NSAIDs are a diverse group of drugs broadly categorized into salicylates (e.g., aspirin), propionic acid derivatives (e.g., ibuprofen), acetic acid derivatives (e.g., ketorolac), oxicam derivatives (e.g., piroxicam), and the heterocyclics (e.g., celecoxib). See Table 15.1 for a list of common NSAIDs and their properties.

All NSAIDs act by inhibition of prostaglandin synthesis. Prostaglandins have important physiologic functions including the mediation of the inflammatory response, the transduction of pain signals, as well as a central antipyretic effect. Prostaglandins are derived from arachidonic acid via a reaction catalyzed by cyclooxygenase (COX) enzymes. By inhibiting COX enzymes NSAIDs block production of prostaglandins from arachidonic acid.

The COX enzymes are known to exist as three isoforms: COX-1, COX-2, and COX-3. COX-1 and COX-2 are the isoforms nonspecifically targeted by the traditional NSAIDs, while the more selective COX-2 inhibitors preferentially block COX-2. COX-1 is involved in normal physiologic functioning such gastrointestinal mucosal protection and hemostasis. COX-2, on the other hand, is inducible during physiologic stress by agents including pro-inflammatory cytokines, neurotransmitters, and growth factors. Although both enzymes are structurally similar and act in the same fashion, their respective gene expression profiles and selective inhibition can determine NSAID side-effects and toxicity.

Studies investigating short term NSAID use in cancer pain management consistently demonstrate a dose-related improvement in pain relief compared to placebo with no increase in side effects. However, the longer-term efficacy and tolerability of NSAIDs is not well established. In modern oncology practice, with often-prolonged survivorship, the sequelae of longer-term NSAID use, such as renal toxicity, gastric ulceration, and increased cardiovascular risk, should be considered [8].

Side by side comparison of NSAIDs, including the COX-2 inhibitors, fails to demonstrate superiority of pain relief from any one formulation. These studies also fail to show significant difference in side effects across the NSAID spectrum. Assuming there are no contraindications to NSAID use, including renal insufficiency or active gastrointestinal ulceration, NSAIDs should be trialed in most cancer pain patients. After initiating an NSAID, regular screening for efficacy and toxicity should be implemented, with discontinuation of the drug if inefficacy or toxicity is observed. Celecoxib, the only remaining COX-2 inhibitor on the US market, has less gastrointestinal toxicity during short term use, yet controversy remains about whether there is any difference, compared to nonselective NSAIDs, beyond 6 months. Finally, celecoxib enjoys the advantage of lacking an antiplatelet effect, but does have similar renal toxicity compared to nonselective NSAIDs.

The use of NSAIDs in management of cancer pain may be precluded by comorbidities or concurrent treatments such as chemotherapy. Side effects and toxicities of NSAIDs present in the general population are often augmented in patients being treated for cancer due to additional treatments or overall poor state of health. In particular, NSAIDs should be prescribed with caution in patients at increased risk for renal, gastrointestinal, or cardiac toxicities as well as those with bleeding disorders or thrombocytopenia [9].

All NSAIDs can transiently decrease renal function in selected patients, resulting in hypertension, edema, and even acute renal failure. Patients at elevated risk for renal toxicities from NSAID treatment include: age greater than 60 years old, compromised fluid status, interstitial nephritis, papillary necrosis, and concurrent administration of nephrotoxic drugs, including cyclosporine and cisplatin. Further, any chemotherapy drugs excreted renally elevate the risk for toxicity.

If the patient’s serum creatinine is elevated or shows a trend towards elevation, the NSAID should be discontinued.

The chronic use of NSAIDs inhibits production of prostaglandins that maintain normal gastrointestinal mucosal integrity, and results in gastric and colonic mucosal damage including erosion and ulceration. Patients at increased risk for GI toxicity include: age greater than 60 years old, history of peptic ulcer disease or significant alcohol use, major organ dysfunction (including hepatic), and use of high-dose NSAIDs for a long duration.

If a patient develops mild to moderate gastric symptoms (dyspepsia, abdominal pain, nausea) discontinuation of NSAIDs should be considered. Alternatively, the patient could be switched to a COX-2 inhibitor due to their lower incidence of GI side effects. Drugs that decrease gastric acidity including antacids, H2 receptor antagonists, or proton pump inhibitors may ameliorate GI side effects.

Certain gastrointestinal conditions should prompt immediate discontinuation of NSAIDs. These include presence of gastrointestinal peptic ulcers, gastrointestinal hemorrhage, and an increase in liver function studies 1.5 times the upper limit of normal.

All NSAIDs, including COX-2 specific agents, can increase the risk of serious cardiovascular thrombotic events such as myocardial infarction and stroke. In addition, NSAIDs may increase blood pressure, and this is likely linked to the increased cardiovascular risks associated with their use. Patients at increased risk for cardiac toxicities include: prior history of cardiovascular disease or those at increased risk for cardiovascular disease or complications due to factors such as a history of smoking or known family history. Further, patients currently using anticoagulants are at significantly increased risk for bleeding complications when placed on concurrent NSAID therapy—NSAID use should be avoided in these patients.

Bone healing is dependent upon an inflammatory response involving numerous cytokines and fibroblast growth factor, so it should not be surprising that an agent that disrupts normal cytokine function may impair bone homeostasis and repair [10]. In fact, this inhibitory healing response has been used therapeutically to prevent heterotrophic bone formation after arthroplasty [11].

However, data on detrimental effects of NSAID use in the perioperative period is somewhat conflicted and controversial [12]. The issue of bone healing and NSAIDs has been addressed most thoroughly in the spinal fusion literature. A retrospective analysis of 288 patients who underwent instrumented spinal fusion from L4 to the sacrum demonstrated a fivefold higher nonunion rate when ketorolac was used in the immediate postoperative period [13]. In direct contrast to this, another retrospective study was performed in which 405 consecutive patients who underwent primary lumbar spinal fusion—a subset of these patients who received ketorolac 30 mg intravenously every 6 h for 2 days had similar fusion rates to a group that had no NSAIDs [14]. A metanalysis of 5 retrospective studies explored the relation of ketorolac dose and successful spinal fusion rates, and concluded that high dose ketorolac (dose > 120 mg/day) may be associated with poor outcomes, whereas standard dose ketorolac (<120 mg/day) was not [15]. Considering the absence of any prospective or randomized studies as well as the high morbidity associated with bony nonunion, use of perioperative nonselective NSAIDs in spinal fusion cases should be considered carefully, particularly when other risk factors for poor bone healing (i.e., smoking) exist. In non-spine orthopedic surgery there is good evidence of NSAID analgesic efficacy without significant compromise of bone healing [12, 16].

Before initiating opioid therapy in any patient it is important to establish goals of care, i.e., short-term postoperative use versus longer-term management for chronic, cancer -related, pain. Patients must be screened for a personal history of drug abuse and serious mood problems. Urine drug toxicology should be considered if there is a suspicion for illicit drug use. Patients must be educated on the safe use, storage, and disposal of their medication [17].

Typically, short-acting opioids are prescribed first, on an as-needed basis. In many countries the short-acting opioids, such as hydrocodone and oxycodone, are formulated with acetaminophen so it is imperative that the total daily dose of acetaminophen does not exceed 4000 mg/day (see Table 15.2). If the acetaminophen dose ceiling is approached, changing to a non-acetaminophen containing product is advised.