A prospective analysis of morphometry, electrode position in STN-DBS, and its motor outcomes in advanced Parkinson's disease -- An institutional study.

Introduction: Parkinson's disease (PD), a progressive neurologic disorder that affects more than 1% of the population over age 65 years, is characterized by varying intensities of tremor, rigidity, and bradykinesia.[1,2] PD is primarily treated medically, especially in the early stages of disease treatment. Carlsson and Cotzias introduced oral levodopa/carbidopa as the "gold standard" medical therapy in 1968.[3] Surgical treatment of PD began in the 1940s with resection of premotor and motor cortices with the hope of alleviating parkinsonian tremor.[4] The resultant improvement was weighed against significant iatrogenic motor deficits, while no effect was seen on either rigidity or bradykinesia.[4] In 1990, Bergman et al. demonstrated in the nonhuman primate model that the induction of parkinsonism was associated with excessive and abnormally patterned discharges in the subthalamic nucleus (STN) and ablation of the nucleus alleviated all symptoms.[5] Based on this work, Benabid et al. implanted the first chronic subthalamic stimulator for PD in the early 1990s and subsequently documented alleviation of all cardinal motor signs of PD in a case series in 1998.[6,7] Deep brain stimulation (DBS) involves placement of electrodes deep in basal nuclei and delivering electrical current to the target. STN-DBS alleviates most of the parkinsonian symptoms. DBS has been shown to be safe and cost effective, conferring up to 40%--70% clinical improvement at 5 years and significantly improving quality of life compared to other treatment modalities.[1,2,8,9--16] DBS is associated with several advantages over other ablative procedures, with the most striking one being reversibility and adjustments in stimulation programs.[17,18] Precise targeting of the subthalamic nucleus has been identified as one of the major factors for the successful outcome of DBS surgery.[19] Hence, knowledge of the anatomical variation of the nucleus carries an important role in the treatment of PD. With the availability of directional leads and multiple programming options including white matter stimulation, there might be a paradigm shift in the near future in this concept of lead position. However, till such time, the final electrode position remains a significant factor determining the outcomes of STN-DBS. The present study aims to investigate the volume of STN and red nucleus (RN) on 3 T magnetic resonance imaging (MRI) susceptibility weighted imaging (SWI) sequences and its possible correlation with disease progression in patients with advanced PD. This study also aims to measure the accuracy of electrode targeting in a three-dimensional plane and motor outcomes of STN stimulation. Aims: 1. To study the 3D morphometry of STN and RN 2. To study the electrode position in patients of advanced PD undergoing DBS 3. To correlate the active electrode position with effect on motor symptoms of PD Materials and Methods This prospective study included all consecutive patients surgically treated for STN-DBS at the Department of Neurology and Neurosurgery at Nizam's Institute of Medical Sciences, Hyderabad, from January 2019 to March 2020. The study was approved by the institutional ethics committee. Final follow-up was done at 6 months. Inclusion criteria All idiopathic PD cases found eligible by Core assessment program for surgical interventional therapies in Parkinson's disease (CAPSIT-PD) protocol 17 with advanced stages of the disease and severe motor manifestations, who underwent DBS at our center were included in this study. Exclusion criteria Patients who required removal of implants or died within 6 months of surgery and could not complete the final follow-up at 6 months were not included in the analysis. Preoperative evaluation Preoperative evaluation included detailed history and examination to assess disease, comorbidities, and ongoing medications. The quantum of gross motor symptoms of disease was assessed by measuring the Unified Parkinson's Disease Rating Scale Part III (UPDRS III) by a qualified UPDRS III specialist in "OFF" phase after depriving the patient of his anti-parkinsonian drugs for at least 12 h. UPDRS III in "ON" phase was measured once the patient achieved