Variation in selection of definitive versus spacer-based exchange for knee periprosthetic joint infection

Purpose To examine selection of coded same-admission definitive versus spacer-based exchange for knee periprosthetic joint infection (PJI), and variation remaining after adjustment for recorded administrative characteristics. Methods Adults undergoing complete implant removal for side-specific knee PJI were identified in the 2020–2022 Nationwide Readmissions Database. Discharge-weighted logistic regression estimated associations. A separate unweighted random-intercept model described heterogeneity among sampled hospitals within each year. Annual hospital identifiers cannot be linked across years. Future-admission indicators and retrospective annual volume were excluded from the principal statistical models. Ten-fold validation kept hospital clusters together; stratified cluster bootstrapping quantified uncertainty. Results Among 15,425 patient-year episodes, 4,667 involved definitive exchange and 10,758 spacer treatment; weighted definitive-exchange use was 29.4%. The C-statistic was 0.576 (95% CI 0.564–0.587). Across 3,115 hospital clusters, adjusted random-intercept variance was 1.0228, median odds ratio (MOR) 2.62 (95% CI 2.45–2.81), and intraclass correlation 23.7% (95% CI 21.2%-26.3%). MORs were 2.64 with a common 90-day lookback, 2.63 in 2022 alone, and 2.49 among clusters with at least 11 eligible cases. Restriction to observed mixed-strategy clusters reduced the MOR to 1.79; this outcome-conditioned comparison was descriptive. Conclusion Recorded administrative characteristics weakly discriminated coded treatment selection. Residual heterogeneity among sampled hospital clusters may reflect unmeasured clinical eligibility, referral case mix, coding, or practice. These data cannot establish institutional capability, persistent hospital preferences, or an appropriate clinical selection score.

Authors

Institutions

Publication Details

Journal
Langenbeck s Archives of Surgery
Published
2026-11-21
DOI
https://doi.org/10.1007/s00423-026-04388-6
Primary Topic
Orthopedic Infections and Treatments
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Variation in selection of definitive versus spacer-based exchange for knee periprosthetic joint infection

Yaniv Steinfeld, Yaron Berkovich, David Maman
Langenbeck s Archives of Surgery
Orthopedic Infections and Treatments
article

Variation in selection of definitive versus spacer-based exchange for knee periprosthetic joint infection

Yaniv Steinfeld, Yaron Berkovich, David Maman
article en

Abstract

Purpose To examine selection of coded same-admission definitive versus spacer-based exchange for knee periprosthetic joint infection (PJI), and variation remaining after adjustment for recorded administrative characteristics. Methods Adults undergoing complete implant removal for side-specific knee PJI were identified in the 2020–2022 Nationwide Readmissions Database. Discharge-weighted logistic regression estimated associations. A separate unweighted random-intercept model described heterogeneity among sampled hospitals within each year. Annual hospital identifiers cannot be linked across years. Future-admission indicators and retrospective annual volume were excluded from the principal statistical models. Ten-fold validation kept hospital clusters together; stratified cluster bootstrapping quantified uncertainty. Results Among 15,425 patient-year episodes, 4,667 involved definitive exchange and 10,758 spacer treatment; weighted definitive-exchange use was 29.4%. The C-statistic was 0.576 (95% CI 0.564–0.587). Across 3,115 hospital clusters, adjusted random-intercept variance was 1.0228, median odds ratio (MOR) 2.62 (95% CI 2.45–2.81), and intraclass correlation 23.7% (95% CI 21.2%-26.3%). MORs were 2.64 with a common 90-day lookback, 2.63 in 2022 alone, and 2.49 among clusters with at least 11 eligible cases. Restriction to observed mixed-strategy clusters reduced the MOR to 1.79; this outcome-conditioned comparison was descriptive. Conclusion Recorded administrative characteristics weakly discriminated coded treatment selection. Residual heterogeneity among sampled hospital clusters may reflect unmeasured clinical eligibility, referral case mix, coding, or practice. These data cannot establish institutional capability, persistent hospital preferences, or an appropriate clinical selection score.

Langenbeck s Archives of Surgery
Carmel Medical Center (IL), Rappaport Family Institute for Research in the Medical Sciences (IL)
Orthopedic Infections and Treatments
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon