Synovial NGS Concordance with ICM Criteria for PJI Diagnosis

Synovial NGS Concordance with ICM Criteria for PJI Diagnosis

Periprosthetic joint infection (PJI) is one of the most consequential complications of hip and knee arthroplasty, and it is also one of the hardest to confirm. No single test settles the question on its own. The International Consensus Meeting 2018 (ICM-18) definition addresses this by combining major and minor criteria, yet major criteria such as two phenotypically matching positive cultures or a sinus tract communicating with the joint are not met in up to 29 percent of cases (Sigmund et al., 2022). Culture-negative PJI compounds the problem, occurring in roughly 28 to 46 percent of cases and leaving the surgeon without organism-level guidance for antimicrobial therapy (Klement et al., 2018; Goswami et al., 2022).

This is the gap that molecular diagnostics are meant to narrow. A central question for surgeons and infectious disease specialists is whether broad-range microbial sequencing of synovial fluid actually agrees with the biomarker-based criteria clinicians already trust, or whether its added sensitivity simply inflates the false-positive rate. A new peer-reviewed study answers that question directly. Tipton et al., published in Frontiers in Microbiology in 2026, analyzed 2,011 synovial fluid specimens with matched next-generation sequencing (NGS) and synovial biomarker results, then measured how closely NGS positivity tracked the ICM-18 minor criteria for infection (DOI: 10.3389/fmicb.2026.1816780). The findings speak to a practical concern in revision arthroplasty: a positive diagnosis before surgery gives the surgeon clarity and a direction of treatment, rather than modifying management after the fact.

What the study measured

The analysis drew on MicroGenDX records of synovial fluid specimens submitted by outpatient surgical clinics between December 2020 and November 2022. Each specimen carried matched data: NGS microbial profiling and three synovial biomarkers used in the ICM-18 minor criteria, namely C-reactive protein (CRP), white blood cell (WBC) count, and polymorphonuclear neutrophil percentage (PMN%). Specimens were scored against ICM-18 minor criteria thresholds, with a cumulative score of four or greater treated as high probability for infection. The cohort skewed toward the knee, which accounted for 86 percent of specimens, with the hip making up the remaining 14 percent. The study was reviewed by the Advarra Center for IRB Intelligence and certified as IRB exempt.

The molecular workflow used targeted amplification and sequencing of the V1-V2 regions of the 16S rRNA gene for bacterial profiling and the ITS3-4 locus for fungal profiling, run on the Illumina MiSeq platform. A standardized positive-or-negative reporting protocol was applied, with contamination assessed against negative extraction and no-template controls and low-abundance signals removed, a design intended to limit false-positive reporting. The test evaluated is the OrthoKEY laboratory-developed test, which the authors note is in good standing with College of American Pathologists (CAP) accreditation, Clinical Laboratory Improvement Amendments (CLIA) licensing, and New York State’s Clinical Laboratory Evaluation Program (CLEP).

Concordance with ICM minor criteria

Across all 2,011 specimens, NGS positivity showed substantial agreement with ICM-diagnosed PJI. The kappa statistic was 0.68, and overall accuracy reached 88.7 percent (95 percent CI, 0.873 to 0.90). Sensitivity was 76.4 percent (0.723 to 0.801) and specificity was 92.3 percent (0.909 to 0.935). Performance was consistent between joints, with knee accuracy at 88.9 percent and hip accuracy at 87.5 percent, and a positive predictive value of 74.3 percent against a negative predictive value of 93.1 percent.

The specificity finding is the one most relevant to the over-diagnosis concern. NGS was more specific to PJI than synovial CRP, which came in at 89.4 percent, though it did not exceed the specificity of WBC (97.9 percent) or PMN% (94.7 percent). NGS produced positive results in 7.7 percent of ICM-negative samples, a false-positive rate the authors describe as comparable to established biomarkers rather than evidence of runaway over-calling. For context, a recent meta-analysis of synovial fluid culture reported 96 percent specificity alongside a pooled sensitivity of 63 percent (Watanabe et al., 2024), against which the study’s NGS sensitivity of 76.4 percent compares favorably.

