Guest posting: Dr Patrick Harris, our roving correspondent at the European Congress on Clinical Microbiology and Infectious Diseases (ECCMID) Conference in Vienna.
At last, an RCT tells us the answer…
For many years the evidence-base for things we do routinely in infectious disease has been somewhat sketchy. We are often forced to formulate our recommendations in the absence of high-level evidence, such as large randomised clinical trials. Frequently our evidence is derived from a handful of observational studies (often poorly designed or underpowered), expert opinion or extrapolation from laboratory studies. As such, there can be great variation in clinical practice, with everyone defending their own pet theories!
For instance, here is a sobering thought. Despite S. aureus bacteraemia being one of the most common clinical conditions we see in infectious disease, in the last 50 years fewer than 1500 patients with this condition have been enrolled in a clinical trial to specifically study treatment options for this disease entity. It presents a huge disease burden and is associated with high mortality, but is almost never the subject of our most powerful research tool: the double-blind randomised trial.
I have just returned from the international ECCMID meeting in Vienna, where I was struck by efforts that are ongoing to fill these gaps in our knowledge. The results from a couple of fantastic examples of large, well-designed, pragmatic randomised trials to answer ‘real-world’ questions were presented at the conference. Importantly these were designed by and for clinicians, without reliance on industry funding.
One of these was the ARREST Trial. Coming from a large consortium of clinical investigators in the UK, they attempted to answer a common clinical question: in a patient with Staphylococcus aureus bloodstream infection, does adding rifampicin improve patient outcomes? There have been several reasons to think that this might be the case. Rifampicin has activity against bacteria in biofilms, may kill bacteria located within white cells and may help clear the organism from the blood more rapidly and reduce the risk of metastatic infection. So adding to standard antibiotic treatment may help. More antibiotics are better than one, right? Well, the answer in this case is a firm ‘no’.
They randomised more than 750 patients with S. aureus bacteraemia, to receive standard drug therapy such as flucloxacillin (for MSSA) or vancomycin (for MRSA) plus placebo, compared with standard treatment in combination with rifampicin, for at least 2 weeks. The take home message was that it made no difference in outcome, including hard endpoints such as mortality or rapidity of bacteraemia clearance. Furthermore, adding rifampicin was associated with more drug interactions and non-severe side effects. There was a small reduction in the rate of relapsed infection with rifampicin, but this effect was not significant.
It is really encouraging to see these large multicentre RCTs, funded by academia or other public bodies, to help optimise outcomes for patients with common but serious infections. Although it was a ‘negative trial’, in the sense that the intervention did not show benefit, this information is extremely helpful to allow us to stop using ineffective interventions and seek out and test alternative strategies.