The use of meprobamate in the treatment of heroin withdrawal symptoms


Although addiction to narcotic drugs is common in many areas of the world, ethnological, economic and social factors present Hong Kong make this a problem of considerable importance in this locality. Prior to 1945 the use of narcotics in this area was confined mainly to the smoking of opium. An analysis of the amounts and manner in which opium was employed and the effects produced by its use suggests that many individuals did not consume sufficient opium to become severely addicted; and, in consequence, the use of opium did not constitute as grave a problem as that at present produced by the consumption of heroin. The smoking of opium has gradually been replaced by the use of heroin, which is now the chief narcotic employed in this area. This changehas been accompanied by a marked increase in the use of narcotics by members of the lower income bracket. Numerousaddicts spend between 75% and 90% of their income on heroin, and are, therefore, unable to afford adequate food, clothing and shelter for themselves and their families.


Author: Carl C. Gruhzit, C.O. Lee
Pages: 6 to 14
Creation Date: 1959/01/01

The use of meprobamate in the treatment of heroin withdrawal symptoms [1,] [2,] [3]

Carl C. Gruhzit Lecturer in Pharmacology, University of Hong Kong
C.O. Lee Medical Officer, H.M. Prison, Victoria, Hong Kong
From the Department of Physiology, University of Hong Kong, Hong Kong, B.C.C.

Although addiction to narcotic drugs is common in many areas of the world, ethnological, economic and social factors present Hong Kong make this a problem of considerable importance in this locality. Prior to 1945 the use of narcotics in this area was confined mainly to the smoking of opium. An analysis of the amounts and manner in which opium was employed and the effects produced by its use suggests that many individuals did not consume sufficient opium to become severely addicted; and, in consequence, the use of opium did not constitute as grave a problem as that at present produced by the consumption of heroin. The smoking of opium has gradually been replaced by the use of heroin, which is now the chief narcotic employed in this area. This changehas been accompanied by a marked increase in the use of narcotics by members of the lower income bracket. Numerousaddicts spend between 75% and 90% of their income on heroin, and are, therefore, unable to afford adequate food, clothing and shelter for themselves and their families.

Various reasons are given by addicts to explain their initial use of heroin. Some claim that they originally took heroin to relieve mild chest or abdominal pains. Others state that the use of heroin enables them to "work harder for longer periods of time". Still other addicts claim they were introduced to the use of heroin by "bad friends ". It is recognized that many of the factors which prompted an individual to begin the use of heroin are also responsible, in part at least, for the continued use of this drug. However, the occurrence of marked withdrawal symptoms following cessation of drug intake represents an additional factor which promotes the further use of heroin. It is felt that, if the withdrawal symptoms could be prevented or markedly alleviated, many individuals might be induced to discontinue the use of narcotics.

Many agents and methods have been employed in the attempt to decrease the severity of the abstinence syndrome (Goodman & Gilman, 1955). These include substitution therapy with methadone, the use of pyrogens, antipyretic analgesics, sedatives (such as paraldehyde, barbiturates or bromides), stimulants (caffeine or amphetamine), anti-cholinesterases, purgatives, anti-diarrhoeal agents, hypertonic glucose or sucrose,calcium or magnesium ions, blistering various hormonal preparations including insulin, parathormone, adreno-corticotropic hormone and cortisone. The use of so many different substances, many of which have opposing actions, indicates that previous methods of treatment failed to provide rapid relief of the withdrawal suffering.

Recently, both phenothiazine derivatives (Aivazian, 1955; McMahon, 1957; Friegdood & Ripstein, 1955; Hedqvist 1954; Sainz, 1957) and meprobamate (Thimann & Gauthier, 1956) have been employed to treat narcotic withdrawal symptoms. Unfortunately, the published results do not clearly indicate the value of these drugs in the treatment of the abstinence syndrome, since, in general, other drugs were administered simultaneously, and the symptoms, were not evaluated quantitatively; nor were adequate controls employed. For example, Friedgood & Ripstein (1955) used chlorpromazinc in the treatment of five morphine and two demerol (pethidine) addicts whose addiction had followed the medical use of these agents. They reported that the administration of relatively large doses of chlorpromazine following the abrupt withdrawal of the addicting drug prevented the " phenomena that constitute the abstinence syndrome". The patients slept most of the time during the first three days of therapy, and were given intravenous fluids and vitamins concurrently. Although it is likely that the marked "psychomotor-sedative" effect of chlorpromazine was of value, the lack of controls makes it difficult to determine the degree of benefit afforded by this treatment.

In Malacca a regimen has been adopted for the treatment of opium addicts which includes the use of chlorpromazine (McMahon 1957). In addition, phenobarbitone, butobarbitone, thiamine and amphetamine are employed in the treatment. Although the schedule utilized "proved adequate for most of the patients seen ", the lack of data regarding the severity of the symptoms, the use of multiple drugs and the omission of controls make it difficult to evaluate the effects of chlorpromazine in this study.

Thimann & Gauthier (1956) noted "moderate" improvement in three out of four heroin addicts treated with meprobamate after gradual withdrawal of the narcotic. Controlled studies employing placebos in heroin addicts were not reported, so it is difficult to ascertain whether the improvement noted was due to the effects of meprobamate or merely due to the passage of time. It was therefore deemed desirable to conduct a controlled study of the effect of meprobamate in the treatment of the abstinence syndrome.


