Search

OBJECTIVE. The diagnosis of early prosthetic joint infection (PJI)-defined as within 6 weeks after a total knee arthroplasty (TKA)-can be difficult because of expected postsurgical changes and elevated inflammatory markers. The role of radiographic evaluation in this situation carries unclear clinical significance. This study had three primary aims: first, to determine when soft-tissue gas is no longer an expected postoperative radiographic finding; second, to determine whether soft-tissue gas is predictive of early PJI; and, third, to determine whether the presence of soft-tissue gas correlates with specific patient characteristics and microbiology culture results. MATERIALS AND METHODS. This retrospective study was of patients who underwent TKA from 2008 to 2018 with available imaging between 5 days and 6 weeks after TKA and no interval intervention before imaging. All confirmed early PJIs were included ( n = 24 cases; 15 patients). For comparison, patients who underwent TKA but did not have a PJI ( n = 180 cases; 150 patients) were selected randomly. Radiographs were reviewed by two readers. A two-tailed p < 0.05 was considered significant. RESULTS. Soft-tissue gas was identified on postoperative radiography of 13 of 24 (54.2%) cases (mean ± standard error of the mean [SEM], 28.3 ± 2.3 days after TKA) with early PJI and four of 180 (2.2%) cases (mean ± SEM, 15.3 ± 7.3 days after TKA) without PJI ( p < 0.0001; odds ratio, 52.0 [95% CI, 14.7-156.9]). The presence of soft-tissue gas on radiography 14 days after TKA had a sensitivity of 0.54 (95% CI, 0.35-0.72) and specificity of 0.99 (95% CI, 0.97-1.00) for early PJI. Staphylococcus species were the dominant organisms; cases with soft-tissue gas showed a wider variety of microbiology species ( p < 0.01). CONCLUSION. Postoperative soft-tissue gas present on radiography performed 14 days or more after TKA is predictive of early PJI and is associated with a wider spectrum of microorganisms.