Imagine this: You’re practicing medicine and a patient comes to you with an illness. You make the diagnosis and then say to the patient, “I can see that you are very sick. And there is a highly effective treatment for your condition that would make you feel a lot better. It’s simple and it isn’t even expensive. But, you know what? I’m not going to give it to you! You’re not sick enough. Come back tomorrow. If you’re sicker tomorrow—well, if you’re sick enough—I will treat you then. But not right now.”
Crazy, right? We’d never do such a thing.
But . . . the problem is, we frequently do that exact thing with our heroin withdrawal patients. I’m not singling out correctional medicine practitioners here. I think that, in general, heroin withdrawal is treated better in correctional settings than it is in the community. Nevertheless, it is a fact that heroin withdrawal is often not properly treated in jails and prisons. I have seen it.
I believe that there are four main reasons that some facilities do not appropriately treat heroin (and other opioid) withdrawal.
This is a Myth: Cold turkey withdrawal is not dangerous
The first is the belief that cold turkey withdrawal from heroin does not hurt anyone. I myself was taught in my residency “no one dies from opiate withdrawal.“ This is a very commonly held belief to this day.
The problem is that this is simply not true. It is a myth to believe that no one ever dies during withdrawal. People do.
Think of the situation of heroin withdrawal this way: No one disputes that patients can get very sick when going through heroin withdrawal. And maybe young and healthy patients can tolerate being that sick with no lasting problems. But what about someone who isn’t that healthy to begin with? Say someone who has asthma and heart disease? Or maybe they have an underlying sepsis acquired from sharing needles? What if this patient is also malnourished and dehydration from not eating? Could such a person already weakened by these conditions end up dying when the physiological stress of withdrawal is piled on? Of course they could!
And they do. I personally know of cases where patients did indeed die while going through opiate withdrawal. But even if they don’t, people get very sick from heroin withdrawal. To assume that opiate withdrawal is a benign condition is a serious fallacy.
Clonidine is an effective treatment for heroin withdrawal
The second reason that opioid withdrawal is often not treated is the mistaken belief that the only effective treatment is more opioids. And jail practitioners are reluctant to prescribe opioids for opioid withdrawal for various reasons. I understand this.
But the belief that the only effective treatment for opioid withdrawal is more opioids is also a myth. There is indeed a highly effective non-narcotic treatment for opioid withdrawal: clonidine.
I should point out here that I am talking here about using clonidine as a short-term treatment for acute heroin withdrawal in a correctional facility. I am not talking about treating the underlying opioid addiction itself. The treatment of opioid addiction itself commonly uses long term prescriptions of Suboxone or methadone, known as Medication Assisted Treatment (MAT).
I am a big fan of addiction MAT (see my comments about continuing outpatient MAT in jails here), but, treating withdrawal is different than treating addiction. We are talking here about withdrawal, most commonly from heroin. When heroin addicts are booked into a jail for, say, a few days, it is simply not logistically possible get them enrolled into a MAT program. These patients are going to experience withdrawal. Let’s treat that first—using clonidine.
Clonidine has been validated as an effective treatment for opioid withdrawal in several studies. Here is the Cochrane Review of several of these.
I also have my experience of successfully treating literally hundreds of patients for opioid withdrawal with clonidine. I can tell you from long experience, it works and it works well.
I understand the reluctance to use opioids in a correctional facility for heroin withdrawal, but I do not understand any reluctance to use clonidine, especially since clonidine is now in common use as a treatment for all sorts of other conditions, such as PTSD and nightmares.
Don’t use Benadryl to treat heroin withdrawal!
The third reason that heroin withdrawal patients are not treated adequately for withdrawal is that some facilities use ineffective treatments such as diphenhydramine (Benadryl) or hydroxyzine (Vistaril).
Let me be simple and clear: hydroxyzine is not by itself an effective treatment for heroin withdrawal! There is no medical literature to support using hydroxyzine in this role.
And why would anyone prefer Benadryl over an effective medication like clonidine, anyway? One works, one doesn’t. Benadryl is, at best, an adjunctive therapy. If you want to add Benadryl to a clonidine regimen for heroin withdrawal, I have no objection to that. Just don’t use Benadryl as the main therapeutic agent
A word about diarrhea. During heroin withdrawal, patients commonly have diarrhea and abdominal cramps. Many facilities treat this with loperamide. I also have no problem with the use of loperamide as an adjunctive therapy to clonidine, but since the cause of the diarrhea is withdrawal, a more effective treatment for the diarrhea would be to simply give more clonidine. Clonidine treats the underlying cause of the diarrhea. My opinion. Also remember that loperamide has a high risk of abuse (as reported in this article).
