Axial Postural Abnormalities in Parkinsonism: Gaps in Predictors, Pathophysiology, and Management

Axial postural abnormalities are common and disabling motor complications of Parkinson's disease (PD) and parkinsonism. They consist of abnormal trunk or neck postures in the upright position, often interfering with daily life activities.1-5 A high number of patients with PD may develop one or more axial postural abnormalities, which may begin as minor forms, almost universal in persons with parkinsonism (eg, minor flexed posture of the trunk and lower limbs), and progress to severe forms such as camptocormia, antecollis, or Pisa syndrome in over 20% of patients,1 although the definitions of severe axial postural abnormalities were purely based on subjective/expert opinion. Drug-induced parkinsonism or progressive supranuclear palsy may also present axial postural abnormalities, and in multiple system atrophy these symptoms are more frequent than in PD and occur earlier in the disease course.2, 6 Because axial postural abnormalities are associated with an increased risk of falling, pain, and diminished quality of life, their proper prevention and management is warranted.2, 4, 5, 7, 8 However, to date, recommendations for multidisciplinary management and prevention remain an unmet need.3 Diagnosis is typically based on a simple clinical examination, whereas treatments (eg, pharmacological, physical therapy, and surgical treatments) have been evaluated only in small single-center studies that failed to show consistent, long-lasting improvement.2, 5, 8-10 The absence of reliable protocols for the assessment, treatment, and prevention of axial postural abnormalities in parkinsonism is probably due to the largely obscure pathophysiology and uncertainty about prognostic factors needed to estimate beneficial responses to therapy (eg, duration of axial postural abnormalities).2, 5, 8, 11-13 The aim of this viewpoint is to critically analyze the literature on axial postural abnormalities, identify the current issues and gaps, and formulate proposals for current clinical management and future research exploration. Research on clinical predictors and risk factors for the onset of axial postural abnormalities in parkinsonism was mainly limited by the low quality of the methodology in the available literature (Supporting Information).2, 5, 8 Indeed, only a few studies were aimed primarily at the analysis of risk factors, and most studies were cross-sectional, lacking adequate longitudinal follow-up. In most cases the patient sample (usually PD) was heterogeneous for disease severity and stage. Clinical predictors were investigated for all axial postural abnormalities; however, the dependent variable in many studies was the presence of more than one axial postural abnormality, rather than a singular one. This approach renders it difficult to identify clinical predictors specific for each axial postural abnormality, which may vary between camptocormia, Pisa syndrome, and antecollis. Also, the cutoffs defining an axial postural abnormality differed across studies and were not stated in many cases. This generated possible bias in the prevalence rates due to differences in the clinical characteristics of patient samples. The identification of clinical predictors in patient samples presenting different degrees of trunk/neck flexion is challenging and limits interpretation of the study results. For example, patients with greater trunk flexion may also have more motor complications (ie, motor fluctuations) that make it difficult to identify the clinical predictors for mild axial forms (ie, anterior trunk flexion),3 which are of major clinical interest for prevention strategies. Finally, the variability in tools and reference bone points used for quantifying the degree of axial postural abnormalities renders comparison of study results arduous. The pathophysiology of axial postural abnormalities in PD is not clear, but several causes have been proposed,2, 5, 8 including central mechanisms (eg, imbalance in basal ganglia functioning leading to dystonia/rigidity, altered sensory–motor integration, higher cognitive function deficits) and/or peripheral mechanisms (ie, alteration in the musculoskeletal system).2, 5, 8 The main pathophysiological mechanisms considered in the literature are discussed here (Supporting Information). Dystonia, or a more general hyperactivity of paraspinal and nonparaspinal muscles, is the most common explanation for axial disability based on findings