When nonpharmacological interventions are failing or do not provide adequate pain control, initiating oral medications, such as acetaminophen, is typically a first-line therapy for symptom relief (i.e., pain, stiffness, swelling) related to knee osteoarthritis.² The American College of Rheumatology (ACR) and the Osteoarthritis Research Society International (OARSI) guidelines for nonsurgical management of knee osteoarthritis recommend the use of acetaminophen for those patients who have failed nonpharmacologic treatment, have no relevant comorbidities, or have no contraindications to use.³,⁴ One should consider short-term use of acetaminophen due to the potential for adverse events, such as elevated liver enzymes, hepatotoxicity, and organ failure.⁵ If acetaminophen is not providing adequate pain relief, other oral agents should be considered.

Nonsteroidal anti-inflammatory drugs (NSAIDs) have been used for treatment of osteoarthritis for several decades. They are recommended for the management of knee osteoarthritis after use of acetaminophen and nonpharmacologic interventions have failed. NSAIDs function to reduce inflammation and pain by reversibly inhibiting cyclooxygenase (COX), the enzyme involved in the pathway of prostaglandin synthesis.⁶ NSAIDs can be categorized into nonselective and COX-2 selective. Nonselective NSAIDs inhibit both COX-1 and COX-2, whereas COX-2 selective NSAIDs inhibit only COX-2.²,⁶ Examples of nonselective NSAIDs include ibuprofen, naproxen, indomethacin, and diclofenac; COX-2 selective NSAIDs often include the suffix “-coxib” such as celecoxib and rofecoxib.

It is generally agreed upon by many organizations that NSAIDs are an appropriate treatment for knee osteoarthritis as they provide both analgesic effects and decrease inflammation.²,³,⁴,⁷ However, when treating elderly patients, providers should consider initiating a topical NSAID due to the higher potential for adverse events in this age group.⁴,⁸

NSAIDs have the potential to cause side effects related to the gastrointestinal (GI), renal, and cardiovascular (CV) systems. In pertaining to the gastrointestinal system, patients taking NSAIDs can have an increased risk of gastritis, abdominal pain, ulcers, GI bleeding, and liver toxicity.²,⁹,¹⁰ Patients may complain of dyspepsia, burning sensation, gastric reflux or they can be asymptomatic and present with hematemesis or hematochezia.⁸,⁹,¹¹,¹²,¹³ The gastric lining contains COX-1, and therefore nonselective NSAID therapy can have higher potential for adverse GI events in comparison to COX-2 selective NSAIDs.¹⁰ In addition to age, risk factors for potential GI-related adverse events include use of anticoagulation, prior gastric ulcer, concurrent use of corticosteroids and/or
aspirin.¹⁰,¹³ Concomitant use of a proton pump inhibitor should be considered in patients with comorbidities to reduce the risk of an adverse event.³,⁴,¹⁰ COX-2 selective NSAIDs when first brought to market showed great promise in reducing GI complications while providing pain relief.⁸,¹⁰,¹⁴,¹⁵,¹⁶,¹⁷ High rates of cardiovascular events (myocardial infarction, cerebrovascular events, and death) were noted and subsequently two of the COX-2 selective medications were withdrawn from the market.⁸,¹⁰,¹⁸,¹⁹ The FDA has extended the potential for CV events to include all NSAIDs including both selective and nonselective NSAIDs.¹⁰,²⁰,²¹ The risk and benefits of NSAID use should be weighed when prescribing these medications especially in patients who have or suspected to have cardiovascular disease.⁸,¹⁰,²¹ In those patients with comorbidities alternative therapeutic modalities should be considered.³,¹⁰ Furthermore, NSAID therapy should be avoided in patients with known renal disease or who are at risk of renal toxicity.⁴,⁶ When administered in appropriate doses and durations, NSAID utilization remains a first-line therapy for management of knee osteoarthritis.

There are numerous oral medications that have been marketed and are indicated for the treatment of pain associated with knee osteoarthritis including duloxetine, tramadol, and other oral opioids. Duloxetine, a serotonin-norepinephrine reuptake inhibitor (SNRI), has demonstrated clinical benefit in treatment of chronic pain, including chronic knee osteoarthritis.³,²² SNRI, as a class of medications, has the potential for numerous drug interactions and an extensive list of common and serious adverse reactions. Common side effects can include fatigue, nausea, constipation, dizziness, and dry mouth, while more serious include seizures, hypertensive crisis, and suicidality.³,²³,²⁴,²⁵ Duloxetine can be considered for therapy in patients who fail to respond to other oral medications (such as acetaminophen and NSAIDs) or in those patients who have contraindications to use of these medications.

Tramadol, a nontraditional opioid, has therapeutic indications for symptom management in knee OA. Studies have reported similar therapeutic responses as oral NSAIDs, including reduction of joint stiffness and pain.²,⁴,⁷,²⁶ Although tramadol and other opioids may be considered in patients who failed or have contraindication to use of other first-line oral agents, many organizations do not make recommendations for or against the use of opioid agents for analgesia.³,⁴,⁷ This is based on limited data to recommend for or against opioid use. Additionally these medications have a potential for side effects, abuse, and addiction.²⁷ The modern opioid epidemic in the United States has led to a push to discourage the use of these medications to treat osteoarthritis. Furthermore, preoperative utilization of opioids can be associated with poor outcomes and a greater chance for increased and sustained use of these drugs in the long-term.

