- Email: [email protected]
LONG-TERM TREATMENT OF PARKINSONISM WITH BROMOCRIPTINE
735
because a visit to the clinic is a period of inactivity and a break with normal routine. The admission to hospital of patients with unstable control for blood-glucose profiles is wasteful of beds and artificial since life in hospital is unlike normal life. Selfmonitoring overcomes these problems and saves the patient unnecessary hospital visits since changes in treatment can be made by telephone. Home blood-glucose monitoring can also be used to assess long-term control. Its advantage over measurement of HbAlya-’6 is that it not only detects poor control, but also shows how to correct it.
We thank the special trustees of Nottingham University Hospital for grant to R. B. T.; and our secretaries, Mrs Jane Richards and Mrs Judith Halford, for their help. a
Requests for reprints should be sent to R. B. T. REFERENCES 1. Engerman, R., Bloodworth, J. M. B. Jr., Nelson, S. Diabetes, 1977, 26, 760. 2. Mauer, S. M., Steffes, M. W., Sutherland, D. E. R., Najarian, J. S., Michael, A. F., Brown, B. M. ibid. 1975, 24, 280. 3. Fox, C. J., Darby, S. C., Ireland, J. T., Sönksen, P. H. Br. med. J. 1977,
ii, 605. Karson, K., Kjellmer, I. Am. J. Obstet. Gynec. 1972, 112, 213. 5. Essex, N., Pyke, D. A., Watkins, P. J., Brudenell, J. M., Gamsu, H. R. Br. med. J. 1973, iv, 89. 6. Cahill, G. F., Etzwiler, D. D., Freinkel, N. New Engl. J. Med. 1976, 296, 4.
1004.
impressed that blood-glucose could be maintained at all times near or within the physiological range without dangerous hypoglycaemia even in relatively long-standing diabetics in whom endogenous insulin secretion is presumably lacking. 17 patients were included in our study because they were thought to be well controlled on clinical criteria, and of these 14 had initial blood-glucose profiles where no value exceeded 10 mmola. 18 more patients were brought within this level of control by adjustment of their treatment with the aid of a Reflomat. Case 1 illustrates how a patient whose control had always been poor achieved excellent bloodglucose values within a week and maintained them for many months. Control in the other 5 patients with severe complications was also greatly improved, although it is too soon to say whether complications will We
were
regress. Strict diabetic control in pregnancy may be difficult to achieve by standard methods of monitoring, especially if the renal threshold for glucose is low. Frequent and often protracted clinic visits are necessary, and many centres advocate admission during the last trimester.5 However, in one study, strict control in pregnant diabetics could be maintained on an outpatient basis up to the 39th week although patients had to be admitted for 24-hour profiles." Self-monitoring, as reported by West and Lowy,t8 is more convenient and we have confirmed its value for the outpatient management of pregnant diabetics. We did
not encounter
any serious difficulties with the
Reflomat. Diabetic patients are accustomed to needles and the need for repeated finger pricks has not been a
major difficulty. In many patients a representative profile could have been obtained with fewer blood-samples and hence less discomfort. The meter we used has drawbacks: it is mains operated and hence limits the patient’s mobility (a smaller battery-operated model would be an advantage); the range of blood-glucose measurement (3-20 mmol/1) is limited-the upper limit is of little practical significance, but the lower limit means that asymptomatic hypoglycxmia cannot be detected nor symptomatic hypoglycaemia confirmed; the machine is expensive to buy and to run-on the other hand, each machine can be used by many patients and its use is cheap when compared with hospital admission for stabilisation. Sonksen
et
al.19
using
a
different reflectance meter,
have reached similar conclusions about the value of self-
monitoring of blood-glucose. Like them, we believe that this approach has advantages over and is a useful supplement
to
other methods used
diabetic control.
to assess
and monitor
7. 8.
Drash, A. J. Pediat. 1976. 88, 1074. Siperstein, M. D., Foster, D. W., Knowles, H. C., Levine, R., Madison, L. L., Roth, J. New Engl. J. Med. 1977, 196, 1060. 9. Sommer, R., Herbinger, W. Dt. med. Wschr. 1975, 100, 1967. 10. Jarret, R. J., Keane, H., Hardwick, C. Diabetes, 1970, 19, 724. 11. Brunton, W. A. T., Steel, J. M., Percy-Robb, I. W. Clin. chim. Acta. 1977, 75, 359. 12. Knowles, H. C. Jr., in Diabetes Mellitus: Theory and Practice. (edited by M. Ellenberg and H. Rifkin); p. 666. New York, 1970. 13. Malone, J. I., Rosenbloom, A. L., Grgic, A., Weber, F. T. Am. J. Dis. Child. 14.
1976, 130, 1324. Koenig, R. J., Peterson,
C. M., Kilo, C., Cerami, A., Williamson, J. R. Diabetes, 1976, 25, 230. 15. Koenig, R. J., Peterson, C. M., Jones, R. L., Saudek, C., Lehrman, M., Cerami, A., New Engl. J. Med. 1976, 295, 417. 16. Gonen, B., Rochman, H., Rubenstein, A. H., Tanega, S. P., Horwitz, D. L. Lancet, 1977, ii, 734. 17. Lewis, S. B., Murray, W. K., Wallin, J. D., Coustan, D. R., Daane, T. A., Tredway, D. R., Navins, J. P. Obstet. Gynec. 1976, 48, 260. 18. West, T. E. T., Lowy, C. Br. med. J. 1977, i, 1252. 19. Sönksen, P. H., Judd, S. L., Lowy, C. Lancet, 1978, i, 729.
