93
RANDOMISED CONTROLLED TRIAL OF YOGA AND BIO-FEEDBACK IN MANAGEMENT OF HYPERTENSION
ratio of practolol to propranolol of 23/1 is in fair agreement with that of 25/1obtained by Bodem et al.46 and of 28/1 obtained in a previous investiga-
tion.9
conclusion, the results of this within-subject study using single intravenous injections suggest that: 1. When the cardiac (3-blocking activity of propranolol, acebutolol, and practolol was nearly equivalent (as judged by reduction in exercise heart-rate) their mean plasma concentrations were in the ratio 1/8/23. 2. Propranolol was not cardioselective (as judged by F.E.V.I, resting and exercise P.F.R. measurements). 3. Acebutolol was not as cardioselective as practolol (as judged by exercise P.F.R. measurements). In
We thank our colleagues at the School of ,Pharmacy, London, who assisted us in this study, May and Baker for supplying acebutolol, Imperial Chemical Industries for supplying propranolol and practolol, and the Board of Governors of St. Bartholomew’s Hospital and the British Heart Foundation for financial support. Requests for reprints should be addressed to J. H., Department of Cardiology, St. Bartholomew’s Hospital, London EC1A 7BE.
CHANDRA PATEL
Croydon, Surrey W. R. S. NORTH M.R.C./D.H.S.S. Epidemiology and Medical Care Unit, Northwick Park Hospital, Harrow, Middlesex
Summary
34
REFERENCES 1. 2. 3. 4.
5. 6. 7. 8.
9. 10. 11. 12. 13. 14.
15. 16. 17. 18. 19.
20. 21. 22.
23. 24.
Gillam, P. M. S., Prichard, B. N. C. Br. med. J. 1965, ii, 337. Aronow, W. S. Am. Heart J. 1972, 84, 834. Wilhelmsson, C., Vedin, J. A., Wilhelmsen, L., Tibblin, G., Werkö, L. Lancet, 1974, ii, 1157. Fox, K. M., Chopra, M. P., Portal, R. W., Aber, C. P. Br. med. J. 1975, i, 117. Jackson, G., Atkinson, L., Oram, S. ibid. p. 708. MacDonald, A. G., McNeill, R. S. Br. J. Anœsth. 1968, 40, 508. Levy, B. J. Pharmac. exp. Ther. 1973, 186, 134. Vaughan Williams, E. M., Bagwell, E. E., Singh, B. N. Cardiovasc. Res. 1973, 7, 226. Kumana, C. R., Marlin, G. E., Kaye, C. M., Smith, D. M. Br. med.J. 1974, iv, 444. Thompson, M. E., McDonald, R. H., Leon, D. F., Shaver, J. A. Clin. Pharmac. Ther. 1974, 16, 750. Wiseman, R. A. Postgrad. med.J. 1971, 47, suppl. 2, 68. Rowland, M. G. M., Stevenson, C. J. Lancet, 1972, i, 1130. Wright, P. Br. med. J. 1974, ii, 560. Brown, P., Baddeley, H., Read, A. E., Davies, J. D., McGarry, J. Lancet, 1974, ii, 1477. Felix, R. H., Ive, F. A., Dahl, M. G. C. Br. med. J. 1974, iv, 321. Amos, H. E., Brigden, W. D., McKerron, R. A. ibid. 1975, i, 598. Windsor, W. O., Kurrein, F., Dyer, N. H. ibid. p. 68. Wright, P. ibid. p. 595. Basil, B., Jordan, R., Loveless, A. H., Maxwell, D. R. Br.J. Pharmac. 1973, 48, 198. Daly, M. J., Flook, J. J., Levy, G. P. ibid. 1975, 53, 173. Zaid, G., Beall, G. N. New Engl. J. Med. 1966, 275, 580. MacDonald, A. G., Ingram, C. G., McNeill, R. S. Br. J. Anœsth. 1967, 39, 919. Richardson, P. S., Sterling, G. M. Br. med. J. 1969, iii, 143. Turner, P., Burman, J., Hicks, D. C., Cherrington, N. K., MacKinnon, J., Waller, T., Woolnough, M. Arch. int. Pharmacodyn.
