Access to the PPM Journal and newsletters is FREE for clinicians.
13 Articles in Volume 10, Issue #5
An Osteopathic Approach to Fibromyalgia
Co-Morbid Psychological Disorders in Interventional Pain Management
Compliance Monitoring and Effective Risk Mitigation Strategy
Cultural Differences and Pain Management
Electronic Prescription of Controlled Substances
Kinetic Chain from the Toes Influences the Craniofacial Region
Non-responsive Pain Patients with CYP-2D6 Defect
Platelet Rich Plasma for Hamstring Tears
The Iontophore
The Treatment of Achilles Tendonitis Using Therapeutic Laser
Thoracic Facet Injections
Urine Drug Testing as an Evaluation of Risk
Vitamin D Levels In Pain and Headache Patients

Non-responsive Pain Patients with CYP-2D6 Defect

Pain patients who do not respond to the analgesic properties of the most commonly used opioids have a significant chance of being genetically incapable of generating the clinically-active metabolite of these medications.

The clinical report presented here contains the only collection of pain patients with CYP-2D6 deficiencies that have been systematically studied. There is a practical salient point in this article for pain practitioners. Any patient who doesn’t respond to hydrocodone or low dosages of oxycodone may have this genetic defect.

Since 2005, I have been testing patients for cytochrome P450-2D6 (CYP-2D6) genetic abnormalities. This was when the blood test was first available for use by us clinicians in private practice. An abnormality of the CYP-2D6 enzyme may be referred to elsewhere in this paper as a deficiency or defect. All told, I’ve tested 15 patients who I have suspected—on a clinical basis—of having a CYP-2D6 genetic defect and, indeed, all have shown an abnormality (see Table 1). This includes myself and my brother. Included in this cohort is a single patient who has CYP-2D6 gene duplicity and is termed a rapid- or ultra-metabolizer. The clinical history of these persons is quite typical and I share this information to help practitioners identify pain patients who may have a genetic defect of CYP-2D6. They will likely not respond to some opioids. These patients may require high dosages of opioids or may be falsely accused of drug-seeking behavior.


There are more than 50 enzymes within the human intestinal tract and liver that are involved in digestion of food and other substances.1-4 The cytochrome-P450 super-family of liver/intestinal enzymes makes up a large portion of the enzyme system that is responsible for the disposal of potentially harmful substances. These enzymes are referred to as ‘CYP450’ enzymes. They are heme-related enzymes that use iron to oxidize substances to usually—but not always—less harmful and water-soluble metabolites that are then excreted into the urine and eliminated from the body. To enhance the water-solubility of the substances even further, the liver often also attaches a group such as a glucuronide or sulfate moiety to the metabolite. This is referred to as conjugation. The CYP450–mediated oxidation is referred to as ‘Phase 1’ metabolism and the conjugation step is referred to as ‘Phase 2’ metabolism.

CYP450 enzymes are responsible for:

  • 1) synthesis of cholesterol and cholesterol-based substances (e.g., sex hormones, cortisol, vitamin D3, bile acids);
  • 2) catabolism of ingested food substances (xenobiotic metabolism with implied protection against plant alkaloids/poisons);
  • 3) conversion of arachidonic acid to certain prostaglandins (e.g., thromboxane-A synthetase (CYPS), Prostacyclin synthetase (CYPS); and
  • 4) ‘Phase 1’ metabolism of drugs—either activation of pro-drugs or inactivation of active-parent drugs.

All of the P450 enzymes have been identified and have been classified into various CYP families and subfamilies. These enzymes appear in the greatest concentration within the liver, but some of them (e.g., CYP-3A4) can also be found within the walls of the small intestine.2

In terms of CYP-related drug metabolism, there are nine CYP-enzymes of known clinical importance and they are referred to as CYP-1A2, CYP-2B6, CYP-2C9, CYP-2C18, CYP-2C19, CYP-2D6, CYP-2E1, and CYP-3A4.3 Of those drugs that un-dergo liver metabolism, a specific CYP enzyme or a specific combination of CYP enzymes is responsible for each drug’s metabolism. For instance, CYP-2D6 is completely responsible for some drugs’ metabolism. It appears completely re-sponsible for the metabolism of the antihypertensive/B-blocker, metoprolol (Lopressor®/Toprol®), while the antihypertensive/B blocker, propranolol (In-deral®) is metabolized by CYP-2D6 (42%), CYP-1A2 (41%), and 17% by non-CYP metabolism. CYP-3A4 is involved in the metabolism of the largest percentage (>50%) of presently-available drugs followed by CYP-2D6 (>25%). While there doesn’t appear to be any polymorphism of any frequency with the CYP-3A4 enzyme that results in a significant clinic impact, the CYP-2C9, CYP-2C19, and CYP-2D6 enzymes have shown polymorphism.

