by Dr David J Sinclair with Dr Roy Eskapa and Michael Sinclair Explanation of alcoholism.
There is a widespread misunderstanding about how and when opioid antagonists such as naltrexone, naloxone and nalmefene suppress the craving for alcohol. The preclinical and clinical evidence reviewed here show that craving is not reduced simply by the presence of the antagonists in the body. Instead, they work by the mechanism of extinction. Alcohol drinking is learned through reinforcement involving the opioid system. Drinking alcohol while an opioid antagonist blocks the reinforcement starts extinction of the drinking behaviour and craving.
Evidence for this includes the fact that naloxone caused no reduction in the rate of lever pressing for alcohol by rats at the beginning of the first session. Similarly, prescribing naltrexone to abstinent alcoholics has not delayed significantly the resumption of drinking. Both rats and humans that are given opioid antagonists show little or no effect at first. Subsequently, both craving and drinking decrease progressively as a function of the number of sessions when alcohol was consumed while the antagonist was present. Almost all of the mean variation is explained by the theoretical extinction curve. Moreover, the motivation for alcohol remains suppressed long after all of the antagonist has been removed from the body. The results suggest that the primary influence of the antagonists on craving and drinking is not directly from the medicine itself but instead is produced by repeated extinction sessions in which alcohol is drunk while reinforcement is blocked by an opioid antagonist.
One of the strongest effects of using naltrexone in the treatment of alcoholism is the progressive decrease in the craving for alcohol. For example, the significance of the effect was p=0.00000000001 when we analysed the data from the first 210 patients at ContrAl Clinics (Figure 1 below). Patients themselves often comment after a couple months of treatment that to their surprise, they no longer are particularly interested in drinking. Previously, alcohol had been the focus of their lives, the main thing they were thinking about. Somehow the treatment had gotten rid of the obsession.
Perhaps because of these results, some people have gotten the mistaken idea that naltrexone itself reduces the craving. Alcoholics are writing in response to Roy Eskapa’s book (1), stating that they can only find doctors who will prescribe naltrexone for the purpose of suppressing craving while abstaining. A recent video on YouTube, although positive about naltrexone, said that it ‘blocks the craving and the high’. The practice of giving naltrexone in a targeted manner (ie. taking the medicine only on drinking days) is often described as telling patients to take naltrexone only when they need it to suppress a particularly high crabbing. Another clinician tells patients to take a double dose of naltrexone on days when they are having very high craving. Most of the theories proposed for how naltrexone works (eg. that alcoholism is caused by too little (2) or too much (3) opioidergic activity), aside from the initial one that it causes extinction (4, 5), predict that the antagonists directly block craving. Certainly one would expect that if the dose of antagonist was sufficiently high to make the person or animal feel different, ie. changing the stimulus situation from that in which alcohol drinking was learned there would be a direct reduction in craving before the first sip of alcohol was consumed.
As shown in this review, most of the available evidence suggests that naltrexone and other antagonists do not directly suppress craving. Or course, one cannot prove there is no such effect, perhaps the experiments have not been powerful enough to demonstrate it. One can conclude, however, that if there is any direct effect on craving, it is too small to be relevant clinically.
On the other hand, the reduction in craving during naltrexone treatment is one of the most powerful and remarkable effects observed. The craving definitely is suppressed, but it is suppressed after drinking alcohol while on naltrexone, that is, after the mechanism of extinction has had an effect. Before the first sampling of alcohol, however, the opioid antagonists produce no significant reduction in craving. The incorrect belief that opioid naltrexone alone blocks craving has probably played a major role in its being prescribed with instructions that make it ineffective, and thus in the low rate of prescribing the medicine.
The material here is from the annual presentations that Dr Sinclair made before the medical students of the University of Helsinki.
Dsm 5 definition of alcoholism
Responding during the first 2 minutes of the SECOND NALOXONE and THIRD NALOXONE session, however, reduced progressively. The motivation for alcohol was reduced after pairing of naloxone with lever pressing and alcohol drinking ie. after triggering extinction.
Published cumulative response patterns from two experiments with different doses of naloxone 30 minutes before access to lever pressing for alcohol also show that naloxone does not directly reduce motivation for alcohol. The AA rats began responding immediately for oral ethanol with naloxone having no effect on responding in either experiment during the first 2 minutes and no significant effect during the first 5 minutes. Higher doses of naloxone did, however, suppress responding later in the 1 hour sessions.
Clinical evidence with opiate addicts
The first naltrexone clinical trial was for treatment of opiate addiction (9). Once the patients began self-administering opiates whilst on naltrexone, the reported craving began decreasing and by the end of the trial, their craving was significantly lower than that reported by the patients receiving the placebo. Among the majority of patients who never self-administered opiates, however, naltrexone had no effect on craving. Naltrexone itself did not reduce craving for opiates.
The explanation given for these results was that naltrexone is thought to work through the mechanism of extinction. Extinction is the mechanism that weakens a learned response after a response is made and then is not followed by reinforcement. In the best known example, Pavlov’s dogs, conditioned to salivate to the sound of a bell by having a bell ring before getting food, had the conditioned response extinguished by not getting food after salivating to the bell. Extinction requires that the response first be made – and then not produced any reinforcement. Consequently, only the patients who made the response of taking opiates while naltrexone blocked the reinforcement had their craving weakened by extinction.
