Gabapentin for Parkinsonism: A Double-Blind, Placebo-controlled, Crossover Trial

Gabapentin for Parkinsonism: A Double-Blind, Placebo-controlled, Crossover Trial Walter L. Olson, MD, Michael Gruenthal, MD, PhD, Marguerite E. Mueller, MD, William H. Olson, MD, Louisville, Kentucky

PURPOSE: Gabapentin is a recently available anticonvulsant whose mechanism of action remains unknown. We suspected efficacy from serendipitous observations of gabapentin in patients with parkinsonism. This led us to a double-blind, placebo-controlled, crossover trial. PATIENTS AND METHODS: We administered gabapentin in a placebo-controlled, doubleblind, crossover trial to 19 subjects with advanced parkinsonism. We measured the effect of placebo and gabapentin on subjects’ symptoms with the Unified Parkinson’s Disease Rating Scale, the Webster Scale, and the Hoehn and Yahr Scale. We assessed tremor with surface-recorded electromyography. RESULTS: Total Unified Parkinson’s Disease Rating Scale improved with gabapentin compared with placebo (P ! 0.0005). Likewise, activities of daily living and examination subscore of the Unified Parkinson’s Disease Rating Scale improved with gabapentin compared with placebo but did not achieve statistical significance. Webster Scale showed improvement but neither Hoehn and Yahr Scale nor Webster Scale changes reached statistical significance. Tremor as measured by the Unified Parkinson’s Disease Rating Scale improved with gabapentin but the use of the root mean square of the rectified electromyography as a measure of tremor activity was not statistically significant. CONCLUSIONS: This study demonstrates that gabapentin improves rigidity, bradykinesia, and tremor of parkinsonism including both Parkinson’s disease and Parkinson’s syndrome. The rigidity and bradykinesia of parkinsonism improve on the drug even when the effects of

From the Departments of Neurology (WLO, MG, WHO) and Internal Medicine (MEM), Division of Physical Medicine and Rehabilitation of the University of Louisville, Louisville, Kentucky. Requests for reprints should be addressed to Walter L. Olson, MD, Department of Neurology, Health Science Center 113 A, University of Louisville, 500 South Preston Street, Louisville, Kentucky 40292. Manuscript submitted June 14, 1996 and accepted in revised form October 16, 1996.

60

gabapentin on tremor are discounted. Am J Med. 1997;102:60–66.

P

arkinsonism is a common progressive, disabling movement disorder.1,2 Parkinsonism is caused by multiple mechanisms but its treatment is practically confined to interventions in the acetylcholinergic and dopaminergic systems. L-Dopa, dopamine agonists, and centrally-acting anticholinergic drugs are all associated with declining efficacy over time. Eventually the adverse effects of the available treatments are greater than the therapeutic benefit that can be achieved. Thus there is a pressing need to find medications that treat parkinsonism by alternate mechanisms. Gabapentin was recently released for use in epilepsy,3,4 though its mechanism of action has yet to be elucidated.5 A patient with leg cramps and parkinsonism placed on 400 mg gabapentin TID noticed that the tremor improved and that leg cramps and rigidity resolved. Observing similar benefit in other patients with parkinsonism led to this study to investigate the role of gabapentin in the treatment of parkinsonism.

PATIENTS AND METHODS Design Study design was a single-dosage placebo-controlled, double-blind, crossover study of gabapentin (Neurontin, Parke-Davis) in the treatment of symptoms of parkinsonism. Whereas other studies measure increased duration of ‘‘on-time’’ estimated by the subjects, this study was designed to probe for improved rating scale responses assessed by a blinded neurologist rater. Patients Men and women, aged 35 to 85, with clinical symptoms of advanced parkinsonism, Hoehn and Yahr stages 2.5 to 5, who were not pregnant, and who were able to give informed consent were eligible for the study. Potential study subjects were identified in the patient population of one of the investigators or self-identified from the membership in Parkinson disease support groups. Exclusion criteria included kidney disease, liver disease, bone marrow dyscrasia, and women of child-bearing capacity not on birth control. In order to have the study population as homogenous as possible, we excluded patients with

Q1997 by Excerpta Medica, Inc. All rights reserved.

