Abstract
It has been suggested that occult infection of joint prostheses contributes to a proportion of aseptic loosening. The aims of the study were to determine the incidence of occult infection in a sample of patients undergoing revision surgery for aseptic loosening and examine the role of ultrasound sonication in its detection.
A prospective trial was conducted at Christchurch and Burwood Hospitals. At the time of revision surgery, intra-operative tissue and fluid samples were taken. The removed prosthesis was immersed in saline solution in a sterile plastic container, and then sonicated. The sonicate fluid underwent prolonged routine cultures (14 days) to increase the rate of detection of slow growing organisms. The cases were patients undergoing revision surgery for aseptic loosening or infection. The control group was comprised of patients having revision surgery for any other indication. These implants were subjected to the same protocol as the study group.
A total of 122 patients were included in the study; 54 in the Aseptic Loosening [AL] group, 15 Infections and 53 controls. There were significantly more smokers in the AL group and less smokers in the control group (p=0.04 and p=0.04 respectively). The mean age for revision in the Infection and Periprosthetic fracture groups was less than those of other groups (p=0.007 and p=0.02) respectively. There were 18 cases with positive intra-operative cultures. Eight of those were in the aseptic group (i.e. 14.8% of the group). Conventional sampling techniques were positive in 17 of 18 cultures (94%). Sonication was only positive in 10 out of the 18 cultures (56%). Sonication was concordant with the conventional sampling techniques in half of the positive cultures in the AL group and overall. The only bacteria to be isolated from sonicate cultures were Staphylococcus Aureus and Coagulase Negative Staphylococci. Diabetes Mellitus was the only risk factor to have a significant association with having a positive culture result (p=0.03). There was also a significant association with having raised pre-operative Neutrophil differential count or inflammatory markers with having a positive culture (p=0.0001). However this association was not present when the AL group was examined separately.
There was a significant rate of positive culture results in the aseptic loosening group of around 15%. Ultrasound sonication was less sensitive than current sampling techniques with no apparent added benefit. This paper does not support the hypothesis that occult infection is a significant driver of aseptic loosening.
