The polyene antifungals exploit a fundamental biochemical difference between fungal and mammalian cell membranes: fungi synthesize ergosterol as their primary membrane sterol, whereas mammalian cells rely on cholesterol. This structural distinction is the pharmacological basis for selective antifungal toxicity and, paradoxically, for the dose-limiting toxicity that defines clinical use of this drug class.
Ergosterol is a 28-carbon sterol that serves the same structural and functional role in fungal cell membranes that cholesterol serves in mammalian cells: it modulates membrane fluidity, regulates the activity of membrane-bound enzymes, and is essential for cell wall synthesis and integrity. The biosynthetic pathway to ergosterol diverges from the mammalian cholesterol pathway at the level of lanosterol, with fungal cells employing a series of enzymes including lanosterol 14-alpha-demethylase (encoded by ERG11), delta-8 to delta-7 isomerase, and squalene epoxidase to arrive at the final ergosterol product. This pathway is also the target of the azole and allylamine antifungal classes, but the polyenes act at a completely different step: they do not inhibit ergosterol synthesis but instead bind directly to ergosterol already incorporated into the fungal cell membrane.1
Amphotericin B (AmB) is a macrolide polyene antibiotic produced by Streptomyces nodosus, first isolated in 1955. Its molecular structure features a large macrolide lactone ring with a hydrophilic polyhydroxyl chain on one face and a rigid, hydrophobic polyene chain (seven conjugated double bonds in AmB) on the opposite face. This amphipathic architecture is directly responsible for membrane interaction. When AmB encounters a membrane containing ergosterol, the hydrophobic polyene face inserts into the lipid bilayer and the drug molecules self-assemble around ergosterol molecules into barrel-shaped oligomeric pore structures. These transmembrane pores are approximately 0.4 to 0.8 nanometers in diameter and allow passive, non-selective flux of monovalent cations (potassium, sodium, hydrogen) and other small molecules across the membrane.2
The resulting loss of membrane electrochemical gradient has several consequences. Potassium leaks out of the fungal cell down its concentration gradient, leading to rapid cellular depolarization. The loss of membrane integrity also disrupts the proton motive force required to drive active transport systems that import nutrients and export metabolic waste products. At concentrations achieved at the standard clinical doses, AmB is fungicidal against susceptible yeasts and fungistatic against most molds. The distinction matters clinically: candidemia and cryptococcal meningitis, conditions requiring fungicidal activity, are settings where the superior killing kinetics of AmB compared to azoles have direct therapeutic implications. An oxidative damage hypothesis, in which AmB generates reactive oxygen species through its interaction with membrane sterols, has been proposed as a secondary mechanism contributing to cell death but remains less well established than the primary pore-forming mechanism.1,2
Selectivity and Its Limits. AmB binds ergosterol with approximately 10-fold higher affinity than it binds cholesterol, which is the structural basis for selective fungal toxicity. However, this selectivity is relative, not absolute. At therapeutically necessary concentrations, AmB does interact with cholesterol in mammalian cell membranes, and this interaction underlies the nephrotoxicity, infusion reactions, and electrolyte disturbances that dominate the toxicity profile of conventional amphotericin B deoxycholate (AmBd). The renal tubular epithelium is particularly vulnerable because it is exposed to high drug concentrations from the tubular lumen and because renal tubular cells express cholesterol-rich apical membranes. The degree of nephrotoxicity is concentration-dependent: the higher the free plasma drug concentration, the greater the interaction with renal tubular cholesterol. This observation is the pharmacological rationale for lipid-based delivery systems, which were developed specifically to reduce the free plasma concentration of AmB while maintaining delivery to fungal infection sites.3
AmB binds ergosterol in the fungal cell membrane and self-assembles into transmembrane pores, causing non-selective ion flux, loss of membrane potential, and cell death. The drug is fungicidal against susceptible Candida species and Cryptococcus neoformans, and fungistatic against Aspergillus and most molds. Selectivity for fungal over mammalian membranes is real but incomplete, accounting for both the clinical utility and the inherent toxicity of the polyene class.
