Deep brain stimulation for Parkinson’s disease–related postural abnormalities: a systematic review and meta-analysis

Deep brain stimulation (DBS) has become a well-established treatment modality for Parkinson’s disease (PD), especially regarding motor fluctuations, dyskinesias, and tremor. Although postural abnormalities (i.e., Camptocormia [CC] and Pisa syndrome [Pisa]) are known to be a major symptom of PD as well, the influence of DBS on postural abnormalities is unclear. The objective of this study is to analyze the existing literature regarding DBS for PD-associated postural abnormalities in a systematic review and meta-analysis. In compliance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines, we conducted a systematic review and meta-analysis of 18 studies that reported the effect of DBS regarding postural abnormalities. After screening of 53 studies, a total of 98 patients (44 female, 53 males, 1 not reported; mean age: 62.3, range 30–83 years) with postural abnormalities (CC n = 98; Pisa n = 11) were analyzed from 18 included studies. Of those patients, 94.9% underwent STN-DBS and 5.1% had GPi as DBS target area. A positive outcome was reported for 67.8% with CC and 72.2% with Pisa. In the meta-analysis, younger age and lower pre-operative UPDRS-III (ON/OFF) were found as positive predictive factors for a positive effect of DBS. DBS might be a potentially effective treatment option for PD-associated postural abnormalities. However, the level of evidence is rather low, and definition of postoperative outcome is heterogenous between studies. Therefore larger, prospective trials are necessary to give a clear recommendation.


Introduction
Parkinson's disease (PD) affects 1% of the population over the age of 60, with increasing prevalence, and is therefore the second most common neurodegenerative disease worldwide [1]. Beside characteristic features of PD (bradykinesia, rigidity, and tremor), one of the most noticeable signs of PD patients is abnormality in their posture, with functional alterations of the spine [2]. The clinical phenotype of postural abnormalities is variable: while Camptocormia (CC) presents in the sagittal, Pisa syndrome (Pisa) is observed in the coronal plane [3]. CC is defined by an abnormal thoraco-lumbar spinal flexion which presents while standing and walking and is alleviated in a recumbent position. Most authors define CC by an arbitrary angle of at least 30-45° flexion of the thoraco-lumbar spine in the standing position [4,5]. Pisa is defined as a reversible lateral deviation of the spine (> 10°) with a corresponding tendency to lean to one side. It is not to be confused with scoliosis, in which lateral bending is caused by an S-shaped curvature and rotation of the spine. Like CC, Pisa occurs when standing and disappears in the recumbent position [6,7]. It is obvious that these postural abnormalities subsequently lead to back pain and degenerative alterations of the spine. With the increasing prevalence of PD and the associated degenerative spinal conditions, the demand for spinal surgery in patients with PD increases rapidly. However, the results of spinal surgery for degenerative spinal conditions in patients with PD are disappointing with failure rates of 25.8-100% [8][9][10][11][12]. As an alternative treatment strategy for the underlying pathomechanism, some authors have described the influence of deep brain stimulation on PD-related postural abnormalities [13][14][15][16][17][18][19][20][21][22][23][24][25][26][27][28][29][30]. We aim to present a systematical review of the current literature. Moreover, we conducted a meta-analysis to evaluate predictive factors for a successful outcome of DBS.

Search strategy
To collect fundamental data, the systematic review was done in accordance with the criteria outlined in the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2009 guidelines [31]. Institutional Review Board approval and/or patient consent were not required. Two reviewers (PS and YA) conducted a computerized search between January 2002 and January 2022 on the PubMed and Web of Science databases. The following algorithm was developed as search strategy: (deep brain stimulation [Title/Abstract]) AND (spine [Title/Abstract]) AND (Parkinson [Title/Abstract]), Filters: Humans, English for the PubMed search and ((TS = (deep brain stimulation)) AND TS = (spine)) AND TS = (parkinson) for the Web of Science search. The reference lists of the included studies were searched to obtain additional articles.