/ 220a 0275 Mp

60

Friday Dec 27 09:00 AM

0002-9343/97/$17.00 PII S0002-9343(96)00381-6

EL–AJM (v. 102, no. 1)

0275

GABAPENTIN FOR PARKINSONISM/OLSON ET AL

TABLE I Concurrent Medications Drug Sinemet† Sinemet CR† Amantadine Selegiline Bromocriptine Pergolide Trihexyphenidyl

Number

Average Dosage*

Dosage Range*

9 6 5 9 8 6 4

688.9 483.3 220 7.7 23.75 0.925 8.5

100 300 200 5 5 0.05 6

2,000 800 300 10 80 1.5 12

*Dosages expressed in mg/day. † Drug dosage expressed is L-dopa.

dystonia, Huntington’s disease, hemiballism, tardive dyskinesia, essential tremor, and cerebellar ataxia in whom parkinsonism features were also present.

Methods Initially we evaluated the subjects with the Unified Parkinson’s Disease Rating Scale,6 The Webster Disability Rating Scale,7 and the Hoehn and Yahr Scale,8 all standardized assessments of disability in parkinsonism. To allow unbiased assessment of an independent blinded evaluator should it prove necessary, we videotaped the examination subscale of the Unified Parkinson’s Disease Rating Scale. In patients with tremor we obtained surface electromyography recordings from the most involved agonist/antagonist pair between 4 and 6 PM (day 1). The independent pharmacist randomized the subjects to one of two treatment arms. We gave one half the subjects 6 doses of gabapentin 400 mg orally and the other half 6 doses of placebo. The subjects took the first dose on the evening of day 1, then 3 divided doses on day 2, and 1 dose in the morning and 1 dose in the afternoon of day 3. We reevaluated and recorded them between 4 and 6 PM on day 3. Then they had an 11day wash-out (days 4 to 14). After a third evaluation and recording between 4 and 6 PM (day 15), we administered the alternate therapy. The subjects who received placebo at first were given active drug and vice versa. Following 2 days of drug treatment, we evaluated and recorded them for the last time between 4 and 6 PM (day 17), and discontinued study medication. The examination subscale of the Unified Parkinson’s Disease Rating Scale, the Webster Scale, and the Hoehn and Yahr scale were all administered by one investigator with training and experience in their use. We maintained the double blind throughout the study. The subjects receiving symptomatic treatment for their parkinsonism were advised to remain on these medications unchanged during the study period. Subject 18, treated by a neurologist not a member of the investigator team, had the dosing of concomitant

medication changed during the washout period between the placebo arm and drug arm of the study, and so was rejected from the study. No subject changed either the timing or dosage of medication during the study sequences of placebo or gabapentin treatment. Table I describes concurrent medication of study subjects. After obtaining Institutional Review Board approval (University Human Studies Committee, University of Louisville UHSC 291-94), we obtained informed consent from all subjects prior to enrolling them in the study.

Statistical Analysis A pharmacist who had no direct contact with any of the investigators or subjects and no knowledge of the subject’s clinical condition randomly assigned the study subjects to receive either gabapentin or placebo as initial treatment. Neither the subjects nor the investigators were informed of treatment assignments until the study was completed. The physical characteristics of the administered gabapentin and placebo were identical. We designed the study as a cross-over study to use the subjects as their own matched controls for all analyses. We utilized the SPSS 6.1 statistical computer program for data analysis. Statistical analysis was by Wilcoxon matched pairs ranked sum test of all ordinal data. We did t test analysis on the electromyography data. We compared baseline value with treatment value for both active drug and placebo to minimize the effects of possible variations in the baseline data as a function of the 11-day washout period. To investigate possible differences in the scores as a function of treatment order, we made unpaired comparisons between subjects assigned to receive placebo first and gabapentin first using the Mann-Whitney U test. All comparisons were twotailed. Owing to the large number of statistical comparisons needed, we sought to limit the type I error to 5% across the entire study. We accomplished this by adjusting the a level for each comparison according to the formula: a(study) Å 1 0 (1 0 a(each comparison))c, where C is the number of individual comparisons. Thus for 18 comparisons, the a level needed to consider any individual comparison significant was 0.0028.9

