Knee Pain Drugs: Medications That Cause or Treat It

NSAIDs: The Most Prescribed Knee Pain Drug Class

Nonsteroidal anti-inflammatory drugs remain the pharmacologic backbone for knee osteoarthritis (OA) and acute knee injuries. The 2019 American College of Rheumatology (ACR) guidelines conditionally recommend oral NSAIDs for knee OA when non-pharmacologic measures alone do not control symptoms [1].

Ibuprofen (400-800 mg three times daily), naproxen (250-500 mg twice daily), and celecoxib (200 mg daily) all demonstrate moderate-to-large effect sizes for OA knee pain. A Cochrane review of 72 trials (N=22,802) found that NSAIDs reduced pain intensity by a mean of 12.4 mm on a 100-mm visual analogue scale compared with placebo [2]. Celecoxib, a selective COX-2 inhibitor, showed equivalent analgesic efficacy to nonselective NSAIDs in the CONDOR trial (N=4,484) while producing fewer upper gastrointestinal events (0.9% vs. 3.8%, P<0.001) [3].

Topical NSAIDs deserve separate mention. The ACR strongly recommends topical diclofenac for knee OA specifically because the knee is a superficial joint with good transdermal drug penetration [1]. A 12-week randomized trial of diclofenac sodium 1% gel (N=492) found that the topical formulation reduced WOMAC pain scores by 44% from baseline vs. 28% with vehicle gel [4]. Systemic absorption is roughly 5-10% of the oral dose, which makes this option particularly attractive for patients over 75 or those with renal concerns.

GI bleeding risk rises with oral NSAID duration. For patients who need long-term therapy, co-prescribing a proton pump inhibitor or switching to a COX-2 selective agent is standard practice per ACR recommendations [1].

Acetaminophen: Limited but Still Used

Acetaminophen was once a universal first-line choice. That changed. The ACR's 2019 guidelines conditionally recommend against acetaminophen monotherapy for knee OA, citing small effect sizes that often fail to reach clinical significance [1].

A patient-level meta-analysis published in The Lancet (N=58,556 across 74 trials) confirmed that acetaminophen at standard doses (up to 4 g/day) produced clinically insignificant pain reduction in knee and hip OA, with an effect size of 0.18 (95% CI 0.11-0.25) [5]. That is roughly one-third the effect of oral NSAIDs.

Still, acetaminophen retains a role. Patients who cannot tolerate NSAIDs, those on anticoagulants where bleeding risk is elevated, and individuals with stage 4 or 5 chronic kidney disease may have no better oral option. The ceiling dose is 3 g/day in older adults (down from 4 g/day) to protect against hepatotoxicity [6].

Intra-Articular Corticosteroid Injections

A corticosteroid shot into the knee joint delivers high local drug concentrations with minimal systemic exposure. The ACR conditionally recommends intra-articular corticosteroids for knee OA [1].

Triamcinolone acetonide (40 mg) and methylprednisolone acetate (40-80 mg) are the two most commonly injected formulations. Onset of pain relief typically occurs within 24 to 48 hours. Duration varies widely. A randomized trial in JAMA (N=140) that administered triamcinolone 40 mg every 3 months for 2 years found significant pain improvement at each injection cycle but also detected a small amount of additional cartilage volume loss (0.21 mm greater than saline over 2 years) on MRI [7].

That cartilage finding raised concern. The clinical significance remains debated because pain scores improved and functional outcomes were equivalent between groups. Most rheumatologists now limit injections to 3 or 4 per joint per year, spacing them at least 3 months apart.

"Intra-articular corticosteroids are not disease-modifying, but they are an effective bridge when patients need rapid relief while titrating other therapies," according to the ACR's 2019 guideline document for the management of osteoarthritis of the hand, hip, and knee [1].

Duloxetine: The Overlooked Oral Option

Duloxetine (Cymbalta) holds FDA approval for chronic musculoskeletal pain, including knee OA. It is not an analgesic in the traditional sense. As a serotonin-norepinephrine reuptake inhibitor (SNRI), it modulates descending pain inhibition pathways in the central nervous system.

A 13-week randomized trial (N=256) published in the International Journal of Clinical Practice showed duloxetine 60 mg/day reduced BPI average pain scores by 1.8 points versus 0.9 points with placebo (P<0.001) in patients with knee OA who had inadequate response to NSAIDs [8]. The ACR conditionally recommends duloxetine for knee OA, placing it alongside NSAIDs and intra-articular corticosteroids as a pharmacologic option [1].

