Scolaris Content Display Scolaris Content Display

Deep brain stimulation of subthalamic nucleus for Parkinson's disease

Esta versión no es la más reciente

Contraer todo Desplegar todo

Abstract

This is a protocol for a Cochrane Review (Intervention). The objectives are as follows:

We assessed the efficacy and safety profile of STN DBS surgery compared to the best available medical treatment or other deep brain surgery procedures in patients with idiopathic PD.

Background

Parkinson's Disease (PD) is a progressive, disabling, degenerative neuronal disorder due to death of cells in the substantia nigra and other brain regions. This disease affects 1% of the population over the age of 65 and is clinically characterised by bradykinesia, tremors, rigidity, and postural instability. Despite new insights into its pathogenesis, there is no definitive cure for PD. Levodopa continues to be the most effective drug for symptomatic treatment (Jankovic 2000) and failure to respond is often used as evidence against the diagnosis of PD.

Drugs other than levodopa, such as dopamine agonists, can be used as monotherapy to improve symptoms in early PD and delay levodopa treatment, but polytherapy, including apomorphine infusion, is often necessary in advanced stages. Unfortunately pharmacological treatment becomes unsatisfactory in a large proportion of patients. Within five years of levodopa treatment, about 40% of patients experience motor fluctuations (e.g. wearing off and on‐off phenomenon) and dyskinesias (Ahlskog 2001a, Nutt 1990) (IntPeerRev_A5). These cause functional disability, are difficult to manage with available drug strategies, and have a major impact on the patient's quality of life.

Surgical treatment of PD was described as early as 1940 (Bronstein 2011). In PD animal models neuronal activity is increased in the STN or in the GPi. Lesion of these nuclei resulted in marked improvement of motor function (Wichmann 1994, Benazzouz 1995). First human studies reported similar effects but used an ablative surgical approach (e.g. pallidotomy) which resulted in destructive brain lesions related to an increased risk of permanent neurological deficits (Hariz 2000).

In the late 1980s there was a renewal of interest in neurosurgery of basal nuclei facilitated by new technologies. First we were able to better identify brain anatomical dysfunctional sites taking advantage of magnetic resonance and modern electrophysiology. Second we moved to a functional surgical approach which is based on the implantation of electrodes into the brain attached to a pacemaker which sends electrical impulses to the brain tissue. This approach is called deep brain stimulation (DBS) and offered advantages than ablative surgical treatment because it is reversible, selective and programmable (controlling frequency and strength of the stimulation). DBS was approved by Food and Drug Administration (FDA) in 1997 as treatment for essential tremor and in 2006 for PD (U.S. Department of Health and Human Services).

Most common sites that can be targeted by deep brain functional neurosurgery are: the subthalamic nucleus (STN), the globus pallidus interna (GPi) and the thalamus. STN DBS surgery was described as highly promising in many small and non‐controlled trials, in terms of motor improvement and reduction of levodopa adverse events (Kumar 1998, Limousin 1998).

GPi was used to relieve tremor, and the effect tend to be long lasting. Rigidity or stiffness may be helped somewhat in 40% of patients but the major disabling problems associated with bradykinesia or walking do not really respond at all. Significant complications of thalamic and pallidal lesions such as weakness on the opposite side of the body, muscular contractions, slurred speech, confusion, eye deviations, mental changes (confusion), infections, or even increased tremor on the same side of the body as the lesion (Duvoisin 2001) limited their diffusion.

A meta‐analysis published in 2006 investigated the effect of PD surgery in randomised and observational studies, showing a significant improvement in activities of daily living (ADL) and motor scores in the surgical compared to the medical group (Kleiner‐Fisman 2006), The primary studies reported variable levels of neurological improvement and different rates of severe complications. Psychiatric sequelae were reported as common in the surgical group.(IntPeerRev_A4)

More recently other reviews (Bronstein 2011, Benaibid 2009) recommended timely referral of potential STN DBS surgical candidates ‐ patients with PD who have developed severe motor complications refractory to the available pharmacological interventions ‐ to a neurosurgery centre for evaluation.

Although not curative, DBS can help manage some of its symptoms and subsequently improve the patient’s quality of life. Major questions that clinicians and patients have during their clinical decision making regard the risk/benefit profile of surgery compared to medical treatment, selection of the target population (early or advanced PD stages), and the best surgical target and approach (Lang 2002). It is important to conduct a systematic search to point out the effects of DBS surgery in experimental settings.

