Are Psychological Predictors of Chronic Postsurgical Pain Dependent on the Surgical Model? A Comparison of Total Knee Arthroplasty and Breast Surgery for Cancer

Abstract

Anxiety, depression, and catastrophizing are generally considered to be predictive of chronic postoperative pain, but this may not be the case after all types of surgery, raising the possibility that the results depend on the surgical model. We assessed the predictive value of these factors for chronic postsurgical pain in 2 different surgical models: total knee arthroplasty for osteoarthritis (89 patients, 65% women, age = 69 ± 9 years, baseline pain intensity = 4.7 ± 2.1) and breast surgery for cancer (100 patients, 100% women, age = 55 ± 12 years, no preoperative pain). Data were collected before surgery, then 2 days and 3 months after surgery. Anxiety, depression, and catastrophizing were measured with the Spielberger State-Trait Anxiety Inventory, Beck Depression Inventory, and Pain Catastrophizing Scale, respectively. Pain was assessed with the Brief Pain Inventory. Neuropathic pain was detected with the DN4 questionnaire. Multivariate logistic regression analyses for the total knee arthroplasty and breast surgery models considered together indicated that the presence of clinically meaningful chronic pain at 3 months (pain intensity ≥3/10) was predicted independently by age (P = .04), pain intensity on day 2 (P = .009), and state anxiety (P = .001). Linear regression models also showed that pain magnification, one of the dimensions of catastrophizing, independently predicted chronic pain intensity (P = .04). These results were not affected by the surgical model or by the neuropathic characteristics of the pain. Thus, state anxiety and pain magnification seem to constitute psychological risk factors for chronic postsurgical pain relevant in all surgical models. PerspectiveThis prospective study performed in patients with total knee arthroplasty or breast surgery for cancer shows that state anxiety, amplification of pain, and acute postoperative pain independently predict postsurgical pain at 3 months and that this does not depend on the surgical model.