THIAMINE AND MAGNESIUM NOT
FOUND TO AFFECT COGNITIVE
FLEXIBILITY IN EATING DISORDERS
DOMINIQUE WELTEVREDEN | 12161160
DR. R. OP ’T LANDT- SLOF | GGZ RIVIERDUINEN EETSTOORNISSEN URSULA
Bachelor thesis for Psychobiologie of the University of Amsterdam | 30 EC
1
Thiamine and magnesium not found to
affect cognitive flexibility in eating
disorders
Abstract
Magnesium and thiamine levels are found to be deficient in patients with eating disorders. These micronutrients are associated with cognitive function; patients with eating disorders reportedly suffer from impaired cognitive function, specifically cognitive flexibility. Cognitive rigidity can diminish treatment effectiveness. This paper aimed to study differences in magnesium and thiamine status and cognitive flexibility in patients with binge-purging and restricting eating disorders and controls, as well as the relationship between these variables. Micronutrient levels and cognitive flexibility were expected to be lower for both groups of patients with eating disorders than controls. A positive relationship between cognitive flexibility and micronutrient levels was expected. The clinical sample for this study consisted of patients with anorexia nervosa, bulimia nervosa, binge eating disorder and other specified feeding or eating disorder. Controls were matched on sex, age and education level. Cognitive flexibility was measured with neuropsychological tasks (WCST, TMT) and a questionnaire (BRIEF-A-SR). Micronutrients were measured intravenously. Patients also filled in a questionnaire on depressive symptomatology. There was no difference between groups for
performance in the neuropsychological tasks or the micronutrient levels. Both eating disorders groups did show impaired cognitive flexibility compared to controls on the questionnaire. Impaired cognitive flexibility was associated with depressive symptomatology. No relationship was found between the micronutrients and cognitive flexibility. Future research should focus on the effect of different micronutrients and comorbid mood disorders on cognitive flexibility in eating disorders, as this could affect treatment success in people afflicted with eating disorders.
Keywords: Eating Disorders, Cognitive flexibility, Thiamine, Vitamin B1, Magnesium, Wisconsin Card Sorting Test, Trail Making Test
2
Introduction
Eating disorders (anorexia nervosa, bulimia nervosa, binge eating disorder and other specified feeding or eating disorders) affect millions of people worldwide, with a lifetime prevalence as high as 8.4% (3.3-18.6%) for women and 2.2% (0.8-6.5%) for men (Galmiche et al., 2019). Eating disorders are characterized by disturbed eating behaviour that affects physical health or psychosocial functioning (American Psychiatric Association, 2013). Patients with an eating disorder have an elevated mortality risk, particularly when diagnosed with anorexia nervosa (Chesney et al., 2014; Fichter & Quadflieg, 2016). In addition, recovery is difficult: only two-thirds of patients with bulimia nervosa (BN), anorexia nervosa (AN) and binge eating disorder (BED) recover over the span of ten or more years (Eddy et al., 2017; Fichter et al., 2008). As such, resources must be invested in developing better treatment for eating disorders.
Due to the disturbed eating behaviour of patients with an eating disorder (ED), micronutrient status often deviates from the norm; for example, as much as 92.8% of the patients with AN had at least one micronutrient deficiency in the study by Hanachi et al. (2019). Two micronutrients that are closely associated are magnesium and thiamine (vitamin B1), as magnesium is required for thiamine activation (Johnson, 2001). These nutrients are found to be deficient in patients with EDs (Hall et al., 1988; Winston et al., 2000).
Magnesium is a trace element that is important for metabolic processes and neurotransmission (Eastwood, 2003; Gropper et al., 2005). Patients with an ED have an additional risk for magnesium deficiency, as deficiency can develop from excessive vomiting or abuse of laxatives, apart from inadequate intake (Gropper et al., 2005). Studies found 16-25% of patients to be hypomagnesemic, with 16-60% suffering from hypomagnesemia during hospital admission (Birmingham et al., 2004; Hall et al., 1988; Raj et al., 2012). Magnesium deficiency can cause cognitive problems: Hall et al. (1988) found that magnesium deficient patients with an ED suffer from recent memory and
concentration problems. In elderly, healthy participants, 12 weeks of magnesium treatment resulted in better scores on various tasks measuring cognitive ability (Liu et al., 2016). Animal models also suggest that elevating brain magnesium improves learning and memory (Slutsky et al., 2010). Another micronutrient that is often deficient in patients with EDs is thiamine. Thiamine is related to various processes in the human body, notably in processes related to energy metabolism, ion transport and neurotransmission (Gropper et al., 2005). Thiamine deficiency has been detected in patients with AN: for 15-38% of patients a thiamine deficiency was demonstrated, both intravenously and when measuring food intake (Chiurazzi et al., 2017; Hanachi et al., 2019; Winston et al., 2000). Lower thiamine levels are associated with problems in glucose metabolism. Impaired glucose
metabolism is linked with Alzheimer’s disease and cognitive deficits: Alzheimer’s patients have lower thiamine and glucose metabolism levels than controls and these levels are strongly correlated (Sang et al., 2018). In animal models, a thiamine deprived diet was further linked to problems with glucose metabolism (Sang et al., 2018). As such, deficient thiamine levels can severely affect brain function. Research has shown that patients with an ED suffer from impaired cognitive function, particularly executive function (Smith et al., 2018). Executive function can be divided into inhibition, updating (also known as the working memory) and cognitive flexibility, per the model of Miyake et al. (2000). Cognitive flexibility, also called set-shifting, is the ability to shift between tasks and mental sets (Miyake et al., 2000). It allows adaptation to new rules, seeing things from a different perspective and adjustment to changing circumstances (Huizinga et al., 2017). Diamond (2013) states that cognitive flexibility is largely built on the other two main executive functions, as inhibition is needed
3 to deactivate the previous mental set and the working memory is used to activate the new mental set.
Cognitive flexibility is impaired in patients with EDs when compared to the general population (Roberts et al., 2007; Smith et al., 2018; Wang et al. 2019). The patients with EDs can be divided into different subtypes: a binge-purging subtype (AN – binge-purging subtype, BN, BED & other specified eating disorder (OSFED) – binge-purging subtype) and a restricting subtype (AN– restricting subtype, OSFED – restricting subtype): disorder-typical characteristics differ vastly between these two
categories. Findings regarding problems with cognitive flexibility are mixed when splitting the ED group by subtype: some studies do not find any problems with set-shifting in patients with binge-purging EDs, while others do (Wu et al., 2014, Aloi et al., 2015). Because of these mixed findings, more research is still needed to determine if there is a difference between ED subtypes for cognitive flexibility.