Objectives

We assessed the efficacy and safety profile of STN DBS surgery compared to the best available medical treatment or other deep brain surgery procedures in patients with idiopathic PD.

Methods

Criteria for considering studies for this review

Types of studies

Randomised controlled trials (RCTs) and non‐randomised controlled trials (nRCTs) with prospective design were included. We did not include studies in which patients were their own controls (before/after design and on/off stimulation studies). Uncontrolled observational trials were also excluded.

Types of participants

Patients with a clinical diagnosis of idiopathic PD and the indication for functional deep brain surgery. Any duration and severity of symptoms and motor complications were considered. All ages were included.

Types of interventions

STN DBS surgery was compared with the best pharmacological treatment available or other deep brain functional surgery (subthalamic nucleus lesion, internal globus pallidus and thalamus ventro‐intermediate nucleus stimulation).

Types of outcome measures

We included studies reporting data on outcomes related to efficacy or safety of STN DBS. We included the following outcome measures.

Primary outcomes

Efficacy

  1. improvement of disability (rated with Unified Parkinson's Disease Rating Scale ‐UPDRS‐ part II or equivalent scales);

  2. improvement of motor function (rated with UPDRS part III or equivalent scales);

  3. reduction of motor complications (rated with UPDRS part IV for dyskinesias and motor fluctuations, or equivalent scales). (ExtPeerRev_B6)

Secondary outcomes

Efficacy

  1. reduction of total levodopa‐equivalent daily dosage;

  2. improvement of quality‐of‐life (rated with EuroQol EQ‐5D, or equivalent scales).

The efficacy outcomes were evaluated at the last follow‐up, which had to be at least three months long.

Safety

  1. complications of surgery (mortality, haemorrhage, infarction, infection, seizure);

  2. complications of stimulation (dysarthria and hypophonia, diplopia, paraesthesias, tonic contraction, confusion, hallucination, maniac/depressive episodes, cognitive disorders, weight gain).

Data on working days lost, caregivers' time, number of hospital admission days in hospital and cost events during the follow‐up were also collected, when available.

Search methods for identification of studies

We applied the criteria described in the Cochrane Handbook for Systematic Review (Higgins 2011) and running the Collaborative Movement Disorders Review Group specific strategy for identification of studies about PD. Details of search strategies are provided in Appendix 1. We adapted the search strategy for all database. The search is not restricted to any languages.(IntPeerRev_A10)

Electronic searches

Relevant trials were identified searching in the following database:

  1. MedLine (1966‐July 2013):

  2. Embase (1988‐July 2013): we adapted the search strategy to that outlined above.

  3. Cochrane Central Register of Controlled Trials (Issue 7, 2013)

  4. clinical trials databases to find on‐going studies (Clinicaltrials.gov)

Searching other resources

Additional strategies for identifying trials included searching the reference lists of review articles and included studies, searching the main congress' abstract books, consulting experts in the field of Parkinson disease, contacting the authors of trial reports, Enquiring of companies that sell deep brain stimulation electrodes and drug manufacturers of main dopaminergic agonists.

Data collection and analysis

Selection of studies

Three review authors (CM, LC and VP) independently screened titles and abstracts from the search and judged whether trials fit the inclusion criteria. Disagreements were resolved by consensus. All potentially relevant articles were retrieved for the assessment of the full publication. The reasons for exclusion were documented.

Data extraction and management

Three authors (CM, LC and VP) independently extracted information about:

  1. study characteristics: design, duration, randomisation, blinding, withdrawals;

  2. characteristics of participants: mean age, sex, type of Parkinson disease duration of disease;

  3. interventions: type of surgery, levodopa dosage and other drugs

  4. outcomes: disability, motor function, motor complication, l‐dopa equivalent, quality of life, mortality

  5. severe psychiatric adverse events

When a trial reported pre‐ and post‐treatment scores, we calculated the mean change from baseline and SD. When only post‐treatment data were available, we analysed them.

Assessment of risk of bias in included studies

Three review authors (CM, LC, VP) independently examined the studies following the criteria specified in using the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). Disagreements were resolved by discussion among the reviewers. We considered these items as the domains for the risk of bias assessment: sequence generation, allocation concealment, blinded assessment of outcome, incomplete data outcome, intention‐to‐treat (ITT) analysis, similar baseline characteristics and competing of interest.