LONG-TERM TREATMENT OF PARKINSONISM WITH BROMOCRIPTINE DONALD B. CALNE ADRIAN C. WILLIAMS ANDREAS NEOPHYTIDES
CHARLES PLOTKIN JOHN G. NUTT PAUL F. TEYCHENNE*
Experimental Therapeutics Branch, Intramural Research Program, National Institute of Neurological and Communicative Disorders and Stroke, National Institutes of Health, Bethesda, Maryland 20014, U.S.A. 92 patients with parkinsonism have been treated with bromocriptine for up to 30 months. 48 continue to receive bromocriptine with benefit; of these, 35 take bromocriptine (mean dose 53 mg daily) in combination with levodopa and 13 take bromocriptine (mean dose 45 mg daily) without levodopa. In those who were originally on levodopa, addition of bromocriptine allowed a mean 41% reduction in the dose of levodopa; the largest group of patients to benefit from bromocriptine entered the study because of excessive dyskinesia or "on-off" phenomena induced by levodopa. In 40 patients bromocriptine was stopped because of adverse reactions, absence of therapeutic response, or non-compliance with the protocol. The main problems were psychiatric disturbance (8 patients) and erythromelalgia (7 patients); these effects tended to occur late (mean 6 months and 10 months, respectively) and with high dosage (mean 66 mg and 115 mg daily). Other fre-
Summary
*Present address:
Washington, D.C.
George Washington University
Medical Center,
736 quent adverse effects were dizziness and nausea; these began considerably earlier (at 2 months and 1 month) and with much lower dosage (31 mg and 12 mg daily). 4 patients died, for reasons apparently unrelated to
therapy. Introduction BROMOCRIPTINE is
an ergot derivative with dopaminergic properties.’ Since 1972 it has been employed therapeutically to reduce serum-prolactin, for which the dose does not usually exceed 10 mg daily.2.1’ In acromegaly up to 60 mg daily has been given to suppress growth-hormone secretion. To alleviate motor deficits in parkinsonism doses as high as 150 mg daily have been used but only in small groups of patients treated for short periods;4-swe report here our experience with 92 parkinsonian patients treated for up to 30 months with high doses of bromocriptine.
comitantly. The patients receiving bromocriptine without levodopa were made up of newly diagnosed cases and those who had either experienced major adverse reactions to levodopa or had not responded to this drug. Because of the inclusion of newly diagnosed (mildly affected) subjects, the neurological deficits in the group of patients receiving brpmocriptine alone were less than those in our total parkinsonian population. This probably accounts for the lower mean dose of bromocriptine in this group. The 4 patients who died (see below) had severe parkinsonism and were receiving a high mean dose of bromocriptine. The indications for experimental bromocriptine therapy are shown in table ii together with a summary TABLE II-RELATION BETWEEN MAIN REASON FOR ENTERING STUDY AND CONTINUANCE ON BROMOCRIPTINE
Patients and Methods The patients had been referred to the National Institutes of Health for experimental therapy. Any with serious hepatic, renal, or psychiatric disease were excluded, but cardiovascular disorders, if stable, were not regarded as an absolute contraindication to bromocriptine therapy. As a group, our patients differed from a random selection of the parkinsonian population in having a higher prevalence of problems with levodopa therapy. 90 patients had idiopathic parkinsonism, 1 had postencephalitic parkinsonism, and 1 had calcification of the basal
ganglia. Patients were started on a 1 mg test dose of bromocriptine because higher initial doses occasionally cause serious hypotensive reactions; intake was increased to optimal levels (usually 50-90 mg daily) over 4 weeks. The dose of levodopa was reduced as necessary over this build-up period, to control adverse reactions. Patients who were receiving anticholinergic agents and amantadine continued on these drugs throughout the study, without any changes. In the clinic each month, neurological status was reviewed and evidence of drug toxicity was sought. Blood-pressure and pulse rates were recorded with the patient lying and standing, and blood-samples were taken for estimation of haemoglobin, red-cell count, red-cell indices, total and differential white-cell count, and serum protein, electrolytes, urea-nitrogen, creatinine, uric acid, glutamic-oxaloacetic and glutamic-pyruvic transaminases (s.G.o.T., S.G.P.T.), alkaline phosphatase, and bilirubin. The urine was examined at each attendance. Patients with angina, old myocardial infarcts, or hypertension were admitted to hospital for careful cardiovascular monitoring while starting bromocriptine. Patients were kept on treatment as long as the benefits of bromocriptine outweighed any ill-effects.
Results
experience with the 92 patients is summarised in iv. 52% of the patients continue to take bromotables and most of these are receiving levodopa concriptine Our
i
-
TABLE I-MEAN DOSES AND DURATIONS OF TREATMENT
BROMOCRIPTINE
*Patients who simply wanted to try any new treatment.
of how many patients benefited. The most useful results with bromocriptine were achieved in patients who were troubled by prominent dyskinesia or "on-off" phenomena. A satisfactory response was also obtained in both the patient with postencephalitic disease and the patient with calcified basal ganglia. Of the patients who had a
previous levodopa, a minority improved substantially when bromocriptine was added, although many remain on combined therapy. Of the 35 patients receiving bromocriptine with levodopa, all had previously been taking higher doses of levodopa. In these patients the dosage of levodopa was reduced by an average of 41% when bromocriptine was added. This figure derives from the group of 35 patients who continue on a combination of bromocriptine and levodopa; when a patient was taking carbidopa this was allowed for by multiplying the levodopa dose by four. About the same reduction in levodopa was achieved in the 40 patients who stopped bromocriptine as in the patients who continued on it. 29 patients were taken off bromocriptine because of adverse reactions. In another 8 patients, bromocriptine was stopped because there was no net benefit. A further 3 patients did not comply with the protocol.
