Levodopa Response and Prognosis in Parkinson's Disease

Abstract & Commentary

By Claire Henchcliffe, MD, DPhil, Assistant Professor, Department of Neurology, Weill Medical College, Cornell University. Dr. Henchcliffe reports no financial relationship to this field of study.

Synopsis: Patients with Parkinson's disease takinglevodopa were divided into groups with (i) moderate-severe fluctuations and (ii) stable response. Despite a longer disease course in moderate-severe fluctuators, progression in the late stages and autopsy findings were similar between groups, suggesting non-linear disease progression.

Source: Kempster PA, et al. Patterns of levodopa response in Parkinson's disease: a clinico-pathologic study. Brain. 2007;130:2123-2128.

The authors identified 97 autopsy-proven Parkinson's disease (PD) cases from the Queen's Square Brain Bank between 2001 and 2006, all with long-term levodopa treatment and rigorous clinical documentation. Clinical features were evaluated by chart review. Cerebral pathology was determined by: synuclein antibody-mediated identification of Lewy bodies, in the substantia nigra, transentorhinal and cingulate cortex, and neocortex of the second frontal gyrus, superior temporal gyrus, and inferior parietal cortex, as well as examination for Alzheimer-type pathology and other changes. Cases comprised 32 women and 65 men, with mean age at diagnosis 61.7 ± 10.3 years (range 43-80). Motor fluctuations developed in 64%, and dyskinesias in 62%. Of the 35 cases without motor fluctuations, only 14% developed dyskinesias. Cases were divided into non-fluctuators, mild fluctuators, and moderate-severe fluctuators. Moderate-severe fluctuators had younger age of onset (57.9 ± 10.7 vs 66.4 ± 8.3 years, p < 0.001), and longer disease course (17.6 ± 5.9 vs 9.9 ± 6.4 years, p < 0.001), with similar age at death (75.4 ± 9.4 vs 76.3 ± 8.3 years) compared with non-fluctuators. Time from diagnosis to starting levodopa was not significantly different between the groups, but moderate-severe fluctuators took higher maximum levodopa doses than non-fluctuators (1041 ± 450 vs 549 ± 267 mg daily, p < 0.001). Incidence of pre-designated milestones of late disease, comprising frequent falls (36% vs 31%), visual hallucinations (64% vs 57%), cognitive disability (53% vs 63%), and moving to residential care (44% vs 63%), did not significantly differ between groups. Mean interval from any of these milestones to time of death was similar between groups (frequent falls: 3.9 vs 3.6 years; visual hallucinations: 4.5 vs 3.9 years; cognitive disability: 3.2 vs 3.0 years; residential care: 2.9 vs 2.5 years). Autopsy examination revealed evidence of moderate-severe loss of substantia nigra neurons and associated Lewy bodies in all, and neocortical Lewy bodies in 67%. Patterns of Lewy body distribution were not different between the groups. Alzheimer-type pathology was seen in 54% cases, and cerebrovascular pathology and amyloid angiopathy were also observed in both groups.


One aim of the present study was to address whether differences in pathology might explain the well-documented heterogeneity in levodopa response seen in PD. However, the authors found no pathologic correlates with the distinct patterns of levodopa response they chose to examine, defined by motor fluctuations. As they point out, post mortem study may simply be too late in the disease to discern differences between groups that might have been significant earlier. In addition, confining study to Lewy body pathology and to this restricted set of brain regions might be insufficient to detect more subtle differences between PD subtypes. That being said, this article presents some fascinating data that require further understanding, and it underlines gaps in our concept of PD progression. Moderate-severe fluctuators had longer disease duration, implying slower progression, yet late disease milestones occurred at a similar time interval before death as in the non-fluctuators. This suggests that in late disease, progression is similar in both groups (i.e., speeds up in non-fluctuators). Importantly, this challenges the concept of a linear disease course, and the authors suggest a likely interaction with the aging process. Despite limitations inherent in this type of study, these findings raise major questions about our understanding of PD pathophysiology. The study also emphasizes the need for improving treatment in late disease, when many symptoms are not dopamine-responsive.