Evolving Concepts on Bradykinesia

Matteo Bologna; Giulia Paparella; Alfonso Fasano; Mark Hallett; Alfredo Berardelli


Brain. 2020;143(3):727-750. 

In This Article

Bradykinesia Pathophysiology

The Role of Basal Ganglia

According to the classical basal ganglia model the main input area of basal ganglia is the striatum, which receives afferents from many cortical areas and from the intralaminar nuclei of the thalamus (Albin et al., 1989; DeLong, 1990). The major output regions of the basal ganglia are the globus pallidus pars interna (GPi) and the substantia nigra pars reticulata (SNr), which project to the thalamus modulating the activity of cortical regions and to the brainstem (Albin et al., 1989; DeLong, 1990; Obeso et al., 2000). The so called 'direct pathway' inhibits the GPi/SNr, and thereby facilitates the motor centres targeted by these nuclei (Albin et al., 1989; DeLong, 1990; Benarroch, 2016; Grillner and Robertson, 2016). The so-called 'indirect pathway' includes projections via the inhibitory globus pallidus pars externa (GPe) and the excitatory STN, targeting the output nuclei of the basal ganglia (GPi/SNr). The net effect of the indirect pathway is to inhibit the motor centres that are innervated by these nuclei. Regardless, studies with optogenetic techniques in mice showed a concurrent activation of striatum from both pathways in one hemisphere during the initiation of specific movements, (Cui et al., 2013). A third hyperdirect cortico-subthalamic pathway is considered a inhibitory network implicated in outcome optimization (Nambu et al., 2002; Aron et al., 2016). As recently demonstrated, the hyperdirect and indirect pathways both projecting on the STN are differentially involved in cognitive aspects of motor preparation and control of motor performance (Neumann et al., 2018). Data from rats have identified a central role of STN in generating bradykinesia, demonstrating that the delivery of negative constant current into STN dramatically ameliorates locomotor deficits in parkinsonian rats, while delivering of positive constant currents to STN induces Parkinson's disease-like locomotor deficits in normal rats (Tai et al., 2012). Concerning the role of dopamine in motor control, experimental studies have shown that dopaminergic neurons are transiently and rapidly activated before spontaneous movement (Howe and Dombeck, 2016) with transient dopamine release triggering movement onset. Dopaminergic activity is likely to be critical for the modulation of movement onset and vigor of future movements (da Silva et al., 2018; Yttri and Dudman 2018). Dopamine secretion is rapid, transient and has a high initial release rather than a sustained secretion (Liu et al., 2018), thus explaining why chronic administration of levodopa does not restore specific motor functions, such fine motor tasks.

Bradykinesia in patients with Parkinson's disease is classically associated with a significant (around 50–60%) dopamine striatal depletion (Fearnley and Lees, 1990; Ehringer and Hornykiewicz, 1998; Kalia and Lang, 2015; Liu et al., 2018), which mainly involves the vulnerable ventrolateral cell groups of the substantia nigra (nigrostriatal pathway). Loss of dopamine reduces the activity of D1 receptor direct pathway activating the striatum and increases the D2 receptor indirect pathway inhibiting the striatum (Wichmann and DeLong, 1996). While direct pathway neurons reduce their firing rate, those in the indirect pathway increase their activity. As a result, the firing in GPe neurons is reduced, which leads to disinhibition of STN neuronal activity changes and, subsequently, excessive excitation of STN targets (GPi and SNr) (Pan and Walters, 1988; Bergman et al., 1994; Hutchison et al., 1994; Sterio et al., 1994; Hassani et al., 1996; Wichmann et al., 1999; Mallet et al., 2006; Kita and Kita, 2011). The increased activity of GPi and SNr neurons is further reinforced by the lack of inhibition from direct pathway neurons. This leads to excessive inhibition of thalamo-cortical and brainstem motor systems, thus interfering with both preparation and execution phases of the voluntary movement (Wichmann and DeLong, 1996). Dopaminergic treatment produces its effects on converging basal ganglia pathways. Namely, it has been shown that dopamine increases the direct pathway medium spiny neurons activity, critically contributing to D1 agonism's motor stimulation in parkinsonian animals (Sagot et al., 2018). The relative hyperactivity of STN and GPi has been advocated to explain the symptomatic effect of ablations or DBS of these targets (functional inhibition hypothesis) (Udupa and Chen, 2015).

