Under General Anesthesia and Assisted Ventilation:
Experience with 510 Patients in Four Countries
The Royal College of Psychiatrists
Colin Brewer, MRC Psycho.; Mary Laban, MRCA; Charles Schmulian, FFA, (The
Stapleford Centre, 25a Eccleston St., London SW1W 9 NP); Lance Gooberman, MD
(ROD Treatment Center, Merchantville, New Jersey); Yiannis Kasvikis, MRC Psych .(Centre for Mental Health, Athens); Nabil Abdel Maksoud, MD (Cairo University)
This paper describes some modifications of the original Vienna method of rapid opiate detoxification under general anesthesia (RODA) and naltrexone induction. We use muscle relaxants and assisted ventilation, with propofol, isoflurane and thiopentone. Octreotide greatly reduced gastrointestinal secretions. There were no significant anesthetic complications. Most patients were fit for discharge within 24 hours. Heroin dose did not correlate with speed of recovery. Abstinence rates as high as 76% at four months were achieved. Patients were successfully withdrawn from as much as 200 mg of methadone daily. Post-detox management is discussed.
Loimer et al (1988a; 1988b) first described the technique of opiate withdrawal precipitated and accelerated by opiate antagonists while the patient is anesthetized for a few hours. They used methohexitone or thiopentone for anaesthesia. Patients were intubated but not paralyzed. Initially, they used a naloxone infusion. Later, patients receiving methadone at doses up to 120 mg daily were detoxified and transferred to full doses of naltrexone in 4-6 hours (Loimer et al, 1991a).
The Vienna group originally speculated that barbiturates, such as thiopentone, had a particular ability to suppress precipitated withdrawal symptoms but it appears that virtually any standard anesthetic agent can be used for this technique. Brewer (1989a) first described the use of propofol, which gives a very rapid recovery. Clonidine was given as a premedication because of its effectiveness in other techniques of precipitated withdrawal using sedation rather than anesthesia (Charney et al, 1986; Brewer et al, 1988; Senft, 1991). Naltrexone was administered during anesthesia via the nasogastric tube.
RODA has become more widely used in the last three years. This is due partly to its commercialization and promotion by the Spanish-Israeli CITA group, who are attempting to patent the procedure (Brewer, 1996b). However, there is a growing realization although it is neither appropriate nor necessary for all patients and may attract many patients who find conventional withdrawal difficult and/or unpleasant. Some will not even contemplate withdrawal, so great is their fear (Milby et al, 1986). Furthermore, although rapid opiate detoxification under oral sedation (RODOS) is often a satisfactory alternative technique, some patients are difficult to sedate and may present nursing problems (Brewer et al, 1988; Brewer, 1989b; Senft, 1991). It is difficult to identify these patients in advance. We present some data on 510 patients having RODA at clinics in London (80 cases), Merchantville, New Jersey (355 cases), Athens (25 cases) and Cairo (50 cases), using similar techniques.
PATIENT SELECTION AND CHARACTERISTICS
The clinics in London and Athens offered several withdrawal techniques, and the final choice was largely left to the patient. In London, where methadone maintenance was also an option, several patients were advised to try a period of stabilization on methadone before reconsidering detoxification. Cairo patients
were discouraged from detoxifying if they were thought to be motivated largely by external pressures. The New Jersey clinic offered only RODA. In Athens, of 60 consecutive patients seeking withdrawal, 35 (60%) chose other methods. All patients were told repeatedly that detoxification was usually only a stage in
the treatment of addiction and that after-care, including supervised naltrexone was important.
Few patients suffered from conditions which contraindicated general anesthesia. Two known HIV positive patients were treated in London, one with a CD4 count of 120. Of the Athens patients, 18 (72%) had hepatitis C.
All Athens patients were heroin users (mean daily use 0.7 grams) as were most of the New Jersey and Cairo patients. In contrast, 30% of London patients were detoxified from methadone. Thirty-six percent of Athens patients and a similar proportion in London were also dependent on benzodiazepines.
