Improvement of Advanced Parkinson’s Disease Manifestations with Deep Brain Stimulation of the Subthalamic Nucleus: A Single Institution Experience

We present our experience at the University of Illinois at Chicago (UIC) in deep brain stimulation (DBS) of the subthalamic nucleus (STN), describing our surgical technique, and reporting our clinical results, and morbidities. Twenty patients with advanced Parkinson’s disease (PD) who underwent bilateral STN-DBS were studied. Patients were assessed preoperatively and followed up for one year using the Unified Parkinson’s Disease Rating Scale (UPDRS) in “on” and “off” medication and “on” and “off” stimulation conditions. At one-year follow-up, we calculated significant improvement in all the motor aspects of PD (UPDRS III) and in activities of daily living (UPDRS II) in the “off” medication state. The “off” medication UPDRS improved by 49.3%, tremors improved by 81.6%, rigidity improved by 50.0%, and bradykinesia improved by 39.3%. The “off” medication UPDRS II scores improved by 73.8%. The Levodopa equivalent daily dose was reduced by 54.1%. The UPDRS IVa score (dyskinesia) was reduced by 65.1%. The UPDRS IVb score (motor fluctuation) was reduced by 48.6%. Deep brain stimulation of the STN improves the cardinal motor manifestations of the idiopathic PD. It also improves activities of daily living, and reduces medication-induced complications.