Fifty-one male patients were included in this study. The individuals were selected during the initial medical examination of inmates shortly after their admission to H. M. Prison,Victoria, Hong Kong. Since approximately forty to fifty addicts are admitted to prison daily, rigid criteria could be adopted for the selection of subjects. Except for six healthy non-addicts employed to evaluate the effects of large doses of meprobamate in normal individuals, all subjects selected had been addicted to heroin for more than eighteen months.The maximum duration of heroin consumption encountered among these subjects was fifteen years, while the average period of use was six years and a half. Nearly half of the addicts selected had used opium for periods ranging between three and forty years (average, eleven years) prior to their employment of heroin.

All addicts selected spent at least HK $4 (5 s.) per day for heroin. The maximum daily expenditure by a member of this group was HK $15 (19 s.), while the average outlay for heroin by the addicts included in this study was about HK $7.50 (8 s. 6 d.) per day. The cost of heroin varies with the availability of the drug and the quality of the product. The various grades employed in smoking (No.2 and No. 3 powders) (" yee ho faan, saam ho faan") usually cost HK $5-7 per g. Chemical analysis of "medium grade" samples indicates that this material contains approximately 87%-92 % diacetylmorphine hydrochloride (Gruhzit, 1958). Therefore the "smokers" in this study used an average of about 1.1 g per day of this heroin salt. It is estimated that, in the process of smoking, between 50% and 75% of the heroin is absorbed (Gruhzit, 1958). More purified grades of heroin employed for intravenous use cost approximately HK $8-10 per g, but probably provide the user with a greater intake per dollar than that obtained by smokers who use cheaper grades of heroin.

The addicts who participated in this study had been in police custody and had appeared before the appropriate judicial official prior to their medical examination. Therefore, all addicts who were selected had abstained from the use of narcotics for 46-72 hours prior to the onset of treatment. Although this time lapse made the testing of the ability of drug therapy to prevent withdrawal symptoms infeasible, the marked abstinence syndrome which usually occurs within two to three days after heroin withdrawal facilitated the selection of patients.

Marked withdrawal symptoms were exhibited by all addicts selected for meprobamate evaluation. Individuals who met the criteria discussed and who also, in the opinion of the observers, appeared to be suffering from marked abstinence symptoms, were asked to describe their illness. Only those who voluntary mentioned the majority of the "major" symptoms were chosen for study. Although it is recognized that some otherwise suitable subjects may have been excluded by this method of selection, it was felt that "prompting" might result in false answers. After selection the patients were extensively questioned regarding the severity of their various symptoms.

As far as could be determined, all persons included in this study were, except for withdrawal symptoms, in relatively good health. Many of the addicts also suffer from tuberculosis, chronic nephritis, gastric ulcers and other major illnesses. Individuals who had a history of any major illness or who showed signs of any serious disease were excluded as subjects for this evaluation. As a further precaution, physical examination and routine urinalysis were done on all subjects.

Care was also taken to select, so far as possible subjects who would answer questions truthfully. Because many individuals prefer hospitalization to hard labour, and because individuals with even moderate withdrawal symptoms, desire treatment, there was a tendency for some of thoseinterviewed, to exaggerate their amount of heroin consumption. This was noted especially after this study had been in progress for several weeks, and the information had spread among the inmates that " kai pak faan yuen ", or "anti-heroin drug" was available. However, the veracity of the addicts could be checked in various ways. The amount of money claimed to be spent on heroin was compared with the individual’s income and family status. The average unskilled labourer earns about HK $3-6 per day. Therefore, unless he has an outside source of income, it is impossible for him to spend more than this amount on narcotics. Unfortunately, many of the addicts do spend up to 90% of their income on heroin, The hands and fingers of the individuals were also examined. carefully. Addicts who smoke heroin, as will be described later, usually have marked staining of the fingers and signs of frequent burns. The failure of addicts who claimed to smoke large amounts of heroin to exhibit these signs cast doubt on their truthfulness. The arms of addicts who employed heroin intravenously were examined for signs of sclerosed veins. Lack of sufficient damage to the veins excluded from participation in this study individuals who claimed to have taken heroin intravenously for several years. Since practically all addicts in this area who consume large amounts of heroin for long periods of time suffer a loss of weight ranging from 15 lb to 40 lb, obesity also disqualified a person as a subject in this experiment. Finally, although there is considerable variation in the severity of withdrawal symptoms among individuals who have consumed similar amounts of narcotics, experience gained in the observation of thousands of addicts during the withdrawal period enables one to estimate the severity of addiction. Individuals were not selected if there was significant discrepancy between the estimated and claimed intake of heroin.

Various methods were employed by the addicts for the administration of heroin. Approximately 10% of the subjects selected self-administered heroin intravenously. This percentage is at least twice that encountered in the general narcotic group in this area, but the increased expenditure and the increased severity of withdrawal symptoms which intravenous users experience influenced the selection of subjects.

