The effect of the addition of a narcotic antagonist on the rate of development of tolerance and physical dependence to morphine

Sections

Plan
Deviations from and extension of plan
Case material
Results in main project
Results in collateral project
Acknowledgements

Details

Author: Nathan B. Eddy 2, Modeste Piller 3, Leo A. Pirk 4, Otto Schrappe 5, Sigurd Wende 6
Pages: 1 to 16
Creation Date: 1960/01/01

The effect of the addition of a narcotic antagonist on the rate of development of tolerance and physical dependence to morphine 1

Nathan B. Eddy 2
Modeste Piller 3
Leo A. Pirk 4
Otto Schrappe 5
Sigurd Wende 6

It was demonstrated by Isbell (1) that the daily administration of the narcotic antagonist nalorphine on a progressively increasing dosage schedule for twenty-eight days did not produce physical dependence in former opiate addicts. The subjects disliked the medication, and when it was abruptly withdrawn, signs of abstinence were not observed.

Lasagna & Beecher (2) found that in man nalorphine in single doses of 10 and 15 mg is a potent analgetic. Were it not for its unpleasant side effects, nalorphine might have proved to be the "non-addicting" analgetic so earnestly sought for a long time. These workers also reported that a mixture of 10 mg of morphine and 2 mg of nalorphine produced pain relief indistinguishable from that elicited by 10 mg of morphine. Thus, at the ratio used the narcotic antagonist did not interfere with morphine analgesia. Furthermore, Lasagna & Beecher suggested the possible usefulness of mixtures of this type in chronic pain, in the light of Isbell's findings that morphine and nalorphine in ratios of 10 to 1 and 5 to 1 produce a milder degree of physical dependence in former addicts than equivalent doses of morphine. This work, of which we had knowledge (3) at the time when we embarked on the programme to be described in this paper, included studies of the effects of chronic administration of such mixtures to post-addicts on an "addicting" and a "therapeutic" schedule. In the former case, narcoticantagonist combinations in proportions of 10 to 1, 5 to 1, and 3 to 1, respectively, were given q. 3-4 hours for 28-30 days in progressively increasing doses. Following withdrawal of such mixtures, the abstinence phenomena were much milder than those observed after comparable periods of addiction to morphine in the same patients. When "therapeutic" levels were used (10 mg of morphine and 1 mg of nalorphine given q. 4 hours for 31 days), the intensity of abstinence signs upon withdrawal was equal to that following abrupt discontinuance of morphine alone given in the same amounts to the same subjects for the same period of time.

1

The work to be described was carried out in New York, under the direction of Dr. Daniel Laszlo (deceased); in Hamburg, under the direction of Prof. Dr. Hans Burger-Prinz; and in Berlin, under the direction of Prof. Dr. Helmut Selbach. We are greatly indebted to these three principals for the guidance, help and encouragement they afforded us from the inception of the study to its end.

2

Chief, Section on Analgesics, Laboratory of Chemistry, National Institute of Arthritis and Metabolic Diseases, National Institutes of Health, Bethesda, Md., U.S.A.

3

Research Fellow, Division of Neoplastic Diseases, Montefiore Hospital, New York, N.Y., U.S.A. Dr. Piller's present address: F. Hoffmann-La Roche & Company, Ltd., Basle, Switzerland.

4

Director, Department of Clinical Investigation, Hoffmann-La Roche Inc., Nutley, N.J., U.S.A.

5

Assistent, Psychiatrische und Nervenklinik, Universitäts-Krankenhaus Eppendorf, Hamburg, Germany. Privat-Dozent Dr. Schrappe's present address: Universitäts-Nervenklinik, Marburg/Lahn, Germany.

6

Assistent, Psychiatrische und Neurologische Klinik der Freien Universität Berlin, Berlin, Germany

However, the effects of narcotic-antagonist mixtures administered in therapeutic doses to patients without previous narcotic experience had not been studied. This appealed to us as a worth-while programme, since it was conceivable that the rate of development of physical dependence would be different in former addicts and in subjects who had no narcotic experience.

One of the difficulties in studying morphine-nalorphine mixtures is the considerable difference in the duration of action of the two agents. The effects of nalorphine last for not more than three to four hours, but some of the effects of morphine persist much longer. Consequently, it seemed desirable to employ a narcotic antagonist which is distinguished by longer duration of action. Fraser & Isbell (4) reported levallorphan to have this quality, as compared with nalorphine. Therefore levallorphan was thought to be the more suitable agent for combination with morphine.

There were three groups of investigators who happened to be interested at the same time in studying the problem of whether the addition of a narcotic antagonist to morphine would alter the "addictiveness" of the latter. Two of these groups were European and one was American. They were brought together because of their mutual interest, 7 and agreed upon a common protocol which was formulated by one of us (N.B.E .), assisted by one of the other authors (L.A.P.), who was also responsible for the correlation of all data. The clinical work was carried out in Berlin, Hamburg and New York.

Plan

The protocol for all three centres (Berlin, Hamburg and New York) will be the same.

Selection of Patients

Individuals with persistent, fairly constant pain of sufficient severity to require parenteral narcotic medication, which will probably be needed for a prolonged period, will be selected. However, patients with cardiac or circulatory disease are to be excluded. Moreover, only subjects with little previous narcotic experience will be eligible for admission to the study - subjects in whom the presence of physical dependence has been ruled out by the negative outcome of a screening allyl test (see below).

