Neurological soft signs and olfactory dysfunction in patients with borderline personality disorder

,


Introduction
According to the 10th revision of the International Classification of Diseases and Related Health Problems (ICD-10), borderline personality disorder (BPD) is characterized by instability of self-image, interpersonal relationships, and affect, as well as marked impulsivity (World Health Organization, 2019).BPD is a serious disorder with a lifetime prevalence of 2.7-5.9%(Grant et al., 2008;Tomko et al., 2014), it often presents with other psychiatric and physical comorbidities (Skodol et al., 2002;Jacobi et al., 2021).BPD places a high burden on the healthcare system due to its high direct and indirect costs, among other factors (Bode et al., 2017;Sveen et al., 2023;Wagner et al., 2022;Bourke et al., 2021;Hastrup et al., 2019;van Asselt et al., 2007).It also has a significant impact on the social and occupational lives of those affected (Fertuck et al., 2018;Bohus et al., 2021), with BPD patients experiencing a reduced quality of life (Soeteman et al., 2008;Cramer et al., 2006;Guillén et al., 2021) and suicide rates ranging from 3 to 10% (Paris, 2019).However, evidence-based therapies such as dialectical behavior therapy have been shown to reduce symptom burden, improve social skills, and improve the long-term prognosis of BPD patients (Zanarini et al., 2018;Guillén et al., 2021;Bartak et al., 2011;Flynn et al., 2017).Therefore, it is highly relevant to better understand and recognize the disease early on.
BPD patients are thought to have altered neuroplasticity, the brain's ability to reorganize itself by forming new neural connections throughout life (Cattane et al., 2017).This process involves the brain's ability to adapt and change in response to various experiences, influences, environmental factors, learning, and injury on different levels (e.g., synaptic, structural, or functional plasticity).Neuroplasticity can stabilize and compensate for disease-related changes in the brain (Smith, 2013).Therefore, abnormalities in clinical neurological examinations could be useful as indicators of psychopathological vulnerability in this patient group (Levit-Binnun and Golland, 2012).For this reason, neurological soft signs (NSS), which are thought to correlate with network alterations in the brain, could be considered as clinical indicators to be examined (Kong et al., 2020;Wang et al., 2019).
NSS are defined as subtle abnormalities in neurological functioning (e.g., deviations in motor coordination, balance difficulties, gait abnormalities, and sensory perception), that can be determined by clinical assessments and may refer to a non-specific brain dysfunction (Dazzan and Murray, 2002;Alamiri et al., 2018).While NSS may not lead to significant functional impairment on their own, they can provide valuable insights into underlying neurological processes and could provide important information for further diagnostic evaluation and treatment planning.NSS are often associated with various psychiatric diseases, such as schizophrenia, autism spectrum disorder, attention deficit hyperactivity disorder (Kong et al., 2020;Tani et al., 2006;Gong et al., 2015).Significant alterations in the NSS (sub-)scores in participants with BPD have already been found in a few studies (Zhang et al., 2015;De la Fuente et al., 2006;Arbabi et al., 2016;Gardner et al., 1987).In the subscales of NSS testing, significant differences were noticeable for example in sensory integration (Zhang et al., 2015) and motor coordination in the group comparison between BPD and healthy controls (Arbabi et al., 2016).However, there are contradictory results, for instance there are also studies with a lack of significant differences in motor coordination (Zhang et al., 2015) or in the NSS total score (Stein et al., 1993).
Associations between psychiatric disorders and NSS were also found in common psychiatric comorbidities of BPD, such as post-traumatic stress disorder (PTSD) (Belrose et al., 2020) and major depressive disorder (Schülke et al., 2023).Especially, in female patients with depression and childhood maltreatment, which plays a remarkable role in the pathogenesis of BPD (Carlson et al., 2009), significant differences have been found in frontal area related NSS-subscales (Zhao et al., 2015).
These clinical neurological abnormalities can be partially reconciled with the current state of neuroimaging, in which review articles have repeatedly identified multiple structural changes in the brain of BPD patients (Krause-Utz et al., 2014;Yang et al., 2016).The literature has highlighted a strong focus in the orbitofrontal cortex, which may play a major role in the symptomatology of BPD (Berlin et al., 2005;Hanson et al., 2010;Chanen et al., 2008).In patients with childhood trauma, subsequent studies have discovered an alteration in the depth of the olfactory bulb, which is located near the orbitofrontal cortex, and have discussed this as another indicator of vulnerability to BPD (Takahashi et al., 2019;Croy et al., 2013).Since recent findings suggest that neurogenesis occurs precisely in the olfactory bulb (Fares et al., 2019;Lazarini and Lledo, 2011;Lötsch et al., 2014), this could be used to draw conclusions about altered neuroplasticity in affected individuals.
Reduced functioning of the olfactory bulb has already been suggested in patients with depression (Pause et al., 2001;Kohli et al., 2016), childhood maltreatment (Croy et al., 2013), schizophrenia (Zou et al., 2018;Kamath et al., 2018) and schizotypy (Mathur et al., 2019).In contrast, no significant difference has yet been found in personality disorders (Ruocco et al., 2009).Ruocco and colleagues investigated a total of 56 patients with a Cluster B personality disorder (71% of whom had BPD), 19 patients with a Cluster C personality disorder, and 61 healthy controls regarding their ability of smell identification (Ruocco et al., 2009).However, the researchers did not further differentiate between the specific types of personality disorders within the Cluster B group.Additionally, they used the "University of Pennsylvania smell identification test" (UPSIT), a screening instrument primarily focused on olfactory identification.Given the constraints of the solitary study exploring olfactory function in BPD patients, our objective is to examine patients diagnosed with BPD.We aim to enhance the validity of the findings by using a screening test that evaluates not only olfactory identification but also olfactory threshold and discrimination.
Olfactory function and NSS have not been concurrently studied despite their relative straightforward examination methods.Both assessments offer insights into sensory perception and neurological functioning and might even elucidate aspects of brain plasticity without requiring sophisticated instrumentation.Enhancing understanding of their interaction could yield insights into the underlying mechanisms of sensory processing and suggest potential ways for intervention or rehabilitation.The aim of this study was to investigate possible associations of BPD (− severity) and alterations in NSS and olfactory function.We hypothesized that neuroplasticity is altered in BPD patients and consequently patients have lower neurological and olfactory function levels.Additionally, we aim to shed light on the correlation between NSS and olfactory function.Even though there is a body of literature regarding neuroplasticity changes detected by neurological imaging, clinical screening would require low-threshold testing options, such as NSS and Sniffin' sticks used in this study.