Most of the remaining subjects either "chased the dragon" [4] (" chui loong ") or "played the mouth organ" (" chui hau kum "). Both of these methods of heroin usage are relatively similar. Several granules of No. 2 or No. 3 powder containing approximately 87%-92% diacetylmorphine hydrochloride are placed, together with approximately four times as much " base powder" (" dai faan "), on a trough formed from cigarette package tinfoil. The material is heated from below with matches or tapers and the ensuing fumes are inhaled. Originally a hollow tube about eight inches long and approximately one-quarter to three-eighths inches in diameter formed from bamboo or rolled-up paper was used to inhale the vapours which arose from the heated mixture. Because of the tendency of the molten mass to flow up and down the tinfoil trough, the source of the fumes also migrates. The moving trail of fumes resembles the classical Chinese concept of the undulating dragon's tail, and the term "chasing the dragon" is therefore employed to describe this method of smoking heroin.

From the addict's standpoint this method offers various advantages - the equipment required is easily obtained and cannot in itself be used to incriminate him. However, although it is estimated that the more adroit addict can inhale approximately 50%-75% of the fumes evolved from the mixture, the use of a narrow tube is relatively inefficient. A small matchbox cover is now usually substituted for the hollow tube, since its increased cross-sectional area facilitates the inhalation of a greater percentage of vapours. The resemblance of a matchbox cover to the musical instrument has resulted in this method's being known as "playing the mouth organ".

Since heroin granules have a tendency to char and decompose when heated, it is necessary to employ a base powder, or " dai faan ", as a vehicle when heroin is smoked in the methods described above. Barbitone or, less frequently, phenobarbitone are commonly employed as base powder. These substances are readily volatilized and inhaled together with the heroin fumes. Previous investigation (Gruhzit, 1958) indicates that, in general, insufficient amounts of the base powder are inhaled by the average heroin addict to produce concomitant barbiturate addiction. However, individuals who smoke largeamounts of heroin, and therefore employ large amounts of base powder, may have combinedheroin-barbiturate addiction. It is likely that the increased incidence of "non-epileptic" withdrawal convulsions which has been observed since the introduction of barbitone as a base powder is the result of barbiturate addiction.

The remainder of the individuals selected for study (less than 10%) combined heroin usage with cigarette smoking. Granules of heroin imbedded in the tip of a cigarette are volatilized and inhaled as the cigarette is smoked. The cigarette must be held in a vertical position to prevent the heroin granules from falling out. Because of this position the use of heroin in this manner is known as " firing the ack-ack gun " (" dah fay gay "). Recently, various modifications in this technique have been noted. Some individuals employ a pipe-type cigarette holder which allows the cigarette to remain in a vertical position without tilting the head. In addition, the use of relatively pure heroin granules has been superseded in part by the use of a "cut" mixture which contains a smaller percentage of heroin but readily adheres to the tip of the cigarette.

Although all the addicts included in this study employed one of the methods described above, other procedures for the utilization of heroin were used by some of the individuals interviewed. "Red pills" (" hung yuen "), containing a relatively small quantity of heroin together with a variety of other ingredients, are smoked or, quite rarely, taken orally. Occasionally one encounters an addict who injects heroin intramuscularly or subcutaneously. The relatively slow absorption occurring with oral, subcutaneous or intramuscular administration of heroin makes these methods less popular than intravenous injection or the inhalation of heroin fumes.

Following selection, the patients were hospitalized and given bed rest, routine diet and water ad libitum. A quantitative estimate of the abstinence syndrome was made by grading the severity of the withdrawal signs and symptoms as absent (0), minimal (+, -), mild (+), moderate (++), severe (+++) or very severe (++++). Two points were given for each "plus" associated with "major" withdrawal symptoms - i.e., those symptoms which produce great discomfort; nausea, vomiting, diarrhoea, abdominal pain, anorexia, insomnia, twitchings and headache. Each "plus" associated with a "minor" complaint - i.e., increased perspiration, increased pilomotor activity, lacrimation, coryza, sneezing, hoarseness, dyspnoea and a feeling of cold, was assigned a value of one point each. Although the latter group of signs and symptoms are useful in evaluating the severity of the abstinence syndrome, they do not cause serious discomfort. Weakness, especially of the extremities, also represents a major complaint during the withdrawal period; however, this symptom was not included in the quantitative evaluation of the abstinence syndrome, since meprobamate itself, in the doses employed, produces weakness of the extremities ( vide infra). Nocturnal seminal emissions may also cause the individual considerable concern. Their incidence, however, is relatively low, and this complaint, although noted, was therefore omitted from the quantitative evaluation.

Since withdrawal symptoms will abate merely with the passage of time, the effects of meprobamate administration were compared with those produced by the administration of placebos. Following the initial studies in which a suitable dose of meprobamate was determined, patients were selected at random to receive either meprobamate or placebo therapy. A "double blind" study - i.e., a study in which neither the evaluators nor the subjects are informed of the nature of the medication administered - was not employed owing to the lack of trained assistants and because those patients receiving meprobamate therapy could be readily recognized, not only by the alleviation of their symptoms, but also by the changes in the tendon reflexes which occurred ( vide infra). Drug therapy was initiated at approximately 10 a.m. on the day of selection and was continued at four-hour intervals (except for the 2 p.m. dosage) during the course of therapy, which normally lasted five days. Detailed analysis of the patients’ response to therapy was made twice daily, at 10 a.m. and 4 p.m. Interim reports by hospital orderlies, warders and dressers were made at more frequent intervals.

