Acta Neurol Scand 1998: 97: 175-183 Printed in U K - all rights reserved

Copyright 0 Munksgaard 1998 ACTA NEUROLOGICA SCANDINAVICA ISSN 0003-6314

Long-term intraduodenal infusion of a water based levodopa-carbidopa dispersion in very adGanced Parkinsons disease
Nilsson D, Hansson L-E, Johansson K, Nystrom C, Paalzow L, Aquilonius S-M. Long-term intraduodenal infusion of a water based levodopa-carbidopa dispersion in very advanced Parkinsons disease. Acta Neurol Scand 1998: 97: 175-183. 0 Munksgaard 1998.

D. Nilsson, 1.-E.Hansson, K. Johansson. C. Nystrom3. 1. Paalzow3, S.-M. Aquilonius

Departments of Neurology, 2Surgery and 3Pharmacy, University of Uppsala. S-751 85 Uppsala, Sweden

Objective - To evaluate the effects of continuous duodenal infusion of levodopa over time on the disabling fluctuations in motor performance in advanced parkinsonian patients. It has earlier been demonstrated that these fluctuations can be reduced by keeping the plasma concentration of levodopa constant. Material and methods - In view of the low water solubility of levodopa a stable dispersion of the drug was developed and used for continuous intraduodenal infusion in patients with advanced Parkinsons disease. Nine patients were evaluated with respect to an optimal oral treatment, during nasoduodenal infusion by a portable pump and then followed for 6 months to 2% years when treated via transabdominal infusion. Upon each test occasion, over 2 non-consecutive days, objective movement analysis by means of an opto-electronic system was applied every 15-20 min and video recordings performed twice every h. On several test occasions plasma levodopa concentrations were analysed every 15 min. Results - The patients showed improvement and decreased variance of their motor function. In the 2 patients followed over a period of 2% years levodopa plasma concentration showed reduced fluctuations on infusion and the levodopa consumption as well as mean levodopa plasma concentration decreased. Conclusion - Continuous duodenal infusion of levodopa is an alternative treatment strategy for patients with advanced Parkinsons disease when conventional therapy has failed.

Key words. Parkinsons disease, levodopa; duodenal infusion; movement analysis; pharmacokinetics Dag Nilsson. Department of Neurology, University Hospital, S-751 85 Uppsala. Sweden Accepted for publication October 15, 1997

Most patients with Parkinsons disease (PD) when treated orally with levodopa develop fluctuations in motor performance. After 3-5 years of treatment one third, after 5-7 years about half, and after 10-12 years nearly all patients suffer from this problem according to earlier reports (1-2). Also recent long-term studies using slow release formulations of levodopa show a high prevalence of motor fluctuations after 5 years of therapy (3-4). It has earlier been demonstrated that motor fluctuations in advanced PD are at least partially related to variations in blood levodopa concentrations and that such fluctuations can be markedly reduced by keeping levodopa plasma concentrations constant ( 5 , 6 for review). The active uptake of levodopa is most effective in the proximal small

intestine, but due t o erratic gastric emptying orally administered levodopa tablets or capsules create uneven plasma concentrations (7). Also dietary related competition for intestinal uptake mechanisms could influence the biological availability of levodopa (8-9). With this background knowledge, systems for continuous intravenous or intraduodenal infusion of levodopa have been presented (10-14). Up to now, due to the low solubility of levodopa, patients treated with continuous levodopa infusion have carried substantial volumes of infusate. To overcome this problem we introduced a water-based dispersion of levodopa and the decarboxylase inhibitor carbidopa. With a portable pump for intraduodenal delivery this formula was shown effective
175

Nilsson et al.

in a short-term study on advanced parkinsonian patients (15). The present report describes the clinical outcome with respect to 9 patients treated during daytime with this method for periods of 6 months to 2% years. To enable a statistic evaluation of motor performance an objective opto-electronic technique was used during the test days, in parallel to video recordings. Also the plasma levodopa concentrations over time were analysed in 2 of the patients.
Material and methods
Patients

