OUP user menu

Prospective randomized trial of lisuride infusion versus oral levodopa in patients with Parkinson’s disease

Fabrizio Stocchi , Stefano Ruggieri , Laura Vacca , C. Warren Olanow
DOI: http://dx.doi.org/10.1093/brain/awf214 2058-2066 First published online: 1 September 2002

Summary

Motor complications are a major source of disability for patients with advanced Parkinson’s disease. Surgical therapies provide benefit to some, but these treatments are expensive and associated with adverse effects. Current research indicates that motor complications are associated with abnormal, intermittent, pulsatile stimulation of denervated dopamine receptors using short acting dopaminergic agents such as levodopa. Retrospective studies suggest that the use of longer‐acting more continuous dopaminergic therapies can improve both motor fluctuations and dyskinesia. We performed a prospective, long‐term (4‐year) trial comparing patients randomized to receive subcutaneous infusion of the dopamine agonist lisuride versus conventional therapy with oral levodopa and dopamine agonists. We demonstrate that patients receiving lisuride infusions experienced a significant reduction in both motor fluctuations and dyskinesia compared with patients receiving standard dopaminergic therapies. Benefits persisted for the 4‐year duration of the study. Mean Unified Parkinson’s Disease Rating Scale scores in ‘ON’ and ‘OFF’ states did not significantly change between baseline and 4 years for patients in the lisuride group, but deteriorated in patients in the levodopa group. This study indicates that continuous lisuride infusion can be beneficial for patients with advanced Parkinson’s disease and reverse established motor fluctuations and dyskinesia.

  • Keywords: lisuride; Parkinson’s disease; randomized trial; prospective trial; dyskinesia
  • Abbreviations: AIMS = Abnormal Involuntary Movements Scale; MPTP = 1‐methyl‐4‐phenyl‐1,2,3,6‐tetrahydropyridine; UPDRS = Unified Parkinson’s Disease Rating Scale

Introduction

Chronic levodopa therapy is associated with the development of disabling motor complications in the majority of Parkinson’s disease patients (Obeso et al., 2000a; Schrag and Quinn, 2000). These can be difficult or impossible to treat satisfactorily with currently available medical therapies (Stocchi et al., 1997; Olanow et al., 2001) and frequently lead to surgical interventions despite their potential for serious adverse events and high cost (Olanow and Brin, 2001). Recent studies suggest that motor complications are due to a sequence of events that involve abnormal pulsatile stimulation of striatal dopamine receptors with consequent plastic changes in downstream neurones and alterations in the firing patterns of basal ganglia output neurones (Olanow and Obeso, 2000). Oral formulations of levodopa are thought to be prone to cause pulsatile stimulation of dopamine receptors because the drug has a relatively short plasma half‐life. Under normal circumstances, dopamine receptors in the striatum are exposed to relatively constant levels of dopamine. With advancing Parkinson’s disease, there is a progressive loss in striatal dopaminergic terminals and in their capacity to store and regulate the release of dopamine. This could result in a loss in their ability to buffer fluctuations in the plasma concentration of a short‐acting dopaminergic agent and result in striatal dopamine receptors being alternately exposed to pathologically high and low levels of activation. This has led to the concept that longer‐acting or more continuous delivery of a dopaminergic agent might prevent or reverse motor complications in Parkinson’s disease (Chase et al., 1989).

