Eur J Clin Pharmacol (1992) 43:357-363 EuropeanJournalof ( ~ ] ~ ( ~ ( ~ r l

@Springer-Verlag t992

Bromocriptine lessens the incidence of mortality in L-Dopa-treated parkinsonian patients: prado-study discontinued* H. P r z u n t e k ~, D . W e l z e l ; , E. B i i i m n e r 3, IN. D a n i e l e z y k 4, H. L e t z e l 3, H.-J. Kaiser 2, P. H . Kraus 1, P. R i e d e r e r 5, D . S c h w a r z m a n n 2, H . W o l f 2, and K. [Jberla 6

Neurologische Klinik der Ruhr-Universit~itBochum,FRG 2Sandoz AG Nt~rnberg,FRG 3Staticon Medizinische ForschungsgesellschaftmbH Planegg, FRG 4Geriatrisches Krankenhaus Lainz, Wien,Austria s Klinische Neurochemie, PsychiatrischeKlinik und Poliklinik, Wt~rzburg,FRG 6Institut for Med. Informationsverarbeitung, Biometrie und Epidemiologie der UniversitfitMt~nchen,FRG Received: March 17,1992/Accepted:April 27, 1992

Summary. L-Dopa supplemented by a peripheral decar-

boxylase inhibitor is considered the most potent therapeutic regimen prolonging active life in Parkinsonian patients. The long-term benefit of therapy is limited by adverse effects, such as dyskinesia and on-offphenomena, which can be mitigated by the concomitant administration of dopamine agonists, such as bromocriptine. In order to quantify the beneficial impact of early combination therapy, a controlled clinical trial (PRADO: PRAvidell+ DOpa) in patients with early Parkinson's disease was carried out, whereby L-Dopa monotherapy (in a fixed combination with benserazide (DoBe) was being compared with the same combination plus bromocriptine (DoBeBro). Patients were recruited and treated by 101 practising neurologists in the Federal Republic of Germany and in Hungary. 'Twenty seven clinical university centers cross-checked the patients at regular intervals. The trial started with 3 months of DoBe monotherapy (median dose of 375 mg L-Dopa for both randomized groups) followed by gradual substitution of DoBe by bromocriptine over 3 months in one of the groups (250 mg L-Dopa/10mg bromocriptine). The target medication was maintained from study months 6 to 54. Parkinsonian symptoms were classified according to the Webster rating scale, the Hoehn and Yahr scale and the Zung Self-Rating Depression Scale. Adverse events and life status were checked at regular intervals. Special emphasis was given to motor performance tests. 587 patients (302 in the DoBe group and 285 in the DoBeBro group) were available for intention-to-treat analysis. Both groups were homogeneous at baseline in all observed parameters. DoBe and DoBeBro proved equi-effective in terms of antiparkinsonian activity after the substitution phase (P II) had been completed. In September 1991, after a median observation period on target medication of 38.4 months in the DoBe group and 40.1 months in the DoBeBro group, 18 versus 8 deaths had been registered.

The Logrank test as well as analysis using the Cox model, both adjusted for age and sex, showed P-values of 0.018 and 0.021, respectively. The mortality risk associated with L-Dopa therapy was reduced by more than 50 % by its combination with bromocriptine. The study was terminated due to this difference in mortality. The causes of death were classified by the treating physicians and constfltants. At the time of study termination 152 patients in the DoBe group and 121 in the DoBeBro group had already discontinued study medication. Of those further 26 patients had died by the date of the final evaluation, t5 on DoBe and 11 on DoBeBro. The results imply that combination therapy with bromocriptine should be preferred over L-Dopa monotherapy from the very beginning. K e y words" Parkinson's disease, Bromocriptine; LDopa/benserazide, early combination therapy, long-term therapy, mortalib; cardioprotection

At present L-Dopa in combination with a peripheral decarboxylase inhibitor still represents the mainstay of treatment for Parkinson's disease. Early therapy improves the quality of life and it leads to a considerable increase in the life expectancy of parkinsonian patients [1, 2]. However, after some years of successful treatment, therapeutic problems arise, such as declining therapeutic response, fluctuation in disability (wearing-off deterioration, on-off-phenomena) and involuntary movements, e.g. dyskinesia and dystonia) [3-5]. L-Dopa itself may contribute to the progression of the disease [6], e. g. especially high doses may lead to neuronal damage due to enhanced generation of free oxygen radicals, whose elimination is impaired by a deficiency of free radical scavenger systems in the basal ganglia of parkinsonian patients

[7-10]. * Dedicated to E.D. Schneider on the occasionof this 60th birthday. l Pravidel = brand of Bromocriptine

Thus, sparing of L-Dopa by partial substitution with a dopamine agonist such as bromocriptine appears a logical therapeutical algorithm.