The findings above lead to the conclusion that thiamine and magnesium deficiencies are present in patients with EDs and that deficiencies are associated with cognitive impairment. On the other hand, cognitive flexibility is diminished in patients with eating disorders. Therefore, it is important to study the effect of magnesium and thiamine deficiencies on cognitive flexibility in patients with EDs. Examining cognitive inflexibility in patients with EDs is important, as cognitive rigidity can cause patients to remain stuck in familiar thought patterns and diminish the effect of treatment (Johnco et al., 2013). Furthermore, when looking at the data regarding thiamine and magnesium status in patients with ED, BED and other specified eating of feeding disorder (OSFED) patients are rarely studied. In addition, studies on differences between cognitive flexibility in ED subtypes have led to mixed findings. This study examined magnesium and thiamine status and cognitive flexibility in patients with restricting EDs, patients with binge-purging EDs and healthy controls, as well as the relationship between magnesium and thiamine and cognitive flexibility in these groups. It was hypothesized that magnesium levels, thiamine levels and cognitive flexibility would differ between all groups and that there was a positive relationship between thiamine and magnesium and cognitive flexibility.
In this cross-sectional observational study, 98 patients with binge-purging EDs (AN, BN, BED and OSFED) were compared to 50 patients with restricting EDs (AN, OSFED) and 37 healthy participants matched on education level, gender and age. Micronutrient status was measured intravenously. Cognitive flexibility was measured via neuropsychological tasks (Wisconsin Card Sorting Test & Trail Making Test) and a questionnaire (Behavior Rating Inventory of Executive Function). Cognitive flexibility was expected to be impaired in both groups of patients with an ED, resulting in worse performance on the neuropsychological tasks and a higher score on the questionnaire. Lower levels of magnesium and thiamine were expected in the binge-purging and the restricting group. As various papers have shown that magnesium and thiamine are involved with cognition and deficiencies lead to cognitive impairment, a positive relationship between micronutrient levels and parameters of cognitive flexibility was expected.
Materials & methods
Recruitment
The clinical group was recruited at GGZ Rivierduinen Eetstoornissen Ursula (Leiden, Netherlands). Patients were diagnosed by clinicians following DSM-V and consequently divided into a restricting (AN – restricting subtype, OSFED – restricting subtype) and binge-purging (AN – binge-purging
subtype, BN, BED, OSFED – binge purging subtype) group. Control subjects were recruited via internal channels of communication at GGZ Rivierduinen and with flyers and posters at various
4 establishments in Leiden. All participants were over the age of 18. Controls were matched for
education level, sex and age with the clinical group; controls were excluded for having a (history of an) ED. Inclusion factors for the clinical group were having AN, BN, BED or OSFED.
Participants received comprehensive information about the study before signing an informed consent form. The study was approved by Medisch-Ethische Toetsingscommissie Leiden Den Haag Delft (the Medical-Ethical Committee Leiden Den Haag Delft). Control subjects received
compensation of €10 for participation and received the lab results of the blood test.
Procedure
The experimental procedure consisted of three parts: completing a questionnaire, neuropsychological tasks conducted at GGZ Rivierduinen and venepuncture at LUMC.
Participants first filled in a battery of questionnaires; the clinical group filled in a physical version and the control group an online version. All questionnaires were in Dutch, as this was the native language of the participants.
Behavior Rating Inventory of Executive Function Adult Self-Report version (BRIEF-A-SR)
The BRIEF-A-SR is a self-report inventory of executive function. The inventory consists of 75 items that measure 9 scales. A higher score on a scale indicates more difficulties with this aspect of executive functioning. In the current study, the score on the shift scale was used, as this is a measure for cognitive flexibility. The BRIEF-A-SR has high internal consistency, with a Cronbach’s alpha between 0.93 and 0.96 (Roth & Gioia, 2005).
Eating Disorder Examination Questionnaire (EDE-Q)
The EDE-Q is a self-report questionnaire for eating disorder psychopathology. It consists of 22 items divided over 4 subscales, rated on a forced-choice scale between 0 and 6. A higher global score reflects more severe eating disorder psychopathology; the questionnaire can thus be used as an index for eating disorder severity (Fairburn & Beglin, 2008). The EDE-Q has high internal validity, with a Cronbach’s alpha of 0.95 (Aardoom et al., 2012). The global score is a highly accurate way of discriminating between individuals with and without an ED, with a clinical cut-off score of 2.17 being calculated in Dingemans & Van Furth, 2017. This cut-off was used as an exclusion criterion for the healthy control group in the current study.
Inventory of Depressive Symptomatology – Self Rated (IDS-SR)
The IDS-SR is a self-report questionnaire for depressive symptoms consisting of 28 items. The score can range from 0 to 84, with a higher score indicating more depressive
symptomatology (Rush et al., 1986). Cronbach’s alpha was found to range between 0.76 and 0.94 (Rush et al., 1996). The IDS-SR was used to monitor for mood disorders, as they could be a factor influencing cognitive flexibility (Abbate-Daga et al., 2015).
On-site, participants were subjected to a short interview about age and body height and weight, as well being asked about possible conditions that could affect testing. Height and weight were used to calculate BMI. After the interview, the Nederlandse Leestest voor Volwassenen was conducted.
Nederlandse Leestest voor Volwassenen (NLV)
The NLV is the Dutch version of the National Adult Reading Test (NART) and consists of 50 words with an irregular pronunciation. The number of words correctly pronounced is converted to an IQ score (Schmand et al., 1991). IQ might predict performance on executive function tasks; these should be comparable for controls and patients, as there is no evidence for impaired intellectual functioning in patients with an ED (Steinglass & Glasofer, 2011).
5 This was followed by the completion of four neuropsychological tasks on a computer to determine cognitive flexibility. Participants performed the Wisconsin Card Sorting Test and the Trail Making Test. The two other tasks assessed other areas of executive functioning and are thus not included in results. The tasks were carried out using CDLJava (Buro Tester, 2006). Each task was prefaced by an oral explanation of the task in Dutch by the experimenter.
Wisconsin Card Sorting Test (WCST)
The WCST is a rule switching task that measures cognitive flexibility. Participants were asked to sort a stack of cards displaying an image into four categories. Cards could be sorted based on colour, shape and number. Participants were able to find out the correct rule based on feedback. The sorting rule changed during the task; participants thus had to switch between different approaches during the task (Heaton et al., 1993). WCST performance was quantified by the number of perseverative errors; perseverative errors are sorting errors that remain after feedback was received that indicated the rule had changed (Smith et al., 2018). For an illustration of the WCST, see appendix B.
Trail Making Test (TMT)
TMT is a task-switching task that measures cognitive flexibility. This study used a modified version of the TMT. In part A1, participants had to connect a series of circles containing a number in numerical order. In part A2, participants connected a series of circles containing a letter in alphabetical order. In part B, participants had to alternate between numbers and letters: 1-A, 2-B, etc. The ratio between the time taken for part A1 and part B was used as a measure for set-shifting ability (Reitan, 1953).