Various doses of meprobamate were employed in the first ten patients, in an effort to determine the regimen which would produce the maximal relief of withdrawal symptoms without inducing significant side effects. As a precautionary measure against hypersensitivity, the initial dose was limited to 0.4 g. Subsequent administration of 1.6 g. every four hours produced dramatic alleviation of the withdrawal symptoms in several patients, but the maintainance of this dosage schedule caused one of these individuals to become semi-comatose within forty-eight hours after the initiation of therapy. Meprobamate administration was discontinued and twenty-four hours later this patient appeared relatively normal, but experienced a recurrence of his withdrawal symptoms. Maintenance doses of 0.44-0.8 g every four hours produced marked relief in several patients, but only slight to moderaterelief in others. As a result of these initial studies, a regimen as selected which employed an initial dose of 0.4 g followed by a "priming" dose of 1.6 g and the subsequent administration of 1.2 g of meprobamate at four-hour intervals during the remainder of the five days of treatment. The 2 a.m. dosage was omitted if the patient was asleep. Fifteen addicts receiving this dosage were compared with a comparable group of fifteen patients who received placebos. Patients on placebo therapy received the same number of tablets looked and tasted like the meprobamate tablets but contained inert substances. Unless otherwise noted, all numerical evaluations of results are based on these thirty patients.


As shown in fig. 1, six hours after the initiation of therapy patients who had received meprobamate had approximately a 60% reduction in the severity of their abstinence syndrome. As would be expected, patients who received placebo therapy and its associated hospitalizationand bed rest also had some alleviation of their complaints. The average percentage recovery in the placebo group at this time, however, was less than a quarter of that present in the meprobamate group. Five days of placebo therapy were required to produce a similar degree of recovery as that observed after one day of meprobamate treatment. When the severity of specific withdrawal symptoms is compared in the two groups (table 1) it is apparent that meprobamate markedly alleviates those abstinence symptoms associated with the gastro-intestinal and neuromuscular systems. Daily measurements of body weight were made in all subjects in an effort to obtain, additional evidence of the relief of gastro-intestinal complaints.

FIG.1. Average severity of heroin withdrawal symptoms in patients receiving meprobamate or placebo therapy

Full size image: 17 kB, FIG.1. Average severity of heroin withdrawal symptoms in patients receiving meprobamate or placebo therapy


During the short period of hospitalization, the maximum weight change which occurred in any patient under study was less than 6 lb. Nevertheless, the average weight gain in the meprobamate group was nearly twice that in the placebo group.

The effect of meprobamate on respiratory symptoms was less dramatic. During the period of treatment there was no significant difference in the incidence or severity of sneezing and hoarsencess in the two groups. Coryza and "air hunger" were less intense in the group treated with meprobamate, but the degree of benefit was not marked. During the course of observation no significant variations in the resting respiratory rate were observed in patients who participated in this study. The severity of "air hunger ", which is experienced by most addicts during withdrawal, could not be closely correlated with changes in respiratory rate. This suggests that the shortness of breath is not due to abnormalities in pulmonary function, but rather involves higher brain centres.

Although headaches, when they occur, represent, from the standpoint of the patient, a serious withdrawal symptom, their incidence was not sufficiently great in either group to allow an evaluation of the effect of meprobamate on this symptom. It was likewise impossible to evaluate" the, effect of' meprobamate on perspiration. Normally, increased perspiration occurs during the withdrawal period. This Study, however, was conducted during the summer months, and even the investigators themselves showed hyperhydrosis.


Comparison of the effects of meprobamate and placebo therapy on the severity of heroin withdrawal symptoms.

6 hours 24 hours
6 hours 24 hours
Major symptoms
3.2 0.9 0.4 3.1 3.1 2.8
2.8 1.3 0.8 2.4 2.0 1.7
2.8 0.8 0.4 2.6 1.8 1.1
1.8 0.4 0.3 2.1 1.1 0.8
1.9 0.1 0.2 2.1 1.2 1.1
Abdominal Pain
2.0 0.6 0.6 1.8 1.2 1.0
2.3 1.0 0.7 2.2 2.1 1.8
0.4 0.3 0.3 0.5 0.4 0.2
Minor symptoms
1.3 0.6 0.4 1.5 1.2 0.7
1.3 0.8 0.6 1.2 1.0 0.9
1.3 1.2 0.8 1.3 1.3 0.9
1.0 0.8 0.7 1.0 0.9 0.8
1.4 0.8 0.7 1.4 1.2 1.1
2.3 1.3 0.7 2.0 2.1 1.9
1.2 0.6 0.2 1.0 0.9 0.7
1.8 1.6 1.3 1.7 1.6 1.4

Average of fifteen subjects in each group before treatment and six and twenty-four hours after initiation of therapy. Symptoms graded as absent (0), mild (1), Moderate (2), severe (3) or very severe (4).

Slight decreases (5-20 mm of Hg) in systolic and diastolic blood pressures occurred in approximately 75% of the addicts during the course of therapy. However, since the average decrease in addicts, who received placebos was similar to that observed in patients who received meprobamate, it is likely that the observed changes in blood pressure are due in part at least, to hospitalization and bed rest and not to a specific drug action. Although Himmelsbach (1936) noted that a slight rise in blood pressure frequently occurs during the abstinence syndrome, all individuals included in this study had initial blood pressure readings within the range usually observed in non-addicts of comparable age and racial characteristics. It is possible, however, that the observed decreases in blood pressure during the course of therapy might be due in part to the warning of the abstinence syndrome.