Amphotericin B deoxycholate, the original formulation of the drug, uses sodium deoxycholate as a micellar solubilizing agent to create a colloidal dispersion suitable for intravenous administration. Understanding its absorption, distribution, metabolism, and excretion (ADME) properties is essential for interpreting its clinical dosing, monitoring requirements, and the rationale for its eventual replacement by lipid formulations in most indications.
Absorption and Administration. AmBd is not absorbed from the gastrointestinal (GI) tract and must be administered intravenously for systemic infections. The drug is formulated at a concentration of 0.1 mg/mL in 5% dextrose and infused over two to four hours. The standard dose for most systemic fungal infections is 0.5 to 1.0 mg/kg per day, with the upper end of this range (1.0 to 1.5 mg/kg per day) reserved for life-threatening infections such as cryptococcal meningitis induction therapy and invasive mold infections. A test dose of 1 mg infused over 20 to 30 minutes was historically recommended prior to the first full infusion to assess for severe infusion reactions, although this practice has fallen out of favor at many centers given that premedication strategies effectively attenuate most acute reactions and the test dose delays initiation of therapy in seriously ill patients.4
Distribution. AmBd is highly protein-bound in plasma (greater than 90%), primarily to lipoproteins including low-density lipoprotein (LDL) and high-density lipoprotein (HDL). The volume of distribution is large (approximately 4 L/kg), indicating extensive tissue penetration. AmB distributes well into the liver, spleen, lungs, kidneys, and adrenal glands. Penetration into the central nervous system (CNS) is poor from the systemic circulation, with cerebrospinal fluid (CSF) concentrations typically less than 4% of plasma concentrations following intravenous (IV) administration of AmBd. Despite this limited CSF penetration, AmBd was the standard treatment for cryptococcal meningitis for decades and retains efficacy in that setting, likely because high drug concentrations are achieved in the choroid plexus and meninges themselves. Intrathecal or intraventricular administration of AmBd has been used for CNS infections refractory to systemic therapy but is associated with severe local neurotoxicity and is rarely employed today.4,5
Metabolism and Elimination. AmB is not significantly metabolized by cytochrome P450 (CYP) enzymes, which means it has no clinically significant pharmacokinetic drug-drug interactions mediated through CYP inhibition or induction. This is a pharmacokinetic advantage over the azole antifungals, which are potent CYP inhibitors. Elimination is slow and complex: approximately 40% of a dose is excreted unchanged in the urine over several days to weeks, and the drug can be detected in tissues for weeks to months after discontinuation. The terminal elimination half-life is approximately 15 days, reflecting slow release from deep tissue compartments. Biliary excretion accounts for a significant fraction of non-renal elimination. Dose adjustment is not required for renal impairment because drug elimination does not depend on glomerular filtration rate (GFR); however, AmBd itself causes nephrotoxicity and worsens renal function in a dose-dependent manner, creating a clinical management challenge in patients with pre-existing renal disease.4
Spectrum of Activity. The antifungal spectrum of AmB is broad and encompasses most clinically important fungal pathogens. AmB is active against Candida species (including most azole-resistant strains, though Candida lusitaniae demonstrates intrinsic resistance and Candida auris may have elevated minimum inhibitory concentrations), Cryptococcus neoformans and Cryptococcus gattii, Aspergillus species (fungistatic), the agents of mucormycosis (the Mucorales order, including Rhizopus, Mucor, and Lichtheimia), dimorphic fungi (Histoplasma capsulatum, Blastomyces dermatitidis, Coccidioides immitis, Paracoccidioides brasiliensis), and Sporothrix schenckii. Notable gaps in spectrum include Fusarium species (variable susceptibility), Scedosporium species (largely resistant), and Trichosporon species (intrinsic resistance due to low ergosterol content).5
Current Clinical Indications for AmBd. In high-income settings with reliable access to lipid formulations, AmBd has been largely displaced as first-line therapy due to its nephrotoxicity. However, it retains a defined clinical role. It remains an option for patients with short anticipated treatment courses where nephrotoxicity risk is low, for settings where cost considerations make lipid formulations inaccessible, and as an intrathecal agent (rarely) for refractory CNS infections. In resource-limited settings, AmBd combined with flucytosine remains the standard induction regimen for cryptococcal meningitis. The World Health Organization (WHO) 2022 guidelines for cryptococcal meningitis recommend a single high dose of liposomal amphotericin B (10 mg/kg) combined with flucytosine and fluconazole for 14 days as the preferred induction regimen, replacing the prior recommendation of one week of AmBd plus flucytosine. Where liposomal AmB is unavailable, AmBd (1.0 mg/kg/day) plus flucytosine for 7 days followed by fluconazole remains an acceptable alternative regimen, given demonstrated superiority of combination therapy over fluconazole monotherapy in clinical trials.6
Standard dose: 0.5–1.0 mg/kg/day IV. Cryptococcal meningitis induction: 1.0 mg/kg/day. Mucormycosis: 1.0–1.5 mg/kg/day (lipid formulation preferred). Infusion: dilute in D5W (not saline — causes precipitation), infuse over 2–4 hours protected from light. No dose adjustment for renal impairment, but monitor renal function, potassium, and magnesium daily. Sodium loading (500 mL normal saline before each infusion) reduces nephrotoxicity risk and should be routine unless contraindicated by fluid overload.