7

We are indebted to F. Hoffmann-La Roche & Company, Ltd., of Basle, Switzerland, and Hoffmann-La Roche Inc., of Nutley, N.J., U.S.A. for initiating intercommunication of the interested groups and for financial support of the investigative programme which followed.

Drugs and Drug Administration

For pain relief two drugs will be used - viz, solutions of morphine sulfate 8 and of a 50 to 1 morphine-levallorphan (Lorfan ®) tartrate 8 mixture. These solutions will be supplied randomized under ten different code designations ( A to J) which need not be identically applied in the three centres. The meaning of the code will be unknown to all workers, so that the study will be carried out in a double-blind fashion. However, all investigators will be aware of the morphine concentration and the fact that each of the ten drug solutions furnished contains the same amount of morphine.

The coded drugs will be assigned to successive patients in alphabetical sequence. After ten individuals have been introduced, the eleventh subject will again receive drug A, and so on. Once analgetic medication is instituted in a given patient, he will continue to receive the same drug solution for the duration of the study.

The recommended dose of analgetic is 10 mg of morphine (alone or in combination with levallorphan) per 60 kg body weight, with a dose adjustment of 0.2 mg of morphine per 1.2 kg weight variation from 60 kg.

According to the patient's complaints and the relief obtained, the supervising physician will adjust, throughout the study, the dose of the assigned drug to give adequate analgesia. If 10 mg of morphine/60 kg body weight is less or more than satisfactory, more or less than 10 mg/60 kg will be administered. The principle to be followed at all times is the smallest dose which will give adequate relief. The dose will be raised only in accordance with this principle, and repeated only when recurrence of pain demands additional narcotic medication. Drug administration will be subcutaneous. Analgetic and sedative drugs other than the coded materials will be avoided as far as possible.

The identity of nalorphine (Nalline®) hydrochloride 9 supplied as commercially available, and of physiologic saline solution used in the allyl tests and placebo tests, respectively, will be known to the investigators, but not to the patients. The details of the administration of these tests will be described below.

Development of Tolerance

The total amount of morphine given to each patient per week will be calculated. An increase of the weekly narcotic requirements of a given patient may be due to a change for the worse in his clinical condition. However, such an increase may also suggest the development of tolerance. If there is considerably greater or faster increase of narcotic requirements in the entire group of patients receiving morphine only, as compared with the entire group of patients receiving the morphine-levallorphan mixture, this will indicate that the addition of the narcotic antagonist to morphine prevented, deferred or attenuated the development of tolerance. Conversely, levallorphan, on prolonged administration, might to some extent interfere with morphine prevented, deffered or attenuated the development of tolerance. Conversely, levallorphan, on prolonged administration, might to some extent interfere with morphine- induced analgesia, and a considerably greater increase of narcotic requirements in the entire groups of patients on the combined medication will be interpreted accordingly.

8

To be reffered to hereafter without salt designations.

9

See footnote 2.

Another attempt at determining the development of tolerance will be made by estimating the degree of each patient's pain before administration of both the initial and one weekly dose of the assigned drug and the degree of pain after these injections until the next analgetic dose is given. In carrying out these estimations, the patients' statements with respect to severity of pain will be recorded by a specially trained nurse observer. The degree of pain will be given the following numerical values: 4, for agonizing pain; 3, for severe pain; 2, for moderate pain; 1, for slight pain; and 0, for no pain. The degree of pain relief will be expressed at half-hour intervals as the difference between the degree of pain present before and that at the time of observation after narcotic medication. Subsequently, the total post-medication pain relief - the pain relief score - will be determined as sum of these differences. A diminishing pain relief score, with a maintained or increasing individual dose of assigned drug, could be indicative of development of tolerance.

Development of Physical Dependence: the Allyl Test

The development of physical dependence will be judged by the appearance of abstinence phenomena following subcutaneous injection of nalorphine (allyl test). The signs to be looked for in these tests and in the placebo tests (see below) and the point scores assigned to the individual abstinence phenomena are listed in table 1. These are based on the scoring system of Himmelsbach (5) with some modifications.

TABLE 1

Criteria for physical dependence

Abstinence signs
Symbol
Numerical value
Systolic blood pressure
BP
1 point for each 2 mm Hg maximal rise, total not to exceed 10 points
Rectal temperature
T
1 point for each 0.1% rise
Respiratory rate
R
1 point for each respiration per minute increase
Yawning
Y
1 point
Lacrimation
La
1 point
Rhinorrhea
Rh
1 point
Perspiration
P
1 point
Goose flesh
G
3 points
Mydriasis
My
3 points
Tremor
Tr
3 points
Restlessness
Re
5 points
Emesis
Em
5 points for each act of emesis
Nausea
Na
3 points

A total point score of 10 or more will be considered indicative of the presence of some degree of physical dependence.

The allyl test will be performed initially as a screening procedure (screening allyl test) before each patient's admission to the study and afterwards at two-week intervals (routine allyl tests). Except for the screening allyl test, which may be carried out at any time of day, the allyl tests and the placebo tests must be administered on all occasions and to all patients in the same time relationship (2 hours) to the preceding dose of analgetic.

