Impaired neurocognitive and psychomotor performance in patients with inflammatory bowel disease

Limited evidence exists regarding cognitive and psychomotor function in patients with inflammatory bowel disease (IBD). Therefore, we aimed to compare the neurocognitive and psychomotor function of 60 IBD patients with 60 age/sex-matched controls. Computer-based instrument Complex Reactinometer Drenovac (CRD) was used for assessment of cognitive domains: convergent thinking (simple mathematical tasks; CRD-11), perceptive abilities (light signal position discrimination; CRD-311) and sophisticated operative thinking (complex psychomotor coordination; CRD-411). The most important analyzed parameters were total test solving time (TTTS); minimal time of particular test solving (TMIN) and total number of wrong reactions (NER). Performance in all three cognitive tests showed statistically significantly longer TTTS and TMIN in IBD patients (P < 0.05), while there was no significant difference in NER. Aforementioned findings were adjusted for BMI, age and duration of education. Our study has shown impaired neurocognitive and psychomotor function in IBD patients compared to controls, especially in mental processing speed and mental endurance of perceptive abilities, convergent thinking and complex operative thinking.

The Mayo endoscopic score (MES) is a reliable instrument for UC staging based only on endoscopic evaluation. Disease can be classified in several categories: 0 -inactive disease (normal mucosa); 1 -mild disease (erythema, decreased vascular pattern, and mild friability); 2 -moderate disease (marked erythema, absence of vascular patterns, friability and erosions); 3 -severe disease (spontaneous bleeding and ulceration) 28 .
The Mayo score/disease activity index (Mayo/DAI) is a clinical score used for the assessment of UC activity. The summary score provides good correlation with UC severity: <2 remission (with no individual subscore >1); 3-5 mild; 6-10 moderate; and 10-12 severe 29 .

Assessment of cognitive and psychomotor function. The Complex Reactionmeter Drenovac (CRD)
is a well-established psychodiagnostic instrument, specifically designed for the detailed evaluation of cognitive and psychomotor function. The concept of CRD-series tests is based on chronometric principles [30][31][32][33][34] . We have utilized 3 representative tests in our study, namely CRD-311 (light signal position discrimination), CRD-11 (simple mathematical tasks) and CRD-411 (complex psychomotor coordination tasks). The aforementioned tests enable a focused assessment of convergent thinking, perceptive abilities and sophisticated operative thinking, respectively 30 .
The CRD-311 test contains 60 single tasks in which subjects must provide an accurate answer by pressing a switch below the corresponding randomly illuminating light-emitting diode. The CRD-11 consists of mathematical symbols (addition or subtraction), randomly illuminating light-emitting diodes and corresponding numbers positioned on the test dashboard, which are used for the construction of simple mathematical tasks, and requires test solving by pressing the button that corresponds to the correct result of the constructed mathematical operations. Finally, the CRD-411 test requires simultaneous limb coordination (arm and leg) in reaction to the random illumination of two light-emitting diodes, each corresponding to one limb 30 .
The most important parameters derived from the CRD-series test include basic time variables (total test solving time [ . The aforementioned parameters are descriptors of mental processing speed and mental endurance (T TTS ; T AVER ; T 1 ; T 2 ; T 3 ; T 4 ), maximal cognitive potential (T MIN ), accuracy of mental processing and alertness (N ER ), as well as "wasted time"/stability/fluctuation (TB) of cognitive function 30,31 . Higher test scores (T TTS ; T MIN ) indicate poorer cognitive and motor performance 31 .
We focused on testing specific cognitive domains that were previously tested in the IBD population as well in other chronic diseases [30][31][32][33][34] . Before the test initiation, all subjects received detailed instructions and were able to perform a trial test. The test was conducted during the morning in a sufficiently illuminated, translucent and quiet room, to eliminate possible distractions. The test sequence was conducted according to the test complexity, from the most simple to the most complicated. Testing was conducted by an experienced technician who was "blinded" to the group type (patients or control subjects) during the experiment. Importantly, all subjects were instructed to sleep at least 8 hours during the night prior to the testing. Similarly, smoking and alcohol consumption was forbidden on the testing day for all subjects.
