Saturday 21 August 1976

selves do

the blood-brain barrier, central should not be impaired.5 6 Two decarboxylase inhibitors have been extensively investigated and preparations of each in combination with levodopa have been marketed: carbidopa (1-K methyldopa hydrazine) in a ratio of 1/10 with levodopa as ’Sinemet’ and ’Madopar’ as a 4/1 combination of levodopa and benserazide (DL-2-amino-hydroxy-2- [2, 3, 4, 4-trihydroxybenzyl] propionohydrazide). Both these preparations reduce nausia and vomiting associated with levodopa and allow a reduction of levodopa dose to 60-80% of that when the aminoacid is used alone.’-9 However, as carbidopa has been studied mainly in the United States and Great Britain and benserazide in Europe and Canada, there have been only a few uncontrolled comparisons of the comparative efficacy of the two preparations and their range of side-effects.10 We report a blind randomised crossover trial of the commercially available formulations of levodopa with carbidopa or benserazide in patients with idiopathic parkinsonism.

COMPARISON OF LEVODOPA WITH CARBIDOPA OR BENSERAZIDE IN PARKINSONISM A. COXON J. L. REID

J. K. GREENACRE A. PETRIE

not

cross

decarboxylation

Department of Clinical Pharmacology and Medicine, Royal Postgraduate Medical School, Hammersmith Hospital, London W12

The therapeutic efficacy and side-effects of two preparations of levodopa with extracerebral decarboxylase inhibitors have been compared in 19 patients with idiopathic parkinsonism in a blind randomised crossover trial. The mean daily dose of levodopa was 658±64 mg/day (mean±S.E.M.) when given together with carbidopa 66 mg/day and 605±59 mg/day when levodopa was combined with benserazide 151 mg/day. There was no significant difference between the treatment regimens either in beneficial effects on parkinsonian symptoms and signs or in the adverse effects of levodopa assessed by a clinical observer unaware of the treatment given. Of the 19 patients studied, 9 preferred the carbidopa preparation, 8 preferred the benserazide preparation, and 2 had no preference. It is concluded that there is no significant difference in therapeutic effects or adverse reactions between the two commercially available decarboxylase-

Summary

Patients and Methods Patients

patients with idiopathic parkinsonism (14 males and 7 females) whose ages ranged from 51to 78 years (63-9+1-88 years, mean±s.E.M.) were investigated. All had had parkinsonian symptoms for 2 years or more which ranged in severity from only mild symptoms affecting one side (5 patients) to severely disabled and chairbound (3 patients). 14 patients were receiving anticholinergic drugs, 4 were on tricyclic antidepressants, and 1 was on amantadine at the start of the study. These drugs were maintained at the same dose level throughout the study. Before the trial 5 patients were receiving levodopa and 16 levodopa together with carbidopa. All patients had derived therapeutic benefit from these levodopa preparations. 21

inhibitor-containing preparations. Central-nervous-system actions and side-effects depend on the daily dose of levodopa, regardless of the different ratios of decarboxylase inhibitors to levodopa. Introduction THE clinical features of parkinsonism are associated uth depletion of striatal dopamine and the degeneration of nigro-striatal dopaminergic neurons which participate n extrapyramidal motor control. 1 levodopa, the aminoacid precursor of dopamine, when given in large doses iratly, improves parkinsonian symptoms and signs, parucularly hypokinesia.13 The therapeutic benefit of levoMpa is dependent on its entry into the central nervous :_,stem and decarboxylation to form dopamine. However, a considerable proportion of oral levodopa is :ecarboxylated in the gastrointestinal tract, liver, kidand other peripheral sites leaving only small counts available to enter the central nervous system.4 Peripheral decarboxylase inhibitors will prevent extradecarboxylation of levodopa, but if they them-

Trial

Design

were randomly allocated to one of two treatment regimens consisting of maximally tolerated doses of either tablets containing 100 mg or 250 mg of levodopa with 10 mg or 25mg of carbidopa, respectively, (regimen A) or unmarked capsules containing 100 mg and 200 mg of levodopa and 25 mg and 50 mg, respectively, of benserazide (iegimen B). After randomisation, drug intake was adjusted until a stable, optimum level was obtained. Patients then entered a 6 wk period in which therapeutic effects and adverse effects including bloodpressure and heart-rate were measured by an observer unaware of the treatment given or the trial design. After 6 wk (3 fortnightly outpatient assessments) patients were changed to the other regimen and after a similar build-up period titrated

Patients

7982

©

382 intake and assessed over a further 6 wk period. 11 patients received regimen A first and the remaining 10 started to

optimal

on

regimen B.