Studies in 1‐methyl‐4‐phenyl‐1,2,3,6‐tetrahydropyridine (MPTP)‐treated monkeys have shown that long‐acting dopamine agonists are associated with a reduced frequency and severity of motor complications in comparison to regular formulations of levodopa (Bédard et al., 1986; Pearce et al., 1998). Similarly, prospective, double blind, controlled trials have demonstrated that patients randomized to initiate symptomatic therapy with a long‐acting dopamine agonist have a delay in onset and reduced frequency of motor complications in comparison to standard oral formulations of levodopa (Parkinson Study Group, 2000; Rascol et al., 2000). There is less data with respect to the potential of long‐acting dopaminergic therapies to benefit patients with established motor complications. A recent study demonstrated that chronic administration of the long‐acting dopamine agonist cabergoline reversed established dyskinesias in MPTP‐treated monkeys (Hadj Tahar et al., 2000). In Parkinson’s disease patients with motor fluctuations and dyskinesia, several open label uncontrolled studies have reported benefits with continuous administration of dopamine agonists or levodopa (Vaamonde et al., 1991; Colzi et al., 1998; Sage et al., 1988; Stibe et al., 1988; Nutt et al., 2000). Here, we report the results of a 4‐year prospective, randomized, open label trial in patients with advanced Parkinson’s disease complicated by motor fluctuations and dyskinesia comparing treatment with continuous infusion of the dopamine agonist lisuride versus conventional oral dopaminergic therapies (levodopa and dopamine agonists). This is the first randomized and controlled study to compare conventional and continuous infusion therapies in patients with advanced Parkinson’s disease. We demonstrate that patients receiving lisuride infusions experienced a significant reduction in both ‘OFF’ time and dyskinesia in comparison to patients receiving oral doses of standard dopaminergic therapies.

Methods

The study was conducted as a four‐year prospective, randomized, open label trial in patients with advanced Parkinson’s disease. Parkinson’s disease was diagnosed according to the criteria established by the London Brain Bank (Hughes et al., 1992). All patients had levodopa‐responsive motor features complicated by motor fluctuations and dyskinesia that could not be satisfactorily controlled with levodopa or other anti‐parkinsonian medications. Exclusion criteria included atypical or secondary parkinsonism, previous neurosurgical therapy for Parkinson’s disease, the use of neuroleptic drugs, dementia that precluded providing informed consent, history of hallucinations or other psychotic features, and unstable medical or laboratory conditions. Patients who met all entry criteria and signed an institutional review board‐approved informed consent according to the Declaration of Helsinki were assigned to one of two treatment groups according to a computer‐generated randomization schedule. The study was approved by the ICCS ‘Neuromed’ Ethical Committee. One group was randomized to receive treatment with subcutaneous lisuride infusions administered by a minipump continuously throughout the day (the lisuride group). The second group was treated with standard oral levodopa combined with a decarboxylase inhibitor (either carbidopa or beneserazide) plus such other anti‐parkinsonian medications as were deemed appropriate (the levodopa group). Patients were evaluated at baseline and at 6‐monthly intervals throughout the study, but could see the physician for dosage adjustment at any time. Neither patients nor physicians were blinded, but separate investigators performed study evaluations (L.V., S.R.) and therapeutic adjustments (F.S.). All evaluations were performed by the same investigator. At each visit, patients were hospitalized for formal testing. Evaluations included:

(i) The Unified Parkinson’s Disease Rating Scale (UPDRS) in the practically defined ‘OFF’ state (∼12 hours after stopping anti‐parkinsonian medication the evening before) and in the best ‘ON’ state (∼1–2 h after starting morning anti‐parkinsonian medication when patients were fully on) (Fahn et al., 1987; Langston et al., 1992).

(ii) A dyskinesia assessment using the Abnormal Involuntary Movements Scale (AIMS) rating scale to record the most severe dyskinesia that occurred during best ‘ON’ state (Simpson et al., 1979).

(iii) Home diary assessments recording time in ‘ON’ and ‘OFF’ states during the 2 days prior to each visit. Patients and/or their carer were trained in the completion of the diary prior to entry into the study.

In addition, nocturnal akinesia and dystonia were each assessed on a 0–4 scale (0 = none and 4 = worst) in the morning prior to performing practically defined ‘OFF’ evaluations.

The primary outcome variable was the change between baseline and final visits in the number of hours spent in ‘OFF’ time during the waking day. Secondary endpoints included change from baseline to final visit in AIMS dyskinesia score and in UPDRS motor score during ‘ON’ and ‘OFF’ stages. Statistical comparison between groups was performed using the Wilcoxon Rank Sum Test. The Wilcoxon Signed Rank test was used to compare paired data.

