- Email: [email protected]
The Management of Intractable Pain in Patients with Advanced Malignant Disease
0022-534 7/78/ 1206-0720$02. 00/0
THE
Vol. 120, December Printed in U.S.A.
JOURNAL OF UROLOGY
Copyright © 1978 by The Williams & Wilkins Co.
THE MANAGEMENT OF INTRACTABLE PAIN IN PATIENTS WITH ADVANCED MALIGNANT DISEASE B. M. MOUNT, R. MELZACK
AND
K. J. MACKINNON
From the Royal Victoria Hospital, Montreal, Quebec, Canada
ABSTRACT
The Brampton mixture is a highly effective, flexible, safe and convenient means to control chronic pain of malignant disease. The mixture is a solution containing morphine, the dose of narcotic varying with the need for analgesia, and is given regularly, usually every 4 hours, with a phenothiazine. The main aims of therapy are prevention of pain rather than treatment, an unclouded sensorium and a normal effect. Terminally ill cancer patients were given the Brampton mixture and a phenothiazine in an attempt to control their pain. The mixture was administered to patients in 3 hospital environments: 1) a palliative care unit, 2) general wards and 3) private rooms. Pain was measured in 92 patients with the McGill-Melzack pain questionnaire. The Brampton mixture controlled pain in 90 per cent of patients in the palliative care unit and in 75 to 80 per cent of patients in the wards or private rooms. The differences in pain scores between patients in the palliative care unit and the other groups were significant. The mixture produced substantial decreases in the 3 major dimensions of pain: 1) sensory, 2) affective and 3) evaluative. Comparison of these results with data obtained in an outpatient pain clinic showed that the Brampton mixture was strikingly more effective than the traditional methods of managing cancer pain. A significant percentage of patients with advanced malignant disease has intractable pain. With moderate to severe chronic pain only the narcotic analgesics provide adequate control. Milder analgesics should be tried for less severe pain and may be helpful in combination with more potent drugs. A wide variety of agents is available. 1 When it is localized moderate to severe pain may be managed adequately by irradiation, nerve block or some form of ablative neurosurgical procedure. In some situations, such as cases ofprostatic carcinoma, specific forms of therapy, including endocrine manipulation, hypophysectomy or radioactive isotope of phosphorus, may give transient relief. However, the response frequently is short lived and not experienced by all patients treated. We are all familiar with the all-too-frequent failures in pain control encountered in managing these unfortunate patients. THE BROMPTON MIXTURE
Background. Although generations of British physicians have gained familiarity with variants of the oral narcotic mixture bearing the name of the Brompton Chest Hospital it was not until 1973 that this formulation was recognized in the "British Pharmaceutical Codex". 2 The Brompton mixture is used for pain of advanced malignant disease when non-narcotic and milder narcotic preparations are ineffective. The important experience of Saunders and Twycross at St. Christopher's Hospice and St. Joseph's Hospice has led to a refinement and standardization of approach that have been associated with greatly increased effectiveness. 3• 4 On reviewing this experience with 418 patients who received narcotics for pain of malignant disease Twycross concluded 1) although most patients received parenteral treatment during . the last 12 to 48 hours because of increasing debility, the majority can be maintained on orally administered diacetylmorphine before this time, 2) there is no single optimal dose or maximum effective dose of diacetylmorphine, 3) the preAccepted for publication August 25, 1978. Supported in part by Grant A7891 from the National Research Council of Canada. 720
scription of diacetylmorphine does not, by itself, lead to impairment of mental faculties, 4) tolerance is not a practical problem, 5) psychological dependence does not occur and 6) physical dependence may develop but appears not to prevent the downward adjustment of the dose of diacetylmorphine when considered clinically feasible. 4 In a subsequent prospective randomized double-blind trial of great significance Twycross found that apart from an oral analgesic potential ratio of 1.5:1 (that is 15 mg. morphine is equivalent to 10 mg. diacetylmorphine orally) diacetylmorphine and morphine administered by mouth were equipotent and equitoxic. 5 Thus, the only advantage of diacetylmorphine is its greater solubility, which allows for smaller volume iajections in the occasional patient requiring parenteral nar0 cotics in high doses. Administration. The standard oral narcotic mixture used at the Royal Victoria Hospital is shown in table 1. The contributions to the effectiveness of this mixture of the small amounts of cocaine and ethyl alcohol are uncertain. Further elucidation must await the results of trials aimed at simplifying the mixture. For most patients analgesia is obtained with 5 to 20 mg. morphine per dose of the mixture but small or elderly patients may require as little as 2.5 mg. The usual sequential doses of morphine given are 2.5, 5, 10, 15, 20, 30, 40, 60, 90 and 120 mg. This standard elixir always is given with a phenothiazine. The phenothiazines are potent antiemetics and, thus, are useful in countering vomiting, a frequent side effect of narcotics. Experience also suggests that they potentiate the narcotic analgesia. Prochlorperazine, 5 mg. in 5 ml., is usually highly effective as an antiemetic with little sedative effect. If restlessness or agitation is a feature 10 to 25 mg. chlorpromazine may be substituted. Twycross and Gilhooley have shown a shelf life of sustained effectiveness that exceeds 8 weeks. 6 Lengthy anticipated survival is not a contraindication since the mixture may be used for many months to several years without dose escalation with care in adjusting the dosage to meet the patient's needs. In adjusting the dosage of narcotic and phenothiazine to
MANAGEMENT OF INTRACTABLE PAIN IN PATIENTS WITH ADVANCED MALIGNANT DISEASE TABLE
1. Royal Victoria Hospital oral narcotic mixture Morphine, 5 mg. or more Cocaine, 10 mg. Ethyl alcohol (98%), 2.5 cc Syrup, 5 cc Chloroform water Ad, 20 cc given with Prochlorperazine, 5 mg. or more (or chlorpromazine, 10 n,g. or.more)
achieve a pain-free state without sedation a number of factors are important. The morphine elixir should be given in 20 ml. doses with the phenothiazine every 4 hours around the clock because the serum half-life of morphine taken orally is about 4 hours. Occasionally, a patient may require a 3-hour schedule. The night dose is omitted only when the patient can sleep through the night free of pain. Careful attention to exact dosage and timing will pay dividends in results. For most patients a pain-free state can be achieved by giving sequential increments in narcotic dose. 7 To treat excruciating pain one may elect to start with a higher narcotic dose then make sequential decrements until analgesia without sedation is achieved. In determining the dosage required dose alterations should be made at intervals of 48 to 72 hours. During this period further analgesia can be achieved with the use of supplemental analgesics as required. In general, it is wise to change only 1 variable, the narcotic or the phenothiazine, at a time. Since the phenothiazines and morphine are synergistic great care must be exercised. Small changes in either variable may produce profound changes in analgesia and sedation. Initiation of narcotic therapy usually will produce transient sedation lasting 48 to 72 hours. It is important to reassure patient and family that pain can and will be controlled and that the initial drowsiness is temporary. Their confidence that control can be achieved will promote analgesia. Dispensing the morphine mixture and the phenothiazine syrup separately allows greater flexibility in adjusting dosage. Once a continuous pain-free state is achieved they may be combined in dispensing for greater ease of administration. Careful observation of the patient's condition during a complete 24-hour period may suggest augmentation of 1 or 2 specific doses at periods of peak activity. If parenteral medication becomes necessary the equivalent dose of morphine is half the previous oral dose. Thus, a patient whose pain has been controlled with 30 mg. orally would then receive 15 mg. intramuscularly. 4 MEASUREMENT OF PAIN