Clinical Assessments

TABLE

I-ESTIMATED BENEFIT OF REGIMEN B OVER REGIMEN A OS SCORES FOR THERAPEUTIC EFFECTS

°

Clinical features of parkinsonism were assessed by a "blind" observer using a fixed-interval rating scale as previously described7 in which 0 represented normality and 4 was the maximurn deficit for 16 symptoms and signs. In addition, finger dexterity and flexion and extension at the wrist were assessed in timed tests. Adverse effects were determined using a standard questionnaire and a similar rating scale for severity. Patients had their blood-pressure measured after resting supine for 3 min, and after standing for 1 min, with an automatic ultrasonic sphygmomanometer (’Arteriosonde’ 1217, Roche Electronics Ltd), and their heart-rate was counted at the radial pulse for 1 min. In addition, patients were invited to rate their own symptoms using bipolar visual analogue scales. A 100 mm line was marked at one end for normality and the other for maximum disability for writing, tremor, speech, and walking. At the end of the study patients were asked to express a preference for tablets (regimen A) or capsules (regimen B) and to mark a similar analogue scale with tablets and capsules marked at the

Positive values indicate higher scores on regimen A and therefore greater dhand thus a greater chnical improvement on regimen B. Negative value: indicate a higher score and thus more disability on regimen B and indicate nr provement on regimen A. Analysis by a modified paired t test revealed no significant difference in am effect measured.

ability

TABLE II-ESTIMATED BENEFIT OF REGIMEN B OVER REGIMEN A ON SCORES FOR INDIVIDUAL ADVERSE EFFECTS AND FOR TOTAL ADVERSE COMPLAINTS ON EACH REGIMEN.

cpposite poles.I Statistical Analysis The mean score of each symptom and sign or adverse effect and time for the timed test for the three assessments on each regimen was determined. The mean value of each assessment for individual patients on regimen B was subtracted from the mean score on regimen A. The estimated treatment effect, for each item was expressed as the "benefit" (positive or negative) of regimen B over regimen A.7 The test of significance of difference between the two regimens was thus reduced to testing whether, for each item, this benefit was significantly different from zero and the appropriate test was a modification of the paired t test. The analysis was designed to take into account any possible effect of the order of drug administration.

B, the lowest dose of the lower

Results

patients who entered the trial, 2 were uncomplete the protocol. In both cases, on regimen

Of the 21

able

to

FUNCTIONAL PHYSICAL SIGNS

Positive values indicate a greater frequency of the effect on regimen A and negative values indicate the effect was more prominent on regimen B. The differenct were not significant when analysed by modified paired t test.

DISABILITIES

strength capsule (one

morning and one night) caused severe orofaciall-and limb dyskinesia for 3-4 h after dosing. These side-effects were so severe that the trial was terminated. Both patients were maintained satisfactorily on regimen A given as half a lower strength tablet twice or three times a day, indicating the greater flexibility of the tablet formulation at the lower end of the dose range. The analysis of the results is based on the observations in the 19 patients who completed three assessments on each regimen. Drug Doses The average

daily dose of levodopa (meanis.E.M,) on 605+59 mg/day (range 200-1000 mg regimen and was similar on regimen A at 658+64 mg/day (rangz 200-1375 mg/day). The mean daily intake of benserazide was 151 mg and of carbidopa 66 mg. B was

Therapeutic Effects Clinical evaluation of a range of parkinsonian svmp and signs and functional disabilities did not rel() any significant treatment benefit of regimen B 0B( regimen A or vice versa for any of the items taken sepa" ately (table i) or when the symptoms and signs or fun-’ tional disabilities were analysed together (fig. 1). The 3’sessments in the present study did not reveal ar significant difference between the two regimens in re> tion to therapeutic efficacy in parkinsonism. toms

Fig. I-Average total score of clinical assessments for individual patients on regimen A or regimen B for the 9 physical signs of parkinsonism (left) and 9 functional disabilities (right).

383

I

MLE III-MEAN SYSTOLIC AND DIASTOLIC BLOOD-PRESSURE AND AFTER 3 MIN REST SUPINE AND STANDING FOR

- HRI-RAn

(iS.E.M.)

1

MIN ON REGIMEN A AND REGIMEN B

or liver-function tests in blood examined before the trial and after each treatment regimen.

lytes,

Self Assessments Mean disability assessed by the patients on self-rating visual analogue scales and expressed as mm from normality was compared for writing, tremor, speech, and walking (fig. 2). The average disability assessed in this way was not great, although the range was considerable. There

marked difference in the mean assessthe two regimens. Qualitatively, self assesscorrelated well with the physician’s blind clinical was no

ments on

ment assessments.