best clinical response following administration of 200 mg levodopa + 50 mg carbidopa. Quantification of dopaminergic treatment was done using levodopa equivalent daily dose (LEDD) score with an appropriate levodopa equivalent dose conversion factor for each drug[18] (conversion factor: immediate release levodopa dose- 1, controlled release levodopa dose- 0.75, entacapone- 0.33, tolcapone- 0.5, pramipexole- 100, amantadine- 1). For example, patient with intake of levodopa 125 mg twice daily, controlled release levodopa 250 mg once daily, and tab pramipexole 1 mg once daily had a LEDD score of 450 ([100 x 2 x 1] + [200 x 0.75] + [1 x 100] = 450). All patients were assessed for dysgraphia by analysis of handwriting in ON phase by a single investigator on bedside testing (once the patient achieved the best clinical response following administration of 200 mg levodopa + 50 mg carbidopa orally). Handedness of patient was noted and hemicorporal UPDRS III of that side was derived. Patients were asked to write their names and places of stay on a paper. Clinical details regarding the quantum of disease and other features were not revealed to the reviewer who was appointed for analysis of handwriting. Subjective clinical analysis of patient's handwriting was done by the reviewer. Handwriting that was small in vertical and horizontal axes was considered as vertical and horizontal micrographia, respectively. Writing that fatigued and decreased in size as it progressed was considered as progressive micrographia. Different features of micrographia, that is, (a) vertical, (b) horizontal, and (c) progressive, were noted in ON phase in the preoperative period. In cases of disparity, patients were asked by the investigator if they had noted any change in handwriting. For quantification of change in micrographia, vertical length of the first letter and width of the word written were calculated. An analysis of improvement in legibility of handwriting was also done for all patients using Fahn--Tolosa--Marin Tremor Rating Scale (FTMTRS). Preoperative image acquisition All subjects were scanned with a uniform protocol involving T1W, T2W, T1W-contrast, GRE, and SWI sequences in 3 T MR (SIEMENS 3T Skyra) with 0.5-mm cuts and zero spacing for planning and targeting selection. MR scanner, image acquisition protocol, and methodology were uniform and done in ON phase without any need for general anesthesia. Volumetric analysis of STN and RN was done using SWI sequences. Data acquisition SWI acquisition Gradient echo sequences, using the following parameters: 80 slices, Field of view (FoV) = 240 mm, FoV phase 88%, Time to repetition(TR)/Time to echo (TE) = 28/20 ms, flip angle of 15°, with fat suppression and flow compensation, resolution matrix of 352 x 248 with slice thickness of 0.6 mm, Anterior-posterior (AP) and Right-left (RL) phase encoding directions for axial and coronal orientation, respectively, were used for the acquisition. Acquisition time was 7 min and 14 s for acquiring each SWI dataset. The volumes of STN and RN were measured on SWI coronal images by tracing the borders of STN and RN manually on each slice and then interpolating them. Use of axial sections and axial minimum intensity projection (min IP) images was made in cases where margins of the nuclei were not clearly delineated on coronal images. Target selection and surgery Target localization, based on the borders of the RN, was planned on a fused contrast MRI using Stealth Station software. For the frame coordinates, a preoperative MRI with frame was performed and fused. Cosman-Roberts-Wells (CRW) frame was used for all cases. The left STN was targeted first in all cases, followed by the right side. The surgical procedure included five-channel microelectrode recording for probable superior surface of STN and macrostimulation at low and high currents for response and side effects, respectively. The tract with the best response and the least side effects was chosen. Lead 3389 (Medtronic Inc., Minneapolis, MN, USA) was placed based on the depth obtained in the microelectrode recordings, with the help of rigid lead guide tube, keeping one in contact with the zona incerta and two in contact with the STN under fluoroscopic guidance by a single neurosurgeon. Postoperative 3 T MRI in T1 and T2W images was obtained before connecting the electrodes