What Is Extended-Spectrum Beta-Lactamase?

Extended-spectrum beta-lactamase, usually shortened to ESBL, is an enzyme produced by certain bacteria that breaks down a wide range of common antibiotics, including many penicillins and cephalosporins that doctors rely on every day. The name sounds technical, but the concept is straightforward: some gut bacteria have evolved a molecular tool that chews up antibiotics before the drugs can kill them. These enzymes are not a disease in themselves. They are a resistance mechanism carried by bacteria that cause familiar infections like urinary tract infections, bloodstream infections, and pneumonia, making those infections much harder to treat.¹

How the Enzyme Works

Beta-lactam antibiotics, the family that includes penicillin, amoxicillin, and cephalosporins, share a chemical structure called a beta-lactam ring. That ring is the part of the molecule that kills bacteria by interfering with their cell walls. Regular beta-lactamase enzymes, which have existed for decades, can break that ring on older antibiotics. ESBLs go further. They evolved from older enzymes through small genetic mutations that reshaped the area around their active site, giving them the ability to hydrolyze (essentially cut open) newer, more powerful antibiotics like third-generation cephalosporins and aztreonam.¹ That is where the “extended-spectrum” label comes from: the enzyme’s reach extends well beyond the antibiotics its predecessors could destroy.

One useful detail is that ESBLs can still be blocked by certain enzyme inhibitors, particularly clavulanic acid (the ingredient in amoxicillin-clavulanate, often sold as Augmentin). This quirk matters for both diagnosis and treatment, as labs use clavulanic acid to confirm the presence of ESBLs and distinguish them from other resistance mechanisms.

Which Bacteria Carry ESBLs

ESBLs are found almost exclusively in a group of bacteria called Enterobacteriaceae, the large family of rod-shaped bacteria that live in the human gut. The two species that cause the most trouble are Escherichia coli and Klebsiella pneumoniae. Together, they account for the overwhelming majority of ESBL-producing infections worldwide, driving both bloodstream infections and urinary tract infections in hospitals and in the community.²

These are not exotic organisms. E. coli is the most common cause of urinary tract infections in otherwise healthy people, and K. pneumoniae is a leading cause of hospital-acquired pneumonia and bloodstream infections. The bacteria themselves are familiar; it is their resistance toolkit that has changed. In a study of bacteremic pneumonia cases caused by ESBL producers, K. pneumoniae was responsible for about two-thirds and E. coli for the remaining third.³

The genes that code for ESBLs sit on plasmids, small loops of DNA that bacteria can pass to one another like trading cards. A specific family of plasmids called IncF types is the most common vehicle, and one particularly successful gene variant, CTX-M-27, has spread globally by hitching a ride on these plasmids, often inside a highly virulent strain of E. coli known as ST131.⁴ This ability to jump between bacteria is a major reason ESBL genes have spread so rapidly across the globe.

Who Is at Risk

Anyone can develop an ESBL-producing infection, but certain factors raise your odds substantially. The strongest risk factor is recent antibiotic use. In children, having taken any antibiotic in the 30 days before infection roughly doubled the risk of the infecting bacteria being ESBL-positive compared with children who had no recent antibiotic exposure.⁵ In adults, a study of community-acquired urinary tract infections found that repeated UTIs, having a urinary catheter at the time of admission, and antibiotic exposure within the prior three months were all independently associated with ESBL-producing infections.⁶

Hospitalization itself is a risk factor, especially long stays involving invasive devices like central lines or ventilators. But ESBL infections are no longer confined to hospitals. They are increasingly picked up in the community, particularly among people with chronic urinary problems or those who have recently been treated with broad-spectrum antibiotics. The common thread is antibiotic pressure: the more antibiotics you have been exposed to, the more opportunity ESBL-carrying bacteria have to outcompete their susceptible neighbors in your gut.

Carrying ESBLs Without Being Sick

Many people carry ESBL-producing bacteria in their intestines without ever developing an infection. This silent colonization is more common than most people realize. When colonized patients were tracked over several years after a urinary tract infection, about six in ten still carried ESBL producers at four months, about four in ten at roughly a year, and roughly one in five after two or more years.⁷ Some people who tested negative later became positive again, suggesting reacquisition from the environment or from other carriers.

The practical concern with gut colonization is that it serves as a reservoir. If a colonized person later gets sick or needs surgery, their own gut bacteria can cause an infection that is already resistant to first-line antibiotics. Among hospitalized patients carrying ESBL-producing K. pneumoniae in their gut, the rate of subsequent hospital-acquired infection was higher than among those carrying ESBL-producing E. coli.⁸ In other words, colonization with K. pneumoniae seems to carry a somewhat higher risk of progressing to actual disease than colonization with E. coli.