When

patients were asked their preference at the end study, 9 preferred regimen A, 8 regimen B, and 2 expressed no preference. These choices were further assessed on the analogue scale of preference (fig. 3). The mean preference (is.E.M.) in mm from the tablet end (regimen A) was 49.2+7-4. This score is very close to 50

of the

which would be the midpoint between capsules and tablets and confirms that the patients, in addition to the assessing physician, observed little difference on the two

mm

Adverse Reactions

The results of analysis of the questionnaire designed to elicit symptoms of adverse effects of levodopa are shown in table n. There was no difference between the two regimens in relation to any side-effect measured, including gastrointestinal reactions, dyskinesia, and confusion. There was no influence of treatment order on side-effects and no difference emerged if all side-effects were pooled and analysed together. Blood-pressure and heart-rate lying and standing are shown in table in. Supine blood-pressure was 134/84 on regimenand 133/82 on regimen B. After standing for 1 min the pressures were 125/84 and 122/82 respectively. There was the expected increase in heart-rate on both regimens on assuming the erect posture. Differences in supine and orthostatic blood-pressure on the two regimens were thus very small and unlikely to be clinically significant. There were no differences in haemoglobin, white blood-cell count, urea, creatinine, electro-

regimens. Discussion The therapeutic efficacy in parkinsonism of levodopa in combination with the peripheral decarboxylase inhibi-

3-Preference of 19 for capsules (regimen

Fig.

scales of

Ltemor, speech, and walking of 19 parkinsonian patients.

?’?"men a ’(0); regimen B (8).

The bar represents

s.E.M.

writing,

or

The closed circle and bar represent mean±s.E.M.

carbidopa7

and benserazide9 is well established. Most workers agree that the benefit of the combined regimens derives from improvement in gastrointestinal side-effects of levodopa, rather than a marked augmentation of extrapyramidal motor effects.’ The two decarboxylase inhibitors at present available have not previously been compared under controlled conditions in the same patients. An open study in which levodopa with carbidopa was substituted for benserazide and, levodopa suggested that the range of therapeutic effects was similar.1O Differences in the pharmacological properties of the two inhibitors have been reported and these could be clinically relevant. Benserazide in very high doses, unlike carbidopa, may enter the central nervous system and impair central decarboxylation in animals6 and man." Both inhibitors are available in fixed, but different, proportions to levodopa: 1/4 in the case of benserazide and 1/10 in the case of carbidopa. These differences may underlie clinical reports of the effects of carbidopa and benserazide on important adverse reactions to levodopa. Chase and Watanabe8 reported little or no effect of carbidopa on the hypotensive effect of levodopa, while Barbeau et awl. observed that bensera-

tors

"! 2-Individual self-assessments on visual analogue

parkinsonian patients for tablets (regimen A) B) on self-assessment analogue scale.

384 zide greatly attenuated the blood-pressure-lowering action. Similar contradictory reports have appeared on the frequency of induction of new involuntary movements or dyskinesia, a common dose-limiting effect of

levodopa.12

13

In the present study, we found no difference in the therapeutic effect or side-effects when the two combination treatments were compared in a blind crossover trial in 19 patients. The similarity of response was evident

only on assessments by the clinician, but also on the self-assessment analogue scales. Self-assessment methods have been most widely used in psychiatryl4 15 but the close correspondence with clinical assessments indicates that self-rating systems may be more widely applicable. There was no difference in the incidence or severity of dyskinesia. Blood-pressure and heart-rate, supine and standing, was similar on the two regimens. It is relevant that the dose oflevodopa used in the two parts of the study was also of the same order. These results suggest that in the amounts used, both carbidopa and benserazide cause a similar, substantial degree of peripheral decarboxylase inhibition and confirm that under the conditions studied, therapeutic benefit and central dose-limiting side-effects depend on the daily intake of levodopa. The only difference between the two regimens was at the lower end of the dose range. For 2 patients, the tablet formulation of carbidopa with levodopa was easily broken in half and permitted a more flexible dosing schedule, avoiding the severe dyskinesia which followed twice daily dosing with 100 mg levodopa and 25 mg benserazide in the lowest capsule dose. In conclusion, both commercially available combinations of levodopa and a peripheral decarboxylase inhibitor have a similar beneficial effect on parkinsonian symptoms and signs and their range and severity of sideeffects is similar at the same daily intake oflevodopa.

not

6

We thank Dr S. Rao

(Merck Sharp & Dohme Ltd) for supplies of levodopa and carbidopa (sinemet) and Dr M. Duffus (Roche Products Ltd) for levodopa and benserazide (madopar). We are grateful to Mrs J. Daniel for help in measuring blood-pressure and patient care, to Miss L. Kirby and Miss B. Edinborough for secretarial assistance, and to the patients and staff of the Parkinsonian Clinic, Hammersmith Hospital. J. K. G. is a Research Fellow of the Medical Research Council of Great Britain and J. L. R. is supported by the Wellcome Trust.