Corticosteroid injections (CSIs) have been widely used as a mainstay treatment of knee osteoarthritis for several decades.⁴⁰ Hollander et al. documented the anti-inflammatory effect CSIs had on inflammatory/rheumatoid arthritis in 1951.⁴¹ Despite the long-standing use of CSIs, the exact mechanism in which these intra-articular injections alleviate symptoms is not fully known. However, it is suggested that use of CSIs results in a reduction of inflammatory processes at the site of administration by interacting with cytokines typically recruited for the immune resp onse.⁴⁰,⁴²,⁴³

Several formulations of injectable corticosteroids exist. These are separated into two categories, water-soluble (nonparticulate) and water insoluble (particulate). Soluble corticosteroids are nonester preparations and these do not form microcrystals in water. These have a quicker onset but shorter duration of action. Sodium phosphate formulations are included in this category; examples include betamethasone sodium phosphate and dexamethasone. Insoluble corticosteroids are ester preparations and these form microcrystals in water. These have a slower onset but longer duration of action because these preparations remain in the synovial fluid longer. Acetate formulations are included in this category; examples include betamethasone acetate, methylprednisolone, and triamcinolone. Insoluble injections tend to have a lower potency and therefore require a higher dose to reach similar response compared to soluble injections.⁴⁰,⁴³,⁴⁴ According to the American College of Rheumatology, the preferred injectable corticosteroids are methylprednisolone acetate, triamcinolone hexacetonide, and triamcinolone acetonide.⁴²,⁴⁵ Refer to Table 22-1 for a list of common corticosteroid preparations.

Triamcinolone acetonide extended release (ER) is a new corticosteroid formulation. The steroid is packaged within a biodegradable bead, which allows the steroid to remain in the joint for an extended period of time in comparison to immediate release injectable corticosteroids.⁴⁶,⁴⁷ In 2019, Spitzer et al. conducted a study that demonstrated treatment with triamcinolone acetonide ER provided symptom relief without damage to cartilage with repeated injections.⁴⁶ While it has only been on the market for therapeutic use since 2017, the extended release injection shows promising results in its relatively early use in treatment of osteoarthritis.⁴⁸

The duration of benefit from intra-articular CSIs is another topic of debate due to discrepancies in several studies in the literature. One meta-analysis reported significant efficacy in intra-articular CSIs at 3 to 4 weeks postinjection but not at 6 to 8 weeks postinjection compared to control.⁴⁹ In a different meta-analysis, patients were noted to have benefit from intervention lasting anywhere from 1 to 24 weeks in duration, although higher doses of steroid may be needed in order to have benefit from weeks 16 to 24.⁴²,⁵⁰

Compared to the potential complications that are associated with surgical intervention, CSIs have been and continue to remain a relatively safe conservative therapeutic intervention for patients with knee osteoarthritis. That being said, there have been adverse side effects attributed to use of corticosteroids. The most common side effect is a postinjection flare, which has been documented to occur in anywhere from 2% to 25% of injection administrations.⁴⁰,⁴³,⁵² This results in an acute pain crisis due to increased inflammation, typically occurring within hours of injection and can last up to 48 to 72 hours in duration.⁵³ It has been suggested that an aspiration to rule out infection should be considered if an intra-articular infection is suspected or if symptoms of postinjection flare persist longer than 24 hours,⁴⁰,⁴³ Along with postinjection flare, postinjection pain is a commonly noted side effect.

The most concerning adverse event of intra-articular CSIs is septic arthritis due to the potential of high morbidity and mortality. However, studies have shown that the actual occurrence of postinjection septic arthritis is as low as 0.01% to 0.03%.⁴⁰,⁴³,⁵⁴ Additional complications that may occur include local tissue atrophy, skin hypopigmentation, and fat atrophy.⁴²,⁵⁵ This can be due to improper placement of an intra-articular injection (not actually placed in the joint) or if the solution exits the joint space out along the pathway from the needle.⁴⁰,⁴³ A concern that is frequently discussed is the thought that intra-articular CSIs could potentially lead to acceleration of cartilage loss and damage compared to placebo. Numerous studies have been performed with conflicting data. Several long-term studies and randomized control trials have been published indicating there were no significant negative effects to the knee anatomical structures or articular cartilage observed with long-term administration of intra-articular CSIs.⁴²,⁴⁶,⁵¹ Furthermore, intervention with intra-articular CSIs had been noted to provide significant improvement in pain, nighttime symptoms, and joint stiffness.⁵¹ In 2017, McAlindon et al. conducted a study comparing intra-articular CSIs in the knee joint to saline injections every 12 weeks for 2 years and reported greater cartilage volume loss in the intervention group compared to the control group.⁵⁶ It should be noted that the protocol of every 12-week injections is not typically done in practice and draws question to clinical relevance of the study conclusions.