Polymyxin B
polymyxin B sulfate, PMB
Polymyxin B is a cationic cyclic lipopeptide antibiotic isolated from Paenibacillus polymyxa in the late 1940s. It kills Gram-negative bacteria by tearing open their outer membrane. Largely abandoned in the 1970s because of kidney and nerve toxicity, it returned to use as a last-line option against carbapenem-resistant organisms when nothing else works.
Mechanism
Polymyxin B carries five positively charged diaminobutyric acid residues. These engage electrostatically with the negatively charged phosphate groups of lipid A, the membrane anchor of Gram-negative lipopolysaccharide, competitively displacing the divalent calcium and magnesium ions that normally bridge and stabilise adjacent lipopolysaccharide molecules. The fatty acyl tail and the hydrophobic D-Phe6-Leu7 segment then insert into the lipid bilayer. The result is a disordered, permeabilised outer membrane, described in recent single-molecule work as long-lived super-stoichiometric clustering driven by many individually weak lipid A contacts. Once the outer membrane is breached, polymyxin B reaches and disrupts the inner membrane, and cell contents leak out.
Because the primary target is a structural lipid rather than a protein, polymyxin B also binds and neutralises free lipopolysaccharide, which is the rationale behind polymyxin B-immobilised haemoperfusion cartridges used in some countries for endotoxic shock. Resistance arises when bacteria modify lipid A, most often by adding 4-amino-4-deoxy-L-arabinose or phosphoethanolamine to reduce its negative charge. These modifications are usually chromosomally regulated through the PmrAB and PhoPQ systems, but the plasmid-borne mcr genes encoding phosphoethanolamine transferases have made this resistance transmissible between bacteria, which is why polymyxin stewardship matters.
What the research shows
There is no modern randomised controlled trial of polymyxin B as an antibacterial treatment. The clinical literature is dominated by observational cohorts in patients with carbapenem-resistant Acinetobacter baumannii, Pseudomonas aeruginosa and Enterobacterales, in whom crude mortality is high because the underlying illness is severe. Comparative reviews of polymyxin B against colistin consistently note that polymyxin B reaches therapeutic plasma concentrations faster and more predictably because it is administered as the active drug, whereas colistimethate must first hydrolyse to colistin. That pharmacokinetic argument is the main reason some centres prefer polymyxin B for bloodstream infection, while colistin is preferred for urinary tract infection because polymyxin B achieves poor urinary concentrations.
Nephrotoxicity is the dominant clinical constraint. Reported acute kidney injury rates vary widely with the definition used and the population studied, from roughly one patient in five in some series to more than half in cohorts applying RIFLE or AKIN criteria; the draft figure of 20-25% has been corrected because it represents the low end of a much wider range rather than a consensus estimate. Neurotoxicity, usually perioral and peripheral paraesthesiae, is much less common. Mechanistic work has clarified why the kidney is the target organ: after glomerular filtration, proximal tubular cells reabsorb filtered polymyxin from the apical surface by a saturable, carrier-mediated route in which megalin has been implicated, so the drug concentrates in the renal cortex and triggers time- and concentration-dependent apoptosis.
Polymyxin B haemoperfusion, in which the peptide is immobilised on a cartridge to adsorb circulating endotoxin, is a separate use and should not be read as evidence for the antibiotic. It is licensed in some countries, but the largest randomised trial in septic shock, EUPHRATES, did not demonstrate a mortality benefit in its overall population, and the technique remains contested.
Evidence assessment
Limited evidence
Downgraded from strong on audit. Every citation that survived verification is either a narrative or structure-activity review, a laboratory mechanism study, or a randomised trial of a haemoperfusion device rather than of the antibiotic. There is no randomised controlled trial of systemic polymyxin B for any infection in this record, and none exists in the wider literature. The United States approval is real but is a legacy registration predating modern efficacy requirements, so treating it as evidence of demonstrated benefit would misrepresent the position; that is why the labelling criterion has not been applied mechanically here. The modern clinical evidence is observational cohorts and pharmacokinetic-pharmacodynamic modelling. Polymyxin B is used because the alternatives for some pan-resistant organisms are worse or absent, not because it has been shown in a trial to be better than anything.
Tiers are applied consistently across the library and re-checked when new trials read out. Read the grading method.
Key studies
Polymyxin B versus colistin: an update Preclinical only
Polymyxin B and colistin share mechanism, spectrum and toxicity but differ importantly in formulation and pharmacokinetics. Polymyxin B is given as the active drug and achieves target plasma exposure more reliably; colistimethate is an inactive prodrug with slow, variable conversion. Polymyxin B achieves poor urinary concentrations and is not preferred for urinary tract infection.