358 Table

1. Study design

Inclusion/exclusion criteria

~ " D e novo" parkinsonian patients

+ Randomisation Therapy Phase I Dosage build up 0-3 months

Build-up dose of L-Dopa/benserazide (6 weeks) and maintain treatment of both groups for 3 months

Therapy Phase II Substitution 3-6 months

Monotherapy L-Dopa Combination therapy: plus benserazide" Reduction of L(DoBe) Dopa/benserazide gradually replaced by bromocriptine b

Therapy Phase III Target medication 6-54 months

L-Dopa plus benserazide a (DoBe)

L-Dopa/benserazide plus bromocriptineb (DoBeBro)

The participating neurologists (n = 101) went through comprehensive inter-rater training and had to adhere to the standard rules of Good ClinicalPractice. Twenty seven clinical universitycentres or special institutions cross-checked the patients at regular intervals every 6 months. An independent scientific advisory board monitored the study The trial started with 3 months treatment with DoBe (therapy Phase I) for both groups, followed by a 3 months gradual substitution of DoBe by bromocriptine in half of the patients (therapy Phase II; Table 1). The equivalent doses of both drugs, required for the gradual substitution of DoBe, were derived from previous studies, which had demonstrated that 5 mg bromocriptine roughly equalled 50.0 mg L-Dopa in (fixed) combination with 12.5 mg benserazide (Madopar®) [11, 13-15]. The latter was used as the study medication (capsules of 62.5 mg or 125.0 mg, respectively). Bromocriptine (Pravidel®) was available as tablets (2.5 mg) and capsules (5.0 or 10.0 mg). After the substitutionphase, steady state therapy in both study groups was maintained up to the 54th study-month, with

MADOPARR; b PRAVIDEL R Table 2. Base-line clinical characteristics in the two study groups

There is circumstantial evidence that combined treatment with bromocriptine and L-Dopa/benserazide started in an early stage of the disease may offer some long-term benefit relating to superior efficacy and tolerability [11, 12]. Conclusive results, from long-term randomized studies are not available. In order to quantify the beneficial impact of early combination therapy a controlled clinical trial has been undertaken, in which LDopa monotherapy (DoBe) was compared with its combination with bromocriptine (DoBeBro). The assumption to be tested was that after 41/2year period motor side effects would be less frequent and less severe after DoBeBro than DoBe. The long-term therapeutic benefit of the combined treatment was also expected to be superior and functional impairment of fine motor skills to be less after DoBeBro therapy. These hypotheses will be analysed in a forthcoming paper. In September 1991 the P R A D O trial was interrupted since the mortality rate in patients treated with the combination was significantly lower than in those treated with L-Dopa alone.

Patients and methods

Study design The aim of this multicentre randomized prospective study in patients with early Parkinson's disease was to compare DoBe and the DoBeBro in terms of overall therapeutic benefit, fine motor skills, and the incidence and extent of all motor side effects. The cause and justification of each case of discontinuation or dropout was meticulously cross-checked. Death, however, was not considered a main endpoint in the protocol. De novo parkinsonian patients, i.e. patients previoulsy treated for up to a maximum of 6 months with L-Dopa were included. Patients with concomitant medication (MAO-B inhibitors, neuroleptics reserpine, a-methyt-dopa, flunarizine derivatives or metoctopramide) were excluded; other exclusion criteria were severe internal or psychiatric diseases. The planned sample size was 500-600 parkinsonian patients, to be randomized into 2 treatment groups. Sample size estimation (or= 0.05, [3= 0.20) was based on the expected difference in motor adverse effects (30%). The planned number took into account loss due to the naturally increased morbidity and mortality of such an elderly population.

Characteristic

L-Dopa/Bense- L-Dopa/Benserarazide (DoBe) zideplus Bromon = 302 criptine (DoBegro) n = 285 % of group

Sex male

51.0

56.5

Hoehn and Yahr staging Stage 1 Stage 2 Stage 3 Stage 4

21.5 38.4 34.1 6.0

19.3 42.5 33.7 4.6

Depression (Zung scale) No depression Mild depression Moderate depression Severe depression

56.0 19.2 16.9 6.6

59.6 22.8 9.8 5.6

Pretreated with L-Dopa + Benserazide Patients with concomitant diseases Cardiovascular disease - Cerebrovascular - Ischaemic heart disease Hypertension Diabetes mellitus Bronchopulmonary disease Gastrointestinal disease

31.5

30.5

50.3 20.5 4.0 9.0 12.0 5.0 2.0 3.3

49.8 28.1 3.0 10.0 16.0 7.0 1.0 1.8

Concomitant medication unreIatedto Parkhason's disease

45.4

46.0

Median Webster rating scale Duration of Parkinson at study onset (months) Duration of L-Dopa pretreatment (months) ~ Age (y) male Age (y) female Height (cm) Weight (kg) BP syst. (mmHg) BP diast. (mm Hg) Heart rate (min i)