Participants
The current study is a monocentric observational study.For our study, two groups were recruited.The first group consisted of 39 female subjects diagnosed with BPD according to the criteria of the 10th revision of the International Classification of Diseases and Related Health Problems (ICD-10) in the inpatient setting of Hannover Medical School (MHH).At the time of examination, 34 of the subjects were undergoing inpatient treatment and five were undergoing outpatient treatment.Nineteen had a comorbid diagnosis of post-traumatic stress disorder (PTSD).Patients with current drug use (excluding nicotine), an acute medical disorder, schizophrenia spectrum disorder, acute upper respiratory illness, post-traumatic brain injury with loss of consciousness lasting more than one hour, or a neurological disease were excluded.The second group comprised 19 healthy female control subjects (HC).The assessment of each study participant lasted approximately two hours in total and was conducted at the Department of Psychiatry, Social Psychiatry, and Psychotherapy of MHH.
All participants gave written informed consent before study inclusion.The subjects continued their previous medication.Initially, we tested 40 BPD and 20 HC subjects, but we had to exclude one patient from each group during the course, because one HC subject was diagnosed with panic disorder and one BPD subject with multiple sclerosis shortly after study inclusion.The study received approval from the ethics committee of MHH (Vote No: 8201_BO_S_2018) and was conducted in accordance with the Declaration of Helsinki (1964) and its later amendments.

Neurological Soft Signs
For the examination of NSS a comprehensive scale developed by Gurvits et al. (Gurvits et al., 1993, Gurvits et al., 2000, Gurvits et al., 2006) was used as previously described (Schülke et al., 2023).This test battery consisted of 50 items, each scored on a scale of 0-3 points, resulting in a total score ranging from 0 to 150 points.A higher score indicated a higher error rate.The items were categorized into superordinate categories, including gait and station (8 items), motor coordination (13 items), frontal lobe (4 items), parietal lobe (7 items), apraxia (1 items), temporal lobe (2 items), and constructional apraxia (15 items).The NSS examination took an average of 60 min and was performed in a standardized manner by trained raters with the same order of items and instructions.The raters were not blinded to the respective research group.

