TCR: Dr. Fishman, you are boarded in both internal medicine and psychiatry, and have gone on to specialize in pain medicine. Can you outline your basic approach to pain symptoms in a psychiatric patient?
Dr. Fishman: Sure. The problem that physicians of all disciplines have is that when you see patients with chronic pain who have comorbid affective disorders, it is very easy to assume that the affective disorder is causing the pain and that the patient doesn't "really" have physical pain, or on the other hand to assume that the depression is secondary and ignore it, but the fact is that the two drive each other. Treating both the depression and the pain is critically important to patients's affective well-being and also their pain state. The interesting part for those of us who specialize in pain management is that we often see doctors who try to treat both the pain and depression with one drug with the idea that antidepressants can be analgesics. It is often a mistake because not all antidepressants are independently analgesic.

TCR: Are some antidepressants more analgesic than others? There certainly has been a lot of publicity about duloxetine lately, for example, touting its analgesic efficacy.
Dr. Fishman: There's a great quote about drugs never being as efficacious after release as they were pre-release. The jury is out on duloxetine but the data is very strong and the hype is even stronger--we've seen this kind of thing before. The bottom line is that the antidepressants that have been shown to be independently analgesic are probably effective not because of their dual reuptake properties, but because they are local anesthetics. This is primarily true of the tricyclics which are well known to be sodium channel blockers. This sodium channel blockade is what causes the local anesthetic effect and also why these drugs, unlike other classes of antidepressants, affect the ECG intervals and why overdose requires ICU level cardiac monitoring.

“The bottom line is that the antidepressants that have been shown to be independently analgesic are probably effective not because of their dual reuptake properties, but because they are local anesthetics.”
- Scott Fishman, M.D.

TCR: So the analgesic effect of tricyclics is due mainly to their local anesthetic effect.
Dr. Fishman: Right. To date, there are no antidepressants that are not also sodium channel blockers that have shown independent analgesia in well-controlled studies.

TCR: What about Effexor (venlafaxine) and the SSRIs? Occasionally you run across an article showing some analgesic effect with these.
Dr. Fishman: Yes, there are many small studies that suggest analgesia with SSRIs or Effexor; however, any lessening of depression in a patient with pain will be analgesic since depression magnifies pain. But in general, the studies that have carefully excluded patients with depression have not shown independent analgesic efficacy in neuropathic pain for non-tricyclic antidepressants. The interesting thing about Effexor is that it has properties that are different from any other antidepressant in that it may have direct opioid effects. There have been a couple of very small studies in normal volunteers that have shown that venlafaxine has raised the pain threshold; whether that is clinically meaningful or not, we don't know.

TCR: What about the SSRIs and SNRIs for comorbid pain and depression?
Dr. Fishman: Depression is such a magnifier of somatic sensations that treating the depression may be just as analgesic as treating the pain with a direct analgesic. So that secondary effect shouldn't be unappreciated. The point is that if you give an SSRI to someone who is not depressed, you are probably not going to see a lot of analgesia.

TCR: But if you have somebody who has depression and is also complaining of headaches, shoulder pain, back pain, etc., chances are that if you adequately treat the depression with an SSRI, some of those pains will go away as well.
Dr. Fishman: Correct. That's why you want to use a good antidepressant to treat the depression maximally, and then you can use a separate agent for independent pain relief. Whether you use a tricyclic for this has more to do with safety, tolerability and ease of use of the drug rather than its efficacy since all analgesics for neuropathic pain appear to be equally effective.

TCR: We see a lot of primary care doctors and psychiatrists using very low doses of Elavil - 10-25 mg QHS -- sometimes as an adjunct to an SSRI for pain relief. Is that reasonable?
Dr. Fishman: It is reasonable to add a tricyclic, but the myth about the tricyclics and pain is that they work in low doses, and that has been shown to be wrong. They may give you some efficacy at low dose, but they maximize their effect over dose increases. I recommend pushing the dose up to the point of tolerability of side effects. However, side effects with tricyclics are common and not only limit how much we can give, but have relegated tricyclics below other choices for neuropathic pain such as some of the newer (safer) anticonvulsants.

TCR: Another medication, Neurontin, is a medication that psychiatrists are comfortable prescribing for a variety of off-label psychiatric uses, and it is effective for pain.
Dr. Fishman: Yes, and I think the choice to use Neurontin is notable not because it is better than any of the other neuropathic analgesics or anticonvulsants, but because it is the safest. It has no P450 hepatic metabolism and it is not protein-bound, so the risk of drug-drug interactions is nil. But it is not the easiest to dose and it is often less tolerable than generally assumed.

TCR: Is it effective for pain?
Dr. Fishman: Yes, but like antidepressants, the response is idiosyncratic. Incidentally, the "son" of Neurontin--Pregabilinwill be approved soon.

TCR: For what indications?
Dr. Fishman: For post-herpetic neuralgia, fibromyalgia, and probably generalized anxiety. It will be similar to if not stronger than Neurontin, but will be approved for BID dosing instead of the QID dosing of Neurontin.

TCR: What about opioid analgesics? Patients on opioids tend to get stigmatized as being "addicted" and there is confusion in the medical community about whether the patient is really "an addict" or is using the medication appropriately. Any ideas about how to help sort that out in a given patient?
Dr. Fishman: The definition of "drug addiction" is the compulsive use of a drug that causes dysfunction, and the continued use of the drug despite that dysfunction. And the definition of "pain relief" is an improvement in functioning due to analgesic treatment. Pain is essentially an alarm that gets your attention and, with enough pain, makes it impossible to do anything beyond attending to the pain. If you have been in pain day in and day out, you lose function in your life. When you treat the pain, their functioning improves. And that is the complete opposite of what we see with addiction. If a patient is taking an opioid and has improved function, you really can't make the argument that he or she is addicted unless there is increased dysfunction elsewhere in their life. There may be physical dependence or tolerance, but these are pharmacological properties of the drugs, and must not be confused with addiction.

TCR: How do you recommend we evaluate patients on chronic opioids?
Dr. Fishman: It requires that you look at functional outcomes. When I put someone on a chronic opioid, the question I ask them is, what are you going to do with this medication that you can't do now? And then we target those things. And if they can't come up with things that are meaningful, then I say, well I have good news for you, you don't really need this medicine. Basically, if you are successfully treating pain, people's lives should get better and there should be objective evidence of that and that objective evidence is diametrically the opposite of the evidence you would get with addiction.

TCR: And so as psychiatrists, when we are not directly treating the pain, but we are treating an emotional problem, we can basically go ahead and treat these patients as we would any other patients and assume that they are going to respond in the same ways.
Dr. Fishman: Yes, I would agree to that and I would add that the psychiatrist can play a critical role in looking at functional outcomes. Because if I am prescribing an opioid and think that the patient is functioning better, and you as the psychiatrist see functional or psychological decline, I need to know that. It doesn't necessarily mean the patient is addicted, but it means that treatment is not going in the right direction.

TCR: Thank you Dr. Fishman.