Because there have not been adequate tools to measure pain it has been difficult to gauge the magnitude of the effect of the Brampton mixture, its relative effects in different hospital environments or how it compares to traditional methods of pain control in cancer patients. The recent development of the McGill-Melzack pain questionnaire to measure clinical pain has given us an opportunity to examine these aspects of its effect on pain in cancer patients. 8 The 3 major indexes measured are present pain intensity or over-all pain intensity, measured on a scale of Oto 5 (0-none, 1-mild, 2-discomforting, 3-distressing, 4-horrible and 5-excruciating); number of words chosen from 20 sets of qualitative words with 2 to 6 words per set that describe the sensory, affective and evaluative properties of pain, and pain rating index, the sum of the rank values of all the words chosen (based on the positions of the words in each set (see figure). PROCEDURE
Each patient selected for the study was informed that he was being given a drug, the Brampton mixture, to control pain and that it was important to evaluate the intensity and kind of pain he felt in order to determine the proper dosage of
the drug. Each patient was instructed as follows: "I will read several groups of words that express feelings and sensations. If any of these words describe what you feel right now, please tell me and I will make a mark at the side of the appropriate word. Choose only 1 word in each group, the one that best expresses how you feel. Omit any groups that are not appropriate." The questionnaire generally was completed in about 10 minutes. At first more time was required to explain the meaning of some of the words. Occasionally, patients asked to have certain groups reread to be certain of the decision that a word was appropriate. When the patient felt no pain a zero was recorded and the list of descriptors was not read. PATIENTS
Palliative care unit. During this study 143 patients were registered in the palliative care unit at the Royal Victoria Hospital, a ward specializing in the care of the terminally ill. 9 Of these patients 90 received the Brampton mixture. Of the remainder many had relatively little pain, which could be kept under control with codeine or another mild or mediumstrength, orally administered analgesic. Others were unable to receive oral medication for a variety of medical reasons and, indeed, some died within hours of arrival. Of the 90 patients who received the Brampton mixture the pain could not be controlled in 8 (9 per cent), including 1 who had severe bladder spasms, 2 with sharp nerve-root pain that radiated into the legs and 5 with severe pain, a major component of which was their despair and anguish at their impending death. These patients were treated with additional or other methods in an attempt to achieve physical and psychological comfort. Of the 90 patients 82 received the Brampton mixture on a continuing basis and 26 were chosen at random to answer the pain questionnaire. The mean number of questionnaires given to the patients during their stay in the palliative care unit was 4. 7 (range 1 to 16). Wards and private rooms. During the study 238 patients on general hospital wards or in private rooms were given the Brompton mixture. The ward rooms consisted of 4-bed standard accommodations on the general medical and surgical wards. The private rooms consisted of single-bed accommodations. As in the palliative care unit group the pain required additional or other treatment in about 10 per cent of the patients. The remainder received the Brampton mixture on a continuing basis and 66 (46 ward patients and 20 private patients) were chosen at random to answer the pain questionnaire. The mean number of questionnaires given to the ward patients was 4.6 (range 1 to 20) and to the private patients, 7. 7 (range 1 to 49). While the standard Brompton regimen was followed in the palliative care unit setting there was greater variability in time and dose scheduling in the wards and private rooms. After the first 8 months of the study an analysis of the data indicated differences in pain control among the palliative care unit, ward and private patients. Consequently, it was impressed on the staff associated with all 3 groups of patients that it was essential to administer the Brompton mixture every 4 hours, even if the patient claimed he was not in pain or if it meant waking him to provide medication. RESULTS
Questionnaire scores. The mean present pain intensity and pain rating index scores for the palliative care unit, ward and private patients are shown in table 2. Pain scores for all 3 groups were remarkably low. Since a present pain intensity score of 1 represents mild pain it is clear that most ward patients had mild pain, private patients had mild to discomforting pain and patients in the palliative care unit had mild pain or none at all. The Brampton mixture, therefore, had a powerful effect on perceived pain in patients in all groups but
722
MOUNT, MELZACK AND MACKINNON
McGill - Melzack Pain Questionnaire Patient's Name
Analgesic(s)
Date Dosage Dosage
Time Time Given Time Given
Analgesic Time Difference (hours): +4 +l +2 +3 PRI: s A E M(S) M(AE) (1-10) (11-15) (16) (17-19) 1 FLICKERING_ QUIVERING PULSING THROBBING BEATING POUNDING 2 JUMPING FLASHING SHOOTING 3 PRICKING BORING DRILLING STABBING LANCINATING 4 SHARP CUTTING LACERATING 5 PINCHING PRESSING GNAWING CRAMPING CRUSHING 6 TUGGING PULLING WRENCHING 7 HOT BURNING SCALDING SEARING 8 TINGLING ITCHY SMARTING STINGING 9 DULL SORE HURTING ACHING HEAVY 1 0 TENDER TAUT RASPING SPLITTING
-
-
-
-
-
-
11 TIRING EXHAUSTING 12 SICKENING SUFFOCATING 13 FEARFUL FRIGHTFUL TERRIFYING 14 PUNISHING GRUELLING CRUEL VICIOUS KILLING 15 WRETCHED BLINDING 16 ANNOYING TROUBLESOME MISERABLE INTENSE UNBEARABLE 17 SPREADING RADIATING PENETRATING PIERCING 18 TIGHT NUMB DRAWING SQUEEZING TEARING 19 COOL COLD FREEZING 20 NAGGING NAUSEATING AGONIZING DRF.ADFUL TORTURING PPI 0 No pain 1 MILD 2 DISCOMFORTING_ 3 DISTRESSING 4 HORRIBLE 5 EXCRUCIATING
-
PPI _ _
(20)
am/pm am/pm am/pm M(T) _ _ _PRI(T) ( 17-20) (1-20)
fOMMENTS:
-
-
-
...,, ......
I\
-
-
-
r
CONSTANT PERIODIC BRIEF
1 ACCOMPANYING SYMPTOMS: NAUSEA HEADACHE DIZZINESS DROWSINESS _ CONSTIPATION _ DIARRHEA COMMENTS:
-
SLEEP: GOOD FITFUL CAN'T SLEEP COMMENTS:
INTAKE:
COMMENTS: S:
SOME LITTLE NONE
McGill-Melzack pain questionnaire, adapted for study of Brompton mixture. Descriptors fall into major groups: sensory- I to 10; affective-11 to 15; evaluative-16 and miscellaneous-17 to 20. Rank value for each descriptor is based on its position in word set. Sum of rank values is pain rating index (PRI). Present pain intensity (PPI) is based on scale of Oto 5.