Introduction
The deep brain stimulation (DBS) system consists of a lead that is implanted into a specific deep brain target. The lead is connected to an implantable pulse generator (IPG), which is the power source of the system. The lead and the IPG are connected by an extension wire that is tunneled under the skin between both of them. This system is used to chronically stimulate the deep brain target by delivering a high-frequency current to this target [1,2].
James Parkinson was the first to describe Parkinson's disease (PD) in 1817; he described it as a combination of tremor, rigidity, postural abnormalities, and bradykinesia [3]. The main step that marked the onset of stereotactic surgery and the surgical treatment of different movement disorders was in 1947, when Ernest Spiegel and Henry Wycis invented the first frame-based stereotactic apparatus  At the end of the scan, we chose an axial T2 image (or two adjacent images) in which both the AC and the PC are seen (Figure 3). With simple arithmetic equations based on the Leksell frame coordinates system, we were able to calculate the stereotactic coordinates of the mid-commissural point (MCP), and the STN directly from the MRI coordinates of the AC and the PC (Figure 4). Based on the known anatomical relationship of the STN to MCP from the previous anatomical studies and stereotactic atlases [36,[47][48][49][50][51][52][53][54], we selected the STN target at 12 mm lateral, 3 mm posterior, and 6 mm inferior to the MCP.   At the end of the scan, we chose an axial T2 image (or two adjacent images) in which both the AC and the PC are seen (Figure 3). With simple arithmetic equations based on the Leksell frame coordinates system, we were able to calculate the stereotactic coordinates of the mid-commissural point (MCP), and the STN directly from the MRI coordinates of the AC and the PC (Figure 4). Based on the known anatomical relationship of the STN to MCP from the previous anatomical studies and stereotactic atlases [36,[47][48][49][50][51][52][53][54], we selected the STN target at 12 mm lateral, 3 mm posterior, and 6 mm inferior to the MCP. At the end of the scan, we chose an axial T2 image (or two adjacent images) in which both the AC and the PC are seen (Figure 3). With simple arithmetic equations based on the Leksell frame coordinates system, we were able to calculate the stereotactic coordinates of the mid-commissural point (MCP), and the STN directly from the MRI coordinates of the AC and the PC (Figure 4). Based on the known anatomical relationship of the STN to MCP from the previous anatomical studies and stereotactic atlases [36,[47][48][49][50][51][52][53][54], we selected the STN target at 12 mm lateral, 3 mm posterior, and 6 mm inferior to the MCP. Brain Sci. 2016, 6, 58 5 of 22     The second method we used to calculate the STN coordinates was direct visualization of the STN on a T2 weighted MRI ( Figure 5) [55]. The STN is the almond-shaped hypointense structure located lateral and anterior to the red nucleus. We identified an axial T2 image that showed the largest red nuclei circumference, and then we drew a line from the midline, medial to lateral, along the anterior edge of RN. The center of the STN was chosen at the extension of this straight line about 12 mm from the midline. Then the coordinates were calculated using the same Excel worksheet. Another method of the STN coordinates localization was done in the OR, using the FrameLink software, which is a part of the StealthStation navigation system (Medtronic, Minneapolis, MN, USA) ( Figure 6). The software compensates for head and frame tilt in any direction. It allows calculation of the STN coordinates and planning of suitable entry point and trajectory of the DBS electrode that avoid going through the cortical sulci, the ventricles, or any cerebral blood vessels. The final coordinates for the procedure were derived from all the previous techniques and subsequently adjusted using intraoperative electrical microrecording and macrostimulation.
In the operating room, the patient was placed on the operating table with a Leksell frame secured to the table using a Mayfield adapter. The C-arm was placed around the patient in order to use intraoperative fluoroscopy for electrode tracking and positioning ( Figure 7). We used transparent sterile drapes to allow easier communication with the patient and observation of the patient's symptoms during this awake procedure ( Figure 8). Two semicircular incisions were made on both sides of the midline ( Figure 9). Then we drilled two burr holes, one on each side, 1 cm anterior to the coronal suture and 2-3 cm lateral to the midline. We started the surgeries with the left side and then shifted to the right side.
We performed microelectrode recording (MER) of the brain activity using a NeuroNav microelectrode recording system (AlphaOmega, Nazareth, Israel) ( Figure 10). Fluoroscopic confirmation of the target approach was obtained at 5 mm intervals, 2 mm above the target, and at the target ( Figure 11). The second method we used to calculate the STN coordinates was direct visualization of the STN on a T2 weighted MRI ( Figure 5) [55]. The STN is the almond-shaped hypointense structure located lateral and anterior to the red nucleus. We identified an axial T2 image that showed the largest red nuclei circumference, and then we drew a line from the midline, medial to lateral, along the anterior edge of RN. The center of the STN was chosen at the extension of this straight line about 12 mm from the midline. Then the coordinates were calculated using the same Excel worksheet. Another method of the STN coordinates localization was done in the OR, using the FrameLink software, which is a part of the StealthStation navigation system (Medtronic, Minneapolis, MN, USA) ( Figure 6). The software compensates for head and frame tilt in any direction. It allows calculation of the STN coordinates and planning of suitable entry point and trajectory of the DBS electrode that avoid going through the cortical sulci, the ventricles, or any cerebral blood vessels. The final coordinates for the procedure were derived from all the previous techniques and subsequently adjusted using intraoperative electrical microrecording and macrostimulation.
In the operating room, the patient was placed on the operating table with a Leksell frame secured to the table using a Mayfield adapter. The C-arm was placed around the patient in order to use intraoperative fluoroscopy for electrode tracking and positioning ( Figure 7). We used transparent sterile drapes to allow easier communication with the patient and observation of the patient's symptoms during this awake procedure ( Figure 8). Two semicircular incisions were made on both sides of the midline ( Figure 9). Then we drilled two burr holes, one on each side, 1 cm anterior to the coronal suture and 2-3 cm lateral to the midline. We started the surgeries with the left side and then shifted to the right side.
We performed microelectrode recording (MER) of the brain activity using a NeuroNav microelectrode recording system (AlphaOmega, Nazareth, Israel) ( Figure 10). Fluoroscopic confirmation of the target approach was obtained at 5 mm intervals, 2 mm above the target, and at the target ( Figure 11).               After identification of the STN borders and depth by the MER, we started high-frequency macrostimulation. The aim of the stimulation was to confirm the optimal target, which provided adequate control of the Parkinsonian symptoms (specifically tremors), without undesirable effects from stimulation below 4 volts. Once we reached our desired target, we removed the microelectrode and replaced it with a standard four-contact (0-3) deep brain stimulation electrode (Medtronic DBS lead 3389). Generally, we placed the deepest electrode contact (0) at or just beyond the target point. We repeated the testing using this electrode in order to confirm the reproducibility of the effects. We locked the electrode in place using a Stimloc device (Medtronic, Minneapolis, MN, USA) ( Figure 12). The excess of the electrode was coiled around the burr hole to create a strain relief loop ( Figure 13). Then the same procedure was repeated on the right side.
The patient returned to hospital after one week for the second-stage surgery, in which the IPG was implanted in the sub-clavicular region under general anesthesia. After surgery, the IPG was interrogated. We checked the impedance of all eight contacts and programmed the pulse width, frequency, and amplitude of stimulation. By the end of the programming, we confirmed that the amplitude was set at zero and that the voltage of the battery was in the expected range.  After identification of the STN borders and depth by the MER, we started high-frequency macrostimulation. The aim of the stimulation was to confirm the optimal target, which provided adequate control of the Parkinsonian symptoms (specifically tremors), without undesirable effects from stimulation below 4 volts. Once we reached our desired target, we removed the microelectrode and replaced it with a standard four-contact (0-3) deep brain stimulation electrode (Medtronic DBS lead 3389). Generally, we placed the deepest electrode contact (0) at or just beyond the target point. We repeated the testing using this electrode in order to confirm the reproducibility of the effects. We locked the electrode in place using a Stimloc device (Medtronic, Minneapolis, MN, USA) ( Figure 12). The excess of the electrode was coiled around the burr hole to create a strain relief loop ( Figure 13). Then the same procedure was repeated on the right side.
The patient returned to hospital after one week for the second-stage surgery, in which the IPG was implanted in the sub-clavicular region under general anesthesia. After surgery, the IPG was interrogated. We checked the impedance of all eight contacts and programmed the pulse width, frequency, and amplitude of stimulation. By the end of the programming, we confirmed that the amplitude was set at zero and that the voltage of the battery was in the expected range. After identification of the STN borders and depth by the MER, we started high-frequency macrostimulation. The aim of the stimulation was to confirm the optimal target, which provided adequate control of the Parkinsonian symptoms (specifically tremors), without undesirable effects from stimulation below 4 volts. Once we reached our desired target, we removed the microelectrode and replaced it with a standard four-contact (0-3) deep brain stimulation electrode (Medtronic DBS lead 3389). Generally, we placed the deepest electrode contact (0) at or just beyond the target point. We repeated the testing using this electrode in order to confirm the reproducibility of the effects. We locked the electrode in place using a Stimloc device (Medtronic, Minneapolis, MN, USA) ( Figure 12). The excess of the electrode was coiled around the burr hole to create a strain relief loop ( Figure 13). Then the same procedure was repeated on the right side.
The patient returned to hospital after one week for the second-stage surgery, in which the IPG was implanted in the sub-clavicular region under general anesthesia. After surgery, the IPG was interrogated. We checked the impedance of all eight contacts and programmed the pulse width, frequency, and amplitude of stimulation. By the end of the programming, we confirmed that the amplitude was set at zero and that the voltage of the battery was in the expected range.