Olfactory dysfunction
The assessment of olfactory dysfunction took an average of 40 min and was conducted using Sniffin' Sticks from Burghart Messtechnik GmbH in Holm, Germany (Item No.: LA-13-00134).The test followed the manufacturer's protocol and included a threshold test, a discrimination test, and an identification test.The sum of the scores obtained from each test yielded a total score between 0 and 48, with a maximum of 16 points available for each test.The threshold test measured the olfactory threshold of the participants using a forced staircase method, while the discrimination test required participants to distinguish between different odors, and the identification test used a multiple forcedchoice method to assess their ability to recognize specific odors.Normosmia was defined as a score > 30, while a score of ≤30 indicated hyposmia, and a score of ≤15 was indicative of anosmia.While normosmia refers to the normal ability to smell, hyposmia describes a reduced sense of smell, and anosmia denotes a complete loss of the ability to smell.Notably, the raters were not blinded to the participants' research group during the examination.

Severity score -BPD
BPD severity was measured using the borderline symptom list 23 (BSL-23) version of the Department of Psychosomatic Medicine and Psychotherapy of the Central Institute of Mental Health in Mannheim, Germany (Bohus et al., 2009).The BSL-23 is a self-rating questionnaire designed to assess symptom severity, with higher score indicating more severe symptoms.It comprises three parts: In the first part, patients rate their well-being over the past week using a scale of 0 to 4 (0 = not at all; 4 = very strong) in response to 23 questions.The second part involves completing a visual analog scale from 0 to100 (0 = very bad; 100 = excellent) to evaluate the overall quality of their personal state of the past week.The third part consists of questions about self-injurious behavior during the past week on a scale of 0 to 4 (0 = not at all; 4 = several times a day).For the analysis of the well-being and self-injurious behavior, we computed both mean values and sums for each subject, referred to as BSL-23 mean, BSL-23 sum, self-injury mean and self-injury sum.The overall personal state is indicated as BSL-state.

Statistical analysis
To compare categorical values between groups, we used the Chi-Square test.The Shapiro-Wilk test was utilized to test for normality in NSS and Sniffin' Sticks data, but the results indicated a non-normal distribution.Therefore, further analyses were conducted using nonparametric tests, specifically the Mann-Whitney U test.Correlational analysis was performed using Spearman's rank correlation coefficient (rho).Baseline demographic information is presented as mean ± standard deviation.A p-value of <0.05 was considered significant.The statistical analyses were conducted using SPSS version 28.The figures were created with R version 4.3.1.Because this was an exploratory study, we did not correct for multiple testing.

Results
All participants included in the study were female.The healthy controls (HC) were matched by age.However, the BPD group had a significantly higher body mass index (BMI) than the HC group.BPD patients were more likely to be smokers and indicated more pack-years than HC (Table 1).

Higher NSS in group comparison
Overall, the BPD group had higher total scores on the NSS assessment than the HC group.When NSS scores were analyzed in subgroups, BPD patients had significantly higher NSS scores in 5 out of 8 subcategories (all p < 0.05, Table 2 and Fig. 1).

No alterations in the total score of the olfactory function
We did not find a significant difference in the total score of olfactory function between HC and BPD when examined with Sniffin' Sticks (Fig. 2D).While the threshold test and discrimination test showed no significant differences, BPD subjects scored higher on the smell identification subscale (Table 2A-C).The total score of NSS was found to have a negative correlation with olfactory function within the BPD group (r = − 0.401, p = 0.011) (Fig. 3 and Table S2 in the supplementary material).

More NSS when taking antipsychotics
Next, we investigated whether the presence of NSS might be influenced by antipsychotic medication.BPD patients who were taking antipsychotic medication (n = 17) had significantly higher total NSS scores than those who were not taking antipsychotic medication (n = 22, U = 288.5,Z = 2.878, p = 0.003) (Table S3 in the supplementary material).However, we did not find any significant differences between the two groups regarding olfactory function (p > 0.05).

No significant influence of PTSD concerning the NSS or olfactory function
We also found no significant difference in the group comparison of

BSL-23 appears to correlate with NSS
Furthermore, we performed a correlational analysis.A significant correlation was found between the BSL-23 sum score and the NSS motor coordination score (r = 0.566, p ≤0.001) as well as the NSS temporal lobe score (r = 0.356, p = 0.039), and between the BSL-23 for self-injury and the NSS frontal lobe score (r = 0.359, p = 0.04) as well as the NSS construction apraxia score (r = 0.37, p = 0.034) (Table S4 in the supplementary material).