Ther. 1971, 191, 104. 25. Nicolaescu, V., Racoveanu, D., Manicatide, M. Eur. J. clin. Pharmac.
1973, 6, 3. 26. Ziskind, M., Weill, H., Bailey, W., Jackson, H., Houser, J. Clin. pharmac. Ther. 1973, 14, 137. 27. Kumana, C. R., Kaye, C. M. Eur. J. clin. Pharmac. 1974, 7, 243. 28. Marcelle, R., Bottin, R., Juchmes, J., Lecomte, J. Acta Allergo-
logica, 1968, XXIII,
hypertensive patients
were
assigned
random either to six weeks’ treatment by yoga relaxation methods with bio-feedback or to placebo therapy (general relaxation). Both showed a reduction in groups blood-pressure (from 168/100 to 141/84 mm. Hg in the treated group and from 169/101 to 160/96 mm. Hg in the control group). The difference was highly significant. The control group was then trained in yoga relaxation, and their blood-pressure fell to that of the other group (now used as controls). at
Introduction PREVIOUS studies 1,2 using yoga and bio-feedback have shown that modification of behaviour designed to maintain a low level of arousal is effective in
reducing resting blood-pressure (B.P.) over a long period. The present study was designed to establish the efficacy of the treatment more formally by all-0cating patients at random to "yoga" or a placebo treatment. As the result of previous experience the total period of training was reduced to six weeks, with two sessions per week, and the training technique was improved by repeating the instructions at each session and by fuller explanation of the disease process and the
treatment.
Patients
hypertensive patients under seventy-nve group practice who had not participated in the previous trials were asked to take part. Hypertensives " were defined as those under pharmacological treatment for at least six months, with initial diastolic B.P. levels of at least 110 mm. Hg on two separate days. They were invited to come to the surgery twice a week All 43 known
years of age in
a
"
for half-hour sessions, at times convenient to themselves between 1 P.M. and 8 P.M., and they were urged to keep to the same time for each session. 37 patients agreed to take part, but 1 woman was excluded before the trial began because her B.P. returned to normal. 2 others were excluded after the trial had started-1 in the control group because of a change in therapy due to severe side-effects
11.
29. Beumer, H. M., Hardonk, H. J. Eur.J. clin. Pharmac. 1972, 5, 77. 30. Cuthbert, M. F., Collins, R. F. Br. J. clin. Pharmac. 1975, 2, 49. 31. Shand, D. G., Nuckolls, E. M., Oates, J. A. Clin. Pharmac. Ther. 1970, 11, 112. 32. Coltart, D. J., Shand, D. G. Br. med. J. 1970, iii, 731. 33. Paterson, J. W., Conolly, M. E., Dollery, C. T., Hayes, A., Cooper, R. G. Pharmac. clin. 1970, 2, 127. 34. Fitzgerald, J. D., O’Donnell, S. R. Br.J. Pharmac. 1971, 43, 222. 35. Robinson, B. F., Epstein, S. E., Beiser, G. D., Braunwald, E. Circulation Res. 1966, 19, 400. 36. Chamberlain, D. A., Turner, P., Sneddon, J. M. Lancet, 1967, ii, 12. 37. Carruthers, S. G., Kelly, J. G., McDevitt, D. G., Shanks, R. G., Walsh, M. J. Br. med. J. 1973, ii, 177. 38. Carruthers, S. G., Kelly, J. G., McDevitt, D. G., Shanks, R. G. Clin. Pharmac. Ther. 1974, 15, 497.
Briant, R. H., Dollery, C. T., Fenyvesi, T., George, C. F. Br. J. Pharmac. 1971, 43, 468P. 40. Baird, J. R. C., Linnell, J.J. Pharm. Pharmac. 1972, 24, 880. 41. Briant, R. H., Dollery, C. T., Fenyvesi, T., George, C. F. Br. J. Pharmac. 1973, 49, 106. 42. Leary, W. P., Coleman, A. J., Asmal, A. C. S. Afr. med.J. 1973, 47, 39.
1245.
