Polymyxin B / Colistin: evidence, FDA status and research

They are last-resort antibiotics for germs that resist other drugs. They were approved before modern trials, and the one big trial of 406 people showed adding another antibiotic did not help.

Sold for

Polymyxin B and colistin are the antibiotics of last resort for Gram-negative bacteria that have beaten carbapenems.

Studies say

The polymyxins kill multidrug-resistant Gram-negative bacteria and remain in use for lack of alternatives, but their approvals predate controlled trials and the one large RCT showed that adding meropenem does not improve outcomes.

Why grade B. The polymyxins were approved decades before randomized trials were required, so their efficacy rests on bacteriology and observational experience. The modern controlled evidence is mostly about how to use them: the AIDA trial randomized 406 patients with carbapenem-resistant Gram-negative infections to colistin alone or colistin plus meropenem and found no difference in clinical failure. Last-resort use is well established; what is missing is trial-grade proof of benefit against modern comparators. ¹²

FDA. Both are FDA-approved antibiotics. Injectable polymyxin B is approved for serious infections caused by susceptible Gram-negative bacteria, with a boxed warning about kidney toxicity, nerve toxicity and the risk of respiratory paralysis; colistimethate sodium for injection, first approved as Coly-Mycin M, is approved for acute or chronic infections due to susceptible Gram-negative bacilli, mainly Pseudomonas aeruginosa. Drugs@FDA lists the first polymyxin B injection approval in 1964 and Coly-Mycin M in 1970, and both drugs are now sold as generics.

Written from 3 cited sources. Fact-checked 2026-10-27: 12 claims checked, 0 corrected. Last refreshed 2026-10-27. Not physician reviewed.

What this means for you

What people take it for: Serious infections that other antibiotics cannot treat
Can a US doctor prescribe it?: Yes, FDA approved for serious infections from germs they can kill.
Properly tested in people?: Partly: some human trials, with gaps. The main trial had 406 people.

Evidence checklist

Randomised controlled trial AIDA randomised trial of colistin combinationsyes
Human study over 100 people AIDA, 406 patientsyes
More than one research group Article does not say who ran the trialno information
Clinical outcomes, not surrogate markers Clinical failure and mortality assessedyes
Human safety data beyond case reports Boxed warning; kidney and nerve toxicity knownyes

Drawn from the cited studies below. A dash means the sources give no information.

What it is

Polymyxin B and colistin (polymyxin E) are ring-shaped peptide drugs that kill bacteria. Researchers first found them from the soil bacterium Paenibacillus polymyxa in the late 1940s. They differ by a single amino acid. Doctors give polymyxin B as the active drug. Colistin is usually given as colistimethate sodium. This is an inactive prodrug, and the body converts it to colistin. Drugs@FDA lists the first polymyxin B injection approval in 1964 and Coly-Mycin M in 1970. Both drugs are now sold as generics. ³ Doctors largely dropped them for whole-body use in the 1970s. They caused kidney damage, and safer drugs had arrived. ²

They returned in the 2000s. Acinetobacter, Pseudomonas and Enterobacterales resistant to carbapenems were spreading. For many of those germs, the polymyxins were the only drugs still active. Polymyxin B is also in over-the-counter skin products such as Neosporin. Toxicity is not a problem there. The body does not absorb it. ¹²

How it works

Polymyxins are charged molecules with a fatty tail. The ring carries a positive charge. It binds to lipopolysaccharide, which carries a negative charge. This molecule forms the outer membrane of Gram-negative bacteria. The drug pushes out the calcium and magnesium ions that hold the membrane together. The fatty tail then pushes into the membrane. The membrane leaks, and the cell dies. The target is the outer membrane itself, not an enzyme. So resistance developed slowly. But plasmid-borne mcr genes now spread resistance between species. They change lipopolysaccharide. The same membrane damage in kidney tubule cells explains the main toxicity. ¹²

What human research shows

The largest randomized trial, AIDA, enrolled 406 patients in Israel, Greece and Italy. All had severe infections. The germs were Gram-negative bacteria resistant to carbapenems. Most were Acinetobacter baumannii. Patients got colistin alone or colistin plus meropenem. At 14 days, treatment had failed in 79 percent of the colistin group. In the combination group it was 73 percent. That difference was not significant. Deaths at 28 days were similar. The trial settled a long argument. Adding a carbapenem as routine does not help when the germ resists it. ¹

The approved uses in the labels rest on older case series and lab data. They do not rest on controlled trials. That was the standard at the time. ²

Study	Model	Finding
AIDA trial	Humans, n=406, randomized, open-label, colistin versus colistin plus meropenem	Adding meropenem did not reduce clinical failure or mortality in carbapenem-resistant Gram-negative infections. ¹
Polymyxin B label clinical basis	Humans, uncontrolled series and bacteriologic data from the 1950s and 1960s	Approval and indications rest on susceptibility data and case experience, not randomized trials. ²

Safety

Kidney injury is the main problem. The polymyxin B label carries a boxed warning for kidney and nerve toxicity. Doctors must dose the drug to kidney function. They must check kidney tests daily. Nerve effects start with tingling around the mouth and dizzy spells. Rarely, they reach paralysis of the breathing muscles from neuromuscular blockade. That is why the label warns against combining polymyxins with other drugs that harm nerves or kidneys. The AIDA trial used colistin. Failure and death rates were high. The drug played a part. But the patients were also very sick. ¹²

Questions for your clinician

Are there alternatives to polymyxins for my infection?
How often will kidney function be checked?
Does adding a second antibiotic help, given the AIDA result?

Questions people ask

What is the difference between polymyxin B and colistin?

They are near-identical ring-shaped peptides from the same bacterium. They differ by one amino acid. Polymyxin B is given as the active drug. It reaches steady blood levels quickly. Colistin is given as colistimethate, an inactive prodrug. It converts slowly and in uneven ways. That makes colistin harder to dose. The kidney clears colistimethate. So it is preferred for urinary infections. Polymyxin B is often preferred for blood infections. ¹²

Why are they called drugs of last resort?

They damage the kidneys. So doctors set them aside once aminoglycosides, cephalosporins and carbapenems were available. The drugs came back only when bacteria resistant to all of those appeared. Now doctors save them for infections where nothing else works. Newer drugs such as ceftazidime-avibactam and cefiderocol have since taken over some of that role. ¹²

Should colistin be combined with another antibiotic?

The AIDA trial was the largest randomized test of that question. Adding meropenem to colistin gave no benefit when the germs resisted carbapenems. Test-tube work had shown the two drugs kill germs better together. That did not carry over to patients. ¹

Is the polymyxin B in Neosporin dangerous?

No. Intact skin does not absorb topical polymyxin B in useful amounts. So the kidney and nerve toxicity that limit the injectable form do not apply. ²

Citations

Paul 2018, Lancet Infect Dis: colistin alone versus colistin plus meropenem for carbapenem-resistant Gram-negative infections (AIDA) doi.org
Polymyxin B for injection prescribing information drugs.com
openFDA Drugs@FDA: Coly-Mycin M (NDA 050108) approval record api.fda.gov