Sample collection and laboratory analysis. Venous blood samples were taken after the performance of the CRD-series test through a polyethylene catheter inserted into a forearm vein. Hemoglobin levels (Hb) and high sensitivity C-reactive protein (hs-CRP) were analyzed with routine laboratory methods in the same biochemical laboratory and by the same specialist in medical biochemistry. Faecal calprotectin (FC) concentrations had been measured from stool samples after appropriate collection, following standard laboratory protocols, within one week after psychometric testing. Statistical analysis. MedCalc statistical software (Ostend, Belgium; version 11.5.1.0) for Windows was used for statistical data analysis. We assessed the normality of data distribution graphically and by the Kolmogorov-Smirnov test. Data were expressed as means ± standard deviation (SD) for continuous parametric variables and as whole numbers and percentages for categorical variables. The Student's t-test was used for the analysis of independent continuous data. We used the Chi-square test for the comparison of sex differences, smoking status, alcohol and coffee consumption between the IBD and control group. Moreover, we compared differences in disease activity, extraintestinal manifestations and surgical history between CD and UC subgroups by the Chi-square test. Multiple linear regression analysis with a stepwise selection algorithm was used to determine the relative importance of independent variables (Hb, hs-CRP, FC, BMI, age, duration of education, CDAI, Mayo/ DAI) in the prediction of CRD parameters. Variables that did not reach statistical significance were excluded from the model. Analysis of covariance (ANCOVA) was used to compare group differences in CRD parameters controlling for important covariates including age, duration of education and body mass index (BMI). The aforementioned variables were included in the ANCOVA model due to their relevance to cognitive function. Levene's test for the equality of variances was performed prior to the ANCOVA analysis. All assumptions for the ANCOVA analysis were confirmed. Finally, to investigate the difference in test solving times within and between groups according to the different test phases, a two-factor (group; test phase) repeated measures ANOVA was carried out. We estimated sphericity using the Greenhouse-Geisser method. If the estimate of sphericity (epsilon) wass greater than 0.75, the assumption of sphericity was met. If not, we used the Greenhouse-Geisser correction factor. The statistical significance was defined as P < 0.05.

Test of convergent thinking (CRD-11 -simple mathematical tasks).
The CRD-11 test consists of simple arithmetic operations and is a measure of decision making speed, problem solving ability and handling difficult situations. Compared to the control group, IBD patients achieved statistically significantly worse values in most parameters of the CRD-11 test, except for N ER which did not reach statistical significance. Moreover, after controlling for covariates, all parameters retained statistical significance, with the exception of TB ( Table 3).

Test of sophisticated operative thinking (CRD-411 -complex psychomotor coordination).
The CRD-411 measures speed of perception, complex mental processing, detection accuracy, spatiovisual orientation, concentration and complex psychomotoric reaction times. All results from the CRD-411 test panel were statistically significantly deteriorated in the observed IBD sample compared to control subjects, except for T3 and N ER which lacked statistical significance. Importantly, after controlling for confounding variables, none of the parameters lost statistical significance. Additionally, N ER proved to be higher in IBD patients (10.64 ± 8.34 vs. 8.25 ± 7.90, P = 0.040) ( Table 3).

Analysis of the IBD subgroups.
There was no difference in CRD test parameters between patients with CD and UC, except in T MIN of CRD-311 which was longer among patients with UC (0.47 ± 0.08 vs. 0.43 ± 0.08, P = 0.045) ( Table 4).  The within and between-group differences in test solving times regarding different test phases of CRD tests. There was a significant difference between the IBD and control group regarding time responses in different test phases (P < 0.05). Moreover, there was a significant within-group difference in time responses between different test phases (P < 0.001) ( Table 8).

