Advertisement

Long-term directional deep brain stimulation: Monopolar review vs. local field potential guided programming

Open AccessPublished:April 28, 2022DOI:https://doi.org/10.1016/j.brs.2022.04.015

      Highlights

      • Intraoperative LFP beta activity predicts the most effective directional contact.
      • Intraoperative LFP can guide programming of long term directional stimulation.
      • Combination of physiological guidance with directionality individualize DBS therapy.
      • Directional stimulation maintains clinical efficacy at 3 years of follow up.

      Abstract

      Background

      Directional subthalamic stimulation in Parkinson's disease can increase stimulation threshold for adverse effects and widen the therapeutic window. However, selection of programming settings is time consuming, requiring a thorough monopolar clinical review. To overcome this, programming may be guided by intraoperatively recording local field potential beta oscillations (13–35 Hz).

      Objectives

      1) Evaluate whether the power of beta oscillations recorded intraoperatively can predict the clinically most effective directional contacts; and 2) assess long-term directional stimulation outcomes between patients programmed based on clinical monopolar review and patients programmed based on beta activity.

      Methods

      We conducted a non-randomized, prospective study with 24 Parkinson's disease patients divided into two groups. In group A (14 patients, 2016–2018), we investigated whether beta activity in the directional contacts correlated with clinical efficacy. Stimulating parameters were selected according to clinical monopolar review and mean follow-up was 27 months. In group B (10 patients, 2018–2019), stimulating parameters were selected according to beta activity and mean follow-up was 13 months.

      Results

      Neurophysiological results showed a strong correlation between clinical efficacy and the low-beta sub-band. Contacts with highest beta peaks increased the therapeutic window by 25%. Selecting the two contacts with highest beta peaks provided an 82% probability of selecting the best clinical contact. Clinical results showed similar improvements in group A (motor score, 72% reduction; levodopa-equivalent daily dose, 65% reduction) and B (72% and 63% reduction, respectively), maintained at long-term follow-up.

      Conclusions

      Our results validate the long-term efficacy of directional stimulation guided by intraoperative local field potential beta oscillations.

      Keywords

      Abbreviations

      AUC
      area under the curve
      DBS
      deep brain stimulation
      LFPs
      local field potentials
      UPDRS
      Unified Parkinson's Disease Rating Scale
      STN
      subthalamic nucleus

      1. Introduction

      Deep brain stimulation (DBS) of the subthalamic nucleus (STN) has proven to be an effective treatment for Parkinson disease [
      • Krack P.
      • Batir A.
      • Van Blercom N.
      • Chabardes S.
      • Fraix V.
      • Ardouin C.
      • et al.
      Five-Year follow-up of bilateral stimulation of the subthalamic nucleus in advanced Parkinson's disease.
      ,
      • Timmermann L.
      • Jain R.
      • Chen L.
      • Maarouf M.
      • Barbe M.T.
      • Allert N.
      • et al.
      Multiple-source current steering in subthalamic nucleus deep brain stimulation for Parkinson's disease (the VANTAGE study): a non-randomised, prospective, multicentre, open-label study.
      ,
      • Deuschl G.
      • Schade-Brittinger C.
      • Krack P.
      • Volkmann J.
      • Schäfer H.
      • Bötzel K.
      • et al.
      A randomized trial of deep-brain stimulation for Parkinson's disease.
      ]. DBS-related side effects occur frequently because of current leaking outside the sensorimotor region of the STN. Different techniques have been developed to avoid such side effects, including bipolar stimulation, interleaving or, more recently, directional stimulation [
      • Volkmann J.
      • Moro E.
      • Pahwa R.
      Basic algorithms for the programming of deep brain stimulation in Parkinson's disease.
      ,
      • Ramirez-Zamora A.
      • Kahn M.
      • Campbell J.
      • DeLaCruz P.
      • Pilitsis J.G.
      Interleaved programming of subthalamic deep brain stimulation to avoid adverse effects and preserve motor benefit in Parkinson's disease.
      ].
      Current steering with segmented electrodes increases the side-effect threshold and therefore the therapeutic window (TW) [
      • Contarino M.F.
      • Bour L.J.
      • Verhagen R.
      • Lourens M.A.J.
      • De Bie R.M.A.
      • Van Den Munckhof P.
      • et al.
      Directional steering: a novel approach to deep brain stimulation.
      ,
      • Pollo C.
      • Kaelin-Lang A.
      • Oertel M.F.
      • Stieglitz L.
      • Taub E.
      • Fuhr P.
      • et al.
      Directional deep brain stimulation: an intraoperative double-blind pilot study.
      ] compared to ring stimulation [
      • Steigerwald F.
      • Müller L.
      • Johannes S.
      • Matthies C.
      • Volkmann J.
      Directional deep brain stimulation of the subthalamic nucleus: a pilot study using a novel neurostimulation device.
      ]. However, the wide range of stimulation parameters available can increase the time required to optimize programming [
      • Wagle Shukla A.
      • Zeilman P.
      • Fernandez H.
      • Bajwa J.A.
      • Mehanna R.
      DBS programming: an evolving approach for patients with Parkinson's disease.
      ]. Thus, neurophysiological markers may provide guidance to reduce the programming burden.
      In patients with Parkinson's disease, local field potential (LFP) recordings from implanted electrodes in the STN have demonstrated high-amplitude, synchronized neuronal oscillations at beta-band frequencies (13–35 Hz) [
      • Brown P.
      • Oliviero A.
      • Mazzone P.
      • Insola A.
      • Tonali P.
      • Di Lazzaro V.
      Dopamine dependency of oscillations between subthalamic nucleus and pallidum in Parkinson's disease.
      ,
      • Hammond C.
      • Bergman H.
      • Brown P.
      Pathological synchronization in Parkinson's disease: networks, models and treatments.
      ] in the dorsolateral area of the STN [
      • Zaidel A.
      • Spivak A.
      • Shpigelman L.
      • Bergman H.
      • Israel Z.
      Delimiting sub territories of the human subthalamic nucleus by means of microelectrode recordings and a Hidden Markov Model.
      ,
      • Zaidel A.
      • Spivak A.
      • Grieb B.
      • Bergman H.
      • Israel Z.
      Subthalamic span of beta oscillations predicts deep brain stimulation efficacy for patients with Parkinson’s disease.
      ]. Beta oscillatory activity is suppressed by both levodopa treatment and high-frequency DBS [
      • Wingeier B.
      • Tcheng T.
      • Koop M.M.
      • Hill B.C.
      • Heit G.
      • Bronte-Stewart H.M.
      Intra-operative STN DBS attenuates the prominent beta rhythm in the STN in Parkinson's disease.
      ,
      • Giannicola G.
      • Marceglia S.
      • Rossi L.
      • Mrakic-Sposta S.
      • Rampini P.
      • Tamma F.
      • et al.
      The effects of levodopa and ongoing deep brain stimulation on subthalamic beta oscillations in Parkinson's disease.
      ,
      • Eusebio A.
      • Thevathasan W.
      • Doyle Gaynor L.
      • Pogosyan A.
      • Bye E.
      • Foltynie T.
      • et al.
      Deep brain stimulation can suppress pathological synchronisation in parkinsonian patients.
      ], which correlates with improvements in bradykinesia and rigidity [
      • Kühn A.A.
      • Kupsch A.
      • Schneider G.H.
      • Brown P.
      Reduction in subthalamic 8-35 Hz oscillatory activity correlates with clinical improvement in Parkinson's disease.
      ,
      • Neumann W.J.
      • Degen K.
      • Schneider G.H.
      • Brücke C.
      • Huebl J.
      • Brown P.
      • et al.
      Subthalamic synchronized oscillatory activity correlates with motor impairment in patients with Parkinson's disease.
      ]. Since the sensorimotor region of the STN is mainly located in the dorsolateral portion of the nucleus [
      • Monakow KH von
      • Akert K.
      • Künzle H.
      Projections of the precentral motor cortex and other cortical areas of the frontal lobe to the subthalamic nucleus in the monkey.
      ,
      • Nambu A.
      • Takada M.
      • Inase M.
      • Tokuno H.
      Dual somatotopical representations in the primate subthalamic nucleus: evidence for ordered but reversed body-map transformations from the primary motor cortex and the supplementary motor area.
      ,
      • Rodriguez-Oroz M.C.
      • Rodriguez M.
      • Guridi J.
      • Mewes K.
      • Chockkman V.
      • Vitek J.
      • et al.
      The subthalamic nucleus in Parkinson's disease: somatotopic organization and physiological characteristics.
      ,
      • Romanelli P.
      • Esposito V.
      • Schaal D.W.
      • Heit G.
      Somatotopy in the basal ganglia: experimental and clinical evidence for segregated sensorimotor channels.
      ], it is plausible to suggest that oscillating beta activity can mark the sensorimotor region in the STN of parkinsonian patients [
      • Trottenberg T.
      • Kupsch A.
      • Schneider G.H.
      • Brown P.
      • Kühn A.A.
      Frequency-dependent distribution of local field potential activity within the subthalamic nucleus in Parkinson's disease.
      ]. Analysis of beta-power activity can be used to guide programming, and adaptive stimulation has shown promising efficacy [
      • Rosin B.
      • Slovik M.
      • Mitelman R.
      • Rivlin-Etzion M.
      • Haber S.N.
      • Israel Z.
      • et al.
      Closed-loop deep brain stimulation is superior in ameliorating parkinsonism.
      ,
      • Little S.
      • Pogosyan A.
      • Neal S.
      • Zavala B.
      • Zrinzo L.
      • Hariz M.
      • et al.
      Adaptive deep brain stimulation in advanced Parkinson disease.
      ,
      • Rosa M.
      • Arlotti M.
      • Ardolino G.
      • Cogiamanian F.
      • Marceglia S.
      • Di Fonzo A.
      • et al.
      Adaptive deep brain stimulation in a freely moving parkinsonian patient.
      ,
      • Little S.
      • Beudel M.
      • Zrinzo L.
      • Foltynie T.
      • Limousin P.
      • Hariz M.
      • et al.
      Bilateral adaptive deep brain stimulation is effective in Parkinson's disease.
      ,
      • Little S.
      • Tripoliti E.
      • Beudel M.
      • Pogosyan A.
      • Cagnan H.
      • Herz D.
      • et al.
      Adaptive deep brain stimulation for Parkinson's disease demonstrates reduced speech side effects compared to conventional stimulation in the acute setting.
      ,
      • Arlotti M.
      • Marceglia S.
      • Foffani G.
      • Volkmann J.
      • Lozano A.M.
      • Moro E.
      • et al.
      Eight-hours adaptive deep brain stimulation in patients with Parkinson disease.
      ]. A recent study in 10 patients showed that LFPs could successfully guide DBS programming using directional leads [
      • Tinkhauser G.
      • Pogosyan A.
      • Debove I.
      • Nowacki A.
      • Shah S.A.
      • Seidel K.
      • et al.
      Directional local field potentials: a tool to optimize deep brain stimulation.
      ].
      In this study, we present our results with STN directional leads in 24 Parkinson's disease patients. Our study has two objectives: 1) evaluate whether the power of beta oscillations in LFP recorded intraoperatively from directional leads can predict the clinically most effective segmented contacts; and 2) compare the clinical outcomes at long-term follow-up between the group of patients with clinical monopolar review-based programming and those with beta-based programming. Overall, our aim is to validate the use of intraoperative LFP beta oscillations to guide the initial programming of long-term directional DBS in Parkinson's disease.