9.1 20 2.53 62.5 65 168 71.0 140 85 76

10.0 21 2.70 63.0 67 168 71.0 140 85 75

With a concomitant Dopa decarboxylase inhibitor

359 Table 3, Disease status on L-Dopa/Benserazide dosage build-up (Therapy Phase I) and after partial substitution by Bromocriptine (Therapy Phase II)" Month L-Dopa/ L-Dopa/BenBenserazide serazide plus (DoBe) Bromocriptine (DoBeBro) Webster rating scale Study start 0 End of L-Dopa build-up 3 End of partial substitution 6

9.1 7.0 6.0

10.0 6.1 6.0

Zung rating scale Study start 0 End of L-Dopa buiIdmp 3 End of partial substitution 6

48.7 43.4 41.6

46.3 42.5 41.3

a Median values some allowance only for individual dose adaption (therapy Phase III). Further details of the study plan were reported in a preceeding paper on the PRADO trial.16

Observed parameters During the first year the following parameters were assessed every 3 months, and later every 6 months: Parkinsonian symptoms according to the Webster Rating [1'7] and the Hoehn and Yahr [18] scales. The Zung Self-Rating Depression Scale [19] was used to evaluate depressive symptoms. Adverse events, including life status, were regularly checked with special attention to the development of motor side-effects, such as dyskinesia, dystonia and on-off phenomena. In addition, patients used a personal diary to record their mobility.

Statistical methods A descriptive data analysis at baseline comparing both groups was performed. Life status data were analysed by the Cox model and Logrank test both adjusted for age and sex. Life-table analysis were performed by means of PC-SAS and PC-BMDP statistical packages [20, 21].

Results

From September 1986 to June 1988 674 patients were enrolled in the trial, of whom 3 died during the Phase I dosage build-up with DoBe and 75 were unsuitable for the second trial phase (9 patients were intended as a pretest and were not randomised). The reasons were: in adequate compliance (n = 42), protocolviolation (n = 5), side effects (n = 22), e. g. nausea,palpitation,vertigo, vomiting or sleep disturbances, and concomitant complicating disease (n = 6). The remaining 587 patients entered Phase II and were randomised to receive either DoBe or DoBeBro. They proved to be homogeneous in all relevant aspects of their baseline characteristics (Table 2). They were also comparable in motor performance tests, etc. What is especially important is that, there was acceptable homogeneity of both groups in terms of their antiparkinsonian response to L-Dopa monotherapy during the build-up phase and also at the end of the partial substitution with bromocriptine (Phase II; Table 3). At the end of Phase I the median dose of L-Dopa/benserazide in both groups was 375 mg, which was kept constant in the corresponding (DoBe) group throughout Phase II. In the DoBeBro group, i.e. after partial substitution of L-Dopa by bromocriptine, the median dosage of L-Dopa/benserazide amounted to 250 mg complemented by 10 mg bromocriptine. Following an interim analysis in Spring 1991, the advisory board focused special attention on life status monitoring, with the result that the trial was terminated on 19th September 1991 due to an increased number of deaths in those on L-Dopa monotherapy. The median observation time on target medication, i.e. DoBe and DoBeBro, was 38.4 months in the DoBe and 40.1 in the DoBeBro group. The life status of patients up to endpoint or on 19 September 1991, is shown in Table 4. There were 18 deaths in those on DoBe treatment and 8 deaths in those on DoBeBro. One of the patients dying during treatment with the latter regimen had only received a very small dose of bromocriptine (1.25 rag). The Logrank test and the Cox model both showed a crude p-value (double-sided) of 0.07; adjusted for age and sex P was approximately 0.02

Table 4. Mortality, discontinuations and dropouts L-Dopa/Benserazide (DoBe)

L-Dopa/Benserazide plus Bromocriptine (DoBeBro)

Total

Patients entering Phase II

302

285

587

On treatment at end of study September 19, 1991

163

167

330

18

8

26

152

121

273

34 15 24 12 40

7 40 18 10 38

41 55 42 22 87

.r

Deaths while on treatment Dropouts or discontinuations of study medication Because of Therapy not effective Side effects Concomitant diseases Protocol violations (Major) Incompliance Deaths after discontinuation or dropout

15

11

Life status after disconthluation/dropout unknown as of September 19,1991

72 (47,3 %)

58 (47,9%)

26 1.30

360 Table 5. Statistical evaluation of mortality in patients on treatment

L-Dopa/ L-Dopa/Benserazide Total Benserazide plus Bromocriptine (DoBe) (DoBeBro) Patients allocated to therapy (Phase II)

