Levodopa and dopa decarboxylase inhibitors

Dyskinesias

[spi]change_bulletLevodopa-induced dyskinesia is very common. Neurologists from Geneva have examined the frequency and characteristics of a subtype of dyskinesia, abnormal involuntary eye movements [ ]. They studied 32 patients with advanced Parkinson’s disease, all of whom had moderate to severe symptoms and dyskinesias, and had had the disease for at least 5 years. Five, of whom two were women, had had abnormal involuntary eye movements for 7–18 years; two were also taking dopamine receptor agonists. The abnormal movements all occurred during the “on” phase of therapy and had a common pattern. They consisted of stereotyped and repetitive movements, upwards, sideways, or both. Some were phasic and others more sustained and tonic; in both cases they were usually (4/5) towards the side of the body more affected by the disease. The authors commented that this should be recognized as a particular type of dyskinesia, and one that may have a negative effect on postural stability.

Although it seems intuitively likely that the dosage of levodopa will have a significant effect on the incidence of dyskinesias, this topic has received surprisingly little attention. At the same time there is evidence that patients in whom ropinirole or pramipexole are used initially are less likely to develop dyskinesias. Neurologists from England and France have used data from two comparisons of levodopa and ropinirole (056 and REAL-PET) in 430 patients [ ]. These trials confirmed the much higher risk of dyskinesias in patients who started on levodopa rather than ropinirole—35% of patients taking levodopa developed dyskinesia in 2–5 years compared with 5% of those taking ropinirole. Among those taking only levodopa there was a highly significant difference in dosage between those who developed dyskinesia and those who did not: 9.2 mg/kg versus 6.9 mg/kg. Logistic regression analysis showed that the only variables associated with the emergence of dyskinesia were age and levodopa dosage per kilogram.

Dyskinesias occurring during long-term use of levodopa have been classified in various ways. One system presents them as falling broadly into three groups:

• “On” dyskinesias—These coincide with periods of clinical response to the drug; they include chorea, myoclonus, and dystonic movements. They are enhanced by dopamine receptor agonists and reduced by dopamine receptor antagonists.

• “Off” dyskinesias—These coincide with periods of poor response and comprise mainly dystonic postures, affecting in particular the feet; they are inhibited by both dopamine receptor agonists and antagonists.

• “Dysphasic dyskinesias’”—These occur at the beginning and end of “on” periods. They involve repetitive stereo-typed movements of the lower limbs, and they too react well to both dopamine receptor agonists and antagonists.

There is reason to believe that dyskinesias during prolonged levodopa treatment result from an interaction between the levodopa and the underlying condition rather than from one or the other. In many cases of severe Parkinson’s disease one can speak of a “peak dose dyskinesia,” which is only problematic when the patient is taking a relatively high dose, which tends to be most marked on the most severely diseased side, and which responds to dosage reduction [ ].

The long-term consequences of levodopa therapy have been considered in a review by predominantly Canadian authors, but whose senior author was Oleh Hornykiewicz, the doyen of the dopamine concept of Parkinson’s disease [ ]. Their conclusions were not at all unexpected. Levodopa was given to 42 patients (30 men, mean disease duration of 16 years, mean follow-up about 9 years). There were adverse effects in over 70%: dyskinesias in 62%, on–off effects in about 17%, and end-of-dose wearing-off in about 7% (the last perhaps surprisingly low). Dyskinesia was not only the most common but also the earliest adverse effect. It should be noted that the mean levodopa dosage in this population was rather low, only 500 mg/day.

An interesting case report has drawn attention to the fact that severe response fluctuations can occur soon after the start of treatment in patients with severe disease [ ].

• Two men aged 76 and 72 years presented with advanced symptoms of Parkinson's disease and were treated with levodopa. Disabling dyskinesia occurred within days of reaching maintenance dosages of levodopa (1–1.5 g/day with benserazide).

Clearly, the disease status of the patients was the determinant for this adverse reaction, not the duration of levodopa therapy.

In 17 parkinsonian patients three tests of proprioception were carried out 1 hour after the administration of levodopa or a dopamine receptor agonist [ ]. Although data were not provided for individual patients, there was an overall 11–31% deterioration in the mean scores in all three of the tests. There was no difference between patients with and without dyskinesias, but the authors suggested that abnormal proprioception may be a factor in drug-induced dyskinesia.

There is some evidence that the atypical neuroleptic drug clozapine can alleviate levodopa-induced dyskinesia while itself providing additional relief in Parkinson’s disease [ ]; clozapine may also relieve levodopa-induced psychosis [ ].

“Start hesitation” (paradoxical akinesia) is the name given to sudden episodes during which the patient feels a sensation of extreme heaviness of the feet and finds himself unable to start walking; the legs tremble and the patient falls forward; the condition can improve if the levodopa dosage is reduced.

