CORR Insights®: Intraocular Pressure Increases After Intraarticular Knee Injection With Triamcinolone but Not Hyaluronic Acid

Where Are We Now?

In the current study, Taliaferro and colleagues [11] show that all interventions, even seemingly low-risk corticosteroid injections for knee arthritis, can have unanticipated side effects. The well-known risks of intraarticular corticosteroid injection include infection, bleeding, steroid flare reaction, glycemic lability in patients with diabetes, skin discoloration, and skin/subcutaneous fat atrophy [10]. Although intraocular pressure elevation is not a well-described side effect of intraarticular corticosteroid injection, Taliaferro and colleagues found that nearly 30% of patients undergoing intraarticular knee corticosteroid injections demonstrate increased intraocular pressure (> 7 mm Hg) at 1 week after the injection, which remains elevated in about half of those patients at 1 month [11].

This is important because the number of adults in the United States with knee osteoarthritis (OA) is increasing. In fact, about 37% of the US adult population older than 60 years of age have radiographic signs of knee OA, and some 12% develop symptoms [5]; stated another way, approximately 26.9 million adults in the US are living with knee OA [8]. Currently, approximately 1.2 million knee replacements are performed each year and by 2030, the annual number of TKAs are expected to grow to 3.5 million per year [7]. Approximately 30% of patients undergoing TKA have attempted nonsurgical treatment with a corticosteroid injection [9].

The current American Academy of Orthopaedic Surgeons (AAOS) Nonoperative Management of Knee OA Clinical Practice Guidelines recommendations regarding the use of intraarticular corticosteroid injections are inconclusive [6], whereas the guidelines strongly recommend against hyaluronic acid largely due to lack of demonstrable clinical efficacy in methodologically sound randomized trials [2, 6].

Where Do We Need To Go?

The rise in intraocular pressure observed in a substantial number of patients receiving intraarticular knee corticosteroid injections is concerning given the number of people receiving this procedure. Are orthopaedists, rheumatologists, and primary care doctors missing a potentially important side effect? While the risk of sustained corticosteroids causing visual damage in patients with glaucoma is well described in ophthalmology [1, 3, 4], the risk of a single corticosteroid injection to cause permanent visual damage is unknown and has not been previously described, which suggests that the risk is small. The current study does an excellent job describing the problem, but we still need to determine the risk of visual damage with a knee corticosteroid injection.

Primary open angle glaucoma affects more than 66 million individuals worldwide [12]. Patients with glaucoma other than primary open-angle glaucoma were excluded from the current study, which is interesting because patients with primary open-angle glaucoma are at increased risk of steroid induced intraocular pressure elevations compared to patients with closed-angle glaucoma [4, 12]. Prior studies have shown that steroid-induced intraocular pressure elevations may take 10 days to 14 days to manifest, and so by measuring intraocular pressure at 7 days, it is possible that the current study design missed additional steroid responders.

There are several questions that arise after reading this excellent article. What is the risk of a corticosteroid injection causing sustained elevated intraocular pressure and optic nerve damage in patients with glaucoma? Is it possible for an intraarticular corticosteroid injection to induce glaucoma in a susceptible patient? As this is unreported in the medical literature, one can assume the risk is low. Finally, since the risk of intraarticular corticosteroid injections in glaucoma patients is unknown, should patients be screened for glaucoma prior to corticosteroid injection for knee OA? The current evidence does not support this, but it may be a topic for future studies.

The current study design measured intraocular pressure [11], which is a surrogate for risk of optic nerve injury and visual field damage. Historically, ophthalmologists screened for optic nerve damage using slit lamp examinations and subjective visual field testing, both of which are not sensitive for detecting early damage. Newer tests for optic nerve damage use ocular coherence tomography to detect earlier damage by measuring retinal nerve fiber layer thickness at the optic disc, and ganglion cell analysis to measure the thickness of ganglion cells in the macula.

How Do We Get There?

Future studies should focus on the risk of optic nerve damage in patients receiving corticosteroid injection for knee OA. A cohort study that incorporates both an ocular coherence tomography exam as well as intraocular pressure measurement would be helpful to assess for early optic nerve damage following a corticosteroid injection for knee OA. Ideally, this larger study would include an arm of patients with glaucoma, both primary open-angle glaucoma and closed-angle glaucoma, to determine whether the risk of optic nerve damage/visual disturbance is elevated in glaucoma patients. The current study describes a previously unknown side effect of intraarticular knee corticosteroid injections in a substantial number of patients. While the clinical effects of elevated intraocular pressure following an intraarticular knee corticosteroid injection remain unknown, given the prevalence of knee arthritis, the topic warrants further investigation.