Eight men and 1 woman with very advanced idiopathic PD took part after giving their informed consent. The group consisted of patients with early onset of the disease, long duration of levodopa therapy and no signs of dementia. This selected group of very advanced patients was referred from neurologists within the reception area of our hospital, with a population of about 1.9 million inhabitants, during a period of 2% years. Patients were considered by several experienced neurologists to be on an optimal oral therapy. Clinical data and earlier treatment strategies are presented in Table 1. The study was approved by the Ethics Committee of the Faculty of Medicine, Uppsala University.
Testing procedure

The patients were first tested on 2 non-consecutive days with respect to the optimal oral therapy, and during a similar period with respect to a continuous nasoduodenal infusion of the levodopakarbidopa dispersion. Thereafter the patients were provided

with a permanent tube (Freka@ PEG standard duodenal Ch 9, Fresenius Ltd, Cheshire, UK) to the duodenum via a gastrotomy (Fig. 1).The first 2 operated patients (A and B) were then tested on another 4 occasions on 2 non-consecutive days during a 2% years period. Because of the promising results another 7 patients (C-I) were included, tested in the same way as the first patients, operated and tested on another 3 occasions during 6 months. On each test occasion the catheter tip was situated in the duodenum as shown by X-ray. The test protocols are presented in Fig. 2. On infusion the patients started with a loading dose into the tube in the morning and then immediatly started the pump. In the evening the infusion was stopped at bedtime and the tube was rinsed with water. Oral levodopa treatment (Table 1) in daytime was replaced by the infusion, but most patients continued with unchanged concomitant medication. Up to now the use of scoring protocols has been the standard to evaluate motor performance in Parkinsons disease. In our experience, however, this approach offers a low sensitivity to quantify performance and to detect rapid motor fluctuations. Therefore we have followed the patients very closely from morning to evening on the test days with an opto-electronic system (MacReflex, Qualysis AB, Gothenburg, Sweden), in order to measure a standardized sequence of movements, the PLM test. The PLM test has been described in detail elsewhere (16, 17). In short, markers with reflector tape on the patient and on an object, to be lifted from the floor and forwarded onto a shelf, are registered by a video camera, with an internal infrared light emitting diode, connected to a computer and the total movement time (MT),

Table 1 Patients characteristics. clinical data Symptoms of P.D. since (year) 10 20 Levodopa since (year)

Fluctuations since (year)

6 15

Patient

Age (years)

Hoehn & Yahr at worse

Levodopa mg/day

Other treatment lactual/earlier) none/selegiline. biperiden none/amantadine. orphenadrine, brornocriptine. selegiline. thalamotomy brornocripttne/amantadine, orphenadrine, selegiline. apornorphine orphenadrine, bromocriptine. selegiline/none brornocriptine/selegiline, thalamotorny none/orphenadrine. selegiline bromocriptine. selegiline/orphenadrine brornocriptine/selegiline. apomorphine bromocriptine, selegiline/orphenadrine. thalarnotorny

63 56

9
18

800
600-900

47

10

>4

2500

58
66 48

14 24 12 26 16 23

14 21 6 19
15 20

10 17

700

>5
>3 12 > 14

69 53 62

1100 800-1 200 500 1500 1200

176

Duodenal infusion of levodopa

Fig. 1. The infusion system for levodopa.

a postural- (P), a locomotor- (L) and a manual- (M) phase can be quantitatively described. The PLM test was applied every 15-20 min from 8 a.m to 5 or 7 p.m. In addition, every 30 min the patients were videofilmed when performing a standardized sequence of motor tasks, piano playing, alternating hand movements, rising from a chair and walking. These videos are primarily used to evaluate any change in time spent in different motoric slates (off, on, hyperkinetic). In patients A and B, studied over a period of 2 % years, several rating scales to disclose progress (Hoehn and Yahr, Webster and North Western University disability scale) were applied upon test occasions on mornings when the patients had been drug-free overnight. Further, in these 2 patients plasma samples for analysis of levodopa concentration were taken every 15 min during the days of oral and nasoduodenal treatment, and on the last occasions of transabdominal intraduodenal delivery for analysing levodopa plasma concentration and its variation. In patients C-I, studied over a period of 6 months, the UPDRS rating scale (Historical ADL, Complications of Therapy and Modified Schwab and England ADL scale), was applied during oral therapy and upon the last test occasion involving intestinal infusion (18). Self rating scales to be completed at home were also distributed to the 7 patients, but all but 1 failed to complete the diaries.

Case A and B
1.

