Expectations and outcomes

Parkinson's disease

Some of the initial attempts at long-term stimulation for chronic disease states were attributed to Shealy in the USA [ ], and Bechtereva in Russia [ ]. However, before stimulation efforts, other surgical measures had been used to effect modulation of the nervous system including the ligation of the choroidal artery for movement disorders [ ] and targeted lesioning. Although DBS is currently used most often for movement disorders, there are many other indications. In this setting, implantation of stimulation devices generally targets the subthalamic nuclei (STN), globus pallidus internus (GPi), or ventralis intermediate nucleus (Vim) of the thalamus.

The exact placement of the stimulator has significant impact on the patient, resulting in symptom relief that is unique for each target. In terms of motor improvements, the targets and symptom relief can vary. For example, the target could include the GPi in cases of dystonia [ , ] or the Vim in cases of non-parkinsonian tremor. In patients with Parkinson's disease (PD), targeting this subnucleus (Vim) does relieve the parkinsonian tremor but fails to modify the other chief symptoms of the disease [ ]. Although debate exists regarding the ideal target for various movement disorders, the target is largely based on the type of expected symptom relief. Target selection is then matched to the nature of symptoms experienced by the patient. To ensure the optimal surgical outcomes for patients with movement disorders, the medical team should carefully select the patient, exhaust the available medical treatments, choose the best target and surgical technique [ ], optimize the DBS settings [ , ], and properly manage the postsurgery medication regimens.

DBS for PD patients is complicated in part because of the spectrum of symptoms that are not just movement related. The stimulation effect reaches other aspects of the disease and does so to varying degrees depending on the target. Both the GPi and STN nuclei not only improve parkinsonian symptoms but also reduce drug-induced dyskinesias [ ] but, in addition, the STN is thought to also reduce medication burden. In the 2009 COMPARE trial, the authors demonstrated that motor score improvements were similar whether targeting the GPi or STN [ ], the same as the results shown by Odekerken et al. in the NSTAPS study [ ], and Weaver et al. [ ]. Zahodne et al. also noted that neither motor nor mood scores differed by these two targets, but the GPi target did demonstrate greater improvements in the subscale ratings related to mobility, activities of daily living (ADL), stigma, and social support. This relatively new claim warrants further research because the impact on patient quality of life is significant. With the negative impact of the disease on patient's cognition, Heo et al. believed that bilateral STN stimulation might lead to slightly more detrimental effects on frontal lobe function and memory [ ]. The authors inferred that improved outcomes with cognition might be obtained via GPi or unilateral targeting. The full impact of GPi versus STN targeting though is still an area of controversy, especially with respect to overall patient outcomes.

Results in quality-of-life studies have varied regarding the effects of surgical treatment in patients with movement disorders. And, the numbers of studies are sparse. Outcomes for movement disorders, particularly PD, have used the Unified Parkinson's Disease Rating Scale (UPDRS) part III, a standardized scale that primarily demonstrates the motor benefits subsequent to surgical intervention. Such measures can in part reflect subjective improvements from the point of view of the patient: for example, regaining motor dexterity and control (e.g., holding a cup of coffee without fear of spilling it). Admittedly, the patient's relief in these circumstances is subjective, difficult to measure and appreciate. In a long-term follow-up study of STN-DBS, Krack et al. found that there were significant improvements in the postoperative UPDRS, specifically ratings were 59% lower at 3 months and 54% lower at 5 years [ ]. In this same study, the authors reported improvement in ADL functions by 49% compared with baseline functioning; they furthermore noted that, before surgery, most patients had depended on others to some degree but after surgery nearly all enjoyed independence throughout the entire follow-up period.

In a similar study assessing the psychological impact of DBS, Schupbach et al. demonstrated comparable results, that is, the UPDRS ratings showed 54% improvement at 5 years [ ]. The authors noted that 10 of 20 patients were able to withdraw completely from all parkinsonian medications. Although the measures of neuropsychological and mood assessments were unchanged by the surgical procedure, cognitive decline was marked. However, this decline was attributed to the natural progression of the PD rather than the intervention performed. Another important aspect of this study demonstrated that daily dosing of levodopa was reduced by 58% and very likely significantly impacted the patient's perception of the disease. When medication regimens are simplified [ ], patients can not only better control their disease symptoms but also notice significant impact on quality of life and cost-effectiveness.

