Clinical and Laboratory Hematologic Findings in Patients Receiving Repeated-Dose Injectable HPβCD-Diclofenac for Acute Postoperative Pain: Pooled Analysis of Two Randomized Controlled Phase III Clinical Trials

Objective: While the non-steroidal anti-inflammatory drugs (NSAIDs) represent an important option for the management of acute postoperative pain, their use can be limited due to potential safety concerns, including bleeding risks. This study examined the bleeding-related safety of injectable diclofenac formulated with hydroxypropyl-β-cyclodextrin (HPβCD-diclofenac) when used for postoperative pain management. Methods: Data from two randomized, double-blind, placeboand active comparator-controlled phase III trials were pooled. Patients in both studies received HPβCD-diclofenac, placebo, or the active comparator ketorolac via intravenous injection every 6 hours for ≤ 5 days following abdominal/pelvic or orthopedic surgery. Bleeding adverse events (AEs) were evaluated through the treatment period and follow-up (≤ 37 days), and relative bleeding AE risks (RR) were estimated. Changes in hematology laboratory values were also assessed. Results: Overall, 608 surgical patients received ≥ 1 dose of study medication. Bleeding AEs occurred in n=9/318 (2.8%) patients receiving HPβCD-diclofenac, n=8/142 (5.6%) patients receiving ketorolac, and n=4/148 (2.7%) patients receiving placebo. Over the period examined, HPβCD-diclofenac was not associated with increased bleeding AE RR versus placebo (1.05 [0.33, 3.35]; p=0.93), nor was ketorolac (2.08 [0.64, 6.77]; p=0.22). Bleeding AEs were predominantly mild or moderate in severity. No treatment-related bleeding AEs occurred in the HPβCDdiclofenac group (1 in both the placebo and ketorolac groups). Among the subset of patients receiving concomitant anticoagulants, bleeding AEs occurred in n=3/60 (5.0%) patients receiving HPβCD-diclofenac, n=2/29 (6.9%) patients receiving ketorolac, and n=0/35 patients receiving placebo. In the HPβCD-diclofenac group, postsurgical shifts to low hematocrit and hemoglobin occurred in 35.7% and 28.3% of patients, respectively (versus 31.4% and 21.5%, respectively, with placebo). Postsurgical shifts in platelet count were uncommon (<3.0% across treatment groups). Conclusions: While follow-up studies in larger populations are warranted, this analysis suggests that HPβCDdiclofenac may not present a significant incremental bleeding AE risk versus placebo when used for acute postoperative pain management.