After Pain From TKA and THA Is Gone, Opioid Use Remains

Study suggests factors other than pain may influence continued use.

Patients who are prescribed opioids to manage their joint pain after total knee and total hip arthroplasty (TKA and THA) procedures may remain on opioids long term, well after they have ceased to experience improvements in their pain, according to collaborators at the University of Michigan, in Ann Arbor, and the University of Cincinnati.

Publishing in Pain (2016;157:1259-1265), the investigators hypothesized that patients may continue taking opioids for other reasons, including pain elsewhere in the body, self-medication for emotional pain and opioid dependence.

The investigators conducted a secondary analysis of a prospective outcome study of 574 patients undergoing TKA and THA procedures between March 2010 and May 2013. The participants completed a series of validated self-report questionnaires preoperatively on the day of their surgery and were followed longitudinally to assess surgical outcomes, with postsurgical data collected at one, three and six months. Patients who reported no opioid use on the day of surgery were considered opioid-naive. The self-report questionnaires included standardized assessments of joint pain, overall body pain, depression and anxiety symptoms, and pain catastrophizing.

Using a multivariate logistic regression model, the investigators predicted the patients’ opioid use at six months from change in knee or hip pain and overall body pain from baseline to six months. Covariates included in the model were baseline pain scores, age, race, sex, opioid use at baseline and type of surgery.

The investigators found that, while overall body pain at baseline (adjusted odds ratio [aOR], 1.71; P<0.001) and opioid use at baseline (aOR, 9.27; P<0.001) were predictive of opioid use at six months, changes in knee or hip pain were not (aOR, 1.01; P=0.900). Moreover, they found that decreased overall body pain was associated with lower odds of being on opioids at six months (aOR, 0.68; P<0.001).

On the day of surgery, patients who persisted on opioid use at six months reported worse pain in their surgery site (P<0.001), greater functional impairment (P=0.002), more stiffness (P=0.001), increased overall body pain (P<0.001), more symptoms of depression (P=0.006) and higher levels of pain catastrophizing (P=0.001), according to a univariate analysis. A univariate analysis further revealed the probability of opioid use at six months increased with preoperative opioid dose, as measured in oral morphine equivalents.

First author Jenna Goesling, PhD, a clinical psychologist in the Department of Anesthesiology at the University of Michigan, said she and her co-authors were surprised by the findings.

“After a successful surgery, as pain in the affected joint improves, we would expect that opioid use would no longer be needed,” Dr. Goesling said. “Counter to our hypothesis, we did not find an association between changes in joint pain and persistent opioid use.”

Dr. Goesling said the next step is to replicate this study in other pain populations. “It is certainly possible that we will see similar results,” she said. “Perhaps we will find even higher rates of persistent use among patients undergoing procedures for low back pain.”

Maria Inacio, MD, a senior research fellow in the School of Pharmacy and Medical Sciences at the University of South Australia, in Adelaide, who has extensive research experience in this area—and more so in the area of joint replacement—said the study’s findings did not surprise her.

“The authors do a good job in concluding that persistent opioid use is much more complex than just pain in the joint undergoing surgery,” Dr. Inacio said. “Patients undergoing joint replacement have a lot of comorbidities, some that can lead to pain and affect their opioid intake.”

She added that the study’s conclusion that overall pain is associated with opioid use has been demonstrated in other patient populations, and that the same can be said of “most” of the risk factors the authors identified for continued opioid use at six months. “The conclusion that in order to address opioid use you must address the patient as a whole and not just the pain in one location likely holds true in other patient populations,” she said.

Dr. Inacio offered some criticisms of the study as well, suggesting that important differences between TKA and THA patients might warrant studying each population separately, and that better attention might have been paid to that and other potential confounding factors, such as comorbidities, body mass index and use of other analgesic medications. However, she said overall, “this is a well-written study and does a good job presenting a lot of complex information.”

Dr. Goesling said while opioids “are certainly an important part of acute pain management,” teaching patients how to manage pain with behavioral-based interventions is “an important next step in reducing opioid use following surgical interventions.”