2.
4.

1. 2.
4.

1.

2. 4.

1. 2. 4.

1. 2. 4.

1. 2. 4.

5.

5.

5.

Case C-l
Oral Nasod

OD

I.
2.
3.

1.

Duodenal 1.

Duodenal 1.

Duodenal 1.

2.

2.

2.

2. 3.

-3(-2) -2(-1)

mnths

Fig. 2. Time schedule for test procedures of oral, nasoduodenal (nasod) and transabdominal duodenal infusion of the levodopa/ carbidopa dispersion. Op 4 indicates time of gastrotomy. 1. PLM test. 2. Video scoring. 3. UPDRS. 4. Disability scales (Hoehn and Yahr, Webster, North Western University Disability Scale). 5. Levodopa analysis. 6 Test day. Each day the patients were tested from 8 a.m. to 5 or 7 p.m.

177

Nilsson et al.
Table 2. Case A and 8. Movement time (MT)(sec)fvariance (var)(sec*) the 2 days at each test occasion Test occasion Day 1 Case A Tablets Nasoduodenal'""' Infusion 8 months"""' Infusion 12 months***' Infusion 18 months"""' Infusion 31 months*** Case 8 Tablets Nasoduodenal" Infusion 8 months"' Infusion 10 months*^ Infusion 18 months" Infusion 31 months**
(A)

M T fvariance

Test occasion Day 2

M T fvariance

2.49 f0.36 2.12 f0.10 1.75 f0.08 1.88 f0.09 1.82 f 0.07 2.09 f0.20 2.73 f 0.25 2.47 f0.28 2.31 fO.12 2.20 f O . 1 1 2.16f0.15 2.32 f0.13

44 45 28 28 27 45 43 40 28 28 27 45

Tablets Nasoduodenal***^ Infusion 8 months***A Infusion 12 months***' Infusion 18 months***' Infusion 31 months*** Tablets Nasoduodenal*^ Infusion 8 months Infusion 10 months"' Infusion 18 months" Infusion 31 months"

2.36 & 0.37 2.15k0.14 1.89i o . 1 0 1.88fO.09 1.80 f0.05 2.06 f0.18 2 55f0.23 2.39 f 0.15 2.51 f0.33 2.19 fO.12 2.33 fO.20 2.41 fO.18

45 43 28 28 28 45 44 43 28 28 27 45

P<O 05 indicates significant reduction in motor variance compared to oral therapy. (*) P<0.05, ("""I P<O.OOl indicates significant improvement in M T compared to oral therapy. n=number of PLM tests.

The drug delivery system and the dispersion

The water-based dispersion contained 20 m g / d and 5 mg/ml of micronized levodopa and carbidopa respectively in methylcellulose. The dispersion for this study was produced ex tempore by the Hospital Pharmacy, University Hospital, Uppsala, Sweden. Chemical stability was tested by sampling from batches of the dispersion during a 2-week period, and through measuring the concentrations of levodopa and carbidopa by a HF'LC-method with EC-detection (15). The dispersion was found stable with respect to both levodopa and carbidopa concentrations. The dispersion was dispensed in cassettes of 50100m1, for 1-3 days' treatment, connected to a portable pump (CADD-PLUS@, Pharmacia Deltec, Stockholm, Sweden). The delivery rate of the dispersion from the pump showed a maximum variation of 10%. From the Hospital Pharmacy cassettes for 1week consumption were delivered to the local pharmacy of each patient. The cassettes were stored in the refrigerator and patients were instructed to turn over the cassettes twice a day to maintain a homogenous dispersion.
Statistics

In patients A and B, MT (movement time) on infusion was compared to corresponding MT during oral therapy (the 1st and the 2nd day, separately) and the non-parametric Kruskal- Wallis test was applied for statistical analysis (19). To obtain a comparison of MT variability the Squared Ranks test for variances was used (19). For each patient C-I the median MT values of the individual 1st and 2nd day on oral, and 1st and
178