302

285

Deaths in patients on treatment

18 8 Double sided P-values Crude Adjuste@ 0.076 0.018 Log rank test 0.072 0.021 Cox model 2.10 Risk ratio RR (crude) Risk ratio RR (adjusted) 2.71 Risk reduction (crude) 52.2 % Risk reduction (adjust63.2 % ed) Risk reduction =

587 26

(1- 1/~).100%

For age and sex (Table 5). Thus, for ethical reasons, the P R A D O trial had to be terminated. The causes of death in the 26 cases, mainly due to cardiovascular complications, are listed in Table 6. A n autopsy was p e r f o r m e d in 3 cases. Besides the incidence of mortality in these 26, cases which ultimately led to termination of the P R A D O trial, other important events are given in Table 4. D r o p o u t or discontinuation had occurred in 152 and 121 cases, respectively, for the reasons listed, all of them carefully crosschecked between the university centres and the treating physicians. The side effects mainly comprised gastro-intestinal complaints, nausea, palpitation, sleep disturbances, which interferred with strict pursuance of the study medication and with patient compliance. More extensive information about the side effects will be published in the subsequent clinical paper. T h e r e was a higher incidence of mortality after discontinuation of study medication in the D o B e than in the D o B e B r o group (15 vs 11), with the limitation that the life status of this subset of patients was only known in 50 %. T h e cases of known overall crude mortality rate was 33/302 = 10.9% in the D e B o group and 19/285 = 6,7 % in the D o B e B r o group. Altogether, the risk of mortality was particularly enhanced in males, irrespective of the therapeutic group. This was true for males in general (Logrank test adjusted for age: P < 0.01) and for males in the D o B e group compared to subjects in all groups combined (males on D o B e Bro therapy, females on D o B e and on combination t h e > apy; Logrank test adjusted for age P < 0.001).

Discussion

L - D o p a with a dopa-decarboxylase inhibitor not only prolongs active life in parkinsonian patients but also reduces excess mortality [22]. Nonetheless, the m a n a g e m e n t of Parkinson's disease has b e c o m e m o r e complicated by the different types of fluctuations associated with long-term L - D o p a medication. Although its combination with bro-

mocriptine has a definite place in coping with established Dopa-related difficulties, such as end-of-dose disturbances, dyskinesias and on-off phenomena, it remains to be shown whether early use of the combination lessen these difficulties in a preventive manner. Retrospective and follow-up data can provide some guidance in this respect [12, 23, 24]. T h a t was the starting point of the P R A D O project, which incorporated essential features of controlled clinical trials e.g. defined patient selection, central randomisation, elimination of bias by cross-checks with university hospitals and control of confounding by considering known prognostic p a r a m e ters. The outcome criteria centred upon fine m o t o r skills Table 6. Patients dying on treatment Age Sex

Time of death (month of treatment) L-Dopa/Benserazide 65 F 7 76 M 9 52 M 11 69 M 15 46 M 15 78 F 17 70 M 19 76 F 20 72 M 23 79 M 27 72 F 28 73 F 29 71 M 31

Cause of death

Pulmonary embolism Myocardial infarction Climbing accident Sudden cardiac death Road accident Pulmonary embolism Carcinoma (prostate) Sudden cardiac death Sudden cardiac death Cardiovascular failure (infection) Pulmonary embolism Cardiac failure, arrhythmia Cardiovascular failure (ischaemic heart disease) 70 M 35 Cardiovascular failure (ischaemic heart disease) 85 M 39 Cardiovascular failure (ischaemic heart disease) 83 M 40 Cardiovascular failure (ischaemic heart disease) 68 M 42 Myocardial infarction 80 M 44 Cardiovascular failure (ischaemic heart disease) L-Dopa/Benserazide plus Bromocriptine 75 F 7a not ascertainable 68 F 7b Myocardial infarction (suspected) 75 M 26 Myocardial infarction 49 M 31 Cardiovascular death (ischaemic heart disease, diabetes mellitus) 85 F 38 Myocardial infarction 66 M 42 Sudden cardiac death (ischaemic heart disease) 79 F 45 Cardiovascular failure (guess to intestinal bleeding, duodenal ulcer) 83 M 47 Myocardial infarction End of therapy Phase II, bromocriptine substitution 5 mg (L-Dopa dose 187,5 mg) b End of therapy Phase II, bromocriptine substitution 1,25 mg (L-Dopa dose 250 mg)