poor response
to
The adverse reactions which most often led to withdrawal of bromocriptine are summarised in table in. Table iv shows that unwanted effects tended to fall into two categories-those arising early in the treatment, before a high dose of bromocriptine could be built up (nausea and dizziness), and those arising after a longer period of treatment, at higher dosage (psychiatric reactions and erythomelalgia). Psychiatric reactions to bromocriptine were similar to those induced by levodopa-namely, confusion, frequent vivid dreams, hallucinations, and delusions. The hallucinations were usually visual (occasionally auditory or olfactory) and the delusions generally paranoid. These psychotic features were usually more florid and persisted for longer
737 TABLE III-ADVERSE REACTIONS LEADING TO WITHDRAWAL OF BROMOCRIPTINE
onary disease, he had entered the study with an old myocardial infarct and had a second during treatment with
bromocriptine. The only patient on whom an necropsy performed died from mesenteric thrombosis; he had extensive atheroma in his large arteries. was
TABLE IV-DOSES AND DURATION OF TREATMENT AT TIME OF
WITHDRAWAL FOR MAIN ADVERSE REACTIONS
Discussion While most of the observations reported here relate to adverse reactions, we wish to stress that the 48 (out of 92) patients who continue to take bromocriptine are doing so because this drug, usually in combination with levodopa,seems the best means of controlling their neurological symptoms. The patients most likely to benefit from bromocriptine are those who have prominent dyskinesia or "dn-off" reactions. It is notable that we have not, so far, observed the development of "on-off" phenomena as a new problem in any patients receiving bro-
mocriptine.
than those associated with levodopa; they eventually cleared in all cases, though in 2 patients they continued for six weeks after withdrawal of bromocriptine. When erythromelagia developed (red, tender, warm, oedematous lower extremities), withdrawal of treatment led to recovery within a week. In 1 patient the erythromelalgia was accompanied by transient livedo reticularis, polyarthralgia, nausea, abdominal pain, and headache. In 3 out of 8 patients with erythromelalgia the therapeutic response to bromocriptine had been so beneficial that the drug was restarted at lower dosage (70 mg daily); the response was satisfactory and symptoms did not recur.
Several patients had blurring of vision and diplopia, and 1 had to stop taking bromocriptine for this reason. Full ophthalmological examination of these patients revealed no consistent pattern of drug-induced external ocular paresis or other abnormality. Nasal congestion, another common symptom, forced withdrawal of bromocriptine in 1 patient. In addition to exacerbation of angina, dyspnoea was occasionally seen, but we have been unable to identify any pulmonary or cardiac dysfunction. Similar symptoms have been seen with levo-
dopa. Symptomless increases in serum transaminase levels (up to a fourfold rise) occurred in 6 patients without concomitant changes in bilirubin or alkaline phosphatase. In most patients the enzyme changes were transient, but in 1 patient the dose of bromocriptine had to be reduced, and in another the drug had to be stopped. Other laboratory tests revealed no significant abnormality. Patients sometimes become hypotensive on bromocriptine, particularly when starting treatment; on occasion
have encountered an acute and severe fall in after a single dose of 2.5mg.Another exceptional problem is induction of a photosensitive erythematous macular skin eruption. 4 patients died during the study. 2 died in their sleep, both having been severely dysphagic in the last few weeks of their lives; aspiration was suspected but not proved. 1 patient died after open-heart surgery for cor-
The adverse reactions to bromocriptine commonly encountered early in treatment (nausea, dizziness, hypotension) are usually transient, and treatment can generally be continued after temporary reduction of dosage. Erythomelalgiaand psychiatric reactions tended to occur later and at higher doses. Erythromelalgia is less likely to arise if the dose of bromocriptine is kept below 100 mg daily. The risk of psychiatric disturbances can be reduced by avoidance of bromocriptine in patients with a history of major psycbiatric disease; when psychotic features occur we have found it necessary to stop
bromocriptine. The increased transaminase levels in 6 patients are an indication that liver-function tests should be monitored in patients receiving long-term bromocriptine. It is noteworthy that another ergot derivative, lergotrile, induces disturbances in liver function which are much more frequent and severe.9 In addition to psychiatric and hepatic disease, we regard severe angina and myocardial infarction as contraindications to bromocriptine therapy. Others have suggested that the drug should be avoided in patients with peripheral vascular disease10 or peptic ulcer.11 An increased incidence of uterine tumours has been reported in rats treated with high doses of bromocriptine for two years.12 This observation is hard to interpret in terms of human risk for three reasons: these tumours result from endocrinological effects, and hormonal changes in the elderly female rat are different from those in women; two years represents most of a rat’s lifespan; and uterine examination in 88 women treated with bromocriptine for up to six years (for endocrinological disorders) has revealed no abnormality. 13 At present we recommend annual gynaecological examination for all women
receiving long-term bromocriptine therapy.
We thank Ms P. Barnicoat and Ms B. Shryock for administrative support, Mrs M. Gillespie and Ms H. Krebs for their help as nurses, and Ms A. Miller for secretarial assistance. Bromocriptine was supplied by Sandoz Ltd.
we
blood-pressure
Requests for reprints should be addressed to D. B. C., Building 10, 6D20, National Institutes of Health, Bethesda, Maryland 20014,
room
U.S.A. REFERENCES
1. Corrodi, H., Fuxe, K., Höfelt, Pharmac. 1973, 25, 409.
T., Lidbrink, P., Understedt, U. J. Pharm.