Post-Operative Patient Assessment
Immediately after the first stage of surgery, all patients had a CT scan of the head to rule out hemorrhage. All of them had an MRI of the brain on the same day of surgery or the next day before discharge to confirm accurate electrode placement.
An experienced neurologist performed the first postoperative programming session one month after surgery. This interval was given to allow the brain to recover from the surgery and the microlesioning effect of the electrode placement. The patient was programmed to the best setting that gave the best clinical improvement at the lowest stimulation intensity and largest therapeutic range before inducing undesirable effects. At the same time, drug doses were reduced.
Comprehensive neurological evaluation was done 12 months after surgery. At this visit, each patient came after 12 h without medical treatment. The patient was reassessed for parts I, II, and IV of the UPDRS. After this, each patient was assessed using the UPDRS III and video recorded in four conditions: "on" stimulation and "off" medication; then "off" stimulation and "off" medication, after switching off the stimulation for at least 1 hour; then "off" stimulation and "on" medication, after administration of a supra-threshold dose of levodopa and waiting for the "best on state"; then "on" stimulation and "on" medication, after turning on stimulation using the chronic stimulation parameters. We also calculated the axial score, MMSE and SERS [40]. We recalculated the LEDD for the patient's recent anti-Parkinsonism medications. We documented any surgical-, device-, or stimulation-related undesirable effects.