361 motor side-effects and overall therapeutic benefit. Every effort was made to monitor patients at planned time intervals. The monitoring approach proved effective in evaluating the basic therapeutic response in both groups of patients (Webster and Zung rating scale), as they were entirely comparable after the run-in-phase with LDopa/benserazide, as well as after partial substitution by bromocriptine, The essential equivalence of both drug regimens regarding their symptomatic activity was established. It was the differential death rate that brought the PRADO-trial to premature discontinuation; the Risk Ratio of L-Dopa monotherapy compared to combination therapy was 2.7, which is a risk reduction of 63 %. From a biometric viewpoint an a posteriori hypothesis had to be tested, The unexpected finding, however, was one that could hardly be explained by chance or bias, since it was not the result of multiple testing without or-adjustment. There was no study hypothesis to test upon mortality differences between the 2 groups, which excludes a statistical test of such differences in the confirmatory sense. The patients mainly died from cardiovascular complications, although some allowance must be made for the inaccuracies of attribution of the causes of death not based on autopsy. Great emphasis was laid upon precise, retrospective analysis of each fatal event by scrutinising all the material available through contacting the treating physicians and consultants. To discuss the differences in death rates means to consider the differential cardiovascular pharmacology of the drugs involved. Interest primarily focuses on the catecholaminergic functions, which are thought to be important in precipitating cardiac arrhythmias and myocardial ischaemia. It is known that higher doses of L-Dopa, as were normal in the pre-dopa-decarboxylase-inhibitor era, can induce manifest cardiac irregularities [25]. The dose of L-Dopa in the present patients dying on the monotherapeutic regimen amounted to 437 mg + 125 daily, which is too low to produce overt cardiotoxic effects. It nevertheless remains in the critical range for potential side-effects, since cardiac arrhythmias may also be observed in patients on combined LDopa/dopa-decarboxylase inhibitor regimens [26-28]. Moreover, the cardiac activity of L-Dopa can be enhanced by concomitant disease or "sensitising" agents, e. g. cyclopropane and the halogenated hydrocarbones known to potentiate the arrhythmogenic effects of catecholamines [29]. Thus, a pathogenetic interaction of dopamine with endogenous catecholamines may have contributed to the overall cardiac risk in such an elderly population. It is noteworthy that dopamine levels in man given L-Dopa rise independently of whether or not a peripheral dopadecarboxylase inhibitor is concomitantly applied [30]. Bromocriptine exhibits a cardiovascular profile clearly different of that of L-Dopa. Due to presynaptic stimulation of peripheral dopamine receptors, it decreases neural norepinephrine spill-over, causing a profound reduction in blood levels, as has been documented in various species and men [31-33]. The inhibition of catecholamine output can be abolished by domperidone, which indicates that bromocriptine reduces norepinephrine release by periph-

eral activation of DA2-receptors [34]. Such depression of sympathetic activity per se could explain part of the cardioprotective potential of bromocriptine. In fact, experimental data show that the liability to ventricular arrhythmias is decreased, which led B. Lown et al, to suggest "a possible future clinical use of the drug, particularly in those patients in whom increased sympathetic tone is demonstrated to be related to ventricular arrhythmia" [35]. That suggestion gains broader significance through recent evidence that major ventricular arrhythmias are associated and perhaps caused by sustained and selective cardiac sympathetic activation [36]. Given the background of the arrhythmogenic potential of L-Dopa, bromocriptine may exert functional antagonism of the latter with the effect of enhanced cardiac safety of the combined drug regimen. The overall implication of P R A D O surely does not mean that L-Dopa monotherapy puts the average parkinsonian patient at major cardiac risk. As a matter of fact, it needed such a large, controlled study as the P R A D O trial to demonstrate the problems involved in L-Dopamonotherapy. Moreover, it remains open whether bromocriptine per se is more cardioprotective than L-Dopa is primarily harmful. Independent of the question of cardiac safety and its modification by bromocriptine, the questions of neuroprotection by dopaminergic agonists and its impact on lifespan should also be considered. There are positive results from various animal experiments bearing on these aspects. The results of the P R A D O trial favour the adoption of combination treatment from the beginning, all the more since the relative reduction in mortality due to such strategy is substantial. Whether bromocriptine should be employed as the first-line therapy and complemented by LDopa to ensure optimal antiparkinsonian efficacy, or whether the two agents should be administered together from the onset of dopaminergic therapy, remains open to discussion.

