Polypharmacy in Pain Management: A Clinical Primer to Drug Interactions

A patient is taking more than one medication, what could possibly go wrong?

The moment a patient takes a medication, there will be an effect, which we classify as efficacy and side effects. With every additional medication, the odds of a side effect, and/or a drug-drug interaction increase.

In 2016, headlines reverberated across the US regarding the investigative reporting of the Chicago Tribune, which found area pharmacies to have missed potentially dangerous drug combinations half of the time. Members of the reporting team strolled into pharmacies with paper prescriptions for a statin and an antibiotic – a combination that could potentiate muscle soreness as part of the clinical presentation of the possibly fatal condition of rhabdomyolysis.¹

Federal budget legislation enacted in 1990 (OBRA) included a requirement for licensed pharmacists to offer patient counseling to all persons receiving a dispensed medication.² Yet, more than three decades later, most pharmacy customers are simply asked to “sign here” when picking up their medication, thus waiving their right to such counseling. Unfortunately, what could have led to proactive opt-out counseling, has created an opt-in scenario.

Considering the federally mandated corresponding legal responsibility for prescribers and dispensers alike to ensure a proper diagnosis and scope of practice,³ one can easily see how these concerns extend to all healthcare professionals, and obviously, to every single patient. Fortunately, drug interaction screenings are on the rise, often occurring before a prescription is even written. This process includes reviewing a patient’s entire medication profile, alongside their medical history, sex, age, and more.

Herein, I review how we can engage polypharmacy going forward – at its best.

To properly screen for drug-drug interactions, it is important to be up to date on potential considerations. Table I lists these considerations broadly, each described below.

Phase 1 metabolism includes oxidation with the CYP-450 enzyme system, reduction, and hydrolysis. Phase 2 includes methylation, glucuronidation, acetylation, sulfation, and conjugation. Phase 3 wraps things up with ATP-binding cassette (ABC) and solute carrier (SLC) transporters moving substances out of the body.

Hepatic cytochrome P450 (CYP-450) enzymes exist in the liver, small intestine, lungs, placenta, and kidneys. The CYP-450 enzymes of particular concern with multiple pain medications, such as opioid medications, include 2D6 and 3A4. These are synonymous with numerous other non-pain management and adjuvant pain medications and, thus, need to be specifically considered.⁵

NSAIDs

NSAIDs are generally known to be metabolized by the CYP-450 system, primarily by CYP-2C9, yet that is not the case for all NSAIDs. NSAIDs metabolized by CYP-2C9 include celecoxib, flurbiprofen, ibuprofen, indomethacin, and piroxicam; along with diclofenac (also metabolized by CYP-3A4), meloxicam (also metabolized by CYP-3A4), and naproxen (also metabolized by CYP-1A2). NSAIDs metabolized by means other than CYP-450 include etodolac, ketoprofen, ketorolac, and oxaprozin, along with the prodrugs of sulindac and nabumetone.⁵

Thus, when ACEs/ARBs are combined with NSAIDs, there is a pre- and post-kidney arterioles reduction in GFR, which can be concerning for some patients. In cases where ACEIs/ARBs and NSAIDs are deemed to be appropriate for a respective patient, there will be a recommended close monitoring for edema, blood pressure changes, and potassium changes.⁵

TCAs

Tricyclic antidepressants (TCAs) are classified as secondary and tertiary amines based on their chemical structures. Tertiary amine TCAs, including amitriptyline and imipramine, tend to produce more anticholinergic effects relative to their respective CYP-2C19 active metabolite secondary amine TCAs, including nortriptyline and desipramine. Each of these TCAs is ultimately further metabolized by CYP-2D6 to inactive metabolites.⁵ Careful review of any potential CYP-2C19 and/or CYP-2D6 inhibitors is warranted for optimizing patient care.

SNRIs

Serotonin norepinephrine reuptake inhibitors (SNRIs) have expanded as a medication class in recent years beyond venlafaxine and duloxetine, yet all still exhibit their effects via the two serotonin and norepinephrine neurotransmitters, as compared to only affecting serotonin for selective serotonin reuptake inhibitors (SSRIs). In potential CYP-450 drug-drug interaction, one must consider the respective CYP-450 inhibition of SNRIs. Venlafaxine and desvenlafaxine are both mild CYP-2D6 inhibitors, while duloxetine is both a mild CYP-1A2 inhibitor and moderate CYP-2D6 inhibitor, and levomilnacipran is a moderate CYP-3A4 inhibitor.⁷ Careful review of any potential CYP-2D6, CYP-1A2, and CYP-3A4 related medications is warranted for optimizing patient care.

Prescription opioids have sedative drug-drug interactions to consider (more on this below). They also come with CYP-450 related drug-drug interaction concerns primarily involving CYP-2D6 and CYP-3A4.

