The diagnostic accuracy of radionuclide arthrography for prosthetic loosening in hip and knee arthroplasty.

PURPOSE OF STUDY

Diagnosis of prosthetic loosening in hip and knee arthroplasty remains a challenge. Although there are a number of diagnostic tools, no single test or combination is 100% sensitive or specific. There has been a recent interest in the use of radionuclide arthrography (RNA) for detection of prosthetic loosening.

METHODS

A retrospective review of 45 consecutive RNA scans from 2005 to 2010 was conducted. RNA findings were compared with intraoperative findings at revision and/or serial radiographic examinations to confirm loosening. A component was considered loose if sequential radiographs demonstrated macromotion, gross subsidence, or progressive radiolucency.

RESULTS

There were 26 females and 17 males, with mean age at RNA of 71 years (range of 53-89 years) and mean time from index surgery, 6.4 years (range of 0.5-23 years). There were 23 total knee replacements (TKR) (19 primary and 4 revision) and 20 total hip replacements (THR) (11 primary and 9 revision). 15 patients underwent revision surgery following RNA. Strict inclusion criteria allowed 27 patients for further analysis. Sixteen RNA scans were suggestive of loosening, of which 14 were confirmed loose. Eleven scans were suggestive of a stable prosthesis, of which 10 were confirmed well fixed. RNA had a sensitivity of 93%, specificity of 83%, positive predictive value of 88%, and negative predictive value of 91%.

CONCLUSION

Radionuclide arthrography should be considered a useful adjunct in the diagnosis of prosthetic loosening in the challenging patient.