Saatci et al. reported STN-DBS improved olfactory dysfunction in 45 PD patients within 3 months post-op, as well as motor symptoms [ ]. Hacker et al. suggest the possibility that DBS in early PD may slow rest tremor progression; UPDRS-III “off” rest tremor score change from baseline to 24 months was worse in patients receiving ODT (optimal drug therapy) versus DBS + ODT ( P = .002). Rest tremor slopes from baseline to 24 months favored DBS + ODT both “off” and “on” therapy ( P < .001, P = .003, respectively). More ODT patients developed new rest tremor in previously unaffected limbs than those receiving DBS + ODT ( P = .001) [ ].

Dafsari et al. found that the location of the electrode contacts have impact on outcomes after STN-DBS, more anterior, medial, and ventral STN-DBS is significantly related to more beneficial nonmotor outcomes in PD patients [ ]. There is also a question whether the outcome of the patient is better if the DBS surgery is done under local anesthesia or if it is done under general anesthesia, thus Holewijn et al. [ ] and Fluchere et al. [ ] found that there is no difference in the short term or long term whether the intervention is performed under local or general anesthesia.

Dembek et al. reported in 10 patients that stimulation in the individual best direction resulted in significantly larger therapeutic windows, higher side-effect thresholds, and more improvement in hand rotation than using circular DBS leads. There was no difference in motor efficacy or stimulation amplitudes between directional and circular DBS in the short-term crossover. Follow-up evaluations 3–6 months after implantation revealed improvements in motor outcome and medication reduction comparable to other DBS studies [ ].

Witte et al. reported that urinary incontinence and frequency improved after both GPi-DBS and STN-DBS at 12 months postoperatively, but this was only statistically significant for the STN-DBS group ( P = .004). The improvements after DBS were present in both men ( P = .01) and women ( P = .05). Nocturia and night-time incontinence did not improve significantly after any type of DBS, irrespective of sex. At 12 months, none of the patients had a Foley catheter [ ].

Essential tremor

Neurostimulation for tremor can significantly benefit a patient's quality of life. Even while undergoing intraoperative testing, patients may feel tearfully happy about the prospect of better motor function. As discussed in other chapters, Vim or STN targeting is most often used in neurostimulation for essential tremor. Many of the research findings regarding tremor are included with information for PD. Sydow et al. documented long-term relief of tremor 6 years after surgery in the 19 of 37 patients available for follow-up; significant reduction in tremor score and improvement in ADL were found compared with baseline or in the stimulation-off mode [ ]. Zhang et al. cited an 80.4% reduction in tremor and 69.7% improvement in handwriting in 34 patients with an average 56-month follow-up [ ]. Interestingly, between 57-month and 90-month follow-up, no statistical difference was found in functional ability when evaluating tremor and handwriting. Subtle adjustments in programming, primarily increases in voltage, were needed by many patients during the 5-year follow-up. At 7 years postoperatively, Hariz et al. noted decreases in the efficacy of DBS for tremor; however, a notable positive impact on quality of life and ADL functioning remained [ ], especially for the patient's ability to eat and concerns with social life. The authors stated the more significant declines in the effects of DBS in most other areas, which began 6–8 years postoperatively, were likely due to aging, aging comorbidities, and disease progression. At this endpoint, tremors significantly worsened when stimulation was turned off. Patients with tremors who had been unable to write their names or perform other common tasks found significant rewards with the renewed ability to function in ways often taken for granted by others.

Mehanna et al. reported five patients who failed to significantly improve with a single implanted lead, and thus those five patients underwent a second lead implantation to treat tremor (two patients with tremor secondary to multiple sclerosis, and three patients due to ET). There was improvement with the second lead or double ON in four patients compared to stimulating the Vim alone. Four patients had 17%–60% tremor improvement after the implant of the second lead on double-blinded evaluation [ ].

Blomstedt et al. reported four patients with ET, implanted with two ipsilateral electrodes, one in the STN and one in the cZI (caudal zona incerta). STN and cZi both proved to be potent targets for DBS in ET. DBS in the cZi was more efficient, since the same degree of tremor reduction could be achieved at lower energy consumption. Three patients became tremor-free in the treated hand with either STN of cZi, while the fourth had a minor residual tremor after stimulation in either target [ ].

