Acute Pain Crisis Caused by Tramadol Remdesivir Drug–Drug Interaction

Introduction

Drug–drug interactions are a concerning problem in the United States and worldwide. Drug–drug interactions can be explained as pharmacodynamic and pharmacokinetic events. The former is a change in the sensitivity of body cells or tissues to a drug (often related to competition at the cellular or molecular sites of action), and the latter is a change in absorption, distribution, metabolism, or excretion of a drug due to an interaction. ¹ Interactions are a common challenge with tramadol, at least partially due to its reliance on the cytochrome P450 enzymes for metabolism. ²

During the COVID-19 pandemic, on May 1, 2020, the antiviral drug remdesivir gained emergency use authorization from its prior status as an investigational drug for infected patients hospitalized with severe COVID-19 infection. On August 28, 2020, the federal government granted expanded authorization to treat patients with remdesivir even without severe disease. On October 22, 2020, the Food and Drug Administration approved remdesivir for use in adult and pediatric patients (>12 years) hospitalized with COVID-19. ³ Once within cells, remdesivir's active metabolite competes with the adenosine triphosphate substrate utilized by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) RNA-dependent RNA polymerase, thereby interfering with normal replication of viral RNA. ⁴ Except for chloroquine and hydroxychloroquine, there are no data regarding known drug–drug interactions. ⁵

Tramadol is a centrally acting analgesic that is frequently utilized in the treatment of cancer and noncancer pain. Unlike other analgesics that have focused mechanisms of action, tramadol is multifaceted in function. It is a partial nonselective μ-opioid receptor agonist and inhibits norepinephrine and serotonin reuptake. ⁶ There are renal and hepatic considerations for dosing adjustments. ⁷ Tramadol is metabolized extensively by CYP2D6 and CYP3A4.^8,9 The manufacturer notes the following: “The effects of concomitant use or discontinuation of CYP3A4 inducers, 3A4 inhibitors, or 2D6 inhibitors with tramadol are complex. Use of CYP3A4 inducers, 3A4 inhibitors, or 2D6 inhibitors with tramadol hydrochloride requires careful consideration of the effects on the parent drug, tramadol, and the active metabolite, M1.” ¹⁰

In times of global emergencies, such as the COVID-19 pandemic, the federal government is authorized to sanction emergency use of drugs that are otherwise considered investigational or lacking clear established indications. Understandably, these drugs typically have less research conducted to assess for drug–drug interactions than drugs that follow the usual pathway of approval without emergency use authorization. ¹¹

Herein, we describe a case of a drug–drug interaction between tramadol and remdesivir. The potential interaction between remdesivir and tramadol has not been previously documented.

Case Description

A 75-year-old man presented to the emergency department with respiratory symptoms five days after exposure to a person with a known COVID-19 infection. The evaluation revealed COVID-19 pneumonia, and he was admitted to the hospital floor.

His past medical history was significant for spinal cord injury with incomplete quadriplegia, neurogenic bladder, hyperlipidemia, hypothyroidism, and chronic pain (due to spasticity and osteoarthritis). His home medications included a long-standing stable dose of tramadol (50 mg four times daily) for the preceding four years with no dose adjustments and adequate pain control. At baseline, he performed most activities of daily living (ADLs) independently, requiring assistance with bathing. He had difficulty ambulating but was able to stand and pivot. He used an electric wheelchair for most mobility. He was initiated on a five-day course of remdesivir 200 mg on day 1 followed by 100 mg daily thereafter for COVID-19 pneumonia.

On day two of remdesivir treatment, he developed severe pain in his legs bilaterally unresponsive to his usual tramadol regimen. He described the pain as typical in character to his usual chronic pain (a gnawing, aching sensation in bilateral thighs, knees, and calves). He rated the pain as 10/10 and reported no response to the 50 mg of tramadol that previously had been effective. An increased tramadol dose of 100 mg was also ineffective. He was no longer able to perform ADLs due to intense discomfort. Imaging showed osteolytic lesions in the lumbar spine consistent with disuse osteopenia related to his incomplete quadriplegia. Monoclonal gammopathy screening was negative. His renal and hepatic function was intact (Cr: 0.79 mg/dL estimated glomerular filtration rate [eGFR 88], total bilirubin 0.3 mg/dL, direct bilirubin <0.02 mg/dL, aspartate aminotransferase [AST] 40 U/L, alanine aminotransferase [ALT] 20 U/L).

Palliative medicine consultation was requested for assistance with pain management. On examination, he was acutely distressed and appeared to be in a pain crisis. He was tearful, moaning in pain, stating that tramadol no longer seemed to work. His legs did not demonstrate spasticity or tenderness to palpation or movement. He denied abdominal cramping and had no piloerection on examination. Based on the abrupt resurfacing of previously well-controlled pain with no alternate etiology found, we considered the possibility of drug interaction between tramadol and remdesivir. He had received no other medications known to interact with tramadol. He received two doses of intravenous fentanyl with brief mild improvement in his pain leading up to the palliative medicine team being consulted.