After the tasks, subjects were redirected to the nearby hospital for venepuncture. Blood samples were drawn at the phlebotomy clinic of the Leids Universitair Medisch Centrum (LUMC) by way of venepuncture. Samples were two 4ml whole blood EDTA tubes and one 8.5 ml SST tube.
Micronutrient levels were then assessed by the Department of Clinical Chemistry of the LUMC. Thiamine was measured as whole blood vitamin B1, magnesium was assessed in serum. Whole blood is a reliable way to measure vitamin B1, as only a small part of total thiamine is present within serum (Gropper et al., 2005).
Data analysis
Data analysis was performed in SPSS Statistics for Windows (version 22.0). To determine if there were differences in cognitive flexibility and micronutrient status between participant groups (restricting EDs (RES), binge-purging EDs (BP) and controls), a multivariate ANOVA (MANOVA) was used. Participant group was used as the categorical predictor variable; magnesium and thiamine level, perseverative error on the WCST, ratio in the TMT and BRIEF-A-SR cognitive flexibility score were used as continuous outcome variables. If the MANOVA was significant, a follow up univariate ANOVA would indicate which variables significantly differed by group. This was then followed by the Bonferroni’s post hoc test to show which of the three groups differed from each other. The
Bonferroni’s post hoc is used, as this test is powerful when a small number of means is compared (Field, 2013). The Bonferroni’s post hoc test was valid on the condition that the assumption of homogeneity of covariances and variances was met. This was tested using a Box’s test and a Levene’s Test, respectively. If the assumptions were not met, the MANOVA would be followed by the Games-Howell post hoc test.
Micronutrient data was partly incomplete for thiamine levels. Therefore, a second MANOVA on the differences between participant groups for magnesium and cognitive flexibility measures (TMT ratio, WCST perseverative errors, BRIEF-A-SR score) was performed to negate the data loss in the original
6 MANOVA. The follow-up MANOVA allowed for inclusion of more participants and offers a better alternative than several univariate ANOVA’s, as these would not preserve interaction effects. To examine the relationship between micronutrient status and executive function, a multiple regression analysis was performed. Thiamine or magnesium levels were used as the independent variable; perseverative errors in the WCST, the TMT ratio, or the BRIEF-A-SR cognitive flexibility score were used as the dependent variable. Apart from micronutrient level, IDS score and age were also incorporated in the model as possible confounding factors: mood disorders and age have been found to influence cognitive flexibility (Abbate-Daga et al., 2015; Wecker et al., 2005). Incomplete data was excluded using pairwise exclusion. Predictors were entered using a blockwise entry, starting with the micronutrient, followed by age and IDS score. For the multiple regression, residual terms should be uncorrelated: this was tested with the Durbin-Watson test. Additionally, the assumption of
homoscedasticity and no multicollinearity must be met. If the assumption of homoscedasticity was not met, the weighted least squared regression would be used.
As a regression analysis was performed with all 3 cognitive flexibility measures for both
micronutrients, a total of 6 tests was performed. In addition to the MANOVA’s, an adjusted α level of 0.0063 (0.05/8) was used to control for Type I error.
Results
Participants
228 individuals took part in this study. A total of 43 participants were excluded from the analyses: for 37 participants, magnesium and thiamine levels were not available, 2 patients had an avoidant restrictive food intake disorder and 4 controls had an EDEQ score above the clinical cut-off. The final study sample consisted of 50 RES patients (0 male; 32 AN-RES and 18 OSFED-RES), 98 BP patients (4 male; 22 AN-BP, 21 OSFED-BP, 26 BN, 29 BED) and 37 healthy controls (1 male). Mean age, BMI and IQ for these groups are shown in Table 1. An ANOVA indicated that there was no significant
difference between the groups for IQ (F(2, 178) = .198, p = .82). Age did significantly differ between the groups (F(2, 178) = 5.13, p = .007), as did BMI (F(2, 178) = 19.2, p < .001) and IDS score (F(2, 151) = 93.2, p <.001). The RES group (27 ± 8.4 (mean ± SD)) was significantly (p = .006) younger than the control group (34 ± 15.2), but groups did not significantly differ from the binge-purging group (30 ± 10.3). The RES group (17 ± 4.0) also had significantly (p = <.001) lower BMI than the control group (23 ± 3.4), but the BP group (25 ± 9.6) did not significantly differ from other groups. IDS score was significantly (p < .001 for both) higher than the control group (7 ± 7) for both the RES (36 ± 13) and BP group (36 ± 11).
Table 1
Means and standard deviation for different participant groups for age, BMI, IQ and IDS score.
Diagnosis N Age BMI IQ IDS
RES 50 27 ± 8.4# 17 ± 4.0# 99 ± 9 36 ± 13#
BP 98 30 ± 10.3 25 ± 9.6 100 ± 10 36 ± 11*
Control group 37 34 ± 15.2# 23 ± 3.4# 100 ± 12 7 ± 7#*
Note. RES = patients with a restricting eating disorder. BP = patients with a binge-purging eating disorder. IDS =
Inventory of Depressive Symptomatology – Self Rated. Mean ± standard deviation is shown. # = difference between the RES and control group at p < .05.
7
Differences between groups
For the comparison of the differences between the RES, BP and control groups for B1, Magnesium, BRIEF-A-SR cognitive flexibility, TMT ratio and WCST perseverative error, data was complete for 66 BP, 39 RES and 37 controls. The homogeneity of variances was equal for all variables, except for the BRIEF-A-SR cognitive flexibility scores (F(2,139) = 3.95, p = .02). The homogeneity of covariances was not equal (M = 47.1, F(30, 40885) = 1.48, p = .04). Assumptions were therefore not met, and the Games-Howell post hoc test was used. The multivariate ANOVA resulted in a Wilk’s lambda that showed a significant effect of participant group on the micronutrient levels and the cognitive
flexibility measures (Λ = 0.625, F(10, 270) = 7.15, p < .001). The univariate ANOVA showed that there was only a significant difference for the BRIEF-A-SR cognitive flexibility score between groups (F(2, 139) = 32.7, p < .001). Means, standard deviations and significance levels for differences between groups for the variables are shown in Table 2. The Games-Howell post hoc test revealed that the BP group (65.9 ± 11.0) and the RES group (65.3 ± 12.6) scored significantly higher than the control group (47.2 ± 8.08, p < .001 for both) for the BRIEF-A-SR cognitive flexibility score. The BP and RES group did not significantly differ from each other (p = .591). The results for these BRIEF-A-SR scores are
illustrated in Figure 1. Table 2
Descriptive data and significance level for differences in thiamine, magnesium, TMT, WCST and BRIEF-A-SR per group from the first MANOVA.