Data Collection and Statistical Analysis
The collected data were coded, tabulated, and statistically analyzed using the IBM SPSS statistics software version 22.0 (IBM Corp., Chicago, IL, 2013). Descriptive statistics were done for quantitative data as minimum & maximum of the range, median and first & third inter-quartile range, as well as mean ± SD (standard deviation). We calculated the number and percentages for qualitative data. Inferential analyses were done for quantitative variables using the two-tailed paired t-test for two dependent groups with parametric data, and the Wilcoxon signed rank test for two dependent groups with non-parametric data. Correlations were calculated using the two-tailed Pearson correlation for numerical parametric data, the two-tailed Spearman's rho test for numerical non-parametric and qualitative data, and the two-tailed partial correlation when controlling for a HYRS. The level of significance was taken at p value <0.05.

Results
Twenty patients were included in this study; six were men (30%), and 14 were women (70%). The age of the patients at the time of surgery ranged from 48 to 82 years, with a mean age of 61.1 years. The duration of illness before surgery ranged from five to 30 years, with a mean duration of 11.7 years. The mean ± SD of HYRS of PD stage was 3.6 ± 0.7 with a range from 2.5 to 5.0. No Figure 13. The final appearance of the electrode fixed using the Stimloc device; the excess of the electrode is coiled around the burr hole to create a strain relief loop.

Post-Operative Patient Assessment
Immediately after the first stage of surgery, all patients had a CT scan of the head to rule out hemorrhage. All of them had an MRI of the brain on the same day of surgery or the next day before discharge to confirm accurate electrode placement.
An experienced neurologist performed the first postoperative programming session one month after surgery. This interval was given to allow the brain to recover from the surgery and the micro-lesioning effect of the electrode placement. The patient was programmed to the best setting that gave the best clinical improvement at the lowest stimulation intensity and largest therapeutic range before inducing undesirable effects. At the same time, drug doses were reduced.
Comprehensive neurological evaluation was done 12 months after surgery. At this visit, each patient came after 12 h without medical treatment. The patient was reassessed for parts I, II, and IV of the UPDRS. After this, each patient was assessed using the UPDRS III and video recorded in four conditions: "on" stimulation and "off" medication; then "off" stimulation and "off" medication, after switching off the stimulation for at least 1 hour; then "off" stimulation and "on" medication, after administration of a supra-threshold dose of levodopa and waiting for the "best on state"; then "on" stimulation and "on" medication, after turning on stimulation using the chronic stimulation parameters. We also calculated the axial score, MMSE and SERS [40]. We recalculated the LEDD for the patient's recent anti-Parkinsonism medications. We documented any surgical-, device-, or stimulation-related undesirable effects.

Data Collection and Statistical Analysis
The collected data were coded, tabulated, and statistically analyzed using the IBM SPSS statistics software version 22.0 (IBM Corp., Chicago, IL, 2013). Descriptive statistics were done for quantitative data as minimum & maximum of the range, median and first & third inter-quartile range, as well as mean ± SD (standard deviation). We calculated the number and percentages for qualitative data. Inferential analyses were done for quantitative variables using the two-tailed paired t-test for two dependent groups with parametric data, and the Wilcoxon signed rank test for two dependent groups with non-parametric data. Correlations were calculated using the two-tailed Pearson correlation for numerical parametric data, the two-tailed Spearman's rho test for numerical non-parametric and qualitative data, and the two-tailed partial correlation when controlling for a HYRS. The level of significance was taken at p value <0.05.