Discussion
The main finding of our study examining NSS and olfactory function in patients with BPD was the significantly higher rate of NSS in patients with BPD compared to HC, while there were no differences in the total olfactory score.Elevated NSS levels have been reported in patients with mental disorders such as PTSD (Belrose et al., 2020), major depressive disorder (Schülke et al., 2023), schizophrenia (Petrescu et al., 2023), but also in BPD (Zhang et al., 2015;De la Fuente et al., 2006;Arbabi et al., 2016;Gardner et al., 1987).These findings are consistent with our cohort of BPD individuals.Nevertheless, the scales utilized in the referenced studies shared a similar focus on NSS but varied in the quantity and selection of items.For example, Arbabi et al. used 29 items modified from Quitkin et al. to assess motor coordination, complex motor sequencing, and sensory dysfunction (Arbabi et al., 2016;Quitkin et al., 1976).Gardner et al. similarly employed 12 items based on Quitkin et al. (Gardner et al., 1987).In contrast, Zhang et al. utilized the Cambridge Neurological Inventory, which examined motor coordination, sensory integration, and disinhibition (Zhang et al., 2015).Standardizing the scale levels would be beneficial for future studies.
The abnormalities in the NSS subcategory of motor coordination and gait and station found in our cohort of BPD individuals are coherent with some studies in the current literature (Arbabi et al., 2016, Gardner et al., 1987).This knowledge could prove valuable in streamlining the NSS examination process as part of a clinical application by focusing on few items.Moreover, the alterations in the temporal lobe are also welldocumented in the existing research, particularly with reference to neuroimaging studies using a cohort of BPD individuals (Krause- Utz et al., 2014;Nenadić et al., 2020;Rossi et al., 2015).
Regarding the individual items, there is an overlap with Gurvits et al. in whose cohort of PTSD individuals, interestingly, the item copying of figures shows the greatest difference in the NSS, as is also the case in our cohort of BPD individuals (Gurvits et al., 2006).This similarity between PTSD and BPD should be kept in mind, as trauma also plays a common role in BPD (Porter et al., 2020;Gunderson et al., 2018).
Since childhood trauma is discussed in the literature as a potential cause of altered neuroplasticity (Cattane et al., 2017), and imaging studies have shown a flatter right olfactory sulcus in BPD patients with trauma compared to BPD patients without trauma (Takahashi et al., 2019), it was considered necessary to make a group comparison between patients with and without PTSD.Although no significant group differences were found in NSS and olfactory function, the question remains to what extent the differences are related to BPD, PTSD, or childhood trauma (Krause-Utz et al., 2014).For this reason, it would be useful to integrate a trauma questionnaire into upcoming studies.
It also remains unclear whether the altered neuroplasticity in BPD   patients is an expression of their disease or whether (structural) brain changes in adolescence lead to an altered neurodevelopment as a vulnerability factor for the development of BPD (Crowell et al., 2009;Cattane et al., 2017).Longitudinal studies would be appropriate to explore this, as NSS correlate with altered brain structure already in adolescence (Bonke et al., 2023).We hypothesized that patients with BPD show altered neuroplasticity.Consistent with this, we demonstrated an association between lower olfactory function and a higher number of neurological soft signs, indicating soft neurological deficits.Studies have demonstrated that the olfactory bulb (OB) experiences volume loss in individuals with a history of childhood maltreatment (Croy et al., 2013), a condition often observed in BPD patients (Ibrahim et al., 2018).A reduced size of the olfactory bulb was also found in people suffering from major depression (Rottstaedt et al., 2018), which is a common and strongly interacting comorbidity of BPD (Köhne and Isvoranu, 2021).Consequently, we hypothesized that the OB might also exhibit volume reduction in BPD patients, given that other brain regions show volume reduction in this population: Neuroimaging studies have already indicated volume reductions in the amygdala and hippocampus in BPD patients, both of which are located in the temporal lobe and are critical for affect perception (Cattane et al., 2017).This is particularly relevant since affective instability is a hallmark of BPD.Our cohort's results support this hypothesis by proving a correlation between temporal lobe-associated NSS and BPD severity, as well as a deficiency in temporal lobeassociated NSS in BPD patients.These findings suggest structural changes in brain areas typical for BPD, indicating altered neuroplasticity in this patient group.Considering that impulsivity is another characteristic of BPD, it could be assumed that alterations might also occur in the frontal lobe and the adjacent OB, as suggested by several studies (Cattane et al., 2017;Chanen et al., 2008;Berlin et al., 2005).However, our study did not find significant deficits in frontal lobe associated NSS.This aligns with our findings regarding olfactory function, which even indicate some areas of superiority of BPD patients.These disparate findings warrant further research.Given that the BSL-23 scores for selfinjury correlate with the frontal lobe associated NSS, it raises the question of whether this could represent a new phenotype, suggesting subsequent studies should focus on this aspect.
