Gaia Giannicola

The effects of levodopa and ongoing deep brain stimulation on subthalamic beta oscillations in Parkinson's disease

Local field potentials (LFPs) recorded through electrodes implanted in the subthalamic nucleus (STN) for deep brain stimulation (DBS) in patients with Parkinsons disease (PD) show that oscillations in the beta frequency range (8-20 Hz) decrease after levodopa intake. Whether and how DBS influences the beta oscillations and whether levodopa- and DBS-induced changes interact remains unclear. We examined the combined effect of levodopa and DBS on subthalamic beta LFP oscillations, recorded in nine patients with PD under four experimental conditions: without levodopa with DBS turned off; without levodopa with DBS turned on; with levodopa with DBS turned on; and with levodopa with DBS turned off. The analysis of STN-LFP oscillations showed that whereas levodopa abolished beta STN oscillations in all the patients (p=0.026), DBS significantly decreased the beta oscillation only in five of the nine patients studied (p=0.043). Another difference was that whereas levodopa completely suppressed beta oscillations, DBS merely decreased them. When we combined levodopa and DBS, the levodopa-induced beta disruption prevailed and combining levodopa and DBS induced no significant additive effect (p=0.500). Our observations suggest that levodopa and DBS both modulate LFP beta oscillations.

Cerebellum and processing of negative facial emotions: Cerebellar transcranial DC stimulation specifically enhances the emotional recognition of facial anger and sadness

Some evidence suggests that the cerebellum participates in the complex network processing emotional facial expression. To evaluate the role of the cerebellum in recognising facial expressions we delivered transcranial direct current stimulation (tDCS) over the cerebellum and prefrontal cortex. A facial emotion recognition task was administered to 21 healthy subjects before and after cerebellar tDCS; we also tested subjects with a visual attention task and a visual analogue scale (VAS) for mood. Anodal and cathodal cerebellar tDCS both significantly enhanced sensory processing in response to negative facial expressions (anodal tDCS, p=.0021; cathodal tDCS, p=.018), but left positive emotion and neutral facial expressions unchanged (p>.05). tDCS over the right prefrontal cortex left facial expressions of both negative and positive emotion unchanged. These findings suggest that the cerebellum is specifically involved in processing facial expressions of negative emotion.

Subthalamic Local Field Beta Oscillations during Ongoing Deep Brain Stimulation in Parkinson's Disease in Hyperacute and Chronic Phases

In the past years, local field potential (LFP) signals recorded from the subthalamic nucleus (STN) in patients undergoing deep brain stimulation (DBS) for Parkinson’s disease (PD) disclosed that DBS has a controversial effect on STN beta oscillations recorded 2–7 days after surgery for macroelectrode implantation. Nothing is known about these DBS-induced oscillatory changes 30 days after surgery. We recorded STN LFPs during ongoing DBS in 7 patients with PD, immediately (hyperacute phase) and 30 days (chronic phase) after surgery. STN LFP recordings showed stationary intranuclear STN beta LFP activity in hyperacute and chronic phases, confirming that beta peaks were also present in chronic recordings. Power spectra of nuclei with significant beta activity (54% of the sample) showed that it decreased significantly during DBS (p = 0.021) under both recording conditions. The time course of beta activity showed more evident DBS-induced changes in the chronic than in the hyperacute phase (p = 0.014). DBS-induced changes in STN beta LFPs in patients undergoing DBS in chronic phase provide useful information for developing a new neurosignal-controlled adaptive DBS system.

Conflict-dependent dynamic of subthalamic nucleus oscillations during moral decisions.

