Perioperative Corticosteroids: Not a Straight Shot Answer: Commentary on an article by P.K. Chan, FRCSEd (Ortho), et al.: "Pain Relief After Total Knee Arthroplasty with Intravenous and Periarticular Corticosteroid. A Randomized Controlled Trial".

Abstract

Commentary Perioperative corticosteroids have become an integral part of many postoperative recovery pathways because of their ability to decrease nausea and vomiting without the undesirable side effects of other antiemetics, many of which have anticholinergic activity that increases the risk of postoperative urinary retention. Not surprisingly, corticosteroids such as dexamethasone have largely replaced anticholinergics such as scopolamine as the antiemetic of choice following total hip and total knee arthroplasty. Corticosteroids also decrease postoperative pain, thus also allowing patients to more predictably mobilize early following total joint arthroplasty surgery. For the reasons mentioned above, it is not surprising that dexamethasone is now given perioperatively in approximately two-thirds of total joint arthroplasty procedures performed in the United States1. However, the optimal route of administration remains debated. To answer this question, Chan et al. performed a placebo-controlled, triple-blinded (i.e., patient, surgeon, and researcher), prospective randomized controlled trial to assess the optimal mode of delivery of perioperative corticosteroids. Patients undergoing primary total knee arthroplasty were randomized to receive (1) intravenous dexamethasone, (2) periarticular triamcinolone, (3) intravenous dexamethasone plus periarticular triamcinolone, or (4) a placebo. The authors assessed postoperative pain both at rest and with maximal flexion, opioid consumption, and functional parameters such as knee flexion, muscle power, and distance walked. They found that the combined intravenous and periarticular group consistently beat the placebo group in almost every outcome assessed, whereas the isolated periarticular and isolated intravenous groups were not consistently superior to the placebo group. However, when assessed head-to-head, the combined corticosteroid group was not superior to the isolated periarticular or isolated intravenous group for any of the outcomes assessed. The results of this study should be interpreted within the appropriate context. The authors concluded that combined intravenous and periarticular corticosteroids were associated with “more significant improvements in the rehabilitation parameters” than intravenous corticosteroids alone. However, on close review of the data, the post-hoc analyses failed to demonstrate significant differences between these 2 groups (i.e., no differences were observed between the combined intravenous-periarticular group and the isolated intravenous group). Furthermore, when assessing the post-hoc comparisons across all 3 intervention groups, the combined intravenous-periarticular group was not superior to either the isolated intravenous group or the isolated periarticular group for any of the pain or functional parameters measured, limiting claims about a superior mode of delivery. While the authors demonstrated that periarticular, intravenous, and combined intravenous-periarticular corticosteroids are all likely superior to a placebo, they failed to conclusively determine an optimal mode of delivery. Certainly, more data are needed before this question can be laid to rest. When considering the optimal mode of corticosteroid administration, surgeons should weigh several additional factors not directly assessed in the current study. Perioperative intravenous corticosteroids decrease the acute inflammatory cytokine cascade following surgery and have a stronger anti-thromboembolic effect than corticosteroids administered periarticularly2,3. Furthermore, while intravenous perioperative corticosteroids have the potential to exert a protective effect against complications such as deep vein thrombosis and pulmonary embolism1, periarticular corticosteroids administered locally in the surgical wound may predispose to complications such as wound-healing problems that may place patients at increased risk for deep infection4. While smaller studies regarding periarticular corticosteroid administration appear to demonstrate the safety of this practice, even the aggregated metadata of all randomized controlled trials are inadequately powered to detect differences in rates of periprosthetic joint infection5. In short, intravenous corticosteroids may have systemic effects that are protective, while high concentrations of local corticosteroids may have deleterious effects that increase risk. Larger studies that have adequate statistical power are needed to settle the issue.