Three lipid-based formulations of amphotericin B are available in the United States: liposomal amphotericin B (L-AmB; AmBisome), amphotericin B lipid complex (ABLC; Abelcet), and amphotericin B colloidal dispersion (ABCD; Amphotec). All three were developed to reduce nephrotoxicity and infusion-related reactions compared to conventional AmBd, but they differ substantially in their physical structure, pharmacokinetics, clinical efficacy data, and tolerability profiles. Selecting among them requires understanding these differences rather than treating them as interchangeable.3
Liposomal Amphotericin B. L-AmB consists of small unilamellar liposomes (approximately 80 to 100 nanometers in diameter) in which AmB is intercalated into the phospholipid bilayer of the liposome membrane alongside cholesterol and distearoylphosphatidylglycerol. The liposomal membrane shields the encapsulated AmB from contact with cholesterol in mammalian cell membranes, reducing interaction with renal tubular epithelium while allowing preferential delivery to sites of fungal infection, where the liposomal structure is disrupted by fungal phospholipases and high local concentrations of ergosterol. L-AmB achieves substantially higher plasma concentrations than AmBd at equivalent doses because the liposomal encapsulation prolongs circulation time and reduces renal uptake.3
The standard dose of L-AmB is 3 to 5 mg/kg/day IV. For central nervous system (CNS) infections and for empirical therapy in neutropenic fever, 3 mg/kg/day is standard, while mucormycosis may warrant 5 mg/kg/day or higher. L-AmB is the best-tolerated of all amphotericin B formulations and is the preferred formulation when cost is not a limiting factor.7
Amphotericin B Lipid Complex. ABLC consists of ribbon-like lipid bilayer structures (approximately 1.6 to 11 micrometers in length) rather than closed liposomes; AmB is intercalated between dimyristoylphosphatidylcholine and dimyristoylphosphatidylglycerol lipid bilayers at a 1:1 drug-to-lipid molar ratio. Because of its large particle size, ABLC is rapidly cleared from the circulation by the mononuclear phagocyte system (MPS), resulting in high drug concentrations in the liver, spleen, and lungs but lower peak plasma concentrations than either AmBd or L-AmB. This distribution pattern makes ABLC particularly well suited for infections in these organs. The standard dose is 5 mg/kg/day IV. Nephrotoxicity is reduced compared to AmBd but is greater than with L-AmB at equivalent doses. Infusion reactions occur at a similar or slightly higher frequency than with L-AmB. ABLC is approved for invasive fungal infections refractory to or intolerant of conventional AmBd.4
Amphotericin B Colloidal Dispersion. ABCD consists of disk-shaped cholesteryl sulfate complexes (approximately 120 to 140 nanometers in diameter) in which AmB is complexed with cholesteryl sulfate at a 1:1 molar ratio. ABCD has a pharmacokinetic profile intermediate between AmBd and ABLC. ABCD is associated with the highest rate of acute infusion reactions of the three lipid formulations, including fever, rigors, and hypoxia, which substantially limits its clinical use. Nephrotoxicity is reduced compared to AmBd. ABCD is approved for invasive aspergillosis in patients refractory to or intolerant of AmBd and is rarely used today given the superior tolerability of L-AmB and ABLC. The standard dose is 3 to 4 mg/kg/day.4
