Principles of Injection Therapy

Musculoskeletal injections are common procedures across various medical specialties. The evidence for musculoskeletal injections is varied and is complicated by differing injection techniques, injectates, and landmark versus image guidance. When performed for the proper indication and with correct technique, musculoskeletal injections can be beneficial and rewarding for both patients and physicians. Before proceeding with an injection, it is critical to obtain a proper diagnosis and rule out relevant contraindications. Absolute contraindications include injectate hypersensitivity, infection, uncontrolled bleeding disorder, and fracture. Relative contraindications include corrected bleeding disorder, anticoagulant use, hemarthrosis, diabetes, immunosuppression, prosthetic joint, high risk of tendon rupture, and psychogenic pain. ^,

This chapter addresses the most common nonbiologic injectates used in clinical practice including corticosteroids, local anesthetics, ketorolac, and hyaluronic acid (HA), as well as the evidence for and against their use.

Corticosteroids

Corticosteroid injections (CSIs) first gained popularity in the 1950s and were used to treat patients with joint pain associated with rheumatoid arthritis. Benefits included delivery of a lower dose of steroid compared with oral options, resulting in a lower systemic risk profile. Risks associated with CSIs include infection, flushing, a transient increase in pain, skin hypopigmentation in superficial injections, allergic reaction, and transient hyperglycemia, among other less common complications. ^,

Corticosteroids are small hydrocarbon molecules. At the cellular level, once the amphipathic corticosteroid arrives in the synovial fluid it crosses the lipid bilayer of the cellular membrane and binds to the glucocorticoid receptor before being translocated into the nucleus and binding to the glucocorticoid response element (a short sequence of DNA that binds transcription factors and regulates gene transcription). Corticosteroids affect cellular transcription in two ways: (1) inflammation is suppressed through the inhibition of tumor necrosis factor α and other proinflammatory mediators; and (2) antiinflammatory gene transcription is enhanced while inflammatory gene transcription is suppressed. ^,

When administered orally, these hydrophobic compounds are small enough to be transported via diffusion across the capillary wall; however, diffusion leads to low overall bioavailability in the synovial fluid. Thus the need to provide high doses to achieve an adequate therapeutic effect, and while the effects of oral steroids are theoretically beneficial to the targeted structure they are nonspecific and can have systemic side effects. Short-term systemic corticosteroid use is generally associated with mild side effects, including cutaneous effects, electrolyte abnormalities, hypertension, hyperglycemia, pancreatitis, hematologic, immunologic, and neuropsychologic effects. Long-term systemic corticosteroid use may be associated with more serious sequalae, including osteoporosis, avascular necrosis, adrenal insufficiency, gastrointestinal, hepatic, and ophthalmologic effects, hyperlipidemia, growth suppression, and possible congenital malformations. ^, Thus practitioners sought a more directed means of delivering corticosteroids into the joint space.

Intra-articular Injections

Intra-articular injection of corticosteroids not only improves the concentration within the joint and minimizes systemic effects but also allows for delivery of polymers or salts molecules complexed with the corticosteroid in aqueous solution to improve drug stability and extend the therapeutic window. Despite this, and due to their very low molecular weight (<700 Da), corticosteroids routinely exhibit a half-life of 12 hours or less. ^,

Corticosteroids have a long track record of use in intra-articular inflammatory processes. Matzkin and colleagues showed improvements in pain and function at all time points (3 weeks, 6 weeks, 3 months, and 6 months) after one CSI, with obesity and severity of disease associated with less improvement and duration of response. Several organizations have deemed CSIs safe and temporarily efficacious, including the Osteoarthritis Research Society International, the US National Institute of Health, the American College of Rheumatology, the Arthroscopy Association of Canada, and the American Academy of Orthopaedic Surgeons. An initial Cochrane review of the efficacy of intra-articular CSIs concluded temporary benefit. However, an updated review of the literature suggested the heterogeneity in trial design and variability of studies limit the ability to assess the benefit of these injections and that there is a low likelihood of meaningful benefit 6 months post injection. ^, Despite the poor data quality, the Cochrane review did find that injected corticosteroids improved pain scores by 1 cm on a standard 10-cm visual analog scale compared with control, with a number needed to treat of 8.

More recently, the potential negative long-term effects of intra-articular CSIs have been highlighted. Cartilage exposure to corticosteroids has been associated with such deleterious outcomes as decreased extracellular matrix production, decreased type 1 collagen synthesis, and induction of chondrocyte apoptosis. ^, In a study by McAlindon and colleagues, repeated exposure to CSIs over 2 years for symptomatic knee osteoarthritis was associated with increased cartilage loss compared with control subjects injected with saline.