COWS undertreats withdrawal
The final reason that heroin withdrawal is often not adequately treated is reliance on withdrawal scoring systems that require patients to meet a certain minimum symptom score before they qualify for treatment. A protocol found in Uptodate using the Clinical Opioid Withdrawal Score (COWS) (found here), for example, and does not begin treatment until a patient has a COWS score of at least 8. That means that a particular heroin withdrawal patient could present with anxiety, muscle aches, chills and nausea—and not get treated! We’re back to “Come back tomorrow and if you’re sick enough, I’ll treat you then!”
I suspect that scoring systems like this are modeled after the alcohol withdrawal scoring system CIWA, which also requires patients to hit a set minimum criteria of sickness before starting treatment. I do not understand this approach (I’ll address CIWA another day). We know that most patients who do not score very high initially are going to get worse. We also know that some patients are more stoic than others, and since they don’t complain enough, they will not get treated. And we know that the earlier that you begin treatment for withdrawal, the more effective it will be. So why wait?
Personally, if a patient in one of my jails says that he is a heroin user and is starting to feel sick, I will start that patient on clonidine. I don’t see the point of waiting. The clonidine will predictably make him feel better. And then he’ll be reassessed later to see if the clonidine dose is adequate or if he needs more.
A word about opinions
What I have written here is my own opinion. I arrived at this opinion as a result of my own training, experience and review of the medical literature. I believe that what I have written here is evidence-based and accurate. But I acknowledge that I could be wrong! You may disagree with me based on your training, your experience and your review of the medical literature.
But if you do disagree, please explain why, in comments.
This and a previous article on Methadone are excellent articles on a very important and common issue. Please comment on treatment of the pregnant patient, both long term, short term and especially the one that comes in Friday night already on Methadone, but the Methadone clinic won’t be open until Monday.
Thanks, Charles. You are correct that pregnant women are a special case and should be continued on narcotics for the duration of the pregnancy, if possible. In the case you mention of a pregnant woman on methadone who arrives on a Friday night but her methadone clinic will not be open until Monday, The DEA allows you to legally treat her over the weekend. Here is a quote from the DEA website:
Question: May an inmate enrolled in an NTP have methadone administered by Department of Corrections medical staff, if the facility does not have a separate registration as an NTP?
Answer: Yes. Medical staff of the Department of Corrections may administer narcotic drugs to a person for the purpose of relieving acute withdrawal symptoms when necessary while arrangements are being made to have methadone supplied by the inmate’s NTP. A separate registration would not be required provided that no more than one day’s medication be administered to the person or for the person’s use at one time. Such treatment is limited to three days and may not be renewed or extended. [21 CFR 1306.07(b)].
If your jail is licensed by the state and the DEA as a clinic, you can legally treat her yourself without an NTPO license for the duration of her pregnancy. However, few jails are registered as clinics. Here is the quote from the DEA website:
Question: May a Department of Corrections medical staff administer methadone to incarcerated, pregnant, opioid dependent women during the course of their pregnancy without a separate registration as an NTP?
Answer: Methadone may be administered in such circumstances when the following conditions are met. A practitioner, or authorized hospital staff, may administer or dispense narcotic drugs in a hospital to maintain or detoxify a person as an incidental adjunct to medical or surgical treatment of conditions other than addiction. Pregnancy is recognized as a medical condition by both DEA and FDA, and, therefore, this would be considered medical treatment of a condition other than addiction.
Such medical treatment is allowed “in a hospital” or institutional setting. However, the Department of Corrections must be licensed by both the state and DEA as a clinic, a hospital, or a hospital/clinic. [21 CFR 1306.07(c)]
I read the Cochrane review on Clonidine and lofexidine. The conclusion was the Methadone was safer than Clonidine and lofexidine, and lofexidine had fewer side effects than clonidine. Yet your preference is clonidine. I’d appreciate it if you clarified. Thanks for an timely discussion.