Skinner, C., Palmer, K. N. V., Kerridge, D. F. Br. J. clin. Pharmac. (in the press). 44. Curry, P. Personal communication. 45. Kaye, C. M., Kumana, C. R. Br. J. clin. Pharmac. 1974, i, 169. 46. Bodem, G., Brammell, H. L., Weil, J. V., Chidsey, C. A. J. clin. 43.
94 TABLE I-B.P.
(mm. Hg)
from guanethidine, and 1 diabetic in the treatment group who developed renal complications and later died. Characteristics of the remaining 34 patients, who completed the trial, are summarised in table i. Of the treatment group, 15 were on specific antihypertensive drugs, one on tranquillisers, and one was untreated. All the
control-group patients
were
on
antihypertensive drugs.
Method All patients attended on three separate days to provide baseline B.P. The patients were then assigned at random to yoga and bio-feedback or general relaxation. Pressures were taken in the standing, sitting, and supine positions, and repeated after half an hour’s rest. A random-zero
sphygmomanometer was used.3 All blood-pressures (fifth phase) were measured " blind by the practice nurse. Patients were asked not to change their drug therapy during the trial. The first phase of the trial consisted of twelve sessions, two a week for six weeks, B.P. being again taken in the standing, sitting, and supine positions before and after the half-hour session. Active treatment consisted, first, of meetings during which the patients were shown films and slides about hypertension, the different ways in which emotion affects bodily processes, the physiology of relaxation, the concept of bio-feedback, self-control, and so on. They discussed their queries and problems freely. The rapport created between doctor and patients as well as that between the patients themselves greatly helped to strengthen the programme and ensure cooperation. Next, during individual training, patients lay on a couch or a reclining chair with legs apart and slightly rotated at the hip joint, arms by the side, shoulders flat, jaw limp, and eyes lightly closed; the head was raised to a comfortable position, but they were encouraged to lie as flat as possible. They were first asked to breathe slowly and rhythmically. Next, they were instructed to go over the different parts of the body mentally in aregular sequence, allowing each part to relax completely. Patients were instructed verbally on this methodical relaxation for ten to twelve minutes, and were then encouraged to remain relaxed for the rest of the session. Once the patient had mastered the method of relaxation, a type TABLE II-CHANGE IN B.P.
BEFORE TRIAL
of transcendental meditation was introduced. Throughout the session the patient was connected to one of two biofeedback instruments giving a continuous audio-signal whose pitch fell as the patient relaxed. Generally the instrument measuring electrical resistance of the skin (’Relaxometer’, Aleph One Ltd., Cambridge;’G.S.R. 90’, Biofeedback Systems Ltd., Manchester) was used for the first few sessions followed by an electromyograph (’Myophone’, Aleph One Ltd.). This confirmation of relaxation was intended to encourage further relaxation. Patients were also encouraged verbally, and shown their B.P. records; they were also instructed to practise relaxation and meditation twice a day, and gradually to try to incorporate these habits into routine activities, the methods depending on individual circumstances. For example, each patient had a red disc attached to his watch to remind him to relax whenever he looked at the time, and some were told to relax before answering the telephone. Patients in the control group attended for the same number of sessions and for the same length of time. However, they were asked to relax on the couch or the reclining chair without being given specific instruction, and they were not connected to bio-feedback instruments. At the end of the trial all patients were followed up every two weeks for three months. In case the observed difference had been due to unknown factors not eliminated by randomisation we then treated the control group with yoga and bio-feedback therapy. Phase 2 of the trial began two months after the end of the follow-up for phase 1. The previous control group, now the new treatment group, attended for twelve treatment sessions as in phase 1. 1 patient dropped out because he could not attend sessions regularly. The previously treated group were used as controls, but were seen only at the beginning and end of this phase to give B.P. readings for comparison. Only one initial and one followup reading was made in each group. 2 patients changed their drugs during the trial. 1 woman in the treated group had hydrallazine added to her propranolol therapy, but remained on it for only three weeks in the middle of phase 1. Another woman in the control group was taken off guanethidine at the end of phase 1, and remained off treatment for phase 2.