Appendix List of participants and institutes Dr. Alberti Eckhardt (Schwetzingen), Dr. Albiez Rudolf (Hamburg), Dr. Backhaus D. (ttildesheim), CA Prof. Dr. Balzereit Fritz (Allg. Krankenh. Hamburg-Harburg), Dr. Bammborschke (Med. Einrichtung d. Universit~t KOln), Dr. Baumann Ktaus (Dillingen), Dr. Beyer Dietrich (Heidelberg), Dr. Bittkau (NeurologischeUniv.Klinik Wtirzburg), Dr. B16ssel-Epple(Regensburg), Dr. Blotenberg (Essen), Dr. Bohr Nils (Koblenz), Dr. Bokor Margarete (Mtinchen), Dr. Brosig Arne (Grevenbroich), Dr. Crome Susanne (D%seldorf), Prof. Dr. Csanda (Sote NeurologischeKlinik Budapest), Dr. Dabanka Fr. (Wetter), Dr. Dangel Fr. (Juliusspital Wtirzburg), Dr. Deppe Andreas (Witten), Dr. Dessauer Mauricio (Hamburg), Dr. Dirks Wahraud (Hannover), Dr. Ditzler K, (Bad Krozingen), Dr. Eckert Peter (Mannheim), Dr. Erbguth Frank (Neurolog. Universitfitsklinik Erlangen), Dr. Faust Giselher (Mainz-Gonsenheim), Dr. Gemende Georg (Bezirkskrankenhaus Bernburg), Dr. Gemende Irene (Bezirkskrankenhaus Bernburg), Prof. Girke (Neurologische Klinik Berlin), PD Dr. Glag (Bezirkskrankenhaus Neubrandenburg), Dr. Gmeiner H.J. (Neurolog. Universitfitsldinik Erlangen),

362 Dr. Goldscheider H.-G. (Augsburg), Dr. Grimmer Wolf (ROsselheim), Dr. Gr6zinger (Neurolog. Universitfitsklinik Mannheim), Dr. Grogeloh Gabriele (Essen), PD Dr. Grossmann W. (Stfidt. Krankenh. Miinchen-Harlaching), Dr. Grundmann Fr. (Universit~tsnervenklinik Homburg), Dr. Hagentocher H.-U. (Heidelberg), Dr. Hartmann Fr. (Neurolog. Universit~tsklinik Mainz), Dr. Haupt Gerhard (Hofheim am Taunus), Prof. HeN (Med. Einrichtung d. Universit~it KOln), Dr. Hemmrich Fr. (Neurologische Klinik Berlin), Dr. Henke Roland (Ludwigshafen), Dr. Herrmann (Oberhausen), Dr. HetzeI (Juliusspital Wiirzburg), Dr. Hofmann Werner (Aschaffenburg), Dr. Holtvoeth (Dortmund), Dr. Huber (Med. Einrichtung d. Universitfit KOln), OA Dr. Husstedt (Universitfitsklinik MOnster), Dr. JOstingmeier Michael (Ulm), Dr. Jungmann Friedhelm (Saarbr0cken), Dr. Katzenmeier Friedhelm (Augsburg), OA Dr. Klemm (Altg. Krankenhaus Hamburg-Harburg), Dipl. Psych. Klotz (Neurolog. Universit~itsklinik Bochum), Dr. Kohlhepp Willibald (Neurologische Univ.