Although lesional, neuroimaging, and brain stimulation studies have provided an insight into the neural mechanisms of judgement and decision-making, all these works focused on the cerebral cortex, without investigating the role of subcortical structures such as the basal ganglia. Besides being an effective therapeutic tool, deep brain stimulation (DBS) allows local field potential (LFP) recordings through the stimulation electrodes thus providing a physiological “window” on human subcortical structures. In this study we assessed whether subthalamic nucleus LFP oscillations are modulated by processing of moral conflictual, moral nonconflictual, and neutral statements. To do so, in 16 patients with Parkinsons disease (8 men) bilaterally implanted with subthalamic nucleus (STN) electrodes for DBS, we recorded STN LFPs 4 days after surgery during a moral decision task. During the task, recordings from the STN showed changes in LFP oscillations. Whereas the 14–30 Hz band (beta) changed during the movement executed to perform the task, the 5–13 Hz band (low-frequency) changed when subjects evaluated the content of statements. Low-frequency band power increased significantly more during conflictual than during nonconflictual or neutral sentences. We conclude that STN responds specifically to conflictual moral stimuli, and could be involved in conflictual decisions of all kinds, not only those for moral judgment. LFP oscillations provide novel direct evidence that the neural processing of conflictual decision-making spreads beyond the cortex to the basal ganglia and encompasses a specific subcortical conflict-dependent component.

Subthalamic local field potentials after seven-year deep brain stimulation in Parkinson's disease.

Studies describing subthalamic (STN) local field potentials (LFPs) recorded during deep brain stimulation (DBS) in patients with Parkinsons disease (PD), within the first month after DBS electrode implant, show that DBS modulates specific STN oscillations: whereas low-frequency (LF) oscillations (2-7 Hz) increase, beta oscillations (8-30 Hz) variably decrease. No data show whether LFPs remain stable for longer than one month after DBS surgery. Having long-term information is essential especially for use as a long-term feedback control signal for adaptive DBS systems. To evaluate how STN activity behaves years after prolonged chronic stimulation in PD we studied STN LFPs at rest without DBS and during ongoing DBS, in 11 parkinsonian patients 7 years (7.54±1.04) after STN electrode implantation for DBS (hyperchronic group) and in 16 patients 3 days after STN electrode implantation (acute group). STN LF and beta-band LFPs recorded at rest at 7 years contained almost the same information as those recorded at 3 days. STN recordings showed similar LFP responses to DBS in the acute and hyperchronic stages: whereas during ongoing DBS the LF power band increased for the whole population, beta activity decreased only in nuclei with significant beta activity at baseline. The LF/beta power ratio in all nuclei changed in both study groups, suggesting that this variable might be an even more informative marker of PD than the single LF and beta bands. Because STN LFP activity patterns and STN LFP responses to DBS stay almost unchanged for years after DBS electrode implantation they should provide a consistent feedback control signal for adaptive DBS.

Neurophysiology of deep brain stimulation.

We review the data concerning the neurophysiology of deep brain stimulation (DBS) in humans, especially in reference to Parkinsons disease. The electric field generated by DBS interacts with the brain in complex ways, and several variables could influence the DBS-induced biophysical and clinical effects. The neurophysiology of DBS comprises the DBS-induced effects per se as well as neurophysiological studies designed to record electrical activity directly from the basal ganglia (single-unit or local field potential) through the electrodes implanted for DBS. In the subthalamic nucleus, DBS locally excites and concurrently inhibits at single-unit level, synchronizes low-frequency activity, and desynchronizes beta activity and also induces neurochemical changes in cyclic guanosine monophosphate (cGMP) and GABA concentrations. DBS-induced effects at system level can be studied through evoked potentials, autonomic tests, spinal cord segmental system, motor cortical and brainstem excitability, gait, and decision-making tasks. All these variables are influenced by DBS, suggesting also distant effects on nonmotor structures of the brain. Last, advances in understanding the neurophysiological mechanisms underlying DBS led researchers to develop a new adaptive DBS technology designed to adapt stimulation settings to the individual patients clinical condition through a closed-loop system controlled by signals from the basal ganglia.

Pathological gambling in Parkinson's disease: Subthalamic oscillations during economics decisions

