Thalamotomy has been performed since the 1950s when some surgeons noted excellent relief of tremor compared with pallidotomy (anterodorsal). Hassler et al. (77) reported the successful treatment of a patient by making a lesion in the ventral tier of lateral thalamus, and Cooper also advocated that this was the optimal target (14). The VL region of the thalamus contains at least three important motor nuclei. According to Hassler's classification and running from anterior to posterior, these are the ventral oralis anterior (Voa), ventral oralis posterior (Vop), and ventral intermediate (Vim). In a more recent nomenclature, the Voa and Vop nuclei are grouped together as the nucleus VL anterior (VLa), and the Vim is referred to as the nucleus VL posterior (VLp). Hassler subsequently refined the target to the Vop for tremor and Voa for rigidity (78). It is now generally agreed that the optimal target for tremor control is the Vim nucleus, which receives its input primarily from the cerebellum. Although most surgeons now target Vim, other subthalamic sites, such as the zona incerta and fields of Forel targeted in the past, are again being examined (79). The method of targeting is similar in principle to that of pallidotomy, relying on initial anatomical targeting possibly followed by electrophysiological refinement. Because of current limitations of the imaging techniques in directly visualizing the subnuclei of the thalamus, indirect targeting is generally applied. This is dependent on standard brain atlases to derive the coordinates of the target, which are then translated into the patient's anatomy in relation to the patient's ventricular landmarks obtained by ventriculography, CT, or MRI. Intraoperative electrophysiological evaluation, microelectrode or field potential recording, may be used to take account of interindividual variability in the position of the target. These can correctly identify "tremor cells" and help to avoid the neighboring structures, such as the internal capsule (80-82). Lesion sizes tend to be smaller (about 60 mm3) (83) compared with pallidotomy (about 220 mm3) (24).
There have been few long-term follow-up reports of thalamotomy in PD. Kelly (84) reported the 10-year follow-up of 60 parkinsonian patients who had thalamotomy between 1965 and 1967 and showed the sustained improvement in tremor and rigidity, but the continued progression of bradykinesia. Later, 12 patients were reported who also had a marked improvement or cessation in contralateral tremor without complications (15). A larger series of 103 patients operated between 1964 and 1969, also followed-up for 10 years, showed that overall 87 patients had a "good" effect and in only seven patients were tremor or rigidity not alleviated completely (29). In a more recent series from 1984 to 1989, all 36 patients underwent CT and microelectrode-guided VL thalamotomy and 86% showed complete cessation of tremor, with a further 5% showing a significant improvement up to 68 months of follow-up (85). The antitremor effect was shown to be maintained in one blinded retrospective study, following thalamotomy alone, subthalamotomy alone, or combined thalamotomy and subthalamotomy (86). In another study, the records of 42 patients with PD who underwent thalamotomy were reviewed and 86% were found to have cessation of or moderate to marked improvement in their contralateral tremor, with a concomitant improvement in function which persisted for as long as 13 years. The mean daily dose of levodopa was reduced by 156 mg and lesion location was in the Vim (87). Postop-eratively, rigidity improved by 30%, ipsilateral tremor worsened, and there was no significant change in other features of parkinsonism.
The complication rates of unilateral thalamotomy range from 10% transient confusion and 8% facial weakness or numbness (29) to 58% transient and 23% persistent complications of contralateral weakness, balance deficits, and blepharospasm (16). In the series of Fox et al., 22 of the 36 patients (61%) experienced complications. Half of these cleared by seven days and only 6% were permanent or bothersome, including dysarthria, dyspraxia, or cognitive dysfunction (85). Deficits of speech, language, and verbal memory such as dysarthria, hypophonia, dysfluency, and aphasia have been described following unilateral thalamotomy and are more common after left- than right-sided procedures (88,89).
Schuurman et al. (90) compared the short-term safety and efficacy of Vim thalamotomy with Vim stimulation (90). Marked improvement or tremor resolution was detected in 79% of the lesioned group compared with 90% in the stimulated group. These results were not statistically different, but only 23% of the lesioned group had an improved functional status compared with 53% of the stimulated group. Additionally, the complication rate for the stimulated group was 17% compared with 47% in the lesioned group, but there was one death in the stimulated group. These positive and negative effects of thalamotomy and thalamic DBS are very similar to a retrospective report comparing the two treatments (91), in which it was additionally shown that tremor recurrence occurred in 15% of patients with thalamotomy, but only in 5% of patients with DBS. Furthermore, 15% of patients with thalamotomy required reoperation to achieve good clinical outcome. These studies show the expected improved morbidity of DBS compared with irreversible lesion-ing. However, long-term follow-up (up to 66 months) has highlighted the potential hardware complications of DBS, which include lead fracture, erosion or migration, infection, CSF leak, and short or open circuits or other system malfunctions (92). In one study, hardware-related complications typically appeared late, 12 months or more after the surgery (93). It is estimated that a significant number of patients with DBS may require subsequent surgery to maintain the hardware (92-94). Further long-term clinical and economic comparative studies are necessary to define whether the benefits of DBS outweigh its higher maintenance and costs.
One factor that continues to confound our understanding of thalamotomy is the variability of lesion locations across series. This is more problematic than with pallidotomy since the boundaries of the thalamic nuclei are not anatomically as well demarcated as in the pallidal complex and also the nomenclature of the thalamic nuclei varies across series and also from human to primate. Atkinson et al. reported on accurate localization of 31 lesions performed in patients with tremor-dominant PD. The optimal improvement of tremor occurred with Vim lesions, which also included its posterior boundary with the sensory nucleus ventralis posterior, suggesting the involvement of the proprioceptive thalamus in parkinsonian tremor (95).
These studies support a role for Vim thalamotomy in patients whose predominant symptom is medically intractable asymmetrical tremor and who are not suitable for DBS, for example, because of inability to cope with the stimulator or the potentially demanding follow-up schedule (96). The majority of patients with PD are, however, likely to have progressive bradykinesia, even if this is not present at the time of surgery. This symptom is not modified by Vim thalamotomy, and so thal-amotomy (and Vim DBS) in the treatment of PD has largely been replaced by alternative therapies.
Was this article helpful?