurnal variation as they tend to fluctuate; peaking at 8 in the morning and diminishing throughout the day (Brambilla, Matsumoto, Araujo & McKinlay, 2009). Within our study blood was sampled before noon (12 PM), therefore minimizing the effect of diurnal variation on the measured level of free
testosterone. Obesity also lowers free testosterone levels (Allan & McLachlan, 2010; Kaufman & Vermeulen 2005; Goncharov et al. 2009). While previous studies have failed to include the potential influence of body mass (Shirayama, Hashimoto, Suzuki & Higuchi, 2002;
Akhondzadeh et al., 2006), this study controlled for obesity (i.e. BMI). Lastly, antipsychotic medication increases the level of prolactin in the blood, which in turn may decrease the level of
chlorpromazine equivalent, the confounding influence of antipsychotics on free testosterone was removed.
Subsequently, factors which may influence the measurement of negative symptoms in our sample also need to be considered. Prior studies found significant differences between testosterone levels in patients with moderate to severe negative symptoms, but not in patients with mild negative symptoms (Shirayama, Hashimoto, Suzuki & Higuchi, 2002). Interestingly, another study solely found a significant relationship between testosterone levels and
predominant negative symptoms for patients with a PANSS negative subscale score above 20 (Akhondzadeh et al., 2006). On average, male patients within this study scored 12.83±4.97 out of 49 on the PANSS negative subscale complying with a mild to moderate negative symptom severity. The average score on the BNSS was 14.42±12.05 out of 82 indicating mild negative symptom severity. Mild severity of negative symptomatology within this sample may provide an explanation for the absence of finding a relationship between testosterone levels and negative symptoms.
There are a number of factors which contribute to the difficulty of conducting clinical research on patients with persistent negative symptoms. The most important being the difficulty of recruiting a sufficient number of patients with severe negative symptoms (Buchanan, 2007). Furthermore, the measurement of negative symptomatology is reliant on subjective responses in order to rate negative symptom severity giving rise to several psychometric constraints (Cohen, Alpert, Nienow, Dinzeo & Docherty, 2008). Rating the severity of negative symptoms using the Positive And Negative Symptom Scale (PANSS) and the Brief Negative Symptom Scale
(BNSS) is dependent on Likert-type symptom rating scales. In order to provide inter-rater reliability, raters within this study were properly trained and certified researchers. However, item 2 and item 4 on the negative subscale of the PANSS require reports of functioning from primary care workers or family in order to be rated. Within this study patients were only seen once without consulting primary care workers to rate these two items, which may have constrained the measured severity of negative symptomatology in patients.
Furthermore, this study does not provide evidence for an inverse relationship between antipsychotic medication and testosterone levels. This finding opposes previous research which
hyperprolactinemia; or reduced prolactin levels (Owens et al., 2019; Ko et al., 2007). According to a thorough systematic review, hyperprolactinemia appears to be sex-related, dose-related and differing per type of antipsychotic (Peuskens, Pani, Detraux & De Hert, 2014). Interestingly, this review showed that high prolactin levels can be pre-existing in patients with psychosis who have not previously taken antipsychotics. According to this review, evidence supporting the existence of a relationship between antipsychotics and reduced testosterone levels remains insufficient. The finding of this study indicating no association between antipsychotics and testosterone levels therefore remains meaningful.
4.1 Strengths and limitations
This study has several strengths, including a large sample of male patients with schizophrenia. In addition, this study used multiple negative symptom scales, the PANSS and BNSS, to measure negative symptomatology in patients. The current study also controlled for age, BMI, diurnal variation, and antipsychotic medication when studying the association between free testosterone and negative symptoms.
However, this study was also subject to several limitations. One inherent weakness of this study is the absence of a healthy-matched control group when comparing testosterone levels in male patients with schizophrenia to a healthy sample of males. Unfortunately, data of a healthy-matched control group was not available as this study used the dataset of a different research project. Due to this limitation, age may have exerted a moderate influence on free testosterone whereas BMI may have influenced both testosterone and free testosterone. Another important limitation is the mild severity of negative symptoms within this sample, which may have contributed to finding no association between free testosterone and negative
symptomatology.
