Habituation After Deep Brain Stimulation in Tremor Syndromes: Prevalence, Risk Factors and Long-Term Outcomes 重试    错误原因

Abstract 重试    错误原因

Introduction 重试    错误原因

Tremor is an involuntary, rhythmic, oscillatory movement of a body part (1), with an estimated prevalence of 14.5% in the general adult population (2). The spectrum of tremor extends from the enhanced physiological postural tremor, often only noticeable during states of fatigue and heightened anxiety, to persistent pathological syndromes including essential tremor (ET), where currently available medications are moderately effective at best. For the severe end of the spectrum of tremor syndromes, functional neurosurgical techniques and neuroradiological procedures exists. These have evolved since Cooper (3) made the unintentional observation nearly 70 years ago, that destruction of a portion of the globus pallidus suppressed tremor of a patient with Parkinson's Disease (PD). Stereotactic lesional surgery mainly targeting the ventrolateral thalamus and posterior subthalamic white matter was used since the 1950's. During this period it was discovered that intraoperative high frequency electrical stimulation would suppress tremor and was used to verify the target region prior to thermal abalation (4, 5). These pioneering stereotactic interventions paved the way for the first cases of deep brain stimulation (DBS) performed by Cooper et al. (6). Motivated by a desire to avoid the frequent dysarthria observed following bilateral radiofrequency thalamotomy, DBS of the thalamic ventral intermediate nucleus (VIM) was revisited by Benabid in 1987 for second-side treatment of tremor and popularized following their 1991 publication of a series of 26 PD and six essential tremor (ET) patients treated with VIM DBS who reported to have a sustained tremor response over a 13-month median follow-up period (7). Subsequently large studies confirmed the effectiveness of VIM DBS for ET and PD (8, 9). 重试    错误原因

DBS remains the most common surgical procedure for medication-refractory tremor. However, the long-term benefits of therapy, particularly in ET, are often observed to wane over time, in a variable, unpredictable pattern. A phenomenon of “tolerance” was first described by Benabid et al. in a series of 80 tremor-dominant PD and 20 ET patients, with either uni- or bi-lateral VIM stimulation (10). Regular increase in stimulation to alleviate tremor was required to a final threshold that could no longer be increased due to the induction of side effects. “Tolerance” was associated with eventual loss of functional benefit and was more commonly observed with those with action tremor, in severe syndromes, with higher stimulation intensity, and where continuous 24-h stimulation had been adopted (10). The phenomenon of “tolerance” has continued to be observed in clinical practice and is now usually referred to as “habituation” (11). Attempts to characterize and quantity habituation have revealed conflicting reports in the medical literature and remain the subject of debate. 重试    错误原因

In this narrative review article, the available evidence for habituation after DBS for tremor syndromes is reviewed to reappraise perceptions of expected long term outcomes and factors that may assist in predicting this phenomenon. We also provide some information on possible pathophysiological mechanisms underlying ‘habituation' and approaches to its management. 重试    错误原因

Habituation Definition 重试    错误原因

Habituation in the context of benefit from DBS was first mentioned in the medical literature by Benabid et al. and described as “tolerance”. It was hypothesized a progressively decreased biological response (habituation) of the neuronal network to be a possible mechanism for the phenomena of “tolerance” (10). Recently the term habituation has been proposed to replace “tolerance”, defined by Fasano and Helmich to be the rapid vanishing of DBS efficacy after programming (11). This definition of habituation can be expanded to include delayed, progressive loss of therapeutic benefit for tremor after DBS, in line with the original concept of “tolerance” due to “decreased biological response (habituation) of the neuronal network” as described by Benabid et al. (10). 重试    错误原因

