Abnormal color discrimination has frequently been reported in patients with Parkinson's disease.27 28 In many studies, this impairment has been found to be most prominent in the tritan (blue-yellow) axis.29,30 Abnormalities of color perception have been demonstrated using both bedside clinical testing techniques such as the Farnsworth-Munscll (FM) 100-hue test30 or more elaborate psycho-physical means such as a computer-generated assessment of color contrast sensitivity.29 Haug et al.29 offered an explanation as to why the tritan contrast threshold is most affected in Parkinson's disease. In general, the blue cone system is preferentially affected in retinal disease, because its response range is limited, and it has the greatest vulnerability. The relatively selective involvement in PD can be explained by the fact that these short-wavelength-sensitive cones are relatively scarce in number in the retina and spaced widely apart, such that maintenance of their large receptive fields is dependent on interaction across considerable distances, a function mediated by the dopam-inergic interplexiform and amacrine cells of the retina, the precise retinal elements that are most affected in PD. Involvement of these same retinal cells in PD may result in other forms of visual dysfunction but not necessarily related to the same pattern of impaired cellular connectivity. Pieri et at.31 studied both color discrimination and contrast sensitivity in PD and found impairment of the two forms of visual dysfunction to be independent variables, suggesting that different retinal mechanisms underlie each.
The abnormality of color vision seen in PD can be demonstrated in very early patients who have not yet begun antiparkinson drug therapy. It can be reversed by treatment with levodopa.32,33 Paradoxically, in one case, color vision was worsened after treatment with the dopamine agonist pramipexole.34 Color discrimination testing in untreated, de novo PD patients has shown a significant correlation between the error score of the FM test and the severity of clinical parkinsonian signs as measured by the motor and activities of daily living subscales of the Unified Parkinson's Disease Rating Scale (UPDRS).35 When PD patients are followed longitudinally over time, color discrimination scores decline progressively as the underlying disease worsens,36-37 although, in one study, the decrementing scores only correlated with decline in the UPDRS activities of daily living (ADL) score,37 and in another with both the UPDRS motor and ADL scores.36 Despite the consistent correlation with disease severity by one measure or another, one investigation demonstrated that the magnitude of color vision abnormality in PD does not correlate with dopam-inergic nigral degeneration as reflected by I123 P-CIT single photon emission tomography of the dopamine transporter. This observation s consistent with the prevailing notion that the visual abnormality in PD is largely ext-ranigral in origin.38 A plausible explanation of why color discrimination impairment does not correlate with nigral degeneration, yet parallels the clinical severity of PD, is that retinal dopamine depletion, although independent of nigral dopamine depletion, occurs contemporaneously at a relatively constant pace over time.
Regan et al.39 questioned whether abnormalities uncovered during color discrimination testing in Parkinson's disease patients are just an epiphenomenon related to the motor disability of Parkinson's disease, since the FM test, used to demonstrate impaired color vision in many studies of PD, requires a motor response to correctly identify varying hues of color. They questioned whether it is the manual impairment of PD patients rather than a primary visual disorder that causes PD patients to fail this test and at the same time explains why levodopa, which corrects the motoric abnormality, improves the color discrimination score. These investigators utilized a computer-controlled test of color vision that did not require a motor response and found that PD patients performed as well as a control group. Their hypothesis, however, fails to explain why other investigators utilizing computer testing techniques29 did uncover abnormalities of color vision in PD, or why most studies have revealed a preferential loss in the tritan color axis with little or no abnormality in the protan (red-green) axis, both of which should have been similarly affected were the abnormal test scores simply a reflection of parkinsonian motor impairment. There is additional evidence that supports the validity of a primary color vision abnormality in PD. Abnormalities of the visual evoked response produced by color pattern stimuli are more responsive to levodopa therapy than are those evoked by black-and-white stim-uli.40 Similarly, color contrast sensitivity in PD patients is most impaired along the tritan axis.29 Lastly, other medical conditions characterized by impairment of dopaminer-gic transmission have been associated with abnormalities of color vision. In patients undergoing cocaine withdrawal, a relative hypodopaminergic state exists, and a similar tritan axis deficit in color discrimination has been noted. The same abnormality of color vision was not seen however during their hyperdopaminergic intoxication phase.41 In schizophrenia, on the other hand, color discrimination abnormalities have been found to be general and not hue specific, leading to the hypothesis that axis-specific color discrimination abnormalities are a reflection of depletion of dopamine rather than its general dys-regualtion.42
Was this article helpful?