Turning is a common but challenging feature of daily life,1,2 triggering freezing of gait and falls3 in individuals with Parkinson disease (PD). Although the characteristics of turning impairments in PD (eg, excessive reduction of spatial gait parameters and impaired axial coordination)4 are well documented, few studies have investigated turning stability in PD.
Turning naturally induces instability to the body in the mediolateral direction, as it requires center of mass (COM) to momentarily be outside the lateral boundaries of base of support (BOS).5,6 To maintain stability while turning, it is necessary to regulate BOS by changes in step width with respect to changes in lateral COM displacement.7 Thus, a lateral shift of COM close to the boundaries of BOS hampers mediolateral stability, which provides lower ability to respond to mediolateral perturbations and thereby increasing the risk of instability.8
While numerous studies have shown that individuals with PD turn with narrower step width,9–12 varying results have been found for turning stability. Compared with healthy individuals, Mellone et al12 found that individuals with PD were spending more time with COM outside the lateral BOS during fast walking turns. In contrast, Bengevoord et al6 found that individuals with PD adopted a safe turning strategy characterized by reduced lateral shift of COM with respect to the BOS. Previous studies on turning stability in PD have been limited to preplanned turns (ie, when the walking direction is known in advance). However, in real-life situations, turning is often executed in an unplanned manner (eg, negotiating obstacles and navigating crowded environments),13,14 which is important to take into account while investigating turning.
Dopaminergic medication has a dramatic effect on motor symptoms15 and has shown to improve straight walking in individuals with PD.16–20 Although medication has shown to improve global measures of turning performance (eg, time and number of steps required to turn),10,16,21–23 measures related to mediolateral stability (eg, step width) have shown to be less responsive to medication.10,24 To our knowledge, no previous study has investigated the effects of dopaminergic medication on mediolateral stability during turning in individuals with PD.
The overarching aim of this study was to investigate mediolateral stability during pre- and unplanned walking turns between individuals with PD and healthy individuals and to investigate whether dopaminergic medication improves mediolateral stability. Compared with their healthy counterparts, we expected impaired mediolateral stability in individuals with PD due to narrow step width.6,9–12 Furthermore, we anticipated larger reduction of mediolateral stability in PD for unplanned turns due to its challenging features. Finally, as instability and falls during turning occur in PD despite medication intake,3 we expected limited effects of medication on turning stability.
Nineteen individuals with PD were recruited to this study from participants in a randomized controlled trial investigating the effects of balance training in individuals with PD.25 Inclusion criteria for the randomized controlled trial were 60 years or older, Hoehn and Yahr stage 2 or 3, being treated with oral dopaminergic medication, able to walk indoors without assistance or a walking aid, a Mini-Mental State Examination score of 24 or more, and absence of musculoskeletal and other neurological impairments affecting gait or balance. Specific exclusion criteria for this study were prior brain surgery and severe dyskinesia or freezing of gait. Screening for freezing of gait was based on self-reported assessment using item 14 of the activities of daily life section of the Unified Parkinson's Disease Rating Scale (UPDRS) and clinical assessment of turning. We also recruited 19 healthy individuals (matched for age and gender) without any medical condition affecting gait or balance performance as a reference group. There were no significant differences between individuals with PD and healthy individuals for any characteristics (see the Table). This study was approved by the Regional Board of Ethics in Stockholm, and all participants provided written informed consent prior to their enrolment in the study.
||Individuals With Parkinson Disease (n = 19)
||Healthy Participants (n = 19)
|Body mass index, kg/m2
|Disease duration, y
|Daily levodopa dose equivalency, mg
|Straight walking velocity,c,d m/s
Abbreviations: PD-OFF, Individuals with Parkinson disease off medication; PD-ON, Individuals with Parkinson disease on medication; UPDRS, Unified Parkinson's Disease Rating Scale.
aValues are mean (standard deviation) for all variables except gender and recurrent fallers that are presented as proportion.
bParticipants who had experienced 2 falls or more during the previous 12 months were classified as recurrent fallers.
cSignificant differences (P ≤ 0.025) between the PD-OFF and PD-ON medicated states (Wilcoxon signed rank test).
dSignificant differences (P ≤ 0.025) between the PD-ON and reference groups (Mann-Whitney U test).