2nd day on each test occasion on infusion therapy were calculated. These median values of day 1 and 2 on infusion were used to calculate the mean median MT of all four test occasions, in each patient. These median MT values on infusion therapy were compared to the median MT values of day 1 and 2 of corresponding days during oral therapy by means of ANOVA (repeated measures, with Dunnet's t-test for multiple comparisons). For the analysis of variation in MT on oral and infusion test days in the patients C-I, the variance in MT for each test day was calculated. These variances in MT were then ranked in the individual patient and the exact probability for the obtained ranking of the oral days was calculated in each patient. The probability figure from each of the 7 patients (n=7) were transformed to z-value, the z-values added and divided with the ,In. The motor performance videotaped every 30 min, was non-blindly scored from -3 to +3. To estimate the change in the time the patients spent in different motoric states, a "normal" state was defined to correspond to the function between -1, 0 or +1 on the video ratings, an "off" state to -3 and -2 and a "hyperkinetic" state to +2 and +3. The analysis was done by comparing the time spent in the 3 different motoric states on oral therapy and on infusion therapy at all test occasions. Differences in mean plasma levodopa concentration in patients A and B were compared using the Kruskal-Wallis test. As the levodopa concentration has a tendency to accumulate during the days on continuous therapy, and as the interesting parameter is not only the variation from morning to evening but the frequent and quick plasma fluctuations, the variances in plasma concentration

Duodenal infusion of levodopa

on oral and duodenal days were compared by means of a simple F-test on the variances about regression. Analyses were performed by means of Stat View 4.02 (Abacus Concepts, Inc., Berkeley, CA).
5
co
o
yl

m m m N N N

m m N N
P I N
0 0

m z m
0 0

Results
Motor function

r -

$cng
- N -

m N

r-m

5
c

c
L

ww wwww
PIN

e m
N -

-m e ? -w m N N N N

In the patients A and B, followed for 2% years, the total movement time (MT) in the PLM test and, on most test occasions, the variance in motor performance decreased significantlyupon intraduodenal infusion as compared to the initial oral treatment with levodopa corresponding to improved speed and less variation in motor function (Table 2). However, between the test occasions at 16-18 months and 28-30 months of continuous treatment there was a tendency towards deterioration in both subjects. According to the video evaluations patients A and B spend more time (35%) in normal motoric state on infusion, less time in both off state (31%) and hyperkinetic state (34%) as compared to corresponding figures on oral therapy, 9%, 54% and 38% respectively. No significant disease progress according to disability scales was noted in patients A and B during the 2% year observation period. On oral therapy patients C-I performed faster according to MT on the 1st day as compared to the 2nd. This could be explained by the new inspiring test situation (there was no such statistical difference in MT between the 1st and 2nd day on infusion). As compared to the 2nd day on oral therapy, MT decreased significantly (P=0.05) on both the 1st and 2nd day on infusion. Also the variances in MT decreased on infusion compared to oral therapy (P=O.OOl) (Table 3). These results are well in accordance with those of patients A and B. Ratings from the video films are in accordance with the PLM test results as patients C-I spend more time (54%) in normal motoric state on infusion, less time in both off state (12%) and hyperkinetic state (34%) as compared to corresponding figures on oral therapy, 30%, 17% and 53 YO respectively. According to the historical ADL data of the UPDRS, obtained for oral treatment as compared to the last test occasion on infusion, 5 out of the 7 patients C-I subjectively improved. Six of the 7 had fewer complications (dyskinesia, dystonia and fluctuations) due to therapy (Fig. 3). During infusion 3 patients claimed improvement and 4 were unchanged when at the best whereas 4 patients claimed improvement and 2 deterioration as compared to
179

Nilsson et a].

-30.00

Mentation, Behavior and Mood, ADL.

Complications of therapy

Casec

Case D

OCaseE

CaseF

CaseG

CaseH

i3 Case I

Fig. 3. Percentual changes compared to oral therapy (upward bars indicating improvement) in UPDRS subscales, Mentation, Behavior, Mood, ADL (left) and complications (right) after 6 months of intraduodenal infusion of the levodopakarbidopa dispersion.

oral therapy when at the worst, according to the Modified Schwab and England ADL scale of UPDRS (data missing for 1 patient).
Pharmacokinetics

Daily levodopa doses on oral administration and on nasoduodenal and duodenal infusion in the 9 patients are shown in Fig. 4. The mean daily plasma levodopa concentrations in patients A and B were significantly lower and when the tendency to increased plasma concentration of levodopa with time of the day was taken into account (see Statistics), fluctuations were significantly reduced (P<O.OOl) on infusion therapy as compared to oral treatment (Table 4). Two complete test days as well as a 3 h period were excluded from analysis of variances in plasma levodopa concentrations as infusion rates were changed by mistake or due to distressing dyskinesia.
General experience and complications