The so-called 'firing model' of basal ganglia bears several limitations (Bar-Gad et al., 2001). In fact, besides changes in firing rate, other complex electrophysiological phenomena occur in basal ganglia (Table 4). These include enhanced bursting, exaggerated oscillatory activity particularly in the beta frequency and coherent activity patterns between basal ganglia nuclei (Nini et al., 1995; Raz et al., 1996; Zirh et al., 1998; Levy et al., 2000, 2002a, b; Magnin et al., 2000; Raz et al., 2000; Vila et al., 2000; Brown et al., 2001; Soares et al., 2004; Sharott et al., 2005; Wichmann and Soares, 2006; Hammond et al., 2007; Mallet et al., 2008; Steigerwald et al., 2008; Chan et al., 2011; Tai et al., 2012; Ellens and Leventhal, 2013; Devergnas et al., 2014; Kammermeier et al., 2016; Sharott et al., 2017; Rodriguez-Sabate et al., 2019). As evidenced by DBS studies basal ganglia beta activity (13–35 Hz) is increased in Parkinson's disease (Brown et al., 2001; Brown, 2003) and the amplitude of this activity correlates with motor impairment. The excessive beta activity occurs more in neurons in the indirect pathway (Sharott et al., 2017), and, as this pathway is inhibitory, there is some understanding as to why excessive activity might lead to bradykinesia. Pathological beta activity is characterized by prolonged bursts of activity with an excessive synchronization within bilateral basal ganglia circuits (Tinkhauser et al., 2017). The abnormal synchronization is also seen between hemispheres and possibly correlates with specific signs of Parkinson's disease, such as freezing of gait; it has been argued that the longer the beta burst within the STN, the higher the chance that this synchronizes with the burst of the opposite hemisphere (Tinkhauser et al., 2017). Abnormal activity patterns correlate with compromised motor performance, including various bradykinesia features (Kühn et al., 2009; Jenkinson and Brown, 2011; Oswal et al., 2013; Little and Brown, 2014). STN abnormal activity directly correlates with the severity of hand bradykinesia in Parkinson's disease (Little et al., 2012; Tan et al., 2013). Prolonged beta bursts in the STN correlates with the slowness of movement or reaction time prolongation in Parkinson's disease (Tinkhauser et al., 2017; Torrecillos et al., 2018). Beta bursts and beta power suppression during repeated movements progressively reduce over time and parallel progressive decrement in the frequency, velocity and amplitude of movements (Steiner et al., 2017; Lofredi et al., 2019). Furthermore, the lack of gamma recruitment alters the signal scaling across different movement velocities (Lofredi et al., 2018). Dopaminergic medications, or DBS, improve motor function by modifying basal ganglia circuits' dynamics from persistent synchronized activity to a more dynamic activity pattern (Priori et al., 2004; Hahn et al., 2008; Vitek et al., 2012; Cleary et al., 2013; Müller and Robinson. 2018), which correlates with clinical improvements of bradykinesia (Ray et al., 2008; Kühn et al., 2009; Tai et al., 2012; Trager et al., 2016). The effect of dopaminergic therapy on oscillatory basal ganglia activity has been studied in detail by Tinkhauser et al. (2017) demonstrating that in Parkinson's disease patients with externalized STN DBS electrodes levodopa administration shortens and decreases the amplitude of the prolonged bursts of beta activity; levodopa also makes the bursts of pathological activity less frequent. The reduction of burst activity induced by levodopa correlated with clinical improvement. Similar findings have been obtained with DBS, particularly when stimulating on-demand with an 'adaptive' DBS (aDBS, also known as closed-loop DBS) (Little et al., 2013). The pathophysiological role of beta oscillation is confirmed, although not by all studies, by the observation that 10–20 Hz DBS can worsen bradykinesia, likely pacing or increasing these low-frequency oscillation (Timmermann et al., 2004; Su et al., 2018). On the other hand, studies have confirmed that levodopa induces a reduction of beta activity and synchronization within higher frequencies (70 Hz) in basal ganglia (Brown et al., 2001; Williams et al., 2002; Foffani et al., 2003). Accordingly, DBS using frequencies within the gamma band (60–80 Hz) has been found to improve bradykinesia (reviewed in di Biase and Fasano, 2016). A recent paper found that movement regularity improved during 60-Hz DBS but not during 140-Hz DBS, while both frequencies were able to improve movement velocity, thus suggesting different but overlapping mechanisms explaining the benefit of these DBS frequencies (Blumenfeld et al., 2017). In particular, 60-Hz DBS amplified oscillations in a low beta-band (11–15 Hz) and attenuated oscillations in a high beta-band (19–27 Hz) whereas 140-Hz DBS attenuated oscillations across the beta-band frequency range (Blumenfeld et al., 2017). Again, these findings reinforce the hypothesis that certain low frequency oscillations (i.e. short bursts, lower end of the beta band spectrum) are indeed beneficial to the physiological function of the basal ganglia whereas others (i.e. long bursts, higher end of the beta band spectrum) are pathological underpinnings of bradykinesia. Beyond the aforementioned local effects, the beneficial effect of DBS relies on its property to decouple the excessive synchronization between cortico-subcortical regions within the same hemisphere (de Hemptinne et al., 2015) and between hemispheres (Little et al., 2016).