Authors have attempted to study habituation in the context of progressive loss of DBS benefit with particular attention given to differentiating progression from the natural history of disease. We agree in theory that comparing the tremor severity in the “off” state at two different time points, after allowing for rebound, represent disease progression; whereas tremor severity in the “on” state is determined by both disease progression and the stimulation effect. The difference (delta) between the on-off state, when compared over time, has been assumed by authors to be a measure of changing stimulation over time and attributed to habituation (12, 13). This is based on the premise that over time, other variables, specifically lead location and optimization of programming remain constant; but further, alternative mechanisms are not contributing or causing the phenomena that has been labeled habituation. Given these provisions, we will proceed on the operational hypothesis, reflected by the change in delta over time, from the definition of habituation known previously as “tolerance”; to be the loss of benefit from electrode reprogramming over time in the setting of optimal electrode placement and programming not explained by disease progression of the tremor syndrome. Habituation should not be explained by loss of micro-lesional implant effect or expected progression due to the natural history of the tremor syndrome. In line with the concept of “rapid vanishing of effect” habituation also refers to temporary improvement in tremor severity following increasing electrical field strength or contact adjustment, followed by subsequent paradoxical worsening. 重试    错误原因

Although this definition is useful conceptually, determining if an individual patient is experiencing habituation after tremor DBS remains very difficult because of the following; Firstly, there is no absolute agreed definition of what constitutes optimal lead placement; more troubling though, is the fact not all DBS leads placed within the optimal 2 mm radius of the intended target have a concordant clinical response (14). Secondly, optimal DBS programming is highly operator dependent as evidenced by significant clinical improvements achieved after expert reprogramming (15). Lastly, progression of the underlying tremor syndrome as part of the natural history of the disease must be subtracted from any apportionment of habituation, in itself a very difficult distinction, highlighting the inherent complexity and uncertainties surrounding this topic. 重试    错误原因

Does “Habituation” Really Exist? 重试    错误原因

Loss of Benefit Over Time 重试    错误原因

Habituation has most commonly been associated with ET, possibly reflecting the experience of clinicians in practice. Over 20 studies have been published looking at the long-term clinical efficacy of DBS in this condition; most commonly involving uni-or bilateral VIM stimulation (12, 13, 15–32). When looking at studies with a greater than 3-year follow-up, the long-term effect compared to baseline, ranges from 31.2–88.4% improvement (11, 13). The less traditional target, posterior subthalamic area (PSA)/caudal zona incerta (cZi), in comparison has relatively few follow-up studies; but with a similar range of effect size from baseline: 33–76% improvement (30, 33, 34). Some studies have suggested that the PSA/cZI may be less prone to habituation; however, no superiority has ever been clearly established (33–35). Despite this persistent improvement from DBS in the long-term, the majority of studies have shown that the effect diminishes over time. 重试    错误原因

A recent systematic review and meta-analysis by Lu et al., included 26 studies with 439 patients, looked at potential outcome predictors following VIM DBS in ET. The pooled treatment effect was 60.3% improvement in objective Tremor Rating Scale (TRS) scores at 20 months (+/– 17.3). Correlation with outcome was seen only with pre-operative TRS scores and follow-up time; both negatively correlated with the clinical outcome (36). It had previously been reported that pre-operative cerebellar dysfunction was a risk factor for the development of early “tolerance” (37). Natural disease progression and habituation have been proposed as the most plausible factors contributing to VIM-DBS treatment declining overtime (17, 23, 27, 31, 33). Despite the absence of consensus guide lines, electrode placement beyond a 2–3 mm radius of an intended target have been associated with suboptimal tremor control and can be a correctable cause of DBS “failure” (37, 38). Further, in cases of suboptimal clinical benefit, DBS lead adjustment of only a few millimeters can have a meaningful benefit (39). Other possible co-contributing factors include incorrect pre-operative diagnosis (14), loss of microthalamotomy (9) and increased impedance of brain tissue over time (10) (Figure 1). However, effects of varying tissue impedance are minimized by constant current DBS systems now more widely used. 重试    错误原因

Open in a separate window

Figure 1 重试    错误原因

Factors contributing to decline in clinical benefit of tremor after DBS. 重试    错误原因