-Klinik Wtirzburg), Dr. Korda Wolfgang (Erlangen), Dr. Kotzian Johannes (FOrth), OA Dr. Kraus (Neurolog. Universit~itsklinik Bochum), Dr. Kresse Margarete (Wiesloch), Dr. Krippner Klaus (Witten), Dr. Kristin D. (Augsburg), Dr. KuhI Karl (Bad Kreuznach), Dr. Kuhn (Well), Dr. Kukiolka Henryk (T~bingen), PD Dr. v. Kummer R. (Universit~it Heidelberg), PD Dr. Lang (Neurotog. Universit/itsklinik Erlangen), Dr. Legewie Renate (Dtisseldorf), Dr. v. Lieven Thomas (Augsburg), Prof. Dr. Lipcsey (Janos Krankenh. Budapest), Dr. Luber Gabriete (Ntirnberg), Prof. Lurati Marion (Schweinfurt), Dr. Maas Gabriele (Hamburg), Dr. Mainusch Getrud (Berlin), Dr. Mann Harald (Rottweil), Dr. Mayer Dieter (Heppenheim), Dr. MeiBner J6rg (Coburg), Prof. Dr. Mettens Hans-Georg (Neurologische Univ.-Klinik W{irzburg), Dr. Michler Michael (Hamburg), Dr. Mirghawameddin Ferydun (Saarlouis), CA Dr. Molitor (Juliusspital Wnrz&urg), Dr. MOllner (Straubing), Dr. Mttnnich J.E. (Dossenheim), Dr. Neeb Veronika (Giel3en), Dr. Nippert Martin (Schweinfurt), DI: Oleanu-Nerbe Margit (MOnchen), Dr. P/~tz K.F. (Schweinfurt), Dr. Peres (Janos Krankenhaus Budapest), Dr. Pfister (Zentralklinikum Augsburg), Dr. Philipps Georg (Mtinster), Prof. Przuntek Horst (Neurolog. Universit~itsklinik Bochum), Dr. Rase Eleonore (Koblenz), Dr. Rathay B. (Neurolog. Klinik Karlsruhe), Dr. Reinecke Ch. (Herrenberg), Dr. Rentrop Ulf (Weinheim), Dr. Rettelbach Rudolf (NeuUlm), PD Dr. Reuther Paul (Bad Neuenahr), Dr. Richter Arno (Ohringen), Dr. Richter-Peill H. (Hamburg), Dr. Ricker Helly (Ochsenfurt), Dr. Riebling Hans (Hamburg), OA Dr. Riffel (Zentralklinikum Augsburg), Dr. Rosenberger Klaus (BrOderkrankenhaus St. Josef, Koblenz), Dr. Rothfeder Ulrich/Dr. Ullrich Axel (Mtinchen), Dr. RueB Hans (Landsberg), Dr. Salewski E. (Mtinchen), Dr. Samtleben Till (Wtirzburg), Dr. Sang Siawasch (Bochum), Dr. Sasse K.A. (Hannover), Dr. Schiller J6rg (Reutlingen), Prof Schimrigk (Universitgtsnervenklinik Homburg), Dr. Schirmer G. (Miinchen), Dr. Scholz E. (Neurolog. Universit~tsklinik Ttibingen), Dr. Scholze R. (Worms), Dr. Schuchardt K./Dr. Budenberg Detmar (Darmstadt), PD Dr. Schatz (Neurotog. Poliklinik Giegen), Dr. Schumacher JOrg (Saarbrticken), Dr. Dr. Schwartz Benno (Duisburg), Dr. Schwartz Gottfried/Dr. Zimmermann (Hamburg), Dr. Seiler Jiirgen (ttamburg), Dr. Takats (Sote Neurolog. Klinik Budapest), Dr. ThOrner/Dr. Friedemann (D0sseldorf), Dr. Tripodi Manfred (Berlin), Dr. Troyke Peter (Mannheim), Dr. Wechtler Ute (Holle), Dr. Weghofer Hermann (Regen), Dr. Weidinger Elisabeth (Niirnberg), CA Prof. Dr. Weinrich Wolfgang (Krankenhaus Nordstadt Hannover), Dr. Weif31ogelRainer (NeckarsgemOnd), Dr, Wessels (Duisburg), Dr. Wilke-Burger Helga (Berlin), Dr. Witt Karin (Dtisseldorf), Dr. Wohlauf/Dr. Fuchs (Parkinsonklinik Wolfach), Dr. Zengtein Rainer (Bayreuth).