The “on-off effect” seen after prolonged therapy in some patients is characterized by sudden swings between severe parkinsonian symptoms (with freedom from adverse reactions) and normal mobility (but with marked adverse drug reactions). It is likely that unexplained variations in dopamine concentrations in the central nervous system are responsible; the condition has been seen less often since combinations with decarboxylase inhibitors came into general use.

Hiccups, an unusual form of dyskinesia, have been attributed to levodopa [ ].

• An 80-year-old man with Parkinson’s disease and dementia developed worsening bradykinesia, and the dose of co-careldopa was increased from 12.5/50 mg bd to 25/100 mg bd. Two days later, he developed hiccups, which lasted for 1 week and which ceased when the dosage was reduced to the previous level. Ocular dyskinesias have been described in a 60-year-old Spanish man who showed intermittent upward and leftward deviations of gaze associated with peak concentrations of levodopa, in doses of up to 800 mg/day plus benserazide [ ].

It should be noted that the patient was and apparently still is taking pramipexole and amantadine in addition to levodopa.

Sleep disorders

Daytime sleepiness and nocturnal wakefulness can both occur in relation to dopaminergic drug therapy, but sleep attacks are rare [ ]. Many factors besides drug treatment can lead to sleep disorders in parkinsonian patients, including the disease process itself and episodes of sleep apnea; some are skeptical about the existence of sleep attacks as a phenomenon distinct from increased general drowsiness [ ], although others disagree. Neurologists from King’s College London have concluded that all dopaminergic drugs are capable of causing sleepiness and that in a minority of patients it may take the form of a narcolepsy-like phenotype, possibly amenable to improvement by selegiline or modafinil; they have recommended that affected individuals should not drive [ ]. Although it was their view that this is a class effect of dopamine agonists, they nevertheless suggested that switching from one agonist to another may sometimes be helpful.

Some remarkable case reports have previously been published [ , ] and reports continue to appear, supplemented by prospective studies and other analyses. For instance, 11 studies involving ropinirole or pramipexole in a total of 2066 patients have been reviewed [ ]. Four of these (two each with ropinirole and pramipexole) were placebo-controlled. The pooled relative risk of somnolence was 4.98 compared with placebo: there was a non-significant trend for greater somnolence with ropinirole, but the confidence intervals were much wider than with pramipexole. In the other studies levodopa alone was compared with levodopa plus the newer drugs; the relative risk was 2.06 compared with levodopa alone. It must be borne in mind that somnolence and sleep attacks may be separate phenomena, although this is controversial.

The whole field of sleep disorders in Parkinson’s disease has been reviewed in a consecutive series of 320 patients from Houston, with analysable data from 303 (sex distribution unknown) [ ]. The mean age was 67 years and the mean duration of the disease was 9.1 years. All the patients completed the Epworth Sleepiness Scale and answered specific questions about falling asleep while driving and about the restless legs syndrome. The mean sleepiness score was 11.1, values greater than 10 being regarded as abnormal. As one would expect, just over half the patients had scores at that level. Higher scores correlated with longer duration and greater severity of the disease, with male sex, and with the use of dopamine receptor agonists. There was no apparent difference in this regard between the three most commonly prescribed drugs in these patients (pergolide, ropinirole, and pramipexole, the last being the most frequent). Among those currently driving, 23% reported falling asleep in the car and nearly 20% had features of the restless legs syndrome. This study has therefore reinforced the view that sleep disorders are common in Parkinson’s disease, especially in its advanced stages, and are exacerbated by dopamine agonists as a class. However, one can draw no conclusions about individual drugs from this review.

The effects of pramipexole, cabergoline, and levodopa on daytime sleepiness have been assessed in three groups of patients with Parkinson’s disease [ ]. The first group was 19 patients taking pramipexole (mean age 60 years, mean dosage 4.5 mg/day, 8 as monotherapy, the remainder with levodopa); the second group was 22 patients taking cabergoline (mean age 63 years, mean dosage 4.1 mg/day, 10 as monotherapy); the third group was 14 patients taking levodopa (mean age 69 years, mean dosage 789 mg/day) as monotherapy. The scores on the Epworth Sleepiness Scale were virtually identical (8.0, 8.1, 8.1). Scores of greater than 16, indicating excessive daytime sleepiness, were also evenly distributed across the three groups, and were attributed to only four individuals; there were no reports of sleep attacks.

These results have been supported by a study in 160 patients with Parkinson’s disease (mean age 66 years) divided into four equal groups and taking levodopa alone, levodopa with bromocriptine, levodopa with ropinirole, or levodopa with pramipexole [ ]. They were compared with 40 healthy younger controls (mean age 58 years). All the subjects were evaluated using the Epworth Sleepiness Scales. All the drug regimens were associated with increased sleepiness compared with the controls but there were no distinct sleep attacks. The authors noted that all the drug regimens produced similar levels of sedation, although the data appeared to show a trend toward greater somnolence with ropinirole and more particularly pramipexole.

