Brain Target May Affect Cognition in Parkinson's
By John Gever, Senior Editor, MedPage Today
Published: June 22, 2012
Reviewed by Dori F. Zaleznik, MD; Associate Clinical Professor of Medicine, Harvard Medical School, Boston and Dorothy Caputo, MA, BSN, RN, Nurse Planner
- In this study, Parkinson's disease patients had equivalent motor outcomes but a possible worsened cognitive decline over a 3-year period after receiving deep brain stimulation targeting the subthalamic nucleus (STN) compared with the globus pallidus interna (GPi).
- Note that there was a nearly 50% dropout rate for patients over the course of the study, and that the STN group scored slightly worse than the GPi group at baseline on some cognitive measures.
Some 3-year outcomes of deep brain stimulation for Parkinson's disease were better when the leads were placed in the globus pallidus interna (GPi) rather than the subthalamic nucleus (STN), researchers found.
In a randomized trial, motor symptoms were comparable between the two groups, according to Frances M. Weaver, PhD, of the Hines VA Hospital in Hines, Ill., and colleagues.
But neurocognitive function declined faster on average in patients with leads placed in the STN compared with patients in whom stimulation was delivered to the GPi (P=0.04), the researchers reported online in Neurology.
Deep brain stimulation has become a standard approach to management of Parkinson's disease when drug therapy is no longer adequate to control symptoms. However, there has been debate about the optimal target within the brain for the electrical stimulation, with the GPi and STN each claiming adherents.
Initial reports on GPi-targeted stimulation had suggested that its effectiveness didn't last as long, Weaver and colleagues explained, but more recent randomized studies had indicated the choice was a toss-up.
On the other hand, those trials followed patients for 2 years at most. With clinicians interested in possible differences in longer-term outcomes, Weaver and colleagues conducted a 3-year randomized trial.
A total of 299 patients were initially recruited and randomized, but with deaths, withdrawals, and disappearances in the lengthy study, only 159 were evaluable at the final visit. Of those, 70 patients had had leads placed in the STN and 89 had had leads placed in the GPi.
Mean patient age at baseline was about 61, and most patients were white and male. They had been on drug treatment for an average of 11 years, with mean Hoehn and Yahr scores off medication of 3.3.
Mean medication doses declined and remained lower for both groups. In the GPi group, usage declined by about 250 mg of levodopa equivalents by month 6 and remained stable through follow-up. In the STN group, doses fell by roughly 450 mg at first and were then stable.
Motor function scores on the Unified Parkinson's Disease Rating Scale (UPDRS) showed significant improvements relative to baseline with both types of stimulation while on medication, which remained stable throughout follow-up.
While off medication, both groups also showed major improvements with stimulation, but patients in the STN group showed more deterioration from month 6 to year 3 (mean increase of 4.9 points versus an increase of 1.7 points in the GPi group).
For other symptoms assessed by the UPDRS -- mentation, mood, activities of daily living, and complications of therapy -- there was no difference between groups in outcomes.
Modest increases were seen in UPDRS Part I scores (mentation, behavior, and mood), reflecting disease progression, from month 6 to year 3. Part II scores (activities of daily living) declined sharply at first, but then increased nearly to baseline levels by year 3.
On the other hand, UPDRS Part IV scores (complications of therapy) dropped initially and then remained stable through the follow-up period.
But of concern for STN-targeted stimulation, Mattis Dementia Scales scores in this group worsened in the first 6 months and then declined through the rest of the follow-up period, ending up significantly lower than in the GPi group.
At baseline, Mattis scores averaged 137.8 in the GPi group and 137.1 in the STN group. By month 6, the mean GPi group score remained virtually unchanged, but in the STN group it declined to 134.7.
At the 3-year follow-up, the mean Mattis score dropped further to 130.9 in the STN group, compared with 135.2 in the GPi group (P=0.01).
Scores on the Hopkins Verbal Learning Test were also similar at baseline but significantly worse after 3 years in the STN group versus those assigned to GPi stimulation (P=0.004).
Score trends on six other neurocognitive tests, on the other hand, did not differ between the treatment arms. These included the Beck Depression Inventory, Wechsler Adult Intelligence Scales, Verbal Fluency, Boston Naming Test, Wisconsin Card Sorting Test, and Brief Visuospatial Memory Test.
The investigators noted that the differences in Mattis scores and the Hopkins Verbal Learning Test should be interpreted with caution since there were baseline cognitive differences between groups which favored GPi, and they did not adjust for these differences or other covariates as confounders.
In an accompanying editorial, Michele Tagliati, MD, of Cedars-Sinai Medical Center in Los Angeles, said the findings appear to give an edge to the GPi as the preferred target for deep brain stimulation.
"Motor outcomes being equal, the lack of ON/ON motor and cognitive deterioration over time [with GPi stimulation] is clearly a very desirable long-term benefit for any deep brain stimulation candidate," he wrote.
Tagliati also questioned the clinical relevance of the greater reduction in medication needs in the STN group, given the poorer symptom control and increased cognitive decline.
On the other hand, he added, patients with severe resting tremor might benefit more from STN-directed stimulation.
Either way, Tagliati suggested, the study provided a welcome infusion of hard data into what had been a fuzzy decision-making process for patients and physicians.
"If the art of making medical decisions reflects a continuous struggle between evidence- and preference-based practices, this study will inject more reliable evidence in delicate long-term decisions, based until now almost exclusively on the preference and personal experience of the deep brain stimulation provider," he wrote.
The study was funded by Medtronic, the Department of Veterans Affairs, and the National Institute of Neurological Disorders and Stroke.
Study authors reported relationships with Teva, Ipsen, Medtronic, Adamas, Schering-Plough, GlaxoSmithKline, GE Healthcare, EMD Serono, Boehringer Ingelheim, St. Jude Medical, Impax, Novartis, Biogen Idec, Solvay, Xenoport, Santhera, Eisai, Allon, Acadia, Valeant, Kyowa, Johnson & Johnson, Oxford Press, and the Movement Disorder Society.
Primary source: Neurology
Weaver F, et al "Randomized trial of deep brain stimulation for Parkinson disease: Thirty-six-month outcomes" Neurology 2012; DOI:10.1212/WNL.0b013e31825dcdc1.
Additional source: Neurology
Tagliati M, et al "Turning tables: Should GPi become the preferred DBS target for Parkinson disease?" Neurology 2012; DOI:10.1212/WNL.0b013e31825dce96.
Add Your Knowledge ™
John Gever, Senior Editor, has covered biomedicine and medical technology for 30 years. He holds a B.S. from the University of Michigan and an M.S. from Boston University. Now based in Pittsburgh, he is the daily assignment editor for MedPage Today as well as general factotum on the reporting side. Go Pirates/Penguins/Steelers!