Purpose: The purpose of this paper is to evaluate the results of One or Two-Stage revision surgery, in the treatment of infected Total Knee Arthroplasty.

Materials and Methods: Between 1978 and 2003, 43 cases of infected TKA were operated. Forty-one cases were followed more than one year. Twenty two patients were males and 19 females. Average follow-up (F-U) time was 8 years (range 1–25). Nine were diagnosed during the first six months postop and 32 cases were late infections. All patients were given intravenous antibiotics for 6 weeks according to culture.

One stage procedure was performed in 20 patients with a painful knee without purulent discharge. Of them, 9 patients were infected with Staph Aureus (45%), 2 Enterococcus, 1 Strep Viridans, 1 Brucella. Seven were clinically infected without positive cultures (35%). Average age at revision was 75.6Y, with 1.9 co-morbidities. Average time interval between primary and revision was 45.6 months. Gallium bone scan was positive in 40%, WBC was above 11000 in 20%, increased ESR in 25% and CRP values above 3 mg/L in 80%. Intra-operative loosening was found in 60% of femoral and tibial components.

Two stage procedure was performed in 21 patients, when pus was present during the operation. Interposed cement impregnated with antibiotics was left between bone ends and at least six weeks of IV antibiotics were given. Average age at revision was 68Y,with 2 co-morbidities. Average time interval between primary and revision was 20 months. Positive Gallium bone scan was in 100%, WBC above 11000 in 80%, increased ESR in 80% and CRP values above 3 in 63%. Twelve patients were infected with Staph Aureus (60%), 4 Enterococcus, 1 Strep Haemolyticus, 1 Candida and 3 clinically infected cases without positive cultures (14%). Intraoperative loosening was found in 43% and 79% of femoral and tibial components, respectively.

Objective result was graded according to HSS score. The differences between ‘one-stage’ and ‘two stage’ groups were checked by t-test of variance.

Results: Twenty four patients out of 41 (60%) were available for F-U, 11 among One Stage group and 13 among Two Stage group. Eleven patients died and 6 were lost for F-U. Subjective satisfaction was reported by 80% of patients. Overall, HSS score increased from 48.3 preoperatively to 80 postoperatively, with 30% excellent result, 48% good, 13% fair and 9% poor.

For One Stage group, postoperative subjective satisfaction was reported by 78% of patients and HSS increased from 49.9 to 81.4. Average arc of motion was 93°. There were 4 patients with recurrent infection (20%), 2 of them were treated with antibiotics and 2 needed second revision (10%).

For Two Stage group, postoperative subjective satisfaction was reported by 82% of patients and HSS increased from 47.3 to 78.6. Average arc of motion was 85°. No re-infection was recorded in this group.

No significant difference between the two groups was found in relation to postoperative HSS score and arc of motion. However, remarkable difference is reported regarding reinfection rate.

Discussion and Conclusion: In this series with an average of 8 years follow-up, HSS score in revision of infected TKR is lower (30% excellent and 48% good) than expected result in primary TKR. The ‘Two-Stage’ procedure was more effective in our hands than ‘One-Stage’ as for eradication of infection in Total Knee Arthroplasty.