(mm. Hg) DURING PHASE
1
95 Discussion
Results
Every B.P. used in the analysis is the mean of three readings taken with the patient standing, sitting, and lying. Although measurements were made at the beginning and end of each session, only those at the beginning of sessions have been used here (B.P. at the end of sessions was invariably lower). In addition, figures for " initial " pressures in phase 1 are the means of readings at the three preliminary sessions, and those for "final" pressures the means of the six follow-up sessions. (In phase 2 there was only one preliminary and follow-up session for both groups.) The results for the first phase of the trial are summarised in table ii, showing only the average initial and final B.P.S in each group. Systolic B.P. fell by an average of 26-1 mm. Hg in the treated group and by 8-9 mm. Hg in the controls. The correspond-
ing figures for diastolic pressures are 15-2 and 4-2 mm. Hg. The differences between the groups are highly significant. In addition there was a signifi-
In phase 1 the treated group showed a much larger reduction in B.P. than the control group. Evidence that this was due primarily to the relaxation technique they had learnt is provided by the fact that at the start of phase 2 their B.P. remained at its lower level, while that in the control group had risen to at least the initial level. Bio-feedback therapy has two components. The first is objective, and depends on showing changes in a physiological function as displayed by auditory or visual signals. The second is thought to be a subjective state which cannot be monitored. A subject may evolve his own means of producing a desirable change in a bio-feedback experiment, but to advance this from a simple demonstration of control over an autonomic function in an experimental situation into a meaningful therapeutic procedure in daily life it is necessary to evolve a technique that will ’
produce predictable objective changes. A subsidiary study on these patients4 indicates that the treated patients showed a smaller rise in B.P. than the control patients after being subjected to experimental stress. In view of the possible importance of this type of therapy and the absence of undesirable side-effects, it seems desirable that further trials should be carried out in otherwise untreated patients under single-blind conditions, perhaps in hypertension clinics. We thank Prof. T. R. E. Pilkington for his encouragement this project, Dr W. E. Miall for his help and criticism, Dr Denis Craddock, Dr Barbara Barnes, and Dr Robert Cruthers for allowing their patients to participate, Mrs M. Gould for taking all the B.P. measurements, and Mrs A. Miller for helping with group discussions. This work was supported in part by a grant from the South-West Thames Regional Health Authority. Requests for reprints should be addressed to C. P., 11 Upfield, Croydon CRO 5DR. to start
REFERENCES B.P. changes
(means) during
two
phases of trial of
yoga and
bio-feedback.
fall in both systolic and diastolic pressure in the control group, though this did not apply to all patients. 2 controls showed a rise in both systolic and diastolic B.P., and another 6 had a rise in one of the two. No patients in the treated group showed a rise in pressure. The means of each group week by week in phase 1 are shown in the left-hand half of the figure. At the beginning of phase 2 B.P. was higher in the new treatment group (formerly control) than at the beginning of the first phase; this could be partly because only one reading was taken. B.P. in the former treated group, however, was close to the level at the end of phase 1; this group was not seen again till the end of the trial. The results are shown in table in and the second half of the figure. The results for the phase 2 treated group were similar to those in phase 1 (i.e., pressure fell significantly) while B.P. in the phase 2 control group changed very little. No significance tests were performed for this phase as their validity seems doubtful, but it is clear that there is again a large difference. No member of the treated group showed a rise in B.P.
1. 2. 3. 4.
Patel, C. Lancet, 1973, ii, 1053. Patel, C. ibid. 1975, i, 62. Wright, B. M., Dore, C. F. ibid. 1970, i, 337. Patel, C. Clin. Sci. mol. Med. 1975, 46, suppl. p. 171.
cant
THE CURABILITY OF BREAST CANCER DIANA BRINKLEY * J. L. HAYBITTLE Addenbrooke’s Hospital, Cambridge
Sum ary cer seen
Results are presented of a long-term follow-up of patients with breast canin the Cambridge area from 1947 to 1950.
After 21 years the survival
of the breast-cancer patients parallel expected survival curve of a similar normal population. This suggests that 18 % of all the breast-cancer patients may be regarded as cured of their disease in the sense defined by Easson and Russell.5 An analysis of the causes of death in the study group after 20 years’ follow-up shows an excess of deaths from breast cancer. The number is 16 times more than would be expected in the normal runs
*
to
curve
the
Present address: King’s College Hospital, Denmark Hill, London SE5 9RS.