Codeine, tramadol, hydrocodone, oxycodone, and meperidine are metabolized by CYP-2D6, with codeine being metabolized to morphine, and hydrocodone being metabolized to hydromorphone to a small extent, along with tramadol being a prodrug being metabolized to the active metabolite O-desmethyltramadol. Fentanyl, methadone, oxycodone, and buprenorphine are metabolized by at least in some part by CYP-3A4, often with active metabolites.⁸

CYP-2D6 major inhibitors with a greater than five-fold increase in the area under the curve (AUC) or greater than 80% decrease clearance include bupropion, fluoxetine, and paroxetine to name a few. CYP-2D6 moderate inhibitors with a two-to-five-fold increase in AUC or a 50% to 80% decrease in clearance include duloxetine and terbinafine amongst others. Of note, CYP-2D6 is not inducible.⁹

Careful review of any potential CYP-2D6 and CYP-3A4 related medications is warranted for optimizing patient care for those utilizing prescription opioid medications.

P-glycoprotein (PGP) transporter functions to transport substances from hepatocytes to the intestinal lumen to be eliminated, while organic anion transporting polypeptide (OATP) functions to transport substances from the portal vein into hepatocytes along with assisting in the excretion from intestinal lumen. PGP inhibitors and OATP inhibitors have a net effect of increasing substrate levels.¹⁰ Colchicine and morphine are analgesic-related medications that are PGP substrates, which can be affected by PGP inhibitors and inducers.

• amiodarone
• azole medications
• azithromycin
• clarithromycin
• erythromycin
• felodipine
• quinidine
• ritonavir
• verapamil

PGP inducers include:

• carbamazepine
• phenytoin
• rifampin
• ritonavir
• St. John’s wort

Methotrexate and statin medications are common OATP substrates, which can be affected by OATP inhibitors and inducers.

OATP inhibitors include:

• cyclosporine
• cimetidine
• clarithromycin
• diclofenac
• erythromycin
• probenecid

OATP is not an inducible transporter.⁵ It is also important to realize that some medications can exert effects at both transporters and metabolic enzymes, causing increased and more complicated drug-drug interaction concerns.

A QTc interval is the time from the beginning of the QRS complex (ventricular depolarization) to the end of the T wave (ventricular repolarization). QTc intervals are shorter in individuals with a higher heart rate and longer with a lower heart rate. QTc intervals are prolonged when greater than 440 ms in men or greater than 460 ms in women, and there is a greater risk of Torsades de Pointes when prolonged to greater than 500 ms.

• hypokalemia
• hypomagnesaemia
• hypocalcemia
• hypothermia
• myocardial ischemia
• post-cardiac arrest
• raided intracranial pressure
• congenital long QT syndrome
• medications

Medications that can possibly cause QTc interval prolongation include:^11,12

• antiarrhythmics
• antibiotics
• antifungals
• antivirals
• antidepressants
• antiemetics
• antihistamines
• antipsychotics
• bronchodilators
• cardiac agents
• decongestants
• gastrointestinal stimulants
• oncology drugs
• opioids
• stimulants

Serotonergic medications include antidepressants (MAOIs, TCAs, SSRIs, SNRIs, trazodone, mirtazapine, and bupropion), opioids (methadone, tramadol, buprenorphine, fentanyl, meperidine, codeine, and dextromethorphan), along with linezolid, lithium, levodopa, triptans, metoclopramide, ondansetron, and cyclobenzaprine.¹³

Drug-drug interaction concerns consistently populate in software programs for any medications with serotonergic activity, yet the clinical relevance and mitigation is not as cumbersome as programs portray. Serotonin syndrome symptoms include:

• agitation/restlessness
• confusion
• tachycardia
• hypertension
• dilated pupils
• muscle spasm or spasticity
• sweating
• diarrhea
• headache
• shivering
• goosebumps

Patients can be educated to monitor for serotonin syndrome symptoms while utilizing medications with serotonergic activity and then to communicate with their healthcare professional(s) without hesitation. Essentially, if one is utilizing serotonergic medications and acutely feels like they have a cold on a Friday, they should not wait until Monday to contact their physician.

Sedative drug-drug interactions can be extremely concerning, particularly when compounded by the use of more than two medications with sedative effects. Perhaps the two most discussed drug-drug sedative interaction medications are opioids and benzodiazepines. Studies have shown an increased risk of respiratory depression (ie, opioid overdose) with concurrent utilization of benzodiazepines with opioids, with a 2017 study resulting in a doubling of the risk,¹⁴ and a 2015 study resulting in a tripling of the risk.¹⁵

Take, for instance, a quadruple sedative medication interaction involving oxycodone, alprazolam, carisoprodol, and zolpidem. The absolute requirement to review this multiple sedative medication interaction is undeniable, yet Micromedex classifies these interactions to be “major,” while Clinical Pharmacology classifies them as “moderate/major,” and Lexicomp says to “consider therapy modification,” and still yet, Facts and Comparisons classifies them “potentially severe or life-threatening.” These classifications do not say the same thing, warranting clinical judgment on the clinical relevance and actions to be taken (ie, not whether to take action, rather than what action to take).