Comparative Efficacy. A point of practical importance is that the three lipid formulations have not been demonstrated to be superior in antifungal efficacy to AmBd in well-powered randomized controlled trials; rather, they have been shown to be non-inferior in efficacy with superior tolerability, primarily reduced nephrotoxicity. The landmark AmBiLoad trial comparing L-AmB 3 mg/kg/day with L-AmB 10 mg/kg/day for invasive mold infections showed no efficacy benefit to the higher dose despite substantially increased toxicity.8 For mucormycosis, retrospective and registry data support using L-AmB at 5 mg/kg/day or higher, but head-to-head data against AmBd at maximum tolerated doses are limited. The choice of lipid formulation over AmBd is therefore driven primarily by the need to prevent nephrotoxicity in patients who require prolonged courses, who have baseline renal impairment, or who are receiving concomitant nephrotoxic agents.7,8
L-AmB (AmBisome): 3–5 mg/kg/day — best tolerability, preferred first choice, highest cost. ABLC (Abelcet): 5 mg/kg/day — high tissue concentration in liver/spleen/lung via MPS uptake, moderate tolerability, lower cost than L-AmB. ABCD (Amphotec): 3–4 mg/kg/day — highest infusion reaction rate of lipid formulations, rarely chosen when alternatives available. None of the three has demonstrated superior antifungal efficacy compared to AmBd in controlled trials; the advantage is tolerability and nephrotoxicity reduction.
The two dominant toxicities of amphotericin B are acute infusion-related reactions and cumulative nephrotoxicity. These are mechanistically distinct and require different management strategies. Understanding the pathophysiology of each guides both prevention and treatment, and informs the decision between conventional and lipid-based formulations for any given patient.
Infusion-Related Reactions. Acute infusion reactions occur in up to 70% of patients receiving AmBd and are the most immediately distressing manifestation of polyene toxicity. The syndrome typically begins 15 to 60 minutes into the infusion and includes fever (often with rigors and shaking chills), headache, nausea, vomiting, myalgia, and in severe cases, hypotension and bronchospasm. The mechanism involves AmB-induced release of prostaglandins, interleukin-1 (IL-1), and tumor necrosis factor-alpha (TNF-alpha) from monocytes and macrophages through a toll-like receptor 2 (TLR-2) and toll-like receptor 4 (TLR-4) dependent pathway, as well as complement activation via the alternative pathway. The reaction is not IgE-mediated and is therefore not a true allergic reaction; it does not predict anaphylaxis and does not contraindicate continued AmB use.4,9
Premedication with acetaminophen 650 mg and diphenhydramine 25 to 50 mg administered 30 to 60 minutes before each infusion attenuates the febrile and histaminergic components of the reaction. Meperidine 25 to 50 mg IV is specifically effective for breaking established rigors by acting on mu-opioid receptors in the hypothalamus to reset the thermostat. Ibuprofen, hydrocortisone 25 mg IV, and slowing the infusion rate are additional measures used at various centers. Reactions typically diminish in severity with subsequent infusions as tolerance develops.4,9
Nephrotoxicity: Mechanism. AmB nephrotoxicity involves two distinct mechanisms operating at different nephron sites. First, AmB causes afferent arteriolar vasoconstriction in the renal vasculature, reducing renal blood flow and glomerular filtration rate (GFR). This vasoconstrictive effect is mediated through thromboxane A2 (TXA2) release and direct smooth muscle effects, and it is rapidly reversible with drug discontinuation or dose reduction. Second, AmB directly damages the distal tubular epithelium by forming pores in the apical cholesterol-containing membrane, causing tubular dysfunction manifesting as renal tubular acidosis (type 1, distal), urinary potassium wasting, urinary magnesium wasting, and impaired urinary concentrating ability.9