Participants walked straight at their self-selected comfortable speed along a 9-m walking lane where the turning intersection was indicated by 2 poles (see Figure 1A). One of the following 3 tasks was performed in a randomized order: walking straight, and walking and turning 180° to the right or to the left. Participants started each trial 6.65 m from the turning intersection to allow steady-state straight walking speed before initiating the turn. Participants were instructed to walk and turn without stopping and taking the most direct path to the target positioned 2.5 m to the right/left of the walking lane. Two different types of walking turns (preplanned and unplanned) were performed. In the preplanned condition, the walking direction was provided by a visual signal, located at the end of the walking lane, before participants started to walk, while in the unplanned condition, the same visual signal appeared approximately 1 step length (0.6 m) prior to the intersection point (see squares in Figure 1A). The pre- and unplanned turning condition consisted of 15 trials per subject (ie, 5 trials for straight walking, right and left turning). Before the start of data collection, practice sessions were performed to familiarize participants with the procedure.
The PD group was assessed twice, first after overnight withdrawal of medication (OFF, average off time = 16 hours, range: 12-22 hours) and then approximately 1 hour after taking their usual morning dose of dopaminergic medication (ON). Between OFF and ON testing, all participants with PD were taking levodopa (n = 19) and 8 participants were also taking a dopamine agonist. Since assessments over 2 days were not possible for many of the participants, both test sessions were performed on the same day. A trained physical therapist assessed motor impairments before each test session (OFF and ON) using the motor section of the UPDRS.
Since walking velocity can influence movement characteristics during walking turns,26,27 the reference group turned at a pace that matched the comfortable pace of the subjects with PD. For most healthy individuals, the matched velocity was achieved by instructing participants to walk slower than their comfortable speed. To achieve a matched turning speed, walking speed was assessed with a hand-held stopwatch and by counting the number of steps over a distance of 2 m prior to the turn. In cases where healthy participants deviated from the targeted speed, they were instructed to change their walking speed (ie, to walk faster or slower). Subsequently, trials from the reference group were included if their walking velocity (prior to the intersection position) was within 1 standard deviation of the mean velocity of the individuals with Parkinson disease on medication (PD-ON).
An 8-camera motion analysis system (Elite 2002, version 2.8.4380; BTS, Milano, Italy) was used to record at 100 Hz the position of 9 spherical retroreflective markers located on the spinous process of the seventh cervical vertebrae (C7) and bilaterally on the head, acromion, posterior superior iliac spine, and heel. Three-dimensional trajectories of the markers were reconstructed using a tracking system (Tracklab-BTS, Milan, Italy). Data were processed and filtered (Butterworth low-pass filter: 7-Hz cutoff frequency) using MATLAB software (MATLAB, version 7.4.0, MathWorks, Natick, Massachusetts).
The outcome variables retained for analysis focused on turning performance (ie, turning rotation, speed, and distance) and turning stability (ie, mediolateral stability, step width, and pelvis lateral displacement). We focused our analysis on 5 turning steps since this was sufficient for a majority of the participants to complete 120° or more of the turn. For turning rotation, the magnitude of pelvis at the fifth turning step compared with the laboratory reference axis was retained for analysis. Mean turning speed (ie, the first derivative of the tangential displacement of the C7 marker) and mean turning distance (ie, the cumulative linear displacements of the C7 marker in the horizontal plane) were calculated for the 5 turning strides.