The main difference from previously described methods was that all patients had assisted ventilation, usually with atracurium as the relaxant. Propofol was the usual induction and maintenance agent, but isoflurane in a closed circuit was used in 18% of London cases and thiopentone was used in several cases in
Cairo. Propofol is expensive and the use of muscle relaxants reduces the total amount needed for the procedure. Duration of anesthesia was four hours in New Jersey and four to six hours elsewhere. Pre-medication includes antiemetics, usually droperidol or ondansetron and a benzodiazepine. H2 blockers or proton pump inhibitors are given to reduce acid secretion in case of aspiration. Clonidine was used in all the centres.
Diarrhea is poorly controlled by anti-withdrawal drugs and is not suppressed by anesthesia. Opiates are ineffective in the presence of opiate antagonists. Profuse liquid diarrhea is common in precipitated withdrawal. All centres initially used pre-treatment laxatives and/or enemas, but the growth-hormone
analog octreotide, given IV or subq as a pre-medication greatly reduced diarrhea and bowel preparation is now unnecessary. A full report on octreotide’s effectiveness is in preparation. Monitoring includes constant ECG, respiration, pulse, BP, SaO2, and end tidal CO2. A urinary catheter is inserted but removed before awakening.
ADMINISTRATION OF OPIATE ANTAGONISTS
In all centres, IV naloxone 1.6-2 mg was given initially. About 20 minutes later, naltrexone in doses from 12.5 mg to 25 mg is given as a suspension via the nasogastric tube. After the naloxone, New Jersey patients also receive 2 mg of IV nalmefene, an opiate antagonists with a half life of ten hours. Further
doses of naltrexone are given two to three hours later to a total of 50 to 200 mg. Higher doses may cause more side effects, but by ensuring opiate blockade for up to four days may reduce the likelihood of early relapse.
Following naltrexone administration, there are usually few signs of withdrawal. Piloerection may be seen and, more rarely, sweating. The dose of muscle relaxant is titrated for less than total paralysis and slight movements of the limbs may be seen initially.
EMERGENCE FROM ANESTHESIA
Patients vary in their behavior at this stage. Most are drowsy but all New Jersey patients get out of bed within 30 minutes and walk to the toilet with assistance before returning to bed, if necessary, for a period of observation. A minority of patients are restless and need further sedation. Further doses of
antiemetics and octreotide can be given for nausea and diarrhea. These problems generally settle within a few hours. Octreotide greatly reduces gastric as well intestinal secretions. If necessary, clonidine should be given in adequate doses provided blood pressure is at least 80/50 and the pulse at least 50 per
minute (Charney et al, 1986). Comparative studies are needed to show whether it is more effective if given as part of the premedication or whether it is equally effective if given before extubation.
The speed of recovery is very variable and seems to bear little or no relationship to the normal daily dose of opiate. Some patients are fit to walk out of the clinic unaided an hour after extubation. All New Jersey cases are treated as out-patients and leave the clinic after a few hours to go home or to stay at nearby hotels where they can receive regular medical or nursing visits. In Athens, Cairo and London, the patients stay in the hospital over night. Occasionally a second night seems advisable, especially if home circumstances are less than ideal (the original Vienna patients stayed in the hospital for five to seven days but all were ambulant by the second day and their treatment was paid for by the state health service).
Management varied in the four centres, reflecting both cultural differences and expectations and the clinical background of the physicians involved. All patients were advised to take naltrexone for at least six months under family supervision. In addition, New Jersey patients, treated by an internist, were strongly encouraged to join or rejoin 12-step groups in line with the importance attached to this approach in the US. Athens and London patients, treated by psychiatrists were offered individualized programs using cognitive behavior concepts. In Cairo, where treatment was coordinated by a medical toxicologist, regular urine testing and family involvement were emphasized. If patients seemed to have few underlying problems, management largely involves being generally supportive and available.
WITHDRAWAL SYMPTOM RATINGS
The most complete set of data relates to the Athens patients. Using a 20 item 4-point withdrawal rating scale developed by Bradley et al (1987), patients were rated on admission before RODA and again the following day before discharge. The overall rating was slightly, but not significantly worse after RODA (paired T-tests pre- vs. post- = 8.5 vs. 12.5, t=1.5). However, on symptom by symptom comparison (Wilcoxon matched pairs), diarrhea (7=2.8, p less than 0.005), feeling cold (Z=2.4 p less than 0.01) and hot and cold flushes (Z=2.6, p less than 0.0l) were significantly worse. Neither age of patient, duration of
opiate use, nor usual opiate dose correlated with symptom severity (Pearson correlation).