4.2 Future research
In order to make more robust conclusions regarding the overall level of testosterone and free testosterone in male patients with schizophrenia, future studies should use a
future studies should consider consulting the psychiatric practitioner or primary care workers of patients to get an accurate image of negative symptom severity in addition to using multiple negative symptom scales. For a better understanding of the relationship between free
testosterone and antipsychotics, future studies may also consider grouping antipsychotics from the same class based on their influence on prolactin. In addition, prolactin levels could be taken into account to get a clearer understanding of the interplay between prolactin, testosterone, antipsychotics, and negative symptom severity.
References
Abel, K. M., Drake, R., & Goldstein, J. M. (2010). Sex differences in schizophrenia.
International review of psychiatry, 22(5), 417-428. doi: 10.3109/09540261.2010.515205 Allan, C. A., & McLachlan, R. I. (2010). Androgens and obesity. Current Opinion in
Endocrinology, Diabetes and Obesity, 17(3), 224-232.
doi: 10.1097/MED.0b013e3283398ee2
Aleman, A., Kahn, R. S., & Selten, J. P. (2003). Sex differences in the risk of schizophrenia: evidence from meta-analysis. Archives of general psychiatry, 60(6), 565-571.
doi:10.1001/archpsyc.60.6.565
American Psychiatric Association. (2000). Diagnostic and statistical manual of mental disorders: DSM-IV-TR. Washington, DC: APA.
Akhondzadeh, S., Rezaei, F., Larijani, B., Nejatisafa, A. A., Kashani, L., & Abbasi, S. H. (2006). Correlation between testosterone, gonadotropins and prolactin and severity of negative symptoms in male patients with chronic schizophrenia. Schizophrenia research, 84(2-3), 405-410. doi: 10.1016/j.schres.2006.02.008
Austin, S. F., Mors, O., Secher, R. G., Hjorthøj, C. R., Albert, N., Bertelsen, M., ... & Thorup, A. (2013). Predictors of recovery in first episode psychosis: the OPUS cohort at 10 year follow-up. Schizophrenia research, 150(1), 163-168. doi: 10.1016/j.schres.2013.07.031 Banks, W. A. (2012). Brain meets body: the blood-brain barrier as an endocrine interface.
Endocrinology, 153(9), 4111-4119. doi: 10.1210/en.2012-1435
Baptista, T., Reyes, D., & Hernández, L. (1999). Antipsychotic drugs and reproductive hormones: relationship to body weight regulation. Pharmacology Biochemistry and
Bergemann, N., Parzer, P., Runnebaum, B., Resch, F., & Mundt, C. (2007). Estrogen, menstrual cycle phases, and psychopathology in women suffering from schizophrenia.
Psychological medicine, 37(10), 1427-1436. doi: 10.1017/S0033291707000578 Bia³ek, M., Zaremba, P., Borowicz, K. K., & Czuczwar, S. J. (2004). Neuroprotective role of
testosterone in the nervous system. Pol. J. Pharmacol, 56, 509-518.
de Boer, J., Prikken, M., Lei, W. U., Begemann, M., & Sommer, I. (2018). The effect of raloxifene augmentation in men and women with a schizophrenia spectrum disorder: a systematic review and meta-analysis. npj Schizophrenia, 4(1), 1-6.