Klocke and colleagues subsequently showed a decrease in the concentration of serum biomarkers for cartilage and bone metabolism at 3 weeks post injection, and Lu and colleagues demonstrated protective effects of steroids on glycosaminoglycan loss after a mechanical injury to the cartilage and proinflammatory cytokine exposure. ^, A systematic review from Wernecke and colleagues in 2015 determined that corticosteroids have a time- and dose-dependent effect on articular cartilage, with beneficial effects occurring at low doses and durations and detrimental effects at high doses and durations. Therefore it seems prudent to use the lowest efficacious dose and to limit repeat injections based on current data. More research into the longer-term effects of these changes and clinical implications is warranted.

There is no consensus regarding the volume of injectate or choice of injected corticosteroid. A comparison study showed similar efficacy between injected triamcinolone hexacetonide (40 mg) and methylprednisolone acetate (40 mg). There are data to suggest improved benefit from triamcinolone hexacetonide over triamcinolone acetonide, and the authors prefer the combined effectiveness with mitigation of risk in using 40 mg rather than 80 mg. In addition, the authors prefer the use of ultrasonographic guidance because it is associated with improved accuracy of intra-articular injections. Wyles and colleagues also demonstrated that corticosteroids are cytotoxic to human mesenchymal stem cells in a dose-dependent fashion, with betamethasone being most toxic and dexamethasone being the least toxic. Again, current data suggest that use of the lowest efficacious dose of corticosteroid should be favored.

Management of pain associated with intra-articular inflammatory processes requires an individualized approach, taking into account the patient’s personal goals, medical history, tolerance of risk, and extent of disability. Any intra-articular injection for such an inflammatory process, corticosteroid or otherwise, should only be pursued as part of a multimodal approach that includes mitigating other risk factors for recurrent pain and dysfunction.

Axial Spine Injections

Fluoroscopy-guided CSIs have been used for radicular pain, both diagnostically and therapeutically. ^, There are several options of corticosteroids for epidural steroid injections, and corticosteroids are divided between particulate (triamcinolone, methylprednisolone, betamethasone) and nonparticulate (dexamethasone) preparations. There is limited evidence to support the superiority of particulate or nonparticulate steroids for epidural CSIs, but there is a concern that particulate formulations pose an increased risk of embolism after inadvertent intravascular injection. The US Food and Drug Administration (FDA) has produced a risk assessment of serious neurologic events after epidural glucocorticoid injections and stated that “although many experts believe the risk is greatest with suspensions [or particulate steroids], the available data do not support comparative safety labeling implying that solutions are safer.” Given the lack of strong data favoring the efficacy of particulate steroids and the risk of catastrophic adverse events, the authors prefer nonparticulate corticosteroids for epidural CSIs. ^,

Irrespective of the choice of injectate, the cost-effectiveness and clinical benefit of epidural CSIs are debated. ^, Many reviews and meta-analyses have been written on the topic of epidural CSIs. A 2015 meta-analysis in Annals of Internal Medicine from Chou and colleagues determined that epidural CSIs provided immediate improvements in pain and function. However, the improvements were small and not sustained. Epidural CSIs also did not affect long-term surgery risk. Furthermore, in a retrospective follow-up performed by Jang and colleagues, a small minority of those who underwent a remote lumbar transforaminal epidural CSI achieved complete relief with one injection, while nearly half had relied on repeat epidural CSIs and oral pain medications, and approximately 25% of patients had progressed to surgery. Of note, outcomes from surgery in the cohort were relatively poor, leading the authors to recommend consideration of epidural CSIs as a viable presurgical intervention despite the lack of definitive relief in the majority of cases. In addition to the lack of expected permanent relief, repeat epidural CSIs expose the patient to the risk of significant adverse events including epidural hematoma, infection, and strokes, among others, and repeat epidural CSIs have been associated with adrenal suppression due to systemic uptake. ^,

Soft Tissue Corticosteroid Injections

CSIs have been used for decades to treat the pain and discomfort associated with soft tissue pathology, including injections into the bursae and tendon sheaths. In the upper extremity, CSIs directed at the common extensor tendon for the treatment of lateral epicondylalgia have been shown to lead to short-term pain relief but have been associated with worse outcomes at 1 year compared with placebo. Other potential targets for ultrasound-guided CSIs have better data to support their use and include the first dorsal compartment of the wrist, trigger finger, median nerve in the carpal tunnel in the distal upper extremity, subacromial bursitis, and shoulder impingement when used in conjunction with an active home exercise program focusing on scapular stabilization.