The screening test will be carried out as follows. Systolic blood pressure, rectal temperature and respiratory rate will be determined, and about five minutes thereafter 3 mg of nalorphine will be injected. At about twenty minutes after antagonist administration, blood pressure, temperature and respiratory rate will be re-determined. From the time of the nalorphine injection the patient will be observed for other signs characteristic of the abstinence syndrome. If the total point score reflects the absence of physical dependence, and unless the side effects are very disturbing to the patient, a second (5 mg) nalorphine dose will be given twenty-five minutes after the preceding dose and the whole procedure repeated. Again, if the outcome of the test remains negative, a third (8 mg) nalorphine dose will be administered at the same time-interval, followed by the above-mentioned measurements and observations. Obviously, if the outcome of the screening allyl test is positive the patient should not be included in the study. All subsequent routine allyl tests will be carried out by giving just one 5 mg dose of nalorphine.

In order to avoid drawing the patient's attention to the unusual circumstances surrounding the allyl tests and also to be able to appraise the results in these tests adequately, placebo injections (1 ml. of saline solution) will be given bi-weekly on days other than those on which the effect of the antagonist is observed. In both the routine allyl and placebo tests systolic blood pressure, rectal temperature and respiratory rate will be measured five minutes prior to and at thirty-minute intervals after the administration of nalorphine and saline solution, respectively, for two hours (four post-injection measurements). In addition to these measurements, all other symptoms listed in table 1 will be looked for during the two-hour period of observation. In calculating the total point score, the differences between the maximal four post-injection measurements for blood pressure, temperature and respiratory rate and the corresponding pre-injection values will be taken into consideration.

If the average total point scores in the routine allyl tests in the group of patients on morphine alone increase more markedly or more rapidly, as compared with the group of patients on the morphine-levallorphan mixture, this will indicate that the addition of the antagonist to morphine interfered with the development of physical dependence.

Deviations from and extension of plan

In general, the details of the plan as outlined above were observed with but minor deviations. Of these the following are worthy of note.

Admission of Patients with Positive Screening Allyl Tests and with Previous Narcotic Experience

The total point score in the screening allyl test for patient K. C. (N.Y.5) 10 was 17 (see table 7). This score definitely indicates the presence of physical dependence, but since the patient's initial narcotic requirements were very low (see table 3), she was admitted to the study although this was contrary to the stipulations specified in the section "Plan ".

10

For the purpose of identification, patients in the Berlin, Hamburg and New York centres are designated "B.", "H.", and "N.Y." followed by consecutive numbers.

Similarly, the abstinence signs observed in the screening allyl test for patient E. A. (H. 12) included mydriasis, so that the presence of some degree of physical dependence could not be excluded. However, since the total score in this test was only 7 (see table 6), she too was admitted to the study. Since the initial narcotic requirements were low also in this case (see table 2), the data pertaining to the increase of weekly narcotic requirements of both these cases were utilized in the assessment of the development of tolerance, but the findings in the routine allyl tests were disregarded in the evaluation of the development of physical dependence.

Finally, patient Mc. S. M. (N.Y. 6) prior to admission to this study (main project) had received approximately 22 mg of morphine per day for almost five weeks, during which time she was included in the New York collateral project (see below). Since the total point scores of the routine allyl tests carried out during the latter project were all under 15 11 (see case N.Y.Coll. 7 12 in table 11), and since the patient required continuation of narcotic medication, she was transferred to this study. Again, this constitutes a deviation from the original plan according to which only subjects with little previous narcotic experience were eligible for admission to this study. Since this patient's initial narcotic requirements were low (see table 2), the data concerning the increase of her weekly morphine requirements were included in the evaluation of the development of tolerance. However, the results in the routine allyl tests recorded in the second and third weeks were considered to have occurred in the seventh and eighth weeks, respectively.

Gaps in Narcotic Medication

In a few instances there were gaps in the administration of narcotic medication because the drug was depleted. Pertinent details are indicated in tables 2 and 3.

Nalorphine Doses in Allyl Tests

The nalorphine doses in some of the screening allyl tests and of subsequent routine allyl tests were reduced as an added element of safety, because in some instances the antagonist doses used in accordance with the original plan precipitated rather severe side reactions, such as marked anxiety, dysphoria, shakiness and others. Therefore the following amended dosage plan was adopted in the New York centre midway in the study. In the screening allyl tests, 1, 3 and 5 mg of nalorphine were administered instead of 3, 5 and 8 mg. Furthermore, in the routine allyl tests 1 mg of nalorphine was injected initially. If the total point score, calculated as in the screening allyl tests, reflected the absence of physical dependence, a second (2 mg) dose of nalorphine was given.Thus, 1 mg of antagonist, possibly followed by 2 mg, was injected in place of a single 5 mg dose. This modification required a slight amendment of the time-table of the procedure.

11

Cf. paragraph "Indication for Presence of Physical Dependence" (below)

12

For the purpose of identification, patients in the New York collateral project are designated "N.Y. Coll.", followed by consecutive number designations.

TABLE 2

Individual doses (initially, at end of fourth week and finally), and weekly doses of morphine sulfate (M.S.) administered to patients in group I (morphine alone)

The amended regimen was applied in all allyl tests (screening and routine) done for patient K. C. (N.Y. 5) and in the routine allyl test done for patient M.A. (N.Y. 1) during the fifth week in which she was studied. Similarly, the nalorphine doses used in several of the screening and routine allyl tests carried out for some of the other New York patients and for several of the patients in the Berlin and Hamburg centres were lower than specified in the section "Plan ". Details of these changes, which were also made for safety reasons, may be gathered from tables 6 and 7.