All patients experienced a general improvement after the introduction of continuous treatment and
180

did not want to return to oral therapy, in spite of the expected inconvenience in carrying a pump. There have been minor technical problems with the infusion system. The gastrotomy was infected transiently postoperatively in 2 cases and minor discharge was common but at a well tolerable level. Case A was given a new catheter after 2 years due to increased resistance, and Case D after 2 months due to a cut by mistake. Case B had his pump repaired once. In Case C it was not possible to find a basal infusion rate during the week with nasoduodenal infusion. He also experienced numbness in his feet just before the last test occasion, EMG showed polyneuropathy but the patient chose to continue the infusion and the symptoms gradually disappeared. These test occasions were therefore excluded in the final statistical calculations. Three more test occasions were excluded from the statistical calculations of MT; in Case E the test 1month post operatively due to a period of pump stop and in case I, who fainted when standing waiting for the PLM-test 3 months after operation, the PLM-test was not carried out on this test occasion, neither on the next.

Duodenal infusion of levodopa

1 3000

1500

i
I
0

I
I Nasoduod infusion
I

Oral therapy

6 mnlk infusion

31 months inlusion

0 ~ d i e nA t

H Patient B A Patient c

+Patent

8 Patienl E

Patient F

A Patient G

epatienc H

*patient

Fig. 4. Daily levodopa doses at the test occasions in the 9 patients including tablets taken during night.

Discussion

Alterations in pre- as well as in post-synaptic mechanism seem to be partially responsible for motor instability in advanced parkinsonian patients on levodopa therapy (13), and recent PET-studies demonstrate that levodopa from plasma is rapidly transformed into synaptically active dopamine in these patients (20-22). Thus, a shift in plasma levodopa is instantly accompanied by a change in effect. In accordance with this it is well established that motor fluctuations in advanced parkinsonian patients on levodopa therapy are reduced upon continuous infusion of the drug intravenously ( 5 ,

13), or into the proximal small intestine (10-12, 14), where the main uptake of amino acids takes place. Levodopa plasma concentration changes related to divided oral dosing and to erratic gastric emptying are eliminated by continuous duodenal infusion, while a theoretical influence of competitive amino acids on levodopa transport across intestinal membranes and the blood-brain barrier is unaffected. It also seems plausible that positive effects reported from oral liquid levodopdcarbidopa therapy are obtained by the possibility to adjust dosing more accurately and by a more reliable bypass of the stomach as compared to tablet treatment (23, 24).
181

Nilsson et al.
Table 4. Mean plasma levodopa concentration (pg/l)kvariance (var) (pgz/Iz)when on oral therapy, on nasoduodenal (ND) and direct duodenal (DD) infusion for 31 months Test day 1 Fg/l Case A Oral Var Test day 2

pg/l

Var

ND DD
Case B Oral

1.32 1.54 0.56 0.72 # 0.43

0.12 O.lO^^* 0.1 2^^^

45 45 33(#) 45 45

0.99 1.40 # 1.15 1.22 0.88

0.15 0.13AA # 0.28

45 43

0.1 1
# 0.01^^^

ND DD

0.06^
0.19**^

45 44 44

The *** indicates significant reduction (P<O.OOl) of the mean levodopa concentrations on infusion therapy (after 31 months) compared t o oral therapy. The * indicates significant reduction (P< 0.001) of the levodopa variances when on infusion compared to oral therapy using a R e s t on the variances about regression. #=data discarded (see text).