RESULTS We recruited 24 subjects and enrolled 19, 13 men and 6 women, aged 41 to 78 (average 64), who have clinical symptoms of parkinsonism, and had benefit from initial treatment with L-dopa, duration of disease 9.3 years (range 2 to 21), Hoehn and Yahr stage 2.5 to 5.0 (average 3.1). Five subjects were excluded: January 1997 The American Journal of MedicineT Volume 102

/ 220a 0275 Mp

61

Friday Dec 27 09:00 AM

EL–AJM (v. 102, no. 1)

0275

61

GABAPENTIN FOR PARKINSONISM/OLSON ET AL

Figure 1. Individual subject responses to treatment using Total Unified Parkinson’s Disease Rating Scale scores. Each line represents the response of a subject compared with baseline. Note there is one prominent placebo responder. Response to gabapentin is significantly better (Z Å 03.4672, P Å 0.0005) than placebo.

Figure 2. Individual subject responses to treatment using Unified Parkinson’s Disease Rating Scale activities of daily living scores. Each line represents the response of a subject compared with baseline. Note the majority of subjects had no change to placebo administration. Response to gabapentin appears significant, but only shows a trend toward improvement, P Å 0.0039.

1 had concurrent exposure to gabapentin, 1 was scheduled to be out of town during the study period, 1 was judged unable to give informed consent, 1 selfidentified subject had evidence of essential tremor rather than parkinsonism, and 1 had a change in concomitant medication between the placebo and drug arm of the study.

Unified Parkinson’s Disease Rating Scale Subjects taking gabapentin versus placebo had improved (Z Å 03.4672) total Unified Parkinson’s Disease Rating Scale scores (P Å 0.0005) as shown in Figure 1. None of the Unified Parkinson’s Disease Rating Scale subscales considered in isolation reached statistical significance, although both the activities of daily living (P Å 0.0039), shown in Figure 2, and examination (P Å 0.0073), illustrated in Figure 3, subscales revealed a trend toward improvement with gabapentin. Figure 4 summarizes the results of observations using the Unified Parkinson’s Disease Rating Scale. 62

Figure 3. Individual subject responses to treatment using Unified Parkinson’s Disease Rating Scale examination scores. Each line represents the response of a subject compared with baseline. Note that most subjects had small improvement to placebo administration. While the response to gabapentin appears significant, it is only a trend toward improvement, P Å 0.0073.

Figure 4. Unified Parkinson’s Disease Rating Scale subscales and total response. Each error bar represents the standard deviation of the graphed results. The error bars oriented downward pertain to the associated subscore; those oriented upward pertain to the total Unified Parkinson’s Disease Rating Scale score. Only statistically significant P values are displayed.

Webster and Hoehn and Yahr Scales Webster and Hoehn and Yahr scales tended to demonstrate improvement in the gabapentin-treated subjects versus the placebo-treated subjects. While neither achieved statistical significance, the Webster scale results demonstrated a trend toward efficacy of gabapentin (P Å 0.0061). Table II documents the results of the Webster and Hoehn and Yahr scales. Tremor Because gabapentin may be effective for the treatment of essential tremor,10 we elected to analyze the effect of gabapentin on the tremor-related items of the Unified Parkinson’s Disease Rating Scale (16, 20, and 21) separately and the drug effect on the Scale without these items to see the effect on parkinsonism tremor and on parkinsonism with the feature of tremor removed. The tremor items evaluated sepa-

January 1997 The American Journal of MedicineT Volume 102

/ 220a 0275 Mp

62

Friday Dec 27 09:00 AM

EL–AJM (v. 102, no. 1)

0275

GABAPENTIN FOR PARKINSONISM/OLSON ET AL

TABLE II Hoehn and Yahr and Webster Scale Results Baseline

Treatment

Rating Scale

Drug

Mean

SD

Mean

SD*

P Value

Hoehn and Yahr

Placebo Gabapentin Placebo Gabapentin

3.05 3.16 15.1 14.5

0.57 0.58 4.25 4.60

3.03 3.05 13.3 12.4

0.59 0.66 4.28 5.33

NS NS NS 0.0061 NS

Webster

SD Å standard deviation; NS Å not significant.

