Oral or IV Antibiotics for Bone and Joint Infection
A patient with a bone or joint infection may spend 6 weeks on IV antibiotics through a central line, with the line complications and readmissions that come with it, when pills may cure the infection as well.
Key terminology
- Noninferiority margin
- This preset limit marks how much worse a new treatment may do and still count as no worse. A result counts as noninferior when the whole confidence interval stays inside it.
- PICC line
- This long IV catheter enters an arm vein and ends in a large central vein. It lets patients receive weeks of IV antibiotics at home.
- Propensity score matching
- This method pairs treated and untreated patients who had similar odds of getting the treatment. It balances measured factors in an observational study and leaves unmeasured ones alone.
- Confounding by indication
- This bias appears when the reason a patient got a treatment also affects the outcome. Sicker patients are often steered to the stronger option, which hides its true effect.
- Bioavailability
- This term describes the share of an oral dose that reaches the bloodstream. A drug with high bioavailability gives blood levels close to an IV dose.
- Debridement
- This surgery removes dead, infected or foreign tissue, usually as part of an I&D in the OR. Antibiotics reach dead bone poorly, so it is cut out.
- Open-label trial
- In this design both patients and doctors know which treatment was given. It can bias judgment calls such as when to label a treatment a failure.
Takeaways
- Oral antibiotics were noninferior to IV antibiotics for adult bone and joint infection. OVIVA randomized 1054 adults at 26 UK centers to oral or IV therapy for the first 6 weeks, starting within 7 days of surgery. Treatment failure at 1 year was 13.2% with oral and 14.6% with IV. The difference of -1.4 percentage points sat inside the 7.5-point noninferiority margin. (Li et al, N Engl J Med 2019;380:425-436, PMID 30699315).
- Weeks of IV therapy add line complications without adding cure. Catheter complications occurred in 9.4% of the IV group and 1.0% of the oral group. Serious adverse events were similar (27.7% vs 26.2%). Median hospital stay was 14 days with IV and 11 days with oral. (Li et al, N Engl J Med 2019;380:425-436, PMID 30699315).
- Children with osteomyelitis did as well on oral antibiotics at discharge as on a PICC line. A cohort of 2060 children at 36 hospitals used propensity score matching and found no more failures with oral therapy. 15.0% of the PICC group had a line complication needing an ED visit or readmission. Sicker children may still have gone home with a PICC (confounding by indication), and PICC use ranged from 0% to 100% by hospital. (Keren et al, JAMA Pediatr 2015;169:120-128, PMID 25506733).
- Childhood osteomyelitis can move to oral antibiotics after a few days of IV therapy. In 131 Finnish children with culture-positive osteomyelitis, clindamycin or a first-generation cephalosporin ran IV for 2-4 days, then orally in large doses (high bioavailability). Staphylococcus aureus caused 89% of cases, all methicillin susceptible. Courses of 20 and 30 days did equally well (1 mild sequela per group), stopping once signs subsided and CRP fell to ≤20 mg/L. (Peltola et al, Pediatr Infect Dis J 2010;29:1123-1128, PMID 20842069).
- Childhood septic arthritis follows the same short IV course and early oral switch. In 130 Finnish children, IV therapy ran 2-4 days and the total course was 10 or 30 days. All recovered without sequelae, and surgery beyond joint aspiration was done in 12%. The switch depended on a good clinical response and a falling CRP. The trial tested duration and does not show a septic hip can skip surgical drainage. (Peltola et al, Clin Infect Dis 2009;48:1201-1210, PMID 19323633).
- Antibiotic route is chosen after the pus and dead bone have been dealt with. In OVIVA only 7.6% were managed without surgery, and 60.6% had infected hardware. Every trial in a Cochrane review of chronic osteomyelitis described debridement before antibiotics. Oral and IV routes gave similar remission at the end of treatment (RR 1.04) and at ≥12 months (RR 0.94). (Li et al, N Engl J Med 2019;380:425-436, PMID 30699315; Conterno and Turchi, Cochrane Database Syst Rev 2013;9:CD004439, PMID 24014191).
- Oral therapy needs susceptible bacteria and a gut that absorbs the drug. The OVIVA authors noted oral therapy may not suit patients with poor enteral absorption or pathogens resistant to oral agents. The Cochrane review found no difference by route only when the bacteria were susceptible. It rated that evidence low quality, and most trials were >20 years old. (Li et al, N Engl J Med 2019;380:425-436, PMID 30699315; Conterno and Turchi, Cochrane Database Syst Rev 2013;9:CD004439, PMID 24014191).
- The adult case for oral therapy rests mostly on one large open-label trial. OVIVA could not blind patients, followed them for only 1 year, and widened its margin from 5 to 7.5 percentage points after an interim analysis (the result met either margin). In low-risk S. aureus bacteremia, a smaller trial of 213 patients found an oral switch after 5-7 days of IV noninferior, with complicated bacteremia excluded. (Li et al, N Engl J Med 2019;380:425-436, PMID 30699315; Kaasch et al, Lancet Infect Dis 2024;24:523-534, PMID 38244557).