It has been known since the early days of levodopa therapy for PD that motor fluctuations and dyskinesia were associated with drug therapy (12). Barbeau (61) referred to it as the "long-term levodopa syndrome." At that time, with no alternative treatments available, he indicated that its existence did not counterbalance the great usefulness of the drug. The questions are what causes their onset and progression and what are the key risk factors? The main issues in the debate address whether they are a result of disease progression or primarily levodopa itself (or the manner in which it is delivered to the brain) or both. The answer is not totally clear, but this question has been examined extensively by (i) evaluating patient populations and examining which of the two factors correlates with the onset of fluctuations and dyskinesia and (ii) examining the actual response fluctuations in a controlled setting to determine possible etiologic explanations.
In a retrospective study, Lesser et al. (62) collected data from 131 PD patients relating to severity of disease and late complications and assessed whether these problems were attributed to duration of disease or levodopa therapy. A relationship was seen between the presence of fluctuations and duration of therapy, since those with fluctuations tended to be treated for four years or more. This was not true for dyskinesia. They, therefore, associated fluctuations with levodopa therapy but did not rule out the possibility that those receiving levodopa longer had a more progressive disease. It was recommended that initiation of therapy be delayed until the patient "begins to function unsatisfactorily in occupational or social situations." This is perhaps the most frequently quoted paper on the subject; however, the authors themselves pointed out the flaws in a retrospective study and indicated the need for a prospective evaluation of the problem.
In another retrospective study, de Jong et al. (63) examined 129 PD patients to determine the role of age of onset, predominant symptom (tremor vs. akinetic/rigid vs. all three together), duration of therapy, and disease severity in the occurrence of motor fluctuations. There was no significant effect of age of onset, predominant symptom or duration of disease prior to levodopa therapy (but there was a trend). However, those patients with later therapy showed a lower frequency of fluctuations. Those patients treated in the earlier stages of disease (Hoehn and Yahr stages 1 and 2) did significantly worse with regard to the onset of fluctuations than patients initiating therapy in later stages (Hoehn and Yahr stages 3 and 4). They concluded that levodopa should not be started until stage 3 disease.
Several studies have since been published which contradict these findings. Cedarbaum et al. (64) criticized the papers described earlier by indicating that the patients treated earlier had more severe disease prior to initiating therapy, continued to progress faster, and thus were more prone to the onset of motor fluctuations. They suggested that levodopa was not the cause of the late complications, nor did the drug itself lead to loss of efficacy. In their own retrospective study, 307 patients were surveyed or interviewed with regard to motor fluctuations and various demographic features and records were reviewed. Patients were evaluated as a whole and were divided into several subgroups based on duration of disease and duration of therapy. Analyses failed to show an association between initiation of levodopa therapy and fluctuations or dyskinesia. Both the duration of disease and duration of therapy were longer in the patients with response fluctuations and dyskinesia. Despite these findings, detailed statistical analyses of sub-groups failed to demonstrate that the age of onset and duration of therapy influenced the occurrence of fluctuations and dyskinesia. Mean delay in levodopa therapy was the same for fluctuators and nonfluctuators. However, patients with dyskinesia were more than three times as likely to have had initiation of levodopa delayed more than two years from diagnosis. The authors did not advocate delaying levodopa therapy because it, in fact, increased chances of dyskinesia.
Blin et al. (65) agreed that the apparent acceleration of progression of disease after initiation of levodopa therapy related to the rapidity of disease progression prior to levodopa therapy and not the therapy itself. They also found that delayed initiation of levodopa led to quicker onset of dyskinesia. In a prospective study of 125 patients, Caraceni et al. (66) followed patients for a mean of six years from initiation of levodopa therapy to evaluate any risk factors for motor fluctuations and dyskinesia. To avoid bias, all patients, regardless of the course of progression, were started on levodopa at first diagnosis. Using a multivariable analysis, they found that the risk of late complications was greater in those with akinetic-rigid PD, younger age of onset, greater disability and duration of disease, and longer interval between disease onset and levodopa therapy. Duration and dose of levodopa therapy were not associated with onset of late complications. They concluded that levodopa did not accelerate the appearance of motor fluctuations, that these complications related to the severity and progression of PD, and that there is no need to delay treatment.