Electromyography Data Surface recording of the electromyogram was done in subjects who exhibited continuous tremor using a Viking II (Nicolet Instrument Corporation). We attempted to obtain an objective repeatable evaluation of parkinsonism tremor by using the root mean square analysis of the rectified spontaneous electromyography signal. The data, shown in Table IV, did not reach statistical significance. However, some subjects appeared to show substantial improvement. Figure 7 depicts an example of the electromyography signal obtained from subject 6. Figure 5. Individual subject responses to treatment using tremorrelated items of the Unified Parkinson’s Disease Rating Scale. Each line represents the response of a subject compared with baseline. Note both placebo responders and worsening subjects to placebo administration. There is no statistical difference between the responses.

Figure 6. Individual subject responses to treatment using total Unified Parkinson’s Disease Rating Scale scores with tremor-related items deleted. Each line represents the response of a subject compared with baseline. Note the consistently greater response to gabapentin, which is statistically significant (Z Å 03.2246, P Å 0.0013).

rately tended to show a drug effect, as seen in Figure 5, but did not reach statistical significance. However, the Unified Parkinson’s Disease Rating Scale without the tremor items demonstrated significant drug response (Z Å 03.2246, P Å 0.0013) illustrated in Figure 6. Table III documents the effects of gabapentin on the Unified Parkinson’s Disease Rating Scale and its tremor items in particular.

Double Blind During the study, subjects thought that they could detect whether they had received placebo or gabapentin. The investigator who gathered the side effects and adverse effects data recorded subject’s sequence ranking based on their subjective responses to the first exposure to study medications. Four patients were unable to determine whether they had received placebo or drug first and thus made no ‘‘assignment’’ of their treatment status. Statistical analysis by means of the chi-square test revealed that subjects could not accurately determine the actual sequence.11 Table V documents the sequence assignment compared with actual sequence in the study. Adverse Events The only serious adverse effect experienced through the study period was a fall with head injury and scalp laceration requiring sutures in subject 4. The fall occurred during the placebo arm of the trial. A computed tomography scan at the time of the fall revealed no detectable intracerebral injury, and the subject completed the study without any further adverse effect. Table VI lists the side effects experienced by 11 subjects on placebo and gabapentin. The incidence and profile of adverse effects is no different than that reported in other studies.

DISCUSSION It has been long recognized that degeneration of the substantia nigra is a causative factor for ParkinJanuary 1997 The American Journal of MedicineT Volume 102

/ 220a 0275 Mp

63

Friday Dec 27 09:00 AM

EL–AJM (v. 102, no. 1)

0275

63

GABAPENTIN FOR PARKINSONISM/OLSON ET AL

TABLE III Results with Reference to Parkinson Tremor Baseline Rating Scale Total Unified Parkinson’s disease Rating Scale Tremor* Total Unified Parkinson’s Disease Rating Scale without tremor†

Treatment

Drug

Mean

SD

Mean

SD

P Value

Placebo Gabapentin Placebo Gabapentin

64.2 63.5 5.5 4.9

17.2 20.4 4.50 4.24

58.6 51.5 5.1 3.3

17.8 20.6 5.54 3.38

NS 0.0005 NS NS

Placebo Gabapentin

58.6 58.5

16.8 19.3

53.5 48.3

17.1 19.6

NS 0.0013

* The tremor scale results are the total of the Unified Parkinson’s Disease Rating Scale items 16, 20, and 21 (see text). † The Total Unified Parkinson’s Disease Rating Scale without Tremor is the total Unified Parkinson’s Disease Rating Scale score with items 16, 20, and 21 subtracted. SD Å standard deviation; NS Å not significant.

TABLE IV Results of Root Mean Square of Rectified Electromyography Baseline Drug

Number Mean

Placebo Gabapentin

8 9

Treatment

SD

Mean

SD

P Value

49.44 39.50 26.38 28.07 42.83 38.07 26.39 25.60

NS NS

NS Å not significant.

