What prevents antibiotic resistance? Antibiotics
SuDDICU trial paves the way for antibiotic decontamination in ICUs
Patients receiving mechanical ventilation are particularly vulnerable to hospital-acquired infections, especially by aerobic gram-negative bacteria and yeasts. These pathogens incubate in the stomach and gain resistance after exposure to intermittent antibiotics.
They may then reflux into the respiratory tract to produce pneumonias or travel further into the gut and translocate into the bloodstream. Such infections from resistant bacteria or yeast are increasingly frequent killers, especially of long-stay patients in ICUs and LTACs.
It has been known for years that many of these infections can be prevented by treating mechanically ventilated patients with prophylactic antibiotics. This is done either by “selectively decontaminating” the GI tract with nonabsorbable antibiotics, infusing broad-spectrum antibiotics intravenously, or both.
Dozens of randomized trials and meta-analyses dating back decades have shown that decontamination with early broad antibiotics can prevent some hospital-acquired infections in ICU patients, and seems to reduce mortality as well.
The practice has never caught on with clinicians or been endorsed by professional societies, however, because of the reasonable concern that it could lead to widespread antibiotic resistance.
An intriguing new randomized trial may not eliminate those fears, but could open the door for antibiotic decontamination in the ICU at a wider scale.
The SuDDICU Trial
Investigators from the Australia and New Zealand Intensive Care Society Clinical Trials Group (ANZICS) and the Canadian Critical Care Trials Group teamed up to run this large, multiphase, international, multicenter randomized trial.
Mechanically ventilated patients randomized to the intervention arm received a topical oral antibiotic paste smeared into the mouth every 6 hours, in addition to an antibiotic suspension administered to the GI tract via a gastric tube every 6 hours. The paste contained 10 mg of colistin, 10 mg of tobramycin, and 125,000 IU of nystatin, while the suspension held 100 mg of colistin, 80 mg of tobramycin, and 2x106 IU of nystatin. The paste was administered for as long as they were mechanically ventilated, up to 90 days.
They also received broad-spectrum IV antibiotics for four days (e.g., ceftriaxone or ciprofloxacin).
Control patients got usual care without the extra antibiotics.
In the first phase, reported in JAMA in 2022, 19 ICUs in Australia crossed over at random intervals to provide the intervention to 5,982 patients. Mortality was 2% lower in the antibiotic arm (27% vs 29% in the controls), just missing statistical significance.
Phase two took place at seven ICUs in Canada, enrolling an additional 3,307 patients, with the data for all 9,289 reported out in NEJM 2025. For the whole population, at 90 days of follow-up:
Mortality was numerically lower in the antibiotic arm (27.9% vs. 29.5%), with an odds ratio for death of 0.93 (95% confidence interval 0.84 to 1.05; P=0.27);
There were ~2% fewer bacteremias in the treated group (4.9% vs 6.8%), a number needed to treat of 53 to prevent one bloodstream infection (significant);
Patients in both arms got roughly the same total doses of antibiotics;
The regularly provided antibiotics prevented resistance, though: fewer patients in the antibiotic arm had resistant organisms recovered (~17% vs ~27%). They had numerically fewer C. difficile infections as well.
Adverse events occurred rarely in both groups, with no severe drug reactions reported.
Antibiotics Reduce Antibiotic Resistance?
Patients in both arms received roughly equal doses of antibiotics in total. But continuous suppression of bacteria in the gut in the intervention patients likely led to fewer clinically evident or suspected infections, leading to fewer courses of systemic antibiotics and less selective pressure toward antibiotic resistance. This produced fewer bloodstream infections and (probably) fewer deaths.
Although both phases of SuDDICU were negative, the consistent signal of numerically lower mortality in 32 randomized trials with over 27,000 participants is persuasive, especially with the plausible mechanism (the statistically significant reduction in bloodstream infections).
A 2% absolute difference in mortality is hard to prove: at these incidences (mortality ~28%), it would require enrolling about 16,000 patients, according to ClinCalc. The negative finding in SuDDICU may be a type II error due to underpowering, which does not diminish the achievement of conducting a ~10,000 subject trial across two continents over 10 years.
Time to Mix the Paste?
A key finding of SuDDICU was that increased antibiotic resistance did not emerge from the digestive decontamination protocol at the participating centers.
A 2% mortality reduction (albeit unproven in SuDDICU, or anywhere thus far) would be a big deal, but the reductions in bacteremias and antibiotic resistance are worthwhile objectives in their own right. Reassured by the ecological data, many centers will likely adopt some form of selective digestive decontamination based on the accumulated evidence of benefit. This might be applied to all ventilated patients, or those considered at higher risk (e.g., comatose or elderly).
As the authors point out, antibiotic resistance patterns vary widely by institution, and the participating centers in SuDDICU did not have high rates of resistant bacteria. At a minimum, implementing a new regime of regular antibiotics for all ventilated patients should include vigilance for emerging ecological changes by an established infection control program.
I think you have missed the point! The paste does nothing ....it is the 4 days of intravenous antibiotics that is the effective part of SDD! And remember, apparently the George Institute has some sort of patent over the paste, so I imagine there is an enormous conflict of interest in the publication!