Tubular damage is dose-dependent and cumulative, and at cumulative doses exceeding 2 to 3 grams of AmBd, persistent structural renal damage may occur even after drug discontinuation. Serum creatinine rises in the majority of patients receiving AmBd at standard doses; a doubling of creatinine from baseline is a commonly used threshold for switching to a lipid formulation or discontinuing therapy.9
Sodium Loading and Nephrotoxicity Prevention. The most evidence-based strategy for reducing AmBd nephrotoxicity is sodium loading: administering 500 mL of normal saline (0.9% NaCl) immediately before each AmB infusion. The mechanism involves several complementary pathways: volume expansion reduces tubuloglomerular feedback-mediated afferent arteriolar constriction, sodium delivery to the distal tubule competes with potassium loss, and intravascular volume expansion dilutes free drug concentration and reduces renal tubular exposure. Multiple prospective studies and meta-analyses have demonstrated that routine saline pre-hydration reduces the incidence and severity of nephrotoxicity without compromising antifungal efficacy. Sodium loading is contraindicated in patients with severe heart failure, pulmonary edema, or anasarca, where the volume load is not tolerable; in these patients, a lipid formulation should be used from the outset. Avoidance of concomitant nephrotoxins (aminoglycosides, vancomycin, NSAIDs, IV contrast, tacrolimus, cyclosporine) during AmBd therapy is also essential; when co-administration cannot be avoided, lipid formulations are mandatory.9
Electrolyte Management. Hypokalemia and hypomagnesemia are near-universal with prolonged AmBd therapy and require aggressive replacement. Potassium wasting results from distal tubular dysfunction and can be severe (serum potassium below 2.5 mEq/L) and refractory to standard oral supplementation. Intravenous potassium chloride replacement, often at doses of 60 to 120 mEq per day, is frequently required. Hypomagnesemia amplifies potassium wasting because magnesium is required for the normal function of the renal outer medullary potassium (ROMK) channel responsible for potassium reabsorption in the distal nephron; correction of hypomagnesemia is therefore a prerequisite for effective potassium repletion. Magnesium supplementation (oral magnesium oxide or IV magnesium sulfate) must accompany potassium replacement in patients with concurrent deficits in both electrolytes. Monitoring should include serum electrolytes, creatinine, and blood urea nitrogen (BUN) at baseline and at minimum every two to three days during therapy, with daily monitoring during the induction phase or when nephrotoxicity is developing.4,9
Use lipid AmB as initial therapy (do not start with AmBd) when any of the following are present: baseline creatinine above 2.5 mg/dL; creatinine clearance below 25 mL/min; concurrent nephrotoxin use that cannot be discontinued (calcineurin inhibitors, aminoglycosides); prior AmBd nephrotoxicity; solid organ or stem cell transplant recipient; anticipated treatment duration exceeding 2 weeks. Switch from AmBd to lipid formulation during therapy if creatinine doubles from baseline or potassium/magnesium cannot be maintained despite aggressive replacement.
Nystatin, the second clinically available polyene antifungal, shares the same fundamental mechanism of action as amphotericin B but differs so dramatically in its pharmacokinetic and toxicity profile that it occupies an entirely separate clinical niche. Understanding nystatin's properties alongside the mechanisms and epidemiology of polyene resistance completes the pharmacological picture of this drug class.