In the lower extremity, CSIs for patellar and insertional Achilles tendinopathy have not reliably shown benefit. ^, There are more data for the treatment of iliopsoas bursitis at the hip, trochanteric bursitis in the lateral hip, and pes anserine bursitis. The use of CSIs in the setting of plantar fasciitis has shown mixed results in comparison with autologous blood-derived products and inferiority to dry needling.

Despite widespread use, the rationale for use of CSIs is contentious. Chronic systemic steroid use in the etiology of tendon rupture is well established. ^, These negative effects have been shown to extend to injected corticosteroids as well. In animal models, intratendinous injections have been shown to cause tendon necrosis at the site of injection and a delayed healing response compared with a saline injection. Collagen degeneration and an increase in matrix metalloproteinase-3 (MMP3) have been identified in histologic analysis of tendons exposed to steroid injections. In contrast, other studies have shown a decrease in inflammatory mediators such as MMP2 and substance P. ^,

Potential complications of CSIs include infection, subcutaneous atrophy, skin depigmentation, and tendon rupture . The use of ultrasound guidance for these procedures has resulted in improved accuracy compared with palpation- or landmark-guided injections. The authors encourage image guidance given the increased risk of tendon damage and risk of rupture with intratendinous CSIs. Further studies are certainly warranted to continue to evaluate the effects of corticosteroids on soft tissues and the efficacy of CSIs.

Local Anesthetics

Amide local anesthetics (lidocaine, bupivacaine, ropivacaine, etc.) are commonly used in musculoskeletal injections to relieve pain. ^, All local anesthetics are membrane-stabilizing drugs that decrease the rate of depolarization and repolarization of the cellular membranes of neurons by inhibiting sodium influx through ion channels in the cell membrane, thus inhibiting action potentials and blocking signal conduction. All nerve fibers are sensitive to local anesthetics; however, type C pain fibers are some of the most sensitive.

Beyond membrane-stabilizing effects, local anesthetics are known to have antiinflammatory actions due to their structural similarity to steroid agents. Lidocaine is the most well-studied local anesthetic and is thought to have antiinflammatory effects by inhibiting sympathetic postganglionic neuron–mediated inflammation. Specifically, lidocaine inhibits the release of inflammatory mediators such as leukotriene B ₄ , interleukin (IL)-1α, and histamine and has direct inhibitory effects on proinflammatory SPGN-mediated plasma extravasations through inhibiting synovial EP1 receptors. Thus mechanisms of local anesthetics, such as lidocaine, include stabilizing nerve cell membranes, inhibiting nociceptive signaling, and decreasing the inflammatory response. Treatment with local anesthetics may also provide longer-term benefit beyond the immediate pain relief. Eker and colleagues showed that treatment of knee osteoarthritis with intra-articular injection of 0.5% lidocaine provided improvements in pain and function that continued at the 3-month follow-up. They postulated that this may be due to decreasing neural cell metabolism and stabilizing the neural membrane.

However, it is important to note that local anesthetics are known to have chondrotoxic effects on human cartilage. Jayaram and colleagues published an elegant review on this topic, noting variable chondrotoxic potential of different local anesthetics and that a single injection can result in chondrocyte death. Local anesthetics reviewed were lidocaine, bupivacaine, ropivacaine, levobupivacaine, and mepivacaine. The chondrotoxic effects are dose and duration dependent and are influenced by the type of local anesthetic used. Chondrotoxicity also seems to be exacerbated by co-administration with corticosteroid. Of the local anesthetics studied, ropivacaine was the least chondrotoxic and bupivacaine the most chondrotoxic. The mechanisms leading to chondrotoxicity appear to be increased apoptosis, caspase inhibition, necrosis, mitochondrial dysfunction, extracellular matrix damage, and decreased DNA-normalized glycosaminoglycan expression. Local anesthetics have also been demonstrated to be toxic to human mesenchymal stem cells, which are crucial for healing. Here too, ropivacaine is the least cytotoxic. ^, Furthermore, Zink and colleagues have shown local anesthetics to be toxic to muscle tissue, with bupivacaine causing significantly more tissue damage than ropivicaine. ^,

In summary, amide local anesthetics have general dose- and duration-dependent cytotoxic effects, with ropivacaine having the least toxicity, bupivacaine the most toxicity to chondrocytes and muscle tissue, and lidocaine the most toxicity to mesenchymal stem cells. ^,

Ketorolac

Ketorolac is a nonsteroidal antiinflammatory drug (NSAID) that inhibits the inflammatory cyclooxygenase and lipoxygenase enzyme systems, and the synthesis of prostaglandins and leukotrienes. Ketorolac provides a reasonable alternative to corticosteroids and HA because it has fewer side effects than corticosteroid, has a faster onset of action than HA, and is more economical than both HA and corticosteroid.