Discussion
Our study has shown that patients with IBD have impaired neurocognitive and psychomotor function compared to the control subjects in all measured domains including perceptive abilities, convergent thinking and sophisticated operative thinking. The aforementioned findings were present irrespective of age, duration of education and BMI. To our knowledge, this is the first study that used the CRD-series for cognitive and psychomotor evaluation in IBD patients. Even though there are several studies evaluating cognitive function in patients with IBD, our study adds important findings to the existing evidence due to the important methodological strength of the CRD-series tests. To date, only Berrill et al. performed a similar, fully computer-based study on an entire population of IBD patients. However, they did not establish an independent significant difference in cognitive and psychomotor parameters between the observed group and healthy subjects reassuring the need for further studies 18   Our study utilized CRD tests for detailed neurocognitive and psychomotor assessment. Multiple studies have shown that the CRD-series tests offer validated, reliable, objective and reproducible assessment of an individual's cognitive and psychomotor characteristics [30][31][32][33][34] . Its language independence, versatility and integrated test generator minimise psychometric bias allowing for multiple retesting in different age and ethnic categories 30 . Most importantly, the high sensitivity of CRD tests allows the detection of minimal changes in cognitive and psychomotor performance.
Several questionnaire-based studies assessed cognitive function in IBD patients. Attree et al. conducted a study on a group of IBD patients with interesting results showing a specific deficit in verbal IQ among the studied group. Furthermore, they found deficits in the object recognition domain of cognitive functioning, with paradoxical superiority in the spatial recognition domain among IBD patients 12 . These findings were subsequently confirmed by Dancey et al. 15 . Similarly, Kennedy et al. did not show visuospatial memory impairment among CD patients Yet, they found a cognitive deficit in the Stroop test which partially evaluates the object recognition function 9 . Moreover, a study on adolescent IBD patients showed borderline significant findings of minor verbal deficits in comparison to age-matched patients with juvenile idiopathic arthritis. However, they did not find evidence of global cognitive impairment in adolescents with IBD 19 . Similarly, Wells et al. did not find any cognitive impairment in the IBD sample using the same Wechsler Adult Intelligence Scale as in the aforementioned studies 14 .   16 . Langenberg et al. provided insights that CD patients exhibit subtle cognitive impairment displayed by a prolonged response time in both the mental processing and decision making domain, but not with error rates 17 . Consistent findings for the aforementioned cognitive domains were obtained in our study indicating compensatory prolonged decision making and data processing time, with a preserved overall efficiency and accuracy in simple cognitive tasks.
Finally, a related computer-based study by Berrill et al. did not find any statistically significant difference in cognitive performance between IBD patients and control subjects after adjustment for covariates. Importantly, they did not measure sophisticated operative thinking and complex psychomotor coordination, which we assessed in our study 18 .
Generally, our study has shown impaired overall cognitive and psychomotor function in patients with IBD compared to healthy subjects, irrespective of age, duration of education and BMI. Specifically, our IBD patients exhibited longer T TTS in all measured cognitive dimensions compared to healthy controls. In terms of mental chronometric principles, these results demonstrate the basic inferiority of the IBD group in neurocognitive and   Different complexity of the CRD tests could provide an insight into the level of cognitive impairment. As reported, simple forms of cognitive tests (CRD-311 and CRD-11) revealed a stable performance (TB) and equal accuracy (N ER ) in both study groups. The main deteriorations in the aforementioned tests among IBD patients were found in the mental processing speed, mental endurance and data analysis (T TTS ; T MIN ) domains. However, performance in the most complex test CRD-411, which requires operative thinking and harmonized eye-hand-leg coordination revealed significant additional impairment in answer accuracy, efficiency, attention, alertness (N ER ), mental stability and cognitive fluctuation (TB) domains among IBD patients. Specifically, IBD patients exhibited equal error rates with prolonged response times in comparison to control subjects when analysing the results of simple CRD tests. However, the most complex CRD test revealed increased error rates, cognitive  Table 7. Multiple linear regression analysis of selected variables as independent predictors of CRD parameters in the UC patients T TTS -total test solving time; T MIN -minimal time of particular test solving; N ER -total number of wrong reactions; TB -total ballast time; BMI -body mass index; Hb -hemoglobin; hs-CRP -high sensitivity C-reactive protein; FC -faecal calprotectin; Mayo/DAI -Mayo score/disease activity index for ulcerative colitis; ns -variables were not included in the model due to statistical insignificance (P > 0.1). a Bvalue indicates regression coefficient of the independent variable. b t-value indicates t-statistic value.