738 TABLE I—CHARACTERISTICS OF THE PATIENTS
DOUBLE-BLIND CONTROLLED TRIAL OF ELECTROCONVULSIVE THERAPY (E.C.T.) AND SIMULATED E.C.T. IN DEPRESSIVE ILLNESS C. P. L. FREEMAN A. CRIGHTON
J. V. BASSON
Department of Psychiatry, Edinburgh University, and Royal Edinburgh Hospital
patients prescribed electroconvulsive therapy (E.C.T.) for treatment of a depressive illness were randomly allocated to two
Summary
40
groups. One group had the first two
E.C.T. treatments
simulated E.C.T. on a double-blind basis. The results show that E.C.T. is significantly superior to simulated E.C.T. in the treatment of depressive illness.
replaced by
Introduction THE object of the trial ness, in the treatment of
was to
compare the effective-
depressive illness, of bilateral electroconvulsive therapy (E.C.T.) with a treatment which simulated E.C.T. as closely as possible. E.C.T. is an effective treatment for depressive illness 1-3 and has tended to become the standard against which novel treatments for depression are assessed. But which parts of the treatment are therapeutically active is not clear. Critics of E.C.T. say that it is a crude and unscientific therapy which if it works does so through fear, punishment, or non-specific factors such as increased nursing and medical attention. The evidence at present available does not support this view and points towards the seizure as being the important therapeutic agent.4 The ideal design for such a trial would have been to have compared a full course of simulated E.C.T. with a full course of real E.C.T. We wanted to conduct the trial on severely and very severely depressed patients for whom E.c.T. is primarily indicated, and we wanted to avoid the criticism that the trial showed E.C.T. to be ineffective because the patients were not those who would usually receive it. We felt it ethically unjustified to withhold for a complete course a treatment generally regarded to be effective and to submit patients to perhaps unnecessary general anaesthesia. The method presented here was therefore a compromise. Method
2.
patients
took part in the trial. The details
i.
The criteria for admission
to
the trial
were as
follows.
Inpatients of either sex, aged 20-70 years. A clinical diagnosis of depressive illness defined by the primary manifestation and major symptom of persistent change of mood exceeding customary sadness, accompanied by one or more of the following symptoms: self-deprecation with a morbid sense of guilt; sleep disturbance; hypochondriasis; retardation of thought and action; agitated behaviour. The depression had to be the primary illness and not secondary to other mental illness such as schizophrenia. A minimum score of 15 on both the Hamilton’ and Beckrating-scales for depression. No E.C.T. in the past three months. Absence of a major or progressive physical illness and no symptoms or signs of organic cerebral disease. The patient had to give informed consent to take part in the trial. Treatment units for whom E.c.T. was preconsidered. Patients entering the trial were randomly allocated to one of two groups. The group given real E.c.T. (group R) received bilateral E.C.T. twice weekly from an ’Ectron’ Mk IV machine. This delivered a bidirectional 60% sine-wave current at 400 V peak to peak for 1.5 s. The group receiving simulated E.C.T. (group s) were given two treatments during which the electrodes were applied to the head but no current was passed. The third and subsequent treatments for this group were normal bilateral E.C.T. In effect, E.C.T. was delayed for one week in the control group and replaced by fake All
patients from four acute
scribed
were
E.C.T.
Patients 40
table
are
shown in
Varga, L., Lutterbeck, P. M., Pryor, J. S., Wenner, R., Erb, H. Br. med. J. 1972, ii, 743. 3. Besser, G. M., Parke, L., Edwards, C. R. W., Forsyth, I. A., McNeilly, A. S. Br. med. J. 1972, iii, 669. 4. Kartzinel, R., Perlow, M., Teychenne, P., Gielen, A. C., Gillespie, M. M., Sadowsky, D. A., Calne, D. B. Lancet, 1976, ii, 272. 5. Lieberman, A., Kupersmith, M., Estey, E., Goldstein, M. New Engl. J. Med. 1976, 295, 1400. 6. Parkes, J. D., Debono, A. G., Marsden, C. D. J. Neurol. Neurosurg. Psychiat. 1976, 39, 1101. 7. Godwin-Austen, R. B., Smith, N. H. ibid. 1977, 40, 479. 8. Martinez-Lage, J. M., Delgado, G., Zubieta, J. L., Viteri, C. Excerpta med. 1977, 427, 23. 9. Teychenne, P. F., Pfeiffer, R. F., Bern, S. M., McInturff, D., Calne, D. B. Neurology (in the press). 10. Wass, J. A. H., Thorner, M. O., Besser, G. M. Lancet, 1976, i, 1135. 11. Wass, J. A. H., et al. Lancet, 1976, ii, 851. 12. Griffith, R. W., Br. med. J. 1977, ii, 1605. 13. Besser, G. M., et al. Br. med. J. 1977, ii, 868.
The anaesthetic procedure was identical for both groups. Premedication was with atropine 600 mg and induction with sodium thiopentone (150-300 mg). Muscle relaxation was with suxamethonium chloride. E.C.T. was always given by the same doctor. This doctor (C.P.L.F.), the E.C.T. nurse, and the anaesthetist were the only people not blind to the patient’s group.
Statistical Analysis All significant values are based on Student’s t tests. Independent t tests were used for between-group sampling and dependent t tests within groups.