doi: 10.1038/s41537-017-0043-3
Brambilla, D. J., Matsumoto, A. M., Araujo, A. B., & McKinlay, J. B. (2009). The effect of diurnal variation on clinical measurement of serum testosterone and other sex hormone levels in men. The Journal of Clinical Endocrinology & Metabolism, 94(3), 907-913. doi: 10.1210/jc.2008-1902
Buchanan, R. W. (2007). Persistent negative symptoms in schizophrenia: an overview. Schizophrenia bulletin, 33(4), 1013-1022. doi: 10.1093/schbul/sbl057
Cella, M., Preti, A., Edwards, C., Dow, T., & Wykes, T. (2017). Cognitive remediation for negative symptoms of schizophrenia: a network meta-analysis. Clinical psychology review, 52, 43-51. doi: 10.1016/j.cpr.2016.11.009
Chue, P., & Lalonde, J. K. (2014). Addressing the unmet needs of patients with persistent negative symptoms of schizophrenia: emerging pharmacological treatment options. Neuropsychiatric disease and treatment, 10, 777. doi: 10.1016/j.cpr.2016.11.009 ClinicalTrials.gov. National Library of Medicine (U.S.) (2016, August). Raloxifene
Symptoms and Improve Cognition. Identifier NCT03043820. Retrieved from http://clinicaltrials.gov/ct/show/NCT03043820?order=1
Cohen, A. S., Alpert, M., Nienow, T. M., Dinzeo, T. J., & Docherty, N. M. (2008). Computerized measurement of negative symptoms in schizophrenia. Journal of psychiatric research, 42(10), 827-836. doi: 10.1016/j.jpsychires.2007.08.008
Conley, R. R., & Kelly, D. L. (2001). Management of treatment resistance in schizophrenia. Biological psychiatry, 50(11), 898-911. doi: 10.1016/S0006-3223(01)01271-9 Elias, A., & Kumar, A. (2007). Testosterone for schizophrenia. Cochrane Database of
Systematic Reviews, (3). doi: 10.1002/14651858.CD006197.pub2
Field, A. P., Miles, J., & Field, Z. (2012). Discovering statistics using R.
Gardner, D. M., Murphy, A. L., O'Donnell, H., Centorrino, F., & Baldessarini, R. J. (2010). International consensus study of antipsychotic dosing. American Journal of Psychiatry, 167(6), 686-693. doi: 10.1176/appi.focus.12.2.235
Goncharov, N. P., Katsya, G. V., Chagina, N. A., & Gooren, L. J. (2009). Testosterone and obesity in men under the age of 40 years. Andrologia, 41(2), 76-83.
doi: 10.1111/j.1439-0272.2008.00863.x
Goyal, R. O., Sagar, R., Ammini, A. C., Khurana, M. L., & Alias, A. G. (2004). Negative correlation between negative symptoms of schizophrenia and testosterone levels. Annals
of the New York Academy of Sciences, 1032(1), 291-294. doi: 10.1196/annals.1314.042
Häfner, H., Riecher-Rössler, A., Der Heiden, W. A., Maurer, K., Fätkenheuer, B., & Löffler, W. (1993). Generating and testing a causal explanation of the gender difference in age at First onset of schizophrenia. Psychological medicine, 23(4), 925-940.
Kasai, K., Iwanami, A., Yamasue, H., Kuroki, N., Nakagome, K., & Fukuda, M. (2002). Neuroanatomy and neurophysiology in schizophrenia. Neuroscience Research, 43(2), 93-110. doi:10.1016/S0168-0102(02)00023-8Get
Kaufman, J. M., & Vermeulen, A. (2005). The decline of androgen levels in elderly men and its clinical and therapeutic implications. Endocrine reviews, 26(6), 833-876.
doi: 10.1210/er.2004-0013
Kay, S. R., Fiszbein, A., & Opler, L. A. (1987). The positive and negative syndrome scale (PANSS) for schizophrenia. Schizophrenia bulletin, 13(2), 261-276.
doi: 10.1093/schbul/13.2.261
Kirkpatrick, B., Strauss, G. P., Nguyen, L., Fischer, B. A., Daniel, D. G., Cienfuegos, A., & Marder, S. R. (2011). The brief negative symptom scale: psychometric properties. Schizophrenia bulletin, 37(2), 300-305. doi: 10.1093/schbul/sbq059
Ko, Y. H., Jung, S. W., Joe, S. H., Lee, C. H., Jung, H. G., Jung, I. K., ... & Lee, M. S. (2007). Association between serum testosterone levels and the severity of negative symptoms in male patients with chronic schizophrenia. Psychoneuroendocrinology, 32(4), 385-391. doi: 10.1016/j.psyneuen.2007.02.002
Leung MD, D. A., & Chue MRC Psych, D. P. (2000). Sex differences in schizophrenia, a review of the literature. Acta Psychiatrica Scandinavica, 101(401), 3-38.
doi: 10.1111/j.0065-1591.2000.0ap25.x
Li, J., Xiao, W., Sha, W., Xian, K., Tang, X., & Zhang, X. (2015). Relationship of serum testosterone levels with cognitive function in chronic antipsychotic-treated male patients with schizophrenia. Asia-Pacific Psychiatry, 7(3), 323-329.