Indication for Presence of Physical Dependence

It became evident from the observations made in the placebo tests that the injection of saline solution was accompanied by a rise of systolic blood pressure and respiratory rate in the majority of patients. In fact, the average point score following the placebo tests was 6.4 (see table 10). Consequently, it was decided to consider the outcome of any allyl test to be indicative of the presence of some degree of physical dependence if the total point score was 15 or more (not 10, as originally planned), provided such a score was contributed to by something more than changes in blood pressure and respiratory rate.

Chance Change of Medication in two Patient during Different Course of Treatment

Two patients in the Berlin centre - viz. D. F. and S. H. were admitted to the study twice. On their second admission- after a medication-free interval of 11 and 7 months,respectively- these patients were considered new cases. For their second course of treatment they were thus assigned the drug solutions which were next in the alphabetical sequence of the coded preparations. After decoding the solutions, it became evident that patient D. F. received morphine during the first course of treatment and morphine-levallorphan during the second course of therapy and that patient S. H. was given the drug mixture first, followed by morphine on his second admission. 13 Different medications were therefore used,merely by chance, in both instances on the two occasions.The minor deviation from the original plan to maintain a given patient on the same drug solution for the duration of the study turned out to be a fortunate one, since both cases thus served as their own controls in evaluating the problems under investigation.

Patient S.H., after another medication-free interval of 40 days, received a third course of treatment consisting of morphine alone. Since drug administration was not carried out in a double-blind fashion, the findings during this third course are not included in the tabular summaries covering the double-blind experiments, but they will be discussed in the section "Results".

13

Patient D. F. was designated B. 3 when he was on morphine and B.11 when he received the morphine-levallorphan combination. Patient S. H. was case B.15 when he was given the mixture and B.12 when he was on morphine.

Extension of Original Programme - Collateral Projects

The aforementioned third course of treatment administered to patient S. H. constitutes an extension of the original programme. Furthermore, double-blind studies, deviating in their design from the original plan, were carried out in the Hamburg and New York centres (collateral projects). Details of these studies are presented in two separate communications (6, 7) but a brief discussion of the New York collateral project is included with this report. In the latter programme, morphine alone or in combination with levallorphan (in the 50 to 1 ratio) was administered chronically to subjects who in all but one case did not need narcotic medication and in whom the presence of physical dependence had been ruled out by the negative outcome of a screening allyl test. In terms of morphine, single doses of 10 mg/60 kg with the necessary adjustments were given t.i.d. and the patients subjected at approximately weekly intervals to allyl tests. In general, treatment was continued for four weeks, followed by abrupt discontinuance of medication for observation of spontaneous withdrawal symptoms.

In the screening allyl tests, 3 mg of nalorphine were administered, with the exception of one subject who received three fractional doses of the antagonist (1, 3 and 5 mg) at intervals of 25 minutes. In the routine allyl tests, 3 mg of nalorphine were used for the first two tests and 6 mg for the third and fourth tests, with very minor deviations. Details of these changes may be gathered from tables 11 and 12.

Since the dose of morphine was standardized and since most patients did not require narcotic medication, the development of tolerance could obviously not be studied in this group of patients. The main objective of the New York collateral project was to obtain further data which would indicate whether or not the addition of levallorphan to morphine interferes with the rate of development of physical dependence.

Case material

Main Project

In the three centres a total of 29 patients were originally admitted to the study. However, 10 subjects were dropped from the programme at a very early stage of the investigation, mostly because in these instances the design of treatment could not be followed, for reasons beyond our control.

Thus, the total series consisted of 19 patients - 10 males and 9 females - whose ages ranged from 23 to 70 years (average 50.6 years). The numbers of subjects included in the Berlin, Hamburg and New York centres were 10, 3 and 6, respectively. Thirteen patients were suffering from cancer, and one patient each had an injury of the cervical spinal cord, chronic polyarthritis, a hypernephroma, stump pain, a hygroma of the left hand and syringomyelia.

As mentioned in the section "Deviations from and Extension of Plan ", the two patients in the Berlin centre who were admitted to the study twice were both considered as separate cases during the two courses of treatment they received in a double-blind fashion. Therefore the entire case material comprised 21 cases. After decoding the drugs, the patients were grouped together according to the medication they received. Thus, group I (9 cases) included the patients to whom morphine alone was administered and group II (12 cases) the patients who were given the morphine-levallorphan mixture. In the former group there were 4 males and 5 females (average age 54 years) and in the latter group 8 males and 4 females (average age 46.5 years).

New York Collateral Project

Fifteen patients - 10 males and 5 females - whose ages ranged from 32 to 75 years (average 57.8 years) were included in the above project. 14 Eight were suffering from cancer and one patient each had fibrous dysplasia of bone, polycythemia, peripheral neuropathy, spondylitis deformans, Paget's disease, osteoporosis and a Pancoast's tumour. As in the main project, the patients were grouped together according to the medication they received after the drugs had been decoded. Thus, group I of the collateral project (7 cases) includes the patients to whom morphine alone was administered and group II of the collateral project (8 cases) the patients who were given the morphine-levallorphan mixture. In the former group there were 4 males and 3 females (average age 57.3 years) and in the latter group 6 males and 2 females (average age 58.1 years).