TABLE V Assessment of Maintenance of Blind Sequence Assignment

P ú D*

D ú P†

P Value

2 3 5

1 1 7

NS NS NS

Wrong No Assignment Correct

Figure 7. Surface-recorded electromyography of subject 6. Two-second samples of surface-recorded electromyography from the first dorsal interosseus of the left hand of subject 6.

* P ú D is placebo followed by gabapentin treatment. † D ú P is gabapentin treatment followed by placebo administration.

TABLE VI

son’s disease12 related to loss of dopaminergic supply to the striatum and that levodopa given to patients with parkinsonism is metabolized to dopamine with beneficial effect on the symptoms.13 To date the principal treatments of parkinsonism have been oriented to dopaminergic supply.14 It is known that dopamine from the substantia nigra acting at the striatum15 results in an inhibitory cascade that is GABAergic from caudate, putamen, and globus pallidus externa.16 Gabapentin, (1-aminomethyl)-1-cyclohexaneacetic acid, was synthesized as a structural analogue of GABA.17 However, gabapentin does not act at GABAA, GABAB,18 or GABAC19 receptors, is not metabolized into GABA or a GABA agonist,20 does not inhibit GABA uptake21 nor degradation by GABAtransaminase,22 but it is known to be a competitive 64

Adverse Effects of Medication Experienced by Subjects Gabapentin Adverse Effect

Number

Percent

5 4 4 2 2 2 1 1 1

25 20 20 10 10 10 5 5 5

Drunk/unsteady on feet Dizzy/lightheaded Sleepy Nausea Dry mouth Nightmares Hallucination Confusion Weakness

64

Friday Dec 27 09:00 AM

Percent

1 1 1 1 1

5 5 5 5 5

inhibitor of branched-chain amino acid aminotransferase and stimulates the activity of glutamate dehydrogenase.22 It is known that gabapentin increases

January 1997 The American Journal of MedicineT Volume 102

/ 220a 0275 Mp

Placebo Number

EL–AJM (v. 102, no. 1)

0275

GABAPENTIN FOR PARKINSONISM/OLSON ET AL

brain GABA,23,24 enhances the release of GABA from rat neostriatum,25 and inhibits monoaminergic release from the striatum, but not acetylcholinergic release from the striatum.26 Further, gabapentin does not act at benzodiazepine, glutamate, glycine, Nmethyl-d-aspartate receptors,27 nor does it influence sodium or calcium channels.28,29 It is not a substrate for any of the enzyme systems involved in the synthesis or catabolism of GABA.22 In neurons, the gabapentin-binding site30 is associated with L-neutral amino acid transporter31 on cell bodies.32 We chose a dose of 400 mg three times a day because we noted that a dose of 300 mg three times a day in other patients with parkinsonism has little or no effect. The 11-day washout period was chosen because in a companion study33 we noted that subjects with spinal cord injury did not return to baseline for at least a week after cessation of gabapentin despite a half life (t1/2) of 4 to 6 hours.34 In this trial, 400 mg gabapentin administered TID improved the total Unified Parkinson’s Disease Rating Scale score, suggesting that gabapentin exerts an effect on the symptoms and signs of parkinsonism. In addition, subjects taking gabapentin manifested improvement in their Unified Parkinson’s Disease Rating Scale activities of daily living subscore over placebo therapy, documenting a benefit on the subjective symptoms of Parkinsonism. Because gabapentin has been anecdotally reported to benefit patients with essential tremor,10 we elected to look at gabapentin effects on parkinsonism tremor. Of interest, even when those measures of tremor in the Unified Parkinson’s Disease Rating Scale were removed from analysis, the total Scale scores still demonstrated a significant response to drug. Heretofore, this improvement in nontremor signs and symptoms of parkinsonism has been observed only with administration of dopamine receptor agonists. These study results are the first evidence of significant treatment benefit by an agent having no known cholinergic or dopaminergic effect in parkinsonism. However, since all subjects were taking at least one dopaminergic agent, we cannot rule out a synergistic effect between gabapentin and the concurrent dopaminergic drug(s). The failure of gabapentin to demonstrate a statistically significant effect on the examination scale of the Unified Parkinson’s Disease Rating Scale may be due to the design of the trial itself. The last dose of drug, placebo or gabapentin, was noon of the day of evaluation, and the evaluation was between 4 PM and 6 PM, about 4 to 6 hours after the dose. Gabapentin t1/2 is 4 to 6 hours, which suggests that subjects may have waning of effect. Thus, gabapentin likely affects the examination subscale of the Unified Parkinson’s Disease Rating Scale, but to document