The meprobamate regimen adopted after the initial produced relatively few side effects when it was employed in the treatment of the abstinence syndrome. Three addicts complained of slight dizziness forty-eight hours after the initiation of meprobamate therapy. Two of them also were mildly ataxic. These side effects, which were not severe enough to cause a cessation of drug therapy, subsided during the period of continued meprobamate administration. One individual who had been selected, to participate in this study developed polyopia monopthalmica within twenty-fours hours following the initial dose of meprobamate. Although it is not known whether this condition represents bizarre withdrawal symptom or was due to meprobamate administration, it was deemed advisable to discontinue drug administration and to exclude this patient from further participation in the study. Normal vision returned approximately seventy-two hours later.

Most individuals who received meprobamate, for the treatment, of withdrawal symptoms showed mild evidence of somnolence. It is difficult, however, to estimate whether this increased tendency to sleep, was the direct result of the sedative effect of the drug or was the normal bodily reaction to the elimination of the withdrawal symptoms, which in some cases had preventedsleep for more than forty-eight hours.

One patient on placebo therapy developed an acute psychotic sate on the fourth day of hospitalization. During this period he attempted suicide and later became moderately violent. He had marked delusions of persecution and accused fellow inmatesof attempted extortion and of causing all his relatives to be incarcerated. There was, of course, no substance in fact to these claims. An attempt was made to transfer this patient to meprobamate therapy, but he refused to swallow the tablets. Paraldehyde was therefore used to sedate this patient temporarily and he was subsequently transfered to a psychiatric hospital, where he recovered in approximately one week.

Similar psychotic states have been observed in other untreated individuals during the withdrawal period. From a scientific aspect it is fortunate that this patient had received placebos rather than meprobamate therapy. It is possible that meprobamate therapy might have prevented the development of this psychotic, state; if, however, such a condition had occurred during the course of meprobamte therapy one might be prone to incriminate the drug as an agent which produces suicidal tendencies. In the light of this experience one must also wonder about the validity of statements which suggest that other tranquillizing drugs produce suicidal tendencies.

Four patients, not included in the numerical evaluation of the results,were treated with meprobamate because of convulsive states or fits which occurred during the withdrawal period. These individuals had clonic-tonic convulsions, with unconsciousness lasting from thirty to sixty minutes. Although these convulsions resemble to some extent those observed in grand mal epilepsy, these patients had no previous history of epilepsy. After the cessation of the initial convulsions these patients were given 1.2-1.6 g of meprobamate every four hours for periods up to five days. No further convulsions were observed in any of these patients. Although the relative infrequency and emergency nature of this condition prevented a controlled study employing placebos, previous experience indicates that these convulsive states usually recur frequently during the subsequent twenty-four hours even when barbiturates are employed as sedatives. The absence of additional convulsions following the administration of meprobamate gives additional suggestive evidence of the ability of this drug to alleviate withdrawal symptoms.

Two additional patients were treated with meprobamate because of acute psychotic symptoms attributed to narcotic withdrawal. Apparent recovery occurred within twenty-four hours following the administration of meprobamate. It is not possible, of course, to be certain that this drug therapy shortened the duration of these psychotic states; however, the psychotic symptoms which occur during drug withdrawal normally persist at least four or five days.

FIG.2. Comparison of the effect of meprobamate and placebo administration on the patellar reflex response during the heroin abstinence syndrome. Analysis of fifteen subjects in each group. Reflex response graded as absent (0), very sluggish (1), sluggish (2), normal (3), brisk (4), very brisk (5)

Full size image: 12 kB, FIG

The patellar, achilles, biceps and triceps reflexes were tested in all patients receiving meprobamate or placebo therapy. Marked differences were noted between the tendon reflex responses in these two groups (fig. 2). Initially both groups had hyperactive tendon reflexes which were especially noticeable in the lower extremities. In the placebo group these reflexes gradually returned towards normal during the period of hospitalization. In contrast, reflexes were usually hypoactive twenty-four hours after the initiation of meprobamate therapy. Maximum depression of reflex activity was noted after 48-72 hours of treatment. During this period the patellar reflex could not be elicited in five of the fifteen meprobamate patients even when reinforcement techniques were employed. In four patients a response was obtained only when reinforcement techniques were used. In all but one of the remaining individuals in the meprobamate group the patellar reflex response was markedly diminished.

Other stretch reflexes were also observed, although the changes were not as marked as those observed in the testing of the patellar reflex. The ankle jerk was absent or very sluggish in approximately 60% of the patients during the administration of meprobamate. The biceps and triceps reflex responses were usually sub-normal, but could still be elicited in all except three patients. In two of these the biceps reflex was absent, and in the third patient the triceps reflex could not be elicited. The depression of tendon reflex responses persisted during the course of meprobamate administration, although there was a slight tendency for reflexes to return towards, normal on the fourth and fifth days of therapy.