LKHC, UMC Utrecht. (2020-a). Testosteron. 22 of may 2020, retrieved from http://lkch.nl/bepalingen/Front/Details/1365
LKHC, UMC Utrecht. (2020-b). Vrij Testosteron. 22 of may 2020, retrieved from http://lkch.nl/bepalingen/Front/Details/1367
McGrath, J., Saha, S., Chant, D., & Welham, J. (2008). Schizophrenia: a concise overview of incidence, prevalence, and mortality. Epidemiologic reviews, 30(1), 67-76.
doi: 10.1093/epirev/mxn001
Moore, L., Kyaw, M., Vercammen, A., Lenroot, R., Kulkarni, J., Curtis, J., ... & Weickert, T. W. (2013). Serum testosterone levels are related to cognitive function in men with schizophrenia. Psychoneuroendocrinology, 38(9), 1717-1728.
doi: 10.1016/j.psyneuen.2013.02.007
National Institute of Mental Health (2019, February). NIMH » Schizophrenia.
Retrieved from: https://www.nimh.nih.gov/health/topics/schizophrenia/index.shtml Owens, S. J., Weickert, T. W., Purves-Tyson, T. D., Ji, E., White, C., Galletly, C., ... &
Weickert, C. S. (2019). Sex-Specific Associations of Androgen Receptor CAG
Trinucleotide Repeat Length and of Raloxifene Treatment with Testosterone Levels and Perceived Stress in Schizophrenia. Molecular neuropsychiatry, 5(1), 28-41.
doi: 10.1159/000495062
Parekh, R. (2017, July 1). What Is Schizophrenia? Retrieved March 1, 2020, from
https://www.psychiatry.org/patients-families/schizophrenia/what-is-schizophrenia Peuskens, J., Pani, L., Detraux, J., & De Hert, M. (2014). The effects of novel and newly
approved antipsychotics on serum prolactin levels: a comprehensive review. CNS drugs, 28(5), 421-453. doi: 10.1007/s40263-014-0157-3
RStudio Team (2020). RStudio: Integrated Development for R. RStudio, Inc., Boston, MA URL
http://www.rstudio.com/
San Ang, M., Rekhi, G., & Lee, J. (2019). Validation of the Brief Negative Symptom Scale and
its association with functioning. Schizophrenia research, 208, 97-104.
doi: 10.1016/j.schres.2019.04.005
Sheehan, D. V., Lecrubier, Y., Sheehan, K. H., Amorim, P., Janavs, J., Weiller, E., Hergueta,
T., Baker, R., & Dunbar, G. C. (1998). The Mini-International Neuropsychiatric Interview (M.I.N.I): The development and validation of a structured diagnostic
psychiatric interview for DSM-IV and ICD-10. The Journal of Clinical Psychiatry,
59 (Suppl 20), 22–33.
Shirayama, Y., Hashimoto, K., Suzuki, Y., & Higuchi, T. (2002). Correlation of plasma neurosteroid levels to the severity of negative symptoms in male patients with schizophrenia. Schizophrenia research, 58(1), 69-74.
doi: 10.1016/S0920-9964(01)00367-X
Stearns, E. L., MacDonnell, J. A., Kaufman, B. J., Padua, R., Lucman, T. S., Winter, J. S. D., & Faiman, C. (1974). Declining testicular function with age: hormonal and clinical
correlates. The American journal of medicine, 57(5), 761-766. doi: 10.1016/0002-9343(74)90850-X
Stevens, J. R. (2002). Schizophrenia: reproductive hormones and the brain. American Journal of Psychiatry, 159(5), 713-719. doi: 10.1176/appi.ajp.159.5.713
UMCU, (2016, July 8). Raloxifene Augmentation in Patients with a Schizophrenia spectrum Disorder to reduce symptoms and improve cognition. EU Clinical Trials Register. Retrieved from
https://clinicaltrials.gov/ct2/show/NCT03043820?cond=raloxifene+augmentation&draw =2&rank=1
Van Rijn, S., Aleman, A., De Sonneville, L., Sprong, M., Ziermans, T., Schothorst, P., ... & Swaab, H. (2011). Neuroendocrine markers of high risk for psychosis: salivary testosterone in adolescent boys with prodromal symptoms. Psychological medicine,