In addition to drug-drug, pharmacokinetic, and pharmacodynamic interactions, there can also be interactions due to a difference in the patient’s genetic composition. Pharmacogenomics (PGx) is the study of the role of genetics in the human body’s response to medications. Numerous physiological systems of the human body, such as metabolic enzymes or drug receptors, can exhibit genetic variability resulting in altered drug responses. For instance, if a patient’s genetic composition facilitates a change in how a patient metabolizes a pain medication, then the selection and dosage of the pain medication may need altered.

Two of the most common CYP450 enzymes that have shown pharmacogenomic involvement with pain management medications are CYP-2C9 and CYP-2D6. CYP-2C9 substrates include pain medications such as ibuprofen and celecoxib, and CYP-2D6 substrates include codeine, dextromethorphan, tramadol, duloxetine, venlafaxine, and tricyclic antidepressants. Pharmacogenetic testing can be performed to help forecast a patient’s response to a given medication before initializing the treatment plan. For example, a patient who metabolizes CYP2D6 poorly may not receive adequate analgesia from tramadol, whereas a patient who metabolizes it ultra-rapidly may experience unnecessary side effects because of having more of the active metabolites present. A similar set of PGx concerns exists for the utilization of codeine, as illustrated in Table III.

Pharmacokinetic and pharmacodynamic medication differences exist among the male and female sexes. In respect to pharmacokinetics, lipophilic medications will exhibit a longer duration of action in fatty tissue, which has a higher body percentage in females, yet hydrophilic medications distribute into smaller volumes of distribution (V_d) in women producing greater effects (eg, ethyl alcohol).

Phase 1 metabolism (CYP-450) is more functional in females (eg, warfarin 2.5 mg to 4.5 mg less weekly is needed for females), while phase 2 metabolism is accelerated in males (eg, acetaminophen and caffeine). Female GFRs are typically 10% to 25% slower than male GFRs (eg, gabapentin and pregabalin are excreted unchanged thus have prolonged effects in females).

In respect to pharmacodynamics, females generally respond better to SSRIs compared to TCAs. A 2018 study in mice showed females exhibiting migraine pain while males did not after both were given CGRP medications;¹⁷ and females tend to exhibit a greater opioid analgesic response.¹⁸ All of these considerations contribute to healthcare professional decisions in respect to drug-drug interaction concerns in patient care.

As humans age, there are significant anatomical and physiological changes that occur, which ultimately can affect safe and effective medication utilization.

Anatomical and physiological changes in aging include:¹⁹

• cardiovascular (heart wall thickens, heart rate decreases and systolic blood pressure increases)
• pulmonary (chest wall thickens and central airways widen, leading to decreased pulmonary flow)
• CNS (brain size and blood-brain barrier decrease)
• renal (kidney size and GFR decrease)
• hepatic (liver mass and CYP-450 function decrease)
• immune (entire immune system function decreases)
• gastrointestinal (gastric emptying frequency decreases and gastric emptying time duration increases)
• overall body changes (body water to muscle ratio decreases and body fat increases)

The American Geriatric Society (AGS) Beers Criteria provide guidance on how to best utilize pharmacotherapy in the elderly population (≥ 65 years). The 2023 AGS Beers Criteria are divided into sections with tables including potentially inappropriate medications to avoid (PIMs, Table 2), PIMs due to concurrent disease/syndrome (Table 3), PIMs to be used with caution (Table 4), Drug-drug interactions (Table 5), avoid or reduce dosage (kidney function, Table 6), and updates since last edition (Tables 7, 8, and 9).

• TCAs (all except doxepin ≤ 6 mg/day)
• paroxetine
• benzodiazepines (all)
• z-hypnotics
• meperidine
• spasmodic muscle relaxants (carisoprodol, chlorzoxazone, cyclobenzaprine, metaxalone, methocarbamol, orphenadrine)
• NSAIDs (all except celecoxib)

Of particular interest, as shown in Beers Criteria Table 5, drug-drug interactions to avoid in elderly patients include:²⁰

• anticholinergic medications (cognitive decline, delirium, and falls/fractures)
• opioids as well as benzodiazepines/gabapentinoids
• ≥ 3 CNS active medications (antidepressants, antipsychotics, antiepileptics, benzos, z-hypnotics, muscle relaxants, and opioids)

In respect to pain-management therapies, for many elderly patients, very few medications remain on this list besides acetaminophen and topical medications, emphasizing the need for intense collaboration between patient and healthcare professionals.

When navigating drug-drug interactions, the underlying thought may be one of concern and hesitation. All will be comforted to recall that medications are not the sole, nor need to be the focus point of all treatment options.

Resources

The gold standard resource for PGx medication concerns is the Clinical Pharmacogenetics Implementation Consortium (CPIC).

Additional resources include the FDA’s table of pharmacogenomic biomarkers and the Dutch Pharmacogenetics Working Group (DPWG) resource.

This author also recommends:

• A Clinician’s Guide to Common Drug Interactions in Primary Care (2020) by Eric Christianson
• The Top 100 Drug Interactions: A Guide to Patient Management (2021) by Philip Hansten and John Horn