Postural instability in Parkinson Disease: To step or not to step

Highlights

• Parkinson subjects and Controls required the same force to be pulled off-balance

• All Controls took a step in direction of pull to maintain balance without falling

• 59% of the time Parkinson subjects did not take a step to maintain balance and fell

• Parkinson subjects do not defend against sudden gravitational instability

• Parkinson gait disorder may be due to subjects avoiding gravitational instability

Abstract

Postural instability is a key feature of Parkinson Disease that is associated with falls and morbidity. We designed a pull apparatus to quantitatively measure the force needed to pull subjects off-balance. Thirteen Controls and eight individuals with Parkinson Disease (PD) were evaluated. All individuals with PD reported subjective symptoms of postural instability and were symptomatic for approximately 9.4 years when tested. No significant differences were found between Controls and PD subjects in the magnitude of force required to pull them off-balance. None of the Controls fell and all took a step into the direction of pull to maintain their balance. 59% of the time PD subjects fell because they did not take a step in the direction of pull to maintain their center of mass (COM) over their feet, thus indicating a deficiency in postural reflexes. If they fell on the first pull, PD subjects did not show a learning effect when pulled multiple times in the same direction. The utility of the Pull Test to detect postural instability is related to the subject's behavioral response, not the force needed to pull them off balance. Our findings may also help explain certain features of the PD gait as an attempt by subjects to avoid postural instability by not placing their COM in gravitationally unstable positions.

Introduction

Postural instability (PI) is a cardinal manifestation of Parkinson disease (PD) that is associated with falls and increased morbidity [6], [7], [18], [46]. PI usually becomes problematic approximately a decade after the onset of PD symptoms [48]. PI, along with speech and gait deficits, does not generally respond to dopaminergic medications [4], [8], [32], [47]. PI and falls have been associated with the akinetic–rigid as opposed to the tremor-dominant subtype of PD [30], [43]. A recent quantitative research study found that axial rigidity, especially involving the neck, is associated with balance and gait problems in subjects with PD [15]. In a meta-analysis of six studies, Pickering et al. [38] found that, except for tremor, all items from United Parkinson's Disease Rating Scale (UDPRS; [13], [26], [27]) were related to falls with the strongest association between gait and speech abnormalities.

A number of studies have sought to establish an effective means to clinically evaluate PI. The most common method is the backward Pull Test, which is a standard part of the UDPRS (item 30). The Pull Test [3], [5], [35] and related PI tests [28], such as the Push and Release Test [44] and the Steady Stance Test [46], have been analyzed for ability to predict PI. Some studies have suggested that the Pull Test is not a particularly effective method for predicting PI and falls [7], [29], [44] whereas others have shown relatively good overall predictive accuracy [23], [46], especially when using an unexpected [6], [24] rather than an expected pull [13], [40]. Although the mechanics and purpose of the Pull Test have been refined and standardized in the MDS-UPDS (Movement Disorder Society-UPDS [17]), it is not clear if one of the factors in testing relates to how much force is needed to displace a subject and, if the subject takes a step to maintain balance, does this constitute evidence for mild PI [26]. Finally, because of safety issues, the Pull Test is only done in the backward direction so that the examiner is able to catch a subject should they fall [17], [26], thus it may not capture PI in other pull directions.

Our research objective was to acquire quantitative data using a modified Pull Test apparatus. We expected that the results would lead to a better understanding of the mechanisms of PI in individuals with PD. We also hoped to develop a more reliable clinimetric that also included an assessment of PI in the forward and side directions. Based on previous literature [23], [26], [38] and personal observations, we expected that PI in individuals with PD would 1) be related to a reduced amount of force needed to pull them off-balance compared to Controls and 2) be worse for backward compared to other pull directions.