      2. Material and methods

      2.1 Patients

      We conducted a non-randomized, open label, prospective study in advanced Parkinson's disease patients. All met CAPSIT criteria and underwent DBS surgery for bilateral STN stimulation using directional leads with intraoperative LFPs recordings. Two groups of patients were studied: in Group A (14 patients in 2016–2018, 28 STN), stimulation was programmed based on clinical monopolar review; in Group B (10 patients in 2018–2019, 19 STN), stimulation was programmed based directly on the beta-peak power analysis from intraoperative LFPs. The institutional review board approved the study, and patients provided informed consent according to the Declaration of Helsinki. This study was not registered at Clin.Trials.gov.

      2.2 Target planning and surgical procedure

      All 24 procedures were conducted by the same team. Target coordinates were obtained by merging preoperative 3T MRI sequences with preoperative CT stereotactic images (Leksell Frame G, Elekta). Patients underwent awake surgery, after overnight withdrawal of dopaminergic medication (OFF-medication). Sedation with dexmedetomidine was used but discontinued during microelectrode recordings (MER) [
      • Krishna V.
      • Elias G.
      • Sammartino F.
      • Basha D.
      • King N.K.K.
      • Fasano A.
      • et al.
      The effect of dexmedetomidine on the firing properties of STN neurons in Parkinson's disease.
      ] and LFP recordings.
      Directional DBS leads (Cartesia; Boston Scientific, Valencia, California) were bilaterally implanted (apart from one patient in group B with a unilateral implant). These leads have four contact levels: the two middle levels are split into three segmented contacts, spanning 120°, and the highest and lowest contacts are ring shaped. Electrode and segmented contact final position was determined according to best MER activity and driving recordings. Once electrodes were implanted and fixed (electrode mark facing anterior position), an external connection allowed the LFPs to be recorded. Finally, the electrodes were internalized. Postoperative stereotactic CT was performed to ensure there were no surgical complications and a pulse generator was implanted.
      Optimal lead location was confirmed by fusing postoperative 3D CT and preoperative 3T MRI. Guide™ XT software (Boston SC) was used for lead location and 3D X-ray was used to assess lead orientation (Fig. 1C and D).
      Fig. 1
      Fig. 1LFP recordings and location of subthalamic leads methodology. 1A) Bipolar intraoperative directional LFP recordings methodology. Example of bipolar LFP recording during 300 s, with presence of activity in the beta band (13–35 Hz). 1B) LFP recordings with examples of patients where both B1 & B2 peaks were depicted and only B1 or only B2 peaks were found. 1C) 3D X-Ray with visualization of directional lead orientation. 1D) Location of the STN leads position (fusion of postoperative CT scan with preoperative 3T MRI). Directional leads (pink) in the STN (green), SNr (blue) and RN (red), image generated using Elements software (Brainlab, Munich) and Guide™ XT software (Boston SC, CA). LFP, local field potentials; SNr, substantia nigra; STN, subthalamic nucleus. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)

      2.3 Intraoperative LFP recordings

      LFPs were recorded for 300–500 s using a custom-made external connection, as previously described [
      • Fernández-García C.
      • Foffani G.
      • Dileone M.
      • Catalán-Alonso M.J.
      • González-Hidalgo M.
      • Barcía J.A.
      • et al.
      Directional local field potential recordings for symptom-specific optimization of deep brain stimulation.
      ], with a bipolar montage between the lowest ring electrode and every inferior segmented electrode and between the highest ring electrode and every superior segmented electrode. LFPs were amplified 10,000-fold and filtered at 1–3000 Hz (D-150, Digitimer, Cambridge, UK). Signals were digitized at 10,000 Hz by an analog/digital converter (1401 plus, Cambridge Electronic Design, Cambridge, UK) connected to a PC. Segments with artifacts were discarded, and the remaining segments were available for offline analysis (Fig. 1A).
      Previously to recordings, impedances were checked twice. A difference between the contacts of more than 1000 Ohms was not found in any case.

      2.4 Clinical monopolar review: ranking settings

      In Group A, OFF-medication clinical evaluation was performed from days 14–20 after surgical lead implant. The thorough and systematic monopolar clinical review was assessed by a movement disorder specialist blinded to the STN oscillatory activity findings. Monopolar review was performed by interrogating contacts from most dorsal (8) to most ventral (1) in a non-randomized order. Ring contacts were also evaluated, but not considered for the purpose of this study. Between testing each contact, wash-in time (time for rigidity to return to basal levels) were taken to avoid carry-over effects. Stimulation intensity was increased to 5 mA (0.5 mA steps), since adverse effects were well tolerated. The stimulation was set at a pulse width of 60 μs and frequency of 130 Hz [
      • Volkmann J.
      • Moro E.
      • Pahwa R.
      Basic algorithms for the programming of deep brain stimulation in Parkinson's disease.
      ].
      Standard monopolar stimulation review is focused on establishing the TW [
      • Steigerwald F.
      • Müller L.
      • Johannes S.
      • Matthies C.
      • Volkmann J.
      Directional deep brain stimulation of the subthalamic nucleus: a pilot study using a novel neurostimulation device.
      ], but with directional leads the proximity between segmented contacts may lead to similar TW (Fig. 2). For this reason, we also considered other parameters, such as threshold to dyskinesias [
      • Bouthour W.
      • Béreau M.
      • Kibleur A.
      • Zacharia A.
      • Tomkova Chaoui E.
      • Fleury V.
      • et al.
      Dyskinesia-inducing lead contacts optimize outcome of subthalamic stimulation in Parkinson's disease.
      ], to determine a more extensive clinical ranking for each directional contact, to increase test sensitivity (see Supplementary Material Fig. 1). The main clinical characteristics of monopolar review are depicted in Tables 1 and 2 within the Supplementary Material.
      Fig. 2
      Fig. 2Representation of the therapeutic window in Group A (n = 14).
      The therapeutic window can be highly comparable between contacts due to the close proximity between segmented contacts in directional leads. In this figure, the X-axis represent 14 patients in Group A (n = 28 hemispheres) and the Y-axis depicts the current intensity (mA). Dots represent the therapeutic window (mA) of each directional contact (168 contacts in total). The lines represent electrodes where several contacts shared the same therapeutic window (≥1 mA), (blue = three or more contacts, red = two contacts). There are 14 electrodes with two contacts sharing the same therapeutic window and 10 electrodes sharing three or more contacts; in 85.7% of electrodes, more than one contact shared the same therapeutic window. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)
      In group B, programming settings were selected by the intraoperative LFP oscillatory activity, and clinical outcomes were subsequently evaluated. Patients in group B were programmed based on LFPs, initially selecting the contact with the highest beta peak. Beta peaks were registered in LFPs from all patients in group B. Only in cases when the best beta contact was suboptimal (mainly due to refractory tremor) or caused side effects was the contact with the second highest beta peak chosen. If the clinical outcome was not optimum, monopolar review was conducted to select another segmented contact or apply ring stimulation.
      Orientation of the two best beta contacts were in the same horizontal plane in 53% of cases, 36% were in the same vertical/sagittal plane, and random orientation in 11% of cases.
      Systematic monopolar clinical review was applied, selecting the segmented contact as cathode and the pulse generator as anode in all patients in group A, as well as in all patients from both groups stimulated in directional monopolar configuration.
      At follow up, clinical outcomes were assessed using the Unified Parkinson's Disease Rating Scale (UPDRS)-III, levodopa-equivalent daily dose (LEDD) reduction, type of stimulation and current intensity applied (in mA).