Parameter RES (n = 39) BP (n = 66) Controls (n = 37) p
Thiamine (nmol/L) 126 ± 30.4 139 ± 37.3 139 ± 26.6 .11 Magnesium (mmol/L) 0.881 ± 0.066 0.858 ± 0.070 0.861 ± 0.066 .21
TMT (ratio) 1.37 ± 0.34 1.47 ± 0.45 1.39 ± 0.29 .39
WCST (perseverative errors) 16.1 ± 8.42 18.0 ± 7.66 15.4 ± 6.04 .18 BRIEF-A-SR (cognitive flexibility
score)
65.3 ± 12.6 62.9 ± 11.0 47.2 ± 8.08 <0.001 Note. RES = patients with a restricting eating disorder. BP = patients with a binge-purging eating disorder. TMT = Trail making test. WCST = Wisconsin Card Sorting Test. Mean ± standard deviation is shown. p from the univariate test statistics that show whether groups significantly differ from each other.
Figure 1. Score for the BRIEF-A-SR cognitive flexibility by group. The total score for the cognitive flexibility parameter of the BRIEF-A-SR is shown by BP (binge purging), RES (restricting) and control group.* = p < .001.
The first MANOVA was followed by a MANOVA including only magnesium and the cognitive flexibility measures. Data was now complete for 94 BP, 48 RES and 37 controls. The homogeneity of
8 not equal for the cognitive flexibility test scores (F(2, 176) = 6.08, p = .003), meaning the Games-Howell post hoc test was used. The analyses yielded comparable results to the first MANOVA, with still only the BRIEF-A-SR cognitive flexibility score differing significantly between the groups. The means, standard deviations and significance levels of the differences between groups are shown in appendix A.
Relationship between micronutrient status and cognitive flexibility
Thiamine
Parametric multiple linear regressions were performed to examine the relationship between
dependent variables TMT ratio (Durbin-Watson (DW) = 2.16), perseverative errors on the WCST (DW = 2.16) and BRIEF-A-SR cognitive flexibility score (DW = 1.91) and independent variables thiamine, IDS score and age. In Table 3 the results of the regression analyses are shown. There were no
significant effects of the independent variables for TMT ratio and perseverative errors on the WCST. Table 3
Linear hierarchical regression model of thiamine, age and IDS as independent variables for TMT ratio, WCST perseverative errors and BRIEF-A-SR cognitive flexibility.
b 95% CI p TMT ratio Step 1 (R2 = .005, F (1, 124) = 0.59, p = .44) Constant 1.3 1.01 - 1.59 <.001 Thiamine (nmol/L) 0.001 -0.001 – 0.003 .44 Step 2 (R2 = .015, F (3, 122) = 0.61, p = .61) Constant 1.3 0.85 – 1.61 <.001 Thiamine (nmol/L) 0.001 -0.001 – 0.003 .50 Age (years) 0.003 -0.003 – 0.010 .31 IDS score 0 -0.005 – 0.004 .84 WCST perseverative errors Step 1 (R2 = 0, F (1, 124) = 0, p = .99) Constant 17.0 11.17 – 22.74 <.001 Thiamine (nmol/L) 0 -0.04 – 0.04 .99 Step 2 (R2 = .017, F (3, 122) = 0.61, p = .56) Constant 13.5 5.90 – 21.0 <.001 Thiamine (nmol/L) -.0001 -0.04 – 0.04 .97 Age (years) 0.084 -0.04 – 0.21 .19 IDS score 0.038 -0.05 – 0.13 .39
BRIEF-A-SR cognitive flexibility
Step 1 (R2 = 0, F (1, 124) = 0.045, p = .83) Constant 59.3 49.61 – 69.04 <.001 Thiamine (nmol/L) 0.007 -0.06 – 0.08 .83 Step 2 (R2 = .335, F (3, 122) = 20.5, p = <.001) Constant 44.0 33.58 – 54.47 <.001 Thiamine (nmol/L) 0.017 -0.04 – 0.08 .55 Age (years) 0.014 -0.16 – 0.19 .16 IDS score 0.466 0.35 – 0.59 <.001
Note. 95% CI = 95% confidence intervals. p shows the significance of the b values being different from 0 by t-test. Statistics next to the step names are from the ANOVA that predicts whether the model is better at predicting the outcome than the best guess, as part of the multiple regression analysis.
9 The regression analysis for BRIEF-A-SR cognitive flexibility score demonstrated that IDS score was a significant predictor for this outcome variable (b = 0.466, p < .001). Age and thiamine level did not have a significant effect on cognitive flexibility in the BRIEF-A-SR. Figure 2 offers further illustration of the relationships between the different variables in this regression.
Magnesium
Parametric multiple linear regressions also were performed to examine the relationship between dependent variables TMT ratio (DW = 2.25), perseverative errors on the WCST (DW = 2.02) and BRIEF-A-SR cognitive flexibility score (DW = 2.03) and independent variables magnesium, IDS score and age. The results of these regression analyses are shown in Table 4. There were no significant effects of these independent variables on TMT ratio and perseverative errors on the WCST. Table 3
Linear hierarchical regression model of magnesium, age and IDS as independent variables for TMT ratio, WCST perseverative errors and BRIEF-A-SR cognitive flexibility.
b 95% CI p TMT ratio Step 1 (R2 = .001, F (1, 150) = 0.13, p = .72) Constant 1.6 0.76 – 2.34 <.001 Magnesium (mmol/L) -0.166 -1.07 – 0.74 .72 Step 2 (R2 = .012, F (3, 148) = 0.62, p = .60) Constant 1.5 0.71 – 2.30 <.001 Magnesium (mmol/L) -0.224 -1.15 - .70 .63 Age (years) 0.004 -0.002 – 0.009 .22 IDS score 0 -0.004 – 0.004 .86 WCST perseverative errors Step 1 (R2 = .002, F (1, 150) = 0.32 , p = .58) Constant 21.5 5.69 – 37.27 .008 Magnesium (mmol/L) -5.18 -23.39 – 13.03 .58 Step 2 (R2 = .012, F (3, 148) = 1.08, p = .36) Constant 19.8 3.96 – 35.68 .02 Magnesium (mmol/L) -7.86 -26.29 – 10.57 .40
Figure 2. Matrix scatterplot of B1 level, IDS score, age and BRIEF-A-SR (survey) cognitive flexibility.
Figure 3.
10
Age (years) 0.090 -0.03 – 0.21 .12
IDS score 0.042 -0.04 – 0.12 .29
BRIEF-A-SR cognitive flexibility
Step 1 (R2 = 0.006, F (1, 150) = 0.920, p = .34) Constant 47.5 21.07 – 73.99 <.001 Magnesium (mmol/L) 14.8 -15.71 – 45.35 .34 Step 2 (R2 = .334, F (3, 148) = 24.7, p = <.001) Constant 43.3 21.31 - 65.27 <.001 Magnesium (mmol/L) 3.67 -21.88 – 29.21 .77 Age (years) 0.014 -0.15 – 0.17 .86 IDS score 0.462 0.35 – 0.57 <.001
Note. 95% CI = 95% confidence intervals. p shows the significance of the b values being different from 0 by t-test. Statistics next to the step names are from the ANOVA that predicts whether the model is better at predicting the outcome than the best guess, as part of the multiple regression analysis.