Heel strike events were determined based on the vertical velocity profiles of the heel markers. Step width was calculated as the distance between the foot perpendicular with the line of progression.28 The first turning step was identified as the first heel strike exceeding 2 standard deviations in mediolateral displacement of 5 straight walking trials (computed for each participant) in the turning direction.10
As illustrated in Figure 1B, 2 turning strategies were identified: step strategy (ie, first turning step was ipsilateral to the turning direction) and spin strategy (ie, first turning step was contralateral to the turning direction).29 As a proxy for COM, the center of the pelvis segment was calculated as the average position between right and left posterior superior iliac spine.30 Pelvis lateral displacement was calculated as the distance between the projection of the center of the pelvis segment and the perpendicular line from the line of progression. Pelvis lateral displacement was retained at approximately 80% of the gait cycle (corresponding to midstance) for both straight walking and turning. As the primary outcome for mediolateral stability, we calculated the absolute difference between pelvis lateral displacement and step width (positive values: step width > pelvis lateral displacement) for walking and turning strides. This outcome represents the distance between the pelvis and the mediolateral margin of BOS; larger values indicate that pelvis is further away from the border of the BOS and closer to the line of progression.
Statistical analyses were carried out using IBM SPSS, version 23.0 (SPSS Inc, Chicago, Illinois). Equality of variance and data normality were tested using Levene's test combined with a visual inspection of the residual curve. No differences were found between turns to the right or left or turning strategy (spin or step turn). Thus, right and left trials were collapsed together for further analysis. Mixed-model analyses were used to analyze differences between groups (PD-ON vs healthy participants) and effects of medication (individuals with Parkinson disease off medication [PD-OFF] vs PD-ON) on turning performance (speed, rotation, and distance). We previously showed a nearly 50:50 distribution between step and spin turns in PD and healthy participants and different step width regulation for these turning strategies.10 Therefore, mediolateral stability, step width, and pelvis lateral displacement were analyzed separately for turning strategies (ie, step and spin turns). For mediolateral stability, step width and pelvis lateral displacement, mixed-model analyses were used to analyze differences between groups (PD-ON vs healthy participants) and steps (turning steps 1, 2, 3, 4, and 5) and effects of medication (PD-OFF vs PD-ON) and steps (turning steps 1, 2, 3, 4, and 5). In case of a significant interaction effect (ie, group × steps or medication × steps), Tukey's honestly significant difference tests were performed as post hoc tests. The significance level was set at P ≤ 0.025 due to multiple statistical comparisons. Data are presented as mean and 95% confidence intervals.
Motor Symptoms and Turning Performance
ON medication, individuals with PD improved by 22% in their UPDRS motor score (P = 0.010, see the Table) and by 4% to 7% in turning speed, turning rotation, and turning distance during pre- and unplanned turns (medication: P < 0.025, Figure 2). Although healthy participants turned at the same speed as individuals with PD (group: P = 0.488), they displayed 6% higher turning rotation and 13% greater turning distance compared with PD-ON (group: P < 0.013, Figure 2).
Regulation of Step Width and Pelvis Displacement
As illustrated in Figures 3A-B and 4A-B, there were no differences in step width and pelvis lateral displacement between PD and healthy participants for straight walking. For PDs and healthy participants, a general pattern of alternating step width was observed, that is, widening—narrow for step turns (Figure 3A) and the opposite, narrow—widening for spin turns (Figure 4A). Widening turning steps (ie, step strategy: steps 1, 3, and 5, spin strategy: steps 2 and 4) were characterized by larger pelvis lateral displacement than straight walking while pelvis displacement was similar to or smaller than straight walking for narrow crossover steps (ie, step strategy: steps 2 and 4, spin strategy: steps 1, 3, and 5, see Figures 3B and 4B).
While initiating pre- and unplanned turns with the step strategy, individuals with PD used 12% to 16% narrower step width (group: P < 0.020, Figure 3A) and 22% to 38% smaller pelvis lateral displacement during the widening turning steps compared with healthy participants (group × step interaction: P < 0.001, Figure 3B). While using the spin strategy for preplanned turns, individuals with PD demonstrated 24% to 36% smaller pelvis displacement for widening steps (P < 0.002, Figure 4B). For unplanned turns, individuals with PD demonstrated a 53% to 55% reduction in step width for the first 2 narrow steps (ie, steps 1 and 3) (P < 0.024, Figure 4A) and smaller pelvis displacement during turning steps 2, 3, and 4 compared with healthy participants (P < 0.004, Figure 4B).
