Postoperative fever: differences between elective vs. traumatic hip, knee and shoulder arthroplasty

It is unclear whether there is a difference in the incidence of postoperative fever (POF) between hip, knee and shoulder arthroplasty. The influence of a trauma setting has not been investigated.

A retrospective review was performed on 675 joint replacement surgeries (hip, knee, shoulder arthroplasty) in an elective or trauma setting over a 2 year period (2016-2018). Patient demographics and perioperative/postoperative parameters were investigated. The fever curve was characterized. The results and costs of any fever-related diagnostic workup were reviewed.

A total of 89 patients (13.2%) experienced a POF ≥ 38.0 °C, only 21 patients (3.1%) a POF ≥ 38.5 °C (of the latter: 4.6% of hip arthroplasties, 0.6% of knee arthroplasties, 0% of shoulder arthroplasties). There was a significantly greater risk (OR 3.88) for POF ≥ 38.5 °C in trauma total hip arthroplasty (THA) compared to elective THA (10.6 vs. 3.0%; p < 0.01). Differences in POF rate between the various joint areas were statistically insignificant when comparing only elective and trauma cases with each other, even though there was a trend for higher rates in hip surgery. Patients experiencing POF ≥ 38.5 °C were more often males (p < 0.01) and had an increased intraoperative blood loss (p = 0.03) and longer hospital stay (p < 0.01). There was only 1/89 POF patients developing an early periprosthetic joint infection. The cost of a positive fever workup (3/27 patients, 11.1%) leading to a new diagnosis and treatment was 2045 US$.

POF ≥ 38.5 °C was more frequent in traumatic THA compared to elective THA. A trend of difference between POF rates between the different joint areas was statistically insignificant after separating elective and traumatic cases. The risk of developing an early periprosthetic joint infection was not increased in POF. A fever-related diagnostic workup was rarely helpful in the absence of clinical symptoms.