Objective: To assess the possible effect of intra-articular steroid injections to future TKA.

Materials-Method: We retrospectively studied all 231 patients who underwent AGC (Biomet) TKR in our hospital from February 2002 to October 2004. Twenty notes were not available in medical records and were excluded from the study. Other exclusion criteria were previous surgery (other than knee arthroscopy) on the affected site, a diagnosis of inflammatory arthritis, immunosu-pressed patients, a previous history of infection around the knee, smoking, diabetic patients. Applying these criteria we excluded a further sixty-seven patients.

The remaining 144 patients were separated in to two groups. Group I (n=54) consisted of those patients that received one or more recorded I/A steroid injections in their operated knee in an orthopaedic clinic, rheumatology clinic or general practice setting prior to surgery. Group II (n=90) consisted of those patients with no record of receiving an I/A steroid injection prior to surgery.

Results: We found that all the deep infections (3) were from Group I and had received an I/A steroid injection up to 11 months prior to surgery. The incidence of superficial infection was not significantly different from the control group (Group II).

In addition to those patients with confirmed deep infections, five patients underwent post-operative investigations for suspected deep infection, due to symptoms of persistent swelling or pain. All had received an I/A steroid injection pre-operatively.

The length of time between injection and subsequent post-operative infection leads us to speculate that the steroid agent might not fully dissolve, becoming trapped within the soft tissues or cystic areas of degeneration in the knee joint. Such steroids may become re-activated during operation, leading to catastrophic results. Indeed, there is experimental evidence to suggest an increased risk of infection with the intra-operative administration of steroids.

Conclusion: We conclude that the decision to administer intra-articular steroids to a patient who may be a candidate for knee replacement surgery should not be taken lightly because of a risk of post operative deep infection.

We reviewed 231 patients who had undergone total knee replacement with an AGC (Biomet) implant over a period of 2.5 years. After applying exclusion criteria and with some loss to follow-up, there were 144 patients available for study. These were divided into two groups; those who had received intra-articular steroid in the 11 months before surgery and those who had not.

There were three deep infections, all of which occurred in patients who had received a steroid injection. The incidence of superficial infection was not significantly different in the two groups. Five patients had undergone investigation for suspected deep infection because of persistent swelling or pain and all of these had received an intra-articular injection pre-operatively. We conclude that the decision to administer intra-articular steroids to a patient who may be a candidate for total knee replacement should not be taken lightly because of a risk of post-operative deep infection.