Results
Twenty patients were included in this study; six were men (30%), and 14 were women (70%). The age of the patients at the time of surgery ranged from 48 to 82 years, with a mean age of 61.1 years.
The duration of illness before surgery ranged from five to 30 years, with a mean duration of 11.7 years. The mean ± SD of HYRS of PD stage was 3.6 ± 0.7 with a range from 2.5 to 5.0. No correlation was found between age, sex, family history, or the preoperative associated medical conditions and the one-year follow-up results ( Table 1).
The total surgical time of the first-stage surgery ranged from 133 to 280 min, with a mean ± SD of 214.8 ± 44.3 min. The average number of MER tracks used for mapping the STN at a single side was 1.4. Table 1. Preoperative demographic and medical characteristics.

Variables
Mean

Motor Scores (UPDRS III)
UPDRS III motor score improved by mean ± SD of 20.3 ± 8.9 (49.3%) from the preoperative "off" medication to the one-year follow-up of the "off" medication "on" stimulation score (p < 0.001). Tremors improved by mean ± SD of 8.0 ± 5.9 (81.6%) with p < 0.001. Rigidity improved by mean ± SD of 3.5 ± 2.1 (50.0%) with p < 0.001. Bradykinesia improved by mean ± SD of 5.7 ± 2.7 (39.3%) with p < 0.001. The axial score improved by mean ± SD of 2.0 ± 1.2 (38.5%) with p < 0.001 (Table 2, Figure 14).   We did not find a correlation between the preoperative improvement in the UPDRS III score, bradykinesia, and axial score after levodopa intake and the one-year follow-up values. A positive correlation was found between the preoperative improvement in tremors score (r = 0.563, p = 0.010) and rigidity score (r = 0.485, p = 0.030) with the postoperative tremors and rigidity, respectively.
The "off" medication "off" stimulation score at one year decreased by a mean of 0.4 (1%) compared to the preoperative "off" medication score. This change was found to be non-significant (p = 0.452). The "on" medication "off" stimulation at one year increased compared to the preoperative "on" score by a mean of 1.6 (8.3%) with p = 0.13.   We did not find a correlation between the preoperative improvement in the UPDRS III score, bradykinesia, and axial score after levodopa intake and the one-year follow-up values. A positive correlation was found between the preoperative improvement in tremors score (r = 0.563, p = 0.010) and rigidity score (r = 0.485, p = 0.030) with the postoperative tremors and rigidity, respectively.
The "off" medication "off" stimulation score at one year decreased by a mean of 0.4 (1%) compared to the preoperative "off" medication score. This change was found to be non-significant (p = 0.452). The "on" medication "off" stimulation at one year increased compared to the preoperative "on" score by a mean of 1.6 (8.3%) with p = 0.13.

Activity of Daily Living
The "off" medication UPDRS II improved by mean ± SD of 8.8 ± 4.5 (73.8%) with p < 0.001 ( Figure 15). The "on" medication UPDRS II improved by mean ± SD of 1.7 ± 5.5 (17.9%), but this improvement did not prove to be statistically significant. The "off" medication SERS improved at the one-year follow-up by mean ± SD of 37.5 ± 15.9 (104.2%) with p < 0.001.

Activity of Daily Living
The "off" medication UPDRS II improved by mean ± SD of 8.8 ± 4.5 (73.8%) with p < 0.001 ( Figure  15). The "on" medication UPDRS II improved by mean ± SD of 1.7 ± 5.5 (17.9%), but this improvement did not prove to be statistically significant. The "off" medication SERS improved at the one-year follow-up by mean ± SD of 37.5 ± 15.9 (104.2%) with p < 0.001.

Mental State
The UPDRS I score showed non-significant change from a preoperative score of 1.7 ± 1.2 to a one-year score of 1.9 ± 1.3 ( Table 4). The MMSE also showed non-significant change (Table 4).

Medications
LEDD decreased in the one-year follow-up by mean ± SD of 849.4 ± 448.1 mg/dL (54.1%) with p < 0.001. The UPDRS IVa score improved at the one-year follow-up by mean ± SD of 2.8 ± 2.1 (65.1%) from the preoperative score with p < 0.001. The UPDRS IVb score improved at the one-year followup by mean ± SD of 1.7 ± 1.1 (48.6%) from the preoperative score with p < 0.001 (Table 5, Figure 16).

