Impact of simultaneous subthalamic and nigral stimulation on dysphagia in Parkinson’s disease

Study design

This single-center, controlled, randomized, double-blind, cross-over clinical trial was conducted at the University Medical Center Hamburg-Eppendorf and compares the effect of STN-DBS and combined STN + SNr-DBS as described previously.²⁵ The study was approved by the local ethics committee and was conducted in agreement with the Code of Ethics of the World Medical Association (Declaration of Helsinki, 1967). Written informed consent was obtained from all participants. Patients were seen by an unblinded therapeutic movement disorder specialist, who performed the programing of the stimulation condition and by a blinded rater at three visits with a time interval of 3 weeks in between (baseline recording with DBS switched off (STIM-OFF), phase I and phase II, see Figure 1). Patients were blinded for their stimulation mode (conventional STN stimulation or combined STN + SNR stimulation). All visits were performed in medication-on condition. Examinations and assessments were performed by blinded investigators at the end of phase I and II. After testing thresholds for side effects in the SNr, defined stimulation settings for the STN and STN + SNr stimulation were fixed for the course of the experiment. The STN settings were not different from those before the study. Afterward, stimulation of the STN or combined STN + SNr was set by a nonblinded investigator in a randomized manner for the following 3 weeks; that is, 7 of 15 PD patients received first a conventional STN stimulation (the control stimulation for placebo effects) and after 3 weeks the combined STN + SNr stimulation, while the other eight patients received first the combined STN + SNr stimulation followed by conventional STN stimulation (see Figure 1). After completion of phase I, the second visit was performed with the reprograming of stimulation parameters in a cross-over manner for the following 3 weeks (starting phase II). The third visit was performed after 6 weeks when phase II was completed. In the end, patients were unblinded to their stimulation mode and the preferred stimulation mode was programed as permanent therapeutic stimulation. Medication and stimulation parameters were held constant during the phase I and II of the study. Only in one case, the stimulation amplitude in the SNr had to be reduced after 2 days due to dyskinesias.

Subjects

Participants’ demographic and clinical characteristics are summarized in Table 1 and Table 2. About 15 patients suffering from idiopathic PD participated in the study. The 32 healthy controls were at least 50 years of age (half of them> 75 years) and without the history of swallowing problems (negative six item dysphagia-screening questionnaire) or any known disease of the central nervous system. All patients had a bilateral STN-DBS with the deepest contacts of the electrodes within the dorsal aspects of the SNr along image-based electrode reconstruction (>4.5 mm below AC-PC). The study in- and exclusion criteria, as well as the selection criteria for DBS surgery, have been reported in detail previously.²⁵ The location of the electrode position was controlled by stereotactic coordinates based on MR imaging and intraoperative microrecording. The sensorimotor STN was identified by cell responses to passive and active movements and a high prevalence of oscillating unit activities in the range between 10 and 30 Hz. Differentiation of STN from SNr was based on the decrease of background noise and the replacement of irregular STN unit activity by tonic regular high-frequency spiking of SNr neurons marked the ventral exit of the STN and dorsal aspect of the SNr, respectively. The stereotactic coordinates of the ventral most DBS contact relative to the mid-commissural point (MCP; mean ± standard deviation in mm) were x = 10.4 ± 0.9, y = 2.8 ± 1.3, z = 6.3 ± 1.0 for the left hemisphere and x = 10.1 ± 1.7, y = 2.7 ± 1.5, z = 5.7 ± 1.4 for the right side (x = lateral to midline, y = posterior to MCP, z = inferior to AC-PC level).

Four patients withdrew from the study due to side effects of combined STN + SNr-DBS. All of them terminated the study under a combined STN + SNr-DBS regime, suggesting that for these patients this stimulation mode was not adequate or even disadvantageous. In detail, side effects were worsening of motor functions as well as a lack of beneficial effects of levodopa, akathisia, general uncomfortable feeling, aggressiveness, and increased confusion and hallucinations. From the baseline electrode position, these patients did not differ from the other PD patients who completed the study. The results of the 11 patients, who completed the full study protocol, were considered.