No serious anesthetic complications were encountered using these methods. One New Jersey patient developed bradycardia and first degree heart block, probably related to clonidine which responded to beta-adrenergic agonists.
ABSTINENCE RATES DURING FOLLOW-UP
RODA was first used in Cairo in October of 1995 and the follow-up data are unusually complete. Of the first 30 patients representing all patients detoxified at least four months ago, only five patients have not been regularly followed-up, usually because of living or working abroad. Regular urine tests have been done in the remaining 25 cases. As of July, 1996, only one out of the 25 had relapsed to opiate use, though in four cases, urine has been positive for cannabis. The very high ‘success rate’ in the Cairo patients – 76% even if all patients lost to follow-up are assumed to have relapsed – probably reflects both
the rigorous selection of well-motivated patients and the suitability of close-knit Egyptian family structures for treatment involving family supervised naltrexone, which increases abstinence rates considerably compared with conventional treatment programs (Gerra et al, 1995).
Follow-up data are less complete for the other centres. RODA patients probably don’t have better long-term results than comparable patients who complete conventional in-patient withdrawal programs. However, "It seems likely that a significant proportion of patients who would fail (or have failed repeatedly) to complete conventional withdrawal will succeed with the help of anesthesia or sedation. This is one important reason for putting these techniques on the therapeutic menu." (Brewer, in press).
IS WITHDRAWAL MORE DIFFICULT FROM METHADONE?
Many addicts and addiction physicians believe that withdrawal from methadone is worse than from heroin but patients who receive methadone treatment may not be representative of the majority of heroin abusers. They may take methadone precisely because they have withdrawal symptoms, which are worse than average. Animal studies show that the severity of withdrawal symptoms is, at least in part, genetically determined (Suzuki et al, 1987). The same is probably true of humans. Withdrawal severity correlates poorly with daily opiate dose (Kosten et al, 1989). To answer the question objectively, a group of methadone addicts would have to be randomized to remain on methadone or take equivalent doses of morphine for at least a week before RODA. Some pure methadone addicts recover quickly even from doses up to 200 mg daily. Some pure heroin addicts take longer than average. However, even in the worst cases, patients with jobs can usually return to work within a week.
These results confirm that RODA is not only rapid and humane, but also an effective and acceptably safe method of opiate withdrawal. It is clear that most patients have relatively mild withdrawal symptoms which soon improve and the findings in the Athens group are similar to those previously reported by the Vienna group (Lormier et al, 1991b). However, claims that patients experience no withdrawal symptoms are manifestly untrue (CITA informs its patients that any discomforts they experience on waking are not withdrawal symptoms, but a "a sign that the body’s immune system has begun functioning again). Furthermore, a small minority have persistent, if largely subjective symptoms, in which can be very distressing even if they cannot always be assessed on the standard withdrawal rating scales. Appropriate medication can assist the recovery process. Apart from naltrexone and clonidine or lofexidine, hypnotics were the most widely prescribed class of drugs. Sleep patterns can take several weeks to normalize. Anti-depressants have not been frequently prescribed.
Farrell (1994) claims that because death from uncomplicated opiate withdrawal is virtually unknown, it is not justifiable to introduce the potential hazards of anesthesia. However, the relatively slight hazards of
modern anesthesia in generally young patients (without the added risks of a surgical procedure) must be set against the frequent and sometimes lethal complications of continuing opiate abuse, especially IV abuse. It should equally be said that since nobody dies from bad teeth, an unshapely nose, or the pain of childbirth, it is unjustifiable to offer general anesthesia for the management of these conditions. Provided that the risks are small and adequately explained, patients are surely entitled to take them (Brewer, 1996a). Punitive attitudes to drug addicts apparently make some health professionals feel that they do not ‘deserve’ good symptom relief (Brewer, 1995).
RODA and early discharge mean that, compared with conventional withdrawal programs, many more patients can be treated during a given period. Unconscious patients do not manipulate nursing and medical staff or smuggle opiates into the ward.