from electromyographic investigation.2, 5, 8, 9, 14-20 The hypothesis that dystonia underlies axial postural abnormalities is supported by three main findings: (1) the existence of drug-induced Pisa syndrome or camptocormia, (2) the (rare) existence of sensory tricks to alleviate some postural abnormalities, and (3) evidence of misalignment improvement by botulinum toxin type-A or lidocaine administration in some patients.2, 5, 8 However, in patients with Pisa syndrome, electromyography findings have been inconsistent concerning the side (ipsilateral or contralateral) of hyperactivity of paraspinal and nonparaspinal muscles in relation to trunk bending and PD asymmetry.16, 18, 19 Some studies suggested two different electromyographical patterns: (1) hyperactivity of lumbar and thoracic paraspinals ipsilateral to the trunk-leaning side during quiet stance and voluntary contralateral trunk flexion18, 21 and (2) hyperactivity of nonparaspinal muscles (eg, external oblique muscles, rectus femoris, and iliopsoas) ipsilaterally to the leaning side during quite stance and during voluntary contralateral trunk flexion.17 The paraspinal muscles contralateral to the side of flexion may have a compensatory action to further limit trunk bending.17 Other studies reported tonic and persistent activity in the paraspinal thoracic muscle and/or abdominal oblique muscles ipsilateral to the bending side and less-affected PD side, along with reduced/absent activity of the same muscles on the opposite side.19 In patients with camptocormia, studies suggested that the bilateral abdominal internal and the external oblique muscles, together with the rectus abdominis muscles, may be responsible for camptocormia with thoracic fulcrum,14 whereas rectus abdominis and iliopsoas hyperactivity could be more frequently involved in camptocormia with lumbar fulcrum.14, 20 Another study found hyperactivity of paraspinal muscles (ie, iliocostalis lumborum, longissimus, and multifidus) and, to a lesser extent, the rectus femoris but not rectus abdominis.15 It is probable that some hyperactivities observed in camptocormia, like those of paraspinal muscles that are spine erectors, are compensatory. Patients with antecollis showed hypertrophy and active spasms or hyperactivity in anterior (sternocleidomastoid muscles) and posterior (semispinalis, splenius capitis, trapezius, and levator scapulae) neck muscles.2 Passive head-up maneuvers may be limited by marked bilateral contractions of sternocleidomastoid muscles.2 The paucity of studies and objective measures to identify and quantify dystonia complicates the interpretation of these findings. Needle electromyography (EMG) has limited use because it is invasive and can be challenging when applied in deep muscles. Also, interpretation of needle EMG is relatively subjective as it is technically difficult to perform and has not been standardized for investigating axial muscles, and quantitative normative measures for activation of these muscle groups are not available. Finally, when EMG is performed in those with long-standing symptoms, it remains difficult to reliably separate primary abnormalities (ie, those causing postural abnormalities) from secondary ones (ie, resulting from the chronic changes in posture, or related to compensatory mechanisms). No studies have specifically and systematically evaluated rigidity in patients with axial postural abnormalities. During clinical examination in the supine position, patients demonstrate almost complete resolution of trunk bending, whereas when sitting or standing the paravertebral muscles may appear to have a wooden consistency.2, 15 The association between axial postural abnormalities and a more severe motor impairment and a rigid-akinetic phenotype makes rigidity a process worthy of investigation, albeit difficult to measure it empirically. Patients with PD have difficulty in properly controlling their postural stabilization and orientation (and also their interaction) based on sensory information from muscles and joints.2 PD patients with Pisa syndrome or camptocormia find it more difficult to achieve good postural alignment with gravity and display greater velocity of body sway than patients without axial postural abnormalities.22, 23 Patients with camptocormia with a lower fulcrum may have more severe gait and postural control deficits than patients with a thoracic fulcrum or patients without camptocormia.23 However, no studies have so far investigated the relationship between antecollis and