Must-read: another adverse effect associated with tramadol

February 9, 2015, 9:54 pm


Tramadol and Hypoglycemia: One More Thing to Worry About. Nelson LS, Juurlink DN. JAMA Intern Med 2015 Feb 1;175(2):194-5


Hypoglycemia associated with use of tramadol has been noted previously in scattered case reports, after both overdose and therapeutic ingestion, involving patients with and without diabetes. In this month’s issue of JAMA Internal Medicine, Fournier et al. presented a large case-control study comparing patients started on tramadol for pain with similar patients started on codeine. They found that the tramadol patients had a significantly increased risk of hospitalization for hypoglycemia, especially in the first month after the drug was started.

True, the hypoglycemia cases were rare, with an incidence of 7 per 10,000 per year. However, as Lewis Nelson (@LNelsonMD) and David Juurlink (@DavidJuurlink) point out in this superb editorial accompanying the article, there are a number of other significant adverse effects associated with this problematic drug. The editorial is a must-read short summary of tramadol pharmacology and toxicology.

They start by noting:

” . . . the expectation that analgesics can significantly reduce or abolish pain is often overly optimistic and can lead to the progressive use of higher doses of stronger analgesics without a reasonable benchmark for success or failure. . . . Nonpharmacologic approaches such as physical therapy, meditation, exercise, and weight loss are harder to implement than medication because they are time consuming, labor intensive, and often not covered by insurance, even though each is supported by evidence of safety and effectiveness in elected patients.”

They then go on to make the following crucial points about tramadol:

  • tramadol works through 2 different mechanisms:
      • a metabolite (O-desmedthyltramadol[M1]) binds to μ opioid receptors; and
      • tramadol inhibits reuptake of serotonin and norepinephrine
  • the metabolism of tramadol to M1 is via CYP2D6, an enzyme with extremely variable activity
  • because of the variable metabolism, “Giving a known dose of tramadol is tantamount to giving an unknown dose of opioid”
  • despite earlier beliefs, tramadol is neither safe nor non-addictive
  • significant adverse effects associated with tramadol include serotonin syndrome, drug-drug interactions, and respiratory depression

This short editorial is a must-read.

Related posts:

The top 5 things to know about tramadol

Tramadol and Brugada Syndrome



  1. steve Sheldon Says:


    I agree that the editorial by Jurrlink and Nelson is a good review of the pharmacology of tramadol, I had most of this on my phone or in my copy of Godmin and Gillman.

    Why is this a must read? 7 out of 10,000 people on the drug had an associated diagnosis of hypoglycemia, that is true. There is no evidence for causality and no comment about the benefit that the other 9,993 patients may have gotten.

    Yes, unfortunately tramadol is abused and diverted, but your assessment of the pharmacological relationship of the drug and 2D6 is entirely misunderstood. Opioids and Tramadol, like so many other drugs, are dosed to effect, that is why there are different pill strengths and different dosing intervals.

    Yes 2D6 adds some variability here, but seriously if we were going to avoid drugs because they are metabolized by 2D6 then we would never give almost any antidepressants, many antipsychotics, other opioids like hydrocodone and oxycodone and my favorite drug in heart failure carvedilol.

    Every drug has its inherrent risks and if you dont like tramadol, you shouldn’t give it out, but to damn it based on an association in 7 out of 10,000 prescriptions seems a little weak without assessing its benefits.

    What do you give to your patients on methadone or buprenorphine who have serious pain? How about people with true opoid allergy and long bone fractures.

    Let’s consider the entire picture and not throw out the drug based on an admitted rare association

  2. Leon Says:


    Excellent questions. I would say that my “must-read” recommendation was not so much for the article by Fournier et al on tramadol and hypoglycemia, but specifically to the comment by Juurlink and Nelson for the important points it makes about tramadol. The association with hypoglycemia which — from this article and previous case reports — I believe is real but rare, is interesting but may have limited clinical significance.

    The real question brought up by the editorial is whether — given the specific pharmacology and toxicology of tramadol — the drug possesses any advantage that would make it a reasonable choice among all available analgesics and pain-control regiments. Originally, tramadol was marketed as being uniquely safe and non-addictive. As we’ve learned, it’s really neither.

    I am certainly not against all use of opiates, even for chronic pain, if they’re used rationally and monitored carefully as part of a comprehensive pain-control program. I don’t think anyone would disagree with the statement that the vast majority of patients prescribed chronic opiates do not use them as part of such a program. That is why recently up to 15,000 persons in U.S. per year have died of prescription drug overdose, and many others have gone on to become addicted to heroin.

    The unpredictability of tramadol kinetics means that slow metabolizers will get little or no analgesic effect, while rapid metabolizers will be at increased risk of significant adverse effects. If opiates are truly indicated, other options such as morphine should be more predictable. As Juurlink and Nelson indicate, in many instances non-drug options can be effective. If a patient had severe pain such as a long-bone fracture and needed medication, but was truly allergic to opiates, I’d consult with a pain specialist.

  3. Ibrahim Al-Busaidi Says:

    Very interesting read.

    Maybe I should be careful when prescribing tramadol to my diabetic patients?

  4. Guy Weinberg Says:

    long bone fractures? consider local anesthetics! that is, regional anesthesia. Nerve blocks can be very effective for the acute pain episode and have a very favorable therapeutic index (notwithstanding LAST 😉

  5. Leon Says:

    Dr. Weinberg:

    Local nerve block would be very effective in these situations in the emergency department, and are being use (especially with ultrasound guidance) with increasing frequency. Of course, the problem of outpatient pain management remains.

  6. Leon Says:

    Dr. Al-Busaidi:

    I think there are many reasons for avoiding tramadol. Tramadol-induced hypoglycemia, although rare, can occur in both diabetic and non-diabetic patients. Just another reason to be careful of the drug in any patient.