The Impact of Deep Brain Stimulation on Speech and Swallow Function in Parkinson Disease

Parkinson disease (PD) is among the most common neurodegenerative conditions affecting adults worldwide, with estimated prevalence in the United States to reach 1.2 million by the year 2030. In addition to the hallmark motor symptoms of PD, people with PD also experience changes to functions of communication and airway protection, in particular deficits in voice production, swallowing, and cough, and these contribute considerably to the morbidity and mortality of PD. Deep Brain Stimulation (DBS) is an increasingly used therapeutic option for alleviating many of the motor symptoms of PD; however, the impact of DBS on voice, swallowing, and cough is not clear. DBS targeted nuclei for PD include the subthalamic nucleus (STN) and globus pallidus interna (GPi). While historical reports suggested different therapeutic effectiveness for limb motor outcome across DBS target sites, more recent data suggest equipoise in treatment outcomes for STN and GPi targets. There is less conclusive evidence on the impact of DBS target site or laterality on voice, swallow, or cough, with existing evidence largely limited to bilateral STN DBS outcomes. Further, of the extant literature, interpretation of outcomes is limited by the methods employed that may lack longitudinal objective and validated outcomes pre- and post- DBS surgery, ON and OFF DBS stimulation, or that incorporate neurostimulation parameters or functional connectivity to adjacent brain regions in their design. Thus, although DBS is now routinely used in PD, significant gaps in knowledge remain regarding the impact of this therapeutic adjuvant on communication and airway protection.

The current project proposes to evaluate the effects of STN or GPi DBS placement for treating limb symptomatology, on voice, cough, and swallowing functions. We hypothesize that DBS stimulation at these sites will impact laryngeal motor control affecting voice, cough, and swallow functions due to spread of stimulation to cortibobulbar pathways. Because cranial nerve nuclei receive predominately bilateral upper motor neuron innervation, we posit that DBS effects are exacerbated with bilateral placement and stimulation of either the STN or GPi sites. Evidence that bilateral lesional stereotactic surgeries such as pallidotomy, capsulotomy and/or thalamotomy lead to worsening bulbar function, are in line with this hypothesis. Our supporting preliminary DBS data indicate that metrics of voice and swallowing degrade significantly following DBS to either the STN or GPi target, however measures of cough effectiveness may actually improve. Our innovative, multidisciplinary approach to determining DBS candidacy, target, and laterality provides an ideal opportunity to prospectively study these outcomes in a rigorous, pragmatic and methodical way. Therefore, we seek to address the following specific aims:

1. To compare laryngeal function during volitional voice tasks pre-post DBS, and when DBS placement is bilateral versus unilateral for STN and GPi targets. Hypothesis 1a: Laryngeal function during volitional voice tasks will remain unchanged pre-post unilateral STN or GPi DBS targets. Hypothesis 1b: Laryngeal function during volitional voice tasks will be worsened pre-post bilateral STN or GPi DBS.
2. To compare laryngeal function during volitional and induced cough tasks pre-post DBS, and when DBS placement is bilateral versus unilateral for STN and GPi targets. Hypothesis 2a: Peak expiratory cough airflow and cough volume acceleration will not significantly change for either voluntary or induced cough pre-post unilateral DBS placement to the STN and GPi targets. Hypothesis 2b: Peak expiratory cough airflow and volume acceleration will be increased during volitional cough compared to induced cough in those with bilateral DBS placement to the STN and GPi targets compared to pre-DBS.
3. To compare airway safety associated with laryngeal onset, degree, and duration of maximum closure during swallowing, pre-post DBS, and when DBS placement is bilateral versus unilateral for STN and GPi targets. Hypothesis 1: Swallowing safety, along with the degree, and duration of maximum laryngeal vestibule closure (LCV) during videofluoroscopic evaluation will be significantly reduced pre-post bilateral DBS placement to the STN and GPi targets compared to unilateral STN and GPi DBS treatment.

This prospective clinical trial will include two participant groups: STN DBS group and GPi DBS group. Each group will complete 4 study visits spaced approximately 6 - 12 months apart. For both the STN and GPi DBS groups, the first study visit will be completed following approval for surgery, but prior to DBS surgery. The second visit will follow placement of their first (unilateral) DBS lead, lead activation, and programming, and the third visit will follow placement of the second (bilateral) lead, lead activation, and programming (see Figure 1 for surgical timeline at our institution). The 4th visit will occur 6 - 12 months following the third visit to account for any time-related changes that may occur following DBS. The post-surgical visits will be completed across two days: one with the stimulators on, and the other with them off. The rationale for including stimulator on / off conditions relates to the central hypothesis: we hypothesize that spread of stimulation to corticobulbar fibers leads to changes in laryngeal motor control; therefore, we expect measures made in the off-stimulation condition NOT to differ from the pre-surgical baseline measures. For stimulation-off condition, participants will arrive for their appointment having turned off the stimulators overnight, which is standard care for their DBS follow-up visits. In order to maintain a reasonably low level of participant burden, all study assessments will be completed with participants in the medication 'on' state. However, we will collect data from their DBS programming visits which do assess patients in an off-medication state to document medication-related motor outcomes.

