To investigate treatment options and perioperative points of revision for periprosthetic joint infection following primary total knee arthroplasty(TKA) and to analyse the short term clinical outcome retrospectively.

A series of nine patients (six women and three men) who met the definition of periprosthetic joint infection by the Musculoskeletal Infection Society following primary TKA and received revision in department of orthopaedics, the Central Hospital of Tongji Medical College were reviewed from July 2009 to June 2016. The mean age was(65±6)years. Two cases received single-stage revision and seven cases received two-stage revision. Antibiotic-loaded cement was used in all the revision. Three cases were mobile knee spacers and four cases were static knee spacers in two-stage revision. Paired t-test was used to compare the preoperative and the last follow-up postoperative mean scores and range of motion of affected knee.

All the patients were followed up for an average of(28±25)months. The range of motion was (43±5)° preoperatively and (87±20)° postoperatively(t=-96.6, P<0.01). The visual analogue scale was (3±2) preoperatively and (7±2) postoperatively (t=-11.3, P<0.01). The Hospital for Special Surgery score improved significantly from (42±10)preoperatively to (84±5) at the last follow-up(t=-127.4, P<0.01). None of the patients with revision developed recurrence of infection.

Revision is effective in the treatment of periprosthetic joint infection following primary TKA; careful preoperative assessment in accordance with the condition of the patients and correct operative options, thorough debridement at infected lesions and appropriate application of antibiotic-loaded cement spacers, and standard postoperative anti-infection treatment are the necessary conditions to obtain satisfactory short term clinical effects.