How Labs Detect ESBLs

Figuring out whether a bacterial infection involves ESBL production is not as simple as looking at the bug under a microscope. The classic approach uses something called phenotypic testing, which relies on a clever trick: comparing how well a cephalosporin antibiotic works on its own versus how well it works when paired with clavulanic acid, the inhibitor mentioned earlier. If the antibiotic works dramatically better with the inhibitor present, the bacteria are almost certainly producing an ESBL.⁹

Several versions of this test exist. The double-disk synergy test places antibiotic discs on a plate of bacteria and looks for a telltale expansion of the zone where bacteria cannot grow near the clavulanic acid disc. Including a fourth-generation cephalosporin like cefepime alongside the usual third-generation cephalosporins can catch strains that also produce another resistance enzyme called AmpC, which can otherwise mask ESBL results.¹⁰ These tests are cheap and widely available, but they can produce ambiguous results when bacterial strains sit at the borderline, especially those that produce low levels of enzyme or carry unusual ESBL gene variants.¹¹

Speed matters, because every hour a patient spends on an ineffective antibiotic is a lost hour. Newer approaches use mass spectrometry (MALDI-TOF MS) to watch the antibiotic break down in real time. Researchers have shown that by incubating bacteria with a cephalosporin for as little as 15 minutes, then checking for breakdown products using MALDI-TOF, ESBL-producing strains can be clearly distinguished from susceptible ones.¹² Conventional testing can take a day or more; cutting that to under an hour could help clinicians switch to the right antibiotic much sooner.

Treatment and the Carbapenem Question

The go-to treatment for serious ESBL infections has long been a class of antibiotics called carbapenems, which are structurally resistant to ESBL enzymes and provide reliable coverage against these bacteria.¹³ Drugs like meropenem and imipenem are the heavy artillery that doctors reach for when standard antibiotics fail. The problem is that heavy use of carbapenems drives the emergence of carbapenem-resistant bacteria, a threat considered even more dangerous than ESBLs because treatment options become vanishingly small.¹⁴

This has sparked a search for “carbapenem-sparing” alternatives. One candidate was piperacillin-tazobactam, a broader-spectrum penicillin-inhibitor combination already in widespread hospital use. A landmark randomized trial (MERINO) tested it head-to-head against meropenem in patients with ceftriaxone-resistant E. coli or K. pneumoniae bloodstream infections. The results were sobering: 30-day mortality was about 12% in the piperacillin-tazobactam group versus roughly 4% in the meropenem group, and the trial failed to show that piperacillin-tazobactam was at least as good.¹⁵ A follow-up analysis found that strains carrying both an ESBL gene and a particular additional resistance gene (OXA-1) were responsible for the highest mortality gap, with an absolute risk increase of about 14%.¹⁶

The picture is not entirely black and white, though. In settings where that OXA-1 co-resistance gene is uncommon, a preliminary study found no significant difference in treatment failure between piperacillin-tazobactam and carbapenems for less severe ESBL E. coli bloodstream infections.¹⁷ The takeaway for now is that carbapenems remain the safest bet for serious ESBL bloodstream infections, but the door is open for targeted carbapenem-sparing strategies once doctors have quick access to detailed resistance information.

For uncomplicated urinary tract infections caused by ESBL producers, the situation is friendlier. Older oral antibiotics like nitrofurantoin, fosfomycin, and pivmecillinam still work against the vast majority of ESBL-producing strains. One study found sensitivity rates above 95% for all three.¹⁸ These drugs work well for bladder infections because they concentrate in the urine, though they are not suitable for bloodstream or deep-tissue infections.

Hospital Burden and Mortality

ESBL-producing infections are not automatically more lethal than their antibiotic-susceptible counterparts, but they do complicate care. A Korean study comparing ESBL and non-ESBL bloodstream infections found that hospital stays were about 1.5 times longer for ESBL cases and total costs were significantly higher.¹⁹ A Swedish study found that urinary-source bloodstream infections from ESBL-producing E. coli led to a median hospital stay about two days longer than non-ESBL equivalents, though 30-day mortality was not significantly different after adjustment.²⁰ Effective antibiotics were given within 24 hours to only about 55% of ESBL cases compared with about 87% of non-ESBL cases, highlighting the time penalty created by unexpected resistance.

The true danger lies less in the ESBL enzyme itself and more in the delay it causes. When clinicians start with a standard antibiotic that turns out to be useless, the infection has extra time to worsen. This is why fast detection and awareness of local resistance patterns are so consequential.

The Co-Resistance Problem

ESBL genes rarely travel alone. The same plasmids that carry ESBL genes often carry resistance genes for other antibiotic classes, including aminoglycosides, sulfonamides, tetracyclines, and fluoroquinolones. In one study of E. coli from commercial poultry flocks, about two-thirds of ESBL-positive strains were also resistant to aminoglycosides, sulfonamides, and tetracyclines.²¹ This bundling of resistance genes means that even using a non-beta-lactam antibiotic can select for the entire plasmid, including the ESBL gene, because the plasmid confers survival advantages against multiple drug classes simultaneously.