Finally, plasticity abnormalities at basal ganglia level have been recently suggested as an additional mechanism of bradykinesia (Yttri and Dudman, 2018). In mice, closed-loop stimulation of the basal ganglia induced changes in movement velocity, outlasting the end of stimulation and abolished by dopamine antagonists (Yttri and Dudman, 2016). In patients with Parkinson's disease, subthalamic and pallidal stimulation induced a long-lasting increase of the inhibitory electrophysiological phenomena recorded in GPi and SNr (Milosevic et al., 2019). Notably, lower levels of plasticity were associated with higher severity of motor symptoms (Milosevic et al., 2019). These data support the role of basal ganglia plasticity abnormalities in generating bradykinesia and suggest that plastic changes are influenced by dopamine depletion (Yttri and Dudman, 2018).

In summary, earlier conceptualizations emphasized their modularity property of basal ganglia and the changes in the firing rate of the various nuclei resulting from dopaminergic loss. More recent investigations demonstrate complex electrophysiological phenomena, including abnormal bursting, oscillatory and plasticity changes, and their correlation with various bradykinesia features. A better understanding of these mechanisms is required to interpret the mechanisms of action of dopaminergic medications and DBS.

The Role of Primary and Non-primary Motor Cortical Areas

Over time is becoming clear that besides cell death in the substantia nigra and loss of dopamine in the striatum, changes occur also in M1, playing an important role in generating voluntary movement abnormalities. First evidence came from animal studies. Dopamine depletion in parkinsonian models increases burst-firing in pyramidal tract type neurons of M1 (Goldberg et al., 2002; Pasquereau and Turner, 2011), decreases the magnitude and the temporal pattern of movement-related M1 activity (Watts and Mandir 1992; Parr-Brownlie and Hyland 2005), and interferes with the specific encoding of movement parameters (Pasquereau et al., 2016). Dopamine depletion in parkinsonian animals also leads to plasticity changes in M1, as evidenced by dendritic spine reduction in the lamina 5b pyramidal tract type neurons. Levodopa treatment partially rescued the enhanced spine turnover and the aberrant spine dynamics (Guo et al., 2015). Thus, the dopamine system finely modulates structural plasticity of the layer V animal pyramidal neurons in M1.