A single gene is responsible for the production of an enzyme. With the technology used to map the genetic code of the human, we can now identify each person’s individual gene makeup for some enzymes. We now know what the usual or “wild” genetic makeup of CYP450 enzymes is for the vast majority of individuals. As a result of mutations in the actual sequence of the nucleotide bases (e.g., adenine could be switched for guanine) for these genes, there can be allelic differences or polymorphism that can result in significant differences in the actual functionality of the enzyme for which the gene is responsible. If a person’s gene contains nucleotide alterations, the resultant en-zyme may have either reduced, enhanced, or normal oxidase activity compared to what is considered to be the normal or “wild” gene/enzyme type. This potential for genetic variation of the same gene is called ‘polymorphism.’ In different races and even between individuals of the same race there can be significant allelic differences in the overall activity/potency of a CYP enzyme. These differences can have significant clinical consequences, espe-cially when extrapolating the effects of a drug from one patient to another or to another race.

The CYP-2D6 Enzyme

There are more than 20 possible allelic variations in the code of the CYP-2D6 enzyme.1-4 The frequency of these poly-morphisms varies within the major ethnic groups. It has been shown that up to 10% of whites, 2% of blacks, and 1% of Asians exhibit CYP-2D6 polymorphism.4 Since autosomal chromosomes are paired, everybody has two alleles. People with the usual or “normal/wild” type have the phenotypic allelic designation of CYP-2D6*1/*1 and they are referred to as extensive drug metabolizers. Those individuals with other “non-normal” alleles (e.g., CYP-2D6*4) will not be able to metabolize drugs to the same degree as those people with the normal/wild genotype and are referred to as “2D6-Deficient.” These individuals can be either completely devoid of enzymatic function or have intermediate enzymatic activity, depending upon the allele combination they possess. They are referred to as non-metabolizers and poor metabolizers, respectively.

Those individuals who express poor or a complete lack of enzyme function (non-metabolizers) are predisposed to the accumulation of the parent drug and will achieve excessive serum levels and prolonged half-lives of the drugs. These individuals have a tendency to become toxic on the “usual” doses of medications. An example of a potentially very harmful polymorphism can be seen in those individuals who are CYP-2C9 non-metabolizers and are put on warfarin (Coumadin®). Up to 35% of whites have CYP-2C9 allele variants that result in 50% less-than-normal 2C9 enzyme activity.6 These individuals are not able to effectively clear warfarin, which relies heavily on 2C9 for its normal metabolism and thus may be significantly over-anticoagulated on doses above 1mg qday. Phenytoin (Dilantin®), fluvastatin (Lescol®), glyburide (Diabeta®, Micronase®), glimepiride (Amaryl®), and glipizide (Glucotrol®) are other medications that rely heavily on 2C9 for their metabolism and could lead to significant adverse clinical outcomes in patients who have genetic CYP-2C9 defects.

Table 1. Fifteen Patients with CYP-2D6 Abnormalities
Patient Genotype
G.K. *1/*4
G.P *1/*4
A.E. *1/*4
J.O. *1/*4
R.E. *1/*4
M.G. *2/*3
R.H. *2A/*4
S.R *3/*3
S.K. *4/*4
S.G. *4/*4
P.D. *4/*9
B.P. *4/*35
S.I. *1/*2A
K.D. *1/*2A
R.M. 2D6-Duplicity
All of the patients in this Table, except for R.M. (who has more than two genes), are Under-Metabolizers or Non-Metabolizers and are labeled as “Intermediate” to “Poor” CYP-2D6 Substrate Metabolizers, respectively. R.M. is labeled as a “Rapid”- or “Ultra”-Metabolizer.