Hoehn (67), based on her comparison of patients in pre and post levodopa eras, indicated that a delay in introduction of levodopa but not duration of treatment was associated with a poorer outcome. Horstink et al. (68) examined the relationship of duration of disease and duration of levodopa therapy and onset of peak-dose dysk-inesia in 54 PD patients and found that both duration of disease and levodopa therapy were greater in the dyskinetic group. The two variables were closely linked, so they then studied patients with significantly asymmetric dyskinesia and found dyskinesia to be most prominent on the worst side, suggesting that disease severity is an important risk factor for dyskinesia, not duration of levodopa therapy.
Roos et al. (69) retrospectively studied 89 PD patients and any clinical correlation's that could be found with onset of response fluctuations (age of onset of PD, presenting symptom, duration of PD, stage of PD at initiation of levodopa, and mean and last dose of levodopa). They used survival and covariate analyses. No correlation was found between the dose of levodopa and the onset of fluctuations. However, a rapid increase in levodopa dose rather than the total dose seemed to determine the onset of fluctuations. They suggested that fluctuations occurred in patients with a more rapidly progressive disease requiring a more rapid escalation in levodopa dose. They also concluded that there are no good reasons to delay levodopa therapy if disability dictates its need.
Finally, Kostic et al. (70) recently examined the effect of Hoehn and Yahr stage of disease at the onset of therapy on the length of time between initiation of levodopa therapy and the development of motor complications. Of 40 consecutive PD patients treated <5 years, 17 were treated in stage 1, 13 at stage 2, and 10 at stage 3. They found that severity of disease was an important factor in the onset of fluctuations and dyskinesia. Those patients initially treated at stage 3 developed dyskinesia and fluctuations significantly earlier than patients in stages 1 and 2. However, latencies from disease onset to development of fluctuations and dyskinesia were not different between groups. This result suggested the importance of disease duration and severity in the onset of late complications in PD.
Although questions remain, the majority of studies suggest that disease duration, progression, and severity are important risk factors in the development of motor fluctuations and dyskinesia. In accepting this conclusion, one would agree that, based on the occurrence of motor complications, there is no reason to delay PD therapy. In fact, two of the studies indicate that a delay would increase the likelihood of dyskinesia. These findings are consistent with reports of patients with late-stage PD (12,70,71) or severe parkinsonism secondary to 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) developing fluctuations soon after the initiation of therapy. This has also been seen in MPTP-treated nonhuman primates (72) and postencephalitic parkinsonism (73).
Several groups have studied the mechanism of motor fluctuations. The findings suggest that both duration of disease and levodopa therapy play a role. Fab-brini et al. (74) demonstrated that perhaps the initial feature that leads to the onset of fluctuations is the progression of degeneration of nigral dopaminergic neurons to a threshold level. Once this level is reached, motor fluctuations begin with wearing-off. In their studies, they examined four groups of patients: levodopa naive, levodopa-treated stable responders (nonfluctuators), patients with wearing-off, and patients with unpredictable on/off. They treated each patient with a continuous intravenous infusion of levodopa for 16 hours and then abruptly stopped it. They found that there was no change in pharmacokinetics of levodopa in the more advanced patients. However, it was noted that there was a decay of antiparkinson-ian effect, which worsened significantly as the patients advanced from being lev-odopa naive to having on/off phenomenon. The authors concluded that the wearing-off effect is initiated as a consequence of the marked loss of presynaptic dopaminergic neurons. With loss to a threshold number of neurons, the dopamine system loses its ability to store and release dopamine, and thus buffer fluctuations in serum and cerebral levodopa and dopamine levels. These levels become dependent on peripheral availability of exogenous levodopa. It is believed that levodopa is converted to dopamine in nondopaminergic cells that lack the ability to store and release it in the normally tonic fashion (23). Stimulation at postsynaptic dopamine receptors then becomes intermittent (pulsatile), as a reflection of the peak and trough profile of oral levodopa therapy. It appears that as soon as this intermittent stimulation of dopamine receptors begins, postsynaptic changes are initiated.