      2.5 LFP analysis

      Offline analysis of LFPs was performed using Spike 2 software (version 8; Cambridge Electronic Design). Spectral analysis was performed using the Welch modified periodogram method, with segments of 8192 samples, Hanning window, and 50% overlap [
      • Brown P.
      Cortical drives to human muscle: the Piper and related rhythms.
      ].
      The results were presented as a power histogram (squared voltage/frequency) in the range 0–45 Hz. Distinctive frequency peaks were identified visually. The criterion for identifying a beta peak is based on a modification by Silberstein et al., 2003 [
      • Silberstein P.
      • Kühn A.A.
      • Kupsch A.
      • Trottenberg T.
      • Krauss J.K.
      • Wöhrle J.C.
      • et al.
      Patterning of globus pallidus local field potentials differs between Parkinson's disease and dystonia.
      ]. These peaks were defined as more than two contiguous bins (2.5 Hz wide) with an absolute power two times greater than the adjacent bins preceding and following the selected frequency. To study the correlation between beta activity and clinical outcome (determined by monopolar review), we first analyzed the area under the curve (AUC) for both B1 (13–20 Hz) and B2 (20–35 Hz) ranges; secondly, beta peaks power amplitude (when present) was measured in both the B1 and B2 ranges. Finally, these values were normalized by the mean value of the amplitudes of the beta spike power of the six directional contacts at each electrode, for each patient separately and averaged over all 14 patients (Fig. 1B).

      2.6 Statistical analysis

      For statistical analysis, the IBM SPSS software was used. For sample size calculation in group A, statistical tests were two-tailed with alpha = 0.05 and beta = 0.2; 22 samples were required to detect a difference ≥0.5 mA in the TW. A standard deviation of 0.82 was calculated from clinical ranking outcome scores. A follow-up loss rate of 0% was estimated. We recruited 14 patients (total sample of hemispheres = 28) to analyze clinical outcomes and LFP correlations. In order to perform the statistical analysis, the Kolmogorov-Smirnov test was done. Then, for normally distributed variables, parametric tests (Student's t) were performed, while for the rest and its comparisons, non-parametric tests (Wilcoxon, Spearman) were used.
      To assess whether beta measurements can be used to predict which contact to stimulate to achieve the best clinical effects for individual patients, we first conducted within-subject correlations for each patient using a Spearman test. Then used Student's t-test on the rho-to-z transformed correlation coefficients between subjects.
      To evaluate the power of beta activity in group A, we divided contacts from each hemisphere into two sets: the first set included the two contacts with highest beta peak-power activity, and the second set with the four remaining contacts (with lower beta peak-power activity). Both contact sets were compared regarding to clinical outcomes based on clinical ranking at the same stimulation intensity. Non-parametric Wilcoxon signed-rank tests with statistical significance p = 0.05 were used to compare the clinical outcomes of the two sets of contacts and compared between all 28 hemispheres. We used the parametric student's t-test to analyze the difference between the mean clinical outcome ranking position and the mean TW increase of the two sets of contacts. Spearman's test was applied to evaluate: a) the correlation between total beta power (AUC total beta power, 13–35 Hz) and clinical ranking; b) the correlation between the AUC B1 band and AUC B2 with clinical ranking (regardless of the presence of peaks); c) the correlation between B1 and B2 peaks with clinical ranking.
      In a second time, we conducted a non-inferiority study between the clinical monopolar review group (group A, n = 14) and LFP-guided directional stimulation group (group B). For a balanced comparison, the sample size of group B was calculated based on the group A sample size, as follows. For group B, the main outcome variable considered was UPDRS-III at 6 months. For an alpha risk of 0.05 and a beta risk of 0.2 in a unilateral contrast, given that the average reduction in group A would be 70.5% (66.23–75.11%) and setting the limit of non-inferiority to a difference of 10%, 18 samples would be required to test the hypothesis. A follow-up loss rate of 0% has been estimated. We included 10 patients (total sample of hemispheres = 19, one patient with unilateral implant) in group B. Unfortunately, it was not possible to complete follow up and only five patients were evaluated at 18 months.
      Student's t-test was used to compare changes in UPDRS, LEDD reduction over time, current intensity, surgery and programming time between both groups (n = 28 hemispheres vs. n = 19 hemispheres). Results were considered significant at p < 0.05, and are presented as mean (±standard deviation) in the text, and as box plots (median ± interquartile range) in the figures.

      3. Results

      3.1 Patient characteristics

      Group A included 14 patients (28 STN) (mean age 53 ± 11.08 years; mean disease duration from diagnosis till surgery 9.2 ± 4.44 years; male:female 9:5) whose pre-DBS, OFF-medication, mean UPDRS-III score was 35.9 ± 10.65 and LEDD was 969.5 ± 401.51 mg. Mean follow-up in this group was 27.6 ± 7.4 months. Group B included 10 patients (19 STN) (mean age 58.6 ± 12.98 years; mean disease duration 10.7 ± 4 years; male:female 7:3) whose pre-DBS, OFF-medication, mean UPDRS-III score was 37.5 ± 13.88 and LEDD was 899.0 ± 223.26 mg. Mean follow-up was 13.8 ± 4.7 months. There were no statistically significant differences between both groups (Supplementary Material Table 3).
      In group A, mean surgery time was 326 (±25) minutes and in group B was 319 (±27) minutes (p = 0.29); therefore, the surgical technique learning curve did not have a significant impact.

      3.2 Neurophysiological results

      In Group A, 168 STN contact pairs were recorded. All hemisphere recordings showed at least one identifiable beta peak. Beta-band sub-analysis showed that B1 peaks were present in 22 STN and B2 peaks in 23. Both B1 and B2 peaks were recorded in 17 STN, while only B1 peaks were observed in five cases and only B2 peaks in six cases.
      For clinical correlation, B1 peak was preferentially used to correlate with the clinical ranking (given its more specific responsiveness to levodopa treatment [
      • Priori A.
      • Foffani G.
      • Pesenti A.
      • Tamma F.
      • Bianchi A.M.
      • Pellegrini M.
      • et al.
      Rhythm-specific pharmacological modulation of subthalamic activity in Parkinson's disease.
      ,
      • Marceglia S.
      • Foffani G.
      • Bianchi A.M.
      • Baselli G.
      • Tamma F.
      • Egidi M.
      • et al.
      Dopamine-dependent non-linear correlation between subthalamic rhythms in Parkinson's disease.
      ]). In our study, we observed in contacts recording both B1 and B2 peaks that contacts with the highest peak power amplitude in B1 also presented the highest peak power amplitude in B2 (Spearman's Rho 0.60) (Supplementary Material Figs. 2A and 2B). Accordingly, the B2 peak was used for clinical correlation only in cases where there was no B1 peak.
      In addition, we calculated the correlation between the clinical ranking and the AUC of gamma frequency (60–90 Hz) (Spearman's Rho = 0.17; p = 0.43). It was not possible to calculate the correlation between the clinical ranking and low band (theta-alpha) because of the artifacts often present in this range in many patients.
      In Group B, beta activity was found in all 19 STN. In 13 hemispheres a B1 peak was recorded, and in 17 hemispheres a B2 peak was recorded. A B1-only peak was recorded in two hemispheres, a B2-only peak in six hemispheres, and both (B1 and B2) simultaneously in 11 hemispheres.
      The histogram of peaks distribution is illustrated in Fig. 3A and B within the Supplementary Material, “Neurophysiological Results - Beta peaks identification”.