Side effects include nausea (reported in 23% of patients during the first 2 weeks), dizziness, fatigue, and constipation. Most adverse effects diminish by week 4. Duloxetine requires gradual taper to avoid discontinuation syndrome.

This drug fills an important gap for patients who cannot take NSAIDs and want an oral alternative beyond acetaminophen. It also works well in patients with concurrent depression or generalized anxiety, addressing two problems simultaneously.

Hyaluronic Acid Injections (Viscosupplementation)

Viscosupplementation involves injecting hyaluronic acid (HA) directly into the knee joint to supplement the naturally occurring synovial fluid. Products include Synvisc-One (single injection), Euflexxa (3-injection series), and Gel-One (single injection).

The evidence is mixed, which explains why guidelines diverge. The ACR conditionally recommends HA injections for knee OA [1], while the American Academy of Orthopaedic Surgeons (AAOS) gave a "limited" recommendation against them in their 2021 clinical practice guideline [9]. A network meta-analysis in the Annals of Internal Medicine (N=21,163 across 169 trials) found that HA injections produced moderate pain improvement versus intra-articular placebo (standardized mean difference -0.34 to 95% CI -0.43 to -0.26), with the effect size being somewhat smaller than corticosteroids at 4 weeks but larger at 26 weeks [10].

Peak benefit typically occurs 4 to 8 weeks after injection and may persist for up to 6 months. Patients with Kellgren-Lawrence grade 2-3 OA (moderate disease) appear to respond best. Those with bone-on-bone grade 4 disease are less likely to benefit.

Biologic DMARDs: When Inflammatory Arthritis Causes the Knee Pain

Not all knee pain is osteoarthritis. Rheumatoid arthritis (RA), psoriatic arthritis (PsA), and ankylosing spondylitis all produce inflammatory knee pain that responds to biologic disease-modifying antirheumatic drugs (DMARDs).

Adalimumab (Humira), etanercept (Enbrel), and other TNF inhibitors remain first-line biologics for RA. The PREMIER trial (N=799) demonstrated that adalimumab plus methotrexate produced ACR70 responses (70% improvement in tender and swollen joints, including knees) in 49% of patients versus 27% with methotrexate alone at 2 years [11]. For psoriatic arthritis, secukinumab (Cosentyx), an IL-17A inhibitor, achieved ACR50 responses in 54% of patients in FUTURE 5 (N=996) at 52 weeks [12].

These drugs are not prescribed for isolated knee OA. The distinction matters clinically. If a patient presents with knee swelling, morning stiffness lasting more than 30 minutes, elevated CRP or ESR, and a positive rheumatoid factor or anti-CCP antibody, the treatment pathway shifts entirely from OA management to DMARD initiation.

Drugs That Cause Knee Pain: Statins

Statins (atorvastatin, rosuvastatin, simvastatin, and others) are among the most commonly prescribed medications worldwide. Musculoskeletal complaints, including knee pain, represent their most frequent side effect.

A meta-analysis of randomized controlled trials published in the European Journal of Preventive Cardiology (N=83,880 across 26 statin RCTs) found that statin therapy increased the risk of musculoskeletal pain and arthralgia with an odds ratio of 1.05 (95% CI 1.01-1.09) compared to placebo [13]. In clinical practice, the rate is perceived as higher. Observational studies suggest 5-10% of statin users report myalgia, arthralgia, or joint stiffness. The "nocebo effect" partially explains this discrepancy.

The mechanism is not fully established. Proposed pathways include reduced coenzyme Q10 synthesis, mitochondrial dysfunction in muscle cells, and alterations in cholesterol metabolism within cell membranes. Lipophilic statins (simvastatin, atorvastatin) may produce more musculoskeletal symptoms than hydrophilic agents (rosuvastatin, pravastatin) [14].

Management involves a structured statin rechallenge. The ACC/AHA guideline recommends switching to an alternative statin at a lower dose, trying rosuvastatin or pravastatin, or using intermittent dosing (e.g., rosuvastatin 5-10 mg every other day) before concluding that a patient is truly statin-intolerant [14].

Drugs That Cause Knee Pain: Fluoroquinolone Antibiotics

Ciprofloxacin, levofloxacin, and moxifloxacin carry an FDA black box warning for tendinitis and tendon rupture, added in 2008 and strengthened in 2016 [15]. The Achilles tendon is the most commonly affected site, but knee tendons (patellar and quadriceps) are also at risk.