Nystatin: Mechanism and Structure. Nystatin is a macrolide polyene produced by Streptomyces noursei, structurally similar to AmB but with a tetraene rather than heptaene polyene chain and a slightly different macrolide ring. Like AmB, it binds ergosterol in fungal cell membranes and forms transmembrane pores causing ion leakage and cell death. Its antifungal spectrum against Candida species and other fungi is comparable to AmB. The critical pharmacokinetic difference is that nystatin is essentially completely insoluble in aqueous solution at physiological pH, which means it cannot be formulated for intravenous administration without producing severe systemic toxicity. The original IV nystatin formulations developed in the 1950s caused unacceptable systemic toxicity and were abandoned. A liposomal nystatin formulation showed promise in clinical trials for invasive fungal infections and demonstrated antifungal efficacy with reduced nephrotoxicity, but was not approved by the US Food and Drug Administration (FDA) and remains investigational. Nystatin therefore exists in clinical practice exclusively as a topical and oral non-absorbed formulation.10
Clinical Applications of Nystatin. Oral nystatin suspension (100,000 units/mL) is used for oropharyngeal candidiasis (thrush) at a dose of 400,000 to 600,000 units four times daily, swished around the mouth and then swallowed or expectorated. It is also used as a swish-and-swallow preparation for esophageal candidiasis in patients who cannot tolerate systemic azoles, though fluconazole is significantly more effective and is the preferred agent for documented esophageal disease. Nystatin tablets (500,000 units) taken orally are used to suppress intestinal Candida colonization, although the clinical benefit of intestinal decolonization in preventing invasive candidiasis in hospitalized patients has not been convincingly demonstrated in large trials. Topical nystatin preparations (cream, ointment, powder) are used for cutaneous and mucocutaneous candidiasis including diaper dermatitis, intertrigo, and vaginal candidiasis. Vaginal nystatin suppositories (100,000 units) are an option for vulvovaginal candidiasis in pregnancy, where systemic azoles are generally avoided. Because it is not absorbed from the gastrointestinal (GI) tract, nystatin produces no systemic toxicity when taken orally; local GI side effects (nausea, vomiting, diarrhea) are the only significant adverse effects at standard doses.10
Polyene Resistance: Mechanisms. True resistance to amphotericin B is rare compared to azole resistance and is one of the defining advantages of the polyene class. The primary mechanism of polyene resistance is depletion or structural alteration of ergosterol in the fungal cell membrane, which reduces or eliminates the drug-binding target. This most commonly results from mutations in ergosterol biosynthesis genes, particularly ERG3 (which encodes C-5 sterol desaturase) and ERG11 (which encodes lanosterol 14-alpha-demethylase). ERG3 mutations cause accumulation of 14-alpha-methylfecosterol, a toxic intermediate that cannot substitute for ergosterol and also cannot bind AmB, effectively removing the drug target. ERG11 mutations reduce lanosterol demethylation, leading to accumulation of methylated sterols that do not support AmB pore assembly. Resistance associated with these ergosterol biosynthesis gene (ERG gene) mutations has been documented clinically in Candida glabrata and Candida tropicalis following prolonged azole exposure, because azoles target the same ergosterol pathway and can co-select for mutations that simultaneously confer azole resistance and reduce AmB susceptibility.11
Clinical Epidemiology of Resistance. Candida lusitaniae demonstrates intrinsic AmB resistance through constitutive ERG3 mutations, producing membranes with reduced ergosterol content. This is an actionable clinical point because C. lusitaniae can be misidentified as other Candida species by some automated identification systems, and a patient with C. lusitaniae fungemia treated empirically with AmB will fail therapy. Candida auris, the multidrug-resistant emerging pathogen of concern, shows variable susceptibility to AmB, with minimum inhibitory concentrations (MICs) at or above the susceptibility breakpoint in some clades, and AmB-resistant isolates have been reported from outbreaks in healthcare facilities. Among mold pathogens, Scedosporium species and Lomentospora prolificans are intrinsically resistant to AmB, as are Trichosporon species (which have low membrane ergosterol content) and most dematiaceous fungi causing chromoblastomycosis. Acquired resistance during AmB therapy is uncommon but has been documented with prolonged exposure in immunocompromised hosts, particularly those with Aspergillus or Candida infections.5,11
Candida lusitaniae — intrinsic resistance (ERG3 mutation); consider alternative therapy for all isolates. Candida auris — variable; susceptibility testing mandatory before relying on AmB. Scedosporium apiospermum and Lomentospora prolificans — intrinsically resistant; voriconazole preferred. Trichosporon species — intrinsic resistance; azoles preferred. Fusarium species — variable, often reduced susceptibility. Do not assume AmB susceptibility for any organism without species-level identification.
Translating polyene pharmacology into sound prescribing decisions requires integrating mechanism, formulation differences, toxicity risk factors, and spectrum into a coherent decision framework. The following principles synthesize the preceding sections into clinically actionable guidance.