RANDOMISED CONTROLLED TRIAL OF YOGA AND BIO-FEEDBACK IN MANAGEMENT OF HYPERTENSION
ratio of practolol to propranolol of 23/1 is in fair agreement with that of 25/1obtained by Bodem et al.46 and of 28/1 obtained in a previous investiga-
tion.9
conclusion, the results of this within-subject study using single intravenous injections suggest that: 1. When the cardiac (3-blocking activity of propranolol, acebutolol, and practolol was nearly equivalent (as judged by reduction in exercise heart-rate) their mean plasma concentrations were in the ratio 1/8/23. 2. Propranolol was not cardioselective (as judged by F.E.V.I, resting and exercise P.F.R. measurements). 3. Acebutolol was not as cardioselective as practolol (as judged by exercise P.F.R. measurements). In
We thank our colleagues at the School of ,Pharmacy, London, who assisted us in this study, May and Baker for supplying acebutolol, Imperial Chemical Industries for supplying propranolol and practolol, and the Board of Governors of St. Bartholomew’s Hospital and the British Heart Foundation for financial support. Requests for reprints should be addressed to J. H., Department of Cardiology, St. Bartholomew’s Hospital, London EC1A 7BE.
CHANDRA PATEL
Croydon, Surrey W. R. S. NORTH M.R.C./D.H.S.S. Epidemiology and Medical Care Unit, Northwick Park Hospital, Harrow, Middlesex
Summary
34
REFERENCES 1. 2. 3. 4.
5. 6. 7. 8.
9. 10. 11. 12. 13. 14.
15. 16. 17. 18. 19.
20. 21. 22.
23. 24.
Gillam, P. M. S., Prichard, B. N. C. Br. med. J. 1965, ii, 337. Aronow, W. S. Am. Heart J. 1972, 84, 834. Wilhelmsson, C., Vedin, J. A., Wilhelmsen, L., Tibblin, G., Werkö, L. Lancet, 1974, ii, 1157. Fox, K. M., Chopra, M. P., Portal, R. W., Aber, C. P. Br. med. J. 1975, i, 117. Jackson, G., Atkinson, L., Oram, S. ibid. p. 708. MacDonald, A. G., McNeill, R. S. Br. J. Anœsth. 1968, 40, 508. Levy, B. J. Pharmac. exp. Ther. 1973, 186, 134. Vaughan Williams, E. M., Bagwell, E. E., Singh, B. N. Cardiovasc. Res. 1973, 7, 226. Kumana, C. R., Marlin, G. E., Kaye, C. M., Smith, D. M. Br. med.J. 1974, iv, 444. Thompson, M. E., McDonald, R. H., Leon, D. F., Shaver, J. A. Clin. Pharmac. Ther. 1974, 16, 750. Wiseman, R. A. Postgrad. med.J. 1971, 47, suppl. 2, 68. Rowland, M. G. M., Stevenson, C. J. Lancet, 1972, i, 1130. Wright, P. Br. med. J. 1974, ii, 560. Brown, P., Baddeley, H., Read, A. E., Davies, J. D., McGarry, J. Lancet, 1974, ii, 1477. Felix, R. H., Ive, F. A., Dahl, M. G. C. Br. med. J. 1974, iv, 321. Amos, H. E., Brigden, W. D., McKerron, R. A. ibid. 1975, i, 598. Windsor, W. O., Kurrein, F., Dyer, N. H. ibid. p. 68. Wright, P. ibid. p. 595. Basil, B., Jordan, R., Loveless, A. H., Maxwell, D. R. Br.J. Pharmac. 1973, 48, 198. Daly, M. J., Flook, J. J., Levy, G. P. ibid. 1975, 53, 173. Zaid, G., Beall, G. N. New Engl. J. Med. 1966, 275, 580. MacDonald, A. G., Ingram, C. G., McNeill, R. S. Br. J. Anœsth. 1967, 39, 919. Richardson, P. S., Sterling, G. M. Br. med. J. 1969, iii, 143. Turner, P., Burman, J., Hicks, D. C., Cherrington, N. K., MacKinnon, J., Waller, T., Woolnough, M. Arch. int. Pharmacodyn.