Study selection
Studies were included if they met the following criteria: (1) human patients, (2) English language, (3) peer-reviewed original articles with full-text available (reviews, systematic reviews, and meta-analyses were excluded), (4) reported postural abnormalities (i.e., PS and CC), (5) performance of Deep Brain Stimulation. Studies that did not meet all inclusion criteria were excluded.
To avoid extraction errors, two reviewers (PS and YA) independently extracted data from the eligible articles. Any discrepancies were discussed and resolved with a third reviewer (PV).

Statistical analysis
According to the reported primary outcomes of the included studies, a relative reduction of > 50% of the thoraco-lumbar angle (TLA) respectively and an absolute value of < 30° of the TLA post-DBS were defined as positive outcome (Table 1). Those parameters were dichotomized (i.e., > 50% reduction vs. < 50% reduction and < 30° vs > 30°) and stratified by each of the assessed variables via Student's t-test (with a significance level of p < 0.05) to identify possible associations with outcome (Table 2). For the meta-analysis, the continuous variables which demonstrate a significant difference for the outcome parameters were subjected. Heterogeneity of study outcomes between the included studies was calculated by Cochran's Q and I 2 statistics and consequently it was defined whether fixed or random effects models were appropriate. Inverse variance tests calculated mean difference and confidence intervals of 95%. The results were visualized by forest plot asymmetry.
Unpaired, two-tailored Student's t-tests (with a significance level of p < 0.05) were performed with the use of

Search results and study selection
After removal of the duplicates from the initial database, 53 articles underwent the first screening process of titles and abstracts. Subsequently, 19 potentially relevant articles were identified that underwent full-text review and screening against the inclusion criteria. Ten articles were excluded (review articles and meta-analyses n = 9; no report about postural abnormalities n = 1), resulting in a total of nine articles. From the reference list, nine additional articles which were not found in the initial algorithm-based search were included, resulting in 18 articles that were finally chosen to be included in the analysis. A RRISM flowchart of the study selection process is represented in Fig. 1.

Meta-analysis results
Three studies were included in the meta-analysis [21,22,29] and 15 studies were excluded due to low number of participants and/or missing demographic parameters. The preliminary analysis was analyzed according to the outcome parameters of the studies (improvement through DBS defined as absolute TLA < 30° or amelioration of the TLA > 50% post-DBS). Younger age was found as positive predictor for beneficial effect of DBS on both outcome parameters (mean age 59.7 ± 6.6 vs 67.1 ± 5.8 years, p < 0.0005 for absolute TLA < 30° and 61.4 ± 7.6 vs 66.2 ± 5.9 years, p < 0.05 for relative TLA reduction > 50%) ( Table 2). Inverse variance analysis revealed a difference of 9.1 (95% CI 5.3-13.0) years between patients with absolute TLA < 30° vs. TLA > 30° and of 4.4 (95% CI 0.5-9.3) years between patients with > 50% vs. < 50% relative TLA improvement ( Fig. 3A and B). Lower pre-operative UPDRS-III (ON/OFF) was found as another positive predictive factor (14.

