Higher risk of adverse outcomes
As compared with codeine, tramadol use was significantly associated with a higher risk of mortality (13.00 vs. 5.61 per 1,000 person-years; hazard ratio, 2.31; 95% confidence interval, 2.08-2.56); absolute rate differences (7.37 per 1,000 person-years; 95% CI, 6.09-8.78), cardiovascular events (10.03 vs. 8.67 per 1,000 person-years; HR, 1.15; 95% CI, 1.05-1.27; ARD, 1.36 per 1,000 person-years; 95% CI, 0.45-2.36), and fractures (12.26 vs. 8.13 per 1,000 person-years; HR, 1.50; 95% CI, 1.37-1.65; ARD, 4.10 per 1,000 person-years; 95% CI, 3.02-5.29).
A subgroup and sensitivity analysis showed that the increased mortality risk associated with tramadol was significantly higher in younger persons vs. older ones (HR, 3.14; 95% CI, 1.82-5.41 vs. 2.39; 95% CI, 2.20-2.60]; P < .001 for interaction). In addition, women had a significantly greater risk of cardiovascular events versus men (HR, 1.32; 95% CI, 1.19-1.46] vs. 1.03; 95% CI, 0.9-1.13]; P < .001 for interaction).
Potential for confounding
Weighing in on the data, Daniel Solomon, MD, MPH, chief of clinical sciences, division of rheumatology, Brigham and Women’s Hospital, and professor of medicine, Harvard Medical School, Boston, noted that because it is extremely unlikely that anyone will ever conduct a large, head-to-head safety trial comparing different opioids, the results of this paper are important to consider.
“However, as the authors appropriately caution, this type of analysis is limited by the strong potential for residual confounding,” he said. “In other words, even though the authors used state-of-the-art methods to limit imbalances between the patients initiating tramadol versus codeine, there is strong reason to believe that imbalances that may account for the differences in adverse events exist.”
For example, he noted that if one looks at the distribution of comorbid conditions in the before-matching group, tramadol initiators demonstrate a higher frequency of chronic kidney disease, diabetes, and overall chronic comorbid diseases. “This suggests to me that prescribers apply selection criteria when choosing who to prescribe which opioid,” Dr. Solomon explained.
“While the authors’ use of propensity score matching limits confounding, it only can improve balance for measured confounders,” he said. “Other factors not measured in this type of data set – blood pressure, pain, physical activity, tobacco use, body mass index – may still demonstrate imbalances even after matching.”
But after these limitations are taken into consideration, the results remain concerning, Dr. Solomon emphasized, particularly the all-cause mortality excess of tramadol versus codeine users. “This study did not include cause of death, which would help the reader understand why users of tramadol were dying more frequently,” he added. “It also might help in understanding whether this is a true biologic effect or residual confounding.”
Perceived safety
In an accompanying editorial, Howard S. Kim, MD, MS, and colleagues from Northwestern University, Chicago, write that the greatest risk of tramadol may involve the perception that it is “inherently safer than other opioids.”
“In actuality, the mechanisms of action and variable metabolism of tramadol in a given population create considerable therapeutic uncertainty and introduce additional risk exposure,” they say, as demonstrated in the current study.
Therefore, when clinicians determine that an opioid is needed for pain relief, it may be a better option to select a pure opioid agonist that has a more predictable therapeutic effect and known adverse effect profile, such as morphine or hydrocodone. “This would allow clinicians and patients to more properly weigh the risks and benefits of initiating opioid therapy through shared decision-making and prompt the level of counseling on safe use, storage, and disposal practices that all opioids deserve,” write the editorialists.
The study was funded by the Fundació Institut Universitari per a la recerca a l’Atenció Primària de Salut Jordi Gol i Gurina. The research was supported by the National Institute for Health Research Oxford Biomedical Research Centre. Dr. Reyes has disclosed no relevant financial relationships. Dr. Solomon disclosed salary support from research contracts to his hospital from Amgen, AbbVie, Moderna, the Rheumatology Research Foundation, and National Institutes of Health; and royalties from UpToDate. Dr. Kim reported unrelated grant support from the Agency for Healthcare Research and Quality.
A version of this article first appeared on Medscape.com.