Award-winning study evaluates cost, effectiveness; outlines guidelines

Is a postsurgical fever after total joint arthroplasty (TJA) a sign of an infection or simply the body’s natural response to tissue injury? Are there ways to tell the difference?

At the 2009 annual meeting of the American Association of Hip and Knee Surgeons, Erik Hansen, MD, presented the results of a study showing that, in most cases, a postsurgical fever work-up after TJA is neither necessary nor cost-effective. Moreover, the study authors were able to identify specific indications that can point to the need for a fever work-up and the most effective tests to use. The study, “Cost and effectiveness of diagnostic fever evaluation in arthroplasty patients,” won the James A. Rand Award.

“Routine postsurgical fever evaluation after TJA rarely changes clinical management,” said Dr. Hansen. “But factors do exist that can improve the diagnostic yield of these work-ups and aid in clinical management.”

Retrospective review used
The authors performed a retrospective review of inpatient electronic medical records to identify patients who underwent hip or knee arthroplasties at a single institution during a 2-year period (July 1, 2006–June 30, 2008). Standard demographic data included the type of arthroplasty, the length of hospital stay, and inpatient complications; fever was defined as a temperature greater than or equal to 38.5ºC (101ºF) and further characterized by maximum temperature, day of occurrence, and frequency.

“All patients received appropriate antibiotic prophylaxis, both before and after the procedure,” noted Dr. Hansen, “as well as daily thromboembolism prophylaxis, unless this was contraindicated.”

Complete data were available for 1,100 patients, including 358 primary total knee arthroplasties, 425 primary total hip arthroplasties, 132 revision knee arthroplasties, and 185 revision hip arthroplasties. Most patients were in the 51–70 age group and had a body mass index between 26 and 35. “Knee patients were older and heavier,” reported Dr. Hansen, “and, not surprisingly, revision patients had longer hospital stays.”

Fever—usually a single event occurring during the first 3 days after surgery—was documented in 161 TJA patients. Of these patients, 69 received some form of diagnostic evaluation (urinalysis; urine culture; blood, wound, or sputum culture; chest radiograph), but the tests resulted in a change in clinical management for only 9 patients (Fig. 1).

“Two thirds of the patients for whom a positive evaluation resulted in change in clinical management had prolonged, complicated hospital stays,” said Dr. Hansen. Only three patients had simple urinary tract infections and a relatively uncomplicated stay, illustrating the importance of using corroborating evidence to guide initiation of diagnostic evaluations.

Telling when trouble’s coming
“Additionally, when we specifically looked at the tests, both urinalysis and urine cultures had a 23 percent positive rate,” said Dr. Hansen. “Blood cultures for bacteria had a 6 percent positive rate, and chest radiographs for pneumonia had a 2 percent positive rate (Fig. 2).”

Although the researchers couldn’t identify an independent demographic variable predictive of a positive response, they were able to outline the following characteristics of a febrile response that would predict a positive evaluation:

• fever after postsurgical day 3
• multiple days of fever
• maximum temperature greater than 39ºC (102ºF)

Is the test worth the cost?
The total cost for all fever evaluations for the patients involved in this study was about $74,000, reported Dr. Hansen. The overall cost per positive fever evaluation was approximately $3,300, and the cost per change in treatment management was more than $8,000.

In this study, the low rate of positive fever evaluations and the low diagnostic yield of the individual tests in the absence of pertinent physical findings are similar to the conclusions of studies in other surgical subspecialties. “Given the large volume of TJAs performed each year, the projected increase in the need for total joint replacements in the Medicare population, and the simultaneous decrease in physician reimbursement, it is of particular import to control costs for this procedure,” wrote the authors.

“In an era of skyrocketing healthcare costs and the recent emphasis by Medicare on reasonably avoidable hospital-acquired conditions, the development of cost-effective, evidence-based practice guidelines for the evaluation of the febrile arthroplasty patient is needed to minimize practice variation and limit waste without compromising patient care,” concluded Dr. Hansen. The authors hope the results of this study will be useful in helping establish clinical practice guidelines for the management of postsurgical fever in the TJA patient.

Dr. Hansen’s co-authors include Derek Ward, BA; Steve Takemoto, PhD; and Kevin J. Bozic, MD, MBA. Drs. Hansen and Takemoto report no conflicts. Dr. Bozic serves as a consultant for United Health Care, The Centers for Medicare & Medicaid Services, Pacific Business Group on Health, and Integrated Health Care Association.

Mary Ann Porucznik is managing editor of AAOS Now. She can be reached at porucznik@aaos.org

Bottom line

• Postoperative fever is common after total joint arthroplasty.
• The standard fever work-up is low-yield and rarely changes clinical management.
• Blood cultures or radiographs as part of fever evaluation have limited utility.
• Request a fever evaluation only if the fever occurs after postoperative day 3, continues for multiple days, or exceeds 39ºC (102ºF).