Anxiety Independently Contributes to Severity of Freezing of Gait in People With Parkinson’s Disease

Parkinson’s disease is a neurodegenerative disorder characterized by a wide range of motor and nonmotor symptoms. Such aspects are related not only to the loss of dopaminergic neurons in the substantia nigra pars compacta but also to dysfunction of serotonergic and noradrenergic pathways.^1,2

Nonmotor symptoms such as anxiety and depression are highly prevalent in people with Parkinson’s disease, with prevalence ranging from 26% to 65%, and these symptoms may also coexist.¹ Both symptoms have a marked impact on activities of daily living, balance, gait, and quality of life,^1,2 and they have been shown to be associated with falls.² Furthermore, anxiety has been described as a predictor of activity avoidance due to fear of falling, a factor that may also contribute to gait impairment in people with Parkinson’s disease.³ Gait disorders are among the most disabling features of the disease; 20% to 60% of individuals with Parkinson’s disease present with freezing of gait.⁴ This phenomenon is a brief and episodic inability to generate effective steps^5,6 that typically occurs while starting to walk, turning, reaching a destination, or passing through narrow passages.⁵ It is considered a strong predictor of falls⁷ and has a major impact on mobility⁴ and quality of life.⁸

Besides the association between freezing of gait and Parkinson’s disease severity,^8,9 it has been suggested that nonmotor symptoms, such as anxiety¹⁰ and cognitive impairment,^5,6 not only are related but also may be involved in the pathophysiology of the phenomenon.^5,6 Among mechanisms previously described to elucidate freezing of gait,⁶ the interference model highlights the role of the limbic system. Based on this model, freezing of gait may arise due to a breakdown of concurrent inputs from cognitive and limbic loops during motor tasks, leading to neuronal crosstalk between the basal ganglia circuits and overloading the striatum.⁶ If anxiety represents a limbic overload and triggers freezing of gait,^10,11 it might be argued that anxiety plays a role not only in the frequency of freezing of gait¹¹ but also in its duration.

Therefore, the aims of this study were to identify the association between freezing of gait and anxiety, depression, cognitive impairment, balance confidence, and quality of life; identify the association between anxiety and balance confidence; and determine factors that contribute to the severity of freezing of gait in people with Parkinson’s disease.

Methods

Participants

A cross-sectional study with consecutive participants with idiopathic Parkinson’s disease¹² who were able to walk without assistance of another person, with or without an assistive device, was performed at the Movement Disorders Clinic at the State of Bahia Health Attention Center for the Elderly in Brazil. Exclusion criteria included atypical Parkinsonism, dementia, severe visual or auditory disturbance, and other comorbidities that could affect balance and gait. The study was approved by a local research ethics committee, and all subjects provided written informed consent. Participants were evaluated in the “on” phase of the antiparkinsonian medications.

Assessments

Demographic and clinical data, including sex, age, Parkinson’s disease duration, motor fluctuations, dyskinesias, comorbidities, and antiparkinsonian medications taken were recorded based on self-report complemented by chart review. The levodopa equivalent dose was calculated according to a previously reported conversion factor.¹³ Subjects were classified as recurrent fallers if they had a history of two or more falls in the past 12 months.

Parkinson’s disease severity was assessed by the motor section of the Unified Parkinson’s Disease Rating Scale (UPDRS), with scores ranging from 0 to 108 points, and the modified Hoehn and Yahr scale, which ranges from 0 to 5.¹⁴ Disability was assessed with the UPDRS activities of daily living (ADL) section, with scores ranging from 0 to 56 points, and the Schwab and England scale, ranging from 0% to 100%.¹⁴ The Montreal Cognitive Assessment (MoCA) was used to evaluate cognition, with scores ranging from 0 to 30.¹⁵

The Hospital Anxiety and Depression scale (HADS) was used for screening disorders of mood. It is composed of depression (HADS-D) and anxiety (HADS-A) subscales, each ranging from 0 to 21 points.¹⁶ A cutoff of ≥8 points in each subscale has been used to classify people with Parkinson’s disease as having anxiety or depression symptoms.¹⁷ Self-perceived balance confidence for performing daily activities was measured with the Activities-Specific Balance Confidence scale, with scores ranging from 0% to 100%;¹⁸ health-related quality of life was measured with the 8-item Parkinson’s Disease Questionnaire (PDQ-8).¹⁹

The Freezing of Gait Questionnaire (FOG-Q) was used to evaluate the presence and severity of freezing of gait; higher total scores (range, 0–24 points) indicate more severe freezing of gait.²⁰ The term “feet glued to the ground” was used to describe freezing of gait. The examiner was instructed to demonstrate or show videos of the freezing phenomenon in different situations, such as start hesitation or freezing while walking and turning. Participants were classified as freezers if they scored ≥1 on item 3. Since items 1 and 2 refer to general gait difficulties, the sum of items 3–6 was used to determine freezing of gait severity.²⁰ All scales were administered by the researchers, including the FOG-Q.