Ther. 1971, 191, 104. 25. Nicolaescu, V., Racoveanu, D., Manicatide, M. Eur. J. clin. Pharmac.
1973, 6, 3. 26. Ziskind, M., Weill, H., Bailey, W., Jackson, H., Houser, J. Clin. pharmac. Ther. 1973, 14, 137. 27. Kumana, C. R., Kaye, C. M. Eur. J. clin. Pharmac. 1974, 7, 243. 28. Marcelle, R., Bottin, R., Juchmes, J., Lecomte, J. Acta Allergo-
logica, 1968, XXIII,
hypertensive patients
were
assigned
random either to six weeks’ treatment by yoga relaxation methods with bio-feedback or to placebo therapy (general relaxation). Both showed a reduction in groups blood-pressure (from 168/100 to 141/84 mm. Hg in the treated group and from 169/101 to 160/96 mm. Hg in the control group). The difference was highly significant. The control group was then trained in yoga relaxation, and their blood-pressure fell to that of the other group (now used as controls). at
Introduction PREVIOUS studies 1,2 using yoga and bio-feedback have shown that modification of behaviour designed to maintain a low level of arousal is effective in
reducing resting blood-pressure (B.P.) over a long period. The present study was designed to establish the efficacy of the treatment more formally by all-0cating patients at random to "yoga" or a placebo treatment. As the result of previous experience the total period of training was reduced to six weeks, with two sessions per week, and the training technique was improved by repeating the instructions at each session and by fuller explanation of the disease process and the
treatment.
Patients
hypertensive patients under seventy-nve group practice who had not participated in the previous trials were asked to take part. Hypertensives " were defined as those under pharmacological treatment for at least six months, with initial diastolic B.P. levels of at least 110 mm. Hg on two separate days. They were invited to come to the surgery twice a week All 43 known
years of age in
a
"
for half-hour sessions, at times convenient to themselves between 1 P.M. and 8 P.M., and they were urged to keep to the same time for each session. 37 patients agreed to take part, but 1 woman was excluded before the trial began because her B.P. returned to normal. 2 others were excluded after the trial had started-1 in the control group because of a change in therapy due to severe side-effects
11.
29. Beumer, H. M., Hardonk, H. J. Eur.J. clin. Pharmac. 1972, 5, 77. 30. Cuthbert, M. F., Collins, R. F. Br. J. clin. Pharmac. 1975, 2, 49. 31. Shand, D. G., Nuckolls, E. M., Oates, J. A. Clin. Pharmac. Ther. 1970, 11, 112. 32. Coltart, D. J., Shand, D. G. Br. med. J. 1970, iii, 731. 33. Paterson, J. W., Conolly, M. E., Dollery, C. T., Hayes, A., Cooper, R. G. Pharmac. clin. 1970, 2, 127. 34. Fitzgerald, J. D., O’Donnell, S. R. Br.J. Pharmac. 1971, 43, 222. 35. Robinson, B. F., Epstein, S. E., Beiser, G. D., Braunwald, E. Circulation Res. 1966, 19, 400. 36. Chamberlain, D. A., Turner, P., Sneddon, J. M. Lancet, 1967, ii, 12. 37. Carruthers, S. G., Kelly, J. G., McDevitt, D. G., Shanks, R. G., Walsh, M. J. Br. med. J. 1973, ii, 177. 38. Carruthers, S. G., Kelly, J. G., McDevitt, D. G., Shanks, R. G. Clin. Pharmac. Ther. 1974, 15, 497.
Briant, R. H., Dollery, C. T., Fenyvesi, T., George, C. F. Br. J. Pharmac. 1971, 43, 468P. 40. Baird, J. R. C., Linnell, J.J. Pharm. Pharmac. 1972, 24, 880. 41. Briant, R. H., Dollery, C. T., Fenyvesi, T., George, C. F. Br. J. Pharmac. 1973, 49, 106. 42. Leary, W. P., Coleman, A. J., Asmal, A. C. S. Afr. med.J. 1973, 47, 39.
1245.