a significantly greater effect on patients in the palliative care unit. The mean present pain intensity and pain rating index values for palliative care unit patients were significantly lower compared to ward and private patients, that is p values <0.05 to <0.001 in all cases as measured by 2-tailed tests based on the median test. 10 The percentages of patients with varying degrees of pain are shown in table 3. The mean present pain intensity and pain rating index scores for each patient were assigned to 1 of 4 categories: 1) none, 2) mild, 3) discomforting and 4) unbearable pain. The percentages of patients in each group are approximately the same on the basis of the present pain intensity and the pain rating index scores. On the basis of the present pain intensity it is apparent that all patients in the palliative care unit who received the Brompton mixture had bearable pain. Among the ward and private patients about 10 to 15 per cent of those taking the mixture continued to have unbearable pain. These data exclude the 9 to 10 per cent of patients in all 3 groups whose pain could not be controlled with the Brompton mixture alone and who required additional or other methods to achieve physical and psychological comfort. If these patients
are included in the over-all analysis it may be concluded that the Brompton mixture was effective in controlling pain in 90 per cent of patients in wards or private rooms. The effectiveness of the Brampton mixture is indicated further in table 4. In an earlier study all patients who attended the outpatient pain clinic at the Royal Victoria Hospital were requested to complete the pain questionnaires. 8 There were 16 patients who had pain associated with cancer. Their scores for present pain intensity, number of words chosen and pain rating index are shown in table 4 along with those of the palliative care unit, ward and private patients. All patients who obtained total pain relief with the Brampton mixture were omitted from this analysis. Thus, table 4 permits comparison of treatments among cancer patients who were in pain at the time of the questionnaire. The patients seen at the pain clinic were all ambulatory and received a variety of oral analgesics and tranquilizers. Therefore, their pain levels were influenced by a variety of medications (such as codeine and dextropropoxyphene) and were presumably lower than they would have been in the absence of any medication. Nevertheless, the ward and private patients who received the Brampton mixture had consistently lower scores
723
MANAGEMENT OF INTRACTABLE PAIN IN PATIENTS WITH ADVANCED MALIGNANT DISEASE
than the cancer patients receiving other medications (and presumably in less advanced stages of the disease). The scores of the patients in the palliative care unit were lowest of all. Moreover, the Brompton mixture clearly produced low scores in all 3 dimensions of pain: 1) sensory, 2) affective and 3) evaluative. Morphine dosage. The average dose of morphine in the Brompton mixture was calculated for all the patients in the study. It was found that 78 per cent received morphine doses of less than 20 mg., while 22 per cent received 20 mg. or more. The mean dose for all patients was 12.3 mg. The average dose of morphine was raised usually if the pain was not brought under control. However, dosage was adjusted carefully according to the need and there was no evidence of habituation to the morphine, thus confirming Twycross' observation. 4 Once a satisfactory morphine dose was determined for a patient the pain scores remained virtually constant, with only minor fluctuations, rarely exceeding 1 present pain intensity interval. This constancy was noted particularly in patients who were observed for several weeks and one for as long as 40 weeks. Moreover, sequential assessments at 1, 2, Mean present pain intensity and pain rating index values for patients in palliative care unit, wards and private rooms receiving Royal Victoria Hospital oral narcotic mixture
TABLE 2.
3 and 4 hours after mixture administration showed no fluctuations in the pain. Once the adequate dose was determined and pain was under control the pain remained at the same level. This is consistent with observations of patients who received the Brompton mixture as long as a year without dose escalation. Adverse effects. The Brompton mixture was, in general, well tolerated. Sedation (usually transient), nausea, vomiting and constipation were the problems most commonly encountered, although the multiplicity of concurrent medical problems precluded attributing these symptoms to the mixture in all cases. Significant tolerance, dependence and respiratory depression were not seen. Domiciliary care. Several patients in the palliative care unit who took part in the study were permitted to go home once the pain and other medical problems were brought under control, and they used a home recording card8 to record the pain at prescribed periods 4 times a day. Usually, there was a remarkable consistency in the present pain intensity scores on the cards. If there was a change in the score it could be detected immediately by the visiting nurse. If the score showed a radical change the patient was rehospitalized immediately until the pain was brought under control (assuming there was no other major change in the patient's condition). DISCUSSION
Patient Category Variable Sex: Male Female Age (mean yrs.) Morphine (mean mg.)* Present pain intensity: Mean Range Pain rating index: Mean Range
Private (20 pts.)
Ward (46 pts.)
Palliative Care Unit (26 pts.) 11 15 53 13.3
25 21 54 13.2
7 13 59 10.5
0.3 0-2
1.1
0-4
1.3 0-3
1.0 0-13.6
5.6 0-29
6.9 0-28
* Mean dose of morphine in Brompton mixture. TABLE
3. Percentage ofpatients with mean pain scores within specific ranges* % Patients on Mixture With Pain
Present pain intensity scores: Palliative care unit (26 pts.) Ward (46 pts.) Private (20 pts.) Pain rating index scores: Palliative care unit (26 pts.) Ward (42 pts.)t Private (19 pts.)t
Unbearahie
NoPain
Mild
Discomforting
84 54 50
8 26 15
8 7 25
0 13
92 64 53
4 17 26
4 7 5
0 12 16
10
* No pain-present pain intensity Oto 0.9 and pain rating index Oto 5.9, mild pain-present pain intensity 1 to 1.9 and pain rating index 6 to 11.9, discomforting pain-present pain intensity 2 to 2.9 and pain rating index 12 to 17.9, and unbearable pain-present pain intensity 3 to 5 and pain rating index 18 to 29.9. t Smaller numbers because patients who were able to make a decision on their level of pain on a scale of O to 5 but whose vocabulary was too limited to complete the pain rating index portion of the questionnaire were excluded. TABLE
The current North American anxiety surrounding the use of narcotics for the intractable pain of advanced malignant disease is summarized as follows: "Giving narcotics is bad management. Narcotics give you 'ups and downs' producing addiction and destroying the personality. They depress cortical function." 11 Our experience suggests that with attention to detail none of these statements needs to be true. On the contrary, excellent pain control with continual pain prevention, an unclouded sensorium and a normal effect may be obtained in the majority of cases. The results show clearly that the Brompton mixture provides a powerful tool for the control of pain in most patients with cancer. Its effectiveness was evident in all 3 groups of patients and was especially impressive in the palliative care unit group. Since the doses of morphine and other ingredients were comparable for the 3 groups the reasons for the significantly greater effectiveness of the Brompton mixture in the palliative care unit group require explanation. We believe the primary reason was the psychological impact of the unit itself. The presence of a highly concerned staff and the help of volunteers who provided comfort and good cheer, as well as all the other amenities of the unit, must undoubtedly have had a strong psychological effect on the pain. Since the palliative care unit is comparable to St. Christopher's Hospice the data confirm Twycross' observations4 • 12 of the powerful analgesic properties of the Brompton mixture in the context of a highly supportive and comforting environment. There is convincing evidence that pain perception is not simply a function of the amount of physical injury but also is determined by the level of anxiety, expectation and other psychological variables. i:i Therefore, it is apparent that brain activities subserving these psychological processes are important in determining the quality and intensity of perceived pain. The results of this study are consistent with the gate-control theory of pain in
4. Mean present pain intensity, number of words chosen and pain rating index for cancer patients observed in an outpatient pain
clinic, palliative care unit, wards and private rooms
Pain clinic (16 pts.) Palliative care unit (5 pts.)* Ward (20 pts.)* Private (12 pts.)*
Mean Age (yrs.)
Mean Present Pain Intensity
Mean No. Words Chosen
56 53 54 59
2.8 1.0 1.5 1.6
8.8 2.9 5.1 4.8
Mean Pain Rating Index Sensory 17.3 3.6 6.1 6.4
Affective
Evaluative
Misc.
Total
2.3 0.5 1. 7 1.1
4.1 1.0 1. 7 1.4
2.3 0.3 2.3 2.0
26.0 5.4 11.7 11.0
* Patients (21 from palliative care unit, 26 from wards and 8 from private rooms) who reported total absence of pain are excluded from this analysis.