References 1. Hoehn HM (1983) Parkinsonism treated with levodopa, progression and mortality. J Neurol Transm 19 [Suppl]: 253-264 2. Guillard A, Chastany C, Fenelon G (1986) Etude ~tlong terme de 416 cas de maladie de parkinson. Facteurs de pronostic et implications th6rapeutiques. Rev Neurol (Paris) 142:207 214

3. Muenter MD, Tyce GM (1971) L-Dopa therapy 0fparkins0n disease plasma L-Dopa concentration therapeutic response and side effects. Mayo Clin Pro 46:231-239 4. Fischer PA, Schneider E, Jacobi Pet al. (1973) Langzeitstudie zur Effektivit~it der L-Dopa-Therapie bei Parkinson-Kranken. Nervenarzt 44:128-135 5. Baas H, Fischer PA (1986) Fluktuationen der Beweglichkeit beim Parkinson-Syndrom. In: Fischer PA (ed) Spfitsyndrome der Parkinson-Krankheit. Roche, Basel, pp 213-234 6. Cohen G (1984) Oxy-radical toxicity in catecholamine neurons. Neurotoxicology 5:7782 7. Dexter DT, Carter C, Agid F et al. (1986) Lipid peroxidation as cause of nigral cell death in Parkinson's disease. Lancet II, 639640 8. Sofic E, Moll G, Riederer P, Jellinger K, Gabriel E (1988) Monoaminerge L~ision bei seniler Demenz vom Alzheimer Typ (SD AT):Vorlfiufige B efunde. In:B eckmann H, Laux G (eds) Biologische Psychiatrie, Synopsis. Springer, Heidelberg, pp 151-t57 9. Youdim MBH, Ben-Shachar D, Riederer P (1989) Is Parkinson's disease a progressive siderosis of substantia nigra resulting in iron and melanin induced neurodegeneration? Acta Neurol Scand 126:47-54 10. Olanow CW (1990) Oxidation reactions in Parkinson's disease. Neurology 40 [Suppl 3] 32-37 11. Fischer PA, Przuntek H, Majer M, Welzel D (1984) Kombinationsbehandlung fr0her Stadien des Parkinson-Syndroms mit Bromocriptin und Levodopa. Ergebnisse einer multizentrischen Studie. Dtsch Med Wschr 109:1279-1283 12. Rinne UK (1985) Combined bromocriptine-levodopa therapy early in Parkinson's disease. Neurology 35:1196-1198 13. Calne DB, Plotkin C, Williams AC et al. (1978) Long-term treatment of parkinsonism with bromocriptine. Lancet I: 735-737 14. Le Witt PA, Calne DB (1981) Recent advances in the treatment of Parkinson's disease: the role of bromocriptine. J Neural Transm 51:175 15. Schneider E, Fischer PA (1982) Bromocriptin in der Behandlung der fortgeschrittenen Stadien des Parkinson-Syndroms. Dtsch Med Wschr 107:175 16. Przuntek H, Welzel D, Schwarzmann D, Letzel H, Kraus PH (1992) Primary combination therapy of early Parkinson's disease. Eur Neurol (in press) 17. Webster DD (1968) Critical analysis of the disability in Parkinson's disease. Modern Treatment 5:257-282 18. Hoehn HM, Yahr MD (1967) Parkinsonism: onset, progression and mortality. Neurology 17:427-.442 19. Zung WW (1965) A self-rating depression scale, Arch Gen Psychiat 12:63-70 20. Mantel N (1966) Evaluation of survivaI data and two new rank order statistics arising in its consideration. Cancer Chemother Rep 50:163-170 21. Cox DR (t972) Regression models and life tables. J R Statist Soc 34 [Set B]: 187-220 22. Krutze JF, Murphy FM (1990) The changing patterns of death rates in parkinsonism. Neurology 40:42-49 23. Rinne UK (1986) Dopamine agonists as primary treatment in Parkinsons's disease. In: Yahr MD, Bergmann KJ (eds) Advances in neurology: Raven Press, New York, pp 519-523 24. Rinne UK (1987) Early combination of bromocriptine and levodopa in the treatment of Parkinson's disease: a 5-year follow-up. Neurology 37:826-828 25. Cedarbaum JM, Schleifer LS (1990) Drugs for Parkinson's disease, spasticity, and acute muscle spasms. In: Goodman S, Gilman A (eds) The pharmacological basis of therapeutics, Pergamon Press, Oxford New York, pp 463-484 26. Mars H (t973) Modification of levodopa effect by systemic decarboxylase inhibition. Arch Neuro128:91-95 27. Miller EM, Wiener L (1974) Ro 4-4602 and levodopa in the treatment of parkinsonism. Neurology 24:482486 28. Papavasiliou PS, Cotzias GC, D0by SE, Steck A J, Fehling C, Bell MA (1972) Levodopa in parkinsonism: potentiation of central effects with a peripheral inhibitor. New Engl J Med 285:8-14

363 29. Katz RL, Lord CO, Eakine KE (1967) Anesthetic-dopamine cardiac arrhythmias and their prevention by beta-adrenergic blockade. J Pharmacol Exp Ther 158:40-45 30. Boomsma E Meerwaldt JD, Man in't Veld AJ, Hovestadt A, Schatekamp MADH (1989) Treatment of idiopathic parkinsonism with L-Dopa in the absence and presence of decarboxylase inhibitors: effects on plasma levels of L-Dopa, dopa decarboxylase, catecholamines and 3-O-methyl-dopa. J Neuro1236:223-230 31. Hof RR Hof A (1981) Acceleration of blood in the aorta: a parameter useful for evaluating cardiotonic and afterload reducing substances. J Pharmacol Methods 6:87-95 32. Sowers JR (1981) Dopaminergic control of circadian norepinephrine levels in patients with essential hypertension. J Clin Endocrinol Metab 53:1133-1137 33. Nilsson A, Hoekfelt B (1977) Effect of bromocriptine on blood pressure, plasma and urinary catecholamines and plasma renin activity (pra) in patients with agromegaly. Acto Endocr 85 [Suppl 212] 95 34. Van Loon GR, Sole MJ,Bain J, Ruse JL (1979) Effectsofbromocriptine on plasma catecholamines in normal men. Neuroendocrinology 28:425-434

35. Starke K, Majewski J, Ensinger H et al. (1986) In vivo operation of prejunctional adrenoccptors in the peripheral sympathetic nervous system. In: Grobecker H, Philippu A, Starke K (eds) New aspects of the role of adrenoceptors in the cardiovascular system. Springer, Berlin Heidelberg New York, p 43 36. Falk RH, Desilva RD, Lown B (1981) Reduction in vulnerability to ventricular fibrillation by bromocriptine, a dopamine agonist. Cardiovasc Res 15:175-180 37. Meredith I'I, Broughton A, Jennings GL, Esler MD (1991) Evidence of a selective increase in cardiac sympathetic activity in patients with sustained ventricular arrhythmias. New Engl J Med 325:618-624

Prof. D. Welzel Sandoz A G Deutschherrnstr. t5 W-8500 Nttrnberg 80 FRG