Study identified four risk factors for sepsis after injections for knee pain.
As knee osteoarthritis becomes increasingly prevalent in the aging population, more patients are turning to their physicians for intra-articular injections. But a Chinese retrospective case-control study in Seminars in Arthritis & Rheumatism has underscored several associated risk factors for deep knee infection and chronic low-grade infection, and points to the need for more thorough pre-injection evaluation of patients and stricter training for doctors administering injections.
Foremost among infection risk factors was injection performed by family physicians, which had an odds ratio (OR) of 5.23 (95% CI 2.0-13.67). In addition, a body mass index of ≥ 25 had an OR for infection of 2.3 (95% CI1.1-4.7), and rheumatoid arthritis an OR of 2.61 (95% CI 1.20-5.68). Injection with corticosteroids versus hyaluronic acid had an OR of 3.21 (95% CI 1.63-6.21).
“We underline the importance of the accurate evaluation of clinical history and comorbidities for every IA injection, particularly for patients with obesity and RA and those receiving corticosteroid injections,” wrote Jiying Chen, MD, of People’s Liberation Army General Hospital, Beijing, and colleagues. “Strict training on septic technique is necessary before doctors should perform invasive knee treatments.”
The authors note that post-injection side effects are generally uncommon and mild — approximately 2%-1% per injection — with most complications being injection-site inflammatory reactions such as pain, swelling, and skin or fat atrophy. But though uncommon, deep knee infection can have serious, sometimes life-threatening, consequences, especially with injections on the rise. Chen and associates refer to an epidemiological analysis reporting that in U.S. patients with knee osteoarthritis, over a 5-year period 43.5% of those who eventually underwent total knee arthroplasty were prescribed preoperative intra-articular steroid injections.
The Beijing investigators identified 50 cases of injection-induced knee infection patients undergoing surgery from 2010-2016 and matched them with 250 non-infected controls. The mean age of patients overall was about 67 years and about 65% were female.
There were 21 cases of septic arthritis and 29 cases of chronic low-grade infection, with the former cases showing significantly higher metrics in fever, localized warmth, swelling, resting pain, night pain, limited motion, serum white blood cell counts and C-reactive protein levels.
Patients who developed infection had a slightly higher mean BMI of 26.2 versus 25.2 for controls.
Of those experiencing infection, 60% underwent injection with corticosteroids versus 34.8% of controls, while just 40% had injections of hyaluronic acid versus 65.2% of controls.
In 82% of deep knee infection cases, injection was administered by a general practitioner compared with 43.6% for controls. Just 12% and 6% received injections from an orthopedic surgeon or a rheumatologist, respectively, compared with 28.8% and 27.6%,respectively, for controls.
“This finding suggests that doctors should receive strict training and pass a formal practice exam before they perform invasive knee treatments, such as intra-articular injections,” the authors write.
Bacterial culture determined that Staphylococcus aureus was the by far the most common microorganism in septic arthritis (47.6%), while low-virulence coagulase-negative (CNN) Staphylococcus was most commonly implicated in chronic low-grade infection (31%), followed by Propionibacterium acnes at 24.1%. Other organisms involved in septic arthritis cases were CNN, Streptococcus, Enterococcus, and gram-negative bacilli, all at 9.5%.
“These results completely jibe with what we see in our practice,” Bharat Kumar, MD, of the University of Iowa in Iowa City. “They show that injection technique is extremely important, and part of that is training and part of it is practice.”
Kumar advised physicians to observe strict asepsis measures and evaluate patients carefully before giving injections. “Joint infections don’t happen in a vacuum. There are tissue factors in bone to consider, and you have to look carefully at people who have rheumatoid arthritis, are immunosuppressed, or are sick or weak, as they are at increased risk for infection.”
MedPage Today reported a recent study that suggested corticosteroid injections may also hasten cartilage breakdown.
Among the study limitations reported by Chen et al were its small size, single-center location, and retrospective nature, which introduced the possibility of recall bias. In addition, since some patients injected at the study site may have been followed up at an external referral institution, the investigators were not able to identify all cases of post-injection deep knee infection.