Results Overall Outcome
Table n shows the reasons for terminating E.C.T. in the two groups. The 4 hypomanic patients settled without further treatment and were judged to have made a
because a visit to the clinic is a period of inactivity and a break with normal routine. The admission to hospital of patients with unstable control for blood-glucose profiles is wasteful of beds and artificial since life in hospital is unlike normal life. Selfmonitoring overcomes these problems and saves the patient unnecessary hospital visits since changes in treatment can be made by telephone. Home blood-glucose monitoring can also be used to assess long-term control. Its advantage over measurement of HbAlya-’6 is that it not only detects poor control, but also shows how to correct it.
We thank the special trustees of Nottingham University Hospital for grant to R. B. T.; and our secretaries, Mrs Jane Richards and Mrs Judith Halford, for their help. a
Requests for reprints should be sent to R. B. T. REFERENCES 1. Engerman, R., Bloodworth, J. M. B. Jr., Nelson, S. Diabetes, 1977, 26, 760. 2. Mauer, S. M., Steffes, M. W., Sutherland, D. E. R., Najarian, J. S., Michael, A. F., Brown, B. M. ibid. 1975, 24, 280. 3. Fox, C. J., Darby, S. C., Ireland, J. T., Sönksen, P. H. Br. med. J. 1977,
ii, 605. Karson, K., Kjellmer, I. Am. J. Obstet. Gynec. 1972, 112, 213. 5. Essex, N., Pyke, D. A., Watkins, P. J., Brudenell, J. M., Gamsu, H. R. Br. med. J. 1973, iv, 89. 6. Cahill, G. F., Etzwiler, D. D., Freinkel, N. New Engl. J. Med. 1976, 296, 4.
1004.
impressed that blood-glucose could be maintained at all times near or within the physiological range without dangerous hypoglycaemia even in relatively long-standing diabetics in whom endogenous insulin secretion is presumably lacking. 17 patients were included in our study because they were thought to be well controlled on clinical criteria, and of these 14 had initial blood-glucose profiles where no value exceeded 10 mmola. 18 more patients were brought within this level of control by adjustment of their treatment with the aid of a Reflomat. Case 1 illustrates how a patient whose control had always been poor achieved excellent bloodglucose values within a week and maintained them for many months. Control in the other 5 patients with severe complications was also greatly improved, although it is too soon to say whether complications will We
were
regress. Strict diabetic control in pregnancy may be difficult to achieve by standard methods of monitoring, especially if the renal threshold for glucose is low. Frequent and often protracted clinic visits are necessary, and many centres advocate admission during the last trimester.5 However, in one study, strict control in pregnant diabetics could be maintained on an outpatient basis up to the 39th week although patients had to be admitted for 24-hour profiles." Self-monitoring, as reported by West and Lowy,t8 is more convenient and we have confirmed its value for the outpatient management of pregnant diabetics. We did
not encounter
any serious difficulties with the
Reflomat. Diabetic patients are accustomed to needles and the need for repeated finger pricks has not been a
major difficulty. In many patients a representative profile could have been obtained with fewer blood-samples and hence less discomfort. The meter we used has drawbacks: it is mains operated and hence limits the patient’s mobility (a smaller battery-operated model would be an advantage); the range of blood-glucose measurement (3-20 mmol/1) is limited-the upper limit is of little practical significance, but the lower limit means that asymptomatic hypoglycxmia cannot be detected nor symptomatic hypoglycaemia confirmed; the machine is expensive to buy and to run-on the other hand, each machine can be used by many patients and its use is cheap when compared with hospital admission for stabilisation. Sonksen
et
al.19
using
a
different reflectance meter,
have reached similar conclusions about the value of self-
monitoring of blood-glucose. Like them, we believe that this approach has advantages over and is a useful supplement
to
other methods used
diabetic control.
to assess
and monitor
7. 8.
Drash, A. J. Pediat. 1976. 88, 1074. Siperstein, M. D., Foster, D. W., Knowles, H. C., Levine, R., Madison, L. L., Roth, J. New Engl. J. Med. 1977, 196, 1060. 9. Sommer, R., Herbinger, W. Dt. med. Wschr. 1975, 100, 1967. 10. Jarret, R. J., Keane, H., Hardwick, C. Diabetes, 1970, 19, 724. 11. Brunton, W. A. T., Steel, J. M., Percy-Robb, I. W. Clin. chim. Acta. 1977, 75, 359. 12. Knowles, H. C. Jr., in Diabetes Mellitus: Theory and Practice. (edited by M. Ellenberg and H. Rifkin); p. 666. New York, 1970. 13. Malone, J. I., Rosenbloom, A. L., Grgic, A., Weber, F. T. Am. J. Dis. Child. 14.
1976, 130, 1324. Koenig, R. J., Peterson,
C. M., Kilo, C., Cerami, A., Williamson, J. R. Diabetes, 1976, 25, 230. 15. Koenig, R. J., Peterson, C. M., Jones, R. L., Saudek, C., Lehrman, M., Cerami, A., New Engl. J. Med. 1976, 295, 417. 16. Gonen, B., Rochman, H., Rubenstein, A. H., Tanega, S. P., Horwitz, D. L. Lancet, 1977, ii, 734. 17. Lewis, S. B., Murray, W. K., Wallin, J. D., Coustan, D. R., Daane, T. A., Tredway, D. R., Navins, J. P. Obstet. Gynec. 1976, 48, 260. 18. West, T. E. T., Lowy, C. Br. med. J. 1977, i, 1252. 19. Sönksen, P. H., Judd, S. L., Lowy, C. Lancet, 1978, i, 729.