Adequate sequence generation and allocation concealment arrange the selection bias. We assessed sequence generation to be at low risk when studies clearly specified a method for generating a truly random sequence. We assessed allocation concealment to be at low risk if the method used to ensure that investigators enrolling participants could not predict group assignment was described.

Detection bias were incorporated under the domain "blinding". We assessed this to be low risk for studies that reported blinding of outcome assessor.

We assessed studies as low risk for attrition bias if an adequate description of participant flow through the study was provided, the proportion of missing outcome data was relatively balanced between groups and the reasons for missing outcome data were provided.

We also assessed studies to low risk of bias if the authors performed the ITT analysis, and if the characteristics of participants at baseline ar similar to avoid imbalance between groups.

Finally, we judged low risk of bias if the authors declared to have or not to have conflicts of interest.

Measures of treatment effect

In order to assess efficacy, raw data for outcomes of interest were extracted where available in the published reports. We expressed continuous data as mean difference (MD) or standardized mean difference (SMD), where different scales were used . We reported change‐from‐baseline to follow‐up and follow‐up scores in separate meta‐analyses. For dichotomous data we calculated relative risk (RR), or odds ratio (OR) along with 95% confidence interval. in the case of rare events.

When not enough information was available to calculate the standard deviations for the changes‐from‐baseline, we calculated them on data from some other trial included in the review (using the same comparison, the same outcome measure, and similar time periods). We calculated the correlation coefficient from a trial reported in considerable detail and imputed a change‐from‐baseline standard deviation in another study, using an imputed correlation coefficient. We specifically use this approach for nRCT where it is not obvious that the comparison of final measurements involves the same quantity as the comparison of changes from baseline (Higgins 2011).

Dealing with missing data

We described missing data and dropouts/attrition for each included study in the 'Risk of bias' table. Whenever needed, we attempted to contact the study authors to request missing data. The analysis of the outcomes was planned on an intention‐to‐treat basis, meaning we included all participants randomised to each group in the analyses, regardless of whether or not they received the allocated intervention.

Assessment of heterogeneity

We interpreted I2 as suggested by the latest version of the Handbook for Systematc Review of Interventions (Higgins 2011):

  • 0% to 40%: might not be important;

  • 30% to 60%: may represent moderate heterogeneity;

  • 50% to 90%: may represent substantial heterogeneity;

  • 75% to 100%: considerable heterogeneity.

In addition, a Chi2‐ test of homogeneity was used to determine the strength of evidence that heterogeneity is genuine.

Data synthesis

Primary outcomes are presented in on‐ and off‐medication. We summarized data in a meta‐analysis using a fixed effect model if they were sufficiently homogeneous. A random‐effect model was used when there was substantial heterogeneity. The analyses were done using the Review Manager software.

Summary of findings table

The main outcome measures were synthesized in two ‘Summary of findings’ tables, comparing STN DBS to the best available medical treatment and to other deep brain surgery (i.e. GPi). We assessed the overall quality of the evidence for each outcome using the GRADE approach, as recommended in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011), against five factors: study design limitations, consistency of results, directness (generalisability), precision (sufficient data) and reporting of the results across all studies that measure that particular outcome. The quality starts at high when high quality RCTs provide results for the outcome, and reduces by a level for each of the factors not met.

High quality evidence: there are consistent findings among at least 75% of RCTs with no limitations of the study design, consistent, direct and precise data and no known or suspected publication biases. Further research is unlikely to change either the estimate or our confidence in the results.

Moderate quality evidence: one of the domains is not met. Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.

Low quality evidence: two of the domains are not met. Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.

Very low quality evidence: three of the domains are not met. We are very uncertain about the results.

No evidence: no RCTs were identified that addressed this outcome.

Subgroup analysis and investigation of heterogeneity

To explore possible causes of heterogeneity, we planned subgroup analyses is for trial design, age, treatment duration and severity of motor complications, but in view of the limited number of trials retrieved, subgroup analysis was only done on the basis of trial design.

Sensitivity analysis

Sensitivity analysis was also done to investigate whether the the effect of surgery was maintained after removing trials at high risk of bias.