186 Letters to the Editor J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.54.2.186 on 1 February 1991. Downloaded from 2 l ournier-Lasserve E, Gout A, Gessain A, et al. males; mean age 57-7 (41-69) years; mean dose of bromocriptine and lasting for 115 HTLV-I, brain abnormalities on magnetic mean minutes. resonance imaging, and relations with mul- duration of disease 10-9 (8-15) years; tiple sclerosis. Lancet 1987;ii:49-50. daily dose of levodopa 1014 (500-1400) mg) Using a similar single rising dose substitu- 3 Annunziata P, Fanetti G, Giarratana M, et al. plus peripheral decarboxylase inhibitor tion model, acute anti-Parkinsonian effects HTLV-I associated spastic paraparesis in an (PDI). All but two had never been treated have been demonstrated for the non-ergot Italian woman. Lancet 1987;ii: 1393-4. 4 Gout A, Gessain A, Bogert F, et al. Chronic with bromocriptine. dopamine agonist PHNO4 and also for the myelopathies associated with human T-lym- In one patient concomitant bromocriptine ergot derivative CQA 209-291, which is photrophic virus type I. Arch Neurol (30 mg/d) was discontinued one day before structurally related to bromocriptine.5 The 1989;46:255-60. lack of effect of bromocriptine seen in the 5 MMWR. Licensure of screening tests for the study while a second had received antibody to human T lymphotropic virus type bromocriptine (40 mg/d) as an adjunct to majority ofpatients in this study may indicate 1. Mortality and Morbidity Weekly Report levodopa over a 14 months period until one that the doses employed were subthreshold 1988;37:736-47. year previously. All showed a predictable for anti-Parkinsonian efficacy. 6 Madden DL, Mundon FK, Tzan NR, et al. Antibody to human and simian retrovirus, wearing-off pattern in response to oral It is, however, remarkable that several HTLV-I, HTLV-II, HIV, STLV-III and levodopa and during the study period the first clinical studies have demonstrated the effec- SRV-I not increased in patients with multiple daily dose of 200 mg of levodopa was taken at tiveness of bromocriptine monotherapy with sclerosis. Ann Neurol 1988;23(suppl): 8 am after a minimum drug free period of chronic administration at doses of between S171-3. eight hours and one hour after a standard 12 5 and 25 mg/d,'-3 that is, daily doses in the hospital breakfast. On alternate days this range of the upper dose levels employed in morning dose was replaced by single rising this single dose trial. A possible explanation doses of bromocriptine following the dose for this apparent discrepancy might be that schedule depicted in the figure, and oral bromocriptine is capable of inducing delayed Failure of oral administration of single domperidone (20 mg) was added if nausea effects with chronic treatment, possibly via rising doses of bromocriptine to pro- and/or hypotension had occurred at the modulatory effects on central dopamine duce acute anti-Parkinsonian effects previous dose level. Efficacy assessments of receptors. Such delayed effects have been levodopa or bromocriptine test doses were noted in one patient in this study (case 7, The optimal dose of bromocriptine in the performed using the motor score (section III) who received combined treatment with treatment of Parkinson's disease has of the Unified Parkinson's Disease Rating bromocriptine until 24 hours before the remained controversial. While most investi- Scale (UPDRS) beginning 30 minutes before study). He complained of increased severity gators have used daily doses of above 40 mg dosing with half-hourly ratings until acute of levodopa-induced dyskinesias on the days several reports have since claimed success drug effects had worn offor up to a maximum following bromocriptine challenges of 10, 15, with oral dose regimes of less than of four hours. 