LONG-TERM TREATMENT OF PARKINSONISM WITH BROMOCRIPTINE DONALD B. CALNE ADRIAN C. WILLIAMS ANDREAS NEOPHYTIDES
CHARLES PLOTKIN JOHN G. NUTT PAUL F. TEYCHENNE*
Experimental Therapeutics Branch, Intramural Research Program, National Institute of Neurological and Communicative Disorders and Stroke, National Institutes of Health, Bethesda, Maryland 20014, U.S.A. 92 patients with parkinsonism have been treated with bromocriptine for up to 30 months. 48 continue to receive bromocriptine with benefit; of these, 35 take bromocriptine (mean dose 53 mg daily) in combination with levodopa and 13 take bromocriptine (mean dose 45 mg daily) without levodopa. In those who were originally on levodopa, addition of bromocriptine allowed a mean 41% reduction in the dose of levodopa; the largest group of patients to benefit from bromocriptine entered the study because of excessive dyskinesia or "on-off" phenomena induced by levodopa. In 40 patients bromocriptine was stopped because of adverse reactions, absence of therapeutic response, or non-compliance with the protocol. The main problems were psychiatric disturbance (8 patients) and erythromelalgia (7 patients); these effects tended to occur late (mean 6 months and 10 months, respectively) and with high dosage (mean 66 mg and 115 mg daily). Other fre-
Summary
*Present address:
Washington, D.C.
George Washington University
Medical Center,
736 quent adverse effects were dizziness and nausea; these began considerably earlier (at 2 months and 1 month) and with much lower dosage (31 mg and 12 mg daily). 4 patients died, for reasons apparently unrelated to
therapy. Introduction BROMOCRIPTINE is
an ergot derivative with dopaminergic properties.’ Since 1972 it has been employed therapeutically to reduce serum-prolactin, for which the dose does not usually exceed 10 mg daily.2.1’ In acromegaly up to 60 mg daily has been given to suppress growth-hormone secretion. To alleviate motor deficits in parkinsonism doses as high as 150 mg daily have been used but only in small groups of patients treated for short periods;4-swe report here our experience with 92 parkinsonian patients treated for up to 30 months with high doses of bromocriptine.
comitantly. The patients receiving bromocriptine without levodopa were made up of newly diagnosed cases and those who had either experienced major adverse reactions to levodopa or had not responded to this drug. Because of the inclusion of newly diagnosed (mildly affected) subjects, the neurological deficits in the group of patients receiving brpmocriptine alone were less than those in our total parkinsonian population. This probably accounts for the lower mean dose of bromocriptine in this group. The 4 patients who died (see below) had severe parkinsonism and were receiving a high mean dose of bromocriptine. The indications for experimental bromocriptine therapy are shown in table ii together with a summary TABLE II-RELATION BETWEEN MAIN REASON FOR ENTERING STUDY AND CONTINUANCE ON BROMOCRIPTINE
Patients and Methods The patients had been referred to the National Institutes of Health for experimental therapy. Any with serious hepatic, renal, or psychiatric disease were excluded, but cardiovascular disorders, if stable, were not regarded as an absolute contraindication to bromocriptine therapy. As a group, our patients differed from a random selection of the parkinsonian population in having a higher prevalence of problems with levodopa therapy. 90 patients had idiopathic parkinsonism, 1 had postencephalitic parkinsonism, and 1 had calcification of the basal
ganglia. Patients were started on a 1 mg test dose of bromocriptine because higher initial doses occasionally cause serious hypotensive reactions; intake was increased to optimal levels (usually 50-90 mg daily) over 4 weeks. The dose of levodopa was reduced as necessary over this build-up period, to control adverse reactions. Patients who were receiving anticholinergic agents and amantadine continued on these drugs throughout the study, without any changes. In the clinic each month, neurological status was reviewed and evidence of drug toxicity was sought. Blood-pressure and pulse rates were recorded with the patient lying and standing, and blood-samples were taken for estimation of haemoglobin, red-cell count, red-cell indices, total and differential white-cell count, and serum protein, electrolytes, urea-nitrogen, creatinine, uric acid, glutamic-oxaloacetic and glutamic-pyruvic transaminases (s.G.o.T., S.G.P.T.), alkaline phosphatase, and bilirubin. The urine was examined at each attendance. Patients with angina, old myocardial infarcts, or hypertension were admitted to hospital for careful cardiovascular monitoring while starting bromocriptine. Patients were kept on treatment as long as the benefits of bromocriptine outweighed any ill-effects.
Results
experience with the 92 patients is summarised in iv. 52% of the patients continue to take bromotables and most of these are receiving levodopa concriptine Our
i
-
TABLE I-MEAN DOSES AND DURATIONS OF TREATMENT
BROMOCRIPTINE
*Patients who simply wanted to try any new treatment.
of how many patients benefited. The most useful results with bromocriptine were achieved in patients who were troubled by prominent dyskinesia or "on-off" phenomena. A satisfactory response was also obtained in both the patient with postencephalitic disease and the patient with calcified basal ganglia. Of the patients who had a
previous levodopa, a minority improved substantially when bromocriptine was added, although many remain on combined therapy. Of the 35 patients receiving bromocriptine with levodopa, all had previously been taking higher doses of levodopa. In these patients the dosage of levodopa was reduced by an average of 41% when bromocriptine was added. This figure derives from the group of 35 patients who continue on a combination of bromocriptine and levodopa; when a patient was taking carbidopa this was allowed for by multiplying the levodopa dose by four. About the same reduction in levodopa was achieved in the 40 patients who stopped bromocriptine as in the patients who continued on it. 29 patients were taken off bromocriptine because of adverse reactions. In another 8 patients, bromocriptine was stopped because there was no net benefit. A further 3 patients did not comply with the protocol.