Results in main project

Development of Tolerance

Tables 2 and 3 present a record of morphine administration in groups I and II, respectively. Included are: (1) individual doses of morphine given initially, at the end of the fourth treatment week and finally for each patient, and (2) the total amount of morphine administered to each patient each week.

As is seen from these tables, the final individual dose of morphine (average of all cases) was increased by 78.6% for the group of patients who received morphine alone and by 16.5% for the morphine-levallorphan mixture group. Considering only the cases which received the medications for 4 weeks or longer, the average percentage increase in the size of the individual doses at the end of 4 weeks was 59 for morphine and 15.1 for the mixture. For the same cases the narcotic consumption was on the average 104.6% greater in the fourth than in the first week for morphine, but only 17.8% greater for the morphine-levallorphan mixture. By the same token, there was less tolerance to the analgetic effect of morphine when this narcotic was combined with levallorphan. If the data pertaining to patient N.Y.5, who was admitted to the study in spite of the unquestionable presence of physical dependence, are excluded, the resulting figures differ only insignificantly from those presented above. Tolerance was judged to be definitely, probably or doubtfully evident or absent as listed below.

14

Case N.Y.Coll.7 was later included in the main project as case N.Y.6.

Tolerance
Group I-9 patients Morphine alone (6 patients carried through 4 weeks or longer)
Group II-12 patients Morphine plus levallorphan (9 patients carried through 4 weeks or longer)
At end of 4th week
   
Tolerance
   
Definitely evident
4 2
Probably evident
1 1
Doubtfully evident
-
2
Absent
1 4
Finally
   
Tolerance
   
Definitely evident
6 5
Probably evident
2
-
Doubtfully evident
-
1
Absent
1 6

As is seen, after 4 weeks of treatment tolerance was considered to be definitely evident in 4 of 6 cases (66.6%) which received morphine for 4 weeks or longer, but only in 2 of 9 cases (22.2%) which were on the morphine-levallorphan mixture for that period of time. Conversely, after 4 weeks of treatment tolerance was considered to be absent in only 1 of 6 cases (16.6%) which received morphine for 4 weeks or longer, but in 4 of 9 cases (44.4%) which were given the mixture for that period.

Furthermore, it appears that, at the end of treatment, tolerance was considered to be definitely evident in 6 of 9 cases (66.6%) on morphine and in 5 of 12 cases (41.7%) on the combined medication. Contrariwise, at the end of treatment tolerance was considered to be absent in only 1 of 9 cases (11.1%) which received morphine, but in 6 of 12 cases (50%) which received the mixture. This difference in the incidence of absence of tolerance in favour of the mixture becomes even more impressive if one considers that the average duration of therapy was 4.7 weeks for the patients on morphine, but 6.5 weeks for those who received the morphine-levallorphan combination.

Tables 4 and 5 present a record of the pain relief scores in groups I and II, as determined for each patient.

As can be seen, the pain relief scores fluctuated to a considerable extent during the course of treatment in some of the patients in both groups. Yet, it is obvious that by and large the scores are fairly consistent with the estimates of tolerance. Details may be gathered by comparing the data in tables 4 (group I) and 5 (group II) with the entries in the right-hand columns of tables 2 and 3.

Development of Physical Dependence

The point scores for the signs indicative of physical dependence for each patient and each allyl test (screening and routine tests) are presented in tables 6 (group I) and 7 (group II).Tables 8 and 9 summarize these scores per patient and per test for those cases in the two groups which received the screening and at least one routine allyl test. Thus, patients N.Y. 4 and B.13 who are included in tables 6 and 7, but for whom only screening tests were performed, were omitted from tables 8 and 9. The results of the placebo tests are presented in table 10.

Eighty-six screening and routine allyl tests were done in the course of this study. Of these, 58, 13 and 15 were carried out in the Berlin, Hamburg and New York centres, respectively. Of the 21 screening allyl tests performed, all were negative with the exception of those carried out for patients H. 12 (group I) and N.Y.5 (group II), as pointed out previously in this paper.

Of the 65 routine allyl tests performed, two were done for patient H. 12 and three for patient N.Y.5. As mentioned before, these five tests are excluded from the evaluation. The remaining 60 tests were distributed as follows: One, two, three, four, six, seven and eight tests each were done for three, five, two, three, one, one and two patients, respectively.

Of these 60 tests, 22 were done for seven cases in group I and 38 for ten cases in group II. As is seen in the right-hand columns of tables 3 and 9, physical dependence was demonstrated at some time during the observation period in all patients except case H.7 in the morphine group and cases B.5 and H.11 in the mixture group. Since patient H.7 had her last routine allyl test during the second week of treatment with small doses of morphine, the negative outcome of her last test has little significance. On the other hand, the negative outcome of the last allyl test of patients B.5 and H.11, which were done in the fourteenth and sixth weeks of treatment respectively, is certainly meaningful. Moreover, the average total dose of morphine administered up to the first score indicative of the development of physical dependence was almost 50% higher in the mixture group than in the group which received morphine alone.

