JOA - 2026-11-01 - Journal Article
Diagnostic Accuracy of Alpha Defensin Lateral Flow Assay in Total Hip and Knee Arthroplasty.
Wu M, Bukowski BR, Hickman JA, Theel ES, Patel R, Berry DJ, Abdel MP
Topics
Key Takeaway
Alpha defensin lateral flow assay achieves 97% specificity and 80% overall sensitivity for PJI diagnosis, but sensitivity drops to 57% in spacer aspirations and 67% in THA versus 84% in TKA.
Summary Depth
Choose how much analysis to show on this article page.
Summary
This single-institution retrospective study evaluated ADLFA diagnostic accuracy against 2018 MSIS criteria (excluding ADLFA) across 473 hip and knee arthroplasty aspirations with 142 confirmed PJIs. Overall sensitivity was 80% and specificity 97%, with NPV of 92%; however, recent antibiotic use reduced NPV from 94% to 76% (P<0.001), spacer aspirations reduced sensitivity to 57% versus 82% (P=0.04), and THA sensitivity was 67% versus 84% for TKA (P=0.033). Inflammatory arthropathy reduced TKA specificity to 91% versus 99% in those without (P=0.02), generating false positives.
Key Limitation
Single-institution retrospective design with 22% of patients on antibiotics within two weeks of aspiration introduces uncontrolled variability in antibiotic agent, dose, and duration that cannot be adjusted for in sensitivity analyses.
Original Abstract
BACKGROUND
Accurate diagnosis of periprosthetic joint infection (PJI) remains suboptimal. Of the tests used, one is the alpha defensin lateral flow assay (ADLFA). This study assessed ADLFA for the diagnosis of PJI in total hip arthroplasties (THAs) and total knee arthroplasties (TKAs). We hypothesized that ADLFA would have high specificity and sensitivity for diagnosing PJI, but that certain clinical features would influence test reliability.
METHODS
We reviewed 473 ADLFA tests from 141 THAs and 332 TKAs performed from 2020 to 2023 at a single institution. The mean age was 66 years, 51% were women, and mean body mass index was 32. There were 142 PJIs (2018 Musculoskeletal Infection Society criteria excluding ADLFA). Overall, 49% were primary and 51% revision procedures, 22% were on antibiotics within two weeks of arthrocentesis, 18% had an inflammatory arthropathy, and 7% had an in situ spacer.
RESULTS
The overall sensitivity, specificity, positive predictive value, and negative predictive value of ADLFA were 80, 97, 92, and 92%, respectively. There were 6% (29 of 473) that had false negative results, of which 38% were on antibiotics within two weeks of arthrocentesis and 21% of which were spacer aspirations. The negative predictive value was lower for those who had recent antibiotic use (76%) compared to those who did not (94%; P < 0.001). There was lower sensitivity (57%) for spacer compared to nonspacer aspirations (82%; P = 0.04). Furthermore, sensitivity in THA (67%) was lower compared to TKA (84%; P = 0.033). In TKAs, there was lower specificity (91%) for those who had inflammatory arthropathies (99%; P = 0.02).
CONCLUSIONS
The ADLFA is highly specific in diagnosing PJI and is less sensitive in patients who have antibiotic spacers and those who have THAs compared to TKAs. Additionally, underlying inflammatory arthropathies may decrease specificity, yielding false positive results, specifically in TKAs.