Skinner, C., Palmer, K. N. V., Kerridge, D. F. Br. J. clin. Pharmac. (in the press). 44. Curry, P. Personal communication. 45. Kaye, C. M., Kumana, C. R. Br. J. clin. Pharmac. 1974, i, 169. 46. Bodem, G., Brammell, H. L., Weil, J. V., Chidsey, C. A. J. clin. 43.
94 TABLE I-B.P.
(mm. Hg)
from guanethidine, and 1 diabetic in the treatment group who developed renal complications and later died. Characteristics of the remaining 34 patients, who completed the trial, are summarised in table i. Of the treatment group, 15 were on specific antihypertensive drugs, one on tranquillisers, and one was untreated. All the
control-group patients
were
on
antihypertensive drugs.
Method All patients attended on three separate days to provide baseline B.P. The patients were then assigned at random to yoga and bio-feedback or general relaxation. Pressures were taken in the standing, sitting, and supine positions, and repeated after half an hour’s rest. A random-zero
sphygmomanometer was used.3 All blood-pressures (fifth phase) were measured " blind by the practice nurse. Patients were asked not to change their drug therapy during the trial. The first phase of the trial consisted of twelve sessions, two a week for six weeks, B.P. being again taken in the standing, sitting, and supine positions before and after the half-hour session. Active treatment consisted, first, of meetings during which the patients were shown films and slides about hypertension, the different ways in which emotion affects bodily processes, the physiology of relaxation, the concept of bio-feedback, self-control, and so on. They discussed their queries and problems freely. The rapport created between doctor and patients as well as that between the patients themselves greatly helped to strengthen the programme and ensure cooperation. Next, during individual training, patients lay on a couch or a reclining chair with legs apart and slightly rotated at the hip joint, arms by the side, shoulders flat, jaw limp, and eyes lightly closed; the head was raised to a comfortable position, but they were encouraged to lie as flat as possible. They were first asked to breathe slowly and rhythmically. Next, they were instructed to go over the different parts of the body mentally in aregular sequence, allowing each part to relax completely. Patients were instructed verbally on this methodical relaxation for ten to twelve minutes, and were then encouraged to remain relaxed for the rest of the session. Once the patient had mastered the method of relaxation, a type TABLE II-CHANGE IN B.P.
BEFORE TRIAL
of transcendental meditation was introduced. Throughout the session the patient was connected to one of two biofeedback instruments giving a continuous audio-signal whose pitch fell as the patient relaxed. Generally the instrument measuring electrical resistance of the skin (’Relaxometer’, Aleph One Ltd., Cambridge;’G.S.R. 90’, Biofeedback Systems Ltd., Manchester) was used for the first few sessions followed by an electromyograph (’Myophone’, Aleph One Ltd.). This confirmation of relaxation was intended to encourage further relaxation. Patients were also encouraged verbally, and shown their B.P. records; they were also instructed to practise relaxation and meditation twice a day, and gradually to try to incorporate these habits into routine activities, the methods depending on individual circumstances. For example, each patient had a red disc attached to his watch to remind him to relax whenever he looked at the time, and some were told to relax before answering the telephone. Patients in the control group attended for the same number of sessions and for the same length of time. However, they were asked to relax on the couch or the reclining chair without being given specific instruction, and they were not connected to bio-feedback instruments. At the end of the trial all patients were followed up every two weeks for three months. In case the observed difference had been due to unknown factors not eliminated by randomisation we then treated the control group with yoga and bio-feedback therapy. Phase 2 of the trial began two months after the end of the follow-up for phase 1. The previous control group, now the new treatment group, attended for twelve treatment sessions as in phase 1. 1 patient dropped out because he could not attend sessions regularly. The previously treated group were used as controls, but were seen only at the beginning and end of this phase to give B.P. readings for comparison. Only one initial and one followup reading was made in each group. 2 patients changed their drugs during the trial. 1 woman in the treated group had hydrallazine added to her propranolol therapy, but remained on it for only three weeks in the middle of phase 1. Another woman in the control group was taken off guanethidine at the end of phase 1, and remained off treatment for phase 2.
