Intravenous opioids and the risk of addiction in individuals with cancer

Clinicians worry that requests for or reliance on intravenous (IV) opioids for pain management may indicate drug-liking and evolve into an opioid addiction. In this study, Arthur et al.¹ implemented a high fidelity and well-blinded randomized crossover trial comparing a single dose of fast infusion (2 min) and slow infusion (15 min) of 1 mg of IV hydromorphone and measured drug-liking using a standardized assessment tool (the DEQ-5). The authors found that in patients without a history of opioid-seeking behavior, a single dose of 1 mg of IV hydromorphone had low drug-liking and low abuse potential. Furthermore, the rate of infusion did not impact drug-liking, although patients receiving the faster infusion reported more drowsiness.

The results of this study need to be tempered when applied to a real-world population. First, social desirability bias may have contributed to the negative result. Participants may have reported low scores of drug-liking because of the stigma attached to opioids. Second, although the study suggests a low abuse potential from two doses of IV opioids, it was not designed to measure the effect of IV opioids on drug-liking after repeated dosing, such as occurs during hospitalization. Third, drug-liking scores may be higher in real-world scenarios in which infusion rates are much faster than those in the study. Nurses often push IV opioids over 30 seconds or less,² and patient-controlled analgesia pumps are also programmed to infuse quickly (1 mg of IV hydromorphone is delivered over 24 s).³

Individuals with substance use disorders (SUDs) or a history of nonmedical opioid use (NMOU) were excluded from this sample. Although not sufficiently powered to report as an outcome, the subgroup analysis of the 15 patients at higher risk for NMOU potentially suggests that the abuse potential of IV hydromorphone may be clinically significantly higher for such individuals. The clinically important difference in the DEQ-5 scale is 10.⁴ Those at higher risk for NMOU scored approximately 20 points higher on the drug-liking score compared to the study average regardless of rate of infusion. Clinical practice could be highly impacted by extending this work into patients with moderate and high risk of opioid use disorder (OUD).

This study also highlighted the adverse effect profile of IV hydromorphone. Despite the relatively young age of the study population (average age, 54 years old), the frequency of oxygen desaturation was high but not different between infusion rates (24% slow rate and 28% fast rate). The frequency of drowsiness was also high but higher with the faster infusion rate (44% slow rate and 62% fast rate). One possible explanation may be the dose of hydromorphone in the study. The authors chose 1 mg of hydromorphone which is within—but at the higher end of—the Food and Drug Administration-approved range of starting doses (0.2–1 mg).⁵

This study provides support for the assertion that the majority of patients who receive IV opioids for cancer-related pain never develop addiction. However, we should continue to monitor for addiction in patients with cancer. The prevalence of nonmedical opioid use and opioid use disorder (equivalent to addiction) are higher in the cancer population than the general US population.^6-8 Several factors contribute to this disparity. First, tobacco and alcohol use are more common among those with substance use disorders, and these substances directly cause or contribute to the development of a number of types of cancer.⁹ Furthermore, those with substance use disorders may delay preventative care, leading to late-stage diagnoses.¹⁰ Individuals addicted to one substance are more likely to develop an addiction to prescribed opioids. Additionally, even without prior substance misuse or addiction, the stress of a cancer diagnosis may drive unhealthy coping strategies, sometimes culminating in addiction. Regardless of the presence or absence of addiction, treating cancer-related pain should be a priority in any patient, and opioids continue to be among the most effective treatments for moderate or severe cancer-related pain.¹¹ More research on the best practices in treating pain in cancer patients with a history of substance use disorders should be prioritized.

Oncologists and any clinicians who treat cancer pain should understand that exposure to an addictive substance alone is unlikely to lead to an addiction. Addiction is a multifactorial and multistep process whereby genetic and psychological factors are influenced by social factors, environment, physical or sexual trauma, adverse childhood experiences, and psychiatric conditions such as depression, anxiety, and post-traumatic stress disorder.^{12, 13} All of these influences—over time—contribute to the development of a craving–reward–withdrawal cycle that drives the neurobiological change in the brain and can eventually lead to addiction.¹⁴

Notably, physical dependence is not the same as addiction. Many cancer patients develop physical dependence on prescribed opioids and a minority develops worsening pain and a psychological dependence on opioids for relieving their pain.¹⁵ In both scenarios, opioid cravings are typically absent. In measuring drug-liking, Arthur et al.¹ are measuring a marker of the potential for the development of opioid use disorder, which is characterized by cravings and inability to control use, rather than simply physical or psychological dependence on opioids.

There are reasons other than addiction or drug-liking that hospitalized patients with pain may prefer IV opioids. Intravenous opioids provide faster analgesia than oral opioids. Patients whose pain is intertwined with anxiety, loss of control, or overextended coping strategies may be particularly inclined to prioritize the rapid action of IV analgesics and the ability to control their dosing through IV patient-controlled analgesia pumps. Patients may be experiencing anorexia or nausea that may impair absorption and make oral opioids less effective. Furthermore, individuals with OUD are more likely to have their pain undertreated, so requests for IV opioids may stem from inadequate analgesia rather than drug-liking.

Arthur et al.¹ have added to an important line of research in drug-liking and abuse potential of IV opioids in cancer-related pain. This study suggests that most people with low risk for OUD and cancer-related pain do not experience drug-liking from 1 mg of IV hydromorphone nor experience higher drug-liking with a faster rate of infusion. Taken together, the results suggest that a few doses of IV hydromorphone have a low abuse potential in low-risk patients. This provides some evidenced-based reassurance for our low-risk patients whose pain is better controlled with IV opioids, have a clear medical indication for IV opioids, and are tolerating them without side effects. This does not mean we do not have to worry about opioid addiction in patients with cancer-related pain nor should IV opioids be prescribed without a clear medical indication. With the global incidence and progression-free survival of cancer predicted to rise, the risk of the development of opioid misuse and opioid use disorder in those who have or have had cancer must continue to be addressed. However, everyone—regardless of risk of opioid misuse or history of addiction—deserves to receive standard of care pain management, including IV opioids when medically indicated.

The authors declare no conflicts of interest.