Dysfunction in the M1 has long been thought to play a role in the generation of bradykinesia in patients with Parkinson's disease (Berardelli et al., 2001; Wu et al., 2011). It has been proposed that M1 may develop secondary changes due to the altered pattern of activity it receives from basal ganglia (Bateup et al., 2010; Kravitz et al., 2010; Cui et al., 2013). Subsequent studies demonstrated that intrinsic M1 abnormalities contribute to parkinsonian motor signs. Electrophysiological techniques have been able to probe the activity of M1 in patients with Parkinson's disease (for a review see Burciu and Vaillancourt, 2018). Several studies revealed an abnormal synchronization in the beta rhythm between the cortex and basal ganglia that may underlie bradykinesia (de Hemptinne et al., 2013). Levodopa attenuates the abnormal synchronization and induces a clinical improvement (Brown et al., 2001; Williams et al., 2002; Foffani et al., 2003). Magnetoencephalography data have shown that increased resting-state cortico-cortical functional connectivity in the 8–10 Hz alpha range characterizes the earliest stages of Parkinson's disease, and that, with disease progression, neighbouring frequency bands (beta and theta) become increasingly involved. These findings suggested that changes in functional intra-hemispheric and inter-hemispheric coupling over the course of Parkinson's disease may be linked to the topographical progression of pathology over the brain (Stoffers et al., 2008). Moreover, phase-amplitude coupling is consistently affected by DBS (de Hemptinne et al., 2015).

Transcranial magnetic stimulation (TMS) studies have provided useful information on M1 excitability and plasticity. Although some of the data are controversial overall the results indicate that Parkinson's disease patients have an increased M1 excitability, as tested with single and paired pulses stimulation (see references in Table 5). Similarly, the results overall suggest that Parkinson's disease patients have a reduced plasticity, as tested with repetitive TMS protocols (see references in Table 6). The pathophysiological roles of excitability and plasticity changes of M1 in generating bradykinesia are still debated. M1 magnetic stimulation after a go-signal has shortened reaction time, normalized the abnormal triphasic EMG pattern and increased the pre-movement cortical excitability (Pascual-Leone et al., 1994). Bologna et al. (2018) found that excitability and plasticity TMS measures of M1 correlated with specific objective kinematic measurements of finger tapping in Parkinson's disease. Namely, excitability abnormalities correlated with motor slowness, while plasticity alterations correlated with sequence effect (Bologna et al., 2018). Dopaminergic medications normalized the majority of M1 excitability and plasticity measures (Tables 5 and 6). However, the reported results are variable and there is no clear relationship between levodopa-induced changes in neurophysiological and movement measures, possibly indicating that TMS measures and movement kinematics have different sensitivity to dopaminergic tone (Monte-Silva et al., 2010; Espay et al., 2011; Suppa et al., 2017; Bologna et al., 2018). Interestingly, studies recording from chronically implanted electrodes over M1 have not consistently found specific hallmarks for bradykinesia while reporting that gamma rhythm is a biomarker for levodopa-induced dyskinesias (Swann et al., 2016). More recently a study using a similar (but temporary) experimental set-up, found that pallidotomy improves bradykinesia by 'unleashing' gamma oscillations of M1 (de Hemptinne et al., 2019). An alternative possibility is that some bradykinesia features are not strictly dependent on dopaminergic loss. A recent neurocomputational study suggests that bradykinesia may result from the concurrent effects of low dopaminergic levels and dysfunctional cortico-striatal plasticity. The results show that training under levodopa administration improves bradykinesia. Conversely, training in unmedicated patients, has a detrimental effect on bradykinesia possibly due to dysfunctional corticostriatal plasticity induction (Ursino and Baston, 2018). Considering the role of M1 in generating bradykinesia, epidural or extradural motor cortex stimulation has been proposed as alternative to DBS in the treatment of Parkinson's disease. However, the results are controversial. Extradural motor cortex stimulation has been found to be beneficial in specific features of bradykinesia (speech and gait disorders) in few studies involving Parkinson's disease (Cilia et al., 2007; Bentivoglio et al., 2012) and PSP patients (Piano et al., 2018). Other studies have, however, only reported a transient benefit (Fasano et al., 2008) or no benefit at all (Moro et al., 2011).