IDS score remained a significant predictor for BRIEF-A-SR cognitive flexibility (b = 0.462, p < .001) in the regression analysis for BRIEF-A-SR cognitive flexibility. Age and magnesium level did not have a significant effect on cognitive flexibility in the BRIEF-A-SR. Figure 3 offers further illustration of the relationships between the different variables in this regression.
Discussion
The current study examined magnesium and thiamine status and cognitive flexibility in patients with a binge-purging ED, patients with a restricting ED and healthy controls, as well as the relationship between magnesium or thiamine and cognitive flexibility. It was hypothesized that magnesium levels, thiamine levels and cognitive flexibility would differ between the groups and that there was a positive relationship between thiamine or magnesium and cognitive flexibility. The thiamine levels, magnesium levels and performance in the TMT and WCST did not significantly differ between the patients with a binge-purging ED, patients with a restricting ED and the control group. BRIEF-A-SR cognitive flexibility score did significantly differ between groups, as the scores of both groups of patients with EDs indicated more problems with cognitive flexibility than the control group.
Figure 3. Matrix scatterplot of magnesium level, IDS score, age and BRIEF-A-SR (survey) cognitive flexibility.
Figure 4
11 Furthermore, thiamine and magnesium levels were not associated with cognitive flexibility. Only IDS score was found to be a significant positive predictor of BRIEF-A-SR cognitive flexibility score. These results led to the conclusion that there is no evidence that healthy controls, patients with a binge-purging ED or patients with a restricting ED differ in thiamine or magnesium status. Evidence was found that points to impaired cognitive flexibility in patients with a binge-purging ED and patients with a restricting ED compared to controls, although this was dependent on the method of measurement. No evidence was found for a relationship between magnesium status or thiamine status and cognitive flexibility in patients with a restricting or a binge-purging eating disorder or healthy controls.
The findings of the current study regarding micronutrient status were not in accordance with previous research. For thiamine, Chiurazzi et al., (2017), Hanachi et al., (2019) and Winston et al. (2000) all found evidence for low thiamine levels in patients with AN. A possible explanation for this discrepancy is that this study included all types of EDs, as opposed to just AN. Dietary patterns differ between EDs and this can affect micronutrient status (American Psychiatric Association, 2013). Another explanation can be found in that Chiurazzi et al. (2017) and Winston et al. (2000) used a different way of measuring thiamine deficiency, namely via food intake and via the ETK enzyme. Hanachi et al. (2019) used a similar method as the current study, but did not compare the patient group with healthy controls; instead, they reported the number of patients that were B1 deficient. They used <126 nmol/L as a measurement for deficiency, based on values determined from levels healthy adults; when applying this threshold to the results of the current study, 27%-50% of participants would be considered thiamine deficient. However, this definition of deficiency is not in line with the norms used by the Department of Clinical Chemistry of the LUMC and Lu & Frank (2008), as they use 70-78 nmol/L as the measurement for deficiency. Thus, the discrepancy in findings can be explained.
Regarding previous literature on magnesium levels, hypomagnesemia was specifically recorded after refeeding in Raj et al. (2012), but Birmingham et al. (2004) and Hall et al. (1988) found
hypomagnesemia in patients before refeeding. Hall et al. (1988) do remark that a group of patients only showed hypomagnesemic symptoms once rehydrated with a saline concentration. As this had not happened in the current study, hypomagnesemia might have been present in patients, but not measurable via serum magnesium. Further support to the theory that the patients might have been magnesium deficient, but that this was not measurable is by Jahnen-Dechent & Ketteler (2012). They note that magnesium supply can be depleted in the body while serum magnesium levels are hardly affected. This is because serum levels are balanced by bone magnesium. To prevent this, magnesium status could be quantified by urinary excretion or the magnesium retention test in follow-up studies (Jahnen-Dechent & Ketteler, 2012).
The findings that the clinical group did not have significantly lower micronutrient levels than controls could alternatively be explained by the theory that while the intake of patients with EDs is limited, the intake mainly consists of nutrient-rich, healthy foods. Some support for this theory can be found: Lobera et al. (2009) revealed that patients with AN consumed significantly more vegetables and less sweet and fried foods than controls. Chiurazzi et al. (2017) studied micro- and macronutrient intake in patients with AN and found that while macronutrient intake was significantly less in patients than in controls, the intake of several micronutrients did differ.
This study found that patients with a binge-purging ED and patients with a restricting ED both scored higher on the cognitive flexibility section of the BRIEF-A-SR than the control group, which indicates more problems with cognitive flexibility in the clinical group. This is noteworthy, as measures for
12 cognitive flexibility in the form of neuropsychological tasks did not find differences between groups. A possible explanation for this could be that the BRIEF-A-SR is a self-report inventory and requires participants to reflect on their behaviour, whereas the neuropsychological tasks are less subjective measures. Participants might not be able to properly reflect on their own cognitive flexibility and so cause a discrepancy in findings from different measures. This has been found in literature: Lounes et al. (2011) found no correlation between scores on a self-report cognitive flexibility test and an experimental measure for cognitive flexibility. This theory might also be supported by the finding that a higher IDS score was a predictor for cognitive flexibility dysfunction per the BRIEF-A-SR. Higher IDS scores reflect more severe depressive symptomatology; depressive symptoms like hopelessness and low self-esteem can induce a negative self-evaluation bias, which might cause participants to give their own cognitive abilities a lower rating. This effect has been found in older adults in Crane et al. (2007), where depressive symptoms, specifically negative thinking, were correlated with lower self-assessment scores for memory.
Another explanation for the different findings regarding cognitive flexibility is that the BRIEF-A-SR inventory assesses different parameters than neuropsychological tasks. Several other studies found that scores on the BRIEF-A-SR indicated problems with executive function in patients, while scores on neuropsychological tasks did not (Rouel et al., 2016; Dingemans et al., 2019). Dingemans et al. (2019) suggest that the BRIEF-A-SR might be more accurate than neuropsychological tasks, as these tasks are administered in controlled environments and thus do not reflect problems with executive
function in real life. Furthermore, Dingemans et al. (2019) mention that neuropsychological tasks like the TMT and WCST aim to isolate cognitive flexibility from other executive functions. Per Diamond et al. (2013), cognitive flexibility is built on the other executive functions, inhibition and working
memory. Thus, studying executive functions independent of each other is not ineffable. Future research should invest in a way to study executive function in a realistic environment while recognizing the interdependence of the different constructs of executive function.