As mentioned before, some investigators have reported on therapy with enteral infusion of levodopa but reports on long-term experience are restricted to only 7 patients in total (10-12). As a nasoduodenal route of administration is uncomfortable and socially disabling, a transabdominal port as used in the present investigation is recommended. Another inconvenience is due to the low solubility of levodopa (<5 mg per ml in water) which means that, earlier, patients on this treatment have been carrying large volumes of infusate. In the present investigation the aforementioned problems have been approached by the use of a transabdominal gastro-duodenal system, normally used in clinical routine for enteral feeding, in combination with a new galenic formulation of levodopa/carbidopa. Effects of a new treatment in patients with very advanced Parkinsons disease are difficult to establish. Motor functions vary, seemingly unpredictable during the day and from day to day. Also the effects of treatment are not necessarily uniform with respect to different subsets of symptoms. In trials with continuous long-term nasoduodenal infusion it is technically difficult to use a double-blind procedure. However, one placebo-controlled shortterm evaluation of intraduodenal infusion has been published (14) showing increased functional on hours and a decreased number of c c ~ f f episodes )7 and also a significantly decreased variability in levodopa levels. As the goal for the infusion therapy is to obtain the maximum time with normal motor performance, we chose to analyse time spent in off, normal and hyperkinetic states and not only time in the 011~~/0ff states. Using our evaluation procedure all 9 patients spent increased time in the normal state (-1, 0 and +1) while, if
182

only the on/off states are considered, only 6 of the 9 patients spent longer time of the day in the on state (-1 to +3) when on infusion compared to oral treatment. Our results demonstrate not only increased mobility but also partial normalization. In the present investigation we used, in addition to video-based scoring, an opto-electronic system measuring a complex sequence of movements, the PLM test, for objective non-biased evaluation during 2 whole days upon each test occasion. This enabled us to perform a statistical evaluation of a number of parameters related to motor performance. This equipment provides a practical way for objective assessment, as a complement to the conventional more subjective methodology in testing parkinsonian patients. Obviously, the results from the PLM test and the video recordings are in good agreement, and demonstrate that patients with very advanced Parkinsons disease benefit from long-term continuous duodenal levodopa infusion. Thus, on infusion the patients perform faster in the PLM test and the variations are reduced. Repeated objective quantitation of motor function as used in the present investigation could be an interesting alternative to conventional scoring procedures in future drug trials. However, it is important to emphasize that the measured parameter is time and not the quality of the movement. A more detailed study on the relationship between global evaluation from video and optokinetic recordings is in progress. With further development the positive effects of continuous duodenal infusion can probably be further increased. The tendency toward accumulation of levodopa during the day could be minimized through a higher loading dose in the morning, in order to quickly establish steady-state conditions, and mechanical complications can be handled by proper education of patients relatives and nurses. Based on the present results another 13 patients are now on continuous infusion with this new levodopa/carbidopa dispersion in our clinic, corresponding to about 3 patients operated per lo6 inhabitants in a year.
Acknowledgements
This study was supported by the Swedish Medical Research Council, grant no 4373, the NHR foundation and the Swedish Parkinsons disease Association. Thanks are due to Dr C.-E. Westerberg for statistical advices and to the Hospital Pharmacy, University Hospital, Uppsala, for fruitful collaboration.

References
1. RINNE UK. Problems associated with long-term levodopa treatment of Parkinsons disease. Acta Neurol Scand 1983: 95 (Suppl): 19-26.