In order to complete this project in a pragmatic fashion that respects the best medical management for participants, we will not randomize participants to receive/not receive DBS, or to DBS target. DBS to both the STN and GPi has been FDA approved for use to treat motor features of PD since 2002; it will be used in a manner consistent with the population and indication for which it was approved. The DBS groups will consist of people who are already approved for DBS surgery (i.e., they are NOT receiving DBS because of participation in this study), and the control group will be recruited from over 1700 people with PD currently followed prospectively in our IRB-approved database, who do not have existing DBS and are not currently being considered for DBS. The DBS target site will be determined in the usual manner by the clinical DBS team based on factors such as motor concerns or complications, cognitive status and individual goals of the surgery, as per guidance by the Congress of Neurological Surgeons. Our inclusion criteria specifies that only patients with mild, or moderate voice and swallowing function may participate. Our goal with this approach is to increase the likelihood of well-matched GPi versus STN groups in terms of their pre-surgical voice and swallow functions, but not to interfere with surgical decisions and standard patient care.

Procedures Participant demographics: Demographic information, including age, sex, disease duration, primary goals for DBS surgery, stimulation parameters and levodopa equivalent dose.

Motor outcomes: Because DBS is intended to alleviate the motor symptoms of PD, it will be important to assess whether this primary therapeutic goal is achieved. Specifically, we will record:

• Unified Parkinson's Disease Rating Scale, Hoehn (UPDRS) - Will be used to assess change in motor function of participants. This will be collected both on/off medication and on/off stimulation.
• Hoehn \& Yahr score - To monitor disease stage. Pulmonary function measures: In order to account for physiologic changes in the respiratory system as possible contributing factors, we will analyze pulmonary function in all participants. The forced expired volume in the first second (FEV1) of a forced vital capacity (FVC) maneuver will be measured for each participant using a digital spirometer (Spirovision 3+m, Futuremed; or Koko spirometry, Nspire health). Maximum inspiratory pressure (PiMax) and maximum expiratory pressure (PeMax) will be measured with a manometer (MicroRPM, Micromedical inc). We will use standard procedures, as put forth by the American Thoracic Society, for measuring FVC, FEV1, the ratio of FEV1/FVC, PiMax, and PeMax.

Cognition: There is known impact of DBS and/or PD disease progression on cognitive function 15,50. In order to first screen for the presence of moderate or severe cognitive impairment (part of exclusionary criteria), as well as to account for potential impact cognitive changes on task performance, the Montreal Cognitive Assessment (MoCA) will be administered by a researcher trained and certified to administer the test (per most recent guidelines for administration on mocatest.org). This will be performed at each study visit.

Laryngeal imaging procedures: Indirect laryngoscopy will be performed with a flexible endoscope. Participants will be seated in a dental exam chair in the upright position. Water-based lubricant will be used to coat the endoscope which will be passed trans-nasally. Once the endoscope is in place with the larynx and surrounding structures clearly visible, the participant will be asked to perform the following tasks, three times each: rest breathing, breath hold, laryngeal diadochokinesis (alternating /i/ and brisk sniff), sustained vowel phonation, pitch glides up and down, soft voluntary cough production, and counting from sixty to seventy, and from eighty to ninety. Participants will be instructed to perform these tasks at normal pitch and loudness levels

Audio-perceptual \& Acoustic Procedures: Participant speech samples will be collected in a quiet clinical research space. Participants will be fitted with a head-mounted microphone situated 10 cm and 45° to 50° angle from the corner of the mouth54,55. Participants will also be equipped with electroglottography (EGG) sensors that are placed on the lateral aspects of the neck, on either side of the thyroid notch (Glottal Enterprises EG2-PCX2). The electrodes are held in place by a strap that is placed around the neck and secured with Velcro. Electrode gel may be used to improve the signal fidelity. They will perform various speech tasks.

Voluntary Cough: Participants will be outfitted with a facemask covering the nose and mouth. The facemask will be coupled to a pneumotachograph and differential pressure transducer. The cough airflow signal will be digitized (Power Lab Data Acquisition System) and recorded (LabChart 7; ADInstruments, Inc) to a laptop computer. Participants will be asked to cough 3 times, with the instruction to "cough as if something went down the wrong pipe.

Induced Cough: Participants will be outfitted with the same equipment used for voluntary cough testing, however there will be a side port with a one-way inspiratory valve for nebulizer connection. The nebulizer will be connected to a dosimeter that delivers aerosolized capsaicin solution during inspiration with a delivery duration of 2 seconds. Participants will complete a capsaicin challenge with three randomized blocks of 0, 50, 100, 200, and 500 μM capsaicin dissolved in a vehicle solution. Participants will be given the instruction "cough if you need to" prior to capsaicin delivery.

Swallowing Evaluation: Swallowing function and physiology will be assessed using high resolution digital fluoroscopy at a pulse rate of 30 frames per second, digitally stored for retrieval and offline analysis. Participants will be seated in the lateral viewing plane. Participants will perform trials thin and pudding-thick boluses.

These tasks will be performed at each study visit, and in the on/off DBS conditions post surgically.