The current study had several strengths: the clinical sample was large and included patients with all eating disorders. Patients with BED and OSFED are rarely included in scientific studies on ED, while this is the largest patient group (Galmiche et al., 2019). Additionally, control participants were screened for eating disorder symptomatology, assuring that the control group was representative of the healthy population. Furthermore, this study controlled for factors that could influence
micronutrient status or cognitive flexibility, as control subjects and patients with EDs were matched for education level, sex and age. The presence of mood disorders, a known confounding factor of cognitive flexibility, was also included in analyses and thus controlled for. Additionally,
micronutrients status was measured intravenously, which allowed for more accurate micronutrient status than when studying intake.
Limitations of this study, in addition to the ones mentioned above regarding magnesium
measurement and critique on the neuropsychological tasks, were the relatively small control group. Due to the ongoing COVID-19 pandemic, the recruitment of controls had to be halted during the study. This led to groups not being perfectly matched on age, as well as the control group possibly being too small to be representative of the general healthy population. While the patient group was large, it consisted of patients seeking treatment; this group is not necessarily representative of all people with an ED (Dingemans et al., 2019). Additionally, the number of patients with a binge-purging ED was near double that of patients with a restricting ED. These three factors affect the generalizability of the findings of this study. Future research should take care to use a study sample representative of the general healthy population and the population with EDs. In addition, this study did not control for the use of vitamin supplements or oral contraceptives. Both of these factors could
13 severely impact micronutrient status in participants: oral contraceptives can negatively affect
magnesium levels (Palmery et al., 2013). Thus, it is important to control for, especially considering the majority of patients with ED is female.
The current research findings do not seem to offer support for an association between magnesium or thiamine levels and cognitive flexibility in people with an ED; future research may focus on the relationship between other micronutrients and executive functioning. A small study by Loria-Kohen et al. (2013) already found evidence for positive effects of folic acid supplementation for depressive symptoms and cognitive status in patients with an ED. Moreover, the questionnaire results showed that cognitive flexibility was impaired in patients with EDs as compared to controls. This indicated that patients with EDs suffer from issues with cognitive flexibility in daily life. It is important to treat this, as it impacts quality of life and can hinder ED treatment. A possible line of study for this is the effect of comorbid mood disorders on cognitive flexibility in EDs. Results showed that the clinical group suffered from depressive symptomatology; mood disorders are known to affect cognitive functioning (Abbata-Daga et al., 2015). Furthermore, Giel et al. (2012) has already found that problems with cognitive flexibility found in patients with AN are in part attributable to comorbid mood disorders. In conclusion, studying the effects of different micronutrients and comorbid mood disorders on cognitive flexibility in patients with ED might be a way to improve the recovery rate for people afflicted with EDs.
14
References
Aardoom, J. J., Dingemans, A. E., Slof Op’t Landt, M. C. T., & Van Furth, E. F. (2012). Norms and discriminative validity of the Eating Disorder Examination Questionnaire (EDE-Q). Eating
Behaviors, 13(4), 305–309. https://doi.org/10.1016/j.eatbeh.2012.09.002
Abbate-Daga, G., Buzzichelli, S., Marzola, E., Aloi, M., Amianto, F., & Fassino, S. (2015). Does depression matter in neuropsychological performances in anorexia nervosa? A descriptive review. International Journal of Eating Disorders, 48(6), 736–745.
https://doi.org/10.1002/eat.22399
Aloi, M., Rania, M., Caroleo, M., Bruni, A., Palmieri, A., Cauteruccio, M. A., De Fazio, P., & Segura-García, C. (2015). Decision making, central coherence and set-shifting: a comparison between Binge Eating Disorder, Anorexia Nervosa and Healthy Controls. BMC Psychiatry, 15(1), 6. https://doi.org/10.1186/s12888-015-0395-z
American Psychiatric Association. (2013). Diagnostic and statistical manual of mental disorders (5th ed.). https://doi.org/10.1176/appi.books.9780890425596
Birmingham, C. L., Puddicombe, D., & Hlynsky, J. (2004). Hypomagnesemia during refeeding in anorexia nervosa. Eating and Weight Disorders - Studies on Anorexia, Bulimia and Obesity,
9(3), 236–237. https://doi.org/10.1007/bf03325074
Burotester. (2006). CDLJava [Computer software]. Retrieved from http://www.burotester.nl/cdljava/menu2.html
Chesney, E., Goodwin, G. M., & Fazel, S. (2014). Risks of all-cause and suicide mortality in mental disorders: a meta-review. World Psychiatry, 13(2), 153–160.
15 Chiurazzi, C., Cioffi, I., De Caprio, C., De Filippo, E., Marra, M., Sammarco, R., Di Guglielmo, M. L.,
Contaldo, F., & Pasanisi, F. (2017). Adequacy of nutrient intake in women with restrictive anorexia nervosa. Nutrition, 38, 80–84. https://doi.org/10.1016/j.nut.2017.02.004
Crane, M. K., Bogner, H. R., Brown, G. K., & Gallo, J. J. (2007). The link between depressive symptoms, negative cognitive bias and memory complaints in older adults. Aging & Mental Health,
11(6), 708–715. https://doi.org/10.1080/13607860701368497
Diamond, A. (2013). Executive functions. Annual Review of Psychology, 64(1), 135-168. https://doi:10.1146/annurev-psych-113011-143750
Dingemans, A. E., & van Furth, E. F. (2017). Het meten van verandering tijdens behandeling voor eetstoornissen: een vergelijking van algemene en specifieke vragenlijst [Measuring changes during the treatment of eating disorders: a comparison of two types of questionnaires].
Tijdschrift voor psychiatrie, 59(5), 278–285.
Dingemans, A. E., Vanhaelen, C. B., Aardoom, J. J., & van Furth, E. F. (2019). The influence of depressive symptoms on executive functioning in binge eating disorder: A comparison of patients and non-obese healthy controls. Psychiatry Research, 274, 138–145.
https://doi.org/10.1016/j.psychres.2019.02.033
Eastwood, M. (2003). Principles of Human Nutrition (2nd Edition). Blackwell Science.
Eddy, K. T., Tabri, N., Thomas, J. J., Murray, H. B., Keshaviah, A., Hastings, E., Edkins, K., Krishna, M., Herzog, D. B., Keel, P. K., & Franko, D. L. (2016). Recovery From Anorexia Nervosa and Bulimia Nervosa at 22-Year Follow-Up. The Journal of Clinical Psychiatry, 78(02), 184–189. https://doi.org/10.4088/jcp.15m10393
16 Fairburn, C. G., & Beglin, S. (2008). Eating Disorder Examination Questionnaire (EDE-Q 6.0). In C. G.
Fairburn (Ed.), Cognitive behavior therapy and eating disorders (pp. 309-313). Guilford Press.