MOUNT, MELZACK AND MACKINNON
suggesting the interaction between psychological and physical factors. 14 The gate-control theory provides a conceptual model for incorporating these cognitive contributions to pain experience. The theory suggests that the amount of input transmitted from peripheral fibers to the brain is determined by sensory input from the body as well as by brain activities that exert a descending influence on the gate system. When the amount of input to the brain exceeds a critical level the ensuing pain experience comprises 3 dimensions: 1) the sensory-discriminative dimension of pain, 2) the powerful motivational drive and unpleasant effect that trigger the organism into action and 3) the evaluation or cognitive dimension. 15 Cancer pain is best understood within such a framework. Lesions of different kinds of tissues may produce pain that varies in intensity, duration and temporal pattern. Moreover, the pain is influenced by psychological factors, such as the knowledge of impending death, fear of disfigurement by the disease and worries about financial or family matters related to prolonged illness. 12 • 16 These factors affect the sensory, affective and evaluative dimensions of pain. 15 There is no reason to believe that the results showing greater effectiveness of the palliative care unit than care in wards or private rooms were biased by selection of the patients who received questionnaires. Roughly, 10 per cent of patients in all 3 groups had severe pain that could not be controlled by the Brompton mixture and had to receive additional or other treatment. Of the remaining patients the samples from the 3 groups were chosen at random. The mean ages of the patients, it was found at the conclusion of the study, were approximately the same and the mean doses of morphine also were about the same for the 3 groups. Although the dose of the phenothiazine was not regulated rigidly it varied within a narrow range, too narrow to explain the large differences between the palliative care unit group and the ward and private groups. The questionnaire provides objective measures of pain, and the high correlations between present pain intensity and pain rating index scores found in an earlier study8 also were found in this study. The validity of the results is further indicated by the fact that patients in the palliative care unit had less pain than the ward or private patients considered separately. Selection or sampling bias in these control groups was unlikely, since each came from hospital populations with different socioeconomic and educational backgrounds. Conversely, it is reasonable to assume that the patients in the palliative care unit did not have less pain initially than those in the other 2 groups. Indeed, the patients sent to the palliative care unit tended to be those with the most troublesome pain. These factors should have militated against greater relief in palliative care unit patients than in the 2 other groups. Therefore, the differences are even more striking and underscore the power of the combination of the Brompton mixture and psychological support provided by a hospital service as the palliative care unit. As a liquid the Brompton mixture has advantages over tablets for oral administration: 1) the dosage can be adjusted easily to meet the patient's need and 2) many patients have dysphagia, either functional or owing to local disease, and find a syrup easier to swallow than a tablet. ADVERSE EFFECTS
These basically are the adverse effects common to all narcotics and include the following. Sedation. When narcotic therapy is introduced transient sedation occurs frequently. The phenothiazine may exaggerate this effect. However, patients with advanced malignant disease often have other causes for somnolence (for example hepatic or renal insufficiency, or metastases). Nausea and vomiting. Routine use of a phenothiazine with
the mixture counters this common side effect of all narcotics. If a patient is vomiting before therapy is instituted control should be achieved first with parenteral medication and maintained subsequently with oral medication. Constipation. The combined effects of poor dietary intake, dehydration, inactivity and narcotic therapy almost invariably lead to constipation. This should be prevented by using a combination of a stool softener and a bowel stimulant (for example dioctyl sodium sulfosuccinate and senna concentrate). Tolerance-dependence. Evans17 and Twycross4 reported that dependence (addiction) is not a problem when narcotics are used for the pain of malignant disease. Marks and Sachar stated, "the excessive and unrealistic concern about the danger of addiction in the hospitalized medical patient is a significant and potent force for undertreatment with narcotics" .18 It would seem, rather, that undertreatment with analgesic medication may encourage craving and psychological dependence. Progressive tolerance and escalating dosage requirements are often given as reasons for delaying the onset of narcotic therapy. Our own experience confirms that of Twycross4 that a change in dosage requirements heralds a change in disease status rather than tolerance. Other adverse effects. Extrapyramidal effects, orthostatic hypotension and other side effects of the phenothiazines must be watched for but they occur infrequently with suggested doses. Because of the phenothiazine's synergism with morphine a small dose of the former is often sufficient. Although cocaine may be highly toxic to habitual abusers there is some question whether tolerance to cocaine develops. 19 The dose of cocaine used by us is similar to that used at St. Christopher's Hospice and is half the dose suggested in the "British Pharmaceutical Codex". 2 It has not led to important toxicity. Hypersensitivity reactions to morphine and to the phenothiazines are rare. When used as outlined above the Brompton mixture provides convenient and uniform pain control without important adverse effects. ADDITIONAL MEASURES
The Brompton mixture, to be effective against the total pain of advanced malignant disease, must be used in combination with other therapies. Symptom control may require additional measures, such as radiotherapy, peripheral nerve or intrathecal block, neurosurgery or physical measures, for example splinting and passive exercises. Tricyclic antidepressants, benzodiazepines, anti-inflammatory agents (for example phenylbutazone), corticosteroids and hypnotics can all be useful in attacking the vicious circle of chronic pain. The importance of environmental manipulation has been referred to already. In this regard the resources of an interdisciplinary team available to help in areas of interpersonal, psychosocial and philosophical need are most helpful. Used as outlined morphine, given in the form of the Brompton mixture with a phenothiazine, is a highly effective, flexible, safe and convenient means of controlling the chronic pain of malignant disease. Mr. J. Ofiesh, and Drs. I. Ajemian and J. Scott contributed to this study, and Ms. M. Hornby and Ms. J. Hymovitch provided assistance. REFERENCES
1. Catalano, R.: Medical management of pain caused by cancer.
Semin. Oncol., 2: 378, 1975. 2. British Pha=aceutical Codex. London: Pharmaceutical, p. 669, 1973. 3. Saunders, C.: The management of terminal illness. Hosp. Med., 1967. 4. Twycross, R. G.: Clinical experience with diamorphine in advanced malignant disease. Int. J. Clin. Pharmacol., 9: 184, 1974.
MANAGEMENT OF INTRACTABLE PAIN IN PATIENTS WITH ADVANCED MALIGNANT DISEASE
5. Twycross, R. G.: The equipotent dose ratio of diamorphine and morphine administered by mouth. Brit. J. Pharmacol., 46: 554, 1972. 6. Twycross, R. G. and Gilhooley, R. A.: Letter: euphoriant elixirs. Brit. Med. J., 3: 552, 1973. 7. Lipman, A. G.: Drug therapy in terminally ill patients. Amer. J. Hosp. Phann., 32: 270, 1975. 8. Melzack, R.: The McGill pain questionnaire: major properties and scoring methods. Pain, 1: 277, 1975. 9. Mount, B. M.: The problem of caring for the dying in a general hospital; the palliative care unit as a possible solution. Canad. Med. Ass. J., 115: 119, 1976. 10. Siegel, S.: Nonparametric Statistics for the Behavioral Sciences. New York: McGraw-Hill Book Co., 1956. 11. Drug and pain management, Leinbach composite videotape No. 3, Seattle, University of Washington, CCTV Services, 1974. 12. Twycross, R. G.: Principles and practice of the relief of pain in terminal cancer. Update, 5: 115, 1972. 13. Melzack, R.: The Puzzle of Pain. Harmondsworth: Penguins,
725
1973. 14. Melzack, R. and Wall, P. D.: Pain mechanisms; a new theory. Science, 150: 971, 1965. 15. Melzack, R. and Casey, K. L.: Sensory, motivational and central control determinants of pain: a new conceptual model. In: The Skin Senses. Edited by D. Kenshalo. Springfield, Illinois: Charles C Thomas Publisher, chapt. 20, p. 423, 1968. 16. Kubler-Ross, E.: On Death and Dying. New York: The Macmillan Co., 1969. 17. Evans, R. J.: Experiences in a pain clinic. Mod. Med. Can., 26: 7, 1971. 18. Marks, R. M. and Sachar, E. J.: Undertreatment of medical inpatients with narcotic analgesics. Ann. Intern. Med., 78: 173, 1973. 19. Jaffe, J.: Drug addiction and drug abuse. In: The Pharmacologic Basic of Therapeutics, 4th ed. Edited by L. S. Goodman and A. Gilman. New York: The Macmillan, Co., chapt. 16, p. 295, 1970.