Dystonia

Stimulation of the GPi is the most studied target for dystonia, both primary and secondary. In primary generalized dystonia (PGD), the mean improvement at 3-month to 12-month follow-up ranged from 46% at 3 months and 80% at 12 months follow-up in the Burke–Fahn–Marsden Dystonia Rating Scale (BFMDRS) severity score, and 37% at 3 months, and 69% at 12-month follow-up in the BFMDRS disability score [ ]. At 2-year follow-up, mean improvement ranged from 34% to 82% in the BFMDRS severity scores and 32%–75% in BFMDRS disability scores [ , , , ]. At 3-year follow-up, the motor improvement and quality of life (SF-36 questionnaire) observed at 1 year had been maintained [ ]. Even, the improvement is sustained at 5-year follow-up [ ]. Significant benefits of this therapy were evidenced by the improvements in general health and physical functioning at 12-month follow-up [ ] and a 15% improvement in the unified dystonia rating scale (UDRS) which is statistically significant [ ].

Reese et al. reported a multicenter case series of six patients (six women and six men) with Meige syndrome, followed up to 78 months after bilateral GPi-DBS, with mean disease duration of 8.3 ± 4.4 years. Dystonia severity in BFMDRS showed a mean improvement of 45% at short-term follow-up (4.4 ± 1.5 months; P < .001) and of 53% at long-term follow-up (38.8 ± 21.7 months, P < .001). Subscores for eyes were improved by 38% ( P = .004) and 47% ( P < .001), for mouth by 50% ( P < .001) and 56% ( P < .001), and for speech/swallowing by 44% ( P = .058) and 64% ( P = .004). Mean improvements were 25% ( P = .006) and 38% ( P < .001) on the Blepharospasm Movement Scale and 44% ( P < .001) and 49% ( P < .001) on the Abnormal Involuntary Movement Scale (AIMS) [ ].

In children, the improvement reported at 6-month follow-up was as high as 56% in the BFMDRS motor scores and 42% in the BFMDRS disability scores [ ]. Compared with adults, the better outcomes in children and adolescents were associated with DYT1-positive genetic status and with less motor impairment before surgery [ , ].

Regarding the neuropsychological outcomes, there have been no reported significant changes in measures of mood and cognition after pallidal stimulation in dystonia patients in short- and long-term follow-up [ , , , ]. Some authors reported no significant reduction in the number of errors in the Wisconsin Card Sorting Test (WCST) at 1-year follow-up [ ]. Others showed that bilateral GPi-DBS clearly improved functional abilities and quality of life [ ], and noted some improvements in concept formation, reasoning, and executive functions [ ].

Contact location greatly impacted outcomes: overall clinical improvements were as high as 89% with posteroventral contacts versus only 67% with antero-dorsal contacts in the pallidum [ , ]. There is also a chance of poor outcome because of lead misplacement. In addressing this topic, Ellis et al. reported 12.8% improvement above the already-obtained improvement in the UDRS score after lead relocation [ ].

Factors that predict poor outcome in generalized dystonia seem related to a greater disability from symptoms (a high preoperative BFMDRS score) and long disease duration [ , ]. There was greater improvement in children with the genetic form DYT1-positive than in children with non-DYT1 forms. Markun et al. have shown that the shorter disease duration in young-onset DYT1 dystonia patients the better long-term bilateral GPi-DBS outcome [ ]. The volume of the GPi stimulated also influences the outcomes, the greater the GPi volume, the greater the degree of improvement [ ].