this the evaluation should occur 1 to 3 hours after the dose, not 4 or more hours afterward. We were interested in the possible effects of gabapentin on parkinsonism tremor, so we analyzed a number of measures of this effect. Three items in the Unified Parkinson’s Disease Rating Scale (16, 20, and 21) and one item in the Webster scale (6) measure tremor effects. Tremor activity in the electromyography can be recorded with surface electrodes. While there seemed to be some improvement in each of these measures when subjects took gabapentin, none of the effects reached statistical significance. All subjects had advanced parkinsonism, at least Hoehn and Yahr stage 2.5. Fourteen subjects had response fluctuations to their prescribed dopaminergic agents, but we collected ‘‘off’’ and ‘‘on’’ time activities of daily living data in only 7 subjects. Because meaningful results could not be obtained, we did not analyze our data for ‘‘off’’ or ‘‘on’’ effects. The brevity of the study limited risks that may accrue in long-term treatment with antiparkinsonism medications.35 – 38 It also restricts conclusions regarding long-term efficacy and tolerability. Since parkinsonism is a lifelong illness once manifest, an investigation of long-term effects is needed. We have observed that some patients have received continued undiminished benefit from gabapentin for more than a year. Although the mechanism of action of gabapentin still remains elusive, it seems to increase brain GABA,24 including striatal GABA,23 and to increase GABA release and acetylcholine release while reducing other monoaminergic release.26 Gabapentin also doubles postsynaptic GABA potentials, an effect blocked by GABAA-receptor inhibition.39 GABAA receptors are known to be prominent in the striatum, globus pallidus interna, and substantia nigra reticulata40 whereas GABAB receptors are more commonly seen in the substantia nigra compacta.41 All of the GABA receptors are expressed in the thalamus. Since parkinsonism is associated with relatively increased function of the indirect inhibitory GABAergic pathway, improvement in the condition may result from stimulation of GABAA receptors with the immediate feedback to inhibit further GABA release.25 Alternatively, if the striatal indirect GABAergic pathway is involved, action of GABA on the globus pallidus interna and substantia nigra reticulata will decrease inhibition of target thalamic neurons and increase mobility. Since striatal GABA-mediated postsynaptic inhibitory activity is an action in concert with the action of the dopamine on the striatum in the indirect pathway,42 this may explain the effect of gabapentin on parkinsonism. Regardless, additional work to elucidate the mechanism of action of gabapentin as an January 1997 The American Journal of MedicineT Volume 102

/ 220a 0275 Mp

65

Friday Dec 27 09:00 AM

EL–AJM (v. 102, no. 1)

0275

65

GABAPENTIN FOR PARKINSONISM/OLSON ET AL

adjunctive medicine in parkinsonism needs to be done. One final note of caution seems pertinent. While this study did not analyze the effect of gabapentin on specific forms of parkinsonism, we have observed that 5 of 6 patients with progressive supranuclear palsy, who had not been subjects of this study, had worsening of their disease when given gabapentin, becoming bedfast (Hoehn and Yahr stage 5). Fortunately, this is a reversible effect on stopping the gabapentin. This observation suggests that not all forms of parkinsonism will benefit from gabapentin.

ACKNOWLEDGMENTS The authors gratefully acknowledge the support of the Frazier Rehabilitation Center for providing space, personnel, secretarial, and pharmacy support for the study. Especially the authors wish to thank Victor H. Wood, RN, CRRN, the clinical nurse coordinator, and Judith G. Olson, RN, who videotaped the subjects and offered her editorial skills. Gabapentin samples were provided by ParkeDavis.