However, these are not uncontested findings. In a study that was described as prospective, but was strictly speaking not, 236 patients (106 men, 130 women, mean age 67 years) were questioned [ ]. All but one was in Hoehn-Yahr stages II–IV. Sleep attacks were reported by 72 patients, of whom 70% were considered to have autonomic dysfunction. The authors concluded that the highest risk was attributable to ropinirole (OR = 7.35), followed by bromocriptine (5.78) and lisuride (5.68); for comparison, the figure for levodopa was 0.61. No patients took pramipexole. Again, it is uncertain whether sleep attacks can truly be distinguished from “ordinary” sleepiness, but the results emphasize the sedative potential of dopamine receptor agonists.

On the other hand, these results have been disputed in another report, with authors in common with the previous report. They presented four case studies, three men and one woman, aged 49–87 years, who had had Parkinson’s disease for 8–17 years; the longest duration was in the youngest patient [ ]. All four were taking levodopa (300–1500 mg/day) plus a decarboxylase inhibitor. All four reported sudden irresistible sleep episodes, even during conversations: two also complained of sleepiness distinct from these attacks. Two of the patients also had definite autonomic disturbances. The authors concluded that sleep attacks do occur and may be associated with levodopa as well as with dopamine receptor agonists.

The sedative potential of levodopa has been examined in 16 healthy volunteers (mean age 25 years, eight men), although of course this represents a very different neuropsychological context from Parkinson’s disease [ ]. They were given levodopa 200 mg or a placebo as a single dose, with domperidone to minimize peripheral adverse effects, followed by the active drug or control in a crossover design. Sedation was assessed using a visual analogue scale and reaction time by the response to a light stimulus. There was no overall change in reaction time, but this concealed considerable interindividual variation. Four subjects complained of nausea and one of excitation after levodopa and all had slower reaction times without complaining of sedation. However, seven subjects reported a greater level of sedation with levodopa than placebo, and this correlated (r = 0.7) with their reaction times. In other words, levodopa can cause sedation in normal young people but does not necessarily do so, at least after a single dose. The extent to which these results can be extrapolated to elderly parkinsonian patients is debatable.

In a Canadian survey of 638 parkinsonian patients who were fully independent and with no evidence of dementia, just over half (327) had excessive daytime sleepiness, as assessed by the Epworth Sleepiness Scale and the Inappropriate Sleep Composite Score [ ]. Only 16 patients had fallen asleep while driving and of these 3 had no warning; all had high scores on the scales mentioned. Unfortunately, the structure of the survey did not allow conclusions regarding drug therapy, although all the patients were taking levodopa (80%), or a dopamine receptor agonist, or both.

In another survey, from Texas, Epworth scores were studied in 368 patients with Parkinson’s disease and 243 with other neurological disorders, mostly other movement disorders, including other causes of parkinsonian syndromes [ ]. The former had scores 25% higher; however, it should be noted that matching for age and other parameters was not clearly described. Using a multivariate model the authors calculated that levodopa dosage, the use of dopamine receptor agonists, and the severity of the disease accounted for only a minor extent (about 9%) of the variation in scores in the parkinsonian group. Even so, they concluded that progression of the disease, the dosage of levodopa, and the use of dopamine receptor agonists are contributory factors in increased daytime somnolence.

Several smaller-scale studies and case reports have also been published. An Italian group have reported polysomnographic studies in a 44-year-old woman with a 5-year history of idiopathic Parkinson’s disease [ ]. She was taking modified-release levodopa, to which pergolide 3 mg/day was added, titrating up from a lower dose. After taking this dosage for 5 days she developed irresistible sleep attacks up to 5 times a day with no warning or obvious triggers. Polysomnography showed that these episodes closely resembled narcolepsy. The episodes ceased after pergolide withdrawal.

Six German patients (aged 36–68 years, disease duration 7–12 years, four men) were studied after reporting sleep attacks following the addition of pramipexole or ropinirole to a levodopa-based treatment regimen [ ]. Five had abnormal results in the multiple sleep latency test, and three of these also had increased errors in a test of driving vigilance. The authors concluded that patients should be assessed individually in terms of their safety to drive. They also suggested that sleep attacks are at the extreme end of the spectrum of daytime sleepiness, although it is not clear that such a conclusion can be reached so unequivocally from such a small study.

A French group has tested the effects of ropinirole 3 mg on sleep latency in 20 healthy men (mean age 24 years) compared with placebo [ ]. The time to sleep onset was significantly quicker in the ropinirole-treated subjects (2.6 versus 4.8 minutes), but sleep attacks did not occur. Clearly this is a very different group of individuals from the usually much older group with Parkinson’s disease, so again definite conclusions are elusive.

Overall, the current evidence suggests that sedation is a class effect of all dopaminergic drugs, including levodopa. It may be more severe with the newer synthetic agents pramipexole and ropinirole, but this cannot yet be stated with certainty. The existence of discrete sleep attacks also remains controversial, although on balance one would conclude that they can occur.