Requests tor reprints should be addressed to J. L. R., Department of Clinical Pharmacology, Royal Postgraduate Medical School, Du Cane Road, London W12 OHS. REFERENCES 1. Hornykiewicz, O. Pharmac. Rev. 1966, 18, 925. 2. Cotzias, G. C., van Woert, M. H., Schiffer, L. M. New

Engl. J. Med. 1967, 276, 374. 3. Calne, D. B., Spiers, A. S. D., Stern, G. M., Laurence, D. R., Armitage, P. Lancet, 1969, ii, 973. 4. Wurtman, R. J., Chow, C., Rose, C. M. J. Pharmac. exp. Ther. 1970, 174,

351. 5. Porter, C. C., Warren, L. S., Titus, D. C., Totars, J. A., Byer, S. S. Biochem. Pharmac. 1962, 11, 1067. 6. Bartholini, G., Pletcher, A. J. Pharm. Pharmac. 1969, 21, 323. 7. Calne, D. B., Reid, J. L., Varkil, S. D., Rao, S., Petrie, A., Pallis, C. A., Gawler, J., Thomas, P. K., Hilson, A. Br. med. J. 1971, in, 729. 8. Chase, T. N., Watanabe, A. M. Neurology, 1972, 22, 384. 9. Barbeau, A., Mars, H., Botez, M. I., Joubert, M. Can. med. Ass. J. 1972, 106, 1169. 10. Korten, J. J., Keyser, A., Joosten, E. M. G., Gabreels, F. J. M. Eur. Neurol. 1975, 13, 65. 11. Hodge, J. V., Oates, J. A., Sjoerdsma, A. Clin. Pharmac. Ther. 1964, 5, 149. 12. Rinne, U. K., Sonninen, V., Sirtola, T. Z. Neurol. 1972, 202, 1. 13. Marsden, C. D., Parkes, J. D., Rees, J. E. Lancet, 1973, ii, 1459. 14. Aitken, R. C. B. Proc. R. Soc. Med. 1969, 62, 989. 15. Lader, M. H., Marks, I. M. Acta psychiat. scand. 1974, 50, 112.

SALIVARY PHENYTOIN CONCENTRATIONS IN EPILEPSY AND IN CHRONIC RENAL FAILURE FELICITY REYNOLDS N. F. JONES

P. N. ZIROYANIS S. E. SMITH

Department of Pharmacology, and Renal Laboratory, St. Thomas’s Hospital and Medical School, London SE1

Summary

plasma unbound (free, therapeutically active), and salivary

Plasma total,

phenytoin concentrations have been measured in seventeen epileptics and in seven patients with chronic renal failure on long-term phenytoin therapy. Patients with chronic renal failure had low plasma total, but disproportionately high plasma free, drug concentrations indicating impaired protein binding. Even so, their plasma free drug concentrations never fell within the predicted therapeutic range on conventional phenytoin dosage. Salivary drug concentrations correlated closely with plasma free drug concentrations. These observations suggest that phenytoin therapy could be more appropriately monitored by measurement of salivary, rather than plasma, drug concentrations. Avoidance of bloodsampling would be an added advantage, particularly in children and in uræmia. The present results suggest that the optimal therapeutic range of salivary phenytoin would be 4-10 µmol/l. Introduction MANY studies

indicate that phenytoin (diphenylof epilepsy is aided by routine hydantoin) the of plasma concentration.1-6 The value of monitoring this rests on the assumption of a close correlation between drug levels in the plasma and in the brain,’e and depends on a constant degree of plasma-protein binding between individuals. Such an assumption is invalid when protein binding is disturbed, as in renal failure9-14 or infancy" or because of competition from other drugs.16 In such circumstances valid monitoring can be expected only from measurement of the unbound (free) fraction of the drug in the plasma, a procedure which is impractical for routine purposes. To overcome these limitations, we have measured plasma total and free drug concentrations in epileptics and in patients with chronic renal failure on phenytoin treatment and compared them with salivary concentrations obtained at the same time. Published reports indicate that in epileptics salivary concentrations of phenvtoin are about 10% of those in plasma. 17 We found that they correlated better with free than with total concentrations in plasma. Their measurement is technicallv simple and the procedure is advocated for routine use, now

treatment

Patients and Methods

Epileptic Patients Seventeen epileptic patients, (eight

female and nine mal2 3 55 were to studied. years, They were receiving age range 90-400 mg phenytoin daily, either alone (four patients) or with

other anticonvulsants including phenobarbitone, primidone, ’, carbamazepine, sulthiame, and diazepam. Patients with Chronic Renal Failure Measurements were made on six patients with end-st renal failure, who had had regular haemodialysis treatment f,’ 2 wk-55 months, and on one patient (case 7 in the acco5,