Abstract
The reported success rate after treatment with debridement, antibiotics and implant retention (DAIR) of hip prosthesis infections has been found variable. We evaluated all reoperations performed because of infection and reported to Swedish Hip Arthroplasty Register (SHAR) between 1999 and 2016. The analyses were separated into reoperations performed for the first time and those which had been preceded by at least one previous reoperation performed because of the same reason. The outcome was repeated reoperation performed because of infection.
1,882 were first-time procedures (Group I) and 2,275 had been preceded by at least one reoperation due to infection (Group II). Head and/or liner exchange had been performed in 47% of the cases in group I, and in 22% in Group II. The mean age varied between 70 and 71 years and there was a dominance of males in all groups (52–59%). Compared to all primary THR performed during this period (n=319,813) patients with inflammatory disease, idiopathic femoral head necrosis and sequel after childhood disease were overrepresented for this type of procedure.
Between 1999 and 2016 the number of DAIR procedures increased from 29 to 383 per year corresponding to 21 and 72 % of all reoperations performed due to infection. In first time reoperations the survival was 74.5±3.1% if the head/liner had been exchanged and 46.2±3.2% if only irrigation and synovectomy had been performed. In patients reoperated at least one time previously due to infection the survival rates dropped to 68.6±4.6% and 34.5±2.4%.
Compared to first time reoperation with exchange of femoral and/or liner, synovectomy and irrigation without exchange of any implant part(s) resulted in an almost tripled risk of a second reoperation due to same reason (Hazard Ratio: 2.8, 95% confidence interval: 2.4–3.3). In cases previously reoperated because of infection (Group II) exchange of head/liner and debridement had a 28% increased risk of failure compared to the corresponding first time reoperations (1.28 1.02–1.6). If none of the components were replaced in Group II, the risk ratio for a new failure increased almost 4 times (3.8 3.3–4.4). Presence of a cemented stem increased the risk for further reoperations due to infection (1.14 1.02–1.28), but not presence of a cemented cup (1.06 0.92–1.19). All hazard ratios were adjusted for age, gender, diagnosis and type of fixation.
The comparatively good results observed after exchange of head and liner might indicate that this is necessary to perform a sufficiently radical debridement. This observation could also be biased by a surgeon related factor suggesting that component exchange mainly is performed by surgeons with long experience of revision surgery.