Assessments and outcome measures

All subjects underwent a standardized otorhinolaryngological examination and a flexible-endoscopic evaluation of swallowing (FEES) by experienced otorhinolaryngologists, blinded with respect to the DBS parameters. FEES was performed using a 2.6-mm-diameter high-definition rhino-laryngo-videoscope (ENT-V3, Olympus Medical Systems Corp., Tokyo, Japan) as described before.² During FEES, standardized test boluses were given in a fixed order: first one teaspoon of pudding, second one teaspoon of water, third 90 ml of water to be drunk with a straw (quickly, but not hastily), fourth 1 biscuit (Ø 91 mm, weight 20 g), and fifth half slice of bread with butter (≈94 × 90 × 9 mm, weight 28 g).

For swallow analysis, residues, penetration and aspiration and leakage were evaluated for each consistency and for each participant.¹⁰ Residues were classified from 0 to 5. The occurrence of a build-up phenomenon, a special variant of residues that goes along with an increase of firm consistencies in the valleculae after each swallow, was also recorded. Penetration and aspiration were assessed on the Rosenbek Penetration-Aspiration Scale (PAS) and leakage on the “scale for bolus location”.²⁶ To better illustrate the clinical impact, pharyngeal residues were classified into three categories according to their clinical relevance: "slight" (0-2), "moderate" (3-4), and "severe” (5). The PAS-scores were grouped into three categories: "non-pathological" (PAS 1-2), "laryngeal penetration" (PAS 3-5), and "aspiration" (PAS 6-8). For each patient, a change in PAS and residue category at baseline compared with both stimulation modes was determined. The capability to swallow pills was rated on a 4-point scale (Table S1). Four different placebo-pills were administered to each participant in a random order: a) small round pill (∅ 7 mm), b) big round pill (∅10 mm), c) dragee (∅ 9.8 mm), and d) capsule (length 18 mm). Retained pharyngeal secretions were rated by the "Murray Score" based on the “short version of the four-point Secretion Severity Rating Scale.” Drooling was assessed using the “Drooling Severity and Frequency Scale” (DFSS) ranging from 1 to 5 for severity and from 1 to 4 for frequency. Both values are added to one resulting score (Score 2 = no drooling and score 9 = maximal drooling). A drooling score ≥ 6 was defined as severe drooling.

The swallowing duration of 90 ml of water, of one biscuit and half slice of bread was measured from first to the last gulp. For evaluating the self-perception of swallowing function under the three conditions, the patients were asked to rank themselves on a visual analog scale (VAS) with a vertical line from "no swallowing problems" to "strongest imaginable swallowing disorder."

Statistics

The primary endpoint was “residues of bread” as they are a characteristic pathology in PD patients.³ A sample-size calculation based on the mean values of residues along the 0-5 residue classification was performed with SD of 0.76, an expected effect size of 1 point, and a P-value of 0.05 resulting in a requirement of 11 patients per group to achieve a power of 80 %. Therefore, the number of PD patients assessed in our study seems to be sufficient.

For interval-scaled data, mean values and standard deviations (Mean ± SD) were calculated. In the boxplots median, interquartile range (IQR), whiskers (highest/ lowest values of the data set within 1.5 times of the IQR), and outliers (highest/lowest values of the data set outside the whiskers and within 3 IQR above/below the whiskers) are given. Values above 3 IQR are regarded as extremes. Significant differences in the mean values between patients and controls were tested with the t test for independent samples. For nominal and ordinal-scaled data, frequencies were calculated and compared by Fisher's exact test. The Friedman test was used to detect the influences of the three stimulation conditions on PAS and residues ranking for the patient group. Distribution differences in PAS and residue ranking between patient and control groups were tested by the Mann–Whitney U test. All statistical tests were two-tailed. To compare patients and controls, the significance level (alpha) was set to 0.05. In cases of multiple comparisons between patients, food consistencies and stimulation conditions, a Bonferroni correction was applied to the alpha level.