In patients with knee and first carpometacarpal joint osteoarthritis, the addition of ketorolac to HA injection provided more rapid onset of pain relief without any added complications when compared with HA alone. ^, In addition, when compared head-to-head with corticosteroid in patients with knee and hip osteoarthritis, ketorolac provided similar improvements in pain and function for up to 6 months. ^, Ketorolac has also been shown to be safe and more effective than corticosteroids in the treatment of shoulder impingement and adhesive capsulitis. ^,

In a 2018 review, Sardana and colleagues concluded that NSAID injections had a less severe side-effect profile compared with corticosteroids and provided equivalent, if not better, pain relief for musculoskeletal pathology. Ketorolac has demonstrated no detrimental effects on chondrocytes and tenocytes in animal models. ^, Beitzel and colleagues reported no toxic effects on human chondrocytes and tenocytes; however, other studies have indicated a dose-dependent cytotoxic effect of ketorolac on human articular chondrocytes. ^, In Abrams and colleagues, the methods more closely reflect the exposure duration of a single-dose injection, but the authors note their methods potentially overestimated the chondrotoxicity of ketorolac. Nonetheless, there is no consensus on the cytotoxicity of ketorolac and the risk of ketorolac intra-articular injections. There is a theoretical association between ketorolac injections and bleeding risk; however, there have been no studies documenting a bleeding complication after a ketorolac injection in contact or collision sport athletes.

Hyaluronic Acid

HA is a polysaccharide consisting of repeating units of the disaccharide N -acetylglucosamine and sodium glucuronate. This glycosaminoglycan structure is found within the extracellular matrix of a variety of soft tissue structures and is synthesized by type B synoviocytes, fibroblasts, and chondrocytes. ^, Namiki and colleagues were one of the first groups to evaluate its use of HA for knee osteoarthritis more than 30 years ago. The injection was later approved by the FDA in 1997 for the treatment of knee osteoarthritis.

The term ‘viscosupplementation’ was coined for this injection based on the initial work of Namiki and colleagues, where the proposed mechanism of action was thought to be improving the viscosity of the arthritic joint. However, since that time there have been several studies that have uncovered several possible mechanisms, including the decrease of inflammatory mediators, improvement in the viscosity of the microenvironment, immunomodulation of catabolic enzymes, and downregulation of nociceptors. ^,

Exogenous HA has been shown to decrease inflammation through the suppression of proinflammatory mediators; specifically, IL-β and tumor necrosis factor α. ^, HA has displayed the ability to immunomodulate catabolic processes by downregulating matrix MMP3 expression and inhibiting MMP synthesis, potentially reducing cartilage destruction. ^, Exogenous HA also has the capability of improving viscosity by stimulating the production of endogenous HA in the synovial fluid and restoring homeostasis. ^, Lastly, exogenous HA is able to decrease pain by binding neuropeptides, creating a boundary layer around nociceptors and not allowing the receptors to transmit pain signals.

There are several exogenous HA products currently on the market that differ in their molecular properties, single- or multi-injection therapies, and whether they are derived from animal products or using a genetically modified bacterial source ( Table 4.1 ). These differences have been shown to affect clinical outcomes, and thus all exogenous HA products should not be evaluated as a group. Molecularly, the most significant difference between products is their molecular weight, with products either being low molecular weight (<1500 kilodaltons [kDa]), moderate molecular weight (>1500 kDa and <3000 kDa), or high molecular weight (>3000 kDa). A meta-analysis performed by Altman and colleagues has suggested that high-molecular-weight HA is more efficacious than moderate- or low-molecular-weight products.

TABLE 4.1

Hyaluronic Acid Products Currently on the Market.

Product Name	High, Moderate, or Low Molecular Weight (kDa)	Source	Dosing (mg/mL)	Number of Injections
Synvisc-one	High	Avian derived	48 mg/6 mL	Single
Gel-one	Unknown	Avian derived	30 mg/3 mL	Single
Monovisc	Moderate	Bacterial derived	88 mg/4 mL	Single
Durolane	High	Bacterial derived	60 mg/3 mL	Single
Hymovis	Unknown	Bacterial derived	24 mg/3 mL	Two
Synvisc	High	Avian derived	48 mg/2 mL	Three
Orthovisc	Moderate	Bacterial derived	30 mg/2 mL	Three
Euflexxa	High	Bacterial derived	20 mg/2 mL	Three
Gelsyn-3	Low	Bacterial derived	16.8 mg/2 mL	Three
Hyalgan	Low	Avian derived	20 mg/2 mL	Three or five
Supartz FX	Low	Avian derived	25 mg/2.5 mL	Three or five