Finally, if the average point scores calculated for treatment weeks in which three or more allyl tests were given and which are entered in tables 8 and 9 are amended by omitting cases H.12 and N.Y.5 and by transferring the scores for patient N.Y.6 in the screening and routine allyl tests to the fifth, seventh and eighth weeks, respectively, for the reasons given above, the following figures result:

 
Group I (Morphine alone)
Group II (Morphine- levallorphan)
Average total point scores of screening allyl test
4 5.1
Routine allyl test during
   
week 2
-
8
3 16.4 15.2
4 13.7 5.6
5 17.7 13.5
6
-
10.3
7 15.2 13.3
9
-
15.7

As is seen, the average total point scores of the routine allyl tests, calculated for the individual treatment weeks, are consistently higher in the morphine group than in the mixture group. These averages increase more rapidly and more markedly in the group of patients who received morphine alone. This becomes particularly evident during the fourth and fifth treatment weeks, but in the seventh week the average total point score for the morphine group is only slightly higher than for the mixture group. Again, these findings suggest that physical dependence develops sooner if the narcotic is given alone than if it is administered conjointly with the antagonist.

Cross-over Experiments

Development of tolerance. When patient D. F. (case B.3) was treated with morphine alone, his narcotic requirements, as compared with the first week, increased by 191% and 230% in the second and third weeks of treatment. On the other hand, when the same subject, after a medication-free interval of 11 months, was placed on morphine-levallorphan (case B.11) the percentage increase of his narcotic requirements in the corresponding weeks was only 63.4 and 102.4 and, during subsequent weeks of therapy, the percentage increase never exceeded 129.2 (sixth week), as may be seen from tables 2 and 3.

Similarly, when patient S. H. (case B.5) was treated with the morphine-levallorphan mixture, his weekly narcotic requirements never increased by more than 21.7% (fifth and sixth weeks); but when the same subject, after a medication-free interval of 7 months, was placed on morphine alone (case B. 12) the percentage increase of his narcotic requirements was 111.7, 135.2 and 205.8 in the second, third and sixth weeks, respectively (See tables 2 and 3).

These findings indicate that patient D. F. developed rapidly marked tolerance when he was on morphine alone and tolerance of a far less degree when he was on the narcotic-antagonist combination. Furthermore, it is obvious that patient S. H. developed very little tolerance, if any, to morphine when the narcotic was given in combination with levallorphan, but definitive and marked tolerance when he was on morphine alone.

Development of physical dependence

When patient D. F. (case B.3) had received a total of 1,920 mg of morphine alone during a 24-day period (average daily dose 80 mg) his second routine allyl test was positive. 15 When the same subject, after the medication-free interval, was placed on morphine-levallorphan (case B.11) and after he had thus been given a total of 4,470 mg of morphine in the form of the mixture during a 42-day period (average daily dose 106.4 mg), his sixth routine allyl test was still negative. 16 However, the following test became positive 17 after 5,920 mg of morphine in the form of the combined medication had been administered during a 55-day period (average daily dose 107.6 mg).

Similarly, when patient S. H. (case B.5) had received a total of 6,585 mg of morphine in the form of the narcotic-antagonist combination during a 98-day period (average daily dose 67.2 mg), his seventh and last routine allyl test was still negative. 18 When the same subject, after the medication-free interval, was placed on morphine alone (case B.12) and after he had thus been given 5,690 mg of morphine during a 58-day period (average daily dose 98.1 mg) his eighth and last routine allyl test became clinically positive. 9 The total point score was only 11, but this low value can be explained by the fact that this allyl test was done 36 hours after administration of the last morphine dose.

Chart I presents graphically the total doses of morphine (alone and in combination with levallorphan) which these two patients received up to the individual allyl tests during the courses of treatment carried out in a double-blind fashion. The chart also shows the outcome of these allyl tests. It is evident from these findings that the development of physical dependence is deferred in the same individuals if morphine is given conjointly with levallorphan.

15

See entry in table 6, fourth week, and chart I.

16

See entry in table 7, seventh week, and chart I.

17

See entry in table 7, ninth week, and chart I.

18

See entry in table 7, fourteenth week, and chart I.

19

See entry in table 6, eleventh week, and chart I.

Remarks about notes 9 to 13: The periods of treatment up to the allyl test as indicated above and in the chart are given as total number of days on which drugs were actually administered. In tables 6 and 7 the time of performance of these tests is listed by weeks. Some minor gaps in drug administration account for the seeming differences between the data in the chart and the entries in the tables.

GRAPH I

Outcome of allyl tests in two patients receiving morphine alone and morphine plus levallorphan in cross-over experiments

Full size image: 77 kB, GRAPH I

After a medication-free period of 40 days, patient S.H. was again placed on morphine, this time in a non-blind fashion. During a 56-day period he received a total of 6,710 mg of the narcotic (average daily dose 119.8 mg). In contrast to the two previous courses, morphine was withheld up to 24 hours following the eight allyl tests administered during this third course of treatment. Although these tests, carried out 2 hours after the last morphine dose, were negative by the point scores obtained, withdrawal symptoms (dysphoria, diarrhoea, polyuria, perspiration, restlessness, etc.) frequently occurred several hours after completion of the tests. Since these symptoms always disappeared with the administration of the next morphine dose, there can be no doubt that the patient developed physical dependence.

Results in collateral project

Development of Physical Dependence

The total point scores for the signs indicative of physical dependence for each subject and each allyl test (screening and routine) are summarized in tables 11 (group I) and 12 (group II).