Irrespective of turning strategy and turning condition, mediolateral stability was differently regulated in individuals with PD as compared with healthy participants (group × step interaction: P ≤ 0.008, Figures 3C and 4C). Specifically, individuals with PD demonstrated an alternating pattern of mediolateral stability; that is, the distance between the pelvis and the lateral margin of BOS was greater for widening turning steps and decreased for narrow crossover steps. In contrast, healthy participants sustained a larger distance between the pelvis and boundaries of BOS during turning than straight walking (except while using the spin strategy for preplanned turns) and preserved that margin across the majority of the turning steps.
There were no effects of medication on mediolateral stability when adopting the step strategy for preplanned turns (P > 0.032) or the spin strategy for pre- and unplanned turns (P > 0.091). However, when using the step strategy for unplanned turns, medication led to a 18% greater distance between the pelvis and the lateral margin of BOS (P < 0.001) owing to a 7% wider step width during ON compared with OFF medication (P = 0.005, Figures 2A-B).
This study assessed proactive and reactive turns, analyzed turning stability with respect foot step adjustments, and matched walking speed of healthy individuals to that of subjects with PD. Our results demonstrate that turning stability is compromised in PD specifically during narrow crossover steps. While healthy participants augmented mediolateral stability while turning, it largely fluctuated in individuals with PD due to impaired scaling of pelvis lateral displacement and step width regulation. Finally, we found that dopaminergic medication had limited effects on turning stability in PD, implying that rehabilitation should focus on promoting safe turning strategies with a specific emphasis on sustainment of a wide BOS.
As individuals with PD are more prone to falls while using narrow crossover steps to recover balance,24 narrow turning steps are likely to destabilize individuals with PD and could contribute to falls in everyday living. In contrast, enhanced mediolateral stability during widening turning steps (ie, larger distance between the pelvis and boundaries of BOS) may have a stabilizing purpose, potentially to recover instability induced by more destabilizing crossover steps. Healthy participants, on the other hand, maintained a large difference between pelvis lateral shift and the margins of BOS across wide and narrow turning steps, suggesting a more fine-tuned scaling between pelvis lateral displacement and regulation of step width in healthy older adults. However, while mediolateral stability remained unchanged in healthy participants for most steps during the preplanned condition, a slightly larger variation between steps was found for this outcome during unplanned turns. Unplanned turns could be more perturbing and challenging, as they require quick modifications of the walking pattern. Thus, the more variable pattern of mediolateral stability during unplanned turns suggests a more destabilizing effect of such reactive turning conditions on individuals with PD.
Furthermore, compromised turning stability in PD was a robust finding—independent of turning strategy and turning condition, suggesting a specific scaling deficit of movement amplitude related to basal ganglia dysfunction.18 In line with our findings, individuals with PD have shown to use narrow step width9–12 and reduced lateral displacement of COM while turning.6 This decrease in movement amplitude during turning might reflect an effort to compensate for axial impairments (ie, “en-bloc”)4 or for the increased postural challenges induced by turning itself. Noteworthy, step width and COM are important for efficient turning9,11 and downregulation of these features could constrain turning performance, as seen in our data.
Although participants with PD improved their motor symptoms and turning performance, dopaminergic medication only improved turning stability for unplanned turns while using the step strategy, owing to wider step width ON medication. External cues facilitate motor performance in PD,31 it is therefore plausible that the visual cues used to trigger and guide unplanned turns in combination with the less challenging features of step turns (ie, wider BOS and less axial rotation)32 could have contributed to this effect. As dopaminergic medication has little effect on mediolateral stability, our results are in accord with others regarding the crucial role rehabilitation plays in the development of strategies to improve turning stability in individuals with PD.33 In a clinical perspective, our results suggest that fall prevention programs should focus on step width regulation, with a specific focus on practicing execution of safe crossover turning steps.