THE
Vol. 120, December Printed in U.S.A.
JOURNAL OF UROLOGY
Copyright © 1978 by The Williams & Wilkins Co.
THE MANAGEMENT OF INTRACTABLE PAIN IN PATIENTS WITH ADVANCED MALIGNANT DISEASE B. M. MOUNT, R. MELZACK
AND
K. J. MACKINNON
From the Royal Victoria Hospital, Montreal, Quebec, Canada
ABSTRACT
The Brampton mixture is a highly effective, flexible, safe and convenient means to control chronic pain of malignant disease. The mixture is a solution containing morphine, the dose of narcotic varying with the need for analgesia, and is given regularly, usually every 4 hours, with a phenothiazine. The main aims of therapy are prevention of pain rather than treatment, an unclouded sensorium and a normal effect. Terminally ill cancer patients were given the Brampton mixture and a phenothiazine in an attempt to control their pain. The mixture was administered to patients in 3 hospital environments: 1) a palliative care unit, 2) general wards and 3) private rooms. Pain was measured in 92 patients with the McGill-Melzack pain questionnaire. The Brampton mixture controlled pain in 90 per cent of patients in the palliative care unit and in 75 to 80 per cent of patients in the wards or private rooms. The differences in pain scores between patients in the palliative care unit and the other groups were significant. The mixture produced substantial decreases in the 3 major dimensions of pain: 1) sensory, 2) affective and 3) evaluative. Comparison of these results with data obtained in an outpatient pain clinic showed that the Brampton mixture was strikingly more effective than the traditional methods of managing cancer pain. A significant percentage of patients with advanced malignant disease has intractable pain. With moderate to severe chronic pain only the narcotic analgesics provide adequate control. Milder analgesics should be tried for less severe pain and may be helpful in combination with more potent drugs. A wide variety of agents is available. 1 When it is localized moderate to severe pain may be managed adequately by irradiation, nerve block or some form of ablative neurosurgical procedure. In some situations, such as cases ofprostatic carcinoma, specific forms of therapy, including endocrine manipulation, hypophysectomy or radioactive isotope of phosphorus, may give transient relief. However, the response frequently is short lived and not experienced by all patients treated. We are all familiar with the all-too-frequent failures in pain control encountered in managing these unfortunate patients. THE BROMPTON MIXTURE
Background. Although generations of British physicians have gained familiarity with variants of the oral narcotic mixture bearing the name of the Brompton Chest Hospital it was not until 1973 that this formulation was recognized in the "British Pharmaceutical Codex". 2 The Brompton mixture is used for pain of advanced malignant disease when non-narcotic and milder narcotic preparations are ineffective. The important experience of Saunders and Twycross at St. Christopher's Hospice and St. Joseph's Hospice has led to a refinement and standardization of approach that have been associated with greatly increased effectiveness. 3• 4 On reviewing this experience with 418 patients who received narcotics for pain of malignant disease Twycross concluded 1) although most patients received parenteral treatment during . the last 12 to 48 hours because of increasing debility, the majority can be maintained on orally administered diacetylmorphine before this time, 2) there is no single optimal dose or maximum effective dose of diacetylmorphine, 3) the preAccepted for publication August 25, 1978. Supported in part by Grant A7891 from the National Research Council of Canada. 720
scription of diacetylmorphine does not, by itself, lead to impairment of mental faculties, 4) tolerance is not a practical problem, 5) psychological dependence does not occur and 6) physical dependence may develop but appears not to prevent the downward adjustment of the dose of diacetylmorphine when considered clinically feasible. 4 In a subsequent prospective randomized double-blind trial of great significance Twycross found that apart from an oral analgesic potential ratio of 1.5:1 (that is 15 mg. morphine is equivalent to 10 mg. diacetylmorphine orally) diacetylmorphine and morphine administered by mouth were equipotent and equitoxic. 5 Thus, the only advantage of diacetylmorphine is its greater solubility, which allows for smaller volume iajections in the occasional patient requiring parenteral nar0 cotics in high doses. Administration. The standard oral narcotic mixture used at the Royal Victoria Hospital is shown in table 1. The contributions to the effectiveness of this mixture of the small amounts of cocaine and ethyl alcohol are uncertain. Further elucidation must await the results of trials aimed at simplifying the mixture. For most patients analgesia is obtained with 5 to 20 mg. morphine per dose of the mixture but small or elderly patients may require as little as 2.5 mg. The usual sequential doses of morphine given are 2.5, 5, 10, 15, 20, 30, 40, 60, 90 and 120 mg. This standard elixir always is given with a phenothiazine. The phenothiazines are potent antiemetics and, thus, are useful in countering vomiting, a frequent side effect of narcotics. Experience also suggests that they potentiate the narcotic analgesia. Prochlorperazine, 5 mg. in 5 ml., is usually highly effective as an antiemetic with little sedative effect. If restlessness or agitation is a feature 10 to 25 mg. chlorpromazine may be substituted. Twycross and Gilhooley have shown a shelf life of sustained effectiveness that exceeds 8 weeks. 6 Lengthy anticipated survival is not a contraindication since the mixture may be used for many months to several years without dose escalation with care in adjusting the dosage to meet the patient's needs. In adjusting the dosage of narcotic and phenothiazine to
MANAGEMENT OF INTRACTABLE PAIN IN PATIENTS WITH ADVANCED MALIGNANT DISEASE TABLE
1. Royal Victoria Hospital oral narcotic mixture Morphine, 5 mg. or more Cocaine, 10 mg. Ethyl alcohol (98%), 2.5 cc Syrup, 5 cc Chloroform water Ad, 20 cc given with Prochlorperazine, 5 mg. or more (or chlorpromazine, 10 n,g. or.more)
achieve a pain-free state without sedation a number of factors are important. The morphine elixir should be given in 20 ml. doses with the phenothiazine every 4 hours around the clock because the serum half-life of morphine taken orally is about 4 hours. Occasionally, a patient may require a 3-hour schedule. The night dose is omitted only when the patient can sleep through the night free of pain. Careful attention to exact dosage and timing will pay dividends in results. For most patients a pain-free state can be achieved by giving sequential increments in narcotic dose. 7 To treat excruciating pain one may elect to start with a higher narcotic dose then make sequential decrements until analgesia without sedation is achieved. In determining the dosage required dose alterations should be made at intervals of 48 to 72 hours. During this period further analgesia can be achieved with the use of supplemental analgesics as required. In general, it is wise to change only 1 variable, the narcotic or the phenothiazine, at a time. Since the phenothiazines and morphine are synergistic great care must be exercised. Small changes in either variable may produce profound changes in analgesia and sedation. Initiation of narcotic therapy usually will produce transient sedation lasting 48 to 72 hours. It is important to reassure patient and family that pain can and will be controlled and that the initial drowsiness is temporary. Their confidence that control can be achieved will promote analgesia. Dispensing the morphine mixture and the phenothiazine syrup separately allows greater flexibility in adjusting dosage. Once a continuous pain-free state is achieved they may be combined in dispensing for greater ease of administration. Careful observation of the patient's condition during a complete 24-hour period may suggest augmentation of 1 or 2 specific doses at periods of peak activity. If parenteral medication becomes necessary the equivalent dose of morphine is half the previous oral dose. Thus, a patient whose pain has been controlled with 30 mg. orally would then receive 15 mg. intramuscularly. 4 MEASUREMENT OF PAIN