      3.3 Beta power and clinical ranking correlations

      In group A patients, to assess whether beta peak power could be useful to guide directional stimulation, we first determined that within-subject relationship between beta and clinical ranking was robust across subjects (by conducting a t-test on the rho-to-z transformed correlation coefficients, t = 9.8, mean difference = −0.68, p = 0.0001) (see Supplementary Material Fig. 4).
      We analyzed the correlation between beta power activity (AUC beta power, AUC B1 power, AUC B2 power and beta (B1 or B2) peaks, with the results and clinical ranking detailed in the Supplementary Material). The strongest correlation with clinical ranking was found with the power of the B1 peak (Spearman's Rho −0.45, p = 0.0001) (see Supplementary Material Fig. 5).
      Considering the high spatial proximity between segmented contacts and their similar beta activity power amplitude or even the same TW, to increase test sensitivity we analyzed the clinical efficacy of the two contacts with highest beta activity and compared with the other four contacts in each electrode. Then we selected the two contacts with highest beta peaks per hemisphere and calculated their averaged normalized beta peak power and their mean clinical ranking position. Secondly, we calculated the averaged normalized beta peak power of the remaining four contacts per hemisphere and their mean clinical ranking position. Those 56 averaged results (two sets per hemisphere) showed a strong correlation between segmented contacts with highest beta peaks and clinical ranking (Spearman's Rho −0.68, p = 0.0001) (Fig. 3).
      Fig. 3
      Fig. 3Correlation between averaged beta peak power and clinical ranking. Representation of the strong correlation (r = −0.68, p = 0 0001) between averaged beta peak power and averaged clinical ranking, indicating that contacts with the highest beta peak power also presented the lower (better) clinical score. The X-axis shows the value of averaged beta peak power amplitude of 1° & 2° contacts (two directional contacts with largest B1 peak, or B2 peak when B1 peak is not present) and averaged beta peak power amplitude of 3° to 6° contacts (the other four directional contacts in each hemisphere), obtaining two sets of averaged beta peak power per hemisphere (28 hemispheres = 56 averaged beta power results). The Y axis shows the averaged clinical ranking score for each set of contacts. This strong correlation indicates that contacts with the highest beta peak power also presented the lower (better) clinical score.
      We further analyzed in each hemisphere how significant was the difference between the two sets of contacts established and the clinical ranking score. We found a statistically significant difference in Wilcoxon paired test (Z = −4.32; p = 0.000) between the clinical ranking position achieved with two contacts with the highest beta peaks and the remaining four contacts in each hemisphere, as represented in Fig. 4A. The mean clinical ranking score was 2.4 (±1.45) for the two contacts with the highest beta peaks compared to 4.1 (±1.54) for the third to sixth contacts. The mean clinical ranking difference between the two groups of contacts was 1.7 (±1.71). The size effect was high (d = 1.14), and so the relationship between these variables may be considered strong.
      Fig. 4
      Fig. 4Correlation between contacts with highest beta peaks versus contacts with lowest beta peaks and clinical ranking/therapeutic window in Group A patients.
      Comparison in each 28 hemispheres (Group A) between the two contacts with highest beta peaks (B1 or B2 sub-band) against the other four contacts with lowest beta peaks. 4A). X axis, 28 directional leads implanted; Y axis, clinical ranking (1 = higher clinical improvement, 4,5 = lower clinical improvement). The two contacts with the highest beta peaks (per hemisphere) are displayed as blue squares, with lower scores in clinical ranking; the remaining four contacts are represented by green dots and have higher clinical ranking scores. The blue horizontal line determines the mean clinical ranking for the two contacts with highest beta peaks (2.4) and the green horizontal line shows the mean clinical ranking for the remaining four contacts (4.1) (p = 0.0001). Hemispheres where the B2 peak was used for clinical correlation (as no B1 peaks were identified) are depicted by the red arrows. In four hemispheres, the two segmented contacts with the highest beta peaks had the same clinical efficacy (in rigidity score) as the other four contacts. One patient (25th and 26th hemispheres) showed no difference in rigidity outcome selecting beta peak contacts or the other four, and is currently stimulated in ring mode to treat severe tremor (ρ symbol). The other two hemispheres with equal efficacy between contacts shown in figure (Δ symbol) are stimulated in directional mode with good clinical efficacy. 4B) X-axis, 28 directional leads implanted; Y-axis, therapeutic window (TW; 0 = smaller TW, 6 = larger TW). The two contacts with the highest beta peaks (per hemisphere) are displayed as purple squares, with higher scores in TW; the remaining four contacts are represented by blue dots and have lower TW scores. The purple horizontal line determines the mean TW for the two contacts with highest beta peaks (2.45 mA) and the blue horizontal line shows the TW for the remaining four contacts (2 mA) (p = 0.009). (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)
      Based on these results, we found that the probability of selecting the best clinical contact by depicting the two contacts with greatest beta peaks was 82%.
      When we analyzed only the conventional TW, we found that in 85.7% of electrodes more than one contact shared the same TW (Fig. 2). Despite this, we found a statistically significant difference between the two set of contacts (Z = −2.596; p = 0.009, Wilcoxon paired test). There was an increase of 0.45 mA (25% increase, 95% confidence interval 5%–40%) using the two contacts with the highest beta peaks, statistically significant (p = 0.011, student's T test) (Fig. 4B). Size effect was also high (d = 5.81), thus difference is clinically relevant.
      In addition, we also calculated the correlation between each clinical ranking and beta power and performed a correlation matrix. As expected, the highest correlation was found between the adverse effects threshold with the therapeutic window, since the former is included in the definition. The beta ranking correlated best with the intensity at the maximum improvement (see Supplementary Material Table 4).

      3.4 Clinical outcomes

      We have described the clinical outcomes separately for Group A (stimulated based on the clinical monopolar review), who were followed for a longer period of time than Group B (stimulated directly according to beta peak-power analysis), and provided a comparison during the first 12 months (n = 24 patients). After initial programming, settings were selected, stimulation mode (directional vs. ring) were kept constant over time, and adjustments in LEDD or current intensity were conducted on demand at follow-up appointments.

      3.4.1 Group A (clinical monopolar-guided stimulation)

      In Group A, no immediate postoperative complications were observed. The mean time required to achieve thorough monopolar review and select stimulating parameters was 272 (±48) minutes per patient.
      At 12 and 18 months, respectively, the mean reduction in UPDRS-III (OFF-medication) was 72.3% (±16.5%) and 70.8% (±17.1%), the mean LEDD reduction was 65% (±22.9%) and 61.2% (±26.4%), and the mean increase in current intensity was 2.3 (±0.9) mA and 2.4 (±0.9) mA.
      Regarding the type of stimulation, 22 (75%) hemispheres had continued with directional stimulation after 2 years since lead implant (in four patients (100%) with the longest follow-up (3 years), directional stimulation was still effective). Six hemispheres (25%) had to be stimulated in the ring mode to achieve optimal tremor suppression (four hemispheres were not completely responsive to directional stimulation) and to treat gait disturbances in one patient.

      3.4.2 Group B (LFP-guided stimulation)

      In Group B, no immediate postoperative complications were observed. No time was necessary to conduct monopolar review as programming parameters were selected based on LFP analysis.
      At 12 months, the mean reduction in UPDRS-III (OFF-medication) was 72.1% (±15.3%), the mean LEDD reduction was 63.1% (±10.9%), and the mean increase in current intensity was 2.4 (±0.6) mA.
      Directional stimulation was used in 15 hemispheres (78%) over 12 months of follow-up, and four hemispheres were stimulated in the ring mode. In two patients with severe tremor despite directional stimulation using the two contacts with highest beta peaks, monopolar review was conducted and ring stimulation was needed to achieve optimal tremor control. In one of these patients, a wide bipolar combination using the upper and lowermost contacts in a bipolar fashion (8-1+) had to be used.
      For more details about programming configuration, see Table 5 in the Supplementary Material.

