Intravenous Acetaminophen for Pain after Major Orthopedic Surgery: An Expanded Analysis

Background and Methods: From the time that Sinatra et al. (Anesthesiology. 2005;102:822) was published to FDA apaproval of intravenous (IV) acetaminophen, an expanded analysis of the original raw study data became necessary for the regulatory submission. The following analyses were conducted: (1) sum of pain intensity differences over 24hours (SPID24) using currently accepted imputation methods to account for both missing data and the effects of rescue; (2) efficacy results after the first 6hours; (3) effects of gender, race/ethnicity, age, weight, surgical site, ASA Class, and serotonin antagonists; and (4) a stepwise regression analysis of why adverse events of nausea and vomiting were numerically (although not statistically) higher in the IV acetaminophen group compared with placebo. Results: Sum of pain intensity differences over 24hours using a 0- to 100-mm visual analog scale was statistically significantly (P<0.001) in favor of IV acetaminophen (n=49) compared with placebo (n=52). Time to rescue was found to be 3.9 and 2.1hours, respectively, for total hip and knee arthroplasty compared with 0.8hours for the placebo group. Rescue medication consumption, requests, and actual administration were all significantly lower in the IV acetaminophen group compared with placebo for each dosing interval, except in the 6- to 12-hours interval where a numerical trend was observed. Analysis of various subset variables demonstrated similar efficacy for each variable. A stepwise regression analysis demonstrated that AE reports of nausea and vomiting were most likely due to prerandomization events, particularly opioid consumption and presence of nausea prior to randomization. Conclusion: Repeated-dose 24-hours end points were found to be as robust as previously published results. IV acetaminophen efficacy and safety appeared to be unaffected by specific subset variables.