Because of' the ability of meprobamate therapy to depress reflex responses, it was thought possible that the failure of meprobamate to relieve the general weakness characteristic of the abstinence syndrome might be due to the substitution of a "drug-induced" weakness for that normally present during the withdrawal period. In an effort to substantiate this hypothesis, doses of meprobamate similar to those employed in addicts were administered to six normal volunteers. All of them developed marked weakness, dizziness and ataxia during the first twenty-four hours of drug administration. Meprobamate was discontinued at this time in two individuals who could no longer walk more than a few feet without support. The remaining four persons were able to tolerate these large doses of meprobamate for three days before it was necessary to discontinue this drug. The ankle and knee jerks were absent, and the biceps and triceps reflexes were markedly diminished in all these patients. Reflexes gradually returned to normal and the dizziness, weakness and ataxia disappeared within 72-96 hours following the cessation of meprobamate therapy.


Although the method of analysis employed in this study clearly indicates the beneficial effects of meprobamate in the treatment of withdrawal symptoms, it probably minimizes the actual value of this drug in the therapy of the abstinence syndrome. During the withdrawal period addicts suffer severe "mental torture ". It is difficult to describe and perhaps impossible to quantitate this symptom. De Quincey, in his Confessions of an English Opium Eater, gives some indication of the nature of the mental suffering experienced by addicts. Although we were unable to obtain such adequate descriptions from the subjects employed in this study, their remarks indicated that meprobamate therapy alleviated their mental distress. Most patients had experienced withdrawal symptoms several times previously during periods of incarceration. Subjects treated with, meprobamate stated that "the suffering was much less than on previous occasions and that "at other times the ‘confusion’ lasted about ten days, while now it was gone in less than two days". It is possible that mere hospitalization, and bed rest contributed to the benefit which these patients received. However patients on placebo therapy complained, " Why don't I feel better and why can't I sleep? The other people who are being treated have improved. What is wrong with me?." The patients on placebos, of course, did not know that they were being administered inactive tablets.

Although the use of a constant meprobamate regimen in the study facilitated the comparison ofthe effect of meprobamate and placebo therapy, it is recognized that a more flexible schedule would be advantageous when treating individual addicts. During the abstinence syndrome addicts show hyperactivity of the neuromuscular, gastro-intestinal and secretory systems. Meprobamate administration depresses these systems. The optimal dose of meprobamate, therefore, would be one which reduces the activity of these systems to normal, but does not produce hypofunction. The dose of meprobamate should therefore be graded in accordance with the severity of the withdrawal symptoms and the response of the individual to treatment. All patients selected for inclusion in this study had severe abstinence symptoms. Nevertheless, it is likely that those individuals who developed ataxia would have received significant relief with smaller doses of meprobamate. It is also probablethat certain subjects, especially those who injected large doses of heroin intravenously, would have experienced more rapid improvement if larger doses of meprobamate had been employed. Unfortunaltey, there does not appear to be any simple criterion on which to base meprobamate dosage. Within limits, however, it appearsthat the continued presence of hyperactive tendon reflexes indicates inadequate dosage and the occurrence of ataxia and dizziness suggests supra-optimal dosage. It should be noted that the average weight of the patients included in this study is significantly less than might be expected in a comparable group in some other areas of the world. It is possible that even larger doses of meprobamate would be required to produce similar results in other racial groups who have a significantly larger stature.

The duration of treatment in this experiment was limited to five days, since the majority of the symptoms had disappeared during this period and because it was felt that longer periods of meprobamate therapy, might result in meprobamate addiction (Lemere, 1956). One patient in the present study had grand mal type convulsions approximately twenty hours following abrupt meprobamate withdrawal. Although there were no sequelaefollowing this convulsive episode, it would appear preferable gradually to decrease the dose administered over a period of several days in order to prevent the possibility of meprobamate withdrawal symptoms

Adjunctive therapy was not employed in this study, because of the possibility that it might interfere with the evaluation of the effects of meprobamate. All the individuals selected for study were relatively healthy. Most addicts, however, have sub-standard diets,and it is likely that the anorexia which accompanies withdrawal produces further vitamin deficiencies which may contribute to a portion of the withdrawal symptoms, especially the generalized weakness. The concomitant administration of vitamin therapy should therefore prove beneficial in some cases. Other adjunctivetherapy appears unnecessary in most cases, since adequate meprobamate dosage quickly relieves the withdrawal symptoms.

Although a direct comparison was not made with other methods employed in the treatment of heroin withdrawal symptoms, previous experience and an evaluation of published results indicate that meprobamate appears superior to other types of therapy utilized for this purpose. In general, most methods employ hospitalization with abrupt, rapid or gradual withdrawal of the narcotic with or without prior substitution therapy. Supplementary measures, including the use of sedatives, antispasmodics and anti-diarrhoeal agents, are employed to treat the specific symptoms which occur. Although controversy, exists regarding the specific value of some of the agents employed in the symptomatic treatment of the withdrawal syndrome, it is apparent that many of these measures provide some degree of relief. With most of the regimens employed, however, significant withdrawal symptoms usually persist for several days to several weeks following the initiation of therapy. This is in marked contrast to the meprobamate therapy employed in this study which produced a dramatic reduction in the severity of withdrawal symptoms within six hours. Although Zucker, Machlin and Scott (1958) recently reported that " meprobamate is not of value as an adjunct in the management of the opiate withdrawal syndrome" thedose of meprobamate which they employed (1.6 g. per day) was only approximately one-fourth of that which was selected in the present study. Our initial findings indicated that even 4g per day of meprobamate may fail to provide adequate relief of the heroin, abstinence syndrome. It is likely, therefore, that the failure of Zucker et al. to observe a beneficial effect of meprobamate may be attributed to the dose which they employed. One cannot exclude the possibility, however, that methadone substitution therapy, which they used, may significantly alter the response to meprobamate.

