Arm swing asymmetry in people with Parkinson's disease and its relationship with gait: A systematic review and meta-analysis

Abstract

Background: Individuals with Parkinson's disease present arm swing alterations that can adversely affect their locomotion.

Objective: To identify differences in arm swing asymmetry (ASA) between individuals with Parkinson's disease (PD) and healthy individuals and to investigate the relationship between ASA, temporal-spatial gait parameters, and disease progression.

Methods: A literature search was conducted in PubMed, Scopus, ProQuest, Web of Science, and EBSCOhost up to February 2023. Cross-sectional studies evaluating parameters of arm swing (AS) and ASA were included. Methodological quality was assessed using the Critical Appraisal Checklist, and the quality of the evidence was measured with a modified Grading of Recommendations Assessment, Development, and Evaluation.

Results: Fourteen studies were included in the systematic review (1130 participants). Irrespective of the medication phase (ON or OFF) and the type of walk test employed, the meta-analysis showed moderate-quality evidence that individuals with PD have increased ASA amplitude (SMD = 0.84; 95% CI: 0.69, 0.99; I²= 0%).Very low-quality evidence suggests higher ASA velocity (SMD=0.64; 95% CI: 0.24, 1.05; I²=59%) and lower AS amplitude on both the most affected (ES = -1.99, 95% CI: -3.04, -0.94, I²: 91%) and the least affected sides (ES = -0.75, 95% CI: -1.05, -0.44; I²=66%). Meta-regression indicated that ASA is inversely related to disease duration (Z: -2.4892, P< 0.05) and motor symptoms progression (Z: -2.1336, P< 0.05).

Conclusions: Regardless of the medication phase and the type of walk test employed, individuals with PD exhibited greater ASA and decreased AS amplitude than healthy individuals. ASA decreases as the disease progresses and symptoms worsen.

Keywords: Arm; Arm swing asymmetry; Gait; Parkinson disease.

Fig. 1

PRISMA flowchart.

Fig. 2

Forest plot from the meta-analysis of arm swing asymmetry during gait in people with Parkinson's disease versus healthy individuals. Note: In studies with more than one PD group evaluated each comparison was included in separate pairs, with groups divided approximately equally between the comparisons⁴⁸ Curtze 2015 (a): PD stage II, ON; (b): PD stage III.IV, ON; (c): PD stage II, OFF; (d): PD stage III-IV; OFF Gera 2020 (a): PD GBA ON; (b) PD No GBA ON; (c): PD GBA, OFF; (d) PD No GBA, OFF Liu 2022 (a): PD stage I-II, OFF; (b) PD stage III-IV, OFF Mirelman 2016 (a): PD No gene mutations, ON; (b) PD gene mutations, ON Roggendorf 2012 (a): PD stage I, OFF; (b): PD stage II, OFF.

Fig. 3

A) Forest plot sensitivity analysis showing effect sizes for arm swing asymmetry in relation to ON/OFF medication status in people with Parkinson's disease during walking. B) Forest plot sensitivity analysis showing effect sizes for arm swing asymmetry in relation to the methods used to determine ASA in people with Parkinson's disease during walking. Note: In studies with more than one PD group evaluated each comparison was included in separate pairs, with groups divided approximately equally between the comparisons.⁴⁸ Curtze 2015 (a): PD stage II, ON; (b): PD stage III.IV, ON; (c): PD stage II, OFF; (d): PD stage III-IV, OFF. Gera 2020 (a): PD gene mutations ON; (b) PD No gene mutations ON; (c): PD gene mutations, OFF; (d) PD No gene mutation, OFF. Mirelman 2016 (a): PD No gene mutations, ON; (b) PD gene mutations, ON. Liu 2022 (a): PD stage I-II, OFF; (b) PD stage III-IV, OFF. Roggendorf 2012 (a): PD stage I, OFF; (b): PD stage II, OFF.