REFERENCES 1. Weiner WJ, Lang AE. Parkinson’s disease. In: Movement Disorders: A Comprehensive Survey. Mount Kisco, New York: Futura; 1989:23–115. 2. Weiner WJ, Lang AE. Other akinetic-rigid and related syndromes. In: Movement Disorders: A Comprehensive Survey. Mount Kisco, New York: Futura; 1989:117–219. 3. Leach JP, Brodie MJ. New antiepileptic drugs—an explosion of activity. Seizure. 1995;4:5–17. 4. Andrews CO, Fischer JH. Gabapentin: a new agent for the management of epilepsy. Ann Pharmacother. 1994;28:1188–1196. 5. McLean MJ. Gabapentin. Epilepsia. 1995;36(suppl 2):S73–86. 6. Fahn S, Elton RL, Members of the UPDRS Development Committee. Unified Parkinson’s Disease Rating Scale. In: Fahn S, Marsden CD, Calne DB, et al, eds. Recent Development in Parkinson’s Disease. Vol 2. Florham Park, New Jersey: Macmillan Health Care Information; 1987:153–163, 293–304. 7. Webster DD. Critical analysis of the disability in Parkinson’s disease. Modern Treatment. 1968;5:257–282. 8. Hoehn MM, Yahr MD. Parkinsonism: onset, progression, and mortality. Neurology. 1967;17(5):427–442. 9. Keppel G. Correction for cumulative type I error. In: Design and Analysis: A Researcher’s Handbook. 3rd ed. Upper Saddle River, New Jersey: Prentice-Hall; 1991:163–184. 10. Burrows GT, King RB. Gabapentin in essential tremor. Neurology. 1995;45(suppl 4):A187–188. Abstract. 11. Conover WJ. Contingency tables. Practical Non-Parametric Statistics. 2nd ed. New York: Wiley & Sons; 1980:143–212. 12. Tretiakoff C. Contribution l’etude de l’anatomie pathologique du locus niger. University of Paris; 1919. 13. Birkmayer W, Hornykiewicz O. Der L-3,4-dioxyphenylalanin (L-Dopa)-effect bei der parkinson akinese. Wien Klin Wochenschr. 1961;73:787–788. 14. Wolters EC, Tissingh G, Bergmans PL, Kuiper MA. Dopamine agonists in Parkinson’s disease. Neurology. 1995;45(3 suppl 3):S28–34. 15. Groves PM, Garcia-Munoz M, Linder JC, et al. Elements of the intrinsic organization and information processing in the neostriatum. In: Houk JC, Davis JL, Beiser DG, eds. Models of Information Processing in the Basal Ganglia. Vol 8. Cambridge, Massachusetts: MIT Press; 1995:51–96. 16. Ferre S, O’Connor W, Fuxe K, Ungerstedt U. The striopallidal neuron: a main locus for adenosine-dopamine interactions in the brain. J Neurosci. 1993;13:5402–5406. 17. Satzinger G, Hartenstein J, Herrmann M, Heldt W. Cyclic amino acids. In: Advanced Organic Chemistry. Vol 260. United States: Warner-Lambert Company; 1977:1–8.

