Biais de la perception de la verticale dans la maladie de Parkinson : revue de la littérature et méta-analyse. a systematic review and meta-analysis

Introduction. Postural disorders are frequent and represent a major source of disability in Parkinson’s disease (PD). Their mechanisms remain to be clarified. The possibility of a bias in the construction of the internal model of verticality is emerging. The number of studies investigating the perception of the vertical in PD has recently much increased. Time has come to synthetize them. This was the objective of this systematic review and meta-analysis. Methodology. We systematically searched in Medline and Cochrane databases from 1960 to December 2021 using the terms "Parkinson Disease", "verticality perception" and their derivatives. We followed the PRISMA and Cochrane recommendations. Eligibility for inclusion, data extraction and study quality (Newcastle-Ottawa scale, NOS) were assessed by two investigators (AT, CP), who followed a standardized protocol. We are preparing the Prospero Registration, and will update the study at due time. We analyzed frequency and magnitude of tilts found in verticality perception by random-effects models meta-analysis. We respected the definition of abnormal tilts given in each study. Results. We identified 17 studies (618 individuals; mean age 66.3 years; 55% male; mean disease duration 8.9 years), the majority of which were published between 2010 and 2020. The quality of the studies was adequate, with a mean NOS score of 7.4 (SD=1.14). Only the visual vertical (VV) could be investigated through a meta-analysis, because of too few studies with small sample size on haptic (n=2) and postural verticals (n=4). For VV, the pooled frequency of abnormal tilts was high: 45% (95% confidence interval [CI] [20; 70], I²=94.5%) in all individuals with PD regardless of it severity and duration of disease course, reaching 62% (95% CI [19; 106], I²=92%) in individuals with PISA syndrome (PS). In all individuals tested, the pooled magnitude of VV tilts was 0.2° (95% CI [-0.57; 0.89], I2=0%) in signed values, and 3.6° (95% CI [1.83; 5.34], I2=52.2%) in absolute values. In individuals in whom VV was abnormally tilted, the pooled magnitude of VV tilts was 5.3° in absolute value (95% CI [3.4; 7.3], I²=52.2%). Few studies investigated the concordance between VV tilts and the side where basal ganglia were predominately altered, as well as the concordance with a body tilt if any (PS). When analyzed this concordance is weak. Conclusion. This is the first systematic review of verticality perception in PD, conducted on 17 studies gathering 618 individuals. We show that VV is often abnormally tilted, but with of a mild magnitude. Because the VV investigates the vestibular graviception, these findings might mean that the vestibular information is frequently impaired in PD, but with a moderate asymmetry between left and right basal ganglia. However, directional concordances for VV and PD or PS tilts were not well analyzed, and when analyzed this concordance was weak. These findings suggest the coexistence of two mechanisms. Most abnormal VV tilts seem to be the consequence of the PS on the perception of the vertical, altered by trunk deformity or misorientation with respect to gravity. In some cases, VV tilts reflect an impaired vestibular graviception. In the future, studies will have to test several modalities of verticality perception, especially visual and postural, on large series, and scrutinize concordances in sides of tilts, body, and verticality perception.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.