postural instability. Postural control is a complex system involving the integration of different sensory inputs, including vestibular, visual, and proprioceptive. Some findings support the notion that proprioceptive function/kinesthesia is abnormal in PD.22, 23 Evidence was initially restricted to motor control of the limbs,24 but more recent studies22, 23 showed that proprioceptive defects might also affect axial motor control not only in the yaw plane (turning about the vertical axis)24 but also in the sagittal and coronal planes. Further studies are needed to understand the contribution of defective proprioceptive and kinesthesia in the pathophysiology of axial postural abnormalities.2, 24 Moreover, postural control depends on vestibular information. PD patients with Pisa syndrome may present with a central, peripheral, or mixed vestibular dysfunction2 along with a deviation in the subjective visual vertical,25 which was estimated to occur in about 67% of PD patients.26 A recent case–control study found worse performance on visuospatial ability tasks by PD patients with Pisa syndrome compared to PD controls without axial postural abnormalities.27 Visuospatial deficits are associated with dysfunctional subjective visual vertical, possibly contributing to altered perception of the trunk in space and the development of Pisa syndrome but not camptocormia.27 Whether visual spatial deficits and subjective visual vertical are defective and might also contribute to the development of antecollis has not been systematically investigated yet. Moreover, postural control demands complex motor–cognitive interactions that rely on high attentional resources and conserved executive functions. Some studies reported lower scores on tasks exploring executive functions (ie, divided attention, inhibitory control, and delayed free recall in verbal learning), perceptual visuospatial functions, and language for patients with Pisa syndrome compared to patients with PD without axial postural abnormalities.2, 5, 8, 28 Nevertheless, the association between kinesthetic/proprioceptive and cognitive deficits with axial postural abnormalities remains controversial. Without further details from pathophysiology studies, and also without intervention studies, three possibilities remain: (1) these deficits are a mere association (ie, they merely coincide with those with more severe disease), (2) these dysfunctions are not causal but rather a secondary manifestation (ie, they result as a consequence of the axial postural abnormalities), or (3) they are actually a contributing factor. The bulk of evidence from EMG (eg, fibrillation potentials, small polyphasic motor unit potentials, and weakness),5, 8, 14, 15, 17, 19, 29-34 muscle biopsy (eg, abnormal histology),5, 8, 29-32, 35 and muscle imaging (eg, fatty infiltration of muscles and muscle atrophy)5, 8, 9, 14, 17, 19, 29-31, 36 suggests myopathy as a contributing factor to the development of axial postural abnormalities. The cause of myopathy remains unclear, but ongoing research marks it as a secondary myopathic process, as a consequence of a chronically maintained abnormal posture.34 These phenomena probably reflect a continuum from early myopathic changes (ie, edema and partial swelling, contrast enhancement without degenerative muscle changes) in the early stages of camptocormia to progression to muscle atrophy and fatty degeneration in the end stage as shown by muscle magnetic resonance imaging (MRI) and biopsy. Although the exact cause of these muscular changes is unclear, they may arise due to muscle disuse or denervation secondary to chronic primary trunk misalignment. Camptocormia might be a consequence of overusing paraspinal muscles due to rigidity in patients with PD or a pronounced proprioceptive dysregulation in the context of underlying disease. Although a primary myopathy has been suggested to explain Pisa syndrome,19 no pathological studies have proven this hypothesis. Therefore, the observed muscular changes are probably caused by secondary mechanisms. The atrophy of paraspinal muscles ipsilateral to the bending side may be due to muscle disuse (excluded because of hyperactivity of nonparaspinal muscles) and cannot contribute to the development of Pisa syndrome; contralateral muscle atrophy instead might be secondary to the stress of prolonged stretching, as a consequence of muscle weakness, theoretically leading to Pisa syndrome.17 The mechanical efficiency of muscular activation of trunk extension (especially at the