20 and 25 mg and experienced enhanced 20 mg/d.'2 The United Kingdom Bromocrip- The results are shown in the figure. While dyskinesias for another week after the end of tine Research Group has recently confirmed 200 mg of oral levodopa led to a mean 50% the study when receiving his pre-study daily the puzzling finding that with a fast intro- reduction ofthe UPDRS motor score, usually levodopa regime. But even this patient failed ductory regime higher doses of bromocrip- within 30 to 45 minutes, there were no acute to show acute anti-Parkinsonian effects from tine are needed than with a slower titration anti-Parkinsonian effects following any ofthe single bromocriptine doses as high as 25 mg. schedule to produce similar degrees of single oral doses of bromocriptine employed Other reasons for the observed failure of clinical improvement in de novo patients with in this study. The only exception was the single bromocriptine doses to induce acute Parkinson's disease.3 patient to whom bromocriptine had been effects could lie in the pharmacokinetics of We have studied the acute anti-Parkinson- given as a chronic treatment for up to one year the drug. Thus the extensive first pass effect ian effect of single rising doses of bromocrip- before the study. The patient experienced an of bromocriptine means that only about 8% tine in an open experimental study in seven acute switch on with complete cessation of a of the parent drug will appear in the systemic patients admitted to hospital with fluctuating coarse resting tremor and moderate peak dose circulation, and its high lipophilicity could Parkinson's disease (three females, four chorea 120 minutes following a single 12 5 mg further reduce the free concentration of bromocriptine in the extracellular space. Only after saturation of the lipid compart- 40 ment following repeated dosing might active A A AL A A A biological concentrations be attained within the soluble compartment. Accordingly

bromocriptine exerts acute effects in the http://jnnp.bmj.com/

+ + Ungerstedt rat model only following high + oral doses of above 9 mg/kg, but with repeated dosing response latencies of a given 30* x + + x dose decrease with simultaneous augmenta- aL) 0 0 tion of efficacy (R Markstein, personal com- 0 munication). Whether such pharmacokinetic A 0 0 properties of bromocriptine or pharmaco- 0 0 dynamic receptor changes in the CNS form C,') V the pathophysiological basis for the different on September 27, 2021 by guest. Protected copyright. 0 20 dose requirements with "fast" versus "slow" 0- + O~ introductory regimes of bromocriptine remains unclear. WERNER POEWE LUDWIG SCHELOSKY BIRGIT KLEEDORFER Department ofNeurology, 10 University of Innsbruck, A-6020 Innsbruck, Austria

1 0 Teychenne PF, Bergsrud D, Racy A, Elton R, Vein B. Bromocriptine: Low-dose therapy in Baseline Dopa 2 5 5.0 7.5 10.0 12.5 15-0 20-0 25-0 Parkinson's disease. Neurology 1982;32: ~~~577-83 Bromocriptine (mg) 2 Grimes JD, Delgado MR. Bromocriptine: Problems with Low-Dose De Novo Therapy (n=7) (n=7) (n=6) (In = 7 ) (n = 5) (n = 6) (n = 5) (n=1 ) (n=2) (n1=l ) in Parkinson's Disease. Clin Neuropharm 1985;8:73-7. Figure Bromocriptine single dose challienges in seven patients. Each patient represented by one 3 UK Bromocriptine Research Group. swmbol indicating minimum scores (U]PDRS part III) over a four hour rating periodfollowing Bromocriptine in Parkinson's disease: a each dose. Mean minimum scores for eiach dose level represented by horizontal bars. Effects of a double-blind study comparing "low-slow" single oral dose of 200 mg levodopa aree shownfor comparison. and "high-fast" introductory dosage in de novo * = regimens patients. J Neurol ... p 0 001 for comparison betweeen baseline ("off') and drug scores.scores.Neurosurg Psychiatry 1989;52:77-82. Letters to the Editor 187

4 Grandas Perez FJ, Jenner PG, Nomoto M, nated intravascular coagulation.' 7 Komory Y, Sugihara H. Biological study of J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.54.2.186 on 1 February 1991. Downloaded from et ai. (+ )-4-Propyl-9-Hydroxynaphthox- The electrophysiological signs and the muscle degenerating hemorrhagic factor from azine in Parkinson's Disease. Lancet 1986;i: the venom ofvipera aspis aspis (Aspic Viper). 906. quick improvement ofthe clinical picture also Int JBiochem 1988;20:1417-23. 5 1'cmlett JA, Quinn QN, Marsden CD, Lataste lead us to exclude a neuropathic lesion and to 8 Swift TR, Ignacio OJ. Tick paralysis. Electro- X, Jaton AL. The Antiparkinsonian Activity hypothesise that a transient functional block physiologic studies. Neurology 1975;25: of CQA 206-291, a new D, Dopamine 1130-33. Receptor Agonist. Clin Neuropharmacol of activation of a number of muscle fibres. 