For primary focal and segmental dystonia, improvements in the BFMDRS score are in the order of 64% at 3 months and 75% at 1 year 75% [ ]. At 2-year follow-up, ratings on the Toronto Western Spasmodic Torticollis Rating Scale (TWSTRS) had improved nearly 60% for both disability and pain scores [ , ]. In patients with cervical dystonia, there was a 43% improvement in the TWSTRS severity score and a 59% improvement when both the disability and pain scores were combined [ ]. A pilot study of bilateral pallidal stimulation in idiopathic cranial–cervical dystonia, BFMDRS motor and disability scores improved 72% and 38% at 6 months, respectively. Total TWSTRS scores improved 54% at 6-month follow-up. Although the combined severity and disability subscores (BFMDRS) showed statistical improvement, the pain subscore only showed a trend toward improvement and was not statistically significant [ ]. General health and physical functioning and depression scores improved significantly. Some negative changes in neuropsychological tests (memory and verbal skills) were observed, but did not impact daily life or employment [ ]. For secondary dystonia (i.e., dystonias due to brain injuries), the reported outcomes of the GPi-DBS were less promising than for other dystonias [ ]. In a recent study that seems to refute some of these findings, Loher et al. reported almost the same outcomes for both primary and secondary dystonias [ ]. In tardive dystonia at 3–6 months after surgery, there were improvements of 74% in BFMDRS-M score, 89% in BFMDRS-D score, and 70% in AIMS. In another study, quality-of-life improvements were significant in physical components and affective states [ ].

In a mixed group of secondary dystonia patients (i.e., myoclonic dystonia, tardive dystonia, posttraumatic hemidystonia) who underwent treatment with bilateral GPi stimulation, improvements in the AIMS score ranged widely from 0% to 73.9%; the patient with no improvement had posttraumatic hemidystonia that temporarily improved after surgery but returned to the baseline findings some days after the DBS was turned on [ ]. Some authors suggest that among the secondary dystonias, the drug-induced forms have potentially better outcomes compared with the secondary dystonias. That is, the BFMDRS severity scores improved 47.2% for the drug-induced group and 37% for the other mixed dystonias, and the BFMDRS disability scores were 54.6% and 34.4%, respectively [ ]. In patients with secondary dystonia, it is important to note that anatomical preservation of the basal ganglia is related to surgical outcome [ ].

Previous reports of the usefulness of thalamic DBS to control dystonia have shown questionable results. Since then, GPi stimulation has gained greater acceptance in the treatment of this syndrome [ , ]. In a study of bilateral anterior dorsolateral STN stimulation in patients with predominantly cervical dystonia, significant improvement occurred in the motor, disability, and total TWSTRS scores. Outcomes were better in those who did not have fixed deformities. The mental component score of the SF-36 markedly improved, and neuropsychological function was not negatively affected as a result of surgery. However, there were no differences in the TWSTRS scores between stimulation-on and stimulation-off for the group as a whole [ ]. In another study of patients with writer's cramp who underwent unilateral ventral oralis anterior/ventral intermediate (Voa/Vim) stimulation (one patient underwent both GPi and Voa/Vim DBS), Fukaya et al. showed that BFMDRS scores improved 87.5% when the stimulator was turned on; this improvement was maintained at 2-year follow-up. In the patient with dual DBS targets, the thalamic stimulation was superior to the pallidal stimulation [ ]; however, superior results were obtained with pallidal stimulation in a previously Vim-DBS implanted patient with paroxysmal nonkinesiogenic dystonia [ ]. These last two studies showed some interesting results, but need further study.

In our review of the literature, we found no consensus on programming settings for either primary or secondary dystonia. Usually GPi-DBS required more amplitude than what is required with GPi or STN-DBS for PD, with wide ranges of pulsewidth and frequency settings for all groups [ , , , ]. Some authors suggested that the pulsewidth should exceed 180 ms (in dystonias), and the rate should be between 130 and 185 Hz (high-frequency stimulation) [ ]. Other studies have reported significant improvement with low-frequency stimulation (50–60 Hz) [ ]. In a study focused on the frequency of pallidal stimulation in primary dystonia, optimized stimulation at 130 Hz resulted in a 43% improvement in the BFMDRS score 6–12 months post-surgery. Quality of life measured through PDQ-39, EuroQoL1, and EuroQoL had significantly improved after surgery when measured in all of the scales. However, in this same study, a significant deterioration was observed at lower frequencies (0, 5, 50 Hz) in all patients [ ].

Recently, STN has been studied as a target for different types of dystonia patients, with promising results that seem to surpass GPi-DBS outcomes in dystonic patients [ , ]. Ostrem et al. reported that high-frequency (130 Hz) stimulation is more effective than low-frequency (60 Hz) stimulation in STN-DBS for primary dystonia patients. Patients had 52.3% ( P = .018) and 45.2% ( P = .028) in BFMDRS-M and TWSTRS-S, respectively, at 6 months, and at 12 months, the BFMDRS-M improved 51.8% ( P = .022) and 56% ( P = .034) in TWSTRS-S [ ].