Discussion
After 53 articles have been screened, 18 were included in the analysis, of those three underwent subsequent metaanalysis. The following main findings were detected: (1) Postural abnormalities associated with PD improved, at least partially, in 67.8% (CC) and 72.2% (Pisa) of patients following DBS. (2) Younger age was found as a positive predictive factor for a beneficial effect of DBS. (3) Lower pre-OP UPDRS-III (ON and OFF) was associated with better outcome following DBS. DBS has become a well-established treatment option for PD over the past decades. Especially patients with motor fluctuations, dyskinesias secondary to chronic levodopa and those with refractory and marked tremor benefit from DBS [33]. Although postural abnormalities are known to be a major symptom of PD [2], which in several cases give  rise to spinal deformities [34], there is a gap of knowledge regarding the influence of DBS on postural abnormalities associated with PD. We aimed to summarize the existing literature by performing a systematic review. We found very heterogeneous approaches to describe the influence of DBS on postural deformities: The definition of appropriate outcome parameter varied between the included studies. While some studies examined clinical parameters, other studies used photographs to evaluate the patients' TLA prior and after DBS. Even within this rough subdivision of outcome parameters, there were significant differences. Clinical parameters were rarely expressed by standardized scores, rather than by anecdotal reports, which precludes subjective comparisons. Regarding TLA, which was used as outcome parameter in 72.2% of the studies, some authors defined a positive effect of DBS by a relative improvement of the TLA > 50% [24,29], other authors describe a post-OP absolute TLA < 30° as positive effect [14,21], and the remaining authors reported a positive effect or non-effect, but without defining this precisely. Three studies reported aspects of the paraspinal muscles associated with postural abnormalities: Asahi et al. described a higher density of paraspinal muscle (measured by CT scans) in patients that improved through DBS [14]. This finding was confirmed by Sakai et al., who performed MRI scans of the lumbar spine and detected a larger crosssectional area of paraspinal muscle in patients with positive effect of DBS [21]. Schaebitz et al. found myopathy confined to the paraspinal muscles in a small case series of PD patients with CC [23].
Schulz-Schaffer et al. and Yamada et al. found that a longer duration of CC prior to DBS was associated with less improvement of the TLA [24,29]. This correlation was further analyzed in a meta-analysis by Chan et al. [35] who described a duration of CC < 2 years predictive for better outcomes. We did not confirm those results, since Schulz-Schaffer et al. did not reveal individual patient characteristics and therefore were not eligible four our meta-analysis.

Limitations
There are several limitations to this study. First, the evidence is mostly limited to case series and reports. To avoid reporting artificially too positive results from case reports, we performed an analysis in which patients from case reports were excluded. However, this did not fundamentally change the M male, F female, GPi globus pallidus internus, STN subthalamic nucleus

Conclusion
We systematically reviewed the existing literature regarding the effect of DBS on PD-associated postural abnormalities.
The results suggest that in certain cases, DBS is a potentially effective treatment option for affected patients. However, the level of evidence is low, since the data is primarily based on retrospective studies with small number of cases. Moreover, the definitions of postoperative outcomes are heterogenous between the existing studies, which impedes the comparison of the studies with one another. Controlled, prospective trials with statistical reliable numbers of patients are missing to give a clear recommendation whether DBS is a sufficient treatment of PD-associated postural abnormalities.
Author contribution Philipp Spindler MD: concept and design, acquisition, analysis, interpretation of data for the work; drafting the work; final approval of the version to be published; agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. Yasmin Alzoobi: acquisition and interpretation of data for the work; critical revision for important intellectual content of the work; final approval of the version to be published; agreement to be accountable for all aspects of the work in ensuring that questions related to Fig. 3 Meta-analysis of predictive parameters for positive outcome (i.e., absolute TLA < 30° or relative TLA improvement > 50%). A Mean (± SD) of patient age with respect to absolute TLA after DBS. B Mean (± SD) of patient age with respect to relative TLA reduction after DBS. C Pre-OP UPDRS (ON) with respect to absolute TLA after DBS. D Pre-OP UPDRS (OFF) with respect to absolute TLA after DBS the accuracy or integrity of any part of the work are appropriately investigated and resolved.
Andrea A. Kühn MD: critical revision for important intellectual content of the work; final approval of the version to be published; agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.
Katharina Faust MD: critical revision for important intellectual content of the work; final approval of the version to be published; agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.
Gerd-Helge Schneider MD: critical revision for important intellectual content of the work; final approval of the version to be published; agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.
Peter Vajkoczy MD: concept and design, interpretation of data for the work; critical revision for important intellectual content of the work; final approval of the version to be published; agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.
Funding Open Access funding enabled and organized by Projekt DEAL. We acknowledge support from the German Research Foundation (DFG) and the Open Access Publication Fund of Charité -Universitätsmedizin Berlin.
Data availability All data generated or analyzed during this study are included in this article. Further inquiries can be directed to the corresponding author.

Declarations
Ethics approval An ethics statement is not applicable because this study is based exclusively on published literature.

Consent to participate
A consent to participate statement is not applicable because this study is based exclusively on published literature.

Competing Interests
The authors declare no competing interests.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.