Because there have not been adequate tools to measure pain it has been difficult to gauge the magnitude of the effect of the Brampton mixture, its relative effects in different hospital environments or how it compares to traditional methods of pain control in cancer patients. The recent development of the McGill-Melzack pain questionnaire to measure clinical pain has given us an opportunity to examine these aspects of its effect on pain in cancer patients. 8 The 3 major indexes measured are present pain intensity or over-all pain intensity, measured on a scale of Oto 5 (0-none, 1-mild, 2-discomforting, 3-distressing, 4-horrible and 5-excruciating); number of words chosen from 20 sets of qualitative words with 2 to 6 words per set that describe the sensory, affective and evaluative properties of pain, and pain rating index, the sum of the rank values of all the words chosen (based on the positions of the words in each set (see figure). PROCEDURE
Each patient selected for the study was informed that he was being given a drug, the Brampton mixture, to control pain and that it was important to evaluate the intensity and kind of pain he felt in order to determine the proper dosage of
the drug. Each patient was instructed as follows: "I will read several groups of words that express feelings and sensations. If any of these words describe what you feel right now, please tell me and I will make a mark at the side of the appropriate word. Choose only 1 word in each group, the one that best expresses how you feel. Omit any groups that are not appropriate." The questionnaire generally was completed in about 10 minutes. At first more time was required to explain the meaning of some of the words. Occasionally, patients asked to have certain groups reread to be certain of the decision that a word was appropriate. When the patient felt no pain a zero was recorded and the list of descriptors was not read. PATIENTS
Palliative care unit. During this study 143 patients were registered in the palliative care unit at the Royal Victoria Hospital, a ward specializing in the care of the terminally ill. 9 Of these patients 90 received the Brampton mixture. Of the remainder many had relatively little pain, which could be kept under control with codeine or another mild or mediumstrength, orally administered analgesic. Others were unable to receive oral medication for a variety of medical reasons and, indeed, some died within hours of arrival. Of the 90 patients who received the Brampton mixture the pain could not be controlled in 8 (9 per cent), including 1 who had severe bladder spasms, 2 with sharp nerve-root pain that radiated into the legs and 5 with severe pain, a major component of which was their despair and anguish at their impending death. These patients were treated with additional or other methods in an attempt to achieve physical and psychological comfort. Of the 90 patients 82 received the Brampton mixture on a continuing basis and 26 were chosen at random to answer the pain questionnaire. The mean number of questionnaires given to the patients during their stay in the palliative care unit was 4. 7 (range 1 to 16). Wards and private rooms. During the study 238 patients on general hospital wards or in private rooms were given the Brompton mixture. The ward rooms consisted of 4-bed standard accommodations on the general medical and surgical wards. The private rooms consisted of single-bed accommodations. As in the palliative care unit group the pain required additional or other treatment in about 10 per cent of the patients. The remainder received the Brampton mixture on a continuing basis and 66 (46 ward patients and 20 private patients) were chosen at random to answer the pain questionnaire. The mean number of questionnaires given to the ward patients was 4.6 (range 1 to 20) and to the private patients, 7. 7 (range 1 to 49). While the standard Brompton regimen was followed in the palliative care unit setting there was greater variability in time and dose scheduling in the wards and private rooms. After the first 8 months of the study an analysis of the data indicated differences in pain control among the palliative care unit, ward and private patients. Consequently, it was impressed on the staff associated with all 3 groups of patients that it was essential to administer the Brompton mixture every 4 hours, even if the patient claimed he was not in pain or if it meant waking him to provide medication. RESULTS
Questionnaire scores. The mean present pain intensity and pain rating index scores for the palliative care unit, ward and private patients are shown in table 2. Pain scores for all 3 groups were remarkably low. Since a present pain intensity score of 1 represents mild pain it is clear that most ward patients had mild pain, private patients had mild to discomforting pain and patients in the palliative care unit had mild pain or none at all. The Brampton mixture, therefore, had a powerful effect on perceived pain in patients in all groups but
722
MOUNT, MELZACK AND MACKINNON
McGill - Melzack Pain Questionnaire Patient's Name
Analgesic(s)
Date Dosage Dosage
Time Time Given Time Given
Analgesic Time Difference (hours): +4 +l +2 +3 PRI: s A E M(S) M(AE) (1-10) (11-15) (16) (17-19) 1 FLICKERING_ QUIVERING PULSING THROBBING BEATING POUNDING 2 JUMPING FLASHING SHOOTING 3 PRICKING BORING DRILLING STABBING LANCINATING 4 SHARP CUTTING LACERATING 5 PINCHING PRESSING GNAWING CRAMPING CRUSHING 6 TUGGING PULLING WRENCHING 7 HOT BURNING SCALDING SEARING 8 TINGLING ITCHY SMARTING STINGING 9 DULL SORE HURTING ACHING HEAVY 1 0 TENDER TAUT RASPING SPLITTING
-
-
-
-
-
-
11 TIRING EXHAUSTING 12 SICKENING SUFFOCATING 13 FEARFUL FRIGHTFUL TERRIFYING 14 PUNISHING GRUELLING CRUEL VICIOUS KILLING 15 WRETCHED BLINDING 16 ANNOYING TROUBLESOME MISERABLE INTENSE UNBEARABLE 17 SPREADING RADIATING PENETRATING PIERCING 18 TIGHT NUMB DRAWING SQUEEZING TEARING 19 COOL COLD FREEZING 20 NAGGING NAUSEATING AGONIZING DRF.ADFUL TORTURING PPI 0 No pain 1 MILD 2 DISCOMFORTING_ 3 DISTRESSING 4 HORRIBLE 5 EXCRUCIATING
-
PPI _ _
(20)
am/pm am/pm am/pm M(T) _ _ _PRI(T) ( 17-20) (1-20)
fOMMENTS:
-
-
-
...,, ......
I\
-
-
-
r
CONSTANT PERIODIC BRIEF
1 ACCOMPANYING SYMPTOMS: NAUSEA HEADACHE DIZZINESS DROWSINESS _ CONSTIPATION _ DIARRHEA COMMENTS:
-
SLEEP: GOOD FITFUL CAN'T SLEEP COMMENTS:
INTAKE:
COMMENTS: S:
SOME LITTLE NONE
McGill-Melzack pain questionnaire, adapted for study of Brompton mixture. Descriptors fall into major groups: sensory- I to 10; affective-11 to 15; evaluative-16 and miscellaneous-17 to 20. Rank value for each descriptor is based on its position in word set. Sum of rank values is pain rating index (PRI). Present pain intensity (PPI) is based on scale of Oto 5.