Non-invasive brain stimulation techniques can be also used for the study of altered connectivity between primary and non-primary motor areas (Hallett et al., 2017). For example, a conditioning single-pulse TMS can influence the effect of a test TMS given over M1 (Koch et al., 2007; Karabanov et al., 2013). Effects on M1 can be also tested after repetitive TMS (rTMS) of non-primary motor areas. RTMS delivered over non-primary motor areas have shown that dorsal premotor cortex-to-M1 functional connectivity is abnormal in patients with Parkinson's disease and it is promptly normalized by levodopa administration (Buhmann et al., 2004; Mir et al., 2005; Suppa et al., 2010). Premotor rTMS, however, has no effect on clinical parkinsonian symptoms or motor performance of ballistic wrist movements, regardless of whether patients were ON or OFF dopaminergic medications (Mir et al., 2005).

Abnormal M1 excitability and plasticity have been investigated less extensively in atypical parkinsonisms. PSP and MSA patients showed an increase M1 excitability and a reduction in short intracortical inhibition (Marchese et al., 2000; Kühn et al., 2004). CBS patients exhibit a pattern similar to Parkinson's disease, with a reduced short intracortical inhibition and a normal intracortical facilitation (Frasson et al., 2003; Murgai and Jog, 2018). MSA patients showed a reduction in long term potentiation (LTP)-like effects after paired associative stimulation, not restored after dopaminergic therapy (Kawashima et al., 2013).

The role of primary and non-primary motor cortical areas abnormalities in Parkinson's disease have also been investigated by neuroimaging, showing either grey matter atrophy (Jankovic et al., 1990; Rosenberg-Katz et al., 2013; Shao et al., 2014) or cortical thinning (Wilson et al., 2019). Early studies also showed a reduced activation of the putamen and the medial frontal cortex (Playford et al., 1992). Variable functional M1 changes, ranging from movement related hypo- (Rascol et al., 1992; Catalan et al., 1999; Turner et al., 2003; Tessa et al., 2010; Hanakawa et al., 2017) to hyperactivation (Sabatini et al., 2000; Thobois et al., 2000; Haslinger et al., 2001) have been also described. The two different patterns of functional M1 changes may be due to several technical differences in the experiments performed and also in the clinical features of the patients studied. Overall functional neuroimaging studies performed, controlling several confounding features that might had affected the results, showed a consistent pattern of reduced activation of M1 (Burciu and Vaillancourt, 2018). Likewise, neuroimaging studies consistently revealed abnormalities of non-primary motor cortical areas regions such as the hypoactivation of the supplementary motor area and hyperactivation of the lateral premotor cortex (Playford et al., 1992; Rascol et al., 1992; Jahanshahi et al., 1995; Samuel et al., 1997; Catalan et al., 1999; Sabatini et al., 2000; Haslinger et al., 2001; Rowe et al., 2002; Buhmann et al., 2003; Wu and Hallett, 2005; Eckert et al., 2006; Ukmar et al., 2006; Yu et al., 2007; Tessa et al., 2010; Wu et al., 2010, 2011; Zhang et al., 2015; Criaud et al., 2016; Karunanayaka et al., 2016; Hanakawa et al., 2017; Hu et al., 2017). Neuroimaging is also a suitable tool for investigating altered connectivity between prefrontal cortex or premotor areas and M1 as well as basal ganglia-cortical interactions (Wu et al., 2011, 2016; Esposito et al., 2013; Spay et al., 2019). Studies investigating network analysis showed an abnormal connectivity between motor regions (Rowe et al., 2002) and abnormal connectivity between the cortico-striatal circuit (Helmich et al., 2010). Levodopa administration (Haslinger et al., 2001; Buhmann et al., 2003; Poston and Eidelberg, 2012; Esposito et al., 2013), as well as STN stimulation (Grafton et al., 2006; Akram et al., 2017, Horn et al., 2017) enhanced the sensorimotor network functional connectivity in the supplementary motor area improving the activation responses. Resting state functional MRI studies also showed that there is an increased interaction between cerebral networks with a loss of segregation between functional networks (Kim et al., 2017). Although patients with loss of segregation had more severe motor symptoms it is unclear whether the loss of segregation correlates with any specific symptom including bradykinesia.