Choosing Between AmBd and a Lipid Formulation. The decision to use conventional AmBd versus a lipid-based formulation should be made before initiating therapy, not reactively after nephrotoxicity develops. Lipid formulation use from the outset is appropriate whenever any of the following risk factors are present: baseline creatinine above 2.5 mg/dL or creatinine clearance (CrCl) below 25 mL/min, receipt of concomitant nephrotoxic agents that cannot be discontinued (calcineurin inhibitors, aminoglycosides, intravenous contrast), solid organ or hematopoietic stem cell transplant recipient status, anticipated treatment duration exceeding two weeks, or a history of prior AmBd nephrotoxicity. In these settings, initiating AmBd and planning to switch if nephrotoxicity develops is a suboptimal strategy because tubular damage from even brief AmBd exposure can be cumulative and partially irreversible. In patients without these risk factors and with short anticipated treatment courses, AmBd with sodium loading remains a reasonable and substantially less costly option.9
Choosing Among Lipid Formulations. L-AmB is the preferred lipid formulation in most clinical settings due to its superior infusion tolerability and the largest body of clinical trial evidence supporting its use. Amphotericin B lipid complex (ABLC) is an acceptable alternative, particularly when cost is a significant constraint or when infections of the liver, spleen, or lungs are the primary concern, given its distribution characteristics through the mononuclear phagocyte system (MPS). Amphotericin B colloidal dispersion (ABCD) carries the highest infusion reaction rate of the three formulations and should not be selected when L-AmB or ABLC are available. For mucormycosis specifically, L-AmB at 5 mg/kg/day or higher is preferred based on available registry data, with surgical debridement as an essential adjunct to medical therapy. None of the three lipid formulations has been demonstrated to be more efficacious than AmBd in prospective controlled trials; the selection rationale is tolerability, not superior antifungal activity.7,8
Monitoring Parameters During Amphotericin B Therapy. All patients receiving any amphotericin B formulation require systematic laboratory monitoring. Serum creatinine, blood urea nitrogen (BUN), potassium, magnesium, and sodium should be measured at baseline and at minimum every two to three days during stable therapy, with daily measurement during the induction phase or when nephrotoxicity is evolving. A doubling of serum creatinine from baseline is the standard threshold for switching from AmBd to a lipid formulation or discontinuing therapy. Hypokalemia and hypomagnesemia require aggressive replacement, and hypomagnesemia must be corrected before potassium wasting can be brought under control, given the dependence of renal potassium conservation on normal magnesium levels. Liver function tests should be monitored periodically, particularly with ABLC, which distributes heavily into hepatic tissue. Complete blood count (CBC) monitoring is less routinely required for polyenes than for flucytosine but is appropriate in patients on prolonged courses.4,9
Amphotericin B and Drug Interactions. The polyene antifungals have a favorable drug interaction profile compared to the azoles because they are not metabolized by cytochrome P450 (CYP) enzymes and do not inhibit or induce CYP isoforms. There are therefore no pharmacokinetic drug-drug interactions of the type that make azole prescribing so complex in transplant and oncology patients. The clinically relevant interactions with AmB are pharmacodynamic rather than pharmacokinetic: additive or synergistic nephrotoxicity with aminoglycosides, vancomycin, nonsteroidal anti-inflammatory drugs (NSAIDs), intravenous radiocontrast agents, and calcineurin inhibitors (tacrolimus, cyclosporine). Concurrent use of loop diuretics amplifies the electrolyte wasting caused by tubular toxicity. The combination of AmB with flucytosine (5-fluorocytosine, 5-FC) is pharmacodynamically synergistic against Cryptococcus neoformans and is the standard induction regimen for cryptococcal meningitis; AmB enhances 5-FC uptake by increasing fungal cell membrane permeability, and the combination is fungicidal at concentrations lower than either agent alone.4,6