66

18. Taylor CP. Mechanism of action of new antiepileptic drugs. In: Chadwick D, ed. New Trends in Epilepsy Management: The Role of Gabapentin. London: Royal Society of Medicine Services Limited; 1993:13–40. 19. Satzinger G. Antiepileptics from gamma-aminobutyric acid. Arzneimittelforschung. 1994;44:261–266. 20. McLean MJ. Clinical pharmacokinetics of gabapentin. Neurology. 1994;44:S17–22. 21. Bartoszyk GD, Meyerson N, Reimann W, et al. New anticonvulsant drugs. Gabapentin. In: Meldrum BS, Porter RJ, eds. Current Problems in Epilepsy. Vol 4. London: John Libbey; 1986:147–163. 22. Goldlust A, Ti-Zhi S, Welty DF, et al. Effects of anticonvulsant drug gabapentin on the enzymes in metabolic pathways of glutamate and GABA. Epilepsy Res. 1995;22(1):1–11. 23. Loscher W, Honack D, Taylor CP. Gabapentin increases aminooxyacetic acid-induced GABA accumulation in several regions of rat brain. Neurosci Lett. 1991;128:150–154. 24. Petroff OAC, Rothman DL, Behar KL, et al. The effect of gabapentin on brain gamma-aminobutyric acid in patients with epilepsy. Ann Neurol. 1996;39(1):95–99. 25. Go¨tz E, Feuerstein TJ, Lais A, Meyer DK. Effects of gabapentin on release of g-aminobutyric acid from slices of rat neostriatum. Arzneimittelforschung. 1993;43:636–638. 26. Schlicker E, Reimann W, Gothert M. Gabapentin decreases monoamine release without affecting acetylcholine release in the brain. Arzneimittelforschung. 1985;35:1347–1349. 27. Kelly KM, Rock DM, MacDonald RL. Gabapentin does not affect N-methylD-aspartate receptor currents or voltage-dependent calcium currents in cultured rodent neurons. Ann Neurol. 1991;30(2):292. Abstract. 28. Rock DM, Kelly KM, MacDonald RL. Gabapentin actions on ligand- and voltage-gated responses in cultured rodent neurons. Epilepsy Res. 1993;16:89– 98. 29. Taylor CP. Emerging perspectives on the mechanism of action of gabapentin. Neurology. 1994;44:S10–16. 30. Suman-Chauhan N, Webdale L, Hill DR, Woodruff GN. Characterisation of [3H]gabapentin binding to a novel site in rat brain: homogenate binding studies. Eur J Pharmacol. 1993;244:293–301. 31. Thurlow RJ, Brown JP, Gee NS, et al. [3H]gabapentin may label a systemL-like neutral amino acid carrier in brain. Eur J Pharmacol. 1993;247:341–345. 32. Hill DR, Suman-Chauhan N, Woodruff GN. Localization of [3H]gabapentin to a novel site in rat brain: autoradiographic studies. Eur J Pharmacol. 1993;244:303–309. 33. Gruenthal M, Mueller ME, Olson WL, et al. Gabapentin for the treatment of spasticity in patients with spinal cord injury. Unpublished data 1996. 34. Goa KL, Sorkin EM. Gabapentin. A review of its pharmacological properties and clinical potential in epilepsy. Drugs. 1993;46:409–427. 35. Meunter MD, Sharpless NS, Tyce GM. Patterns of dystonia (I-D-I and D-I-D) in response to L-dopa therapy for PD. Mayo Clin Proc. 1977;52:163–174. 36. Klawans HL, Goetz C, Bergen D. Levodopa induced myoclonus. Arch Neurol. 1975;32:331–334. 37. Barbeau A. Contributions of levodopa therapy to the neuropharmacology of akinesia. In: Siegfried J, ed. Parkinson’s Disease: Rigidity, Akinesia, Behavior. Vol 1. Bern: Hans Huber; 1972:151–174. 38. Barbeau A. The clinical physiology of side effects and long term L-dopa therapy. Adv Neurol. 1974;5:347–365. 39. Kocsis JD, Honmou O. Gabapentin increases GABA-induced depolarization in rat neonatal optic nerve. Neurosci Lett. 1994;169:181–184. 40. Wisden W, Laurie DJ, Monyer H, Seeburg PH. The distribution of 13 GABAA receptor subunit mRNAs in the rat brain. I. Telencephalon, diencephalon, mesencephalon. J Neurosci. 1992;12:1040–1062. 41. Hausser MA, Yung WH. Inhibitory synaptic potentials in guinea-pig substantia nigra dopamine neurones in vitro. J Physiol (Lond). 1994;479(pt 3):401– 422. 42. Alexander GE, Crutcher MD, DeLong MR. Basal ganglia-thalamocortical circuits: parallel substraits for motor, oculomotor, ‘‘pre-frontal’’ and ‘‘limbic’’ functions. Prog Brain Res. 1990;85:119–146.

January 1997 The American Journal of MedicineT Volume 102

/ 220a 0275 Mp

66

Friday Dec 27 09:00 AM

EL–AJM (v. 102, no. 1)

0275