We aimed to investigate turning stability in individuals with mild to moderate PD without freezing of gait; therefore, these results can only be generalized to this specific population. To control for differences in walking speed between PD and the reference group, we instructed healthy participants to walk slower than their comfortable walking speed. This speed-matching procedure could cause a more executive control strategy (ie, increase the awareness on “how” they walk), which in turn could result in abnormal turning behavior. However, in a recent study, we investigated the effects of walking speed on turning stability in older healthy adults and found that the pattern of mediolateral stability was similar while walking in a slow and comfortable pace.34 Participants with PD were also tested on the same day (ie, OFF medication was tested prior to ON medication), which may have led to an underestimation of the medication effect (eg, due to fatigue) or to an overestimation induced by a practice effect from repeated testing. Thus, to address the possible bias of the speed-matching procedure and repeated testing, practice sessions were performed prior to testing to familiarize participants with the procedure. To avoid the risk of fatigue, brief resting sessions were allowed during testing. A further limitation of this study is that we did not include outcomes of axial coordination, which is believed to influence turning stability by reducing the counterrotation between trunk and pelvis.4
Turning stability is compromised in individuals with PD, specifically during narrow crossover steps when compared with healthy participants. Contrasting a more steady-state control of mediolateral stability in healthy participants, individuals with PD fluctuated their mediolateral stability during turning, which could reflect a specific scaling deficit related to basal ganglia dysfunction. As dopaminergic medication had limited effect on turning stability, rehabilitation plays an important role to promote safe turning strategies, with a specific emphasis on sustainment of a wide support base.
The authors would like to thank all the participants who contributed to this work and Niklas Löfgren and Håkan Nero for their support with data collection.
1. Mancini M, El-Gohary M, Pearson S, et al. Continuous monitoring of turning in Parkinson's disease: rehabilitation potential. NeuroRehabilitation. 2015;37(1):3–10.
2. Glaister BC, Bernatz GC, Klute GK, Orendurff MS. Video task analysis of turning during activities of daily living. Gait
3. Bloem BR, Grimbergen YA, Cramer M, Willemsen M, Zwinderman AH. Prospective assessment of falls in Parkinson's disease. J Neurol. 2001;248(11):950–958.
4. Hulbert S, Ashburn A, Robert L, Verheyden G. A narrative review of turning deficits in people with Parkinson's disease. Disabil Rehabil. 2015;37(15):1382–1389.
5. Orendurff MS, Segal AD, Berge JS, Flick KC, Spanier D, Klute GK. The kinematics and kinetics of turning: limb asymmetries associated with walking a circular path. Gait
6. Bengevoord A, Vervoort G, Spildooren J, et al. Center of mass trajectories during turning in patients with Parkinson's disease with and without freezing of gait
7. Patla AE, Adkin A, Ballard T. Online steering: coordination and control of body center of mass, head and body reorientation. Exp Brain Res. 1999;129(4):629–634.
8. Hof AL, Gazendam MG, Sinke WE. The condition for dynamic stability
. J Biomech. 2005;38(1):1–8.
9. Huxham F, Baker R, Morris ME, Iansek R. Footstep adjustments used to turn during walking in Parkinson's disease. Mov Disord. 2008;23(6):817–823.
10. Conradsson D, Paquette C, Lökk J, Franzen E. Pre- and unplanned walking turns in Parkinson's disease: effects of dopaminergic medication
. Neuroscience. 2017;341:18–26.
11. Mak MK, Patla A, Hui-Chan C. Sudden turn during walking is impaired in people with Parkinson's disease. Exp Brain Res. 2008;190(1):43–51.
12. Mellone S, Mancini M, King LA, Horak FB, Chiari L. The quality of turning in Parkinson's disease: a compensatory strategy to prevent postural instability? J Neuroeng Rehabil. 2016;13:39.
13. Shumway-Cook A, Patla AE, Stewart A, Ferrucci L, Ciol MA, Guralnik JM. Environmental demands associated with community mobility in older adults with and without mobility disabilities. Phys Ther. 2002;82(7):670–681.