Spectrum Gaps and Empirical Therapy Limitations. Despite its broad spectrum, several important gaps in AmB coverage must be recognized to avoid empirical therapy failures. Candida lusitaniae, Candida auris (variable), Scedosporium species, Lomentospora prolificans, and Trichosporon species are not reliably covered by AmB. When empirical polyene therapy is initiated for a febrile immunocompromised patient and response is inadequate, failure of microbiological identification to yield an organism does not validate continued AmB therapy if the clinical trajectory is worsening. Species-level identification with susceptibility testing should guide de-escalation or switch decisions. Galactomannan and beta-d-glucan assays provide early surrogate markers for response in invasive mold and yeast infections respectively, and serial monitoring of these biomarkers can support treatment duration and formulation decisions even before culture data are available.5,11
AmB mechanism: ergosterol binding, transmembrane pore formation, fungicidal against yeasts, fungistatic against molds. AmBd dose: 0.5–1.0 mg/kg/day; always sodium-load 500 mL NS before each infusion. Lipid formulation threshold: creatinine doubling, concurrent nephrotoxins, transplant recipients, anticipated duration above 2 weeks. L-AmB preferred lipid formulation; ABLC acceptable alternative; ABCD rarely chosen. No CYP interactions; pharmacodynamic nephrotoxicity interactions with aminoglycosides, calcineurin inhibitors, NSAIDs, contrast. Spectrum gaps: C. lusitaniae (intrinsic), C. auris (variable), Scedosporium, Trichosporon. AmB + flucytosine: synergistic, standard induction for cryptococcal meningitis.
Brajtburg J, Powderly WG, Kobayashi GS, Medoff G. Amphotericin B: current understanding of mechanisms of action. Antimicrob Agents Chemother. 1990;34(2):183-188.
doi:10.1128/AAC.34.2.183Bolard J. How do the polyene macrolide antibiotics affect the cellular membrane properties? Biochim Biophys Acta. 1986;864(3-4):257-304.
doi:10.1016/0304-4157(86)90002-XAdler-Moore J, Proffitt RT. AmBisome: liposomal formulation, structure, mechanism of action and pre-clinical experience. J Antimicrob Chemother. 2002;49(Suppl 1):21-30.
doi:10.1093/jac/49.suppl_1.21Gallis HA, Drew RH, Pickard WW. Amphotericin B: 30 years of clinical experience. Rev Infect Dis. 1990;12(2):308-329.
doi:10.1093/clinids/12.2.308Patterson TF, Thompson GR, Denning DW, et al. Practice guidelines for the diagnosis and management of aspergillosis: 2016 update by the Infectious Diseases Society of America. Clin Infect Dis. 2016;63(4):e1-e60.
doi:10.1093/cid/ciw326World Health Organization. Guidelines for Diagnosing, Preventing and Managing Cryptococcal Disease Among Adults, Adolescents and Children Living with HIV. Geneva: WHO; 2022. ISBN 9789240052178.
who.int/publications/i/item/9789240052178Walsh TJ, Finberg RW, Arndt C, et al. Liposomal amphotericin B for empirical therapy in patients with persistent fever and neutropenia. N Engl J Med. 1999;340(10):764-771.
doi:10.1056/NEJM199903113401004Cornely OA, Maertens J, Bresnik M, et al. Liposomal amphotericin B as initial therapy for invasive mold infection: a randomized trial comparing a high-loading dose regimen with standard dosing (AmBiLoad Trial). Clin Infect Dis. 2007;44(10):1289-1297.
doi:10.1086/514341Heidemann HT, Gerkens JF, Spickard WA, Jackson EK, Branch RA. Amphotericin B nephrotoxicity in humans decreased by salt repletion. Am J Med. 1983;75(3):476-481.
doi:10.1016/0002-9343(83)90353-4Pappas PG, Kauffman CA, Andes DR, et al. Clinical practice guideline for the management of candidiasis: 2016 update by the Infectious Diseases Society of America. Clin Infect Dis. 2016;62(4):e1-e50.
doi:10.1093/cid/civ933Sanglard D, Odds FC. Resistance of Candida species to antifungal agents: molecular mechanisms and clinical consequences. Lancet Infect Dis. 2002;2(2):73-85.
doi:10.1016/S1473-3099(02)00181-0