In summary, the demonstration that intrinsic abnormalities of the M1 play an essential role in generating bradykinesia is one of the most important pathophysiological advances in the last decade. The most compelling evidence came from animal studies. Electrophysiological techniques and neuroimaging have provided in vivo confirmation of the M1 abnormalities in human Parkinson's disease patients. Some of these studies also found a correlation between M1 abnormalities and with various bradykinesia features. The role of non-primary motor cortical area abnormalities in Parkinson's disease, although less compelling, have also been suggested.

The Role of Cerebellum

The hypothesis that cerebellum contributes to bradykinesia in Parkinson's disease is based on the anatomical evidence of reciprocal connections between the cerebellum and basal ganglia (Ichinohe et al., 2000; Bostan and Strick, 2018). Namely, the cerebellum receives a disynaptic glutamatergic projection from the STN. The hypothesis that abnormal signals from the STN in Parkinson's disease results in abnormal cerebellar activation is supported by the observation that STN high-frequency stimulation through DBS electrodes in rats leads to a reduction in the activity of cerebellar Purkinje cells and, as a consequence, disinhibition of cerebellar nuclei as also evidenced by increased cFOS expression in the cerebellar nuclei (Moers-Hornikx et al., 2011). In keeping with these notions, a recent connectivity study in Parkinson's disease patients found that bradykinesia is improved when DBS affects the ascending ipsilateral cerebellar-thalamo-cortical pathway (Strotzer et al., 2019). Moreover, it has been reported that dopaminergic loss is associated with loss of Nissl-stained Purkinje cells, reported in parkinsonian animals (Rolland et al., 2007; Heman et al., 2012).

Neurophysiological studies investigating cerebellar function demonstrate that Parkinson's disease patients, with prominent bradykinetic-rigid symptoms, have deficient short-latency and long-lasting cerebellar-thalamo-cortical inhibitory interactions that cannot be restored by dopaminergic medication (Carrillo et al., 2013). Functional or metabolic neuroimaging studies demonstrated abnormal cerebellar activation in patients with Parkinson's disease while performing various upper limb movements (Rascol et al., 1997; Catalan et al., 1999; Wu and Hallett, 2005; Yu et al., 2007; Wu et al., 2010), including finger movements (Rascol et al., 1997; Cerasa et al., 2006; Yu et al., 2007), motor timing tasks (Jahanshahi et al., 2010), complex sequential movements (Catalan et al., 1999), bimanual two-hand coordinated tasks (Wu et al., 2010) or simultaneous movements (Wu and Hallett, 2008). In addition to abnormal activity in the cerebellum, a series of neuroimaging studies showed an abnormal connectivity pattern of the cerebellum in Parkinson's disease (Wu et al., 2009, 2011, 2016; Jahanshahi et al., 2010). The relationship between cerebellar hyperactivation or abnormal cerebellar connectivity and bradykinesia remains unclear. Interestingly, in patients with progressive micrographia, a manifestation of the sequence effect, a decreased connectivity between the posterior putamen and cerebellum was found (Wu et al., 2016), but this abnormality did not correlate with the severity of progressive micrographia, and normalization of connectivity with levodopa did not improve progressive micrographia. A disconnection among the cerebellum, the pre-supplementary motor area and the rostral cingulate motor area, in addition to the dysfunction of the basal ganglia motor circuit, was demonstrated in Parkinson's disease presenting progressive micrographia. i.e. the sequence effect (Wu et al., 2016). Also relevant in this regard, a recent neuroimaging study of freezing of gait cases induced by discrete brain lesions found that a variety of different involved networks all localize to a dysfunction of the cerebellar locomotor centre (Fasano et al., 2017). The cerebellar involvement in Parkinson's disease is finally supported by the evidence of abnormal visuomotor learning in patients, restored through DBS (de Almeida et al., 2019).