      3.5 Comparison of clinical outcomes between groups A and B

      We compared the clinical results over 12 months of follow-up, as illustrated in Fig. 5A. In both groups, we observed a reduction in the mean OFF-medication UPDRS-III (72.3% and 72.1% in groups A and B, respectively; p = 0.97) and in the mean LEDD reduction (65% and 63.1% in groups A and B, respectively; p = 0.16). The mean current intensity applied was similar between groups (2.5 mA in group A and 2.4 mA in group B; p = 0.59). There were no statistically significant differences between groups at 12 months of follow-up.
      Fig. 5
      Fig. 5Clinical outcomes.
      5A) Comparison of clinical outcomes between Group A (yellow; n = 14 patients, stimulated based on clinical monopolar review) and Group B (blue; n = 10 patients, stimulated based on LFP recordings and beta peak analysis). Representation of Unified Parkinson's Disease Rating Scale (UPDRS)-III median reduction, levodopa-equivalent daily dose (LEDD; mg) median reduction, and median current intensity (mA) applied during follow-up (36 months vs. 18 months in groups A and B, respectively). 5B) Comparison of clinical outcomes between the directional stimulation and ring stimulation groups. Representation of median UPDRS-III reduction, median LEDD (mg) reduction, and median current intensity (mA) applied during follow-up between the directional stimulation group (blue) and ring stimulation group (red) (36 months vs. 18 months in the directional stimulation and ring stimulation groups, respectively). (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)

      3.6 Directional vs. ring stimulation

      Despite not being the main objective of our study, we compared the clinical outcomes of all 24 patients between those stimulated in the directional versus ring mode (Fig. 5B). Directional stimulation was used in 30 hemispheres (69%) at 12 months (n = 22 patients, 43 hemispheres), 16 hemispheres (72%) at 24 months (n = 11 patients, 22 hemispheres), and 11 hemispheres (68%) at 30 months (n = 8 patients, 16 hemispheres).
      Ring stimulation at 6 months (10/47 (21%) hemispheres) was applied mainly to treat severe tremor (in eight (80%) hemispheres), and freezing of gait disturbances (in two (20%) hemispheres). The level used for ring stimulation included the contact showing the highest beta peak in all cases.
      At 12 months, the directional stimulation group showed a mean UPDRS-III reduction of 72.5% (±14.6%), a mean LEDD reduction of 62% (±19.6%), and a mean increase in current intensity of 2.3 (±0.9) mA. In the ring stimulation group at 12 months, there was a mean UPDRS-III reduction of 71.3% (±22.9%), a mean LEDD reduction of 58.9% (±37.4%), and mean increase in current intensity of 2.5 (±0.3) mA.
      There were no statistically significant differences between groups at 12 months of follow up (p = 0.89, p = 0.76 and p = 0.43 for UPDRS-III, LEDD and current intensity, respectively).

      4. Discussion

      4.1 LFP recordings predict the clinically most effective directional contact

      Intraoperative LFP recordings and postoperative sub-band analysis can be used to select optimal electrode contacts in DBS programming [
      • Ince N.F.
      • Gupte A.
      • Wichmann T.
      • Ashe J.
      • Henry T.
      • Bebler M.
      • et al.
      Selection of optimal programming contacts based on local field potential recordings from subthalamic nucleus in patients with Parkinson's disease.
      ].
      In our prospective study, we analyzed in Group A (14 patients) the relationship between the contacts with the highest clinical efficacy and the power of the beta band. Clinical monopolar review showed that 85.7% of electrodes had more than one contact sharing the same therapeutic window (Fig. 2). As our purpose was to achieve the highest discrimination among the segmented contacts, we included the TW and also other complementary clinical parameters, such as dyskinesia-induced current intensity (which can influence the contact efficacy [
      • Bouthour W.
      • Béreau M.
      • Kibleur A.
      • Zacharia A.
      • Tomkova Chaoui E.
      • Fleury V.
      • et al.
      Dyskinesia-inducing lead contacts optimize outcome of subthalamic stimulation in Parkinson's disease.
      ]).
      Correlation of clinical rankings with LFP recordings for directional leads was previously published by Tinkhauser et al. (2018) [
      • Tinkhauser G.
      • Pogosyan A.
      • Debove I.
      • Nowacki A.
      • Shah S.A.
      • Seidel K.
      • et al.
      Directional local field potentials: a tool to optimize deep brain stimulation.
      ]. They studied 10 Parkinson's disease patients and found that stimulating the two contacts with the highest beta activity showed a probability of selecting the best clinical contact in 84% of cases. In our study, we observed that rather than the total amount of normalized power of beta activity, the peaks – mainly in the low beta-band (B1, 13–20 Hz) – were highly correlated with clinical improvement. Therefore, selecting the two contacts with highest peaks led to an 82% probability of finding the best clinical contacts and to a good clinical correlation.
      In patients with severe tremor, it is possible that directional stimulation with the contacts presenting the greatest beta peaks would not achieve optimal tremor control. Previous studies have proved that beta activity is related to rigidity and bradykinesia but not correlated with tremor [
      • Kühn A.A.
      • Kupsch A.
      • Schneider G.H.
      • Brown P.
      Reduction in subthalamic 8-35 Hz oscillatory activity correlates with clinical improvement in Parkinson's disease.
      ,
      • Weinberger M.
      • Mahant N.
      • Hutchison W.D.
      • Lozano A.M.
      • Moro E.
      • Hodaie M.
      • et al.
      Beta oscillatory activity in the subthalamic nucleus and its relation to dopaminergic response in Parkinson's disease.
      ,
      • Kühn A.A.
      • Tsui A.
      • Aziz T.
      • Ray N.
      • Brücke C.
      • Kupsch A.
      • et al.
      Pathological synchronisation in the subthalamic nucleus of patients with Parkinson's disease relates to both bradykinesia and rigidity.
      ,
      • Anzak A.
      • Tan H.
      • Pogosyan A.
      • Foltynie T.
      • Limousin P.
      • Zrinzo L.
      • et al.
      Subthalamic nucleus activity optimizes maximal effort motor responses in Parkinson's disease.
      ,
      • Beudel M.
      • Little S.
      • Pogosyan A.
      • Ashkan K.
      • Foltynie T.
      • Limousin P.
      • et al.
      Tremor reduction by deep brain stimulation is associated with gamma power suppression in Parkinson's disease.
      ]. In our series, approximately 25% of patients had to be stimulated in the ring mode in order to achieve optimal tremor control.
      In group B (10 patients), we applied LFP-guided directional stimulation and confirmed that similar reductions in UPDRS-III (72%) and LEDD (63%) could be achieved at 12 months, with no need to change the type of stimulation or increase current intensity to maintain efficacy. Thus, we believe that LFP guidance is a useful tool to optimize programming and guarantees efficiency with long-term directional stimulation.

      4.2 Directional stimulation at long-term follow-up

      Directional stimulation emerged as a tool to avoid side effects derived from current leaking toward neighboring regions [
      • Contarino M.F.
      • Bour L.J.
      • Verhagen R.
      • Lourens M.A.J.
      • De Bie R.M.A.
      • Van Den Munckhof P.
      • et al.
      Directional steering: a novel approach to deep brain stimulation.
      ,
      • Pollo C.
      • Kaelin-Lang A.
      • Oertel M.F.
      • Stieglitz L.
      • Taub E.
      • Fuhr P.
      • et al.
      Directional deep brain stimulation: an intraoperative double-blind pilot study.
      ,
      • Schüpbach W.M.M.
      • Chabardes S.
      • Matthies C.
      • Pollo C.
      • Steigerwald F.
      • Timmermann L.
      • et al.
      Directional leads for deep brain stimulation: opportunities and challenges.
      ]. Directional leads can focus stimulation toward a confined section in the horizontal anatomical plane [
      • Timmermann L.
      • Jain R.
      • Chen L.
      • Maarouf M.
      • Barbe M.T.
      • Allert N.
      • et al.
      Multiple-source current steering in subthalamic nucleus deep brain stimulation for Parkinson's disease (the VANTAGE study): a non-randomised, prospective, multicentre, open-label study.
      ,
      • Steigerwald F.
      • Müller L.
      • Johannes S.
      • Matthies C.
      • Volkmann J.
      Directional deep brain stimulation of the subthalamic nucleus: a pilot study using a novel neurostimulation device.
      ,
      • Dembek T.A.
      • Reker P.
      • Visser-Vandewalle V.
      • Wirths J.
      • Treuer H.
      • Klehr M.
      • et al.
      Directional DBS increases side-effect thresholds—a prospective, double-blind trial.
      ], which allows a more accurate and individualized volume of tissue activated and may provide adequate stimulation to address clinical impairment during disease evolution in the mid-to long-term follow up [
      • Steigerwald F.
      • Müller L.
      • Johannes S.
      • Matthies C.
      • Volkmann J.
      Directional deep brain stimulation of the subthalamic nucleus: a pilot study using a novel neurostimulation device.
      ,
      • Dembek T.A.
      • Reker P.
      • Visser-Vandewalle V.
      • Wirths J.
      • Treuer H.
      • Klehr M.
      • et al.
      Directional DBS increases side-effect thresholds—a prospective, double-blind trial.
      ].
      In our study, we present results of 24 Parkinson's disease patients implanted with directional leads, which is the largest series with the longest follow-up in the literature. 70% of hemispheres were stimulated in the directional mode and this proportion was maintained for up to 3 years.
      In the literature, the reduction in UPDRS-III in PD patients implanted in the STN was 66% at 12 months and 59% at 24 months with ring stimulation [
      • Krack P.
      • Batir A.
      • Van Blercom N.
      • Chabardes S.
      • Fraix V.
      • Ardouin C.
      • et al.
      Five-Year follow-up of bilateral stimulation of the subthalamic nucleus in advanced Parkinson's disease.
      ,
      • Schüpbach W.M.M.
      • Chastan N.
      • Welter M.L.
      • Houeto J.L.
      • Mesnage V.
      • Bonnet A.M.
      • et al.
      Stimulation of the subthalamic nucleus in Parkinson's disease: a 5 year follow up.
      ,
      • Wider C.
      • Pollo C.
      • Bloch J.
      • Burkhard P.R.
      • Vingerhoets F.J.G.
      Long-term outcome of 50 consecutive Parkinson's disease patients treated with subthalamic deep brain stimulation.
      ,
      • Fasano A.
      • Romito L.M.
      • Daniele A.
      • Piano C.
      • Zinno M.
      • Bentivoglio A.R.
      • et al.
      Motor and cognitive outcome in patients with Parkinson's disease 8 years after subthalamic implants.
      ]. In our series, directional stimulation resulted in a mean UPDRS-III reduction of 73% and 69%, at 12 and 24 months, respectively. LEDD reduction with directional stimulation was 62% and 56% at 12 and 24 months, respectively and mean current intensity in our series was 2.5 mA over 3 years, similar to that reported in the literature [
      • Krack P.
      • Batir A.
      • Van Blercom N.
      • Chabardes S.
      • Fraix V.
      • Ardouin C.
      • et al.
      Five-Year follow-up of bilateral stimulation of the subthalamic nucleus in advanced Parkinson's disease.
      ,
      • Schüpbach W.M.M.
      • Chastan N.
      • Welter M.L.
      • Houeto J.L.
      • Mesnage V.
      • Bonnet A.M.
      • et al.
      Stimulation of the subthalamic nucleus in Parkinson's disease: a 5 year follow up.
      ,
      • Wider C.
      • Pollo C.
      • Bloch J.
      • Burkhard P.R.
      • Vingerhoets F.J.G.
      Long-term outcome of 50 consecutive Parkinson's disease patients treated with subthalamic deep brain stimulation.
      ,
      • Fasano A.
      • Romito L.M.
      • Daniele A.
      • Piano C.
      • Zinno M.
      • Bentivoglio A.R.
      • et al.
      Motor and cognitive outcome in patients with Parkinson's disease 8 years after subthalamic implants.
      ]. Our results demonstrate that directional stimulation is clinically effective, with reductions in UPDRS-III, LEDD, and current intensity that are similar to those published in the literature with ring stimulation, and which were stable and maintained over long-term follow-up.

