Is intravenous better than oral antibiotics for treating osteomyelitis?

A new high-quality randomised controlled trial (RCT) comparing the use of intravenous (IV) and oral antibiotics to treat bone and joint infections has just been published in the world’s most prestigious medical journal, The New England Journal of Medicine.

Put simply, this RCT found oral antibiotics had the same effect as IV antibiotics in resolving osteomyelitis by 12 months. It also found that oral antibiotics resulted in shorter hospital stays and fewer complications than IV. These findings published in this journal are a massive deal and could completely change diabetic foot osteomyelitis management, but as we always say is this the full story?

What do we know before this study?

Standard treatment for osteomyelitis typically involves surgical debridement of the infected bone followed be prolonged courses of IV antibiotics to eradicate any remaining infection. The antibiotics are considered necessary because of the potential for residual bone and soft tissue infection that leads to relatively high osteomyelitis recurrence rates of around 20% after 12 months.

For decades IV has been the first choice of antibiotics route for osteomyelitis as its thought to be superior to oral in bioavailability and bone penetration.  The down side is that IV is also thought to be more expensive, use more hospital resources, have more complications and less preferred by patients than oral. But, in recent years the theory that IV is better than oral has begun to be challenged.

The most recent Cochrane review found no differences between oral and IV antibiotics in recurrence rates for chronic osteomyelitis, but their findings were based on small older low quality trials. The last International diabetic foot infection guidelines recommended that surgical debridement should be considered followed by 6 weeks of antibiotics for osteomyelitis, and that either IV or oral can be selected as long as the antibiotic used has good bioavailability. But, they both recommended large high-quality trials are still needed to confirm these findings. And this is where this new RCT comes in.

What did this new study do then?

This was a large multi-centre RCT of patients that were within 7 days of their diagnosis for a bone or joint infection in 26 UK hospitals. Patients were randomized to receive either IV or oral antibiotics for the first 6 weeks of their treatment. They then followed all patients for one year to see if patients had “treatment failure” (typically recurrence of infection), any complications, their hospital usage, adherence to their antibiotic treatments and what their quality of life was like.

At this point we should say that patients could have had a bone or joint infection essentially anywhere in their body. So this was not specifically for foot osteomyelitis. Additionally, patients could have had IV antibiotics before they started in the study or up to 5 days of IV after for any other infections. So most patients had at least some short course of IV antibiotics in both groups. Lastly, the antibiotic chosen was at the discretion of the treating Infectious Diseases consultant and patients and their doctors knew which treatment they were getting (not blinded). But, any potential treatment failures were referred to an unbiased blinded expert committee to determine if they were actually treatment failures or not.

So what did they find?

They recruited 1,054 eligible patients (527 in each group). Both groups had the same baseline characteristics for a whole range of characteristics, except the oral group (61%) had more males than the IV group (68%). Of interest to the diabetic foot community was that only 20% had their infection in the foot, 20% had diabetes and 6% had peripheral arterial disease. So this wasn’t necessarily a diabetic foot osteomyelitis group, but was typical of populations getting bone and joint infections.

After one year they found no statistical differences in treatment failure between the groups (14.6% IV vs 13.2% oral). They also found no statistical differences after adjusting for a range of potential factors that may have biased this result, including looking at only those that completed all treatments, different centre’s results, different severity of treatment failures etc.

They also found that the IV group had longer hospital stays (14 vs 11 days), poorer quality of life, halted their treatment earlier due to difficulties or complications (18.9% IV vs 12.8% oral), had more IV catheter complications (9.4% vs 1.0%), similar other complications, but slightly better overall adherence to taking all prescribed antibiotics than the oral group (94% v 88%).

What was good or bad about this study?

The main strengths were: i) it was a large multi-centre study of patients with bone or joint infections; ii) they reported similar comprehensive baseline characteristics; iii) they thoroughly followed patients for 12 months; iv) treatment failure was diagnosed by an expert committee that did not know which treatment patients got; and v) they used robust statistical analyses.

Limitations were: i) patients and doctors knew which treatment they were getting (non-blinded); ii) the antibiotic agent wasn’t standardized as it was chosen by the treating Infection Diseases consultant; and iii) they recruited a wide range of people with different bone and joint infections. But, in saying all that these limitations could also be considered strengths as they are very representative of the typical populations who get bone and joint infections and their typical treatments.

So what does that all mean?

Well the authors sum it up nicely, “oral antibiotic therapy was noninferior to intravenous therapy when used during the first 6 weeks in the management of bone and joint infection, as assessed by treatment failure within 1 year. Oral antibiotic therapy was associated with a shorter length of hospital stay and with fewer complications than intravenous therapy.” They also go on to quite rightly say that, “our results thereby challenge a widely accepted standard of care”.

So does that answer the title’s question? Well not entirely but it does go a long way too. This trial is by far the largest and most rigorous trial to test IV against oral antibiotics for osteomyelitis anywhere. Therefore, it is by far the best evidence anywhere in the world to answer this question. However, it did test patients with a whole range of different osteomyelitis and not just diabetic foot osteomyelitis.

Finally, for diabetic foot fans, one of the lead authors of this study was Professor Ben Lipsky. Prof Lipsky is the global guru of diabetic foot infection and chair of the new International diabetic foot infection guidelines being launched in May. Thus, we suggest these findings will at the very least re-inforce the current guideline recommendation that either IV or oral antibiotics can be selected to treat diabetic foot osteomyelitis depending on the patient’s tolerances. This may mean more patients with diabetic foot osteomyelitis will have oral antibiotics in future, leave hospital earlier, have fewer complications and have a better quality of life. And such a situation can only lead us one step closer to our goal of ending avoidable amputations within a generation.

And after all that, if you don’t believe us why don’t you register at our DFA 2019 conference where we will have not one but two expert authors of the upcoming new International diabetic foot infection guidelines – Professor Larry Lavery and Dr Mathew Malone – presenting these new guidelines, how they treat diabetic foot infections and osteomyelitis now and into the future and much, much, much more.