Intensity of perioperative analgesia but not pre‐treatment pain is predictive of survival in dogs undergoing amputation plus chemotherapy for extremity osteosarcoma

Abstract

The purpose of this bi‐institutional retrospective study was to determine whether, in dogs treated with limb amputation and adjunctive chemotherapy for osteosarcoma, oncologic outcomes are impacted by either: (1) baseline cancer pain severity, or (2) the approaches used for perioperative pain management. Data were extracted from the medical records of 284 dogs that underwent both limb amputation and chemotherapy (carboplatin and/or doxorubicin) between 1997 and 2017 for localized (non‐metastatic) osteosarcoma of the appendicular skeleton. Kaplan–Meier survival curves and Cox proportional hazard (PH) models were used to determine the impact that retrospectively scored baseline pain levels (high vs. low) and various analgesic and local anaesthetic treatments had on both metastasis‐free survival and all‐cause mortality. For the entire population, the median disease free interval and median overall survival times were 253 and 284 days, respectively. Baseline pain was rated as “low” in 84 dogs, and “high” in 190 dogs; pain severity had no detectable effect on either metastasis‐free survival or all‐cause mortality. When accounting for the potential influences of known prognostic factors, dogs treated with what was characterized as a high‐intensity perioperative analgesic plan (including both a non‐steroidal anti‐inflammatory drug [NSAID] and a bupivacaine‐eluting soaker catheter placed at the amputation site) had a higher probability of survival than dogs treated with a low‐intensity perioperative analgesic plan (neither an NSAID, nor a soaker catheter); the median overall survival times were 252 and 378 days, respectively (hazard ratio: 2.922; p = .020).