Potential signals for sub-clinical infection

One of the more clinically interesting findings sits inside the ICM-negative group. NGS positivity rose in a near-linear fashion with cumulative ICM score (R-squared = 0.93, p = 0.0028). Among samples scored as low probability for infection, an NGS-positive result was significantly associated with elevated synovial WBC (p < 0.0001) and PMN (p = 0.0011), though not with CRP. In other words, some NGS-positive specimens that fell below the biomarker threshold for PJI still carried quietly elevated cellular markers.

The authors interpret these discrepant results cautiously, suggesting they may reflect infection from organisms that elicit a weak immune response, subclinical infection, or early infection that has not yet declared itself by conventional markers. This is a hypothesis-generating observation rather than a settled conclusion, but it identifies a subpopulation of patients in whom molecular detection may surface infection that biomarker scoring alone would miss, particularly in difficult cases involving fastidious organisms with ambiguous biomarker values.

Uncommon organisms are collectively common

The microbial profile reinforces a familiar message in PJI literature. Eight bacterial species dominated the composition in 68 percent of samples, while the remaining third comprised 46 different dominant species. Staphylococcal species were the most frequent dominant organisms, reported in 51 percent of samples, including S. epidermidis, S. aureus, S. lugdunensis, and S. capitis. Of the NGS-positive specimens, 22.3 percent were polymicrobial.

That long tail matters. Any single uncommon organism, such as Enterococcus faecium or Streptococcus dysgalactiae, is individually rare, yet uncommon species as a group are a routine feature of PJI and are precisely the organisms most likely to be overlooked by limited testing. The study notes that anaerobes such as Anaerococcus, Bacteroides, and Finegoldia magna were frequently observed and are underreported by culture-dependent methods. Cutibacterium acnes was also abundant and is well known to require anaerobic isolation and extended incubation for reliable culture detection. The relevance of atypical pathogens in PJI is further underscored by a recent case study highlighting PJI caused by a rare zoonotic pathogen, Erysipelothrix rhusiopathiae (Abraham et al., 2026). By comparison, commonly used multiplex qPCR joint infection panels report a fixed organism set, with reported sensitivity to PJI in the range of 41 to 56 percent and a known tendency to underreport polymicrobial infection.

A complementary tool, not a replacement

The study’s authors are explicit that NGS is not positioned to replace synovial WBC or PMN, both of which outperformed it on overall sensitivity and accuracy for diagnosing PJI. Instead, the data support a complementary role similar to the one fluid culture occupies: a test that adds value when the diagnosis is unclear by traditional markers, and when capturing the causative organism preoperatively would change management. In culture-negative cases, in patients with a high pretest probability of infection, or when a rare or fastidious pathogen is suspected, broad-range sequencing offers detection breadth that biomarker scoring and limited molecular panels do not provide. That breadth supports targeted antimicrobial therapy and stewardship rather than empiric treatment in the dark.

The Cutibacterium acnes finding illustrates the point. Because C. acnes can be slow and difficult to recover by culture and does not always provoke a strong host response, early molecular confirmation can provide impetus toward treatment rather than leaving equivocal biomarker data unresolved. When detected and treated, C. acnes joint infections have shown high treatment success and favorable two-year survival in the cited literature, indicating that this organism is manageable when it is identified.

Weighing cost against clinical value

Cost remains a real barrier to molecular adoption. An IDSA emerging infections network survey found that 33 percent of infectious disease physicians had been prevented from using molecular diagnostic testing by cost and payor concerns, even though 78 percent considered molecular testing occasionally or often helpful in clinical decision making (Rao et al., 2024). The economics, however, favor the approach described here. The test evaluated in the study is typically priced at ~$400 per test, while an earlier cost-effectiveness analysis concluded that NGS could be cost effective up to ~$3,916 per test, assuming at least 71 percent sensitivity and 94 percent specificity (Torchia et al., 2019). The performance reported by Tipton et al. sits within those assumptions, and the price sits well below the modeled threshold.

What the study does not establish

The findings carry limitations worth stating plainly. NGS results were compared only to ICM-18 minor synovial criteria, not to culture, to serum biomarkers such as D-dimer and sedimentation rate, to alpha-defensin, or to full medical-record context. The analysis was retrospective and based on central-laboratory records, without access to antibiotic exposure at the time of sampling or to downstream treatment outcomes. The authors note that a single-center prospective follow-up comparing matched NGS, culture, and additional data points is underway, which will help define the precise margin of benefit.