      4.3 Future perspectives of directional stimulation

      At present, there are no randomized comparative results between ring and directional stimulation [
      • Timmermann L.
      • Jain R.
      • Chen L.
      • Maarouf M.
      • Barbe M.T.
      • Allert N.
      • et al.
      Multiple-source current steering in subthalamic nucleus deep brain stimulation for Parkinson's disease (the VANTAGE study): a non-randomised, prospective, multicentre, open-label study.
      ], although some ongoing studies have preliminarily shown interesting data in favor of directional stimulation [
      • Schnitzler A.S.
      • Mir P.M.
      • Brodsky M.B.
      • Verhagen L.V.
      • Groppa S.G.
      • Cheeran B.C.
      • et al.
      Directional versus omnidirectional Deep Brain Stimulation for Parkinson's disease: results of a prospective, blinded, multi-center, single-arm crossover study - MDS Abstracts.
      ]. In the study by of Schnitzler et al. [
      • Schnitzler A.S.
      • Mir P.M.
      • Brodsky M.B.
      • Verhagen L.V.
      • Groppa S.G.
      • Cheeran B.C.
      • et al.
      Directional versus omnidirectional Deep Brain Stimulation for Parkinson's disease: results of a prospective, blinded, multi-center, single-arm crossover study - MDS Abstracts.
      ], 66 Parkinson's disease patients were evaluated after 3 months of ring stimulation and another 3 months of directional stimulation. Directional stimulation provided a wider TW in 84.8% of cases, which was 35% greater than the ring TW. They also performed a blind evaluation for both clinicians and patients, and detected a preference for directional stimulation over the ring mode.
      Whether directional stimulation is useful other than avoiding motor side effects and is able to provide additional advantages is yet to be determined. However, in recent years there has been a growing interest in the neuropsychiatric effects related to STN-DBS. Some case studies have described mood and behavioral changes [
      • Berney A.
      • Vingerhoets F.
      • Perrin A.
      • Guex P.
      • Villemure J.G.
      • Burkhard P.R.
      • et al.
      Effect on mood of subthalamic DBS for Parkinson's disease: a consecutive series of 24 patients.
      ,
      • Bordini B.J.
      • Garg A.
      • Gallagher C.L.
      • Bell B.
      • Garell P.C.
      Neuropsychological effects of bilateral deep brain stimulation of the subthalamic nucleus in Parkinson's disease.
      ], while prospective studies have published the influence of DBS on fatigue, apathy, depression, anxiety, impulse control disorders, and weight gain [
      • Barichella M.
      • Marczewska A.M.
      • Mariani C.
      • Landi A.
      • Vairo A.
      • Pezzoli G.
      Body weight gain rate in patients with Parkinson's disease and deep brain stimulation.
      ,
      • Antosik-Wójcińska A.
      • Ł Święcicki
      • Dominiak M.
      • Sołtan E.
      • Bieńkowski P.
      • Mandat T.
      Impact of STN-DBS on mood, drive, anhedonia and risk of psychiatric side-effects in the population of PD patients.
      ,
      • Petry-Schmelzer J.N.
      • Krause M.
      • Dembek T.A.
      • Horn A.
      • Evans J.
      • Ashkan K.
      • et al.
      Non-motor outcomes depend on location of neurostimulation in Parkinson's disease.
      ]. It could be possible that directional stimulation guided by beta activity may offer clinical advantages regarding non-motor symptoms as it provides a more focused and individualized STN stimulation.

      4.4 Limitations

      One limitation of our study is the absence of a direct comparison between ring and directional stimulation, or between conventional programming and beta-activity guided programming. Regarding follow-up, patients were clinically evaluated by different movement disorder neurologists, so there could be a systematic bias. Another limitation may be related to the DBS stun effect, although patients were evaluated after 14–20 days (as is our common clinical practice). Although we report the results in 24 patients implanted with directional leads, a larger number of patients would be needed to support the long-term efficacy. In addition, during monopolar review contacts were interrogated in a non-randomized order; therefore, we cannot exclude a potential order effect. In addition, using 0.5 mA steps, especially from higher intensities, might mean that the cutoff for side effects could be missed. It should also be considered that the sample size decreased over time, affecting the analysis strength at 18 months follow-up and beyond.

      5. Conclusions

      Our results demonstrate the clinical efficacy of directional stimulation over 3 years of follow-up and validate the use of intraoperative LFP beta oscillations to guide the initial programming of long-term directional DBS in Parkinson's disease. LFP-guided programing provides a unique opportunity to adjust the stimulation for each individual patient, and the combination of physiological guidance with directionality will allow us to shape activation volumes tailored to patient needs.

      Funding sources

      This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

      CRediT authorship contribution statement

      Carla Fernández-García: Conceptualization, Methodology, Writing – original draft, preparation. Mariana H.G. Monje: Formal analysis, Resources. Víctor Gómez-Mayordomo: Resources, Visualization. Guglielmo Foffani: Visualization, Writing – review & editing. Rafael Herranz: Formal analysis, Data curation. Maria José Catalán: Resources, Writing – review & editing. Mercedes González-Hidalgo: Resources, Writing – review & editing. Jorge Matias-Guiu: Writing – review & editing. Fernando Alonso-Frech: Conceptualization, Methodology, Writing – original draft, preparation.

      Declaration of competing interest

      None.

      Acknowledgements

      The draft manuscript was edited by Deborah Nock (Medical WriteAway, Norwich, UK). We thank Dr. María Valle Loarte, Dr. Lydia Vela Desojo, Dr. Eva Lopez Valdés and Dr. Rocío García Ramos, who facilitated patient recruitment.

      Appendix A. Supplementary data

      The following is the Supplementary data to this article:

      References

        • Krack P.
        • Batir A.
        • Van Blercom N.
        • Chabardes S.
        • Fraix V.
        • Ardouin C.
        • et al.
        Five-Year follow-up of bilateral stimulation of the subthalamic nucleus in advanced Parkinson's disease.
        N Engl J Med. 2003; 349: 1925-1934https://doi.org/10.1056/nejmoa035275
        • Timmermann L.
        • Jain R.
        • Chen L.
        • Maarouf M.
        • Barbe M.T.
        • Allert N.
        • et al.
        Multiple-source current steering in subthalamic nucleus deep brain stimulation for Parkinson's disease (the VANTAGE study): a non-randomised, prospective, multicentre, open-label study.
        Lancet Neurol. 2015; 14: 693-701https://doi.org/10.1016/S1474-4422(15)00087-3
        • Deuschl G.
        • Schade-Brittinger C.
        • Krack P.
        • Volkmann J.
        • Schäfer H.
        • Bötzel K.
        • et al.
        A randomized trial of deep-brain stimulation for Parkinson's disease.
        N Engl J Med. 2006; 355: 896-908https://doi.org/10.1056/nejmoa060281
        • Volkmann J.
        • Moro E.
        • Pahwa R.
        Basic algorithms for the programming of deep brain stimulation in Parkinson's disease.
        Mov Disord. 2006; 21https://doi.org/10.1002/mds.20961
        • Ramirez-Zamora A.
        • Kahn M.
        • Campbell J.
        • DeLaCruz P.
        • Pilitsis J.G.
        Interleaved programming of subthalamic deep brain stimulation to avoid adverse effects and preserve motor benefit in Parkinson's disease.
        J Neurol. 2015; 262: 578-584https://doi.org/10.1007/s00415-014-7605-3
        • Contarino M.F.
        • Bour L.J.
        • Verhagen R.
        • Lourens M.A.J.
        • De Bie R.M.A.
        • Van Den Munckhof P.
        • et al.
        Directional steering: a novel approach to deep brain stimulation.
        Neurology. 2014; 83: 1163-1169https://doi.org/10.1212/WNL.0000000000000823
        • Pollo C.
        • Kaelin-Lang A.
        • Oertel M.F.
        • Stieglitz L.
        • Taub E.
        • Fuhr P.
        • et al.
        Directional deep brain stimulation: an intraoperative double-blind pilot study.
        Brain. 2014; 137: 2015-2026https://doi.org/10.1093/brain/awu102
        • Steigerwald F.
        • Müller L.
        • Johannes S.
        • Matthies C.
        • Volkmann J.
        Directional deep brain stimulation of the subthalamic nucleus: a pilot study using a novel neurostimulation device.
        Mov Disord. 2016; 31: 1240-1243https://doi.org/10.1002/mds.26669
        • Wagle Shukla A.
        • Zeilman P.
        • Fernandez H.
        • Bajwa J.A.
        • Mehanna R.
        DBS programming: an evolving approach for patients with Parkinson's disease.
        Parkinsons Dis. 2017; 2017https://doi.org/10.1155/2017/8492619
        • Brown P.
        • Oliviero A.
        • Mazzone P.
        • Insola A.
        • Tonali P.
        • Di Lazzaro V.
        Dopamine dependency of oscillations between subthalamic nucleus and pallidum in Parkinson's disease.
        J Neurosci. 2001; 21: 1033-1038https://doi.org/10.1523/jneurosci.21-03-01033.2001
        • Hammond C.
        • Bergman H.
        • Brown P.
        Pathological synchronization in Parkinson's disease: networks, models and treatments.
        Trends Neurosci. 2007; 30: 357-364https://doi.org/10.1016/j.tins.2007.05.004
        • Zaidel A.
        • Spivak A.
        • Shpigelman L.
        • Bergman H.
        • Israel Z.
        Delimiting sub territories of the human subthalamic nucleus by means of microelectrode recordings and a Hidden Markov Model.
        Mov Disord. 2009; 24: 1785-1793https://doi.org/10.1002/mds.22674
        • Zaidel A.
        • Spivak A.
        • Grieb B.
        • Bergman H.
        • Israel Z.
        Subthalamic span of beta oscillations predicts deep brain stimulation efficacy for patients with Parkinson’s disease.
        Brain. 2010; 133: 2007-2021https://doi.org/10.1093/brain/awq144
        • Wingeier B.
        • Tcheng T.
        • Koop M.M.
        • Hill B.C.
        • Heit G.
        • Bronte-Stewart H.M.
        Intra-operative STN DBS attenuates the prominent beta rhythm in the STN in Parkinson's disease.
        Exp Neurol. 2006; 197: 244-251https://doi.org/10.1016/j.expneurol.2005.09.016
        • Giannicola G.
        • Marceglia S.
        • Rossi L.
        • Mrakic-Sposta S.
        • Rampini P.
        • Tamma F.
        • et al.
        The effects of levodopa and ongoing deep brain stimulation on subthalamic beta oscillations in Parkinson's disease.
        Exp Neurol. 2010; 226: 120-127https://doi.org/10.1016/j.expneurol.2010.08.011
        • Eusebio A.
        • Thevathasan W.
        • Doyle Gaynor L.
        • Pogosyan A.
        • Bye E.
        • Foltynie T.
        • et al.
        Deep brain stimulation can suppress pathological synchronisation in parkinsonian patients.
        J Neurol Neurosurg Psychiatry. 2011; 82: 569-573https://doi.org/10.1136/jnnp.2010.217489
        • Kühn A.A.
        • Kupsch A.
        • Schneider G.H.
        • Brown P.
        Reduction in subthalamic 8-35 Hz oscillatory activity correlates with clinical improvement in Parkinson's disease.
        Eur J Neurosci. 2006; 23: 1956-1960https://doi.org/10.1111/j.1460-9568.2006.04717.x
        • Neumann W.J.
        • Degen K.
        • Schneider G.H.
        • Brücke C.
        • Huebl J.
        • Brown P.
        • et al.
        Subthalamic synchronized oscillatory activity correlates with motor impairment in patients with Parkinson's disease.
        Mov Disord. 2016; 31: 1748-1751https://doi.org/10.1002/mds.26759
        • Monakow KH von
        • Akert K.
        • Künzle H.
        Projections of the precentral motor cortex and other cortical areas of the frontal lobe to the subthalamic nucleus in the monkey.
        Exp Brain Res. 1978; 33: 395-403https://doi.org/10.1007/BF00235561
        • Nambu A.
        • Takada M.
        • Inase M.
        • Tokuno H.
        Dual somatotopical representations in the primate subthalamic nucleus: evidence for ordered but reversed body-map transformations from the primary motor cortex and the supplementary motor area.
        J Neurosci. 1996; 16: 2671-2683https://doi.org/10.1523/jneurosci.16-08-02671.1996
        • Rodriguez-Oroz M.C.
        • Rodriguez M.
        • Guridi J.
        • Mewes K.
        • Chockkman V.
        • Vitek J.
        • et al.
        The subthalamic nucleus in Parkinson's disease: somatotopic organization and physiological characteristics.
        Brain. 2001; 124: 1777-1790https://doi.org/10.1093/brain/124.9.1777
        • Romanelli P.
        • Esposito V.
        • Schaal D.W.
        • Heit G.
        Somatotopy in the basal ganglia: experimental and clinical evidence for segregated sensorimotor channels.
        Brain Res Rev. 2005; 48: 112-128https://doi.org/10.1016/j.brainresrev.2004.09.008
        • Trottenberg T.
        • Kupsch A.
        • Schneider G.H.
        • Brown P.
        • Kühn A.A.
        Frequency-dependent distribution of local field potential activity within the subthalamic nucleus in Parkinson's disease.
        Exp Neurol. 2007; 205: 287-291https://doi.org/10.1016/j.expneurol.2007.01.028
        • Rosin B.
        • Slovik M.
        • Mitelman R.
        • Rivlin-Etzion M.
        • Haber S.N.
        • Israel Z.
        • et al.
        Closed-loop deep brain stimulation is superior in ameliorating parkinsonism.
        Neuron. 2011; 72: 370-384https://doi.org/10.1016/j.neuron.2011.08.023
        • Little S.
        • Pogosyan A.
        • Neal S.
        • Zavala B.
        • Zrinzo L.
        • Hariz M.
        • et al.
        Adaptive deep brain stimulation in advanced Parkinson disease.
        Ann Neurol. 2013; 74: 449-457https://doi.org/10.1002/ana.23951
        • Rosa M.
        • Arlotti M.
        • Ardolino G.
        • Cogiamanian F.
        • Marceglia S.
        • Di Fonzo A.
        • et al.
        Adaptive deep brain stimulation in a freely moving parkinsonian patient.
        Mov Disord. 2015; 30: 1003-1005https://doi.org/10.1002/mds.26241
        • Little S.
        • Beudel M.
        • Zrinzo L.
        • Foltynie T.
        • Limousin P.
        • Hariz M.
        • et al.
        Bilateral adaptive deep brain stimulation is effective in Parkinson's disease.
        J Neurol Neurosurg Psychiatry. 2016; 87: 717-721https://doi.org/10.1136/jnnp-2015-310972
        • Little S.
        • Tripoliti E.
        • Beudel M.
        • Pogosyan A.
        • Cagnan H.
        • Herz D.
        • et al.
