The Effect of Surgery on Chronic Opioid Users Surprises

Author: Michael Vlessides

Anesthesiology News

Among chronic opioid users, having surgery seems to be associated with a faster time to opioid discontinuation—contrary to popular belief.

A study by Toronto researchers concluded that several factors were associated with reduced odds of opioid discontinuation in this patient population, including oxycodone use, higher opioid doses, and diagnoses of chronic obstructive pulmonary disease (COPD) and dementia.

“Of patients coming to our operating rooms, 23% will already be on an opioid by the time they see you on the day of surgery, and 3% will be chronic opioid users,” said Naheed Jivraj, MD, an anesthesiology resident at the University of Toronto. “However, we don’t know whether surgery moves the needle at all and changes the trajectory of opioid consumption among this cohort of chronic opioid users.

“Second, it’s unclear whether there are any risk factors associated with prolonged or continued opioid use in this population.”

For purposes of the investigation, chronic opioid use was defined as:

• an opioid prescription that overlapped the index date plus either 120 or more cumulative calendar days of filled opioid prescriptions; or
• 10 or more opioid prescriptions filled in the prior year.

Each surgical patient was then matched to three nonsurgical chronic opioid users by age, sex, Charlson comorbidity index and daily opioid dose. Multivariable models were used to adjust for a host of potential confounding variables and evaluate the association between surgery and time to discontinuation of opioids in the ensuing year.

Opioid discontinuation was defined as the longer of either:

• opioid-free for greater than twice the number of days supplied by the previous prescription; or
• 30 opioid-free days.

Finally, multivariable logistic regression was used to evaluate factors associated with the discontinuation of opioids in the year after hospitalization among surgical chronic opioid users.

As Dr. Jivraj discussed at the 2019 annual meeting of the Canadian Anesthesiologists’ Society (abstract 635242), the final cohort comprised 4,755 surgical patients and 14,265 matched nonsurgical patients, all of whom were identified as chronic opioid users.

“Of this cohort, patients were primarily receiving oxycodone and codeine,” Dr. Jivraj explained. “And nearly 40% of our chronic opioid users were coprescribed a benzodiazepine in the year before surgery.” The most common surgical procedures were carpal tunnel release, laparoscopic cholecystectomy and hysterectomy.

Of note, the study found that a greater proportion of patients discontinued opioids in the year after surgery than their counterparts who did not have surgery (36% vs. 29%; P≤0.001). The median time to discontinuation also was significantly shorter in surgical patients (133 days; interquartile range, 119-150 days) than for nonsurgical patients (236 days; interquartile range, 225-247 days; P≤0.001).

After adjusting for a variety of comorbidities and sociodemographic characteristics, the researchers found that surgery was associated with a 34% increased likelihood of opioid discontinuation (adjusted odds ratio [aOR], 1.34; 95% CI, 1.27-1.42).

Factors found to be associated with reduced odds of opioid discontinuation in the year after hospitalization included:

• average opioid dose greater than 90 morphine equivalents (aOR, 0.39; 95% CI, 0.32-0.49);
• filling a prescription for oxycodone before surgery (aOR, 0.73; 95% CI, 0.56-0.98);
• history of COPD (aOR, 0.75; 95% CI, 0.64-0.88); or
• history of dementia (aOR, 0.60; 95% CI, 0.38-0.95).

“On the other hand, patients who were taking codeine [aOR, 1.80] or tramadol [aOR, 2.32] instead of morphine before surgery, or patients who received an acute pain services consultation during their hospitalization, were more likely to discontinue opioids in the year of follow-up,” Dr. Jivraj explained. Patients at least 65 years of age also were 43% more likely to discontinue opioids in the year after surgery (P<0.001).

“In the same model we looked at all the surgical procedures that patients were undergoing,” he said. This part of the analysis revealed that patients who underwent procedures primarily indicated to reduce pain—such as carpal tunnel release or hemorrhoidectomy—were no more likely to discontinue opioids than those who came for an acute surgical presentation, such as laparoscopic appendectomy.

“Given that our findings were in contrast to the hypothesis that we generated, we had to start thinking about why,” Dr. Jivraj discussed. “The first thing we landed on is that surgery is probably some sort of light bulb event that changes patient behavior.

“Secondly, we didn’t really think about the physicians,” he said. “If you look at the opioid discontinuation literature, patients are almost always discontinued opioids by a physician. And our surgical patients were two or three times more likely to see a physician than those in the general population. And then, of course, surgery itself might just very well be actively treating these patients’ pain.”

The Pendulum Has Swung

Yet as Gregory Hare, MD, PhD, a professor of anesthesia at the University of Toronto, was quick to point out, discontinuing opioids may not necessarily be the holy grail that many clinicians think. “Is discontinuation of chronic opioids a good thing or a bad thing?” asked Dr. Hare, who was not involved in the study. “You may have people who are taking a Tylenol 3 [codeine-acetaminophen, Janssen] and they’re shopping, doing activities, and seeing their friends, and I don’t think that’s a problem.

“So I think these are important data, but it’s important to know who’s being harmed by being a chronic opioid user and who’s actually being helped, because I think the pendulum has swung way too far to one side,” Dr. Hare said.

“I agree with you in terms of the pendulum being swung too far in the other direction,” Dr. Jivraj replied. “We’re seeing that already. For example, we’ve seen that chronic opioid users whose medications have been discontinued will subsequently go to other sources, whether it’s on the street or otherwise, and it’s caused an increase in opioid-related mortality.”