In an article appearing in JAMA in August 2005 entitled, “Anesthesia-Assisted vs Buprenorphine- or Clonidine-Assisted Heroin Detoxification and Naltrexone Induction,” authors Collins, Kleber, Whittington and Heitler conclude that “rapid opioid detoxification with opioid antagonist induction using general anesthesia has emerged as an expensive, potentially dangerous, unproven approach to treat opioid dependence”. On closer look at their methods, results, and interpretations it is clear that this is in fact not the case at all. They further claim that their own data “do not support the use of general anesthesia for heroin detoxification and rapid opioid antagonist induction.” This article will show that the authors failed to demonstrate the conclusions they claim and that in fact when performed using strict guidelines and with appropriate safeguards, anesthesia-assisted rapid opiate detox is a safe, cost-effective method that is a valid alternative for many patients with opiate addiction.
Using the Waismann Method, the procedure as done in our center is safe and effective and addresses the need for a rapid, humane, and well-tolerated opiate reversal method. Our results for our more than 1700 patients using our method (The Waismann Method of rapid detox) demonstrate no life-threatening adverse events during the procedure. One patient had aspiration pneumonia and recovered completely with antibiotics and chest physiotherapy. At 12 months following Anesthesia-Assisted Detox, nearly 70% of our patients remain opiate free, as per Waismann Institute reports (unpublished), based upon patient self-reporting.
As the authors of the Collins, Kleber, et al paper correctly point out, addicted patients and the general public feel a sense of “desperation” about the treatment of opiate dependence. Traditional methods have failed to provide a humane, well-tolerated and effective means of reversing opiate dependence. Drop out rates from traditional methods are extremely high, up to 30% in some reports (Scherbaum). Success rates as defined by long-term abstinence rates after detoxification are disappointing. Patients and their families are eager for the availability of a safe, short, comfortable and effective method of detoxification. We agree that it is imperative that treatment modalities such as anesthesia-assisted opiate detox are subject to rigorous review and are held to the highest standards, however, we question the methods and conclusions of the authors of this study.
With the goal of trying to compare the safety and efficacy of anesthesia-assisted rapid detox with other medication based methods of rapid opiate reversal, the authors of the 2005 study compared anesthesia-assisted to buprenorphine-assisted detoxification and to detox with clonidine alone. Given the unrestricted grant from Reckitt Benckiser, the manufacturer of buprenorphine, provided to Dr Kleber, who was one of the principal investigators of the study, it is hard to imagine an unbiased approach to the study outcome.
For the study a total of 105 heroin-dependent patients were randomized into three inpatient withdrawal treatment groups. Patients were initially included if they simply reported “general good health” and could be classified in the American Society of Anesthesiologists physical classification status I or II, ie “otherwise healthy, no other medical problems” for class I, or “a chronic medical condition that is well-controlled, eg, hypertension, diabetes, asthma, for class II”. It was only after three patients had serious complications during anesthesia assisted detox that inclusion and exclusion criteria were more appropriately defined and restricted.
Patients in the study were randomized into three treatment arms: Patients in the anesthesia-assisted group were maintained under anesthesia for 4-6 hours on the morning of day 1 after a pre-anesthesia dose of clonidine (0.3 mg) and were given a 50 mg dose of naltrexone induction during anesthesia, as well as nalmefene 4 mg over 30 minutes. This was followed by maintenance doses of naltrexone beginning the following day (50 mg daily); the buprenorphine group received an initial sublingual buprenorphine dose on day 0, and began naltrexone induction on day 2; the clonidine group was given no anesthetic agents and no opiate antagonist at all during their inpatient phase and did not begin naltrexone until day 7. Following hospital discharge, all patients were followed for 12 weeks. During the outpatient phase of the study for all three arms, every patient was given daily naltrexone (50 mg) and twice weekly relapse-prevention psychotherapy. At these visits, urine samples were collected for toxicology.