Many people of Ethiopian or Saudi Arabian descent are ultra-extensive 2D6-metabolizers because they possess multiple copies of the normal 2D6*1 gene. These people will inactivate drugs such as fluoxetine (Prozac®), amitriptyline (Elavil®), propafenone (Rhythmol®), and clemastine (Tavist®) at such a rapid rate that they may not achieve high enough serum levels to be afforded the usual clinical benefit of these drugs unless their dosage regimen is significantly altered to address the shortened half-lives that results.

Sometimes these gene abnormalities can lead to clinically-significant advantages. For example, all of the proton pump inhibitors (e.g., omeprazole (Prilosec®), lansopraxol (Prevacid®), pantopraxole (Protonix®), etc.) are largely metabolized by CYP-2C19 and, when given to 2C19-deficient patients (>20% of Asians10), they tend to exhibit substantially greater ulcer healing rates than the norm because they are essentially receiving significantly high-er doses of drug.5,7 Those individuals who are CYP-2D6 deficient will be protected from the neurotoxic metabolite of methylenedioxymeth-amphetamine MDMA/ec-stasy) since this metabolite is normally formed by 2D6.

Cytochrome P450 Genotype Testing

A blood test for some cytochrome P450 genotypes has been available since 2005 so that we can now identify some people who will clear liver metabolized drugs from their body differently than the norm. People with two abnormal alleles will exhibit the least enzymatic activity and are referred to as poor drug metabolizers. Those individuals who have one normal (*1) allele and one abnormal (i.e., *10), are referred to as heterozygote’s and will exhibit intermediate drug metabolism. It is also possible for about 2% of whites and >25% of Saudi Arabians to have duplication of the functional *1 gene and results in ultra drug metabolism. Up to 13 copies of CYP-2D6 in the same person have been reported.1,4

Table 2. When to Suspect a Patient has CYP2D6 Deficiency
1. Patient who reports a past or present history of non-responsiveness to codeine, hydrocodone, oxycodone, or tramadol.
2. Patient who reports better pain relief with Darvocet-N-100 than hydrocodone (Vicodin®) or oxycodone (Percocet®).
3. A patient who gives a history of significant toxicity or adverse reaction when they have taken specific Beta-Blockers (e.g., Labetalol (Trandate®), Timolol (Blocadren®), Metoprolol (Toprol®), anti-depressants(i.e., Fluoxetine (Prozac®), Paroxetine (Paxil®), or anti-psychotics(i.e., Haloperidol (Haldol®), Chlorpromazine (Thorazine® ).

For over the last eight years, the FDA has mandated that drug companies identify which CYP enzyme are responsible for the metabolism of their drug. They must also report whether the drug has any activating or inhibiting effects on the function of these CYP enzymes. This has been an immense aid in being able to identify specific drug-drug interactions. For older drugs, this information may be difficult to find but it can be found for many of them.

CYP-2D6 Polymorphism and Its Clinical Significance With Opioids

CYP-2D6 polymorphism is of particular importance as to whether patients are able to receive the expected analgesic effect from all of the most commonly prescribed oral opioids (e.g., codeine, hydrocodone (Vicodin®), oxycodone (Per-cocet®), tramadol (Ultram®).8-10 This is because all of these agents are pro-drugs that need to be metabolized to be effective and are relatively inactive as the parent drug. Codeine is converted by liver enzymes—particularly by CYP-2D6—to morphine; hence, it is the morphine, not the codeine, that results in the patient’s analgesia.8,9 If the patient is “CYP-2D6 deficient,” they will not be able to convert the codeine to the active morphine component and will not receive any analgesic effect8,9—especially if they are homozygous for an abnormal allele no matter the dose.

The same is true for hydrocodone, which is converted to its active component hydromorphone (Dilaudid®), and for oxycodone, which is converted to its active component oxymorphone.8-10

Tramadol is somewhat active as the parent drug but its 2D6-mediated metabolite, M-1, is six times more potent than the parent drug. Therefore, while 2D6-deficient patients may receive some analgesic effect from tramadol, it is no where close to the effect/potency that “normal/wild” type patients receive. 2D6-deficient patients may require two to four times the “usual” dose to get the same analgesic effect. Table 2 presents findings that reasonably lead to the suspicion that a patient has CYP2D6 deficiency.