Our first recommendation was to subsequently switch to oral hydromorphone (2 mg four times daily) and there was near-complete relief of his pain. We continued hydromorphone until completion of remdesivir with one additional day for washout of remdesivir, and then he resumed his usual tramadol dosing. During his admission, he was initially maintained on his normal home medications at his usual doses of aspirin, baclofen, gabapentin, levothyroxine, polyethylene glycol, and the aforementioned tramadol. His usual home omeprazole what substituted with pantoprazole, pravastatin was switched to atorvastatin, and ranitidine was switched to famotidine per hospital formulary. His usual home vitamins (vitamins C, D, and E) were held during the hospitalization.

New hospital medications he received before the onset of pain while hospitalized included atorvastatin, enoxaparin, remdesivir, topical diclofenac gel, famotidine pantoprazole, ondansetron, tamsulosin, influenza immunization, iohexol, senna-docusate, ceftriaxone, and cefepime. New medications he received after the onset of pain included senna-docusate, bisacodyl suppository, docusate-benzocaine enema, acetaminophen, hydromorphone, fentanyl, ondansetron, and potassium chloride. Phone follow-up after dismissal revealed that he was back on his home tramadol dosing and had returned to his baseline of adequate pain relief.

Discussion

This patient's clinical presentation is most consistent with a drug–drug interaction between tramadol and remdesivir. His previously well-controlled pain was suddenly no longer controlled, with no alternative etiology found to explain the pain exacerbation. His pain management returned to baseline once he was no longer taking remdesivir. Once consultation was requested with our palliative medicine team, we sought to avoid agents with phase 1 metabolism such as tramadol and fentanyl (both previously trialed before our consultation); while both morphine and hydromorphone are metabolized by phase 2, the latter was selected by institutional preference.

Tramadol is a centrally acting analgesic with a complex and incompletely understood mechanism of action. It is believed that the mechanism for analgesia is primarily related to its μ agonism with the serotonin and norepinephrine reuptake inhibition contributing to the descending pain modulation pathways. ¹² Pertinently, tramadol is a prodrug that requires metabolism to be effective.

Tramadol is metabolized to its O-desmethyl-tramadol (M1) metabolite through the CYP2D6 pathway. ¹² Approximately 80% of tramadol is metabolized by CYP2D6. The M1 metabolite is a more potent μ-opioid receptor agonist and is thought to be the primary component responsible for pain management. The M1 metabolite has a strong affinity for the opioid receptor, whereas the M2, catalyzed by CYP2B6 and CYP3A4, and other metabolites have only a weak affinity. However, the M5 metabolite, N,O-desmethyltramadol, is also considered an active metabolite with analgesic effects.^13,14 The M1 metabolite is further metabolized by CYP3A4 and CYP2B6 to the M5 metabolite. ¹⁵

Individuals who have CYP2D6 polymorphisms resulting in a CYP2D6 poor metabolizer phenotype are at risk for decreased tramadol efficacy due to decreased conversion to the active M1 metabolite. In a pharmacokinetic analysis, the bioavailability of the active M1 metabolite was dramatically reduced in CYP2D6 poor metabolizers versus those with normal CYP2D6 metabolism (3% vs. 63%). ¹⁶ CYP2D6 is an enzyme that is easily saturated, making up only 1%–5% of liver cytochrome P450 content, yet it metabolizes approximately one-quarter of medications. ¹⁴

Based on in vitro data, remdesivir is a substrate for CES1, Cathepsin A, CYP2C8, CYP2D6, CYP3A4, OAPT1B1, and P-gp.^5,17 Remdesivir is an inhibitor of CYP3A4, OATP1B1, OATP1B3, BSEP, MATE1, MRP4, and NTCP.^5,18 There are no data regarding in vitro testing to determine if remdesivir also inhibits CYP2D6.^5,17 Therefore, it is unknown if remdesivir or its two metabolites have any effect on CYP2D6. ¹⁰ Given that remdesivir is also CYP2D6 substrate, a possible mechanism of this suspected interaction could be through CYP2D6 inhibition through pharmacodynamic competitive inhibition from remdesivir, thereby reducing the conversion of tramadol to its active metabolite.

It is unknown if this patient had any CYP2D6 partial deficiency (e.g., CYP2D6 intermediate metabolizer phenotype) that may not have been enough to interact with tramadol alone. However, if CYP2D6 were further inhibited, this could be a potential mechanism for exacerbation of the apparent drug–drug interaction. Of note, the patient was not administered any CYP inducers nor any known CYP2D6 inhibitors at the time of remdesivir administration.

Drug–drug interactions increase as polypharmacy increases. In addition, drug–gene interactions may exist at baseline through inherited genetic polymorphisms. These drug interactions possibly may be predicted when detailed preapproval testing for interactions have been performed, genetic polymorphisms of metabolic pathways for a given patient are known (i.e., through pharmacogenetic testing), or when postmarketing reactions have been reported.

For drugs that are new or have had less preapproval testing (such as emergency use approval drugs), one must have a high index of suspicion of drug–drug interaction and should collaborate with pharmacists to help identify potential mechanisms and clinical relevance of pharmacogenomic testing results. In this patient's case, the timing of onset and offset of the proposed interaction is convincing and additionally, although not perfect, more objective tools such as the Adverse Drug Reaction Probability Scale by Naranjo et al. ¹⁹ demonstrates that the interaction is “probable.” To our knowledge, this is the first report of a potential drug–drug interaction between tramadol and remdesivir.