Seventy-two screening and routine allyl tests were done in the course of this study. All 15 screening tests performed were negative. Of the 57 routine allyl tests 25 were done for the seven cases in group I and 32 for the eight cases in group II. As is seen from tables 11 and 12, ten tests in the former group (40%) and 7 tests in the latter group (21.8%) were indicative of the presence of some degree of physical dependence, as judged by a total point score of 15 or higher.

Summary of total scores in screening allyl tests (S.A.T.) and allyl tests (A.T.) during treatment with narcotic, recorded for patients in group I (morphine alone) 1

By the same criterion, physical dependence was demonstrated at some time during the observation period in five of the seven patients (71.4%) who received morphine alone, and in four of the eight patients (50%) who were given the morphine-levallorphan mixture.

Conversely, the allyl tests remained negative in two of the patients (28.6%) treated with morphine alone, but in four of the patients (50%) who received the mixture.

The distribution of the time of appearance of signs of physical dependence was as follows:

Numbers of patients with first point score indicative of physical dependence

Group
During treatment weeks
1
2
3
4
>5
Group II (morphine-levallorphan)
0 0 2 2 4
Group I (morphine)
1 1 2 1 2

All these findings suggest that physical dependence develops more readily if morphine is given alone than if it is administered together with levallorphan.

The average point scores calculated for treatment weeks in which more than three allyl tests were given are correlated below:

 
Group I (Morphine)
Group II (Morphine- levallorphan )
Average total point scores of: Screening allyl test
5.7 5.5
Routine allyl test during week
-
7.8
1    
2 9.3 6
3 13.6 10.8
4 14.6 15.5
>5
13.5 9.6

As is seen, the average total point scores of the routine allyl tests, calculated for the individual treatment weeks, are higher in the morphine group than in the mixture group, with the exception of the fourth week. It is obvious that the average scores in the morphine group show a somewhat earlier tendency to rise than the average scores in the mixture group, and this may be interpreted as an indication of earlier development of physical dependence in the former group. Conversely, the somewhat delayed rise of the average scores in the mixture group may be interpreted as an indication that the addition of levallorphan to morphine delays the development of physical dependence.

The assessment of the presence or absence of withdrawal symptoms following abrupt discontinuance of medication was not always in agreement with the findings in the last allyl test. These differences, which are difficult to explain, are described in detail in one of the separate communications (6).

Occurrence of Side Effects

Euphoria and dizziness occurred more frequently in the patients who received morphine alone and drowsiness and constipation were more often observed in the patients who were treated with the morphine-levallorphan mixture.

It as noteworthy that the distribution of side effects during the four treatment weeks was quite different in the morphine and mixture groups. In the former group the incidence of side effects decreased rapidly from week to week, but in the mixture group the occurrence of reactions was more evenly distributed throughout the treatment period. These observations may suggest that tolerance to the side effects of morphine develops more slowly if the narcotic is combined with levallorphan, but the possibility cannot be excluded that the incidence of side effects from levallorphan increases with continued administration of the antagonist.

Summary and conclusions

Morphine sulfate and a 50 to 1 morphine-levallorphan (Lorphan) tartrate mixture were administered under code designations in a double-blind fashion to 19 patients with chronic pain, most of whom suffered from cancer. Included were cases from three centres - viz. Berlin, Hamburg and

New York. With but minor deviations all patients were treated and all observations and interpretations made in accordance with a protocol formulated in advance.

Morphine doses, which were adjusted throughout the study to the smallest which gave adequate relief, were repeated when recurrence of pain demanded additional medication. A careful record of drug administration was kept. In addition to weekly narcotic requirements, weekly pain relief scores were determined for each patient as a check on the development of tolerance.

Before initiation of coded medication and in general at two-week intervals during such medication, each patient received nalorphine to detect the presence of physical dependence by precipitation of abstinence signs (allyl test). Point scores, based on the scoring system of Himmelsbach, were determined in these tests as well as in placebo tests which were given under the same conditions in alternate weeks.

Two patients in the Berlin centre were admitted to the study twice. As became evident after decoding the medications, one of these patients was given morphine on his first admission and the morphine-levallorphan mixture on his following admission, and in the second case it was the other way around. Thus, different medications were used, merely by chance, in both instances on the two occasions and these patients therefore served as their own controls. Since these subjects were considered as separate cases during their two courses of treatment, the entire case material comprised 21 cases, of which 9 were given morphine alone (group I) and 12 the morphine-levallorphan mixture (group II).

The averages of individual morphine doses, both finally and at the end of 4 weeks of drug administration, were considerably higher in group I than in group II. The average narcotic consumption was 104.6% greater in the fourth than in the first week for morphine, but only 17.8% greater for the morphine-levallorphan mixture.

Tolerance was judged to be definitely evident, both after 4 weeks of narcotic medication and at the end of treatment, in a much higher percentage of patients in group I than in group II. Conversely, tolerance was considered to be absent at these times in a considerably lower percentage of patients in group I than in group II. All these findings strongly suggest that the development of tolerance is deferred and/or attenuated by combining morphine with levallorphan.

Pain relief scores fluctuated considerably in some patients, but on the whole were fairly consistent with the estimate of tolerance.