Treatment regimen

Patients randomized to the lisuride group were administered their regular evening dose of anti‐parkinsonian medication and oral medications were then discontinued. The following morning, a subcutaneous infusion containing lisuride hydrogen maleate (Schering AG, Berlin, Germany) at a concentration of 1 mg diluted in 1 ml of sterile water was initiated using a modified programmable insulin pump (Hoechst MRSI, MRSII, Milan, Italy, Canè CronoPar, Turin, Italy). The infusion was administered daily between the hours of 8 am and 8 pm, and discontinued overnight. The dose was adjusted during the first week of treatment to obtain the best clinical response and could be re‐adjusted at any time throughout the study. In patients who were not completely mobile during the day or who had nocturnal dysfunction, oral levodopa plus a peripheral decarboxylase inhibitor could be added. No other anti‐parkinsonian medications were administered during the course of the trial. The infusion pumps hold 5–10 ml of solution and were refilled every 3–10 days. The needle was placed subcutaneously into the abdominal wall, and could be changed every 2–5 days—although the majority preferred to remove the needle each night. All patients received treatment with oral domperidone (60 mg/day) for ∼3 months to prevent peripheral dopaminergic side effects. Blood pressure and electrocardiograms were monitored during the 24 h preceding infusion therapy, during the first day of the lisuride infusion, and following 1 week of treatment. Patients randomized to the levodopa group were maintained on their pre‐trial levodopa regimen. The levodopa dose and frequency could be manipulated and other anti‐parkinsonian agents such as dopamine agonists could be introduced at any time throughout the study in order to obtain maximal clinical benefit.

Results

Forty patients gave informed consent and were enrolled into the study. Patients in the lisuride group had slightly more severe parkinsonism as evidenced by higher UPDRS scores in the ‘ON’ state (P = 0.04), but ‘OFF’ time and dyskinesia severity were comparable in the two groups (Table 1). There were no other significant differences in baseline demographic variables between the two groups. Two lisuride‐treated patients withdrew from the study during the first 3 months of treatment because of their inability to cooperate with the treatment protocol. The remaining 38 patients completed the 4 years of the trial with no missed visits. The number of hours spent in the ‘OFF’ state over the course of the study in the lisuride and levodopa groups is shown in Fig. 1. In comparison to baseline, mean ‘OFF’ time improved by 59.3% at 4 years in the lisuride group, while it worsened by 21.4% in the levodopa group (P < 0.0001). After 4 years of treatment, patients on lisuride infusion therapy had a mean (± SD) of 1.2 ± 0.7 h of ‘OFF’ time per day compared with a mean of 4.1 ± 1.3 h at baseline (P < 0.0001) while patients in the levodopa group had a mean of 5.1 ± 0.7 h of ‘OFF’ time at 4 years compared with a mean of 4.2 ± 1.1 h at baseline (P < 0.0001). After 4 years of treatment, 12 out of 18 patients in the lisuride group had ≤1 h ‘OFF’ time per day, three had no ‘OFF’ time at all, and none had more than 2 h ’OFF’ per day. In contrast, all patients in the levodopa group had >4 h of ‘OFF’ time per day.

Fig. 1 Mean number of ‘OFF’ hours ± SD during the working day in patients in the lisuride and levodopa groups. Note that at each visit, patients receiving lisuride infusions had a significant decrease in the number of ‘OFF’ hours compared with patients in the levodopa groups (P < 0.0001). There were 20 patients in the levodopa group at all visits. There were 20 patients in the lisuride group at baseline and 18 at each other visit.

View this table:
Table 1

Baseline demographics

Levodopa (± SD)Lisuride (± SD)
Number2020
Age53.8 ± 8.752.6 ± 7.2
Disease duration (years)8.8 ± 2.710.0 ± 2.1
Motor UPDRS ‘ON’21.7 ± 5.125.1 ± 5.2
Motor UPDRS ‘OFF’44.4 ± 10.049.7 ± 6.9
Hoehn–Yahr score (0–4)3.8 ± 0.74.0 ± 0.8
No. with motor complications (%)100100
No. with dyskinesia (%)100100
Hours off during day4.2 ± 1.14.1 ± 1.3
AIMS score (0–4)2.2 ± 0.92.2 ± 0.7
Levodopa dose (mg)675 ± 180.6688.2 ± 133.4
Duration of levodopa therapy (years)8.1 ± 2.49.1 ± 2.3