Case Examples

I first heard about the possibility of CYP-2D6 polymorphism in 1996 and ever since then I have accumulated a list of more than 30 patients whom I believe to be CYP-2D6 deficient. I have also believed myself to be 2D6-deficient since failing to achieve pain relief with codeine, hydrocodone (Vicodin) and oxycodone/aspirin (Percodan®) that I had taken for a painful dental abscess in 1976. Many of the patients I have identified as probably CYP-2D6 deficient have been labeled by previous health care providers to be either “crazy,” “drug-seekers,” or “malingerers” because of their claims that they were still experiencing significant pain despite even high doses of Vicodin or Percocet. Some patients didn’t respond to some opioids, again raising the question of some genetic abnormalities.

Case #1

P.D. is a white 56-year-old female who works as a pharmacist. She presented to our orthopaedic clinic for the first time in March 2002. She had originally injured her left knee while skiing in 1973. She suffered a medial meniscus tear and ACL tear that resulted in her undergoing an open medial meniscectomy soon after the injury. No surgery was done to address her ACL tear and, over the ensuing years, she continued to have occasional instability episodes but sought no medial care for this problem.

She started experiencing activity knee pain in the late 1990s and was diagnosed as having significant arthritis. She underwent left total knee replacement in 2001 and experienced significant post-op pain that did not respond to hydrocodone (Vicodin), and oxycodone (Percocet). We found out about this retrospectively at the time we learned she had difficulty regaining her knee range of motion. Later she required closed manipulation. After this procedure, she gained 110° of flexion but she continued to have significant day-to-day pain due to joint instability from inadequate plastic thickness and ligament imbalance.

In 2002, she required surgical insertion of a thicker and custom asymmetrical tibial plastic insert. Since the patient didn’t inform us pre-operatively of her previous apparent lack of benefit from Vicodin and Percocet, we initially ord-ered these same medications as her breakthrough pain options. Her pain was treated with 48 hours of ketorolac (Tor-adol®) 30mg IV q8h and levorphanol (Levo-Dromoran®) 1mg P.O. q8h, and she responded well to these two medications. Within hours of their discontinuation, however, the patient started experiencing significant pain that was no longer controlled with maximum oral doses of hydrocodone (Vicodin®) or oxycodone (Percocet). The hydrocodone (Vicodin) and oxycodone (Percocet) were discontinued and she was switched to meperidine (Demerol®) 50mg-100mg P.O. q4h under the assumption that she was probably CYP-2D6 deficient. Within two doses of meperidine (Demerol), the patient’s pain was back under control and she had a normal uneventful recovery from that point on and was discharged on the 4th post-op day. The rest of her post-op course was normal and she gained 135° of flexion by 8 weeks post-op, and she did not need any meperidine (Demerol) after the first six weeks.

The patient came back to us two and half years later complaining of pain in her opposite knee. X-rays showed severe lateral compartment arthritis as the explanation for the knee pain she had been experiencing over the previous year. The patient underwent surgery in April 2005 and this time, anticipating her lack of response to hydrocodone (Vicodin) and oxycodone (Percocet), we used different pain medications from the start. We used 48 hours of ketorolac (Toradol®) 30mg IV q8h and Levorphanol (Levo-Dromoran®) 1mg P.O. q8h for baseline pain control and Meperidine (Demerol) 50mg to 100mg P.O. q4h for breakthrough pain. She did well post-op and was discharged on the third post-op day. The remainder of her post-op course was uneventful, and she gained 115° of flexion by the time that she was 5 weeks post-op. Later on, the patient was switched to pentazocine (Talwin-NX®) because of some histamine-mediated itching complaints with meperidine.

A CYP-2D6 genotype test on this patient in May 2005 showed a genotype of CYP-2D6*4/*9. The *4 allele, is the most common (70%) abnormal allele, and is associated with enzymes that are virtually non-functional. The *9 allele is associated with enzymes that exhibit decreased CYP-2D6 function. Therefore, phenotypically, this patient would expect to be a poor to intermediate metabolizer of CYP-2D6 substrates like codeine, hydrocodone (Vicodin), and oxycodone (Percocet). This apparently explains this patient’s past experience with these medications.