Statistical analyses were carried out with the statistical software package SPSS, version 19 (IBM, USA).

Effect of DBS mode on endoscopic swallowing parameters

Residues of bread and biscuit were more frequently and severely present in PD patients compared with controls at baseline and both under stimulation conditions (P < 0.005). Residues of water, as well as penetration and aspiration of all consistencies, were comparable between PD patients and controls (Table 3).

Under STN-stimulation moderate or severe pharyngeal residues of bread were found in 9/11 (80%) patients and of biscuit in eight patients. Three of them also showed penetration/aspiration of water. Affecting 6 of 11 patients also the build-up phenomenon (a special variant of residues which goes along with an increase of firm consistencies in the valleculae after each swallow) was common in the cohort. Simultaneous STN + SNr-stimulation had no additional positive effect on pharyngeal residues and penetration/aspiration. The three patients with penetration/aspiration had PAS-scores of 6, 5, and 8 under STN resp. 8, 5 and 7 under STN + SNr-DBS (no clinically relevant influence). DBS itself (compared with STIM-OFF) did not lead to a clear change in residue severity. Eight of the nine patients with bread residues under STN-stimulation showed no change in combined STN + SNr-stimulation. Considering simultaneous STN + SNr-stimulation, no significant effect on the other endoscopic swallowing parameters like build-up, pharyngeal secretion, leakage, and swallowing capability of pills was found.

Interestingly, PAS-scores of most patients were not influenced by DBS mode. However, the outcome was variable with PAS worsening for water under both stimulation modes in three patients. One of them demonstrated improvement of swallowing biscuit and another patient showed a substantially better PAS (from 8 to 1) under both stimulation modes.

The individual descriptive data, the mean values of residues and PAS under the stimulation modes and the results from Friedmann tests are given in Table 3. There was no significant change in the mean values of swallowing ability at baseline compared with STN-DBS or STN + SNR-DBS. Looking at the individual category change, most patients showed no change in penetration-aspiration ("non-pathological," "laryngeal penetration," and "aspiration") or pharyngeal residues ("slight,” "moderate," and "severe”) due to DBS mode (neither STN-DBS nor STN + SNr-DBS) for all tested consistencies. In both DBS modes, some patients exhibited improvement and some other impairment of swallowing without a significant difference. The influence of DBS mode on category change is illustrated for penetration-aspiration and residues in Figure 2A and B.

The build-up phenomenon was not influenced by DBS mode. In six patients (four with and two without build-up), no change due to DBS was observed. In the other patients, the presence of build-up was highly variable considering the DBS mode.

Most subjects in our cohort did not show relevant leakage and disturbed pharyngeal saliva management. Neither at baseline nor with STN-DBS, any patient exhibited a leakage. Under STN + SNr-DBS, only one patient showed a slight leakage of water and pudding and one patient a mild leakage of biscuit. Slight abnormalities of pharyngeal saliva management were found in only three patients and without relation to DBS mode.

The swallowing capability of pills was disturbed in four patients. All of them had a severe malfunction. One patient had a substantial impairment for all four pills, the remaining three patients had problems with one or two pills regarding all four pills equally. There was no significant impact of the factor DBS mode or pill type (Figure S1).

Effect of DBS mode on self-reported swallowing function and clinical parameters

Almost all patients indicated no or only slight dysphagia on the VAS (mean values: baseline 1.77; STN-DBS 0.94; STN + SNr-DBS 1.6). No significant influence of the DBS mode was found on the subjective impression (Figure 3), but the self-perception was very inconsistent in our cohort. Some patients reported improvement, other impairment of swallowing due to both DBS modes.

At baseline, only two patients showed drooling at all. One patient had moderate (DSFS 6) and one mild (DSFS 4) drooling, which was each absent in both DBS modes. Another patient developed moderate drooling (DSFS 6) only under STN + SNr-DBS.

The time differences for swallowing water, bread, and biscuit were not significantly different for both DBS-modes compared with baseline (Figure S2; Table S2).