P-97. Early Oral versus Prolonged Intravenous Antibiotics for Osteomyelitis (OVIVA-VA): A Single-Center Veterans Affairs Cohort

TL;DR

Switching to oral antibiotics early for osteomyelitis in veterans is as effective as long IV treatment and causes fewer side effects.

Contribution

Demonstrates that early oral antibiotics are non-inferior to prolonged IV therapy in a Veterans Affairs cohort.

Findings

Early oral therapy had 5.3% adverse events vs 24.8% with prolonged IV therapy.

Treatment failure rates were 21.8% for early oral vs 25.3% for prolonged IV.

Hospital stays were shorter with early oral therapy (7.4 vs 12.2 days).

Abstract

The VA manages high burdens of chronic osteomyelitis (OM). The OVIVA trial demonstrated that an early switch to oral antibiotics is not inferior to extended intravenous (IV) therapy in a predominantly civilian population. Whether a similar benefit holds within VHA remains uncertain.

Early Oral Therapy Associated With Fewer Antibiotic-Related Adverse EventsThirty-day cumulative incidence curves demonstrate markedly fewer antibiotic-related adverse events in the early-oral group (5.3%) than in the prolonged-IV group (24.8%). Shaded ribbons depict 95 % confidence intervals; log-rank test p < 0.001.

Early Oral Therapy Associated With Fewer Antibiotic-Related Adverse Events

Thirty-day cumulative incidence curves demonstrate markedly fewer antibiotic-related adverse events in the early-oral group (5.3%) than in the prolonged-IV group (24.8%). Shaded ribbons depict 95 % confidence intervals; log-rank test p < 0.001.

Non-Inferiority Stands Up to Every Sensitivity CheckForest plot of risk-difference estimates (early oral – prolonged IV) for treatment failure across five weighting/adjustment strategies. Point estimates remain close to zero, and all 95 % confidence intervals sit well within the ±7.5-percentage-point non-inferiority margin, confirming the robustness of the primary finding.

Non-Inferiority Stands Up to Every Sensitivity Check

Forest plot of risk-difference estimates (early oral – prolonged IV) for treatment failure across five weighting/adjustment strategies. Point estimates remain close to zero, and all 95 % confidence intervals sit well within the ±7.5-percentage-point non-inferiority margin, confirming the robustness of the primary finding.

We performed a retrospective single-center cohort study of Veterans with confirmed OM from 1/2017 – 12/2021 at the Ann Arbor VA. We compared patients treated with ≥4 weeks of IV antibiotics to those switched to an oral regimen within 7 days with inverse-probability-of-treatment weighting (IPTW). The primary endpoint was treatment failure within 1 year. Secondary endpoints were antibiotic-related adverse events, catheter-related complications, C. difficile infection, and hospital length of stay.

Baseline Characteristics and Covariate Balance After IPTW

Unweighted patient demographics, comorbidities, and infection features (mean ± SD or n/N %) are compared between early-oral and prolonged-IV groups, with χ² or t-test p-values. The final column shows weighted standardized mean differences (SMDs) demonstrating excellent covariate balance (all SMDs ≤ 0.08) after inverse-probability treatment weighting.

We identified 110 patients treated with early oral antibiotics and 162 patients treated with prolonged IV antibiotics. There were higher rates of S. aureus infection lower rates of surgical source control among patients treated with early extended IV antibiotics. Imbalances were eliminated after weighting (weighted SMDs ≤ 0.08 for all covariates). After weighting, treatment failure occurred in 24/110 (21.8 %) patients on early-oral vs 41/162 (25.3 %) patients on prolonged-IV treatment (risk difference –0.9 pp, 95 % CI –4.2 to 2.4). Antibiotic-related adverse events were 5.3 % in the early-oral group vs 24.8 % in the prolonged IV group (–19.5%, 95 % CI –25.2 to –13.8); C. difficile 0 % vs 6.2 % (–6.2%, 95 % CI –8.6 to –3.8). Adjusted length of stay was 7.4 in the early oral group vs 12.2 days in the prolonged IV group (difference –4.8, 95 % CI –6.1 to –3.4). Sensitivity analyses using overlap weights, high-dimensional propensity scores, and weight trimming produced concordant estimates, and doubly robust models confirmed the findings.

In this single-center VHA cohort, an early switch to oral antibiotics was associated with infection outcomes that were non-inferior to prolonged IV therapy and clinically meaningful reductions in antibiotic-related adverse events, catheter-related complications, C. difficile infections, and hospital length of stay. The benefits reported in OVIVA likely translate to routine practice at the VHA, supporting the adoption of an early oral step-down strategy.

All Authors: No reported disclosures

Figures