Object: Pre- and postoperative cognitive function was evaluated in 25 patients with Parkinson's disease (PD) who underwent unilateral posteroventral pallidotomy (PVP) to clarify the effects of this operation on cognitive function.
Methods: Cognitive function was assessed before surgery and 1 month and 3 months afterward by using a battery of neuropsychological tests consisting of orientation, digit span (forward and backward), digit learning, 5-minute memory of five words, similarities, serial sevens, animal name listing, and the kana pick-out test, as well as the Japanese version of the Mini-Mental State. Preoperative evaluation revealed that patients experienced difficulties in performing the neuropsychological tasks of reasoning and abstraction, working memory of numerals, word fluency, and concept formation compared with 36 healthy volunteers (p < 0.05). A simple regression analysis showed strong relationships in a negative fashion between preoperative (r = 0.81, p = 0.002) and 1-month (r = 0.79, p = 0.0059) and 3-month (r = 0.85, p = 0.0016) postoperative gross-total scores and Hoehn and Yahr staging. Preoperative and postoperative scores at 1 month and 3 months were analyzed by analysis of variance, but only the Fisher's post hoc test revealed the source of difference. All tasks except orientation (p = 0.0292) were unchanged in the 10 patients who underwent surgery on the right side. The scores for this task at 1 month postsurgery were significantly lower (p = 0.0203) but improved to preoperative values by 3 months. In the 15 patients who underwent surgery on the left side significant differences among tasks were revealed for serial sevens (p = 0.0471) and animal naming (p = 0.0425). The scores for these tasks were significantly lower at 1 month postoperatively (p = 0.0431 for serial sevens and p = 0.0408 for animal naming), but improved by 3 months after the operation.
Conclusions: The present data revealed that cognitive dysfunction in patients with PD relates to advancement of Hoehn and Yahr stage, but PVP is not associated with significant long-lasting cognitive deficits.