9 DettbarnWD, Higman HB, Rosenberg P, et al. 1989;12:55-9. This could be related to three possible Rapid and reversible block of electrical mechanisms in particular: 1) a neuromuscular activity by powerful marine biotoxins. Science block; 2) a direct action on muscle fibres; 3) a 1960;132:300-1. block of depolarisation in the terminal por- tions ofa number of motor nerve fibres. A neuromuscular block may be related either to a presynaptic site of action of the Neuromuscular paralysis in vipera venom, such as beta-bungarotoxin4 and aspis envenomation: pathogenetic antiacetylcholinesterase, or to a postsynaptic mechanisms site ofaction, like alpha-bungarotoxin.2 None of these substances has been detected in Vipera aspis is the most common agent of vipera aspis and the electrophysiological MATTERS snake envenomation in Italy and Western findings of the reported case are neither Europe.' Its bite affects coagulation and consistent with a presynaptic nor a postsyn- ARI SING causes a shock syndrome with severe car- aptic defect ofneuromuscular transmission. diovascular failure. A direct myotoxic effect ofanimal toxin has Neurotoxicity, clinically characterised by been related to phospholipase A2 activity, external ophthalmoplegia, is uncommon (two which has been detected in all viperidae cases out of 205 patients bitten by vipera Comparison of two methods for venoms so far investigated.2 Moreover some thermal thresholds aspis)' and difficult to explain because overt authors5 suggest that some toxins, like car- measuring neurotoxic substances have not been detected diotoxin of Dendroaspis jamesoni, can induce in vipera aspis venom.2 Our case suggests that In their recent paper,' Drs Levy, Abraham musclefibrenecrosis witha structural damage and Reid compare two techniques for the venom is neurotoxic. of the subneural apparatus. Nevertheless A 20 year old herpetologist was bitten by a measuring thermal thresholds in diabetic myonecrotic action is shown to be confined to neuropathy. On the basis of their results they vipera aspis at the distal extremity of the the site of injection.67 index finger of the left hand. When he was conclude that there is little to choose between The action of the toxin on the terminal the method of limits and the forced choice admitted to the intensive care unit (30 min- portions of motor fibres could transiently utes later) he was unconscious (Glasgow procedure of psychophysical analysis in the block the conduction of a number of motor determination of thermal thresholds. We Coma Scale 7), pale, tachycardic (170 beats/ fibres by preventing their depolarisation. A min), tachypnoeic (50 breaths/min), without believe that their results and-the conclusions lesion in this location is consistent with based thereon are incorrect and a con- detectable peripheral pulses and blood pres- normal tests of neuromuscular transmission sure. There was a metabolic acidosis sequence of their experimental design. and with the rapid recovery of the amplitude When comparing two techniques attempt- (pH 7 26) and disseminated intravascular of the muscle action potential as observed in coagulation. The left hand was oedematous. ing to measure the same parameter it is our case. This mechanism has been hypothes- imperative that all variables are comparable Centrifugal venous compression was applied ised also in the neuromuscular paralysis on the left arm. Shock, metabolic failure and and strictly controlled since they influence induced by tick envenomation5 and by other the accuracy of the final results."7 By their disseminated intravascular coagulation syn- biotoxins such as tetradotoxin.9 drome were treated with fresh frozen plasma, own admission the authors have ignored a Why the neurotoxic action of the vipera number of these variables as follows: albumin, dextran, dopamine and adrenaline, aspis venom appears to remain strictly NaHCO, and heparin iv infusions. Cardio- 1) The reference skin temperatures for the localised in cephalic muscles remains un- Sensortek and Marstock methods are respiratory function, metabolic balance and explained. Peculiar physiological characteris- consciousness returned to normal within the different (30'C and 32'C respectively).23 tics of cephalic motor units might be an Neither is in the optimum range 34-35'C at following three hours. explanation. Neurological examination revealed facial which the variability ofthe thermal threshold