Recently various phenothiazine derivatives have been employed in the treatment of addicts ( vide supra). Unfortunately most of the published results do not give sufficient data regarding the selection of patients, the severity of addiction, the dose schedule employed, the manner of evaluation, the degree of response or the controls employed. It is likely, however, that these "tranquillizing" drugs may produce results comparable to those observed with meprobamate. However, serious toxic reactions, including jaundice (Penber, 1955) and agranulocytosis (Hodges & Lazerte, 1955; Boleman, 1955), have been observed following the administration of phenothiazine derivatives. In contrast, the incidence of serious toxic effects with meprobamate is relatively low (Selling, 1955; Borrus, 1955; Lasagna, 1956) although various allergic reactions have been observed (Borus, 1957). In view of the effectiveness and relative non-toxicity of meprobamate it appears that this agent may represent the drug of choice in the treatment of the abstinence syndrome.

All addicts employed in this experiment had been withdrawn from heroin for more than forty-six hours. Although this circumstance facilitated the selection of subjects for study, it precluded the testing of the ability of meprobamate to prevent withdrawal symptoms. It is felt likely that meprobamate would be effective in preventing abstinence symptoms in heroin addicts if the administration of suitable doses was initiated at the onset of withdrawal. Arrangements are contemplated which will allow the testing of this hypothesis.[5]

From a sociological standpoint, however, the complete prevention of abstinence symptoms might prove detrimental. Although previous reports indicate that the relapse rate among treated addicts is high (Vogel et al., 1948), the experience of withdrawal symptoms might act as a slight deterrent to the resumption of the use of narcotics. The present study has been conducted during a relatively brief period of time (three months) and it is as yet impossible to evaluate the effect of meprobamate treatment on the narcotic relapse rate. Many of the addicts who have been interviewed express an apparently sincere desire to break the habit (" kai tuen yuun"). There are, however, many factors which make it difficult for a former addict to abstain from the use of narcotics following his release from confinement. Many individuals return to their old environment and associates and are under considerable pressure to resume their old habits. Nevertheless, it is felt that in many cases hospitalization and treatment, even with placebos, will act as a definite deterrent to the resumption of narcotic usage. It is likely that if one believes he has received treatment for the " drug habit " this belief will act as a crutch which will, especially in borderline cases, strengthen the will power sufficiently to resist the temptation to resume the use of narcotics.

The marked inhibition of tendon reflexes produced in humans following the administration of large doses of meprobamate may represent a similar type of action of this drug on synaptic transmission as observed in animal experiments. Various investigators (Berger, 1954; Hendley et al., 1954) have shown that in cats meprobamate inhibits polysynaptic reflexes, such as the crossed extension reflex, but has an insignificant effect on monosynaptic reflexes, such as the patellar reflex. Unfortunately, no convenient method was available to test the effect of meprobamate on polysynaptic reflexes in humans. It is possible that the ataxia observed in some of the addicts and in normal volunteers may have been related to the inhibition of polysynaptic reflexes. However, the occurrence of diminished or absent patellar reflexes without evidence of ataxia perhaps suggests that in humans meprobamate has a greater tendency to affect monosynaptic reflexes than polysynaptic reflexes.

Since the physiological basis of withdrawal symptoms is unknown, it is impossible to reach any definite conclusions regarding the mechanism of action of meprobamate in the treatment of withdrawal symptoms. Tatum et al. (1929) and Seevers & Woods (1953) have suggested that the withdrawal symptoms might be due to the persistence of the stimulating effects of narcotics after the depressant effects have worn off. This theory has been criticized on the basis that certain drugs, such as the barbiturates, which have no significant stimulating action, may also produce withdrawal symptoms. The stimulatory effects, however, may not necessarily be due to the direct action of the drug, but rather to various bodily compensatory mechanisms which tend to counteract the depressant effects of the drug. Compensatory adjustments occur frequently following the administration of drugs. These are perhaps most often associated with the use of hormones, where the mechanisms of compensatory action are relatively well understood. They also occur with drugs which alter the blood pressure. Reflex bradycardia and vasodilatation may follow the administration of hypertensive agents. In many cases the compensatory mechanisms persist for a considerable time following the withdrawal of the initial stimulus. Thus hypothyroidism is frequently observed following the cessation of thyroxine therapy. Hypotension may result following the stoppage of adrenaline or noradrenaline infusions. It is possible that the compensatory mechanisms may account, in part at least, for the tolerance which occurs with continued administration of narcotics, and the persistence of compensatory readjustment after the termination of the direct effect of the drug might be responsible for the occurrence of withdrawal symptoms.