poor response
to
The adverse reactions which most often led to withdrawal of bromocriptine are summarised in table in. Table iv shows that unwanted effects tended to fall into two categories-those arising early in the treatment, before a high dose of bromocriptine could be built up (nausea and dizziness), and those arising after a longer period of treatment, at higher dosage (psychiatric reactions and erythomelalgia). Psychiatric reactions to bromocriptine were similar to those induced by levodopa-namely, confusion, frequent vivid dreams, hallucinations, and delusions. The hallucinations were usually visual (occasionally auditory or olfactory) and the delusions generally paranoid. These psychotic features were usually more florid and persisted for longer
737 TABLE III-ADVERSE REACTIONS LEADING TO WITHDRAWAL OF BROMOCRIPTINE
onary disease, he had entered the study with an old myocardial infarct and had a second during treatment with
bromocriptine. The only patient on whom an necropsy performed died from mesenteric thrombosis; he had extensive atheroma in his large arteries. was
TABLE IV-DOSES AND DURATION OF TREATMENT AT TIME OF
WITHDRAWAL FOR MAIN ADVERSE REACTIONS
Discussion While most of the observations reported here relate to adverse reactions, we wish to stress that the 48 (out of 92) patients who continue to take bromocriptine are doing so because this drug, usually in combination with levodopa,seems the best means of controlling their neurological symptoms. The patients most likely to benefit from bromocriptine are those who have prominent dyskinesia or "dn-off" reactions. It is notable that we have not, so far, observed the development of "on-off" phenomena as a new problem in any patients receiving bro-
mocriptine.
than those associated with levodopa; they eventually cleared in all cases, though in 2 patients they continued for six weeks after withdrawal of bromocriptine. When erythromelagia developed (red, tender, warm, oedematous lower extremities), withdrawal of treatment led to recovery within a week. In 1 patient the erythromelalgia was accompanied by transient livedo reticularis, polyarthralgia, nausea, abdominal pain, and headache. In 3 out of 8 patients with erythromelalgia the therapeutic response to bromocriptine had been so beneficial that the drug was restarted at lower dosage (70 mg daily); the response was satisfactory and symptoms did not recur.
Several patients had blurring of vision and diplopia, and 1 had to stop taking bromocriptine for this reason. Full ophthalmological examination of these patients revealed no consistent pattern of drug-induced external ocular paresis or other abnormality. Nasal congestion, another common symptom, forced withdrawal of bromocriptine in 1 patient. In addition to exacerbation of angina, dyspnoea was occasionally seen, but we have been unable to identify any pulmonary or cardiac dysfunction. Similar symptoms have been seen with levo-
dopa. Symptomless increases in serum transaminase levels (up to a fourfold rise) occurred in 6 patients without concomitant changes in bilirubin or alkaline phosphatase. In most patients the enzyme changes were transient, but in 1 patient the dose of bromocriptine had to be reduced, and in another the drug had to be stopped. Other laboratory tests revealed no significant abnormality. Patients sometimes become hypotensive on bromocriptine, particularly when starting treatment; on occasion
have encountered an acute and severe fall in after a single dose of 2.5mg.Another exceptional problem is induction of a photosensitive erythematous macular skin eruption. 4 patients died during the study. 2 died in their sleep, both having been severely dysphagic in the last few weeks of their lives; aspiration was suspected but not proved. 1 patient died after open-heart surgery for cor-
The adverse reactions to bromocriptine commonly encountered early in treatment (nausea, dizziness, hypotension) are usually transient, and treatment can generally be continued after temporary reduction of dosage. Erythomelalgiaand psychiatric reactions tended to occur later and at higher doses. Erythromelalgia is less likely to arise if the dose of bromocriptine is kept below 100 mg daily. The risk of psychiatric disturbances can be reduced by avoidance of bromocriptine in patients with a history of major psycbiatric disease; when psychotic features occur we have found it necessary to stop
bromocriptine. The increased transaminase levels in 6 patients are an indication that liver-function tests should be monitored in patients receiving long-term bromocriptine. It is noteworthy that another ergot derivative, lergotrile, induces disturbances in liver function which are much more frequent and severe.9 In addition to psychiatric and hepatic disease, we regard severe angina and myocardial infarction as contraindications to bromocriptine therapy. Others have suggested that the drug should be avoided in patients with peripheral vascular disease10 or peptic ulcer.11 An increased incidence of uterine tumours has been reported in rats treated with high doses of bromocriptine for two years.12 This observation is hard to interpret in terms of human risk for three reasons: these tumours result from endocrinological effects, and hormonal changes in the elderly female rat are different from those in women; two years represents most of a rat’s lifespan; and uterine examination in 88 women treated with bromocriptine for up to six years (for endocrinological disorders) has revealed no abnormality. 13 At present we recommend annual gynaecological examination for all women
receiving long-term bromocriptine therapy.
We thank Ms P. Barnicoat and Ms B. Shryock for administrative support, Mrs M. Gillespie and Ms H. Krebs for their help as nurses, and Ms A. Miller for secretarial assistance. Bromocriptine was supplied by Sandoz Ltd.
we
blood-pressure
Requests for reprints should be addressed to D. B. C., Building 10, 6D20, National Institutes of Health, Bethesda, Maryland 20014,
room
U.S.A. REFERENCES
1. Corrodi, H., Fuxe, K., Höfelt, Pharmac. 1973, 25, 409.
T., Lidbrink, P., Understedt, U. J. Pharm.