The effect of lisuride infusion versus levodopa treatment on dyskinesia is shown in Fig. 2. In comparing the baseline and 4 year visits, dyskinesia scores decreased (improved) by 49% in the lisuride group and increased (worsened) by 59% in the levodopa group (P < 0.0001). In the lisuride group, dyskinesia scores improved from a mean of 2.2 ± 0.7 at baseline to a mean of 1.3 ± 0.5 at 4 years (P < 0.0001), while in the levodopa group they increased from a mean of 2.2 ± 0.9 at baseline to 3.5 ± 0.5 at 4 years (P < 0.0001). No lisuride patient had a dyskinesia score >2, while all patients in the levodopa group had scores of 3 or 4. The benefit of lisuride infusion versus levodopa on dyskinesia is also demonstrated by differences in results of questions 32–34 of the UPDRS, which measure the duration, intensity and discomfort associated with dyskinesia. Patients in the lisuride group showed an improvement from a baseline score of 4.8 ± 1.0 to a score of 1.2 ± 0.7 at 6 months, while there was almost no change for patients in the levodopa group (baseline score of 4.5 ± 1.1; 6 month score of 4.6 ± 1.2 (P < 0.001). This difference persisted over the duration of the study.

Fig. 2 Mean dyskinesia scores ± SD in patients in the levodopa and lisuride groups. Note that at each visit, patients receiving lisuride infusions had a significant decrease in the number of ‘OFF’ hours compared with patients in the levodopa group (P < 0.0001).

The improvement in ‘OFF’ time in lisuride‐treated patients occurred within hours following initiation of the infusion, but improvement in dyskinesia developed gradually over ∼3–12 weeks.

Mean UPDRS motor scores in ‘ON’ and ‘OFF’ states did not change significantly between baseline and 4 years for patients in the lisuride group, but deteriorated in patients in the levodopa group (Fig. 3). Significant improvements over baseline were detected for scores of nocturnal akinesia and dystonia in the lisuride compared with the levodopa group (Table 2).

Fig. 3 Mean UPDRS motor scores ± SD in the (A) practically defined ‘OFF’ and (B) best ‘ON’ states in patients in the lisuride and levodopa groups. Note that there was no significant change from baseline in UPDRS ‘ON’ or ‘OFF’ scores in patients receiving lisuride infusion (P = 0.18 and 0.24, respectively). In contrast, patients in the levodopa group experienced deterioration in mean UPDRS score that was significant in the ‘ON’ state (P < 0.001) and just failed to reach significance in the ‘OFF’ state (P = 0.06).

View this table:
Table 2

Comparison of baseline and final scores of parkinsonian motor function

Levodopa groupLisuride groupP value*
Baseline48 monthsBaseline48 months
Number of patients20202018
Hours ‘OFF’4.2 ± 1.15.1 ± 0.74.1 ± 1.31.2 ± 0.7<0.001
AIMS2.2 ± 0.93.5 ± 0.52.2 ± 0.71.3 ± 0.4<0.001
Motor UPDRS ‘ON’21.7 ± 5.128.2 ± 2.725.1 ± 5.226.8 ± 1.8ns
Motor UPDRS ‘OFF’44.4 ± 10.049.7 ± 7.449.0 ± 6.946.5 ± 6.4ns
UPDRS 32–34 (sum)4.5 ± 1.15.85 ± 1.04.8 ± 1.01.8 ± 1.2<0.001
Dystonia1.9 ± 0.42.5 ± 0.72.0 ± 0.20.2 ± 0.1<0.001
Nocturnal akinesia2.5 ± 0.33.9 ± 1.62.4 ± 0.61.5 ± 0.8<0.001
Levodopa (mg)675 ± 180.61032.5 ± 144.4688.2 ± 133.4333.3 ± 89.9<0.001
Lisuride (mg/h)0000.9 ± 0.2

*P value comparing change from baseline between the levodopa and lisuride groups; ns = not significant