Mental State
The UPDRS I score showed non-significant change from a preoperative score of 1.7 ± 1.2 to a one-year score of 1.9 ± 1.3 ( Table 4). The MMSE also showed non-significant change (Table 4).

Medications
LEDD decreased in the one-year follow-up by mean ± SD of 849.4 ± 448.1 mg/dL (54.1%) with p < 0.001. The UPDRS IVa score improved at the one-year follow-up by mean ± SD of 2.8 ± 2.1 (65.1%) from the preoperative score with p < 0.001. The UPDRS IVb score improved at the one-year follow-up by mean ± SD of 1.7 ± 1.1 (48.6%) from the preoperative score with p < 0.001 (Table 5, Figure 16).

Complications
All patients reported a postoperative headache that was relieved within a few days. This headache may be attributed to the surgical intervention itself or to the small film of pneumocephalus that appeared on the postoperative CT scans of all the patients. This pneumocephalus was transient and disappeared within days on follow-up CT scans. Three patients complained of memory difficulties; one patient experienced a worsening of his daytime hallucinations, one patient experienced numbness of the scalp at the incision site, and one patient experienced an increase in his depressive symptoms. These adverse events were transient and disappeared within a few weeks. One patient suffered a worsening of speech when turning the stimulation on. One patient developed postoperative right frontal hemorrhage at the site of the cortical penetration by the electrode, and she suffered deterioration of the level of consciousness. The hematoma was evacuated and she achieved complete recovery within one month. One patient suffered from wound dehiscence at the IPG implantation site. The IPG was removed and replaced with a new one in a subcutaneous pocket in the anterior abdominal wall.

Discussion
In this study we present a snapshot of our experience at UIC in the management of advanced PD using STN-DBS by reporting our preoperative and intraoperative methods as well as the clinical outcomes of 20 patients who were operated on from 2013 to 2014 and followed up for a year. Our experience with DBS started in 2001, and over the years we have modified and improved our techniques [55][56][57].

Complications
All patients reported a postoperative headache that was relieved within a few days. This headache may be attributed to the surgical intervention itself or to the small film of pneumocephalus that appeared on the postoperative CT scans of all the patients. This pneumocephalus was transient and disappeared within days on follow-up CT scans. Three patients complained of memory difficulties; one patient experienced a worsening of his daytime hallucinations, one patient experienced numbness of the scalp at the incision site, and one patient experienced an increase in his depressive symptoms. These adverse events were transient and disappeared within a few weeks. One patient suffered a worsening of speech when turning the stimulation on. One patient developed postoperative right frontal hemorrhage at the site of the cortical penetration by the electrode, and she suffered deterioration of the level of consciousness. The hematoma was evacuated and she achieved complete recovery within one month. One patient suffered from wound dehiscence at the IPG implantation site. The IPG was removed and replaced with a new one in a subcutaneous pocket in the anterior abdominal wall.

Discussion
In this study we present a snapshot of our experience at UIC in the management of advanced PD using STN-DBS by reporting our preoperative and intraoperative methods as well as the clinical outcomes of 20 patients who were operated on from 2013 to 2014 and followed up for a year. Our experience with DBS started in 2001, and over the years we have modified and improved our techniques [55][56][57].
We have done our preoperative target calculations using a combination of the standard indirect atlas-based X, Y, and Z coordinates of the STN at 12 mm lateral, 3 mm posterior, and 6 mm inferior to the MCP, and direct visualization of the STN on T2 weighted images.
We believe that intra-operative physiological and clinical confirmation of the target is crucial in the final position confirmation. The initial anatomical and radiological planning is also essential in target selection. Accurate preoperative planning would decrease the intra-operative time needed for the MER and the number of the microelectrodes tracks needed to reach the target, and subsequently decrease the complications. This fact is supported by our results, as our average number of MER tracks was 1.4 tracks and our average surgical time was 214.8 min.

Conclusions
Despite the limitations of this study as the assessments were not blinded to before vs. after surgery, to medication status, or to DBS status, we can conclude that bilateral deep brain stimulation of the subthalamic nucleus significantly improves motor symptoms, activities of daily living, and medication-induced complications in patients with advanced PD.