Neonates and Young Children

ESBL-producing bacteria are an especially serious concern in neonatal intensive care units. Newborns have immature immune systems and frequently need invasive procedures like ventilators and central lines, all of which raise infection risk. A meta-analysis of risk factors in the NICU found that prior antibiotic use, particularly cephalosporins, low birth weight, lower gestational age, and use of parenteral nutrition or mechanical ventilation were all strongly associated with ESBL colonization or infection.²²

ESBL outbreaks in NICUs tend to be devastating. A systematic review of such outbreaks reported a pooled mortality rate of about 31% among infected neonates, and the typical outbreak lasted more than six months. Understaffing was the most commonly identified risk factor for outbreaks spreading, and the usual trigger was the admission of a colonized baby followed by horizontal transmission to others.²³ Mother-to-child transmission during birth has also been documented; in one cohort of colonized infants, about 6% developed late-onset sepsis and nearly 3% died.²⁴

Animals, Food, and the Environment

ESBL-producing bacteria are not limited to hospitals or even to humans. They have been found in livestock, retail meat, pets, and environmental water sources. This has led to concern that the food chain is a major highway for resistant bacteria to reach people. The picture, however, is more nuanced than it first appears. A large pooled analysis of ESBL gene distributions across human, animal, meat, and environmental reservoirs found that while most ESBL gene subtypes turned up in every reservoir to some degree, the gene profiles in livestock and food did not closely match those circulating in human clinical populations. The authors concluded that livestock reservoirs, including poultry, are probably not the main source of ESBL-producing bacteria in humans.²⁵

That said, mobile genetic elements carrying ESBL genes do circulate through agricultural and environmental pathways, and the potential for horizontal gene transfer between bacterial strains in shared environments remains a concern from a One Health perspective.²⁶ Even if farm animals are not the primary source of human ESBL infections today, widespread antibiotic use in agriculture creates a vast selection pressure that keeps resistance genes circulating in the broader ecosystem.

Stewardship and Prevention

Because antibiotic use is the single strongest driver of ESBL spread, antibiotic stewardship programs have become a frontline defense. A meta-analysis of stewardship interventions found that hospitals that implemented structured programs saw a reduction in ESBL-producing infections by about 18%, with a pooled odds ratio of 0.82.²⁷ In practice, stewardship means reviewing whether each antibiotic prescription is necessary, choosing the narrowest-spectrum drug that will work, and stopping treatment as soon as it is safe to do so.

Infection control measures, including hand hygiene, contact precautions for colonized patients, and environmental cleaning, also play a role, especially in settings where patients are densely clustered. A university hospital that launched a multidisciplinary intervention combining infection control with antibiotic stewardship saw large drops in ESBL K. pneumoniae infections, with infection proportions falling from about half to around a quarter in surgical wards.²⁸ Interestingly, the effect on ESBL E. coli infections was not significant, likely because E. coli infections often arise from a patient’s own gut flora rather than from hospital transmission, making them harder to prevent through contact precautions alone.

Experimental Approaches on the Horizon

With standard antibiotic options narrowing, researchers are exploring some unconventional strategies. One is bacteriophage therapy, which uses viruses that naturally prey on bacteria. Laboratory studies have shown that specific phages can completely eradicate ESBL-producing E. coli within hours at optimal doses, and that cocktails of multiple phages targeting different bacterial surface receptors can suppress regrowth for at least 24 hours in vitro.²⁹ Phage therapy is still mostly experimental in humans for this purpose, but phages against ESBL producers have been isolated from environmental water sources, suggesting a rich natural supply to draw from.³⁰

Another line of research involves fecal microbiota transplantation (FMT), better known for treating recurrent Clostridioides difficile infection, as a way to flush ESBL-producing bacteria out of the gut. A proof-of-principle study in 15 colonized patients found that 20% tested negative for ESBL producers after a single transplant, and 40% were clear after a second, with restored microbial diversity moving the gut ecosystem closer to the donor’s composition.³¹ Those are modest success rates, and the approach is nowhere near routine clinical use, but the logic is appealing: rather than trying to kill resistant bacteria with yet another antibiotic, you try to outcompete them with a healthier microbial community.

On the drug development side, newer beta-lactamase inhibitors like avibactam, vaborbactam, and relebactam are designed to block a broader range of resistance enzymes than clavulanic acid can. These are paired with existing antibiotics and have shown promise in clinical trials.³² They represent the most near-term expansion of the treatment toolkit, and some combinations are already approved for specific indications, though their use needs to be just as carefully stewarded as carbapenems to prevent the next wave of resistance from arriving even sooner.