Primary Care Corner with Geoffrey Modest MD: New CDC Guidelines for Opiate Prescribing

By Dr. Geoffrey Modest

The CDC recently came out with their formal guidelines on when to initiate or continue prescribing opiates (see!documentDetail;D=CDC-2015-0112-0002​ ). Of note, it is commented on many times in the publication about the lack of high-quality trials, but that there was an urgency to develop these guidelines given the increasing risks of opioid misuse/adverse effects, and that the “contextual evidence” reviews, complementary information that assists in the translation of clinical research into recommendations, “provides indirect evidence and should be interpreted accordingly”.


  1. Nonpharmacologic therapy and nonopioid meds are preferred for chronic pain; and if opioids are used, they should be combined with these interventions as appropriate. There are minimal data supporting chronic opiate benefits, so hard to recommend given their known risks, except for this little caveat in the guidelines: “no study of opioid therapy versus placebo, no opioid therapy, or nonopioid therapy for chronic pain evaluated long-term (>1 year) outcomes related to pain, function, or quality of life. Most placebo-controlled randomized trials were ≤ 6 weeks in duration”. So, no data really. There is a comment that it’s okay for end-of-life care (commenting that “evidence of long-term opioid therapy for chronic pain outside of end-of-life care remains limited”), which does suggest there may be benefit (and, I’m not sure what the real difference in subjective pain is, comparing those at end-of life and those not). My point is that there are basically no data, that in my experience there are patients with really bad chronic pain who pretty clearly benefit from opiates and sometimes higher doses, and this puts us providers in a bind. There is no question to me that trying nonpharmacologic therapy is really important (PT, weight loss in those with knee pain etc., massage/manipulation, psych therapy and esp. cognitive behavioral therapy, exercise…., and combinations of these). And that non-opioid therapies often help (acetaminophen, etc… though I am concerned that prolonged NSAID use has its very real problems for the GI tract, kidneys, and heart especially, and significant mortality), but I should add that some of these drugs (e.g. salsalate, trilisate) are off the Medicare-approved list for unknown reasons, are much more benign, but in my experience sometimes work really well. And local steroid injections often give reasonably long-term relief (e.g. joint injections, trigger point injections) in my experience, as well as other adjuvant meds such as tricyclics, anticonvulsants (pregabalin, gabapentin, carbamazepine), and SNRIs. I would further reinforce avoiding opiates unless really needed, the above often work, and it is clear that opioids do have significant harms (abuse/overdoses, MIs, car accidents…)


  1. Before starting opioids for chronic pain, prescribers should establish realistic treatment goals with patients in terms of pain and function, including discontinuing them if benefits do not outweigh risks. This applies to pain lasting >3 months or past the time of normal tissue healing


  1. Before starting and periodically during opioid therapy, clinicians should discuss risks and realistic benefits of continued use, as well as the patient and provider responsibilities for managing therapy. This includes safety issues, which might be uncovered by looking at the prescription drug monitoring program (PDMP). Again, the issue is: unknown benefit (i.e., not studied, though the patient may attest to the benefit) but clear risks (and there are data showing that patients who have multiple providers prescribing the opioids are at increased risk). And contextual evidence suggest that many patients do not understand the safety issues.


  1. When starting opioids, use immediate-release ones, not the extended-release ones. The latter have likely increased potential for overdose, and, as I have mentioned in earlier blogs, there really are no data showing that long-acting ones are better, either more effective or safer. Also, if you decide to switch from short to long-acting and are switching opioids, remember that there is incomplete cross-tolerance, so the dose of the long-acting med should be reduced. Also, given the above, they recommend NOT giving long-acting along with short-acting opiates (this is pretty different from the old model: give long-acting to get steady state of opiates, then short-acting for “breakthrough” pain). Also given the [actually not-so-scientific] data finding more deaths with methadone, that should not be the first agent to use for a long-acting one.