The study reported 3 serious adverse events occurring in the anesthesia arm of the study. However, a closer look fails to identify a causal relationship between the Anesthesia-Assisted Opiate Detox and the event. One patient had severe pulmonary edema and aspiration pneumonia necessitating reintubation and intensive care unit management for 5 days. The patient subsequently admitted a prior history of “several prior complicated pneumonias and of possible obstructive sleep apnea”. Following this experience, the investigators recognized that these conditions must be exclusionary for anesthesia-assisted detox. In addition, with appropriate anesthesia monitoring and maintenance of patient ventilation, severe pulmonary edema should not occur. This was not a complication of Anesthesia-Assisted Rapid Opiate Detox, but rather one of anesthesia technique and monitoring. A second patient developed suicidal ideation requiring psychiatric hospitalization 5 days after anesthesia and subsequently revealed a prior diagnosis of bipolar disorder. Again, not only was this occurrence not a complication of the Anesthesia-Assisted Rapid Detox, but significant psychiatric history was then recognized as sufficiently exclusionary for undergoing the procedure. In a third adverse event, an insulin-dependent diabetic presented a difficult glucose management problem post-anesthesia, delaying hospital discharge by one day, and then developed diabetic ketoacidosis (DKA) two days following discharge. A prior history of DKA was elicited that had previously been concealed. Once again, this experience led the investigators to revise their exclusion criteria to specifically exclude patients with glucose greater than 160 or with any history of insulin-dependent diabetes.
Clearly, if rigorous inclusion/exclusion criteria had been applied from the start, each of the adverse events reported in the study would have been avoided. When patients are carefully screened and AON is utilized only for a select population of otherwise healthy individuals, the method is clearly a safe alternative.
Further, in each of these cases, the adverse event was not caused by the Anesthesia-Assisted Opiate Detoxification. Clearly, the psychiatric complication of the bipolar disorder occurring 5 days after anesthesia was not the result of Anesthesia-Assisted Rapid Detox itself but might have occurred given any circumstances. The pulmonary edema and glucose derangements were not the results of the Anesthesia-Assisted Rapid Detox. In our experience, not one patient of more than 1700 has had blood pressure abnormalities or hyper/hypoglycemia.
With our method, patients are placed under anesthesia using propofol and kept sedated for 45 – 60 minutes total. This short duration using “light” anesthesia minimizes the risks of complications associated with the use of anesthetic agents. In the anesthesia-assisted arm of the Collins study, patients were put under general anesthesia with propophol, isoflurane, and midazolam (a much deeper anesthesia) and paralyzed with vecuronium as needed, and then kept under general anesthesia for 4-6 hours It is unclear why they were kept under anesthesia for so long, but clearly the more prolonged the time under anesthesia, the greater the risk of anesthesia-related complications. Patients were only given 0.3 mg of oral clonidine prior to anesthesia induction, which is not enough to counter the noradrenergic hyperactivity associated with opiate withdrawal, and were given an excessively large, rapid dose of nalmephine (4 mg IV over 30 minutes). The selection of these doses may explain the hemodynamic fluctuations faced by the investigators. Using a slower administration of nalmephine over 60 minutes, rather than 30, and a dose of 1.5-2 mg rather than the 4 used in the study, we have found that patients tolerate the procedure extremely well, and have faced no such difficulty with hemodynamic instability.
For the purposes of the JAMA article, withdrawal severity was assessed by standard severity scales (SOWS, OOWS, and COWS). Patients in the anesthesia group demonstrated unusually high withdrawal severity prior to the initiation of the procedure, which the authors attribute to anticipatory anxiety. This is in contrast to our experience. The authors then failed to demonstrate a consistent pattern of withdrawal severity with the three measurement tools selected. It is hard to claim that Anesthesia-Assisted Detox was not better tolerated than other methods of detox given this inconsistency. Clearly, this needs to be evaluated further to demonstrate our firm belief that undergoing the most severe withdrawal symptoms under anesthesia is more comfortable than experiencing them while awake.
For the purposes of the study, patients were considered to have relapsed if opiates or any drug (including marijuana, benzodiazepines, barbiturates, and amphetamines) was found to be present in urine samples. Patients were therefore considered relapsed if they remained abstinent from opiates but tested positive for other substances. Looking just at the mean proportions of patients with urine samples positive for opiates, the authors claim no differences, but the anesthesia-assisted group appears better, with 0.54 compared to 0.62 and 0.73 for the buprenorphine and clonidine groups, respectively. Overall their patient retention results were dismal for all three treatment arms. Patients were considered “dropouts” if they relapsed to opiate dependence or missed two consecutive weeks of outpatient visits. Interestingly, relapse to opiate use was determined by two urines positive for opiates. We would consider any urine positive for opiates to identify opiate relapse.
The retention rate at 12 weeks was only 5 patients each (14%) for both anesthesia and buprenorphine-assisted groups and 2 patients (5.9%) for the clon