At the 4‐year visit, patients in the lisuride group were receiving lisuride at a mean infusion rate of 0.91 ± 0.17 µg/h. All received supplemental levodopa during the day, but none received levodopa during the night and no other anti‐parkinsonian medications were employed. In comparison to baseline, the mean levodopa dose was decreased by a mean of 51.6% for patients in the lisuride group (688.2 ± 133.4 to 333.3 ± 89.9; P < 0.001) while the levodopa dose was increased by 53% for patients in the levodopa group (675 ± 180.6 to 1032.5 ± 144.4; P < 0.001). Fifteen patients in the levodopa group were treated with a supplemental oral dopamine agonist (lisuride: 7 patients, range 1.5–4.5 mg; bromocriptine: 8 patients, range 15–60 mg) and seven were treated with selegiline.

Adverse events are listed in Table 3. For the most part, they are typical of what is seen with dopaminergic therapies and none led to patient withdrawal from the study. Hallucinations and psychiatric complications were more common in the lisuride group, but were not severe. Eleven patients receiving chronic lisuride infusion experienced skin nodules but, in general, these were mild and did not compromise continued therapy. Peripheral oedema was noted in four patients in the levodopa group, but was attributed to concomitantly employed bromocriptine in each case.

View this table:
Table 3

Adverse effects

Lisuride groupLevodopa group
Nausea42
Psychiatric complications31
Hypotension32
Peripheral oedema04
Hypersexuality31
Skin nodules110

Discussion

There has been considerable interest in the possibility that treatment with more continuously acting dopaminergic therapies might provide benefit to patients with advanced Parkinson’s disease (Nutt et al., 2000). We present the first prospective randomized study comparing continuous versus intermittent administration of dopaminergic therapies to patients with advanced Parkinson’s disease. We demonstrate that patients randomized to receive continuous subcutaneous daytime infusions of lisuride experienced significant improvement in both ‘OFF’ time and dyskinesia compared with patients randomized to receive traditional oral levodopa treatment with or without other medical therapies (see Figs 1 and 2). Patients randomized to receive lisuride infusions experienced a significant improvement in both ‘OFF’ time and dyskinesia compared with baseline that persisted throughout the 4‐year duration of the study. Further, there was no decline in UPDRS motor scores during ‘ON’ or ‘OFF’ states, suggesting that tolerance did not develop using this protocol. In contrast, patients randomized to receive oral levodopa experienced significant worsening of both dyskinesia and ‘OFF’ time in comparison to baseline and UPDRS scores tended to worsen.

These changes do not tell the whole story. The majority of patients receiving lisuride infusion experienced a motor response that was near normal and occurred for the most part without complicating dyskinesia. Lisuride patients had virtually no troublesome dyskinesia, while troublesome dyskinesia remained a problem in all patients in the levodopa group. Lisuride benefits began ∼20 min after starting the infusion in the morning and wore off ∼30–60 min after the infusion was stopped at night. Nonetheless, some benefits persisted during the night and measures of nocturnal function were significantly improved compared with those in the levodopa group. Thus, patients treated with continuous daytime lisuride infusions were able to function relatively normally with minimal motor complications throughout the 4‐year duration of the study. These benefits are all the more remarkable when one considers that an entry criterion for this trial was severe motor complications that could not be controlled with medical therapy. Indeed, patients treated with conventional dopaminergic therapies continued to deteriorate over the course of the study despite further increases in levodopa dosage and the introduction of oral dopamine agonists. The severity of disability in patients participating in this study was such that they might otherwise have been considered for a surgical therapy. The benefits obtained with continuous infusion of lisuride are comparable with those observed with surgical procedures, without the incumbent risks and costs (Olanow and Brin, 2000).

It is interesting to consider the mechanism responsible for the benefit associated with continuous infusion of lisuride. Dopamine neurones normally exhibit a relatively constant firing at a rate of ∼3–4 cycles per second (DeLong et al., 1983). Phasic increases in dopamine neuronal firing occur in response to novel stimuli or in anticipation of reward (Strecker and Jacobs, 1985; Ljungberg et al., 1992; Schultz, 1986) and burst firing can be observed in response to glutamate activation (Johnson et al., 1992). However, the consequent increase in dopamine release does not lead to a rise in extracellular concentrations of dopamine, presumably because of the extremely rapid reuptake capacity of presynaptic dopamine terminals (Grace, 1991). Thus, synaptic dopamine concentrations and levels of dopamine receptor activation tend to remain relatively constant under physiologic circumstances.