a significantly greater effect on patients in the palliative care unit. The mean present pain intensity and pain rating index values for palliative care unit patients were significantly lower compared to ward and private patients, that is p values <0.05 to <0.001 in all cases as measured by 2-tailed tests based on the median test. 10 The percentages of patients with varying degrees of pain are shown in table 3. The mean present pain intensity and pain rating index scores for each patient were assigned to 1 of 4 categories: 1) none, 2) mild, 3) discomforting and 4) unbearable pain. The percentages of patients in each group are approximately the same on the basis of the present pain intensity and the pain rating index scores. On the basis of the present pain intensity it is apparent that all patients in the palliative care unit who received the Brompton mixture had bearable pain. Among the ward and private patients about 10 to 15 per cent of those taking the mixture continued to have unbearable pain. These data exclude the 9 to 10 per cent of patients in all 3 groups whose pain could not be controlled with the Brompton mixture alone and who required additional or other methods to achieve physical and psychological comfort. If these patients
are included in the over-all analysis it may be concluded that the Brompton mixture was effective in controlling pain in 90 per cent of patients in wards or private rooms. The effectiveness of the Brampton mixture is indicated further in table 4. In an earlier study all patients who attended the outpatient pain clinic at the Royal Victoria Hospital were requested to complete the pain questionnaires. 8 There were 16 patients who had pain associated with cancer. Their scores for present pain intensity, number of words chosen and pain rating index are shown in table 4 along with those of the palliative care unit, ward and private patients. All patients who obtained total pain relief with the Brampton mixture were omitted from this analysis. Thus, table 4 permits comparison of treatments among cancer patients who were in pain at the time of the questionnaire. The patients seen at the pain clinic were all ambulatory and received a variety of oral analgesics and tranquilizers. Therefore, their pain levels were influenced by a variety of medications (such as codeine and dextropropoxyphene) and were presumably lower than they would have been in the absence of any medication. Nevertheless, the ward and private patients who received the Brampton mixture had consistently lower scores
723
MANAGEMENT OF INTRACTABLE PAIN IN PATIENTS WITH ADVANCED MALIGNANT DISEASE
than the cancer patients receiving other medications (and presumably in less advanced stages of the disease). The scores of the patients in the palliative care unit were lowest of all. Moreover, the Brompton mixture clearly produced low scores in all 3 dimensions of pain: 1) sensory, 2) affective and 3) evaluative. Morphine dosage. The average dose of morphine in the Brompton mixture was calculated for all the patients in the study. It was found that 78 per cent received morphine doses of less than 20 mg., while 22 per cent received 20 mg. or more. The mean dose for all patients was 12.3 mg. The average dose of morphine was raised usually if the pain was not brought under control. However, dosage was adjusted carefully according to the need and there was no evidence of habituation to the morphine, thus confirming Twycross' observation. 4 Once a satisfactory morphine dose was determined for a patient the pain scores remained virtually constant, with only minor fluctuations, rarely exceeding 1 present pain intensity interval. This constancy was noted particularly in patients who were observed for several weeks and one for as long as 40 weeks. Moreover, sequential assessments at 1, 2, Mean present pain intensity and pain rating index values for patients in palliative care unit, wards and private rooms receiving Royal Victoria Hospital oral narcotic mixture
TABLE 2.
3 and 4 hours after mixture administration showed no fluctuations in the pain. Once the adequate dose was determined and pain was under control the pain remained at the same level. This is consistent with observations of patients who received the Brompton mixture as long as a year without dose escalation. Adverse effects. The Brompton mixture was, in general, well tolerated. Sedation (usually transient), nausea, vomiting and constipation were the problems most commonly encountered, although the multiplicity of concurrent medical problems precluded attributing these symptoms to the mixture in all cases. Significant tolerance, dependence and respiratory depression were not seen. Domiciliary care. Several patients in the palliative care unit who took part in the study were permitted to go home once the pain and other medical problems were brought under control, and they used a home recording card8 to record the pain at prescribed periods 4 times a day. Usually, there was a remarkable consistency in the present pain intensity scores on the cards. If there was a change in the score it could be detected immediately by the visiting nurse. If the score showed a radical change the patient was rehospitalized immediately until the pain was brought under control (assuming there was no other major change in the patient's condition). DISCUSSION
Patient Category Variable Sex: Male Female Age (mean yrs.) Morphine (mean mg.)* Present pain intensity: Mean Range Pain rating index: Mean Range
Private (20 pts.)
Ward (46 pts.)
Palliative Care Unit (26 pts.) 11 15 53 13.3
25 21 54 13.2
7 13 59 10.5
0.3 0-2
1.1
0-4
1.3 0-3
1.0 0-13.6
5.6 0-29
6.9 0-28
* Mean dose of morphine in Brompton mixture. TABLE
3. Percentage ofpatients with mean pain scores within specific ranges* % Patients on Mixture With Pain
Present pain intensity scores: Palliative care unit (26 pts.) Ward (46 pts.) Private (20 pts.) Pain rating index scores: Palliative care unit (26 pts.) Ward (42 pts.)t Private (19 pts.)t
Unbearahie
NoPain
Mild
Discomforting
84 54 50
8 26 15
8 7 25
0 13
92 64 53
4 17 26
4 7 5
0 12 16
10
* No pain-present pain intensity Oto 0.9 and pain rating index Oto 5.9, mild pain-present pain intensity 1 to 1.9 and pain rating index 6 to 11.9, discomforting pain-present pain intensity 2 to 2.9 and pain rating index 12 to 17.9, and unbearable pain-present pain intensity 3 to 5 and pain rating index 18 to 29.9. t Smaller numbers because patients who were able to make a decision on their level of pain on a scale of O to 5 but whose vocabulary was too limited to complete the pain rating index portion of the questionnaire were excluded. TABLE
The current North American anxiety surrounding the use of narcotics for the intractable pain of advanced malignant disease is summarized as follows: "Giving narcotics is bad management. Narcotics give you 'ups and downs' producing addiction and destroying the personality. They depress cortical function." 11 Our experience suggests that with attention to detail none of these statements needs to be true. On the contrary, excellent pain control with continual pain prevention, an unclouded sensorium and a normal effect may be obtained in the majority of cases. The results show clearly that the Brompton mixture provides a powerful tool for the control of pain in most patients with cancer. Its effectiveness was evident in all 3 groups of patients and was especially impressive in the palliative care unit group. Since the doses of morphine and other ingredients were comparable for the 3 groups the reasons for the significantly greater effectiveness of the Brompton mixture in the palliative care unit group require explanation. We believe the primary reason was the psychological impact of the unit itself. The presence of a highly concerned staff and the help of volunteers who provided comfort and good cheer, as well as all the other amenities of the unit, must undoubtedly have had a strong psychological effect on the pain. Since the palliative care unit is comparable to St. Christopher's Hospice the data confirm Twycross' observations4 • 12 of the powerful analgesic properties of the Brompton mixture in the context of a highly supportive and comforting environment. There is convincing evidence that pain perception is not simply a function of the amount of physical injury but also is determined by the level of anxiety, expectation and other psychological variables. i:i Therefore, it is apparent that brain activities subserving these psychological processes are important in determining the quality and intensity of perceived pain. The results of this study are consistent with the gate-control theory of pain in
4. Mean present pain intensity, number of words chosen and pain rating index for cancer patients observed in an outpatient pain
clinic, palliative care unit, wards and private rooms
Pain clinic (16 pts.) Palliative care unit (5 pts.)* Ward (20 pts.)* Private (12 pts.)*
Mean Age (yrs.)
Mean Present Pain Intensity
Mean No. Words Chosen
56 53 54 59
2.8 1.0 1.5 1.6
8.8 2.9 5.1 4.8
Mean Pain Rating Index Sensory 17.3 3.6 6.1 6.4
Affective
Evaluative
Misc.
Total
2.3 0.5 1. 7 1.1
4.1 1.0 1. 7 1.4
2.3 0.3 2.3 2.0
26.0 5.4 11.7 11.0
* Patients (21 from palliative care unit, 26 from wards and 8 from private rooms) who reported total absence of pain are excluded from this analysis.