Physical dependence was detectable by the allyl test sooner or later in almost all patients in both groups. However, the average total dose of morphine administered up to the first score indicative of the development of physical dependence was almost 50% higher for group II than for group I. Moreover, the average total point scores of the allyl tests increased more rapidly and markedly in group I than in group II. This is particularly evident in the fourth and fifth treatment weeks. All these findings suggest that physical dependence develops more readily if morphine is administered alone than if it is given combined with levallorphan.

In the cross-over experiments on two Berlin patients tolerance developed rapidly and markedly when the medication was morphine alone and to a far less degree (hardly at all in one patient) when the medication was the morphine-levallorphan mixture. One patient showed evidence of physical dependence after receiving 1.92 gm of morphine over a 24-day period and no sign of physical dependence after 4.47 gm of morphine as mixture in a 42-day period. Similarly, the second patient developed physical dependence after 5.69 gm of morphine in 58 days, but his allyl test was still negative after 6.58 gm of morphine as mixture in 98 days. Thus, these cross-over experiments corroborate the deferment or attenuation of tolerance and physical dependence when morphine is combined with levallorphan.

Furthermore, in the New York centre, morphine and the morphine-levallorphan mixture were administered in a double-blind fashion to 15 patients, most of whom did not require narcotic medication. The dose of morphine was standardized and in general treatment was continued for 4 weeks (N.Y. collateral project). Seven patients were given morphine alone (group I) and 8 the narcotic-antagonist combination (group II).

Before initiation of coded medication and approximately at two-week intervals during such medication, each patient received nalorphine to detect the presence of physical dependence, just as in the main project.

Signs of physical dependence appeared somewhat earlier in group I than in group II. Moreover, the average total point scores of the allyl tests in the former showed a slightly earlier tendency to rise than in the latter. These findings also suggest that physical dependence develops somewhat sooner if morphine is administered alone than if it is given conjointly with levallorphan.

In the morphine group the incidence of side effects decreased rapidly from week to week, but in the mixture group the occurrence of reactions was quite evenly distributed throughout the period of treatment. One possible explanation for these observations is that the addition of levallorphan defers the development of tolerance to the side effects of morphine.

Considering all data presented, there may be an advantage in the combination of morphine with levallorphan. However, the addition of levallorphan seems to defer the emergence of tolerance and physical dependence to morphine for a comparatively short period.

*
 
*
*

ADDENDUM

Since formulating this report we have learned from Drs. Deneau and Seevers through personal communication that they studied the physical dependence liability of mixtures of morphine and levallorphan in monkeys. The animals received subcutaneously 5 mg/kg of the narcotic in combination with different amounts of the antagonist q. 4 hours for 31 days. The ratios of morphine to levallorphan used were 2.5 to 1, 10 to 1 and 40 to 1. Upon withdrawal, abstinence syndromes occurred; but in each case these were less severe than is usually observed when even less morphine is given alone at less frequent intervals. These workers concluded that levallorphan definitely reduced the degree of physical dependence which otherwise would have developed to morphine. These findings in monkeys are in agreement with our observations in man that physical dependence may develop somewhat sooner if morphine is administered alone than if it is given together with levallorphan. When Drs. Deneau and Seevers administered morphine and levallorphan in a 1: 1 ratio to monkeys - the animals received 5 mg/kg of both drugs q. four hours - development of physical dependence was completely suppressed.

Acknowledgements

We wish to express our thanks to Dr. Gisela Zumpfe, who collaborated with Dr. Sigurd Wende in the Berlin phase of the study, and to Mrs. Vernice Vankinscot (Research Nurse), who devoted all her time to the observation of the patients studied at Montefiore Hospital in New York. The coded medications used were provided by F. Hoffmann-La Roche & Company, Ltd., of Basle, Switzerland, and Hoffmann-La Roche Inc. of Nutley, New Jersey, U.S.A.

References

001

ISBELL, H., Personal communication to one us (N. B. E.), followed by publication: Attempted addiction to nalorphine, Fed. Proc. 15, 442, 1956.

002

LASAGNA, L. & BEECHER, H. K., The analgesic effectiveness of nalorphine and nalorphine-morphine combinations in man , J. Pharmacol. and Exper. Therap. 112, 356-363, 1954.

003

ISBELL, H., Personal communication to one of us (N. B. E.)

004

FRASER, H. F. & ISBELL, H., Personal communication to one of us (N. B. E.), followed by publication: Morphine antagonists , Fed. Proc . 14, 340, 1955.

005

KOLB, L. & HIMMELSBACH, C. K.: Clinical studies of drug addiction: Critical review of withdrawal treatment with method of evaluat-

000

ing abstinence syndromes, Amer. J. Psychiat., 94, 759-799, 1938; and Himmelsbach, C. K., "Studies of certain addiction characteristics of (a) dihydromorphine (Paramorphan), (b) dihydrodesoxymorphine-D (Desomorphine), (c) dihydrodesoxycodeine-D (Desocodeine), and (d) methyldihydromorphinone (Metopon), J. Pharmacol. & Exper. Therap. 67, 239-249, 1939.

006

PILLER, M. & LASZLO, D., "Untersuchungen über den Einfiuss eines Opiatantagonisten (Lorfan) auf die Morphinwirkungen bei chronischer Verabreichung am Menschen ", Schweiz. med. Wchnschr., 90, 518-522, 1960.

007

SCHRAPPE, O., "Können Opiatantagonisten die Entwicklung von Physical dependence bei chronischer therapeutischer Morphinanwendung verhindern?" In preparation.