Where OrthoKEY fits

This study is a direct evaluation of OrthoKEY, the MicroGenDX laboratory-developed test for periprosthetic joint infection. OrthoKEY combines qPCR and NGS with ICM-aligned biomarkers for periprosthetic joint infection diagnostics, reporting CRP, WBC, and PMN% alongside molecular results so that clinicians receive both the biomarker picture and organism-level detail from a single synovial fluid aspirate. The test is designed to detect polymicrobial PJI, including culture-negative cases, which is where the breadth of broad-range sequencing is most useful.

OrthoKEY at a glance

• qPCR and NGS with ICM-aligned biomarkers (CRP, WBC, PMN%) reported alongside molecular results
• Detects polymicrobial PJI, including culture-negative cases
• Broad-range 16S and ITS sequencing detects bacterial and fungal species across a 60,000-plus microbe reference database
• Up to 17 antimicrobial resistance (AMR) genes reported across the panel
• qPCR and NGS results typically available within 24 to 48 hours from sample receipt
• Single synovial fluid aspirate; typically priced at ~$400 per test

MicroGenDX is a CAP-accredited, CLIA-certified, CLEP-permitted (NYS DOH) molecular diagnostics laboratory. For transparency, Tipton et al. is MicroGenDX-authored research, published open-access and peer-reviewed in Frontiers in Microbiology.

To review the full study or learn how OrthoKEY supports preoperative PJI diagnosis:
CLICK HERE

References

1. Tipton CD, Ancira J, Tarabichi S, Shahi A, Jarvis K, Omeir K, Sanford N, Tallman NA, White J, Phillips CD, Parvizi J, McPherson EJ. Microbial next generation DNA sequencing of aspirated synovial fluid shows concordance with ICM criteria biomarkers for diagnosing periprosthetic joint infection in hip and knee arthroplasty. Front Microbiol. 2026;17:1816780. doi:10.3389/fmicb.2026.1816780
2. Sigmund IK, Luger M, Windhager R, McNally MA. Diagnosing periprosthetic joint infections. Bone Joint Res. 2022;11:608-618. doi:10.1302/2046-3758.119.BJR-2022-0078.R1
3. Klement MR, Siddiqi A, Rock JM, Seyler TM, Parvizi J, Chen AF. Are all periprosthetic joint infections the same? Evaluating major vs minor criteria. J Arthroplast. 2018;33:1515-1519. doi:10.1016/j.arth.2017.12.010
4. Goswami K, Clarkson S, Phillips CD, et al. An enhanced understanding of culture-negative periprosthetic joint infection with next-generation sequencing: a multicenter study. J Bone Joint Surg Am. 2022;104:1523-1529. doi:10.2106/JBJS.21.01061
5. Watanabe S, Kamono E, Choe H, Ike H, Inaba Y, Kobayashi N. Differences in diagnostic sensitivity of cultures between sample types in periprosthetic joint infections: a systematic review and meta-analysis. J Arthroplast. 2024;39:1939-1945. doi:10.1016/j.arth.2024.03.016
6. Rao PS, Downie DL, David-Ferdon C, et al. Pathogen-agnostic advanced molecular diagnostic testing for difficult-to-diagnose clinical syndromes. Open Forum Infect Dis. 2024;11:ofae395. doi:10.1093/ofid/ofae395
7. Torchia MT, Austin DC, Kunkel ST, Dwyer KW, Moschetti WE. Next-generation sequencing vs culture-based methods for diagnosing periprosthetic joint infection after total knee arthroplasty: a cost-effectiveness analysis. J Arthroplast. 2019;34:1333-1341. doi:10.1016/j.arth.2019.03.029
8. Parvizi J, Tan TL, Goswami K, Higuera C, Della Valle C, Chen AF, et al. The 2018 definition of periprosthetic hip and knee infection: an evidence-based and validated criteria. J Arthroplast. 2018;33:1309-1314.e2. doi:10.1016/j.arth.2018.02.078
9. Abraham M, et al. A common infection with an uncommon organism: periprosthetic joint infection of the knee caused byErysipelothrix rhusiopathiae. 2026.

For Healthcare Professionals. MicroGenDX is a CAP-accredited, CLIA-certified, CLEP-permitted (NYS DOH) molecular diagnostics laboratory. Customer service: (855) 208-0019  |  [email protected]