instability and prolonged response times as well among the IBD group. Preceding findings suggest compensation in simple cognitive tests on the account of prolonged mental processing, problem solving and decision making (T TTS ). Therefore, in terms of accuracy and error rates (N ER ), it can be stated that IBD patients may exhibit latent cognitive impairment in simple cognitive functions and manifested cognitive impairment regarding complex cognitive functions. Aforementioned outcomes could suggest that complex cognitive functions are more affected by the disease process. Similar findings were presented in a related study which found prolonged global cognitive response times, but unaltered error rates, indicating subtle cognitive impairment in patients with CD 17 . Moreover, there was a significant difference between the IBD and control group regarding time responses in different test phases. The latter could indicate fragile cognitive stability, earlier cognitive fatigue and susceptibility to mental exhaustion among IBD patients. However, further studies are needed.
Finally, our study suggests that IBD per se deteriorates cognitive and psychomotor dysfunction, independent of ageing, education level and BMI status. The mechanism for cognitive and psychomotor dysfunction among IBD patients in our study can eventually be understood as a consequence of an intrinsic disease process. However, further studies are necessary for the establishment of a true causal relationship.
The pathophysiological basis for cognitive impairment in IBD patients has been investigated by Petruo et al. 35 . They have demonstrated selective functional alterations at the level of response selection in the fronto-striatal pathway. The aforementioned brain regions are especially important for cognitive flexibility, leading to a task-switching deficit in IBD patients. Interestingly, the neurophysiological analysis did not reveal functional deficits in brain regions important for working memory, the attentional selection process and perceptual categorization 35 . Conversely, our findings of aggravated CRD performance in the used tests indirectly point to a dysfunction in those cognitive components with the exception of memory.
Except for the functional neuropsychological deteriorations, scarcely available studies undoubtedly report morphologic pathologic changes in IBD patients as well. Hollerbach et al. assessed neurophysiological function and morphologic changes in IBD patients using evoked potentials and magnetic resonance imaging. The study had shown prolonged p300 wave latency and focal white matter lesions equivalent to short-term memory dysfunction and sensory perception disturbances. The aforementioned functional changes were mostly present in cortical postcentral regions corresponding to the somatosensory cortex, which is important for sensory data processing leading to perception deficit, which can correspond to the findings from our study 36 . These pathophysiological and morphologic changes are certainly interconnected, possibly by impaired neurogenesis and reduced neuroplasticity in IBD patients.
Whether disease activity plays a role in cognitive dysfunction is not definitely established. Data from our study did not reveal any significant correlation between CRD parameters and disease activity indices in any tested domain. Furthermore, we did not establish any significant correlation between FC and cognitive function. Consistent findings were reported by several studies 14,17,18  Albeit it has been shown that systemic inflammatory parameters do not perfectly provide insight into the degree of intestinal inflammation, epidemiological studies indicate that systemic inflammation can deteriorate cognitive function 37,38 . However, no correlation was generally found between CRD variables and hs-CRP levels in our study. Similarly, a related study did not establish a significant correlation between CRP and global cognition 16 . On the contrary, Langenberg et al. found an independent positive correlation of hs-CRP with both mental processing and decision making time 17 . Nevertheless, levels of hs-CRP levels did not affect cognitive performance in our study.