Duodenal infusion of levodopa

2. MARKHAM DIAMOND Long term follow-up of early CH, SG. dopa treatment in Parkinsons disease. Ann Neurol 1986: 19: 365-72. 3. DUPONT ANDERSEN BOASJ et al. Sustained-release E, A, Madopar HBS compared with standard Madopar in the long-term treatment of de novo parkinsonian patients. Acta Neurol Scand 1996 93: 14-20. G, S, J, 4. BLOCK LISSC, REINES IRR NIBBELINK D and the CR First Study Group. Comparison of immediate-release and controlled release carbidopa/levodopa in Parkinsons disease. Eur Neurol 1997: 37: 23-7. 5. BREDBERG TEDROE J, AQUILONIUS PAALZOW E, S-M, L. Pharmacokinetics and effects of levodopa in advanced Parkinsons disease. Eur J Clin Pharmacoll990: 39: 385-9. 6. HARDERS, BAASH, RIETBROCK Concentration-effect S. relationship of levodopa in patients with Parkinsons disease. Clin Pharmacokinet 1995: 29: 243-56. 7. KURLAN ROTHFIELD WOODWARD et al. Erratic R, KP, WR gastric emptying of levodopa may cause random fluctuations of parkinsonian mobility. Neurology 1988: 38: 419-21. 8. LENNERNASNILSSON AQUILONIUSS-M, AHRENSTEDT H, D, 8, KNUTSON PAALZOW The effect of L-leucine on the L, L. absorption of levodopa, studied by regional jejunal perfusion in man. Br J Clin Pharmacol 1993: 35: 243-50. 9. NUTI JG, WOODWARD HAMMERSTAD WR, JP, CARTER JH, ANDERSON The on-off phenomenon in Parkinsons JL. disease. Relation to levodopa absorption and transport. New Eng J Med 1984: 310: 483-8. 10. SAGE TRoosm S, SONSALLA HEIKKILA JI, PK, R, Duvoisr~ RC. Long-term duodenal infusion of levodopa for motor fluctuations in parkinsonism. Ann Neurol 1988: 2 4 87-9. 11. SAGE I~OOSKIN JI, S, SONSALLA HEIKKILA ExperiPK, RE. ence with continuous enteral levodopa infusions in the treatment of 9 patients with advanced Parkinsons disease. Neurology 1989: 39 (Suppl. 2): 60-3. 12. CEDARBAUM SILVESTRI KUTTH. Sustained enteral JM, M, administration of levodopa increases and interrupted infusion decreases levodopa dose requirements. Neurology 1990: 40: 995-7. 13. MOURADIAN MM, HEUSER IJE, BARONTI CHASETN. F, Modification of central dopaminergic mechanisms by continuous levodopa therapy for advanced Parkinsons disease. Ann Neurol 1990: 27: 18-23. 14. KURTH MC, XTRUD TANNER et al. Double-blind, JW, CM placebo-controlled, crossover study of duodenal infusion of levodopakarbidopa in Parkinsons disease patients with on-off fluctuations. Neurology 1993: 43: 1698-703. 15. BREDBERG NILSON D, JOHANSSON al. Intraduodenal E, K et infusion of a water-based levodopa dispersion for optimisation of the therapeutic effect in severe Parkinsons disease. Eur J Clin Pharmacol 1993: 45: 117-22. 16. STEGG, INGVARSSON JOHNELSVALW THORSELKJS PE, B, M, M. Objective measurement of motor disability in Parkinsons disease. Acta Neurol Scand 1989: 126: 67-75. 17. JOHNELS INGVARSSON HOLMBERGMATOUSEK B, PE, B, M, STEGG. Single-dose L-dopa response in early Parkinsons disease: Measurements with optoelectronic recording technique. Mov Disord 1993: 8: 56-62. 18. FAHN ELTON L and members of The UPDRS DeveS, R lopment Committee. Unified Parkinsons disease rating S, CD, DB, GOLDSTEIN M, scale. In: FAHN MARSDEN CALNE eds. Recent developments in Parkinsons disease, Vol. 2. Florham Park, New Jersey: Macmillan Healthcare Information 1987: 153-63,293-304. 19. CONOVER Practical nonparametric statistics (2nd edn). WJ. New York: John Wiley & Sons, 1980. 20. TEDROFF AQUILONIUS HARTVIG P, LANGSTROM J, S-M, B. Functional positron emission tomographic studies of striatal dopaminergic activity. Changes induced by drugs and nigrostriatal degeneration. Adv Neurol 1996: 69: 443-8. 21. RDROFF J, PEDERSEN AQUILONIUS M, S-M, HARTVIG P, JACOBSSON LANGSTROMLevodopa-induced changes in G, B. synaptic dopamine in patients with Parkinsons disease measured by [C]raclopride displacement and PET. Neurology 1996: 46: 1430-6. J, R, P 22. TEDROFF TORSTENSSONHARTVIG et al. L-dopa n modulates striatal dopaminergic function i v i m A study in non human primates using positron emission tomography. Synapse 1997: 25: 56-61. 23. KURT MC, TETRUD IRWIN LYNESS JW, I, WH, LANGSTON JW. Oral levodopakarbidopa solution versus tablets in Parkinsons disease patients with severe fluctuations: a pilot study. Neurology 1993: 43: 1036-9. 24. PAPPERTEJ, GOETZ CG, NIEDERMAN BSN, LING ZD, STEBBINS CARVEY Liquid levodopa/carbidopa proGT, PM. duces significant improvement in motor function without dyskinesia exacerbation. Neurology 1996: 47: 1493-5.

183