An interesting side note is that in January 2003, this patient’s 16-year old daughter tore her ACL for which we performed surgery for ACL reconstruction. We got a call from the mother post-operatively wondering if it were possible that her daughter had the same problem processing pain meds as she had, since her daughter did not seem to be getting any pain relief with the hydrocodone (Vicodin) or extended-release oxycodone (Oxycontin®). Her mother asked whether it would be okay to give her daughter some of her left over meperidine (Demerol). She was told it would be fine and she called us the next day stating that her daughter noticed a remarkable difference in her pain control after just the first dose of meperidine (Demerol). Although not documented, it would appear that she inherited her mother’s CYP-2D6 deficiency.

Table 3. Listing of Known CYP-2D6 Inhibitors and the Degree
to Which They Induce Inhibition
Degree of Inhibition Agents
Potent Clemastine (Tavist®), Fluoxetine (Prozac®), Nortripyline (Pamelor®), Paroxetine (Paxil®), Quinidine (Quinaglute®), Terbinafine (Lamisil®), Thioridazine (Mellaril®)
Moderate Amiodarone (Cordarone®), Chlorpromazine (Thorazine®), Cimetidine (Tagamet®), Desipramine (Norpramin®), Duloxetine (Cymbalta®), Fluphenazine (Prolixin®), Haloperidol (Haldol®), Hydroxyzine (Vistaril®), Metoclopromide (Reglan®), Perphenazine®, Ticlopidine (Ticlid®), Trifluoperazine (Stelazine®)
Modest Amitriptyline (Elavil®), Chlorpheniramine (ChlorTrimeton®), Citalopram (Celexa®), Cyclobenzaprine (Flexeril®), Delaviridine (Rescriptor®), Imipramine (Tofranil®), Pimizide (Orap®), Sertraline (Zoloft®), Venlafaxine (Effexor®), Ziprasidone (Geodon®)
Minor Buproprion (Wellbutrin®/Zyban®), Clozapine (Clozaril®), Diphenhydramine (Benadryl®), Escitalopram (Lexapro®), Fluphenazine (Prolixin®), Fluvoxamine (Luvox®), Methadone (Dolophine®), Olanzepine (Zyprexa®), Perphenazine (Trilafon®), Ranitidine (Zantac®), Thiothixine (Navane)
Unknown Celecoxib (Celebrex®), Cocaine, Doxorubicin, Ginseng, Hydroxychloroquine (Plaquenil®), Indinavir (Crixivan®), Orphenadrine (Norflex®), Pergolide (Permax®), Propafenone (Rhythmol®), Propoxyphene (Darvon®), Protriptyline (Vivactil®), Ritonavir (Norvir®), Ropinirole (Requip®), Trimipramine (Surmontil®), Yohimbine

Case #2

B.P. is a white 73-year-old female. She presented with a typical history of osteoarthritis affecting her right knee. After years of failing to respond to non-surgical treatment, she had a total knee replacement in 2000. While the patient was in the hospital, she received ketorolac (Toradol) 15mg IV q8h and metha-done 5mg P.O. BID for 48 hours for baseline pain control, and she was given hydrocodone (Vicodin) 5mg q4h prn for breakthrough pain. For the first three days post-op, she did fine and was discharged on the fourth post-op day. She called several days after being home complaining that her pain control no longer seemed to be as good as it was while she was in the hospital and didn’t think the hydrocodone 5mg (Vicodin) was offering her any pain relief. She was switched to hydrocodone 7.5mg (Vicodin-ES®), and she called back two days later stating that she continued to have as much pain as she had with the hydrocodone 5mg. She was also experiencing gastrointestinal upset from the hydrocodone 7.5mg. She was switched at this point to oxycodone (Percocet) but once again called back within days stating that she still lacked any pain control although her gastrointestinal distress was better. It was at this point that she was switched to propoxyphene (Darvocet-N-100®) and she called back several days later stating that this medication con-trolled her pain very well.

