To date, the results of only two short, randomized, single-blinded trials of pallidotomy have been published. The most recent study showed improvements of 75% in contralateral dyskinesia, 45% with respect to complications of therapy (UPDRS IV), 36% in ipsilateral dyskinesia, and 34% in parkinsonism at six months follow-up compared with medical therapy in which these aspects generally worsened (41). In the other study (42), 37 patients, who were matched for age and severity of PD, were randomized to receive either unilateral pallidotomy within one month (n = 19) or maximal medical therapy for six months (n = 18). Although the nonoperated group showed an 8% deterioration of median UPDRS motor scores and no change in dyskinesia, the operated group showed 31% and 50% improvements in parkinsonism and dyskinesia scores, respectively. This group more recently showed that, as expected, unilateral pal-lidotomy is less efficacious in improving parkinsonism than bilateral subthalamic stimulation, reducing the UPDRS by 9.5 points compared with 25 points for STN DBS (43).
There have been only two nonblinded studies of patients treated by pallido-tomy compared with a medically treated group (44), with each study supporting the findings of the randomized single-blinded studies. Numerous other open-labeled nonrandomized trials (27, 45-61) have generally drawn the same conclusion (Table 1) indicating that the most dramatic response is the reduction in contralateral dyski-nesia by 80% to 95% which is sustained for up to 5.5 years (61). Overall, the off UPDRS score improves by 24% to 37% and declines thereafter to about 18%, although this continues to remain significantly improved at 5.5 years from baseline (49,61). Individual items of contralateral tremor, rigidity, and akinesia generally mirror this response, although the magnitude of the antitremor effect (up to 65%) appears greater and more sustained than that of rigidity (43%) or akinesia (falling from 46% at six months to 17% at 5.5 years). Despite these sustained differences in UPDRS subset scores, an initial improvement in activities of daily living of 37% is not sustained (61), but results from patient self-assessments imply that patients continue to benefit
(49). In contrast to contralateral off scores, ipsilateral off scores, and both contralateral and ipsilateral on scores are not significantly sustained, although an initial improvement of up to 27% may occur. Ipsilateral on dyskinesia scores appear to be improved initially by 30%. This effect is also decreased with time and is not significant after 12 months postsurgery (49).
The responses of axial symptoms and gait are variable. Complex analysis of posturography has shown that an improvement in gait and posture may be maintained for up to 12 months (62). It is possible that the gait improvement results from a decrease in dyskinesia when on. Three-dimensional motion capture analysis of walking suggests that the effect is mainly due to an improvement in speed of walking (63). More traditional UPDRS gait/postural instability subset scores, however, show only an initial modest improvement (26-37%), which is lost within subsequent years (49,61). It is possible that the effect of pallidotomy on gait may be mediated in part via descending influences on the brainstem, as well as ascending influences on thalamo-cortical circuits (62). Longer follow-up of complex gait analyses is required before reliable conclusions can be drawn.
The frequency of severe complications is approximately 5%, with transient facial and limb paresis being the most common. Hemianopsia or quadrantanopsia are potential complications of lesioning the nearby optic tract. There is a well-documented consistent feature of a mild but asymptomatic decrease in verbal fluency (30), mostly following left-sided unilateral pallidotomy (64), but permanent cognitive adverse effects did not seem to persist in one small study of 11 patients at four years of follow-up (65). Postoperative weight gain has been described (66). This "side effect" was found in 23% of patients in one study (67). It was highly correlated with the improvement in off motor UPDRS scores, but not with changes in energy intake or dyskinesia scores. Some series have reported a higher overall incidence of major complications. In the controlled trial of de Bie et al., nine of the 19 (47%) operated patients had surgical morbidity (two major, four minor persistent, and three minor transient). Lesion locations were not presented; however, this level of high morbidity has also been documented by other independent groups (27,60). It is likely that the variability of lesion locations and surgical techniques account for these differences.
A systematic attempt to correlate outcome with lesion location has been made. Gross et al. (68,69) studied the variability in lesion location within the ventral pallidum in 33 patients with PD. Lesions were not distributed randomly within the internal pal-lidum but were distributed along a line running anteromedially-posterolaterally, parallel to the lateral border of the posterior limb of the internal capsule. In this cohort, anteromedial lesions were associated with a greater improvement in dyskinesia, whereas central lesions lead to a greater improvement in akinesia scores and gait disturbance (69). This result may partly explain the variable results in resolution of dysk-inesia and akinesia among different neurosurgical centers and demonstrates the precision that is required to perform pallidotomy.
Was this article helpful?