After DBS, mobile, phasic dystonic movements respond rapidly and are predictors of good outcome, whereas fixed postures are less likely to improve and are predictors of poor outcome at 12-month follow-up, mostly due to muscle contractures [ ]. Although tonic components tend gradually to improve, some patients experience rapid improvement shortly after the DBS. Long-lasting benefits were not observed until 6–12 months later in most patients. The presence of microlesion effect immediately after pallidal DBS for dystonia also appears to be a good predictor of optimal clinical outcome, though this remains controversial [ ].

Gruber et al. showed that GPi-DBS is also a valid option for patients with tardive dystonia/dyskinesia, this RCT found that at 3 months postrandomization dystonia severity improved significantly within the neurostimulation group by 22.8% and nonsignificantly within the sham group (12.0%) compared to their respective baseline severity [ ]. In 2014, a randomized, sham-controlled trial of pallidal neurostimulation in 62 patients with medication-refractory cervical dystonia, reported at 3 months, the reduction in dystonia severity was significantly greater with neurostimulation (−5.1 points [SD 5.1], 95% CI −7.0 to −3.5) than with sham stimulation (−1.3 [2.4], −2.2 to −0.4, P = .0024; mean between-group difference 3.8 points, 1.8 to 5.8) in the intention-to-treat population. A total of 21 adverse events (five serious) were reported in 11 (34%) of 32 patients in the neurostimulation group compared with 20 (11 serious) in nine (30%) of 30 patients in the sham-stimulation group. Serious adverse events were typically related to the implant procedure of the implanted device, and 11 of 16 resolved without sequelae. Dysarthria (in four patients assigned to neurostimulation vs. three patients assigned to sham stimulation), involuntary movements (i.e., dyskinesia or worsening of dystonia; five vs. one), and depression (one vs. two) were the most common nonserious adverse events reported during the course of the study [ ].

Regarding the surgical technique, there are different techniques to be used. Classically frame-based DBS technique is the most used worldwide, recently frameless DBS is also used, and most recently robot-assisted DBS seems to be a promising technique with good results [ ]. Starr et al. reported six pediatric patients with primary dystonia (five DYT1 mutation-positive), in whom implanted DBS bilateral leads in GPi (5 patients) and STN (1 patient) through an interventional MRI (iMRI)-guided procedure using general anesthesia without physiological testing. Starr et al. reported a single brain penetration in all the 12 leads, with an error in the actual implanted lead. The planned trajectory of mean difference was 0.6 ± 0.5 mm, and the mean surgical time (leads only) 190 ± 26 min. The mean percent improvement in the BFMDRS movement scores was 86.1% ± 12.5% at 6 months (n = 6, P = .028) and 87.6% ± 19.2% at 12 months ( P = .028). The mean stimulation settings at 12 months were 3.0 V, 83 μs, 135 Hz for GPi-DBS, and 2.1 V, 60 μs, 145 Hz for STN-DBS) with no serious adverse events [ ].

Spinal cord stimulation for pain

As technology developed for electrical cardiac stimulation, these same principles were applied to stimulation of the nervous system, namely, the spinal cord. This new technology found support from the gate theory of pain, which helps explain how the nervous system is affected in pain syndromes [ ]. Although still somewhat controversial, the gate theory premise is the inhibition of small, unmyelinated pain fibers by the activation of large sensory nerve fibers. The spinal cord stimulator, placed over the dorsal columns of the spinal cord at approximately the mid-thoracic level, activates these large fiber neurons thereby to inhibit or diminish pain sensation. Good outcomes were reported with at least 50% reduction in pain; satisfaction was achieved in 47% of study participants 5 years after surgery [ ]; 25% of patients returned to work after the implant. In the same study, many patients reduced or eliminated analgesics for pain and noted improvements in ADL.