A large population-based study using FDA Adverse Event Reporting System data found that fluoroquinolones were associated with a 3.2-fold increased risk of tendon rupture compared to other antibiotic classes [16]. Risk factors that amplify this effect include age over 60, concurrent corticosteroid use, renal impairment, and prior organ transplantation.

The FDA's 2016 safety communication advised that fluoroquinolones should be reserved for conditions without alternative treatment options, specifically excluding uncomplicated urinary tract infections, acute sinusitis, and acute bronchitis from their routine use [15]. Patients who develop new knee pain, swelling, or a sensation of snapping while on a fluoroquinolone should stop the drug immediately and contact their prescriber.

"Fluoroquinolones should be reserved for use in patients who have no alternative treatment options for acute bacterial sinusitis, acute bacterial exacerbation of chronic bronchitis, and uncomplicated urinary tract infections," states the FDA Drug Safety Communication of July 2016 [15].

Drugs That Cause Knee Pain: Aromatase Inhibitors

Aromatase inhibitors (AIs), including letrozole (Femara), anastrozole (Arimidex), and exemestane (Aromasin), are standard adjuvant therapy for postmenopausal women with hormone receptor-positive breast cancer. Arthralgia is their most common reason for discontinuation.

The incidence is striking. The ATAC trial (N=9,366) reported arthralgia in 35.6% of patients receiving anastrozole versus 29.4% on tamoxifen [17]. More recent observational data suggest that up to 50% of patients on AIs experience some degree of joint pain, with the knees and hands being the most frequently affected sites.

The mechanism involves estrogen deprivation. Estrogen receptors exist in synovial tissue, cartilage, and bone. When aromatase inhibition drops circulating estradiol to near-undetectable levels, these tissues lose a protective anti-inflammatory signal. The resulting synovial thickening and increased intra-articular fluid have been documented on MRI and ultrasound studies.

Management strategies include switching between AIs (letrozole to exemestane, for example, since the steroidal and nonsteroidal agents have different binding profiles), adding exercise therapy (which has Level 1 evidence for reducing AI-associated arthralgia), or short courses of NSAIDs. For severe cases, switching to tamoxifen with its different side-effect profile may be preferable to stopping endocrine therapy altogether [17].

Drugs That Cause Knee Pain: Immune Checkpoint Inhibitors

Immune checkpoint inhibitors (pembrolizumab, nivolumab, ipilimumab) can trigger immune-related adverse events (irAEs) affecting joints. Inflammatory arthritis from checkpoint inhibitors mimics rheumatoid arthritis and frequently involves the knees.

A systematic review published in the Annals of the Rheumatic Diseases found that inflammatory arthritis occurred in approximately 5-7% of patients receiving checkpoint inhibitor therapy, with knee involvement reported in over 60% of those cases [18]. The onset is variable, ranging from 2 weeks to over 12 months after treatment initiation.

Mild cases (CTCAE grade 1) are managed with NSAIDs. Moderate cases (grade 2) typically require low-dose prednisone (10-20 mg/day). Severe cases (grade 3-4) may necessitate holding the checkpoint inhibitor and initiating DMARDs, though oncologists weigh this decision carefully against the cancer treatment benefit.

Topical and Emerging Therapies

Beyond oral medications and injections, several topical and newer approaches target knee pain with varying levels of evidence.

Capsaicin cream (0.025-0.075%) applied four times daily depletes substance P from peripheral nerve terminals. A Cochrane review found modest benefit over placebo for OA pain with a number needed to treat (NNT) of 8 [19]. Compliance is the main barrier because the burning sensation during the first 1-2 weeks drives many patients to quit.

Platelet-rich plasma (PRP) injections have generated significant interest. A meta-analysis in the American Journal of Sports Medicine (N=1,423 across 14 RCTs) reported that leukocyte-poor PRP produced greater pain reduction than hyaluronic acid at 12 months for knee OA (weighted mean difference -15.5 mm on 100-mm VAS, P<0.001) [20]. Neither the ACR nor AAOS currently provides a strong recommendation for or against PRP, citing heterogeneity in preparation methods and study quality.

Tanezumab, a nerve growth factor (NGF) inhibitor, completed Phase 3 trials for knee OA. The FDA issued a complete response letter in 2021 due to a signal of rapidly progressive OA requiring joint replacement. This drug class remains investigational.