Effects of Patient and Surgery Characteristics on Persistent Postoperative Pain: A Mediation Analysis.

OBJECTIVE Acute postoperative pain intensity is associated with persistent postsurgical pain risk. However, it remains unclear whether acute postoperative pain intensity mediates the relationship between clinical factors and persistent pain. METHODS Subjects from a mixed surgical population completed the Brief Pain Inventory and pain catastrophizing scale before surgery, and the Brief Pain Inventory daily after surgery for 7 days and at 30 and 90 days after surgery. We considered mediation models using the mean of the worst pain intensities collected daily on each of postoperative days (PODs) 1 to 7 against outcomes of worst pain intensity at the surgical site endpoints reflecting persistent postsurgical pain (POD 90) and subacute pain (POD 30). RESULTS The analyzed cohort included 284 subjects for the POD 90 outcome. For every unit increase of maximum acute postoperative pain intensity through PODs 1 to 7, there was a statistically significant increase of mean POD 90 pain intensity by 0.287 after controlling for confounding effects. The effects of female versus male sex (m=0.212, P=0.034), pancreatic/biliary versus colorectal surgery (m=0.459, P=0.012), thoracic cardiovascular versus colorectal surgery (m=0.31, P=0.038), every minute increase of anesthesia time (m=0.001, P=0.038), every unit increase of preoperative average pain score (m=0.012, P=0.015), and every unit increase of catastrophizing (m=0.044, P=0.042) on POD 90 pain intensity were mediated through acute PODs 1 to 7 postoperative pain intensity. DISCUSSION Our results suggest the mediating relationship of acute postoperative pain on persistent postsurgical pain may be predicated on select patient and surgical factors.