Fichter, M. M., & Quadflieg, N. (2016). Mortality in eating disorders - results of a large prospective clinical longitudinal study. International Journal of Eating Disorders, 49(4), 391–401. https://doi.org/10.1002/eat.22501
Fichter, M. M., Quadflieg, N., & Hedlund, S. (2008). Long-term course of binge eating disorder and bulimia nervosa: Relevance for nosology and diagnostic criteria. International Journal of
Eating Disorders, 41(7), 577–586. https://doi.org/10.1002/eat.20539
Field, A. (2013). Discovering Statistics Using IBM SPSS Statistics (4th ed.). SAGE Publications.
Galmiche, M., Déchelotte, P., Lambert, G., & Tavolacci, M. P. (2019). Prevalence of eating disorders over the 2000–2018 period: a systematic literature review. The American Journal of Clinical
Nutrition, 109(5), 1402–1413. https://doi.org/10.1093/ajcn/nqy342
Gibson, G. E., Hirsch, J. A., Fonzetti, P., Jordan, B. D., Cirio, R. T., & Elder, J. (2016). Vitamin B1 (thiamine) and dementia. Annals of the New York Academy of Sciences, 1367(1), 21–30. https://doi.org/10.1111/nyas.13031
Giel, K. E., Wittorf, A., Wolkenstein, L., Klingberg, S., Drimmer, E., Schönenberg, M., Rapp, A. M., Fallgatter, A. J., Hautzinger, M., & Zipfel, S. (2012). Is impaired set-shifting a feature of “pure” anorexia nervosa? Investigating the role of depression in set-shifting ability in anorexia nervosa and unipolar depression. Psychiatry Research, 200(2–3), 538–543.
https://doi.org/10.1016/j.psychres.2012.06.004
Gropper, S. S., Smith, J. L., & Groff, J. L. (2005). Advanced Nutrition and Human Metabolism. Thomson Wadsworth.
17 Hadigan, C. M., Anderson, E. J., Miller, K. K., Hubbard, J. L., Herzog, D. B., Klibanski, A., & Grinspoon,
S. K. (2000). Assessment of macronutrient and micronutrient intake in women with anorexia nervosa. International Journal of Eating Disorders, 28(3), 284–292.
https://doi.org/10.1002/1098-108X(200011)28:3%3C284::AID-EAT5%3E3.0.CO;2-G
Hall, R. C. W., Hoffman, R. S., Beresford, T. P., Wooley, B., Tice, L., & Klassen Hall, A. (1988). Hypomagnesemia in Patients with Eating Disorders. Psychosomatics, 29(3), 264–272. https://doi.org/10.1016/s0033-3182(88)72362-2
Hanachi, M., Dicembre, M., Rives-Lange, C., Ropers, J., Bemer, P., Zazzo, J.-F., Poupon, J., Dauvergne, A., & Melchior, J.-C. (2019). Micronutrients Deficiencies in 374 Severely Malnourished Anorexia Nervosa Inpatients. Nutrients, 11(4), 792. https://doi.org/10.3390/nu11040792
Heaton, R.K., Chelune, G.J., Tallay, J.L., Kay, G.G. & Curtiss, G. (1993). Wisconsin Card Sorting Test. Psychological Assessment Resources.
Huizinga, M., Baeyens, D., Ouwehand, C., & Burack, J. (2017). Executive functions. In S. Linkenauger, E. Geangu, & B. Hopkins (Eds.), The Cambridge Encyclopedia of Child Development (pp. 297– 303). Cambridge University Press. https://doi.org/10.1017/9781316216491.050
Jahnen-Dechent, W., & Ketteler, M. (2012). Magnesium basics. Clinical Kidney Journal, 5(Suppl 1), i3– i14. https://doi.org/10.1093/ndtplus/sfr163
Johnco, C., Wuthrich, V. M., & Rapee, R. M. (2013). The role of cognitive flexibility in cognitive restructuring skill acquisition among older adults. Journal of Anxiety Disorders, 27(6), 576– 584. https://doi.org/10.1016/j.janxdis.2012.10.004
Johnson, S. (2001). The multifaceted and widespread pathology of magnesium deficiency. Medical
18 Jonides, J., & Nee, D. E. (2005, January 1). A representation of the Wisconsin Card Sorting Task
[Illustration]. https://doi.org/10.1093/schbul/sbi053
Kortte, K. B., Horner, M. D., & Windham, W. K. (2002). The trail making test, part B: Cognitive flexibility or ability to maintain set?. Applied Neuropsychology, 9(2), 106-109.
https://doi:10.1207/S15324826AN0902_5
Liu, G., Weinger, J. G., Lu, Z., Xue, F., & Sadeghpour, S. (2015). Efficacy and Safety of MMFS-01, a Synapse Density Enhancer, for Treating Cognitive Impairment in Older Adults: A Randomized, Double-Blind, Placebo-Controlled Trial. Journal of Alzheimer’s Disease, 49(4), 971–990. https://doi.org/10.3233/jad-150538
Lobera, I. J., & Ríos, P. (2009). Choice of diet in patients with anorexia nervosa. Nutricion Hospitalaria,
24(6), 682–687. https://doi.org/10.3305/NH.2009.24.6.4552
Loria-Kohen, V., Gómez-Candela, C., Palma-Milla, S., Amador-Sastre, B., Hernanz, A., & Bermejo, L. M. (2013). A pilot study of folic acid supplementation for improving homocysteine levels,
cognitive and depressive status in eating disorders. Nutricion Hospitalaria, 28(3), 807–815. https://doi.org/10.3305/nh.2013.28.3.6335
Lounes, N., Khan, G., & Tchanturia, K. (2011). Assessment of Cognitive Flexibility in Anorexia Nervosa – Self-Report or Experimental Measure? A Brief Report. Journal of the International
Neuropsychological Society, 17(05), 925–928. https://doi.org/10.1017/s1355617711000671
Lu, J., & Frank, E. L. (2008). Rapid HPLC Measurement of Thiamine and Its Phosphate Esters in Whole Blood. Clinical Chemistry, 54(5), 901–906. https://doi.org/10.1373/clinchem.2007.099077
Miyake, A., Friedman, N. P., Emerson, M. J., Witzki, A. H., Howerter, A., & Wager, T. D. (2000). The Unity and Diversity of Executive Functions and Their Contributions to Complex “Frontal Lobe”
19 Tasks: A Latent Variable Analysis. Cognitive Psychology, 41(1), 49–100.
https://doi.org/10.1006/cogp.1999.0734
National Institute for Health and Care Excellence. (2017). Nutrition support for adults: oral nutrition
support, enteral tube feeding and parenteral nutrition (NICE Guideline CG32).
https://www.nice.org.uk/guidance/cg32
Oudman, E., Wijnia, J. W., Oey, M. J., van Dam, M. J., & Postma, A. (2018). Preventing Wernicke’s encephalopathy in anorexia nervosa: A systematic review. Psychiatry and Clinical Neurosciences,
72(10), 774–779. https://doi.org/10.1111/pcn.12735
Palmery, M., Saraceno, A., Vaiarelli, A., & Carlomagno, G. (2013). Oral contraceptives and changes in nutritional requirements. European review for medical and pharmacological sciences, 17(13), 1804–1813. https://www.europeanreview.org/article/4579
Raj, K. S., Keane-Miller, C., & Golden, N. H. (2012). Hypomagnesemia in Adolescents With Eating Disorders Hospitalized for Medical Instability. Nutrition in Clinical Practice, 27(5), 689–694. https://doi.org/10.1177/0884533612446799
Reitan, R. (1956). Trail Making Test: Manual for Administration, Scoring, and Interpretation. Indiana University Press.