  1. Use the lowest effective dose of opiates. And especially if increasing the dose to >50 morphine milligram equivalents (MME)/day. And generally “should avoid increasing the dosage to ≥​​ 90 MME/day”. One interesting contradiction is that methadone maintenance programs often have people above 100mg methadone/day (that is, >300 MME) for the long-term. In fact I have a chronic pain patient who is in a methadone program and on 70mg for the past many years. Given the presumed benefit of TID dosing of methadone for chronic pain, I appealed to the Medicaid program in Massachusetts so I could give him 70mg of methadone in divided doses at the health center but was unable to get approval for more than 60mg, the Medicaid max. However, I was told I could give him 60mg of methadone and an almost unlimited amount of oxycodone along with it (?!?). The dosage restriction suggested above, as pointed out in prior blogs, comes from ecological data showing that those on higher doses (e.g. >100 MME/day) have higher risk of overdoses and deaths. But, again, there are NO (as in, zero) randomized controlled studies looking at the benefits of higher vs lower doses. And I certainly have some chronic pain patients who are on high doses for a long time and who are very willing to take risks in order to get “better pain relief and function”, from their perspective (part of the issue in those requiring higher MME’s is that they may have genetic variants in mu receptors and need higher doses to get an effect– see past blogs as listed at the end). Also, we should consider giving naloxone kits to patients on opioids in case there is an overdose, esp. if they are on higher doses of opiates.


  1. Chronic opiate use begins with acute pain therapy. We should also give the lowest dose possible and shortest duration of immediate-release opiates. (And ERs should not blithely prescribe opiates). “3 days or less will often be sufficient; more than 7 days will rarely be needed”. The 3-day suggestion is largely “expert opinion”, though there was a study in patients with acute low back pain showing that there was usually a significant decrease in pain by the 4th day. And another one I blogged on recently (see blogs below) not finding that opiates were in fact better than nonopiates for acute low back pain. And the guidelines reinforce not giving ER/LA preparations for acute pain.


  1. Evaluate benefits and harms within 1-4 weeks of starting opiates and at least every 3 months thereafter, more frequently if a history of substance use disorder, of overdose,  if taking ≥​​ 50 MME/d, or if taking other CNS depressants. There may be utility to using validated scales to assess function, pain control and quality of life (e.g. PEG scale–Pain average, interference with Enjoyment of life, and interference with General activity). The recommended rate of tapering doses is not clear, some suggesting rapid tapers over 2-3 weeks in those with severe adverse events (e.g. overdose), others recommend slower tapers at 10%/week.


  1. Before starting and periodically thereafter, evaluate risk factors for opioid-related harms. Include considering offering naloxone
  • Patients with disordered breathing: the issue is opiate-related respiratory depression. Those with moderate-to-severe sleep-disordered breathing should probably not have opiates
  • Pregnant women: avoid initiating opiates during pregnancy, since they are associated with stillbirth, poor fetal growth, pre-term delivery, neonatal opioid withdrawal syndrome and birth defects. And for those pregnant and on opiates chronically, be careful about tapering (risks to patient and fetus if patient goes into withdrawal). Also a potential issue with breast-feeding: neonatal toxicity and death have been reported when mothers take codeine
  • Patients with renal or hepatic insufficiency — use more caution and increased monitoring, given decreased processing/clearing of drugs
  • >65 yo: opiates may be more dangerous, given reduced renal function. Also, more opiate-related confusion.
  • Mental health issues: untreated depression could lead to overdoses (suicide, or confusion). Anxiety treated with benzos adds toxicity when given together with opiates. And, though not mentioned in the recommendations, those under stress or not sleeping well experience more pain (i.e., best to try to help with underlying issues here)
  • Patients with substance use disorders — illicit drugs and alcohol increase likelihood of opioid-related overdose deaths
  • And, consider giving naloxone to those who are at higher risk of overdose


  1. Review the PDMP to see if patient is receiving high dose opiates or other meds that put him/her at higher risk. This should be done at least every 3 months. (Though I would add that there are a few problems here: the pharmacy data are not updated as quickly as they should be; navigating the website is not easy and one has to click on the same patient many times if they list different addresses; hard to get data on patients who go to other states for opiates or gets them through the VA system; and it really takes a lot of time doing so in a busy primary care session (the issue of chronic pain management really is, to me, the poster-child for team-based care: a clinician just does not have the time to do everything, especially since we typically have so many complex medical and psychosocial issues to take of in mostchronic pain patients. A team-approach with nursing and others involved in the voluminous paper work, checking PDMP, calling patients back for urine tests/pill counts, etc., is really essential to practicing high quality care).