Thanks for the question, Charles. Lofexidine is marketed in Great Britain as a treatment for opioid withdrawal, but is not available in the United States. As you say, methadone is a great treatment for heroin withdrawal. There are two major problems with using methadone in a jail (versus a supervised outpatient clinic). The first problem is that the DEA has strict rules for using methadone to treat the symptoms of acute withdrawal. Unless your jail is a registered Methadone Treatment Program (and very few are), it is too much of a logistical hassle for most jails.
Second, and more importantly, jails do not know how long inmates withdrawing from heroin will be in jail. Many will only be there for a day or two. Methadone is a very long acting narcotic. Once you give it, it will hang around for days. The danger here is that an inmate gets out after, say, only one or two doses of methadone, uses heroin when the methadone is still on board, and this combination results in an overdose.
Unless you can guarantee that a patient in jail will be there long enough to complete a withdrawal program and then transition into an addiction treatment program, it is safer to treat acute withdrawal with clonidine.
Less chance of hypotension with lofexidine.
Lofexidine is marketed in Great Britain as a treatment for opioid withdrawal. It is not yet available in the United States.
What about patients who have very low blood pressure?
Thanks for the question, Kristy. Since clonidine can lower blood pressure, it should not be used in patients with a very low blood pressure. Uptodate quotes a protocol that uses the blood pressure cutoff of 85/55. The Prescriber’s Letter recommends a cutoff of 90/60. It is quite rare to have a blood pressure that low. Interestingly, I think, clonidine is commonly prescribed as a nightmare therapy by mental health professionals who (I am told by patents) do not check a blood pressure prior to prescribing.
First, thanks for provoking a needed discussion
Second, may I suggest that the approach here not have a focus on “…what is best?” but what has greatest safety?
What is best? can only be determined in retrospect, as the information we have or can likely rely upon in the detention setting is typically incomplete …
My experience is within a center that has a liberal prescriptive landscape, including methadone for assisted “detox”… and our issues are on the opposite side and that is how do we insure patient safety.
So I would respectfully change your opening question to: How is withdrawal handled safely, particularly when there may be multiple withdrawal “risks”…… Conservative care may be the safest, particularly when considering how many providers prescribe multiple medications for only “risks”….
No discussion can be complete on this subject without mentioning Buprenorphine (Subutex). Maybe I missed it in Jeff’s article and the subsequent comments. I heard Todd Wilcox’s presentation on this at the ACCP Spring educational Conference March 18, 2017, Atlanta.
Thanks, Dr. Buck! See my most recent post. Buprenorphine is great. It is not the best choice for all jails, though.
Yea, I mean why alleviate an inmates pain right?
I agree that in general opioid detox is undertreated, and there’s wide misconception on use of Clonidine. My providers have not used it even though it is an option on our protocol due to BP concerns.
A comment on Buprenorphine: We taper Benzo during Benzo detox, I’m not sure why we don’t taper Opioids (using partial agonist such as Bupernorphine) during Opioid detox. We already use it for pregnant women, and plan to use this in our detox protocols.
Thanks, Dr. Chiang! See my recent post. Buprenorphine is great, for those who use it properly. What I disagree with are the jails that still allow “Cold Turkey” withdrawal.
I work in a 650 bed county jail in an area with a lot of cheap potent heroin. Needless to say, we see a LOT of opiate detoxers. I would say that, above all, effective and routine monitoring are the most essential component of any detox program. In addition, if you can get an X-license and start using subutex then you should do it as soon as possible. Subutex relieves un-necessary suffering, improves patient outcomes, and basically takes a lot of the “work” out of detoxing people in acute opiate withdrawal.
We monitor all detoxers at 0800, 1400, 2000, 0200. We provide subutex on a tapered dosing regimen over 5 days (if anyone wants the taper we use, please respond to this). I (the PA) personally round on all detoxers every morning that I’m at work. This extra level of provider attention allows you to get ahead of the 8-ball and see your ‘problem’ detoxers before they get too bad off. I think it’s important to realize just how sick detoxers can get.
Thanks for the comment! It sounds like you have a great program!
Our small jail uses Tylenol 3 for the treatment of opiate/heroin withdrawals in pregnant women. Are there any DEA regulations on this use?
Since Tylenol #3 is not a DEA schedule II substance, so, like Tramadol, there are no DEA regulations that I know of that prohibit its use as a treatment for withdrawal. However, I do not think this is optimal medical therapy, especially for a pregnant patient in withdrawal. I personally would not use codeine in this manner.