with the battery. Center of C1 contact (left) and C9 contact (right) of the electrode was taken as the final position. Cartesian coordinates of the final electrode position were compared to Cartesian coordinates of the proposed target, and final deviation was calculated. Any deviation more than 2 mm from the planned target was noted and considered significant. Electrodes within 2 mm epicenter of the proposed target coordinates were considered as being in optimum position. In electrodes with deviation more than 2 mm, final active contact position was labeled as medial or lateral and anterior or posterior, based on the individual deviation in final x and y coordinates, respectively. Follow-up The UPDRS III, LEDD scores were recorded at 6 months postsurgery. All values obtained postoperatively were compared with baseline scores, and outcomes were correlated with the final electrode position. Changes in speech, gait, and bradykinesia were evaluated by giving subscores individually. Improvement in hemicorporal UPDRS III score was correlated with contralateral electrode position. Postoperative analysis of handwriting was done in drug OFF phase after depriving the patient of his anti-parkinsonian drugs for at least 12 h, with the internal pulse generator (IPG) in ON state. Patients were asked to write the same set of words in the OFF phase that were written by the patient in preoperative ON period, and various features of micrographia, such as vertical, horizontal, and progressive components, were noted by the reviewer. For quantification of change in micrographia, vertical length of the first letter and width of the word written were calculated and compared with preoperative dimensions. Statistical analysis Data entry was done in into Microsoft Excel 2007, and statistical analysis was done using IBM Statistical Package for the Social Sciences (SPSS) trial version. One-way analysis of variance (ANOVA), t-test, and chi-square test were applied to test for significance. P value less than 0.05 was considered significant. Results: There were totally 64 patients who underwent surgery in the defined period. All the 128 implanted leads were analyzed till the end of the study period. Demographic parameters Age distribution The mean and median age of patients was 57.23 ± 9.70 and 57. 5 years, respectively (range 36--83 years). Maximum number of patients were in the age group of 60--70 years (n = 28, 43.75%), followed by the age group 50--60 years (n = 20, 31.25%). There were 10 patients (15.63%) in the age group of 40--50 years. Only one patient had age more than 80 years. Gender distribution There were 47 (73.44%) males and 17 (26.56%) females in the study cohort. Age of onset Mean age of onset in the present study was 49.23 + 10.49 years. There were 31 (48.44%) patients who had disease onset before 50 years of age, qualifying them for early-onset PD (EOPD) group, and 33 (51.56%) patients who had disease onset after 50 years of age, qualifying them for late-onset PD (LOPD) group. Disease duration Average duration of disease in the present cohort was 7.96 ± 4.68 years. There were 14 (21.88%) patients who presented to us within 5 years of onset of disease. There were 33 patients (51.56%) who had disease duration of 6--10 years, while 17 (26.56%) patients presented to us with disease duration of more than 10 years. Clinical parameters UPDRS III, hemicorporal UPDRS III, and LEDD Mean UPDRS III in the preoperative period was 55.64 ± 10.75. There were 21 patients (32.81%) who had UPDRS III in the range of 50--60. About 17 patients (26.56%) had UPDRS III in the range of 40--50 and 60--70. Only one patient had UPDRS III more than 80. Mean baseline right and left hemicorporal UPDRS III were 19.01 ± 4.35 and 18.98 ± 4.48, respectively. Mean preoperative LEDD was 699.21 ± 293.62 mg. Maximum number of patients (n = 12, 18.75%) had preoperative LEDD in the range of 600--700 and 700--800. Ten patients (15.63%) had LEDD in the range of 400--500. Two patients (3.13%) had LEDD of more than 1500. Volumetric analysis Volumetric analysis of STN was done for 52 patients. Volumes of right and left STN and RN were calculated from SWI sequences. None of the patients required any form of anesthesia for image acquisition. Average of volumes of nucleus from both sides was calculated. STN volume Mean volume of STN was 103.46 ± 21.17 mm3. Right STN volumes ranged from 60 to 120 mm3, with a mean of 106.15 ± 23.60 mm3, whereas left STN volumes ranged from 70 to 160 mm3, with a mean of 100.76 ± 21.76 mm3. RN volume Mean volume of RN was 321.73 ± 67.66 mm3. Right RN volumes ranged from 190 to 460 mm3, with a mean of 321.73 ± 65.16 mm3, whereas left RN volumes ranged from 130 to 500 mm3, with a mean of 321.73 ± 73.39 mm3. Correlation of STN and RN volumes with demographic and disease parameters Volumetric analysis of STN was done for 52 patients. Among them, 14 (37.84%) patients presented to us within 5 years of onset of disease. There were 20 patients (54.05%) who had disease duration of 6--10 years. Only three (8.11%) patients presented with the onset of disease duration more than 10 years. Disease onset before 50 years of age was seen in 24 (46.15%) patients, qualifying them for EOPD group, and 28 (53.85%) patients had disease onset after 50 years of age, integrating them into LOPD group. Parameters acquired through volumetric analysis were compared with demographic and clinical features. Hemicorporal UPDRS III was compared with contralateral STN and RN volumes. Subthalamic nucleus Disease duration was found to be positively correlated with STN volume, but statistical significance was not achieved. Overall, no difference related to gender was noticed, except in cases with disease duration of less than 5 years. Males had significantly lesser STN volume than females (P = 0.046). STN volume did not differ in EOPD and LOPD groups. No statistical significance was noted between UPDRS III in OFF state and STN volumes. On multivariate analysis as well, age of onset, disease duration, and UPDRS III scores were not found to be associated with any changes in STN volumes. Red nucleus Mean right and left RN volumes were 321.73 ± 65.16 and 321.73 ± 73.39 mm3, respectively. Average RN was 321.73 ± 67.66 mm3. Age was not found to be related to any volumetric change in RN. Weak positive trend was noted between volume of RN and disease duration (Pearson correlation 0.204, P = 0.14). Among patients with disease duration of 5--10 years, the average RN volume was significantly more in males (P = 0.018); however, no overall gender-related differences were noted. Patients with EOPD had significantly more volume of RN compared to patients in the LOPD group (P = 0.014). UPDRS III scores in OFF period did not correlate with nuclei volumes. On multivariate analysis between age of onset, disease duration, and UPDRS III score, only disease duration was associated with increased relative risk; however, significance was not reached (odds ratio [OR] 2.076, P = 0.40) Clinical outcomes post-STN-DBS Patients were followed at 6 months for analysis of outcomes. UPDRS III and LEDD scores of patients were noted. Mean reduction in overall UPDRS III, right and left hemicorporal UPDRS III scores was 32.20% ± 20.12%, 38.03% ± 21.37%, and 36.13% ± 21.01%, respectively. Difference in preoperative and postoperative UPDRS III scores was found to be statistically significant (P = 0.0001). Mean postoperative LEDD was 562.92 ± 277.22, compared to preoperative LEDD of 699.21 ± 293.62. Difference in preoperative and postoperative LEDD scores was statistically significant (P = 0.0079). Correlation of motor outcomes with demographic features Demographic parameters were analyzed for correlation with postoperative outcomes. Males and females had no difference in reduction in postoperative UPDRS III scores (P = 0.84) and LEDD (P = 0.70). No statistical difference in outcomes was noted in EOPD and LOPD groups. However, it was found that patients who were less than 65 years of age at the time of surgery had more significant reduction in UPDRS III, compared to more elderly patients (P = 0.02). More than 30% reduction in LEDD was noted in patients with age less than 65 years (P = 0.01). Disease duration and postoperative reduction in UPDRS III and LEDD scores had no relation with period. Correlation of electrode position with gross motor outcomes Deviation in active contact Mean deviation in all the frame coordinates was less than 2 mm. Least deviation in the final position was seen in the left x coordinate (1.5234 ± 1.2146) [Table 13]. Correlation of deviation in x coordinate and motor outcomes Negative correlation was found between reduction in UPDRS III scores