Moreover, Hb concentrations were generally not correlated with CRD parameters in our study. Similarly, Wells et al. did not find any significant correlation of Hb concentrations and cognitive function. Moreover, they did not observe improvements in cognitive function with the treatment of anemia in IBD patients 14 . However, Golan et al. found a significant correlation between Hb levels and better cognitive performance 16 . On the contrary, a related study on a CD sample has shown an independent correlation of plasma Hb levels with slower mental processing speed 17 . Nonetheless, there is no evidence of the influence of Hb levels on cognitive function in our study. www.nature.com/scientificreports www.nature.com/scientificreports/ Moreover, it was previously shown that ageing deteriorates cognitive and psychomotor function 39 . Consistently, we have established a significant correlation of age with worse cognitive performance in most CRD parameters. However, a related study did not establish any significant correlation between age and cognition 17 13,15 . Importantly, similar to previous studies, our sample did not include geriatric subjects so the aforementioned findings cannot be applied to elder IBD patients 13,15,17 . However, there were no differences in age and gender between the observed groups in our study, so we can state that groups were adequately age-and sex-matched. Finally, it should be noted that age was included as a covariate in the ANCOVA model in our study.
Whether there is a discrepancy in cognitive performance between CD and UC patients is not well established, but the available evidence does not support the conjecture of disparity. Likewise, we have not established a significant difference in cognitive performance among patients with CD and UC, which is consistent with related studies 13,18 .
All of the above findings emphasize the cognitive burden of IBD on daily executive functions such as working abilities, especially in the fields, which require stable performance, strong concentration and repetitive operations. It is necessary to underline that, depending on the complexity, the used CRD tests correspond and simulate usual daily functions -shopping abilities, job tasks, vehicle driving, etc. 32 . Since even subtle impairment in reaction time increases the risk of traffic road and workplace accidents, these findings in the IBD population are of definite clinical significance. However, these assumptions could not be confirmed since appropriate studies are lacking. Nonetheless, self-awareness in these patients could eventually encourage job prequalifications and avoidance of prolonged drive, eventually leading to accident prevention. Additionally, prevention and early treatment could theoretically intercept cognitive deterioration in these patients. Finally, cognitive dysfunction could lead to problems with therapeutic compliance and dietary discipline, and therefore to circulus vitiosus of the IBD and cognitive impairment.
Similar to other studies, our study has several limitations. Firstly, our research design was cross-sectional, so it was not possible to establish any causal relationship or follow longitudinal cognitive alterations. Further, our sample mostly involved patients with IBD of low to moderate activity so we are not able to extrapolate our findings on the entire IBD population, yet we can only speculate. Further, the CRD-series are not specifically designed for the IBD population, but are rather generic instruments for cognitive and psychomotor function assessment. Moreover, our study did not assess verbal function and memory, so we were not able to fully compare these cognitive domains with the few available studies. Additionally, the impact of mood disorders such as depression canot be fully excluded since we haven't used any depression quantification score. Moreover, we cannot fully exclude the influence of anxiety and fatigue on the cognitive performance in our study. This should be additionally evaluated in further studies on an IBD population with CRD utilization. Still, all subjects denied the presence of mood disorder in a self-reported questionnaire and none of them had a previous clinical diagnosis or clinically evident signs of depression, anxiety and chronic fatigue. Finally, normative data for CRD tests are not established, so it is impossible to validate control group results.
Since this is the first study using the CRD instrument on an IBD population, it is difficult to compare our results with other studies. However, the categorization of study results in different cognitive domains alleviates the interpretation and allows for more direct comparison. There is a substantial need for an objective, non-invasive and standardized neuropsychological test battery in the IBD population 18 . With respect to the aforementioned methodological advantages of the CRD tests, it may be desirable to encourage further comprehensive CRD studies on IBD patients [30][31][32][33][34] .
In conclusion, our study has shown impaired neurocognitive and psychomotor function in patients with IBD compared to control healthy subjects, especially in mental processing speed and mental endurance of perceptive abilities, convergent thinking and sophisticated operative thinking, which was independent of age, duration of education and BMI. Further research is necessary to clarify the significance of these findings.

Data Availability
We disclose any restrictions on the availability of data, materials and associated protocols.