14. Musselman KE, Yang JF. Walking tasks encountered by urban-dwelling adults and persons with incomplete spinal cord injuries. J Rehabil Med. 2007;39(7):567–574.
15. Connolly BS, Lang AE. Pharmacological treatment of Parkinson disease
: a review. JAMA. 2014;311(16):1670–1683.
16. Curtze C, Nutt JG, Carlson-Kuhta P, Mancini M, Horak FB. Levodopa
is a double-edged sword for balance and gait
in people with Parkinson's disease. Mov Disord. 2015;30(10):1361–1370.
17. McNeely ME, Duncan RP, Earhart GM. Medication improves balance and complex gait
performance in Parkinson disease
18. Morris M, Iansek R, McGinley J, Matyas T, Huxham F. Three-dimensional gait
biomechanics in Parkinson's disease: evidence for a centrally mediated amplitude regulation disorder. Mov Disord. 2005;20(1):40–50.
19. Shan DE, Lee SJ, Chao LY, Yeh SI. Gait
analysis in advanced Parkinson's disease—effect of levodopa
and tolcapone. Can J Neurol Sci. 2001;28(1):70–75.
20. O'Sullivan JD, Said CM, Dillon LC, Hoffman M, Hughes AJ. Gait
analysis in patients with Parkinson's disease and motor fluctuations: influence of levodopa
and comparison with other measures of motor function. Mov Disord. 1998;13(6):900–906.
21. Hong MN, Earhart GM. Effects of medication on turning deficits in individuals with Parkinson's disease. J Neurol Phys Ther. 2010;34(1):11–16.
22. McNeely ME, Earhart GM. The effects of medication on turning in people with Parkinson disease
with and without freezing of gait
. J Parkinson Dis. 2011;1(3):259–270.
23. Franzen E, Paquette C, Gurfinkel VS, Cordo PJ, Nutt JG, Horak FB. Reduced performance in balance, walking and turning tasks is associated with increased neck tone in Parkinson's disease. Exp Neurol. 2009;219(2):430–438.
24. King LA, Horak FB. Lateral stepping for postural correction in Parkinson's disease. Arch Phys Med Rehabil. 2008;89(3):492–499.
25. Conradsson D, Löfgren N, Nero H, et al. The effects of highly challenging balance training in elderly with Parkinson's disease: a randomized controlled trial. Neurorehabil Neural Repair. 2015;29(9):827–836.
26. Forsell C, Conradsson D, Paquette C, Franzen E. Reducing gait
speed affects axial coordination of walking turns. Gait
27. Akram SB, Frank JS, Chenouri S. Turning behavior in healthy older adults: is there a preference for step versus spin turns? Gait
28. Huxham F, Gong J, Baker R, Morris M, Iansek R. Defining spatial parameters for non-linear walking. Gait
29. Hase K, Stein RB. Turning strategies during human walking. J Neurophysiol. 1999;81(6):2914–2922.
30. Yang F, Pai YC. Can sacral marker approximate center of mass during gait
and slip-fall recovery among community-dwelling older adults? J Biomech. 2014;47(16):3807–3812.
31. Cassimatis C, Liu KP, Fahey P, Bissett M. The effectiveness of external sensory cues in improving functional performance in individuals with Parkinson's disease: a systematic review with meta-analysis. Int J Rehabil Res. 2016;39(3):211–218.
32. Taylor MJ, Dabnichki P, Strike SC. A three-dimensional biomechanical comparison between turning strategies during the stance phase of walking. Hum Mov Sci. 2005;24(4):558–573.
33. Yang WC, Hsu WL, Wu RM, Lin KH. Immediate effects of clock-turn strategy on the pattern and performance of narrow turning in persons with Parkinson disease
. J Neurol Phys Ther. 2016;40(4):249–256.
34. Conradsson D, Paquette C, Franzen E. Medio-lateral stability during walking turns in older adults. PLoS One. 2018;13(6):e0198455.