Statistical Analysis

Data were analyzed using IBM SPSS version 21 (IBM, Armonk, N.Y.). The Kolmogorov-Smirnov test was performed to check normality of data. Comparisons between freezers and nonfreezers were performed using Student’s t test, Mann-Whitney test, chi squared, or Fisher’s exact test, when appropriate. Spearman’s correlation coefficients (two-tailed) were used to assess the correlation between FOG-Q, freezing of gait severity, HADS, HADS-D, HADS-A, and Activities-Specific Balance Confidence scale. The strength of correlations was determined as follows: 0.00–0.30, very weak; 0.30–0.50, weak; 0.50–0.70, moderate; 0.70–0.90, strong; and 0.90–1.0, very strong.²¹

To identify associations between potential contributors and freezing of gait severity, univariate linear regression analyses were performed. Both HADS-A and HADS-D ≥8 points were set as the main potential contributors of mood symptoms.¹⁷ Variables with p<0.05 in univariate analysis were chosen to be entered into the stepwise multiple linear regression model adjusted for age and sex. A significance level of 0.05 was set for all statistical tests.

Results

Participant Characteristics and Comparisons Between Freezers and Nonfreezers

A total of 78 participants with Parkinson’s disease were enrolled in this study. The Kolmogorov-Smirnov normality test indicated that the scores on the measures were not normally distributed, except the MoCA, HADS, HADS-D, Activities-Specific Balance Confidence scale, and UPDRS-ADL and motor sections. The median age was 69 years (range, 60–89), and 42 (54%) were male. The median disease duration was 6 years (range, 1–20), the median Hoehn and Yahr scale stage was 2.5 (range, 1.5–4), and the median levodopa equivalent dose was 650mg/day (range, 150–1,962.5).

Twenty-seven individuals (35%) were classified as freezers. Comparisons between freezers and nonfreezers are shown in Table 1. In brief, freezers had longer disease duration and greater disease severity (UPDRS motor and Hoehn and Yahr scale) and disability (UPDRS-ADL and Schwab and England scale) than nonfreezers. Freezers also showed higher levels of anxiety (HADS-A) and depression (HADS-D), as well as lower balance confidence (Activities-Specific Balance Confidence scale) and reduced quality of life (PDQ-8) (p<0.05). Table 2 shows the frequency and duration of freezing of gait episodes.

HADS, HADS-A, and HADS-D showed a small to moderate correlation with FOG-Q, freezing of gait severity, and Activities-Specific Balance Confidence scale. A moderate correlation was identified between scores on the Activities-Specific Balance Confidence scale and FOG-Q, as well as a small correlation with freezing of gait severity (p<0.05) (Table 3).

Univariate Analysis

Univariate associations between potential contributors and freezing of gait severity are shown in Table 4. Freezing of gait severity was associated with history of ≥2 falls in the past year, disease severity (Parkinson’s disease duration, UPDRS motor, and Hoehn and Yahr scale), Parkinson’s disease symptoms (motor fluctuations and HADS-A ≥8 points), levodopa equivalent dose, and balance confidence (Activities-Specific Balance Confidence scale) (p<0.05). These potential contributors, age, and sex were entered into the multivariate linear model.

Multivariate Analysis

The final multivariate linear model is shown in Table 5. The Hoehn and Yahr scale accounted for 31% of the variance in FOG-Q severity scores. The presence of HADS-A ≥8 points increased the explained variance to 38%, and the full model (reached by adding the levodopa equivalent dose) explained 42% of the variance in freezing of gait severity.

Discussion

This study focused on the association between the presence and severity of freezing of gait and symptoms of anxiety and depression in people with Parkinson’s disease. Nearly one third of participants were classified as having freezing of gait. Disease severity, higher antiparkinsonian medication intake, anxiety, depression, a history of recurrent falls in the past year, disability, and reduced balance confidence were associated with freezing of gait. However, multivariate analysis showed that only disease severity, anxiety, and higher antiparkinsonian medication intake were independent contributors to freezing of gait severity.