Studies have demonstrated a narrowing of the therapeutic window, alteration of threshold for onset of dyskinesia, and steepening of the anti-PD response slope, all which underlie progression toward a more unstable response to levodopa (75,76). On a cellular level, upregulation of preproenkephalin (PPE) mRNA and downregu-lation of dynorphin and substance P mRNA among others were found in parkinsonian rodents and primates (77). There is also a resulting change in firing rate and pattern for neurons in the postsynaptic regions, including basal ganglia, thalamus, and cortex. These changes and those resulting from pulsatile stimulation of postsy-naptic receptors may lead to the development and progression of dyskinesia and fluctuations, particularly that related to PPE (77). In addition, studies have indicated that postsynaptic glutamate receptors become hypersensitive in relation to the development of motor fluctuations and dyskinesia as compared to patients without fluctuations (78). These findings support the involvement of postsynaptic mechanisms, reflecting an increased sensitivity of clinical response to small fluctuations in dopamine levels and differing pharmacological mechanisms for antiparkinson-ian response and dyskinesia. It is, therefore, suggested that it is not the drug but the pulsatile nature of its delivery that is important. This conclusion is based on the reversal of fluctuations with continuous dopaminergic therapy, IV or duodenal levodopa, infusions of apomorphine and lisuride (79-82), but despite these findings, this conclusion remains controversial, especially in relation to the development of motor complications.
In trials comparing levodopa to dopamine agonists (49,53), levodopa therapy leads to earlier onset and more frequent occurrence of dyskinesia and wearing-off. This would suggest that either the agonist prevents the onset of these problems or that levodopa, or its short duration of action, has a role in causing them. It does appear that once the progression threshold is reached, levodopa plays a role, via intermittent stimulation of postsynaptic receptors, in the progression of fluctuations to a more unpredictable pattern. One needs to consider that onset and progression are probably caused by different scenarios. It has been demonstrated that continuous infusion of levodopa and stimulation of the subthalamic nucleus (STN) can reduce motor fluctuations and dyskinesia (79,82,83). Some indicate that the reversibility of fluctuations implicates levodopa in the cause of fluctuations (84,85), but this is not the only interpretation. It can also mean that the role played by levodopa in motor fluctuations is potentially reversible. The dose of levodopa also appears to be important since higher doses probably lead to wider pulses. This was demonstrated in the ELL-DOPA study where the group receiving 600mg/day of levodopa developed earlier, more frequent dyskinesia compared to those on lower doses (59).
Age of onset is one other risk factor independent of disease progression and levodopa that leads to motor complications. Several retrospective studies have demonstrated mixed results, perhaps due to the age range in the populations studied (63,66). Studies of young onset disease (age of onset <40) indicate a dyskinesia rate of 94% after five years (28). One population-based study examined the frequency of dyskinesia by decade starting with the age of onset 40 to see if the risk of dyskinesia changes over the age spectrum (86). The rate of dyskinesia at five years was examined in patients receiving levodopa as first line therapy and had been treated for at least five years. The incidence of dyskinesia declined, as the age of onset increased: for onset at 40 to 59 years, the frequency was 50%; for 60 to 69 years, frequency was 26%; and for >70 years, the frequency was 16%. The authors suggested that patients with onset of PD before 40 have a unique and unknown physiology, leading to substantial dyskinesia that is age-related, not necessarily levodopa-related.
Was this article helpful?