Oral antibiotics may be noninferior to intravenous therapy for fracture-related infections in the modified intent-to-treat analysis; however, noninferiority was not confirmed in the per-protocol analysis, introducing uncertainty regarding absolute equivalence.
Researchers from the Major Extremity Trauma Research Consortium conducted the POvIV randomized clinical trial to evaluate the efficacy of oral vs intravenous (IV) antibiotics in treating fracture-related infections (FRIs). In the study, published in JAMA Surgery, the researchers enrolled 233 patients aged 18 to 84 years who had undergone fracture fixation or arthrodesis with implants and subsequently developed an FRI without radiographic evidence of osteomyelitis. The participants were randomly assigned to receive either oral or IV antibiotics following surgical debridement and were followed-up for 12 months to assess the primary and secondary outcomes.
The primary outcome was the number of study injury–related surgical interventions within 1 year. In the unadjusted modified intent-to-treat (mITT) analysis, oral antibiotics met the noninferiority threshold (mean surgical interventions = 1.3 vs 1.1 for IV, upper bound of 95% confidence interval [CI] = 0.59 vs the 0.67 margin). However, PP analysis did not support noninferiority (upper bound = 0.71 vs 0.67 margin). A post hoc adjusted mITT analysis supported these findings. However, the prespecified adjusted per-protocol (PP) analysis did not confirm noninferiority, as the upper bound of the 95% CI exceeded the noninferiority threshold (0.71 vs the 0.67 margin).
The secondary outcome assessed the recurrence of deep surgical site infections. In the unadjusted mITT analysis, reinfection was observed in 30.4% of patients in the oral group and 32.2% in the IV group (0.44 vs 0.41 infections per person-year). The adjusted mITT analysis estimated reinfection rates at 34.8% (95% CI = 24.4%–46.0%) for oral therapy and 30.5% (95% CI = 21.7%–40.0%) for IV therapy. In the PP analysis, reinfection rates were 37.7% (95% CI = 26.9%–49.3%) for oral therapy and 29.8% (95% CI = 20.8%–39.1%) for IV therapy.
Among the 233 patients, 22.7% were female, and the mean age was 46.0 (13.9) years. The largest fracture site subgroup included tibia or fibula fractures (64.4%), followed by femur (6.9%) and radius or ulna fractures (6.9%). The mean duration of baseline antibiotic coverage was 39.3 (11.0) days in the oral group and 40.7 (18.4) days in the IV group. Notably, 41% of the patients in the oral group received linezolid, whereas 29% of those in the IV group received vancomycin.
Crossovers were observed in 6.1% of the patients assigned to oral therapy who received IV antibiotics and 10.2% of those in the IV group who transitioned to oral antibiotics. Those who crossed from IV to oral treatment exhibited higher rates of reoperations and reinfections. Additionally, crude rates of nonunion at 1 year were similar between the groups (4.3% for oral therapy vs 4.2% for IV therapy).
The findings indicated that oral antibiotic therapy was noninferior to IV therapy in the primary outcome based on mITT analyses. However, discrepancies in the PP analysis and secondary outcomes introduce uncertainty regarding absolute equivalence. Oral antibiotics demonstrated noninferiority in mITT analysis, but further research is needed to address the discrepancy between mITT and PP findings and clarify long-term efficacy.
Full disclosures can be found in the published study.