Roberts, M. E., Tchanturia, K., Stahl, D., Southgate, L., & Treasure, J. (2007). A systematic review and meta-analysis of set-shifting ability in eating disorders. Psychological Medicine, 37(8), 1075– 1084.https://doi.org/10.1017/s0033291707009877
Roth, R. M., & Gioia, G. A. (2005). Behavior rating inventory of executive function--adult version. Psychological Assessment Resources.
20 Rouel, M., Raman, J., Hay, P., & Smith, E. (2016). Validation of the Behaviour Rating Inventory of
Executive Function – Adult Version (BRIEF-A) in the obese with and without binge eating disorder. Eating Behaviors, 23, 58–65. https://doi.org/10.1016/j.eatbeh.2016.07.010
Rush, A. J., Gullion, C. M., Basco, M. R., Jarrett, R. B., & Trivedi, M. H. (1996). The Inventory of Depressive Symptomatology (IDS): psychometric properties. Psychological Medicine, 26(3), 477–486. https://doi.org/10.1017/s0033291700035558
Rush, A. J., Giles, D. E., Schlesser, M. A., Fulton, C. L., Weissenburger, J., & Burns, C. (1986). The inventory for depressive symptomatology (IDS): Preliminary findings. Psychiatry Research,
18(1), 65–87. https://doi.org/10.1016/0165-1781(86)90060-0
Sang, S., Pan, X., Chen, Z., Zeng, F., Pan, S., Liu, H., Jin, L., Fei, G., Wang, C., Ren, S., Jiao, F., Bao, W., Zhou, W., Guan, Y., Zhang, Y., Shi, H., Wang, Y., Yu, X., Wang, Y., & Zhong, C. (2018). Thiamine diphosphate reduction strongly correlates with brain glucose hypometabolism in Alzheimer’s disease, whereas amyloid deposition does not. Alzheimer’s Research & Therapy, 10(1), 26. https://doi.org/10.1186/s13195-018-0354-2
Schmand, B., Bakker,D., Saan, R. & Louman, J. (1991). De Nederlandse Leestest voor Volwassenen: een maat voor het premorbide intelligentieniveau. Tijdschrift voor gerontologie en geriatrie,
22(1), 15-19.
Slutsky, I., Abumaria, N., Wu, L.-J., Huang, C., Zhang, L., Li, B., Zhao, X., Govindarajan, A., Zhao, M.-G., Zhuo, M., Tonegawa, S., & Liu, G. (2010). Enhancement of Learning and Memory by Elevating Brain Magnesium. Neuron, 65(2), 165–177. https://doi.org/10.1016/j.neuron.2009.12.026
Smith, K. E., Mason, T. B., Johnson, J. S., Lavender, J. M., & Wonderlich, S. A. (2018). A systematic review of reviews of neurocognitive functioning in eating disorders: The state-of-the-literature and future directions. International Journal of Eating Disorders, 51(8), 798–821.
21 Steinglass, J. E., & Glasofer, D. R. (2011). Neuropsychology. In I. Frampton & B. Lask (Eds.), Eating
Disorders and the Brain (pp. 106–121). Wiley-Blackwell.
Talbot, A., Hay, P., Buckett, G., & Touyz, S. (2014). Cognitive deficits as an endophenotype for anorexia nervosa: An accepted fact or a need for re-examination? International Journal of
Eating Disorders, 48(1), 15–25. https://doi.org/10.1002/eat.22332
Wang, S. B., Gray, E. K., Coniglio, K. A., Murray, H. B., Stone, M., Becker, K. R., Thomas, J. J., & Eddy, K. T. (2019). Cognitive rigidity and heightened attention to detail occur transdiagnostically in adolescents with eating disorders. Eating Disorders, 1–13.
https://doi.org/10.1080/10640266.2019.1656470
Wecker, N. S., Kramer, J. H., Hallam, B. J., & Delis, D. C. (2005). Mental Flexibility: Age Effects on Switching. Neuropsychology, 19(3), 345–352. https://doi.org/10.1037/0894-4105.19.3.345
Winston, A. P., Jamieson, C. P., Madira, W., Gatward, N. M., & Palmer, R. L. (2000). Prevalence of thiamin deficiency in anorexia nervosa. International Journal of Eating Disorders, 28(4), 451– 454. https://doi.org/10.1002/1098-108X(200012)28:4%3C451::AID-EAT14%3E3.0.CO;2-I
Wu, M., Brockmeyer, T., Hartmann, M., Skunde, M., Herzog, W., & Friederich, H.C. (2014). Set-shifting ability across the spectrum of eating disorders and in overweight and obesity: a systematic review and meta-analysis. Psychological Medicine, 44(16), 3365–3385. https://doi.org/10.1017/s0033291714000294
22
Appendix A
Table A
Descriptive data and significance level for differences in magnesium, TMT, WCST and BRIEF-A-SR per group from the second MANOVA.
Parameter RES (n = 48) BP (n = 94) Controls (n=37) p Magnesium (mmol/L) 0.878 ± 0.069 0.858 ± 0.068 0.861 ± 0.066 .26 TMT (ratio) 1.37 ± 0.38 1.44 ± 0.42 1.39 ± 0.29 .53 WCST (perseverative errors) 16.1 ± 8.33 18.1 ± 7.85 15.4 ± 6.04 .11 BRIEF-A-SR (cognitive flexibility
t-score)
64.8 ± 13.3 62.7 ± 11.20 47.2 ± 8.08 <0.001 Note. RES = patients with a restricting eating disorder. BP = patients with a binge-purging eating disorder. TMT = Trail making test. WCST = Wisconsin Card Sorting Test. Mean ± standard deviation is shown. p from the univariate test statistics that show whether groups significantly differ.
23
Appendix B
Figure B. Illustration of the Wisconsin Card Sorting Test. The bottom card can be sorted into one of the 4 stacks; this can be by colour (blue), number (two) and shape (plus). After sorting, participant will receive feedback on if the way of sorting was correct or not. If participant keeps sorting the card into the same stack after negative feedback, this counts as a