thoracic level) was markedly reduced in PD patients with camptocormia compared to those with PD without postural issues and to healthy controls, which suggests impairment of neuromuscular recruitment.34 The EMG of the paraspinal muscles is technically difficult, but this disadvantage is relative, as MRI and muscle biopsy can show the presence of myopathy. Although the finding of a monomorphic myopathy in the camptocormia of PD patients can be considered corroborated,35 the clarification of its cause remains an important task. Currently, the recording of myopathy with its subdivision into acute and chronic forms of muscle changes provides clues for therapeutic options.12 Postural improvement after high-frequency deep brain stimulation (DBS) of the bilateral subthalamic nuclei37 supports the notion that axial postural abnormalities may result from dysfunction of the basal ganglia. There is limited evidence for the presence of drug-induced axial postural abnormalities in persons with parkinsonism,2, 5, 8 with a dose reduction (or even complete withdrawal) or switching to another dopaminergic medication leading to improvement in posture.2, 8, 38 Finally, patients with axial postural abnormalities may present with a concomitant clinical history of back surgery (eg, laminectomy), degenerative spinal conditions, or trauma, which may have mechanical effects on bones or soft tissue (eg, muscles or fascia) and predisposing to their development.39 In clinical practice, the first step is to objectively measure the degree of trunk and/or neck flexion preferably while the patient is wearing only underwear (Fig. 1) during standing and walking.40 We suggest to investigate the possible presence of mild forms of axial postural abnormalities at each visit and inform and educate patients and caregivers on these symptoms; when actual axial postural abnormalities have been identified, clinicians could consider pharmacological and nonpharmacological interventions for their treatment. Figure 2 shows a practical road map as guidance for their clinical management. Four major issues may have limited the progress of research in this field: (1) different nosology and cutoffs to define axial postural abnormalities, (2) lack of common and objective assessment procedures, (3) heterogeneity of study design, and (4) heterogeneity of patient samples. Future research should focus on their clinical evaluation in terms of early detection and quantification using consensus-based criteria.3 Prospective, observational, long-term studies with large samples are immediately needed to explore the potential risk (causal) factors and pathophysiology involved in their development. From a therapeutic point of view, we call for well-controlled clinical trials: randomized trials using standard criteria for inclusion and endpoint measurement3 are needed to compare the efficacy of different pharmacological and nonpharmacological interventions and their combined effect (ie, botulinum toxin injections + physiotherapy). Such controlled studies could also focus on the merits of dopaminergic therapy withdrawal when these are suspected to have played a role in causing axial postural abnormalities. Moreover, we suggest researchers focus on a single postural abnormality per study, because risk factors, pathophysiology, and response to treatments may vary between camptocormia, Pisa syndrome, and antecollis. Finally, prevention programs using common diagnostic criteria3 should be investigated to study how the development of axial postural abnormalities can be avoided. Educate patients and their families to monitor possible improvement or worsening of posture. Adjustment of pharmacological therapy: remove the prescribed dopaminergic therapy potentially associated with the subacute onset of axial postural abnormality. The removal of therapy needs to be done very cautiously as this may also lead to a worsening of the underlying parkinsonism.8 Other expert opinion–based options include the following: botulinum toxin injection of hyperactive muscles should be cautiously considered for each patient based on EMG findings, in carefully selected patients, under the care of experts; avoiding injections of compensatory paraspinal and nonparaspinal muscles; prompt physiotherapy intervention to strengthen weak/compensatory trunk/neck muscles and improve balance and mobility; and prescription of walking aids (ie, high-frame walker) if needed. When a patient is suitable for DBS aimed at managing motor symptoms, camptocormia, and Pisa syndrome could also improve after surgery. Orthopedic surgery should be considered in very selected cases: patients should be properly counseled regarding the increased risk of operative complications and closely followed for incipient failure. We thank Rossella Targa (librarian) for her assistance in developing the literature search strategy. Open Access Funding provided by Universita degli Studi di Verona within the CRUI-CARE Agreement. (1) Writing of the first draft: C.G., M.T., C.A.A. (2) Data collection: C.G., C.A.A., J.N., C.A., D.G., M.L.D., D.S., Y.L., M.A.