        Adaptive deep brain stimulation for Parkinson's disease demonstrates reduced speech side effects compared to conventional stimulation in the acute setting.
        J Neurol Neurosurg Psychiatry. 2016; 87: 1388-1389https://doi.org/10.1136/jnnp-2016-313518
        • Arlotti M.
        • Marceglia S.
        • Foffani G.
        • Volkmann J.
        • Lozano A.M.
        • Moro E.
        • et al.
        Eight-hours adaptive deep brain stimulation in patients with Parkinson disease.
        Neurology. 2018; 90: e971-e976https://doi.org/10.1212/WNL.0000000000005121
        • Tinkhauser G.
        • Pogosyan A.
        • Debove I.
        • Nowacki A.
        • Shah S.A.
        • Seidel K.
        • et al.
        Directional local field potentials: a tool to optimize deep brain stimulation.
        Mov Disord. 2018; 33: 159-164https://doi.org/10.1002/mds.27215
        • Krishna V.
        • Elias G.
        • Sammartino F.
        • Basha D.
        • King N.K.K.
        • Fasano A.
        • et al.
        The effect of dexmedetomidine on the firing properties of STN neurons in Parkinson's disease.
        Eur J Neurosci. 2015; 42: 2070-2077https://doi.org/10.1111/ejn.13004
        • Fernández-García C.
        • Foffani G.
        • Dileone M.
        • Catalán-Alonso M.J.
        • González-Hidalgo M.
        • Barcía J.A.
        • et al.
        Directional local field potential recordings for symptom-specific optimization of deep brain stimulation.
        Mov Disord. 2017; 32https://doi.org/10.1002/mds.26949
        • Bouthour W.
        • Béreau M.
        • Kibleur A.
        • Zacharia A.
        • Tomkova Chaoui E.
        • Fleury V.
        • et al.
        Dyskinesia-inducing lead contacts optimize outcome of subthalamic stimulation in Parkinson's disease.
        Mov Disord. 2019; 34: 1728-1734https://doi.org/10.1002/mds.27853
        • Brown P.
        Cortical drives to human muscle: the Piper and related rhythms.
        Prog Neurobiol. 2000; 60: 97-108https://doi.org/10.1016/S0301-0082(99)00029-5
        • Silberstein P.
        • Kühn A.A.
        • Kupsch A.
        • Trottenberg T.
        • Krauss J.K.
        • Wöhrle J.C.
        • et al.
        Patterning of globus pallidus local field potentials differs between Parkinson's disease and dystonia.
        Brain. 2003; 126: 2597-2608https://doi.org/10.1093/brain/awg267
        • Priori A.
        • Foffani G.
        • Pesenti A.
        • Tamma F.
        • Bianchi A.M.
        • Pellegrini M.
        • et al.
        Rhythm-specific pharmacological modulation of subthalamic activity in Parkinson's disease.
        Exp Neurol. 2004; 189: 369-379https://doi.org/10.1016/j.expneurol.2004.06.001
        • Marceglia S.
        • Foffani G.
        • Bianchi A.M.
        • Baselli G.
        • Tamma F.
        • Egidi M.
        • et al.
        Dopamine-dependent non-linear correlation between subthalamic rhythms in Parkinson's disease.
        J Physiol. 2006; 571: 579-591https://doi.org/10.1113/jphysiol.2005.100271
        • Ince N.F.
        • Gupte A.
        • Wichmann T.
        • Ashe J.
        • Henry T.
        • Bebler M.
        • et al.
        Selection of optimal programming contacts based on local field potential recordings from subthalamic nucleus in patients with Parkinson's disease.
        Neurosurgery. 2010; 67: 390-397https://doi.org/10.1227/01.NEU.0000372091.64824.63
        • Bouthour W.
        • Béreau M.
        • Kibleur A.
        • Zacharia A.
        • Tomkova Chaoui E.
        • Fleury V.
        • et al.
        Dyskinesia-inducing lead contacts optimize outcome of subthalamic stimulation in Parkinson's disease.
        Mov Disord. 2019; 34: 1728-1734https://doi.org/10.1002/mds.27853
        • Weinberger M.
        • Mahant N.
        • Hutchison W.D.
        • Lozano A.M.
        • Moro E.
        • Hodaie M.
        • et al.
        Beta oscillatory activity in the subthalamic nucleus and its relation to dopaminergic response in Parkinson's disease.
        J Neurophysiol. 2006; 96: 3248-3256https://doi.org/10.1152/jn.00697.2006
        • Kühn A.A.
        • Tsui A.
        • Aziz T.
        • Ray N.
        • Brücke C.
        • Kupsch A.
        • et al.
        Pathological synchronisation in the subthalamic nucleus of patients with Parkinson's disease relates to both bradykinesia and rigidity.
        Exp Neurol. 2009; 215: 380-387https://doi.org/10.1016/j.expneurol.2008.11.008
        • Anzak A.
        • Tan H.
        • Pogosyan A.
        • Foltynie T.
        • Limousin P.
        • Zrinzo L.
        • et al.
        Subthalamic nucleus activity optimizes maximal effort motor responses in Parkinson's disease.
        Brain. 2012; 135: 2766-2778https://doi.org/10.1093/brain/aws183
        • Beudel M.
        • Little S.
        • Pogosyan A.
        • Ashkan K.
        • Foltynie T.
        • Limousin P.
        • et al.
        Tremor reduction by deep brain stimulation is associated with gamma power suppression in Parkinson's disease.
        Neuromodulation Technol Neural Interface. 2015; 18: 349-354https://doi.org/10.1111/ner.12297
        • Schüpbach W.M.M.
        • Chabardes S.
        • Matthies C.
        • Pollo C.
        • Steigerwald F.
        • Timmermann L.
        • et al.
        Directional leads for deep brain stimulation: opportunities and challenges.
        Mov Disord. 2017; 32: 1371-1375https://doi.org/10.1002/mds.27096
        • Dembek T.A.
        • Reker P.
        • Visser-Vandewalle V.
        • Wirths J.
        • Treuer H.
        • Klehr M.
        • et al.
        Directional DBS increases side-effect thresholds—a prospective, double-blind trial.
        Mov Disord. 2017; 32: 1380-1388https://doi.org/10.1002/mds.27093
        • Schüpbach W.M.M.
        • Chastan N.
        • Welter M.L.
        • Houeto J.L.
        • Mesnage V.
        • Bonnet A.M.
        • et al.
        Stimulation of the subthalamic nucleus in Parkinson's disease: a 5 year follow up.
        J Neurol Neurosurg Psychiatry. 2005; 76: 1640-1644https://doi.org/10.1136/jnnp.2005.063206
        • Wider C.
        • Pollo C.
        • Bloch J.
        • Burkhard P.R.
        • Vingerhoets F.J.G.
        Long-term outcome of 50 consecutive Parkinson's disease patients treated with subthalamic deep brain stimulation.
        Park Relat Disord. 2008; 14: 114-119https://doi.org/10.1016/j.parkreldis.2007.06.012
        • Fasano A.
        • Romito L.M.
        • Daniele A.
        • Piano C.
        • Zinno M.
        • Bentivoglio A.R.
        • et al.
        Motor and cognitive outcome in patients with Parkinson's disease 8 years after subthalamic implants.
        Brain. 2010; 133: 2664-2676https://doi.org/10.1093/brain/awq221
        • Schnitzler A.S.
        • Mir P.M.
        • Brodsky M.B.
        • Verhagen L.V.
        • Groppa S.G.
        • Cheeran B.C.
        • et al.
        Directional versus omnidirectional Deep Brain Stimulation for Parkinson's disease: results of a prospective, blinded, multi-center, single-arm crossover study - MDS Abstracts.
        Mov Disord. 2019; 34 (n.d)
        • Berney A.
        • Vingerhoets F.
        • Perrin A.
        • Guex P.
        • Villemure J.G.
        • Burkhard P.R.
        • et al.
        Effect on mood of subthalamic DBS for Parkinson's disease: a consecutive series of 24 patients.
        Neurology. 2002; 59: 1427-1429https://doi.org/10.1212/01.WNL.0000032756.14298.18
        • Bordini B.J.
        • Garg A.
        • Gallagher C.L.
        • Bell B.
        • Garell P.C.
        Neuropsychological effects of bilateral deep brain stimulation of the subthalamic nucleus in Parkinson's disease.
        Stereotact Funct Neurosurg. 2007; 85: 113-120https://doi.org/10.1159/000098526
        • Barichella M.
        • Marczewska A.M.
        • Mariani C.
        • Landi A.
        • Vairo A.
        • Pezzoli G.
        Body weight gain rate in patients with Parkinson's disease and deep brain stimulation.
        Mov Disord. 2003; 18: 1337-1340https://doi.org/10.1002/mds.10543
        • Antosik-Wójcińska A.
        • Ł Święcicki
        • Dominiak M.
        • Sołtan E.
        • Bieńkowski P.
        • Mandat T.
        Impact of STN-DBS on mood, drive, anhedonia and risk of psychiatric side-effects in the population of PD patients.
        J Neurol Sci. 2017; 375: 342-347https://doi.org/10.1016/j.jns.2017.02.020
        • Petry-Schmelzer J.N.
        • Krause M.
        • Dembek T.A.
        • Horn A.
        • Evans J.
        • Ashkan K.
        • et al.
        Non-motor outcomes depend on location of neurostimulation in Parkinson's disease.
        Brain. 2019; 142: 3592-3604https://doi.org/10.1093/brain/awz285