Standard cutoff values for infection not accurate within 6 weeks of surgery

Bedair H. Clin Orthop Relat Res. Published online: June 29, 2010.

The current standard cutoff value for synovial fluid white blood cell count should not be used during the immediate postoperative period, according to Harvard Medical School researchers.

In their study, Hany Bedair, MD, and colleagues at Harvard Medical School, Boston, assessed how well four common laboratory tests — synovial white blood cell count, differential, C-reactive protein and erythrocyte sedimentation rate — detected infection in the first 6 weeks after primary total knee arthroplasty (TKA).

After reviewing 11,964 primary TKAs, the researchers identified 146 with a knee aspiration within 6 weeks of surgery. Nineteen out of 146 knees developed an infection, diagnosed by positive culture or gross purulence.

To determine which tests could diagnose infection within 6 weeks of surgery, Bedair and colleagues compared demographic information, time from surgery and the laboratory test values between infected and uninfected knees. They constructed receiver operating characteristic curves to designate optimal cutoff values for each test parameter.

The optimal synovial white blood cell cutoff for diagnosing infection was 27,800 cells/µL (sensitivity, 84%; specificity, 99%; positive predictive value, 94%; negative predictive value, 98%). The optimal cutoff for differential was 89% for polymorphonuclear cells and 95 mg/L for C-reactive protein.

Results showed that the infected knees had higher synovial white blood cell counts (92,600 cells/µL vs. 42,000 cells/µL), percentage of polymorphonuclear cells (89.6% vs. 76.9%) and C-reactive protein (171 mg/L vs. 88 mg/L).

“With a cutoff of 27,800 cells/µL, synovial white blood cell count predicted infection within 6 weeks after primary TKA with a positive predictive value of 94% and a negative predictive value of 98%,” the authors wrote. “The use of standard cutoff values for this parameter (~ 3000 cells/µL) would have led to unnecessary reoperations.”

The investigators of this study should be commended for their valuable work and insight into a commonly frustrating clinical scenario. The diagnosis of infection after joint arthroplasty can be a tremendous challenge, requiring the interpretation of several ambiguous and often conflicting laboratory tests. Over the last 10 years, many groups have published research that has focused on the standard cutoff laboratory values that optimally differentiate between infected and aseptic TKAs. We have come to know that TKAs with an associated synovial fluid cell count of greater than 2500-3000 cells/µL are likely infected. However, evaluating certain specific clinical scenarios, such as the immediate postoperative knee, continues to frustrate surgeons, as standard cutoff values for infection are inaccurate during postoperative inflammation.

This study has provided surgeons with three important considerations. First, synovial fluid from a sterile TKA within 6 weeks of surgery is inflamed and may have what we would otherwise consider a very high white cell concentration. The standard cutoff for infection of 2500-3000 cells/µL cannot be used during this postoperative period. Secondly, the cutoff value of 27,800 cells/µL appears to provide the best differentiation between infected and sterile knees within 6 weeks of TKA. The use of this cutoff value does not require a calculated adjustment for the red blood cell concentration. Third, the authors found that when a CRP value of greater than 100 mg/L is observed during the postoperative period, there is a high likelihood of infection, suggesting the need for a diagnostic aspiration. This cutoff for the CRP value is almost 10-fold higher than what is generally used to diagnose infection. Therefore, surgeons evaluating a TKA in the immediate postoperative period should beware of misinterpreting standard laboratory tests for infection, but they can be guided by the initial data the authors of this study have provided. Diagnosing infection after joint arthroplasty remains a challenge, but the study authors have provided us with another important pearl.

– Carl Deirmengian, MD
3B Orthopaedics
Director of Arthritis and Joint Replacement Research
The Lankenau Institute for Medical Research
Wynnewood, Pennsylvania