In the 2008 PROCESS study, Manca et al. noted marked improvements in quality of life for patients with spinal cord stimulators [ ]. While overall costs to the health-care system were higher, improvements in quality of life were seen by using the short-form (SF-36) and EuroQol-5D (EQ-5D) [ ]. The EQ-5D consists of five questions, each relating to a different dimension that included mobility, self-care, and ability to undertake usual activity, pain/discomfort, and anxiety/depression. Each dimension has three possible levels of severity described as none or moderate or severe problems. Based on their combined answers to the EQ-5D questionnaire, patients can be classified into one of 243 health states. Each health state has an associated utility score on a 0 (death) to 1 (good health) scale. At mean baseline of EQ-5D, the PROCESS study participant scores were 0.15, which was considerably worse than the 0.31 for patients admitted to the hospital with ischemic strokes [ , ]. Considering the significant toll on the quality of life for patients with failed back syndrome, the potential for improvement in quality of life of these patients should not be underestimated.

Multiple studies have demonstrated superiority of SCS in comparison with other types of management. In comparison with optimal medical management, Kumar et al. demonstrated superior outcomes after SCS for failed back syndrome [ ]. They reported that many patients not only found greater relief of neuropathic pain by implantation of SCS than conventional therapy, but also showed a significant increase in the quality of life. Kumar et al. reported that 24 months after implantation, patients with SCS had greater satisfaction with treatment, and improved functional capacity and health-related quality of life [ ]. Remarkably, 30% of patients returned to work after SCS therapy, including four of 37 patients who had been out of work for more than 2.5 years. In a 2007 comparison of reoperation versus SCS for failed back syndrome, North et al. showed that SCS insertion was more effective and less expensive than re-operation for patients with failed back syndrome who had a previous surgery [ ]. SCS was found to be most effective when patients avoided repeat surgery. Additionally, costs of SCS for patients with failed back syndrome were significantly less than for repeat spinal surgery. Both factors play major roles in a patient's quality of life.

Recently, a paresthesia-free high-frequency SCS (HF10 SCS) has been proved to have better clinical outcomes in patients with chronic intractable low back pain and legs. Al-Kaisy et al. in a prospective, multicenter, observational study reported 88% of successful during the trialing period, thus 88% underwent permanent implantation of the system. Mean back pain was reduced from 8.4 ± 0.1 at baseline to 3.3 ± 0.3 at 24 months ( P < .001). Concomitantly to the pain relief, there were significant decreases in opioid use, Oswestry Disability Index score, and sleep disturbances. Patient satisfaction and recommendation ratings were high. Adverse events were similar in type and frequency to those observed with traditional SCS systems [ , ]. Annemans et al. demonstrated that HF10 SCS is cost-effective and provides a greater number of quality adjusted life years (QALY) compared to conventional medical management, reoperation, traditional nonrechargeable SCS and rechargeable SCS in failed back surgery syndrome patients [ ].

In a randomized controlled trial in patients with chronic back and leg pain, Kapural et al. showed at 3 months, 84.5% of implanted HF10 therapy patients were responders for back pain and 83.1% for leg pain, and 43.8% of traditional SCS patients were responders for back pain and 55.5% for leg pain ( P < .001 for both back and leg pain comparisons). The relative ratio for responders was 1.9 (95% CI, 1.4 to 2.5) for back pain and 1.5 (95% CI, 1.2 to 1.9) for leg pain. The superiority of HF10 therapy over traditional SCS for leg and back pain was sustained through 12 months ( P < .001). HF10 therapy patients did not experience paresthesias [ ]. In a later report, Kapural et al. reported a 24-month follow-up from a multicenter, randomized, controlled pivotal trial, and demonstrated long-term superiority of HF10 therapy compared with traditional SCS in treating both back and leg pain [ ].

Arle et al. reported that at clinical HFS frequencies and pulse widths, HFS preferentially blocks larger-diameter fibers, and concomitantly recruits medium and smaller fibers. These effects are a result of interaction between ion gate dynamics and the “activating function” (AF) deriving from current distribution over the axon. The larger fibers that cause paresthesia in low-frequency stimulation are blocked, while medium and smaller fibers are recruited, leading to paresthesia-free neuropathic pain relief by inhibiting wide dynamic range cells [ ].

Also, HF20 cervical SCS has been reported to be safe and efficacious in patients suffering from headache and migraines [ , ], and other clinical entities such as complex regional pain syndrome [ , ], urinary incontinence in patients suffering from neurological disease or spinal cord injury [ ], and neck and upper limb pain [ ].