Five trials, 410 patients, and MRSA that clears itself half the time
A Cochrane review of five trials found antibiotics may clear MRSA in cystic fibrosis, but the data barely support it.
By The Weekend · · 5 min read
Five trials. 410 people. Six months of treatment, at most. That is the entire evidence base behind how doctors decide whether to hit a "superbug" lung infection with antibiotics in people with cystic fibrosis — and at the end of it, one finding stands out: some patients cleared the infection with no treatment at all.
Methicillin-resistant Staphylococcus aureus, MRSA, is one of the bacteria that colonises the thick, hard-to-clear mucus in cystic fibrosis lungs. It is treated as a genuine threat because it is believed — not proven, believed — to cause more lung damage than ordinary staph infections. The evidence for how to actually get rid of it, reviewed by Cochrane, comes from five studies conducted in Italy and North America, recruiting patients aged 11 to 25 on average, running between eight weeks and six months. That is the entire dataset informing a treatment decision faced by clinics worldwide.
The comparison that should have been simple wasn't
Two of the five trials, covering 106 people, tested oral antibiotics — trimethoprim, sulfamethoxazole and rifampicin, or co-trimoxazole and rifampicin, combined with decontamination creams for skin and nose — against no treatment at all. One trial stopped early because the benefit looked clear enough that continuing seemed unethical. That is normally the strongest signal a trial can send.
But at six months, there was no difference between the groups in how many people still tested positive for MRSA. Some of the untreated patients had cleared the infection on their own. Quality of life, flare-up frequency, side effects, weight change — little to no difference across the board, with one exception: fewer people on antibiotics were hospitalized in the first six months in one study. That is the one number in the oral-antibiotic comparison that survived scrutiny. Everything else washed out.
Inhaled vancomycin: the drug that's supposed to be different
The second comparison, two studies covering 251 people, tested inhaled vancomycin against a placebo — and, in one study, a standard dose against a higher one. Inhaled antibiotics probably make no difference to lung function compared with placebo across both studies. One study did report a small lung-function gain after 20 weeks, but that is one result inside two studies that mostly agreed with each other in finding nothing.
Flare-up frequency: a standard dose of vancomycin probably makes no difference compared with placebo. The lower dose did buy a longer stretch before the next flare-up in one study. The higher dose knocked down MRSA bacteria levels for up to a month in another. Neither study found a difference in quality of life or side effects. A single, smaller study — 25 people — tested oral plus inhaled antibiotics against oral antibiotics plus a placebo inhaler, too small a sample to draw a real number from.
What the numbers say: mostly, "we don't know yet"
Strip away the trial jargon and the picture is this: across three separate comparisons — oral versus nothing, inhaled versus placebo, and combined versus oral alone — the confident findings are thin. The strongest one is the early-stopped oral-antibiotic trial, and even that produced no six-month difference in whether MRSA was still detectable. The honest summary is that early treatment appears possible, but nobody has shown that current routine treatment changes outcomes that matter to a patient living with cystic fibrosis: how often they end up in hospital, how their lungs function years later, how long they live.
This is not a funding-conflict story in the usual sense — there is no supplement company or manufacturer angling for a "wellness" claim here. But it is worth naming who runs the trials and who reviews them: the original studies were conducted by hospital and university research groups in Italy and North America, and the synthesis is Cochrane's own systematic review process, which explicitly grades its confidence in each finding rather than reporting a single verdict. That grading is the reason the review can say "probably no difference" in one place and "may clear the infection" in another, instead of collapsing everything into a single headline number.
Ten more days in hospital, or none — nobody has measured which
The gap in this evidence is not really about MRSA. It's about timing. All five trials predate the widespread use of CFTR modulators — the newer drugs that correct the faulty protein behind cystic fibrosis and let mucus move more normally through the lungs. Those drugs are changing what a cystic fibrosis lung actually looks like: thinner mucus, fewer flare-ups, different bacterial ecology. A trial run in 2010 on lungs full of thick, static mucus may simply not describe how MRSA behaves, or how well antibiotics work against it, in a patient taking a modulator drug today.
That is the mechanism worth sitting with: this isn't a case of bad science producing an ambiguous answer. It's good science answering a question that the underlying disease has since started to change. Cochrane's own conclusion says as much — research is needed to reassess MRSA treatment specifically in the modulator era, because the five trials that exist can't speak to it.
What changes next, and what would prove this wrong
Nothing about routine treatment protocols shifts immediately. Clinics will likely keep prescribing combinations of oral and inhaled antibiotics with decontamination, because the alternative — doing nothing and hoping the infection clears on its own, as it did for some patients in these trials — is not something most clinicians or parents will accept on a coin-flip of evidence. The cost of that caution is real: months of oral antibiotics, nebulized vancomycin routines, and decontamination regimens that show, so far, little measurable difference in the outcomes patients care about most.
What would settle this is a specific kind of trial: one run on patients currently taking CFTR modulators, tracking not just whether MRSA is detectable in a sputum sample but hospital admissions, lung function over years rather than months, and quality of life at a scale large enough to separate a real signal from the noise these five studies couldn't clear. Until that trial exists, the honest answer for a patient asking "will this antibiotic get rid of my MRSA" is: possibly, for a while, and nobody has yet shown it changes what happens to you next.