Considerable evidence now indicates that abnormal intermittent or pulsatile activation of brain dopamine receptors contributes to the development of motor complications in Parkinson’s disease through the induction of plastic changes in striatal neurones and altered neuronal firing patterns (Olanow and Obeso, 2000; Obeso et al., 2000b). In the parkinsonian state, there is a loss of striatal dopaminergic terminals and their ability to store and regulate the release of dopamine. Accordingly, striatal dopamine receptor activation becomes increasingly dependent on the peripheral availability of exogenously administered dopaminergic agents. Under these circumstances, it has been proposed that fluctuations in the plasma level of a short‐acting dopaminergic agent such as levodopa may not be adequately buffered and result in striatal dopamine receptors being exposed to alternating high and low levels of activation (Obeso et al., 2000c; Olanow et al., 2000). Indeed, in rats with unilateral 6‐hydroxydopamine lesions, levodopa treatment produces a 35‐fold increase in extracellular dopamine concentration in the lesioned striatum compared with a 2‐fold increase in the intact striatum (Abercrombie et al., 1990). PET studies in Parkinson’s disease patients using raclopride binding as a marker of synaptic dopamine levels similarly suggest that the extracellular dopamine concentration is higher in the more affected striatum following a dose of levodopa (Tedroff et al., 1996). Studies in MPTP monkeys further demonstrate that non‐physiological, pulsatile stimulation of dopamine receptors induces gene and protein changes in striatal neurons (Calon et al., 2000) and alterations in the neuronal firing pattern of basal ganglia output neurons (Filion et al., 1991) that are associated with the development of motor complications.

These observations suggest that long‐acting dopaminergic agents that provide more continuous stimulation of dopamine receptors might be associated with a reduced risk of motor complications (Chase et al., 1989; Olanow et al., 2000). In MPTP monkeys, several studies have shown that long‐acting dopamine agonists are associated with a reduced frequency and severity of motor complications than regular formulations of levodopa or short‐acting dopamine agonists (Bédard et al., 1986; Pearce et al., 1998; Gomez‐Mancilla et al., 1991; Blanchet et al., 1993; Gomez‐Mancilla and Bédard, 1992; Jenner, 2000). Indeed, the same short‐acting dopamine agonist induces dyskinesia when it is administered intermittently, but not when it is infused in a continuous manner (Blanchet et al., 1995). Prospective double blind clinical trials in untreated Parkinson’s disease patients similarly demonstrate that the risk of inducing motor complications is markedly reduced if symptomatic therapy is initiated with a long‐acting dopamine agonist compared with a short‐acting formulation of levodopa (Parkinson study group, 2000; Rascol et al., 2000).

It is less clear if established motor complications can be reversed with long‐acting or continuous dopaminergic therapies. In MPTP monkeys, cabergoline has been shown to reverse levodopa‐induced motor complications (Hadj Tahar et al., 2000). In Parkinson’s disease patients with motor complications, several open label non‐controlled studies have examined therapies designed to provide more continuous dopaminergic stimulation. Continuous infusions of levodopa, apomorphine or lisuride have been shown to consistently reduce ‘OFF’ time and the severity of motor fluctuations (Sage et al., 1988; Stibe et al 1988; Vaamonde et al., 1991; Hughes et al., 1993; Kurth et al., 1993; Stocchi et al., 1993; Gancher et al., 1995; Nutt et al., 1997; Nilsson et al., 1998; Colzi et al., 1998; Pietz et al., 1998; Syed et al., 1998). In some studies, as with our own, improvement in ‘OFF’ time was associated with a reduction in the severity and duration of dyskinesia (Colzi et al., 1998). As in our study, improvement in ‘OFF’ time occurred with the initiation of infusion therapy indicative of a pharmacological effect, but anti‐dyskinesia benefits worsened at first and improvement only occurred after weeks to months of treatment, suggesting that these benefits are related to a plastic effect that occurs in association with continuous dopaminergic therapy. We do not think the anti‐dyskinesia effect was related to lowering the dose of levodopa, as the medication was discontinued immediately prior to starting the infusion in all patients, yet benefits with respect to dyskinesia did not occur for weeks to months. The use of an infusion pump to deliver a dopaminergic drug to the brain in a relatively continuous fashion may not precisely mirror what occurs in the normal brain, but it may be less abnormal and less prone to induce motor complications than the intermittent administration of short‐acting formulations of levodopa.