Our clinical results with alcoholics in regular practice
The daily intake of alcohol, reported in the drinking diaries of the first 210 alcoholics treated at the Tapiola and Tampere ControlAl clinics (not previously published) is shown in Figure 4. As in the pre-clinical results, the critical data is the effect on the first day of treatment with naltrexone. If naltrexone itself is able to reduce the patients’ craving to the extent that it is clinically relevant for drinking, then there should be a significant reduction in the number of drinks taken on this first day of treatment.
In fact, there was no reduction at all from the mean level of alcohol drinking during the two weeks before treatment. Indeed, the number of drinks consumed on the first day is slightly, although not significantly, higher than the pre-treatment mean. Naltrexone alone did not reduce craving and drinking.
Subsequently, after drinking has been paired with naltrexone and extinction could have an influence, alcohol consumption decreased progressively. The decrease is highly significant.
The most common protocol used in the clinical trials of treating alcoholism with naltrexone has been first to detoxify the patients, and then instruct them to abstain completely from alcohol while taking either naltrexone or placebo every day. If naltrexone directly reduced craving, the patients getting the antagonist would be able to abstain longer than those getting placebo.
Our database has 24 publications using this protocol with either naltrexone of nalmefene. Every one of them found no significant effect of the antagonist on the time to the first drink ie. no significant effect of the antagonist on clinically-relevant craving, (10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33). Reviews also conclude that the evidence for both naltrexone, (34) and nalmefene, (35), shows the antagonists are effective in preventing patients who are drinking from relapsing to heavy drinking, but do not prolong abstinence.
First drink priming from endorphins
In addition to extinction, there is at least one other mechanism through which naltrexone seems to suppress alcohol drinking: blocking of the ‘first drink effect’. The evidence for this is rather weak but one can see possible evidence in Figure 3. The rats in the SALINE session showed a second bout of level pressing for alcohol, approximately 20 to 30 minutes into the hour. The second bout is missing in each of the naloxone sessions, including the first, when the responding in the first 2 minutes was not reduced. A likely explanation is that effects from the first bout of drinking were stimuli for the second bout. One of the effects of the first bout was a release of endorphins, and the effects of the endorphins were among the stimuli for the second bout. Naltrexone blocks the effects of the endorphins, and this removes a stimulus for the second bout of drinking.
There is also anecdotal evidence from clinical practice. Patients frequently report, early in naltrexone treatment, that they are surprised to find that they did not finish the bottle, that they are satisfied by only a few drinks and do not bother taking more. Our explanation has been that the naltrexone is blocking the first drink effect. The one thing that always has been present when the second drink is taken is the effect produced by the first drink. This is generally a stimulatory effect, and it is has been shown in animals that the stimulatory effect of alcohol is blocked by opioid antagonists. (36)
Direct opioidergic stimulation of drinking
In addition to effects relating to learning, there is some evidence that endorphins may have a direct ability to stimulate alcohol drinking prior to any experience with alcohol. Figure 5 shows results obtained with AA rats given a single injection of morphine, the time was distributed throughout the day in order to preclude any circadian influences. A high peak in alcohol drinking was observed 4 hours after the morphine injection, regardless o the time of day when the morphine had been administered. This was followed by a suppression of alcohol drinking, keeping it significantly lower than the rats’ own prior level of intake for several days.
These results were observed in rats that had had prolonged prior experience with voluntary alcohol drinking. Almost identical results, however, had previously been seen in naïve Sprague Dawley rats ie. ones that had never had alcohol before the large morphine injection (30 mg/kp IP). (37). As in the alcohol-experiences rats, there was a very high peak in alcohol drinking during the fourth hour after receiving morphine, followed by a significant suppression, relative to controls, for the next 6 days.
These results indicate that opioidergic activation can stimulate alcohol drinking even without prior drinking experience.
Definition of alcoholic how many drinks
Theory for expecting direct blocking of craving
The learned ability of stimuli related to alcohol to elicit feelings of craving is known to be reduced after drinking alcohol while naltrexone is present. The theory proposing, however, that naltrexone alone can block craving starts with the assumption that when alcohol is consumed, and thus endorphins released, the stimuli present at the time would be classically conditioned to cause a release of endorphins by themselves. In analogy with Pavlov’s experiment, if you repeatedly heard a bell being rung before you drank alcohol, eventually the bell itself would cause endorphins to be released. The binding of these endorphins to their receptors or some effect from it would then be felt as a craving and would help to stimulate more alcohol drinking, just as the morphine in Figure 5 did and as the first drink is believed to work. Naturally, if the opioid receptors were blocked by naltrexone, the craving related to alcohol cues would not be produced and drinking would not be stimulated, just as naltrexone probably blocks the first drink stimulation of subsequent drinking.
The weakest part of the theory is the assumption that alcohol-related cues develop the ability to release endorphins. There is no doubt such cues are able to produce craving and that they help elicit drinking, but to my knowledge there has been no proof that they cause endorphins to be released. If the cues do not cause endorphins to be released, opioid antagonists could not affect whatever effects the cues have: specifically, opioid antagonists could not block craving elicited by alcohol-related stimuli.
Against the hypotheses that the cues release endorphins is the apparent failure of the cues to produce reinforcement. Does watching a bar or seeing advertisements for alcoholic beverages cause reinforcement? Probably not. It does not satisfy the desire for alcohol. In fact, there are other theories claiming such stimuli do not product pleasant reactions but instead produce unpleasant feelings similar to withdrawal.