Unfortunately the mechanisms, if they do in fact exist, involved in the production of compensatory readjustments following the administration of narcotics are unknown. It is tempting to theorize that they involve facilitation of synaptic transmission and that meprobamate is effective because it alters synaptic transmission (Berger, 1954). Although the effect of meprobamate on synaptic transmission in the spinal cord may play a significant role in decreasing twitchings, the relatively uniform suppression of many of the withdrawal signs and symptoms by meprobamate indicates that it must also involve other areas of the central nervous system. The relatively selective effect on meprobamate of the electrical activity of the ventrolateral nucleus of the thalamus, reported by Hendley et al. (1954), suggests that this area might be primarily involved in the action of meprobamate. However, doses of meprobamate comparable to those employed in the present study also produce electro-encephalographic changes in other areas of the brain resembling those obtained after the administration of secobarbitone, phenobarbitone or paraldehyde (Pfeiffer et al., 1957). The effect of meprobamate cannot be explained merely on the basis of its sedative action, because sub-hypnotic doses of barbiturates fail to produce significant relief of many symptoms, especially those involving the gastro-intestinal tract, which occur during withdrawal.


  1. The administration of large doses of meprobamate to fifteen hospitalized heroin addicts suffering severe withdrawal symptoms decreased the average severity of the abstinence signs and symptoms more than 60% within six hours. At the end of five days of meprobamate therapy the withdrawal symptoms had decreased to less than 10% of the initial values. Significantly less improvement occurred in a comparable group of addicts who received placebos.

  1. The administration of large doses of meprobamate to normal volunteers produced marked vertigo, ataxia and weakness. Vertigo and slight ataxia was observed in approximately 20% of the addicts who received meprobamate therapy.

  1. Meprobamate produced marked inhibition of the patellar, achilles, biceps and triceps reflexes in both addicts and normal individuals.

  2. It appears likely that meprobamate represents the drug of choice the treatment of the heroin abstinence syndrome.


ALVAZIAN, G. H. (1955) Dis. Nerv. Syst. 16, 57.

BERGER, F. M. (1954) J. Pharmacol. & Exper. Therap. 112, 413.

BOLEMAN, A. P., Jr. (1955) J. Am. Med. Ass. 157, 364.

BORRUS, J. C. (1955) Ibid. 157, 1596.

----- (1957) Med. Clin. N. Am., W. B. Saunders Co., Philadelphia and Londo, page 327

DENBER, H. C. B. (1955) J. Am. Med. Ass. 158, 586.

FRIEDGOOD, C. E., & RIPSTEIN, C. B. (1955) New Eng. J. Med. 252, 231.

GOODMAN, L. S., & GILMAN, A. (1955) The Pharmacological Basis of Therapeutics, 2nd ed., The Macmillan Co., New York, page 248.

GRUHZIT, C. C. (1958) United Nations Narcotic Bul. In press.

HEDQUIST, H. (1954) Svenska Lekartidn. 51, 1286.

HENDLEY, C. D., LYNES, T. E., & BERGER, F. M. (1954) Proc. Soc. Biol. & Med. 87, 608.

---- (1955) Fed. Proc. 14, 351.

HODGES, H. H., & Lazerte (1955) J. Am. Med. Ass. 158, 114.

LASAGNA, L. (1956) J. Chron. Dis. 3, 122.

LEMERE, F. (1956) J. Am. Med. Ass. 160, 1431.

----- (1956) Arch. Neurol. Psychiat. 76, 205.

McMAHON, J. E. (1957) Med. J. Malaya. 12, 373.

PFEIFFER, C. C., et al. (1957) Anm. N. Y. Acad. Sci. 67, 734.

SAINZ, A. A. (1957) Psychiat. Quart. 31, 275.

SEEVERS, M. H., & WOODS, L. A. (1953) Am. J. Med. 14, 546.

SELLING, L. S. (1955) J. Am. Med. Ass. 157, 1594.

TATUM, A. L., SEEVERS, M. H., & COLLINS, K. H. (1929) J. Pharmacol. & Exper. Therap. 36, 447.

THIMANN, J., & GAUTHIER, J. W. (1956) Quart. J. Stud. Alcohol. 17, 19.

VOGEL, V. H., ISBELL, H., & CHAPMAN, K. W. (1948) J. Am. Med. Ass. 138, 1019.

ZUCKER, A. H., MACHLIN, S. D., & SCOTT, W. (1952) Am. J. Psychiat. 15, 254.


We wish to thank Mr. C. J. Norman, Commissioner of Prisons, and Dr. D. J. M. Mackenzie, Director of Medical and Health Services, for their permission to conduct this study. In addition, we would like to acknowledge the kind co-operation of Superintendent T. G. Garner, Chief Officer E. M. Gernmell and the other officers and staff of H.M. Prison, Victoria, whose assistance materially facilitated this investigation Finally, we would like to express our appreciation to Senior Superintendents A. A. Shaw and T. Cashman and Superintendents A. A Baggott and J. F. Ferrier, of the Police Anti-corruption and Narcotics Bureau, for their aid in the investigation of many facts of the addiction problem.


The meprobamate (" Miltown ") and placebos employed in this study were kindly supplied by the Lederle Laboratories Division of the American Cyanamid Company, New York.


The opinions expressed in this paper are those of the authors, and do not necessarily reflect the views of the Hong Kong Government.


See Bulletin on Narcotics; vol. X, NO. 3.


Preliminary studies now being conducted indicate that meprobamate is capable of preventing many of the abstinence signs and symptoms in heroin addicts undergoing withdrawal.