MOUNT, MELZACK AND MACKINNON
suggesting the interaction between psychological and physical factors. 14 The gate-control theory provides a conceptual model for incorporating these cognitive contributions to pain experience. The theory suggests that the amount of input transmitted from peripheral fibers to the brain is determined by sensory input from the body as well as by brain activities that exert a descending influence on the gate system. When the amount of input to the brain exceeds a critical level the ensuing pain experience comprises 3 dimensions: 1) the sensory-discriminative dimension of pain, 2) the powerful motivational drive and unpleasant effect that trigger the organism into action and 3) the evaluation or cognitive dimension. 15 Cancer pain is best understood within such a framework. Lesions of different kinds of tissues may produce pain that varies in intensity, duration and temporal pattern. Moreover, the pain is influenced by psychological factors, such as the knowledge of impending death, fear of disfigurement by the disease and worries about financial or family matters related to prolonged illness. 12 • 16 These factors affect the sensory, affective and evaluative dimensions of pain. 15 There is no reason to believe that the results showing greater effectiveness of the palliative care unit than care in wards or private rooms were biased by selection of the patients who received questionnaires. Roughly, 10 per cent of patients in all 3 groups had severe pain that could not be controlled by the Brompton mixture and had to receive additional or other treatment. Of the remaining patients the samples from the 3 groups were chosen at random. The mean ages of the patients, it was found at the conclusion of the study, were approximately the same and the mean doses of morphine also were about the same for the 3 groups. Although the dose of the phenothiazine was not regulated rigidly it varied within a narrow range, too narrow to explain the large differences between the palliative care unit group and the ward and private groups. The questionnaire provides objective measures of pain, and the high correlations between present pain intensity and pain rating index scores found in an earlier study8 also were found in this study. The validity of the results is further indicated by the fact that patients in the palliative care unit had less pain than the ward or private patients considered separately. Selection or sampling bias in these control groups was unlikely, since each came from hospital populations with different socioeconomic and educational backgrounds. Conversely, it is reasonable to assume that the patients in the palliative care unit did not have less pain initially than those in the other 2 groups. Indeed, the patients sent to the palliative care unit tended to be those with the most troublesome pain. These factors should have militated against greater relief in palliative care unit patients than in the 2 other groups. Therefore, the differences are even more striking and underscore the power of the combination of the Brompton mixture and psychological support provided by a hospital service as the palliative care unit. As a liquid the Brompton mixture has advantages over tablets for oral administration: 1) the dosage can be adjusted easily to meet the patient's need and 2) many patients have dysphagia, either functional or owing to local disease, and find a syrup easier to swallow than a tablet. ADVERSE EFFECTS
These basically are the adverse effects common to all narcotics and include the following. Sedation. When narcotic therapy is introduced transient sedation occurs frequently. The phenothiazine may exaggerate this effect. However, patients with advanced malignant disease often have other causes for somnolence (for example hepatic or renal insufficiency, or metastases). Nausea and vomiting. Routine use of a phenothiazine with
the mixture counters this common side effect of all narcotics. If a patient is vomiting before therapy is instituted control should be achieved first with parenteral medication and maintained subsequently with oral medication. Constipation. The combined effects of poor dietary intake, dehydration, inactivity and narcotic therapy almost invariably lead to constipation. This should be prevented by using a combination of a stool softener and a bowel stimulant (for example dioctyl sodium sulfosuccinate and senna concentrate). Tolerance-dependence. Evans17 and Twycross4 reported that dependence (addiction) is not a problem when narcotics are used for the pain of malignant disease. Marks and Sachar stated, "the excessive and unrealistic concern about the danger of addiction in the hospitalized medical patient is a significant and potent force for undertreatment with narcotics" .18 It would seem, rather, that undertreatment with analgesic medication may encourage craving and psychological dependence. Progressive tolerance and escalating dosage requirements are often given as reasons for delaying the onset of narcotic therapy. Our own experience confirms that of Twycross4 that a change in dosage requirements heralds a change in disease status rather than tolerance. Other adverse effects. Extrapyramidal effects, orthostatic hypotension and other side effects of the phenothiazines must be watched for but they occur infrequently with suggested doses. Because of the phenothiazine's synergism with morphine a small dose of the former is often sufficient. Although cocaine may be highly toxic to habitual abusers there is some question whether tolerance to cocaine develops. 19 The dose of cocaine used by us is similar to that used at St. Christopher's Hospice and is half the dose suggested in the "British Pharmaceutical Codex". 2 It has not led to important toxicity. Hypersensitivity reactions to morphine and to the phenothiazines are rare. When used as outlined above the Brompton mixture provides convenient and uniform pain control without important adverse effects. ADDITIONAL MEASURES
The Brompton mixture, to be effective against the total pain of advanced malignant disease, must be used in combination with other therapies. Symptom control may require additional measures, such as radiotherapy, peripheral nerve or intrathecal block, neurosurgery or physical measures, for example splinting and passive exercises. Tricyclic antidepressants, benzodiazepines, anti-inflammatory agents (for example phenylbutazone), corticosteroids and hypnotics can all be useful in attacking the vicious circle of chronic pain. The importance of environmental manipulation has been referred to already. In this regard the resources of an interdisciplinary team available to help in areas of interpersonal, psychosocial and philosophical need are most helpful. Used as outlined morphine, given in the form of the Brompton mixture with a phenothiazine, is a highly effective, flexible, safe and convenient means of controlling the chronic pain of malignant disease. Mr. J. Ofiesh, and Drs. I. Ajemian and J. Scott contributed to this study, and Ms. M. Hornby and Ms. J. Hymovitch provided assistance. REFERENCES
1. Catalano, R.: Medical management of pain caused by cancer.
Semin. Oncol., 2: 378, 1975. 2. British Pha=aceutical Codex. London: Pharmaceutical, p. 669, 1973. 3. Saunders, C.: The management of terminal illness. Hosp. Med., 1967. 4. Twycross, R. G.: Clinical experience with diamorphine in advanced malignant disease. Int. J. Clin. Pharmacol., 9: 184, 1974.
MANAGEMENT OF INTRACTABLE PAIN IN PATIENTS WITH ADVANCED MALIGNANT DISEASE
5. Twycross, R. G.: The equipotent dose ratio of diamorphine and morphine administered by mouth. Brit. J. Pharmacol., 46: 554, 1972. 6. Twycross, R. G. and Gilhooley, R. A.: Letter: euphoriant elixirs. Brit. Med. J., 3: 552, 1973. 7. Lipman, A. G.: Drug therapy in terminally ill patients. Amer. J. Hosp. Phann., 32: 270, 1975. 8. Melzack, R.: The McGill pain questionnaire: major properties and scoring methods. Pain, 1: 277, 1975. 9. Mount, B. M.: The problem of caring for the dying in a general hospital; the palliative care unit as a possible solution. Canad. Med. Ass. J., 115: 119, 1976. 10. Siegel, S.: Nonparametric Statistics for the Behavioral Sciences. New York: McGraw-Hill Book Co., 1956. 11. Drug and pain management, Leinbach composite videotape No. 3, Seattle, University of Washington, CCTV Services, 1974. 12. Twycross, R. G.: Principles and practice of the relief of pain in terminal cancer. Update, 5: 115, 1972. 13. Melzack, R.: The Puzzle of Pain. Harmondsworth: Penguins,
725
1973. 14. Melzack, R. and Wall, P. D.: Pain mechanisms; a new theory. Science, 150: 971, 1965. 15. Melzack, R. and Casey, K. L.: Sensory, motivational and central control determinants of pain: a new conceptual model. In: The Skin Senses. Edited by D. Kenshalo. Springfield, Illinois: Charles C Thomas Publisher, chapt. 20, p. 423, 1968. 16. Kubler-Ross, E.: On Death and Dying. New York: The Macmillan Co., 1969. 17. Evans, R. J.: Experiences in a pain clinic. Mod. Med. Can., 26: 7, 1971. 18. Marks, R. M. and Sachar, E. J.: Undertreatment of medical inpatients with narcotic analgesics. Ann. Intern. Med., 78: 173, 1973. 19. Jaffe, J.: Drug addiction and drug abuse. In: The Pharmacologic Basic of Therapeutics, 4th ed. Edited by L. S. Goodman and A. Gilman. New York: The Macmillan, Co., chapt. 16, p. 295, 1970.