-W., M.S., R.W., V.T.L., G.I., and SC. (3) Review and critique: all authors. R.B. receives salary from Chulalongkorn University and stipend from the Royal Society of Thailand, has received consultancy and/or honoraria/lecture fees from Abbott, Boehringer-Ingelheim, Britannia, Ipsen, Novartis, Teva-Lundbeck, Takeda, and Otsuka pharmaceuticals; he has received research funding from the Newton Fund, the UK Government, Thailand Science and Research Innovation Bureau, Thailand Research Fund, Crown Property Bureau, Chulalongkorn University, and the National Science and Technology Development Agency; he holds patents for laser-guided walking stick, portable tremor device, nocturnal monitoring, and electronic Parkinson's disease symptom diary as well as copyright on Parkinson's mascot, dopamine lyrics and teaching video clips for common nocturnal and gastrointestinal symptoms for Parkinson's disease. A.F. reports the following: Consultancies from Abbvie, Medtronic, Boston Scientific, Sunovion, Ipsen, Merz; Advisory Boards of Abbvie, Boston Scientific, Ceregate, Inbrain, Ipsen, Medtronic, Jazz, Biogen/Sage; Honoraria from Abbvie, Medtronic, Boston Scientific, Sunovion, UCB, Ipsen; grants from University of Toronto, Weston foundation, Abbvie, Medtronic, Boston Scientific, Michael J Fox Foundation, European Union, MSA coalition, Dystonia Medical Research Foundation. L.L. reports the following: Consultancies and Honoraria from Bial, Zambon, Abbvie and Grants from University of Turin, Zambon, Abbvie. N.M. reports the following: dvisory Boards for Angelini, GW Pharmaceuticals and Grants from UCB Pharma, Desitin, Eisai, LivaNova, Angelini, GW Pharmaceuticals. M.M. reports the following: Consultant: St Jude/ Abbott, Honoraria: Glaxo Research grants: Glaxo, Allergan, TEVA. Royalties: Springer, Random House, Cambridge University press, Humana Press. Editor honorarium Wiley & Son. Movement Disorders Society.receives salary from Chulalongkorn University and stipend from the Royal Society of Thailand, has received consultancy and/or honoraria/lecture fees from Abbott, Boehringer-Ingelheim, Britannia, Ipsen, Novartis, Teva-Lundbeck, Takeda, and Otsuka pharmaceuticals; he has received research funding from the Newton Fund, the UK Government, Thailand Science and Research Innovation Bureau, Thailand Research Fund, Crown Property Bureau, Chulalongkorn University, and the National Science and Technology Development Agency; he holds patents for laser-guided walking stick, portable tremor device, nocturnal monitoring, and electronic Parkinson's disease symptom diary as well as copyright on Parkinson's mascot, dopamine lyrics and teaching video clips for common nocturnal and gastrointestinal symptoms for Parkinson's disease. A.F. reports the following: Consultancies from Abbvie, Medtronic, Boston Scientific, Sunovion, Ipsen, Merz; Advisory Boards of Abbvie, Boston Scientific, Ceregate, Inbrain, Ipsen, Medtronic, Jazz, Biogen/Sage; Honoraria from Abbvie, Medtronic, Boston Scientific, Sunovion, UCB, Ipsen; grants from University of Toronto, Weston foundation, Abbvie, Medtronic, Boston Scientific, Michael J Fox Foundation, European Union, MSA coalition, Dystonia Medical Research Foundation. L.L. reports the following: Consultancies and Honoraria from Bial, Zambon, Abbvie and Grants from University of Turin, Zambon, Abbvie. N.M. reports the following: dvisory Boards for Angelini, GW Pharmaceuticals and Grants from UCB Pharma, Desitin, Eisai, LivaNova, Angelini, GW Pharmaceuticals. M.M. reports the following: Consultant: St Jude/ Abbott, Honoraria: Glaxo Research grants: Glaxo, Allergan, TEVA. Royalties: Springer, Random House, Cambridge University press, Humana Press. Editor honorarium Wiley & Son. Movement Disorders Society. Data sharing not applicable to this article as no datasets were generated or analysed during the current study Data S1. Supporting Information Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.