and deviation from the proposed target in both right x (Pearson correlation - 0.16, P = 0.18) and left x coordinates (Pearson correlation - 0.21, P = 0.08) [Figures 23 and 24]. Cases with mediolateral deviation of left x less than 3 mm had significant reduction in UPDRS III (P = 0.05) and speech subscore (P = 0.05). Deviation less than 2 mm in left x was significantly associated with more than 50% reduction in gait subscores (P = 0.04). All the three patients with left x deviation more than 3 mm had deterioration in gait subscore. None of the patients with right x deviation more than 4 mm had any reduction in gait subscore. Correlation of deviation in y coordinate and motor outcomes Optimal placement of right y electrode was significantly associated with >30% reduction in UPDRS III in the postoperative period (P = 0.02). Cut-off of more than 30% was taken as better response for LEDD reduction. It was found that anterior deviation of right y electrode was associated with significantly lesser reduction in LEDD (P = 0.02). Negative correlation was found between deviation from the proposed target in left y and reduction in UPDRS III scores (Pearson correlation - 0.10, P = 0.40), right hemicorporal UPDRS III scores (Pearson correlation - 0.13, P = 0.28), gait subscores (Pearson correlation - 0.18, P = 0.13), and bradykinesia subscores (Pearson correlation - 0.02, P = 0.87). Analysis and correlation of fine motor symptoms with the electrode position Different features of handwriting in preoperative and postoperative periods were analyzed for 51 patients. Preoperatively, 28 (54.90%) patients had vertical micrographia, 14 (27.45%) patients had horizontal micrographia, and 24 (47.06%) patients had progressive micrographia. Postoperatively, maximum improvement was noted in vertical micrographia and minimum improvement in horizontal micrographia. After surgery, only 14 (27.45%) patients had vertical micrographia, while 11 (21.15%) patients had horizontal micrographia and 15 (29.41%) patients had progressive micrographia [Table 14]. The prevalence of micrographia was reduced in the postoperative period, but the difference was not statistically significant. The tremor component of handwriting was assessed by FTMTRS grading. There were 8, 15, and 28 patients in grades 1, 2, and 3 respectively, with none in grade 0, in the preoperative period. Postoperatively, only 10 (19.61%) patients had FTMTRS grade 3 handwriting. Postoperatively, four patients (7.84%) had deterioration in FTMTRS grades. Maximum number of patients (n = 22, 43.14%) showed improvement in FTMTRS grades by 1 [Table 14]. Mean preoperative FTMTRS Part B in ON phase was 2.38 ± 0.74 [Figure 26]. Mean post-DBS FTMTRS grade in OFF phase was 1.61 ± 1.03. Difference in pre- and postsurgery scores was found to be statistically significant (P = 0.0001). Increasing age was found to be negatively correlated with reduction in FTMTRS grades (Pearson correlation -0.10, P = 0.46). Patients with age more than 65 years had less improvement in FTMTRS grades (P = 0.03). Weak positive correlation was noted between changes in FTMTRS scores and hemicorporal UPDRS III scores (Pearson correlation 0.12, P = 0.40). A significant positive correlation was found between postoperative LEDD and FTMTRS scores (Pearson correlation 0.29, P = 0.03). Metric parameters of words written were compared in 51 cases. Mean vertical upstroke was 5.19 ± 2.01 and 5.65 ± 2.42 mm in preoperative and postoperative periods, respectively. Mean horizontal width of words was 3.31 ± 1.61 and 3.88 ± 1.58 cm in preoperative and postoperative periods, respectively. The differences in these parameters were not statistically significant. Moderate positive correlation was noted between improvement in FTMTRS grade and increase in horizontal width of words written (Pearson correlation 0.23, P = 0.10) Conclusions: Volume of STN stays consistent as the disease progresses. However, disease duration and early age of onset in PD can be associated with increased RN volume. Therapeutically, STN-DBS results in significantly improved functional outcome, particularly in patients with age less than 65 years. Accurate final electrode position in relation to the proposed target is associated with maximum clinical benefit and improvement in dysgraphia. Area of Research: Clinical aspects [ABSTRACT FROM AUTHOR]