  1. Check urine drug screen prior to starting opiates, and “consider” doing them at least annually thereafter. These are important for a variety of reasons, including patient safety


  1. Avoid prescribing opiates if patients are on benzodiazepines. Based on a lot of observational data, but as pointed out in some prior blogs, those on benzos by themselves may have underlying psych conditions which have significant mortality associated. But the opiates and benzos in combo are likely to produce more respiratory depression. In stopping the benzos, very important to taper slowly (e.g. decrease dose not more than 25% every 1-2 weeks)


  1. Arrange treatment for patients with opioid use disorder (e.g. with methadone or buprenorphine as meds; in combination with behavioral therapies). I believe that all of us who prescribe buprenorphine are very impressed with the results in the majority of patients… I really feel it is one of the few interventions I do which really give patients back their lives. And, given the huge benefits and significantly decreased risks of buprenorphine, I can see no reason why nurse practitioners/physician assistants/medical residents should not be able to prescribe buprenorphine, both because it is so effective in so many people, and, also, ironically, these providers are allowed to prescribe much more potentially dangerous meds anyway (oxycontin, methadone etc.)


So, my bottom line is that there is no doubt that opioids can be dangerous both to the patient and society (through diversion, availability in the streets, overdoses, crime), with cited statistics of 165,000 people dying from overdoses related to opioid pain medication in the US from 1999-2014, and >420,000 ED visits related to opioid pain relievers in 2011. And this danger is very likely increased with higher doses of opiates, or their combos with other meds (e.g. benzos). But there really are very little scientific data to inform these guidelines, making it hard for us in the trenches to accept the “expert opinion” when we have patients in front of us with inadequately treated chronic pain and severe functional impairment from that pain. And, I think, pain is pain, whether it is in cancer patients, those at the end-of-life, or those who fall off a ladder. So, I am a strong advocate for pretty much all of the above guidelines, especially trying to avoid opiates whenever possible, using adjunctive therapies including injections, trying to avoid benzos, giving the lowest opiate dose possible, educating patients on risks and benefits (And emphasizing that the benefit of opiates is rarely complete or near-complete pain relief). And I have even had several patients come off chronic opiates, some having been on them for years. But, there is no question in our practice, this issue of treating chronic pain is remarkably common and remarkably difficult (and remarkably hard to do in the context of a brief primary care visit, where we also deal with their depression/psych issues, hypertension, homelessness or other profound social issues, diabetes, illicit drug use, domestic violence, ……..). And there is no question to me that some patients do need higher doses of opiates to function, whether they have cancer or not.


A couple of other comments on the guidelines:


  • The estimates of the risk of opioid addiction are lower than previously believed/promulgated: in pain clinic settings, the rate of addiction is 2-14%. In primary care settings in patients on chronic opioids it is 3-26%.
  • Although methadone has been singled out as a particularly bad actor, and now requires more stringent prior authorization by Massachusetts Medicaid, the actual data are not so clear: several observational studies (e.g. Oregon Medicaid) found no increased risk of death or overdose, and a VA study found lower overall risk with methadone, despite the concern about prolongation of QTc intervals. another recent study did find increased risk of overdose with methadone, with “twice as many single-drug deaths as any other prescription opioid”


For other blogs: shows that teens given legit prescribed opiates are more likely to misuse opiates later in life. There are also studies in adults having cataract surgery or varicose vein stripping, finding that those still using opioids within 7 days of surgery had higher risk of use after 1 year (cannot show causal relationship in this type of study, but does raise the potential benefit of avoiding or minimizing opioid use unless absolutely necessary, even for short-term acute pain) finding unclear benefit of giving opiates which includes many of my comments about the lack of studies on opiates and the risks of developing strict guidelines in their absence (many more comments than above)​ which looks at some genetic variants (e.g. in the mu receptor) and their effects on individual’s drug use

(Visited 1 times, 1 visits today)