GIOVANNI ANTONINI http://jnnp.bmj.com/ diplegia, pharyngolaryngeal paresis, bilateral MAURIZIA RASURA measurements is minimal.67 ptosis and external ophthalmoplegia, with GIORGIO CONTI* 2) The rate of temperature -change in the complete ocular immobility. CONSALVO MATTIA* Marstock technique is fixed. By comparison Department ofNeurological Scence, in the Sensor- The strength of the trunk, limb and res- Insitute ofAnesthesiology and Resuscitation,* the rate of temperature change piratory muscles, deep tendon reflexes, plan- University ofRome "La Sapienza", tek technique, as described by the authors, tar and abdominal reflexes, and sensory func- Rome, Italy varies not only during the application of a single stimulus but also for different applica- tions were normal. Symptoms were not Correspondence to: Dr Giovanni Antonini, Dipar- modified by iv administration of 10mg of timento di Scienze Neurologiche, III Clinica tions. This is a source of variability.3467 edrophonium. Neurologica, Viale dell'Universita 30,00185 Rome, 3) In the Sensortek technique two stimuli of Neurophysiological studies of the facial Italy. different modalities are applied to the skin on September 27, 2021 by guest. Protected copyright. nerves showed a low amplitude muscle action more or less simultaneously; there is a tactile potential (0 9 mv-nv > 3 mv), with normal stimulus (when the thermode is appliedto the latency. Repetitive stimulation at low and skin) in addition to the specific thermal high frequencies, tetanisation and stimula- 1 Pozio E. Venomous snake bites in Italy: stimulus. It is particularly important that a tion with paired stimuli at stimulus intervals epidemiological and clinical aspects. Trop pure thermal stimulus is applied without Med P.arasit 1988;39:62-66. tactile cues as the latter has been shown to of less than 10 ms gave normal responses 2 Anthony T. Venoms: chemistry and molecular without signs ofneuromuscular transmission biology. New York: John Wiley, 1976. modify thermal sensation.' defects. Blink reflex showed responses with 3 Schwartman RJ, Hill JB. Neurologic complica- 4) The duration of application of the ther- normal latencies. Similar neurophysiological tions of disseminated intravascular coagula- mode is poorly controlled in the Sensortek tion. Neurology 1982;32:791-7. method. This will influence both the amount studies performed on other nerves (median, 4 Dreyer F, Penner R. The action of presynaptic common peroneal and sural) were normal. snake toxins on membrane currents ofmouse and rate of energy transferred to the receptor Five days from the onset of the disease the motor nerve terminals. J Physiol 1987; zone. 386:455-63. 5) The pressure of application of the ther- patient improved considerably and after 10 5 Duchen LW, Excell BJ, Patel R, Smith B. days, neurological examination and neuro- Changes in motor end-plates resulting from mode to the skin is uncontrolled in both physiological tests were noranal. He was dis- muscle fibre necrosis and regeneration: a light techniques. The authors state that the impor- charged after 10 days. and electron microscope study of the effects tance of this factor "in clinical testing has not of the depolarizing fraction (cardiotoxin) of been systematically investigated". This is The lack of clinical involvement of motor, Dendromaspis Jamesoni venom. J Neurol Sci sensory and cerebellar pathways within the 1974;21:391-417.. incorrect.67 brainstem, together with the normal latency 6 Harris JB, Karlsson E, Thesleff S. Effects of an 6) The lack of calibration of heat transfer at isolated toxin from Australian Tiger Snake the thermode-skin interface in both tech- of blink reflex responses in this case, do not (Notechis Scutatus Scutatus) venom at the suggest an involvement of the brainstem mammalian neuromuscular junction. Brit J niques does not allow for the variability ofthe possibly caused by oedema and/or dissemi- Pharnacol 1973;47:141-6. thermal properties of the skin.89