In contrast to the significant results in the NSS, no significant differences in the total olfactory score could be found between BPD patients and HC.Although this differs from the results of other disorders such as depression (Pause et al., 2001;Kohli et al., 2016), childhood maltreatment (Croy et al., 2013) and schizophrenia (Mathur et al., 2019;Zou et al., 2018), it is in line with data from Ruocco et al., who already reported no significant differences in odor recognition in patients with personality disorders (Ruocco et al., 2009).However, the results of our cohort of BPD individuals show that BPD patients tend to smell better, which is underlined by the significant difference in odor identification.Better olfactory function can also be found in the literature concerning patients with panic disorder (Burón et al., 2015).Further investigation is needed to explore the reasons for these similarities, particularly considering that the imaging data reveals a reduction in the orbitofrontal cortex, in the proximity of which the area of odor perception is located, in patients with BPD as well as panic disorder (Krause-Utz et al., 2014;Yang et al., 2016;Roppongi et al., 2010).
When determining the causality of the relationship between BPD and olfactory function, confounding factors must always be considered.For this reason, an exploration of BMI would also be important.The literature tends to support a positive correlation between obesity and olfactory function in the general population, but there is no general agreement on this point (Sollai and Crnjar, 2023;Jacobson et al., 2019;Arikawa et al., 2020;Aschenbrenner et al., 2008;Herz et al., 2020;Velluzzi et al., 2022).In our cohort of BPD individuals, the correlation between BMI and olfactory function was not significant.Nevertheless, the BPD subjects had a significantly higher average BMI than the control subjects.This should be considered as a possible confounding factor, which could be responsible for the better smell identification score in our BPD cohort.
Furthermore, the significantly increased NSS in patients treated with antipsychotics contradict the findings of Schülke et al., who found no influence of antipsychotics on the NSS score in patients with major depressive disorder (Schülke et al., 2023).In patients with schizophrenia, there is even discussion of antipsychotics having a beneficial effect on NSS through their protective influence on brain structure (Bachmann and Schröder, 2018).Further studies are needed to investigate the cause of this difference and to explore whether BPD patients are more susceptible to neurological side effects of antipsychotics.Possibly, BPD patients receiving antipsychotics were more severely affected or could represent a psychopathological subgroup of BPD (e.g., with more hallucinations or intermittent psychotic symptoms) but these features were not assessed in our current study.In contrast to the NSS, the lack of effect of antipsychotics on olfactory function suggested by our results is partially consistent with the current literature on schizophrenia patients (Brewer et al., 2001).
The scores of the BSL-23 correlated significantly with the NSS, while the correlation with the olfactory function was not significant.Correlations with NSS and severity have been found in other mental disorders such as depression (Zhao et al., 2015) and PTSD (Belrose et al., 2020).The same applies to the olfactory function, as correlations with mental disorders such as panic disorder or depression have been reported, which is contrary to our results (Marin et al., 2023).Since our study showed that different components of the BSL-23 have different results regarding their significance, an exploration would be necessary to what extent this list would have to be modified for further studies in order to increase the reliability and validity.
The present study addresses the implication of the need to expand the clinical diagnostics of BPD patients with low-threshold neurological testing.Further standardized investigations may enable not only a better understanding of the pathophysiology of the disorder, but also a better classification of patients within the diagnosis and thus probably more targeted therapy.
There are limitations underlying our study.First, this study was cross-sectional, thus there are no longitudinal data points.Additionally, we did not perform MRI to correlate findings with possible abnormalities.Moreover, the sample size is small, and we have no patients with only a BPD diagnosis.Therefore, it should be considered that results may be influenced by psychiatric comorbidities such as depression and PTSD.The same applies to medication use as well as comorbid somatic conditions.Furthermore, we did not use a clinical interview instead of the BSL-23 self-report.Finally, we did not match the subjects by educational level, the raters were not blinded to the respective research group, and we did not correct for multiple testing.
In conclusion, our study indicates altered NSS in BPD patients.In contrast, BPD subjects tend to have a higher score in olfactory function regarding smell identification.Further studies of this type including trauma questionnaires and MRI are necessary to establish NSS and smell tests as clinical screening tools in patients with BPD and to identify the effects of the disorder on neuroplasticity.

Fig. 1 .
Fig. 1.Group comparison between borderline patients (patients) and healthy controls (controls) with Mann-Whitney U test concerning the neurological soft sign (NSS) total score.

Fig. 2 .
Fig. 2. Group comparison between borderline personality disorder patients (BPD, blue) and healthy controls (HC, red) across the different subtests and total olfactory score using the Mann-Whitney U test.(For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)