738 TABLE I—CHARACTERISTICS OF THE PATIENTS
DOUBLE-BLIND CONTROLLED TRIAL OF ELECTROCONVULSIVE THERAPY (E.C.T.) AND SIMULATED E.C.T. IN DEPRESSIVE ILLNESS C. P. L. FREEMAN A. CRIGHTON
J. V. BASSON
Department of Psychiatry, Edinburgh University, and Royal Edinburgh Hospital
patients prescribed electroconvulsive therapy (E.C.T.) for treatment of a depressive illness were randomly allocated to two
Summary
40
groups. One group had the first two
E.C.T. treatments
simulated E.C.T. on a double-blind basis. The results show that E.C.T. is significantly superior to simulated E.C.T. in the treatment of depressive illness.
replaced by
Introduction THE object of the trial ness, in the treatment of
was to
compare the effective-
depressive illness, of bilateral electroconvulsive therapy (E.C.T.) with a treatment which simulated E.C.T. as closely as possible. E.C.T. is an effective treatment for depressive illness 1-3 and has tended to become the standard against which novel treatments for depression are assessed. But which parts of the treatment are therapeutically active is not clear. Critics of E.C.T. say that it is a crude and unscientific therapy which if it works does so through fear, punishment, or non-specific factors such as increased nursing and medical attention. The evidence at present available does not support this view and points towards the seizure as being the important therapeutic agent.4 The ideal design for such a trial would have been to have compared a full course of simulated E.C.T. with a full course of real E.C.T. We wanted to conduct the trial on severely and very severely depressed patients for whom E.c.T. is primarily indicated, and we wanted to avoid the criticism that the trial showed E.C.T. to be ineffective because the patients were not those who would usually receive it. We felt it ethically unjustified to withhold for a complete course a treatment generally regarded to be effective and to submit patients to perhaps unnecessary general anaesthesia. The method presented here was therefore a compromise. Method
2.
patients
took part in the trial. The details
i.
The criteria for admission
to
the trial
were as
follows.
Inpatients of either sex, aged 20-70 years. A clinical diagnosis of depressive illness defined by the primary manifestation and major symptom of persistent change of mood exceeding customary sadness, accompanied by one or more of the following symptoms: self-deprecation with a morbid sense of guilt; sleep disturbance; hypochondriasis; retardation of thought and action; agitated behaviour. The depression had to be the primary illness and not secondary to other mental illness such as schizophrenia. A minimum score of 15 on both the Hamilton’ and Beckrating-scales for depression. No E.C.T. in the past three months. Absence of a major or progressive physical illness and no symptoms or signs of organic cerebral disease. The patient had to give informed consent to take part in the trial. Treatment units for whom E.c.T. was preconsidered. Patients entering the trial were randomly allocated to one of two groups. The group given real E.c.T. (group R) received bilateral E.C.T. twice weekly from an ’Ectron’ Mk IV machine. This delivered a bidirectional 60% sine-wave current at 400 V peak to peak for 1.5 s. The group receiving simulated E.C.T. (group s) were given two treatments during which the electrodes were applied to the head but no current was passed. The third and subsequent treatments for this group were normal bilateral E.C.T. In effect, E.C.T. was delayed for one week in the control group and replaced by fake All
patients from four acute
scribed
were
E.C.T.
Patients 40
table
are
shown in
Varga, L., Lutterbeck, P. M., Pryor, J. S., Wenner, R., Erb, H. Br. med. J. 1972, ii, 743. 3. Besser, G. M., Parke, L., Edwards, C. R. W., Forsyth, I. A., McNeilly, A. S. Br. med. J. 1972, iii, 669. 4. Kartzinel, R., Perlow, M., Teychenne, P., Gielen, A. C., Gillespie, M. M., Sadowsky, D. A., Calne, D. B. Lancet, 1976, ii, 272. 5. Lieberman, A., Kupersmith, M., Estey, E., Goldstein, M. New Engl. J. Med. 1976, 295, 1400. 6. Parkes, J. D., Debono, A. G., Marsden, C. D. J. Neurol. Neurosurg. Psychiat. 1976, 39, 1101. 7. Godwin-Austen, R. B., Smith, N. H. ibid. 1977, 40, 479. 8. Martinez-Lage, J. M., Delgado, G., Zubieta, J. L., Viteri, C. Excerpta med. 1977, 427, 23. 9. Teychenne, P. F., Pfeiffer, R. F., Bern, S. M., McInturff, D., Calne, D. B. Neurology (in the press). 10. Wass, J. A. H., Thorner, M. O., Besser, G. M. Lancet, 1976, i, 1135. 11. Wass, J. A. H., et al. Lancet, 1976, ii, 851. 12. Griffith, R. W., Br. med. J. 1977, ii, 1605. 13. Besser, G. M., et al. Br. med. J. 1977, ii, 868.
The anaesthetic procedure was identical for both groups. Premedication was with atropine 600 mg and induction with sodium thiopentone (150-300 mg). Muscle relaxation was with suxamethonium chloride. E.C.T. was always given by the same doctor. This doctor (C.P.L.F.), the E.C.T. nurse, and the anaesthetist were the only people not blind to the patient’s group.
Statistical Analysis All significant values are based on Student’s t tests. Independent t tests were used for between-group sampling and dependent t tests within groups.
Results Overall Outcome
Table n shows the reasons for terminating E.C.T. in the two groups. The 4 hypomanic patients settled without further treatment and were judged to have made a