*3.4. Relationship between the Standing Postural Profile and the GI*

We did not find any significant correlation between the postural angles and the GI measurements. However, when not correcting for multiple comparisons, the shank angle was predictive for velocity variables of the IMB phase. The results are shown in Table 5.

**Table 5.** Correlation between the shank angle and gait initiation measurements. Only significant partial correlations between postural angles and gait initiation measurements corrected for the influence of the group variable are shown (Spearman's ρ, *p*-value < 0.05). No correlation was significant after Bonferroni correction for multiple comparisons.


### *3.5. Pattern of Movements during GI*

The overall pattern of segmental movements during GI did not show clear differences between groups (Tables 6 and 7). However, PDF showed shorter times of movement onset for almost all ranked segments (Table 7), possibly suggesting tight inter-segmental coupling [50]. All groups started preferably with the swing or stance arm, especially the swing hand for HCs and PDNF and the stance hand for PDF (Figures S1 and S2). The abdomen was often the last body segment moved by HCs and PDNF, but not by PDF (Figure S1). Of note, we observed a remarkable inter-subject variability of SCoM onset times, especially for PD, as shown by the high value of the standard deviation (Table 6) and the large dispersion of the temporal order of SCoM movement onsets (Figure S1), which probably prevented us from capturing statistically significant differences.

**Table 6.** Onset of segmental movements at gait initiation. Data are shown as mean (standard deviation). Time of movement onset of each segmental center of mass was expressed as the percentage with respect to total gait initiation duration (i.e., from the onset of the anticipatory postural adjustments to the heel contact of the swing foot) and compared across groups (Dunn's test, no difference was significant after Bonferroni correction for multiple comparisons). Abbreviations: HC, healthy controls; PDF, Parkinson's disease with freezing of gait; PDNF, Parkinson's disease with no freezing of gait; ST: stance limb; SW: swing limb.



**Table 6.** *Cont.*

**Table 7.** Onset of rank-ordered segmental movements at gait initiation. Data are shown as mean (standard deviation). Time of movement onset of rank-ordered segmental centers of mass was expressed as the percentage with respect to total gait initiation duration (i.e., from the onset of the anticipatory postural adjustments to the heel contact of the swing foot) and compared across groups (Dunn's test, no difference was significant after Bonferroni correction for multiple comparisons). Abbreviations: HC, healthy controls; PDF, Parkinson's disease with freezing of gait; PDNF, Parkinson's disease with no freezing of gait.


#### **4. Discussion**

This study aimed to evaluate the specific biomechanical alterations of APAs at GI in PD patients with a positive history of FOG, accounting for known confounders such as medication condition, anthropometric measurements, base of support, and initial stance posture. The CoP displacement and velocity during the imbalance phase were altered in both PDNF and PDF patients, but more prominently in the latter group. The CoP displacement along the anteroposterior axis during the unloading phase was impaired only in PDF patients. The order of SCoM movements was unaltered in the two patient groups. The postural profile did not correlate with GI outcome measurements.

Our findings are in line with previous studies in PD that showed an impairment in APAs' production at GI [20,51]. However, a direct comparison with earlier works is limited because we aimed to minimize possible bias from cueing or imposed postural constraints that are known to affect the execution of the GI task [9,11,12,18,22,52–54].

We have now shown that there is a profound alteration of APA execution in PDF patients, which cannot be attributed to specific demographic or clinical features (such as disease severity and duration, medication dose, and efficacy) as the patient groups were matched for all of these features [55].

The IMB phase of APA execution was significantly altered in all PD patients, particularly in PDF (Table 3). Increasing evidence suggests that this GI phase is governed by centrally mediated feedforward signals and involves the cortico-basal ganglia loop, with the SMA-proper and the striatum chiefly contributing to the execution of these preprogrammed movements [6,8,9,11,12,56–60]. In PD, we have previously shown a detrimental effect of striatal dopamine loss in the IMB execution at GI [12]. Recent studies in Parkinsonian patients suggested that striatal dopamine may in part enable normal movement by encoding sensitivity to the energy cost of a movement [61–64]. Therefore, from the perspective of motor planning, especially of patterned and consolidated motor actions such as APAs, a reduced tonic dopaminergic activity could reframe the coding of the expected energetic costs and impair motor control [63].

In our study, we also showed a prominent alteration in the AP displacement of the CoP during the UNL phase in PDF patients. We interpret this result as a possible alteration, mainly of PDF patients, in the processing and integration of somatosensory information prior to stepping [6,14,65,66]. A chief contribution to integrate proprioceptive and voluntary components for a proper weight transfer during GI can be expected from the premotor– parietal–cerebellar loop [14,58,67–71]. An impaired ability to inhibit stance postural control and initiate stepping and poor set-shifting is also included in pathophysiological hypotheses of FOG in PD [5,10,57,72–77].