A key point here is that the patient did poorly with hydrocodone (Vicodin) and oxycodone (Percocet), but did very well with propoxyphene/APAP (Darvocet®). In terms of analgesic potency, propoxy-phene is not normally expected to be as potent as hydrocodone or oxycodone. If a patient has a CYP-2D6 genetic defect, the explanation is readily apparent. Since, unlike hydrocodone (Vicodin and oxycodone (Percocet), propoxyphene (Darvocet) is active as the parent drug and it does not require conversion to an active metabolite.

This patient was tested and showed an abnormal genotype of CYP-2D6*4/*35. As with the first patient, the *4 allele is the most common (70%) abnormal allele and is associated with enzymes that are virtually non-functional. The *35 allele is associated with enzymes that exhibit decreased CYP-2D6 function. Therefore, phenotypically, this patient would be expected to be a poor to intermediate metabolizer of CYP-2D6 substrates like codeine, hydrocodone (Vicodin), and oxycodone (Percocet).

Case #3

S.K. is a 73-year-old white female. During surgery for total knee replacement, she experienced good pain control in the hospital while receiving ketorolac (Toradol) 15mg IV q8h and levorphanol (LevoDromoran®) 1mg P.O. BID for base-line pain control and with hydrocodone (Vicodin) 5mg q4h prn for breakthrough pain. She was discharged from the hospital taking hydrocodone (Vicodin) 5mg q4h prn. On the fifth post-op day, however, she complained of increasing pain that was not responding to the hydrocodone. At this point she was switched to oxycodone (Percocet) 5mg q4h prn pain but she complained bitterly of lack of any pain relief from this opioid. She was switched to fentanyl (Duragesic®) patch-25mcg/hr q3days and Levorphanol 1mg q6-8h prn for breakthrough pain. This combination gave her much better pain control. She ceased pain medication after four weeks.

Genotype testing on this patient showed a CYP-2D6*4*4; homozygous for *4 and means that both alleles are completely non-functional. She represents the most inactive type of phenotype (“non-metabolizer”) for 2D6-substrates. Of the three cases presented, she is the most likely to completely fail to respond to any dose of codeine, hydrocodone, or oxycodone.

CYP-2D6 Enzyme Inhibition by Non-Opioids

There are some non-opioid drugs that can induce a “pseudo-CYP deficiency” by inhibiting the function of “normal” enzymes in those persons who are genetically normal/wild type metabolizers, which results in a “pseudo-CYP retarded-metabolism” state5-7 (see Table 3). This interaction accounts for many, but not all, of the drug-drug interactions seen clinically.3 For example, both the sulfamethoxazole and the trimethoprim components of Bactrim® are potent inhibitors of 2C9 enzymes and leads to the inhibition of the normal clearance of warfarin (Coum-adin), which explains the elevation in the INR seen when patients, who take wafarin (Coumadin), are put on Bactrim to treat an infection.11 When the patient is taken off the Bactrim, their CYP-2C9 enzyme function will return to “normal” and their INR will return to it’s usual level.

There are a number of drugs that could, if administered concurrently, could induce a “pseudo-2D6 deficiency” in genetically “2D6-Normal” individuals and, therefore, lead to the same problems of limited analgesia with codeine, hydrocodone (Vicodin), oxycodone (Percocet), or tramadol (Ultram). The CYP-2D6 enzyme does not appear to be inducible as is the case with CYP-3A4.


Hopefully with all that we now know about the analgesic properties of the most commonly used opioids (e.g., codeine, hydrocodone, oxycodone) and the fact that individual patients have a significant chance of being genetically incapable of generating the clinically active component of these medications, CYP-2D6-deficient patients will no longer continue to suffer in pain due to our ignorance of the existence of this condition. Henceforth, it is hoped that health care providers will now first think of non-responders to be genetically different from the rest of the population, rather than labeling them as malingerers, “crazy,” or drug seekers. These patients deserve to be tested for cytochrome P450-2D6 genotyping before being given either of these labels or be given an opportunity to try an opioid that bypasses the CYP-2D6 metabolic pathway. Such drugs include morphine, meperidine, methadone, and fentanyl.11 Testing for abnormal CYP-2D6 alleles and avoidance of CYP-2D6 inhib-itors is very appropriate in patients who don’t respond to normal dosages of the common opioids, codeine, hydrocodone, and oxycodone.

Last updated on: December 20, 2011
close X