Pathological gambling develops in up to 8% of patients with Parkinsons disease. Although the pathophysiology of gambling remains unclear, several findings argue for a dysfunction in the basal ganglia circuits. To clarify the role of the subthalamic nucleus in pathological gambling, we studied its activity during economics decisions. We analyzed local field potentials recorded from deep brain stimulation electrodes in the subthalamic nucleus while parkinsonian patients with (n = 8) and without (n = 9) pathological gambling engaged in an economics decision‐making task comprising conflictual trials (involving possible risk‐taking) and non conflictual trials. In all parkinsonian patients, subthalamic low frequencies (2–12 Hz) increased during economics decisions. Whereas, in patients without gambling, low‐frequency oscillations exhibited a similar pattern during conflictual and non conflictual stimuli, in those with gambling, low‐frequency activity increased significantly more during conflictual than during non conflictual stimuli. The specific low‐frequency oscillatory pattern recorded in patients with Parkinsons disease who gamble could reflect a subthalamic dysfunction that makes their decisional threshold highly sensitive to risky options. When parkinsonian patients process stimuli related to an economics task, low‐frequency subthalamic activity increases. This task‐related change suggests that the cognitive‐affective system that drives economics decisional processes includes the subthalamic nucleus. The specific subthalamic neuronal activity during conflictual decisions in patients with pathological gambling supports the idea that the subthalamic nucleus is involved in behavioral strategies and in the pathophysiology of gambling.

Increased short latency afferent inhibition after anodal transcranial direct current stimulation.

Transcranial direct current stimulation (tDCS), a technique for central neuromodulation, has been recently proposed as possible treatment in several neurological and psychiatric diseases. Although shifts on focal brain excitability have been proposed to explain the clinical effects of tDCS, how tDCS-induced functional changes influence cortical interneurones is still largely unknown. The assessment of short latency afferent inhibition (SLAI) of motor evoked potentials (MEPs) elicited by transcranial magnetic stimulation (TMS), provides the opportunity to test non-invasively interneuronal cholinergic circuits in the human motor cortex. The aim of the present study was to assess whether anodal tDCS can modulate interneuronal circuits involved in SLAI. Resting motor threshold (RMT), amplitude of unconditioned MEPs and SLAI were assessed in the dominant hemisphere of 12 healthy subjects (aged 21-37) before and after anodal tDCS (primary motor cortex, 13min, 1mA). SLAI was assessed delivering electrical conditioning stimuli to the median nerve at the wrist prior to test TMS given at the interstimulus interval (ISI) of 2ms. Whereas RMT and the amplitude of unconditioned MEPs did not change after anodal tDCS, SLAI significantly increased. In conclusion, anodal tDCS-induced effects depend also on the modulation of cortical interneuronal circuits. The enhancement of cortical cholinergic activity assessed by SLAI could be an important mechanism explaining anodal tDCS action in several pathological conditions.

The Effects of Levodopa and Deep Brain Stimulation on Subthalamic Local Field Low-Frequency Oscillations in Parkinson's Disease

New adaptive systems for deep brain stimulation (DBS) could in the near future optimize stimulation settings online so as to achieve better control over the clinical fluctuations in Parkinsons disease (PD). Local field potentials (LFPs) recorded from the subthalamic nucleus (STN) in PD patients show that levodopa and DBS modulate STN oscillations. Because previous research has shown that levodopa and DBS variably influence beta LFP activity (8-20 Hz), we designed this study to find out how they affect low-frequency (LF) oscillations (2-7 Hz). STN LFPs were recorded in 19 patients with PD during DBS, after levodopa medication, and during DBS and levodopa intake combined. We investigated the relationship between LF modulations, DBS duration and levodopa intake. We also studied whether LF power depended on disease severity, the patients clinical condition and whether LF modulations were related to electrode impedances. LF power increased during DBS, after levodopa intake and under both experimental conditions combined. The LF power increase correlated with the levodopa-induced clinical improvement and the higher the electrode impedance, the greater was the LF power change. These data suggest that the LF band could be useful as a control neurosignal for developing novel adaptive DBS systems for patients with PD.

