By Richard R. Watkins, MD, MS, FACP, FIDSA

Professor of Internal Medicine, Northeast Ohio Medical University; Division of Infectious Diseases, Cleveland Clinic Akron General, Akron, OH

Dr. Watkins reports no financial relationships relevant to this field of study.

SYNOPSIS: In a randomized, controlled trial of adult patients with bone or joint infections, researchers found oral antibiotic therapy was noninferior to intravenous therapy based on treatment failure at one year.

SOURCE: Li HK, Rombach I, Zambellas R, et al. Oral versus intravenous antibiotics for bone and joint infection. N Engl J Med 2019;380:425-436.

The standard management of bone and joint infections for more than 40 years has been surgical debridement followed by a prolonged (i.e., four to six weeks) course of intravenous (IV) antibiotics. However, this paradigm has become increasingly questioned because of recent evidence showing similar efficacy between oral and IV therapy. Li and colleagues sought to clarify outcomes at one year between IV and oral antibiotics administered during the initial six weeks of treatment for bone and joint infections.

The Oral Versus Intravenous Antibiotics for Bone and Joint Infections (OVIVA) trial was conducted at 26 sites in the United Kingdom. Patients 18 years of age and older were enrolled if they had one of the following conditions: native osteomyelitis of the extra-axial skeleton, native joint infection requiring excision arthroplasty, prosthetic joint infection (PJI), orthopedic fixation device infection, or vertebral osteomyelitis with or without associated discitis or soft-tissue infection. The patients were randomized 1:1 within seven days of surgery or the start of antibiotics to either six weeks of IV or oral therapy. The choice of antibiotics was left to the judgment of the infectious disease physician, and adjunctive oral therapy (e.g., rifampin) was permitted for patients in the IV group. Therapy beyond six weeks was permitted for both groups.

The primary endpoint was treatment failure within one year, which was defined as the presence of at least one clinical criterion (draining sinus tract from the bone or prosthesis or the appearance of frank pus adjacent to the bone or prosthesis), microbiological criterion (phenotypically indistinguishable bacteria isolated from two or more deep-tissue samples or a pathogenic organism from a single aspirate or biopsy), or histologic criterion (the presence of characteristic inflammatory infiltrates or microorganisms). Secondary endpoints included probable or possible treatment failure, early discontinuation of the treatment strategy, IV catheter complications, Clostridioides difficile infection, serious adverse events, resource use, health status, the Oxford hip and knee scores, and adherence to treatment.

The modified intention-to-treat (ITT) analysis included 1,015 patients, of whom 639 (61%) had hardware-related infection and 80 (7.6%) were treated without surgical intervention. The identified pathogens were Staphylococcus aureus (38%), coagulase-negative Staphylococcus (27%), Streptococcus species (15%), Pseudomonas (5%), Gram-negatives besides Pseudomonas (16%), and culture negative (16%). Ten percent of the patients had methicillin-resistant S. aureus (MRSA). The most commonly prescribed IV antibiotics were glycopeptides (41%) and cephalosporins (33%), while quinolones (44%) and combination therapy excluding rifampin (14%) were the most common oral agents.

Therapy was continued past six weeks for 805 of 1,049 participants (76.7%). The median total duration of therapy was 78 days (interquartile range, 42 to 99 days) in the IV group and 71 days (interquartile range, 43 to 94 days) in the oral group (P = 0.63). Treatment failure based on clinical, microbiological, or histological criteria occurred in 74 of 506 (14.6%) in the IV group and 67 of 509 (13.2%) in the oral group. Furthermore, the ITT analysis showed a difference in the risk of definitive treatment failure between the oral group and the IV group of -1.4 percentage points (90% confidence interval [CI], -4.9 to 2.2; 95% CI, -5.6 to 2.9), which indicated noninferiority. A post-hoc subgroup analysis did not reveal any advantages of IV therapy over oral therapy.

For the secondary endpoints, there were no significant differences between oral and IV therapy in terms of C. difficile infection rates or the percentage of patients reporting at least one serious adverse event. More patients discontinued therapy earlier in the IV group than in the oral group (18.9% vs. 12.8%, respectively; P = 0.006). Not surprisingly, catheter complications were more common in the IV group (9%) than in the oral group (1%), with the latter reflecting the fact that 10% of those in the oral group received IV therapy at any time during the study. The median hospital stay was significantly greater in the IV group (14 days [interquartile range, 11 to 21]) compared to the oral group (11 days [interquartile range, 8 to 20], P < 0.001). There was no significant difference between the groups in Oxford hip scores, but better Oxford knee scores were observed in the oral group compared to the IV group at days 120 and 365 (P = 0.01 and P = 0.04, respectively). Finally, the use of rifampin did not lead to a significant difference in outcomes (P = 0.22).


The results of the study by Li and colleagues challenge the practice of administering six weeks of IV antibiotics to treat bone and joint infections, which is widely accepted as the standard of care. Although this is not the first study to show the effectiveness of oral antibiotics for these indications, it was well designed and directly compared the two regimens in the same settings and in the same patient population. Indeed, not only was oral therapy noninferior to IV therapy, it also was associated with shorter hospital stays and fewer complications, although the latter was due mostly to adverse events related to catheters. These findings likely will shift the treatment paradigm for orthopedic infections away from prolonged IV therapy.

Another interesting finding was the lack of benefit found from adding rifampin, which many experts view as an important adjunctive agent when treating biofilm. One explanation might be related to the timing, such that in many cases, rifampin was not used during the entire course of therapy.

The study had some limitations. First, there were few cases of MRSA, a highly virulent organism that often does not respond as well as other pathogens in orthopedic infections. Thus, oral antibiotics might be riskier for MRSA, and IV antibiotics (e.g., vancomycin, daptomycin, or ceftaroline) likely should continue to be the recommended therapy. Second, the treatment regimens were heterogeneous, as were the surgical interventions, which does not allow for comparisons of one treatment to another. Finally, the open-label design may have introduced selection bias, although it would have been unethical to give IV placebo.

Has the time come for the use of oral antibiotics to treat bone and joint infections? The answer is probably yes, with some exceptions, such as patients with poor gastrointestinal absorption or those with antibiotic-resistant pathogens, like MRSA. Questions remain about optimal therapy regimens and surgical management. It also seems prudent to recommend close outpatient monitoring with oral therapy and a switch to IV therapy if clinical improvement is suboptimal.