In our study, about one third (35%) of participants had freezing of gait, in line with previous findings.⁴ As expected, disease severity (Parkinson’s disease duration, UPDRS motor, and Hoehn and Yahr scale) was associated with freezing of gait, a well-established risk factor for the onset of the phenomenon.^8,9 Higher antiparkinsonian medication intake, as measured by using the levodopa equivalent dose, was also associated with freezing of gait, in line with previous studies.^8,9 This may reflect disease severity and an attempt to alleviate freezing of gait episodes.²²

Consistent with previous evidence, both anxiety (HADS-A)^2,11 and depression (HADS-D)^2,8 were associated with freezing of gait. Stressful situations, such as moving under time constraints, walking in crowded places, and performing a secondary task, are related to freezing of gait, suggesting its interaction with anxiety.^4,5 Moreover, both anxiety²⁴ and depression²⁵ have been found to be associated with increased stride time variability, an impairment related to freezing of gait.⁶ The link between mood disorders and freezing of gait may be explained by the contribution of limbic circuits to the underlying pathophysiology of the phenomenon.⁶

Interestingly, higher level of anxiety was shown to be associated with lower balance confidence, a factor related to postural instability and falls in people with Parkinson’s disease.^3,25 Although cause and effect conclusions cannot be drawn from this relationship, we argue that anxiety may act as a stressor, leading to feelings of not being able to perform some activities without imbalance or falling. On the other hand, an individual may have to perform some activities even if there is a lack of balance confidence for them, which in turn may contribute to anxiety. Freezing of gait would be likely to occur during both situations, given their stressful nature.

A history of recurrent falls in the past year, disability (Schwab and England scale and UPDRS-ADL), reduced balance confidence (Activities-Specific Balance Confidence scale), and quality of life (PDQ-8) were also associated with freezing of gait. Falls and freezing of gait are two closely related phenomena.⁴ Freezing of gait can suddenly disrupt balance and put people with Parkinson’s disease at risk of falling, being considered a strong predictor of increased fall rates and frequent falls.⁷ Freezing of gait may also lead to reduced balance confidence,²⁵ especially for performing tasks that are more challenging to balance and more likely to elicit freezing of gait, such as sweeping the floor, walking in a crowed place, being bumped, and walking on a slippery surface, as measured by using the Activities-Specific Balance Confidence scale. Therefore, freezing of gait may contribute to greater disability²² and reduced quality of life.⁸

Not surprisingly, most variables associated with the presence of freezing of gait were also associated with its severity. However, motor fluctuations only showed an association with freezing of gait severity. This may be explained by the higher frequency and longer duration of freezing of gait episodes during “off” states.²²

The results of the multivariate analysis showed that disease severity, as measured by using the Hoehn and Yahr scale, was the strongest contributor to freezing of gait severity; it accounted for a high proportion of the variance in FOG-Q severity scores. Anxiety and levodopa equivalent dose were also independently associated with freezing of gait severity. These findings provide additional support for the contribution of anxiety to greater freezing of gait severity,¹¹ taking into account not only the frequency but also the duration of the episodes. Since cognitive-attentional strategies are required to overcome freezing of gait episodes,^5,10 the association between anxiety and freezing of gait severity might be explained by the relationship between anxiety and set-attentional shifting impairment; people with Parkinson’s disease who are anxious may take more time to switch attention between tasks.¹¹ Thus, anxiety may exacerbate freezing of gait and make episodes last longer time.

The main strength of this study is that potential contributors of freezing of gait severity were simultaneously evaluated. Since anxiety may be a modifiable risk factor,¹ our findings support a need to routinely screen for mood disorders in people with Parkinson’s disease who present with freezing of gait. Future trials should evaluate whether treating anxiety with pharmacological and/or cognitive-behavioral therapies may reduce the frequency and duration of freezing of gait episodes. We acknowledge that freezing of gait may be under- or overestimated, as the FOG-Q is a self-report measure. However, the examiners demonstrated to all participants different types of freezing of gait episodes or showed videos to minimize this potential bias. Also, the smaller number of participants with freezing of gait might have influenced the number and strength of contributors that were found.

In summary, anxiety was found to be an independent contributor to freezing of gait severity and should be evaluated in people with Parkinson’s disease who present with freezing of gait. Future prospective studies are needed to explore the relationship between anxiety levels and the onset, frequency, and duration of freezing of gait episodes.