(mm. Hg) DURING PHASE
1
95 Discussion
Results
Every B.P. used in the analysis is the mean of three readings taken with the patient standing, sitting, and lying. Although measurements were made at the beginning and end of each session, only those at the beginning of sessions have been used here (B.P. at the end of sessions was invariably lower). In addition, figures for " initial " pressures in phase 1 are the means of readings at the three preliminary sessions, and those for "final" pressures the means of the six follow-up sessions. (In phase 2 there was only one preliminary and follow-up session for both groups.) The results for the first phase of the trial are summarised in table ii, showing only the average initial and final B.P.S in each group. Systolic B.P. fell by an average of 26-1 mm. Hg in the treated group and by 8-9 mm. Hg in the controls. The correspond-
ing figures for diastolic pressures are 15-2 and 4-2 mm. Hg. The differences between the groups are highly significant. In addition there was a signifi-
In phase 1 the treated group showed a much larger reduction in B.P. than the control group. Evidence that this was due primarily to the relaxation technique they had learnt is provided by the fact that at the start of phase 2 their B.P. remained at its lower level, while that in the control group had risen to at least the initial level. Bio-feedback therapy has two components. The first is objective, and depends on showing changes in a physiological function as displayed by auditory or visual signals. The second is thought to be a subjective state which cannot be monitored. A subject may evolve his own means of producing a desirable change in a bio-feedback experiment, but to advance this from a simple demonstration of control over an autonomic function in an experimental situation into a meaningful therapeutic procedure in daily life it is necessary to evolve a technique that will ’
produce predictable objective changes. A subsidiary study on these patients4 indicates that the treated patients showed a smaller rise in B.P. than the control patients after being subjected to experimental stress. In view of the possible importance of this type of therapy and the absence of undesirable side-effects, it seems desirable that further trials should be carried out in otherwise untreated patients under single-blind conditions, perhaps in hypertension clinics. We thank Prof. T. R. E. Pilkington for his encouragement this project, Dr W. E. Miall for his help and criticism, Dr Denis Craddock, Dr Barbara Barnes, and Dr Robert Cruthers for allowing their patients to participate, Mrs M. Gould for taking all the B.P. measurements, and Mrs A. Miller for helping with group discussions. This work was supported in part by a grant from the South-West Thames Regional Health Authority. Requests for reprints should be addressed to C. P., 11 Upfield, Croydon CRO 5DR. to start
REFERENCES B.P. changes
(means) during
two
phases of trial of
yoga and
bio-feedback.
fall in both systolic and diastolic pressure in the control group, though this did not apply to all patients. 2 controls showed a rise in both systolic and diastolic B.P., and another 6 had a rise in one of the two. No patients in the treated group showed a rise in pressure. The means of each group week by week in phase 1 are shown in the left-hand half of the figure. At the beginning of phase 2 B.P. was higher in the new treatment group (formerly control) than at the beginning of the first phase; this could be partly because only one reading was taken. B.P. in the former treated group, however, was close to the level at the end of phase 1; this group was not seen again till the end of the trial. The results are shown in table in and the second half of the figure. The results for the phase 2 treated group were similar to those in phase 1 (i.e., pressure fell significantly) while B.P. in the phase 2 control group changed very little. No significance tests were performed for this phase as their validity seems doubtful, but it is clear that there is again a large difference. No member of the treated group showed a rise in B.P.
1. 2. 3. 4.
Patel, C. Lancet, 1973, ii, 1053. Patel, C. ibid. 1975, i, 62. Wright, B. M., Dore, C. F. ibid. 1970, i, 337. Patel, C. Clin. Sci. mol. Med. 1975, 46, suppl. p. 171.
cant
THE CURABILITY OF BREAST CANCER DIANA BRINKLEY * J. L. HAYBITTLE Addenbrooke’s Hospital, Cambridge
Sum ary cer seen
Results are presented of a long-term follow-up of patients with breast canin the Cambridge area from 1947 to 1950.
After 21 years the survival
of the breast-cancer patients parallel expected survival curve of a similar normal population. This suggests that 18 % of all the breast-cancer patients may be regarded as cured of their disease in the sense defined by Easson and Russell.5 An analysis of the causes of death in the study group after 20 years’ follow-up shows an excess of deaths from breast cancer. The number is 16 times more than would be expected in the normal runs
*
to
curve
the
Present address: King’s College Hospital, Denmark Hill, London SE5 9RS.