Adverse effects were not a major problem in our study. Only two patients withdrew from the study and this was because they could not comply with the protocol rather than because of any specific side effect. Psychiatric complications have been a problem with continuous administration of dopamine agonists, particularly when the infusion is administered around the clock (Vaamonde et al., 1991; Pietz et al., 1998). This is one of the reasons we administered infusions only during the waking day. Psychiatric side effects were not a major problem with this protocol. Eleven patients had skin nodules, but these were mild and not of the severity reported with apomorphine, which requires infusion of a larger volume of fluid (Hughes et al., 1993; Gancher et al., 1995).

While the lisuride infusion technique is demanding, only two patients withdrew over the course of this 4‐year study. Changing the infusion needle at night and using catheters of appropriate length for each patient helped to make patients more comfortable. We chose to infuse lisuride rather than apomorphine because of our impression that skin problems at the injection sites are less of a problem. Patients in the control group were maintained on best medical therapy despite the superiority of lisuride infusion because no other options were available to them during this time. Surgery was not yet being performed, and lisuride is an experimental drug that was only available in sufficient supply to maintain therapy for those in the infusion arm of the study.

Our study is the first trial to compare infusion of a dopaminergic agent to conventional oral dopaminergic therapies in a prospective, randomized long‐term trial. It would have been preferable to employ a double blind protocol, but this was not practical given the nature and duration of the trial. We did, however, utilize a separate investigator to perform clinical evaluations and the same investigator performed all clinical assessments. This investigator was not involved in any therapeutic decision and only had contact with the patient at the time of the examination. This study was designed as a proof of principle study to determine whether the continuous delivery of a dopaminergic agent could provide benefit to patients with advanced Parkinson’s disease complicated by motor flucutuations and dyskinesia, which could not be controlled with more traditional treatment approaches. The ability to utilize continuous infusion strategies in patients with advanced Parkinson’s disease is difficult to implement, time‐consuming, and may not be practical for routine practice. Neither lisuride nor apomorphine are available in all countries, nor is the peripheral dopamine antagonist domperidone—which may be necessary in order to prevent peripheral dopaminergic side effects. Intrajejeunal levodopa administration requires an invasive procedure and can be difficult to manage for both patient/carer and physician. It remains to be determined if we will be able to simulate the effects of continuous dopamine stimulation with currently available oral dopaminergic agents. Perhaps this can be accomplished through the use of a very long‐acting dopamine agonist such as cabergoline, dopamine agonists such as ropinirole or pramipexole administered frequently throughout the day, or frequent doses of levodopa/carbidopa preparations administered in combination with a catechol‐o‐methyltransferase (COMT) inhibitor to extend the levodopa elimination half‐life. Transdermal delivery of dopamine agonists may provide stable levels of a dopaminergic agent over the course of the day, but none are yet currently available. Further studies to determine whether dosing regimens with any of these approaches can be developed which provide sustained plasma levels. So too are trials of continuous dopaminergic stimulation strategies in animal models. Our study demonstrates that treatments utilizing a more continuous dopaminergic therapy can provide meaningful benefit to patients with advanced disease who suffer disabling motor complications. While this approach is not easy to administer in its present form, further efforts to investigate these approaches are warranted as they potentially represent an alternative to surgery in this patient population.

Acknowledgements

We wish to thank Schering AG for the supplies of lisuride and Drs David Marsden and Reinhart Horowski for their advice during the course of this project. Grant support was provided by the Bachmann–Strauss Dystonia and Parkinson Foundation and the Lowenstein Foundation.

References

  1. 33.
View Abstract