Despite impaired APA execution, the sequencing of the movement did not show major alterations in the PD groups. We speculate that additional inputs from the cerebellum could overcome impaired information processing by favoring internal movement timing [78]. The efficacy of an online compensatory role of the cerebellum [70,78] is suggested in our study by the relatively preserved SCoM temporal movement sequencing [79], which could have also prevented the appearance of any gait freezing episode during our acquisitions. Relative timing of segmental movements was also described as unaltered in patients with PD by Rosin and colleagues (1997), further suggesting a compensatory rather than detrimental role of the cerebellum in Parkinsonian patients with FOG and balance disturbances [60,78,80]. Of relevance, the high variability in the SCoM movement onsets might have prevented us from detecting differences across groups. Further studies with larger cohorts might further explore this aspect to definitively rule out the presence of PD-related alterations in the movement sequencing.

We envisioned a significant impact of postural abnormalities on GI in PD, but our results did not support this hypothesis. Interestingly, our findings instead confirmed previous physiological studies reporting no correlation between APA execution at GI and the natural inclination of the trunk [33] or of a forward leaning up to 30% of the maximum voluntary lean [35].

Our study suffers from some limitations. First, although we reduced as much as possible the influence of known confounders (i.e., initial feet position and posture, anthropometric parameters, and cues), we cannot fully exclude a residual influence of Parkinsonian symptoms such as bradykinesia and rigidity on the task performance [37]. However, in our previous work [12], we showed that levodopa intake, by improving bradykinesia and rigidity, increases the length and speed of the first step at GI, but does not affect the AP shift during UNL. We can thus hypothesize that the alterations in AP displacement during UNL in the PDF group are not related to akinetic-rigid symptoms, but to impairment of the motor program itself. Future studies are needed to better clarify this aspect. Second, the limited sample size and very stringent statistics may have limited the detection of differences between groups (e.g., SCoM onset times). Third, the lack of a brain imaging evaluation in this study prevents any firm conclusions about our pathophysiological interpretation of the kinematic and dynamic findings, but they match well with the brain metabolic activity changes [66,68] and network derangements [81–84] described during actual gait and gait freezing episodes in Parkinsonian patients.

In conclusion, our data demonstrate substantial impairment of feedforward motor programming mechanisms at GI in Parkinsonian patients. The deterioration of the UNL and stepping in PDF patients would suggest an additional impaired integration of postural and locomotor programs subserving gait initiation and modulation, which might be partly compensated by cerebellar mechanisms triggering time-locked models of body movement. Postural alterations seem to play a minor role in GI impairment in patients with PD. Last, but not least, our results suggest the potential clinical utility of recording the CoP displacement during GI, and particularly its AP shift during the UNL to identify patients at risk of FOG and to monitor the efficacy of therapeutic strategies. Future longitudinal studies may support this assumption.

**Supplementary Materials:** The following supporting information can be downloaded at https: //www.mdpi.com/article/10.3390/bioengineering9110639/s1, Figure S1: Rank order of segmental CoM onset times; Figure S2: Rank order of onset times of groups of segmental CoM.

**Author Contributions:** Conceptualization, C.P. and I.U.I.; methodology, C.P. and I.U.I.; software, C.P. and S.H.; formal analysis, C.P., L.B., S.H. and I.U.I.; investigation, C.P. and I.U.I.; resources, C.P., G.P. and I.U.I.; data curation, C.P. and L.B.; writing—original draft preparation, C.P. and L.B.; writing—review and editing, J.V. and I.U.I.; visualization, C.P.; supervision, S.H., J.V. and I.U.I.; project administration, J.V. and I.U.I.; funding acquisition, J.V., G.P. and I.U.I. All authors have read and agreed to the published version of the manuscript.

**Funding:** The study was sponsored by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation)—Project-ID 424778381-TRR 295 and the Fondazione Grigioni per il Morbo di Parkinson. C.P. was supported by a grant from the German Excellence Initiative to the Graduate School of Life Sciences, University of Würzburg. L.B. was supported by a grant from New York University School of Medicine and The Marlene and Paolo Fresco Institute for Parkinson's and Movement Disorders, which was made possible with support from Marlene and Paolo Fresco. This publication was supported by the Open Access Publication Fund of the University of Würzburg.

**Institutional Review Board Statement:** The study was conducted in accordance with the Declaration of Helsinki and approved by the Ethics Committee of the Università degli Studi di Milano (5/16, 15.02.2016).

**Informed Consent Statement:** Informed consent was obtained from all subjects involved in the study.

**Data Availability Statement:** The data presented in this study are available upon request from the corresponding author. The data are not publicly available for privacy reasons.

**Acknowledgments:** We would like to thank all patients and caregivers for their participation. Our special thanks go to Paolo Cavallari for his help in conducting the study and to Monica Norcini for study management and administrative support. The study was sponsored by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation)—Project-ID 424778381-TRR 295 and the Fondazione Grigioni per il Morbo di Parkinson. C.P. was supported by a grant from the German Excellence Initiative to the Graduate School of Life Sciences, University of Würzburg. L.B. was supported by a grant from New York University School of Medicine and The Marlene and Paolo Fresco Institute for Parkinson's and Movement Disorders, which was made possible with support from Marlene and Paolo Fresco. This publication was supported by the Open Access Publication Fund of the University of Würzburg.

**Conflicts of Interest:** The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript; or in the decision to publish the results.

#### **References**