Stormy weather in the human basal ganglia

In this issue of Clinical Neurophysiology, Joundi and colleagues assessed subthalamic beta (15–30 Hz) activity during finger tapping in a group of patients with Parkinson’s disease (PD) (Joundi et al., 2013). Their main finding was that beta activity recorded from the subthalamic nucleus (STN) decreases during finger tapping independently from motor biomechanical variables, whereas beta band activity after movement execution increases and correlates with the finger tapping rate. They, in essence, concluded that beta suppression facilitates continuous movement sequences. Over the past 10 years much interest has centered on beta activity recorded from human basal ganglia (BG) during movement execution. Available data shows that during voluntary movement beta band oscillations change both in the amplitude domain and in the frequency domain (Alegre et al., 2005; Cassidy et al., 2002; Doyle et al., 2005; Foffani et al., 2005; Hsu et al., 2012; Kempf et al., 2007; Kuhn et al., 2004; Levy et al., 2002; Paradiso et al., 2003; Priori et al., 2002; Williams et al., 2005). Focusing on amplitude domain, Joundi and colleagues expanded previous evidence showing that a sequential finger-tapping task induces persistent beta rhythm suppression uncorrelated with movement velocity. Their findings might have several important implications. First, as Joundi et al. themselves suggest, their observation helps to explain the role of beta-band neural oscillations in human motor control, supporting the current hypothesis that, rather than coding biomechanical movement variables, STN beta suppression subserves a global prokinetic function within the motor system. Second, changes in movement-related beta activity provide further support for developing a novel adaptive system for deep brain stimulation (DBS) therapy (Priori et al., 2005a,b). Ample evidence shows that BG oscillations recorded online reflect the patient’s clinical status (Alonso-Frech et al., 2006; Kuhn et al., 2008; Priori et al., 2004; Zaidel et al., 2009). The beta rhythm is among the candidate control signals for an adaptive DBS device (Giannicola et al., 2010, 2012; Marceglia et al., 2007). In fact, in studies conducted 1 month (Rosa et al., 2011) and 7 years (Giannicola et al., 2012) after DBS surgery we found that beta band activity has the advantages of being robust and resistant over time. Beta band activity reflects fluctuating BG oscillations caused by common therapeutic interventions for PD (levodopa, apomorphine and DBS) (Brown and Williams, 2005; Eusebio et al., 2011; Giannicola et al., 2010; Kuhn et al., 2006; Levy et al., 2002; Priori et al., 2004; Rosa et al., 2011). It also correlates at least in part with the patient’s clinical status (Kuhn et al., 2006; Priori et al., 2004; Zaidel et al., 2009). The findings reported by Joundi and colleagues underline that changes in BG oscillatory patterns due to the intrinsic pathogenetic mechanisms in PD cumulate with specific oscillations during movements that our patients execute in their day-to-day life. All these changes ultimately lead to a complex fluctuating oscillatory pattern. Despite these ‘‘stormy’’ fluctuations of the BG neural activity in patients with PD, current DBS systems deliver constant stimulation. The more recently proposed adaptive DBS systems will overcome this limitation by optimizing stimulation online to the BG ‘‘weather’’, thus improving patients’ quality of life (Marceglia et al., 2007). Whether the findings reported by Joundi and colleagues can provide practical help in developing the beta rhythm as a control neurosignal for an adaptive DBS system requires interpreting with caution. First, whereas several groups found the beta rhythm in only some patients (Eusebio et al., 2011; Giannicola et al., 2010; Rosa et al., 2011) others reported it in all their patients (Chen et al., 2006; Trottenberg et al., 2007; Zaidel et al., 2010). Hence beta band activity could lack consistency across different patients and would therefore be suitable as a control signal only in selected cases. A final major research problem that remains unsolved is the causality relationship between BG neural rhythms – including the beta rhythm – and the patient’s motor status. No evidence yet proves whether the beta, or other rhythms in the BG, cause the motor impairment (bradykinesia, akinesia) or vice versa. In theory, musculoskeletal changes, for instance changes in muscle tone, induced by treatments for PD, could cause oscillation changes. Conversely, changes in BG oscillatory patterns could causally and pathophysiologically be responsible for parkinsonian motor impairment. Passive angular joint displacement, with related changes in muscle length and tension, cause neuronal changes in the BG both at single and population level (Delong et al., 1984; Hamada et al., 1990; Johnson et al., 2009; Vitek et al., 1999; Wichmann et al., 1994; Wichmann and Kliem, 2004). This observation theoretically argues in favor of the hypothesis that BG oscillations merely reflect the patient’s peripheral status, with no pathophysiological role in motor impairment. Yet stimulation in the beta-frequency range in the human STN worsens motor conditions (Chen et al., 2011; Eusebio et al., 2008; Fogelson et al., 2005; Moro et al., 2002; Timmermann et al., 2004), thus suggesting that beta activity could have a causal antikinetic role. Obviously, the fact that stimulation in the beta-frequency range worsens patients’ clinical conditions does not per se definitively