Uptake of polymyxin B into renal cells Preclinical only
Uptake of polymyxin B into proximal tubular epithelial cells was saturable and occurred primarily across the apical membrane, indicating carrier-mediated rather than passive accumulation, with megalin proposed as the mediating receptor. This provides a mechanistic basis for the drug's marked accumulation in the renal cortex and its nephrotoxicity.
Potent activity of polymyxin B is associated with long-lived super-stoichiometric accumulation mediated by weak-affinity binding to lipid A Preclinical only
Polymyxin B binds individual lipid A molecules only weakly, but accumulates in long-lived super-stoichiometric clusters on the outer membrane. This revises the simple one-to-one binding model and helps explain the steep concentration dependence of killing.
Structure-activity relationships of polymyxin antibiotics Preclinical only
Maps which structural features of the polymyxin decapeptide drive antibacterial potency and which drive renal accumulation, establishing the diaminobutyric acid charge cluster as essential for lipid A engagement while the fatty acyl tail governs membrane insertion and contributes substantially to toxicity.
Effect of Targeted Polymyxin B Hemoperfusion on 28-Day Mortality in Patients With Septic Shock and Elevated Endotoxin Level: The EUPHRATES Randomized Clinical Trial Preclinical only
Negative for the primary outcome. Polymyxin B haemoperfusion did not significantly reduce 28-day mortality in the overall trial population, though post-hoc subgroup analyses generated continuing debate.
Safety
Dose-limiting nephrotoxicity is the principal risk, typically appearing within the first week and often reversible if the drug is stopped. Neurotoxicity presents as perioral numbness, paraesthesiae, dizziness, ataxia and visual disturbance. Neuromuscular blockade with respiratory arrest is a rare but serious hazard, particularly with concurrent anaesthetics, aminoglycosides or in myasthenia gravis. Intrathecal administration, used occasionally for meningitis, can cause meningeal irritation. Hypersensitivity reactions and, with topical use, contact sensitisation occur. Concurrent nephrotoxins should be avoided where possible and renal function monitored closely.
Regulatory status
| Jurisdiction | Status |
|---|---|
| United Kingdom | Systemic polymyxin B is not a routinely licensed injectable product in the UK; where systemic polymyxin therapy is needed, colistimethate sodium is the usual choice, and any polymyxin B use would generally involve unlicensed or imported product on a named-patient basis. Polymyxin B does appear in some licensed topical, ear and eye combination preparations. It should be used only under specialist infection advice. |
| United States | An approved prescription medicine of long standing, marketed as polymyxin B for injection and as a component of numerous topical, otic and ophthalmic combination preparations. The systemic product carries warnings for nephrotoxicity, neurotoxicity and neuromuscular blockade, and its labelling reflects a pre-modern registration rather than a contemporary efficacy dossier. |
| WADA (sport) | Not prohibited. Polymyxin B does not appear on the WADA Prohibited List in or out of competition. |
Questions
It was largely shelved in the 1970s once safer options such as aminoglycosides and later carbapenems became available, because of its reputation for kidney and nerve toxicity. The spread of carbapenem-resistant Acinetobacter, Pseudomonas and Enterobacterales from the 2000s onwards left clinicians facing organisms with essentially no other reliably active agent. Polymyxin B returned by default rather than by merit, and much of the modern work on it has been an effort to characterise properly a drug that was registered before such characterisation was required.
Chemically they differ at a single position in the ring: polymyxin B has D-phenylalanine at residue 6, colistin has D-leucine. Practically the difference is bigger. Polymyxin B is administered as the active molecule, so it reaches effective plasma concentrations quickly and predictably. Colistin is administered as colistimethate, an inactive sulfomethylated prodrug that must hydrolyse in the body, which makes early exposure lower and more variable. Conversely colistimethate is filtered and converted in the urinary tract, so colistin is the better choice for urinary infection while polymyxin B is generally preferred for bloodstream infection.
Often, but not always. Reported rates of acute kidney injury vary a great deal with the definition applied, from around a fifth of treated patients in some series to more than half in cohorts using formal RIFLE or AKIN criteria. It usually appears in the first week and frequently improves after the drug is withdrawn. The injury reflects accumulation of the peptide in proximal tubular cells by a saturable, carrier-mediated uptake route, followed by apoptosis. Recovery is less complete in patients with pre-existing renal impairment, in older patients and where other nephrotoxic drugs are given at the same time. This is a decision for the treating infection specialist, not something to self-assess.