In summary, cerebellar dysfunction is now undoubtedly considered one of the factors involved in the pathophysiology of bradykinesia. This is supported by the evidence of anatomical reciprocal connections between cerebellum and basal ganglia and by neurophysiological and neuroimaging studies in Parkinson's disease patients. A relationship between cerebellar involvement, micrographia, and some bradykinesia features of the upper limb is present. Cerebellar involvement in Parkinson's disease supports the view of bradykinesia as a network disorder.

The Sensorimotor Function

Converging evidence indicates that defective integration of sensory information at various levels of a complex cortico-subcortical network, including the primary somatosensory cortex (S1) and basal ganglia, may be involved in the pathophysiology of bradykinesia. Several studies on Parkinson's disease patients have indicated abnormalities in sensory processing, including abnormalities of sensory discrimination, proprioceptive integration, kinaesthetic sense of joint displacement (Demirci et al., 1997; Seiss et al., 2003; Konczak et al., 2007; Wright et al., 2010; Patel et al., 2014) and mechanisms of sensorimotor integration (Georgiev et al., 2016). Demirci et al. (1997) also provided evidence of a mismatch between kinaesthetic and visual perception in Parkinson's disease, linking the sensory disturbances to a scaling abnormality. Recent information on the altered sensorimotor integration in Parkinson's disease has been provided by studies assessing the somatosensory temporal discrimination threshold (STDT), (Lacruz et al., 1991; Artieda et al., 1992; Rocchi et al., 2016; Conte et al., 2017a; Lee et al., 2017). The STDT abnormalities in Parkinson's disease reflect disease severity and duration (Conte et al., 2016, 2018). Some studies have reported that higher values of STDT correlate with higher UPDRS part III score (Artieda et al., 1992) while other studies found no correlations (Conte et al., 2010; Lee et al., 2010, 2017; Lyoo et al., 2012; Rocchi et al., 2013). Quantitative measurement of bradykinesia throughout inertial sensors showed a correlation between higher STDT values and increased variability in amplitude and speed, which may reflect the prolonged temporal processing of tactile information and altered sensorimotor integration (Lee et al., 2017). Accordingly, recent findings from studies on movement-related changes of STDT in Parkinson's disease patients further support a link between an altered STDT and finger movement abnormalities in Parkinson's disease patients (Conte et al., 2017b). Levodopa treatment influences STDT values in Parkinson's disease (Artieda et al., 1992; Lee et al., 2005; Conte et al., 2010; Rocchi et al., 2013). In a dopamine transporter PET study, Lyoo et al. (2012) reported that increased STDT values correlated with ligand uptake in both the caudate and putamen. Finally, little is known about STDT and atypical parkinsonisms. One study found higher values of STDT in patients with MSA compared to healthy subjects (Lyoo et al., 2007) and to Parkinson's disease patients (Rocchi et al., 2013).

In summary, a defective integration of sensory information at the cortico-subcortical network is another essential component involved in the pathophysiology of bradykinesia. Evidence comes from the study of sensorimotor integration and STDT. Despite some of these studies reporting correlations between altered sensory processing and bradykinesia, this issue deserves further experimental confirmation.