Our whole state has one jail that does methadone treatment for pregnant opiate addicts. Maricopa county appears to have a great program. Not only do they do methadone, they have a great program set up to ensure proper OB treatment as well as services after release.
NCCHC is opposed to using Tylenol #3 in this situation.
We are a 500 bed jail facility. I have read your article about opiate withdrawal and your experience is similar to ours. I am curious to understand your dosing perimeters for clonidine in opiate withdrawal. Thanks.
Sorry for the delay, Jonathan. For typical heroin withdrawal, we begin with Clonidine 0.1mg po BID for ten doses. That actually is enough for, say, 3/4 of the patients we see. But we recheck every patent periodically. Depending on the need, we can progress to Clonidine 0.2 BID or even TID before tapering off. All of this depends on an adequate blood pressure, for example. Methadone withdrawal has to be treated longer, of course.
Thank you so much for answering my question.
Jonathan Stump
That is the same clonidine order we use here at our jail. We (RN) also do BID withdrawal assessments at med passes. This allows us to keep track of symptoms as well as monitor VS (we have a protocol to hold clonidine if BP is less than 85/70). Most of the inmates appreciate how the clonidine helps with the detox process causing them to push fluids (also good for them) to keep their BP high enough.
I work in a 300 bed facility and our COWS is as follows:
Tylenol #3, Gabapentin 400 BID to TID depending on the level, Imodium, Zofran, Clonidine of course watch BP. Frequency of mediation depends on COWS level. How do you feel about this protocol. Methadone and or Suboxone gets a 10 day monitor. It seems to work for our inmates/patients.
Hi Dorie! This is the first time I have heard of anyone using codeine to treat opioid withdrawal. However, I have heard of protocols that use Tramadol I have also read that gabapentin has been used as an adjunct to treat opioid withdrawal. I have never used either. I just rely upon clonidine. According to the Cochrane Review, clonidine works as well as decreasing doses of methadone to treat opioid withdrawal. Gowing L, Farrell MF, Ali R, White JM. Alpha2-adrenergic agonists for the management of opioid withdrawal. Cochrane Database Syst Rev. 2014;3: CD002024. If my heroin withdrawal patients are still having nausea or diarrhea after beginning clonidine, I usually give more clonidine.
That is not to say that your protocol is bad! If it works, great. My main message is simply to treat opioid withdrawal. I am anti-cold turkey!
It is my understanding that some NCCHC physician surveyors do not accept a T3 protocol as acceptable , especially in pregnant females, even for a short period. There different opinions on this. Perhaps a neonatologist, as well as obstetricians need to be consulted.
One problem with T3 is that a certain percentage of the population (I seem to remember 20%) lack the enzyme that metabolizes codeine to morphine. So for those ~20%, Codeine will be totally ineffective. I think that this is a problem that needs to be anticipated in advance. Each facility should get a committee together, perhaps including a local obstetrician or neonatologist, and decide what will be done when a heroin addicted pregnant patient shows up . . .
Thanks for your thoughts. There seems to be continuing controversy of this issue. I’m not convinced T3 is contraindicated in this setting; some do. There are 3 issues: 1) the mother, 2) the pregnancy & 3) the fetus. Two & 3 are not necessarily the same. I think the primary concern is the fetus. I have not been able to find any literature or studies that show the fetus would be harmed. I was able to find 1 study that suggested the fetus is NOT harmed. I think those who think T3 in contraindicated do not have a basis for that, other what they assume, which may purely be anecdotal. I think you are correct. Perhaps ACCP should present a forum on this subject with neonatologists & obstetricians.
Thank you, everyone, for sharing your experience. As a psychiatrist, recently following up some cases in correctional facilities including about 3000 beds in Gulf area. After Lofexidine FDA approval, would you consider it as first line in opioid withdrawal?. It has been widely prescribed in Egypt for years.
I have never used lofexidine, but have used its cousin, clonidine, successfully for years. Lofexidine reportedly has less effect on blood pressure than clonidine, but I have had very few cases where clonidine’s blood pressure effects were a significant issue. Critical question: How much does lofexidine cost in the US compared to clonidine?
Cochrane summary of the two: http://www.cochrane.org/CD002024/ADDICTN_clonidine-lofexidine-and-similar-medications-management-opioid-withdrawal