β-Lactam Antibiotic and β-Lactamase Inhibitor Combinations

β-Lactam antibiotics act by binding to penicillin-binding proteins (PBPs), thus inhibiting bacterial cell wall synthesis. Increasingly, however, these antibiotics are rendered ineffective because of degradation by β-lactamases. This family of enzymes, which are produced by gram-positive and gram-negative bacteria (including anaerobes) and mycobacteria,^{1 - 2} hydrolyze the β-lactam ring, thereby inactivating the antibiotic molecule prior to binding with PBPs. In response, β-lactamase inhibitors have been developed to conserve the activity and extend the spectrum of any accompanying β-lactam drug against β-lactamase–producing microorganisms.

More than 90% of Staphylococcus aureus strains worldwide now produce β-lactamase, which is liberated extracellularly and hydrolyzes all penicillins other than those in the methicillin group. Most other gram-positive bacteria do not produce β-lactamase, although resistance may occur by other means. In penicillin-resistant Streptococcus pneumoniae (PRSP), for instance, resistance occurs because of PBP mutations rather than β-lactamase production.¹

Globally, more than 90% of Bacteroides fragilis strains produce β-lactamase, and frequencies for other gram-negative organisms have been reported in the range of 30% to 60%.^{1 ,3} The gram-negative β-lactamases are classified by their substrate profile, susceptibility to inhibitors, molecular weight, and genetic determinants.^{1 ,4} They may be coded on plasmid or chromosomal DNA, and more than 1 type may be produced by the same species at the same or different times. Due to carriage on plasmids and promiscuous exchange of such material between bacteria, these resistance genes have spread widely and are also subject to mutation.

Plasmid-mediated β-lactamases, such as those produced by methicillin-sensitive S aureus (MSSA), Haemophilus influenzae, Moraxella catarrhalis, Escherichia coli, Klebsiella pneumoniae, certain species of Enterobacteriaceae, and anaerobes like B fragilis, are susceptible to inactivation by β-lactamase inhibitors. By contrast, chromosomally encoded β-lactamases (which are typically produced by Enterobacter, Pseudomonas, Citrobacter, Morganella, and Serratia species) generally are not.^{1 ,3 - 4} Chromosomally mediated β-lactamases are sometimes inducible, especially following treatment with extended-spectrum cephalosporins.¹ The clinical significance of in vitro activity against mycobacterial β-lactamases remains unclear.²

Clavulanic acid, sulbactam, and tazobactam are the 3 inhibitors currently available for clinical use. Orally, sulbactam is combined with ampicillin as a prodrug and is commercially available in Asia and Europe. In the United States sulbactam is approved only for parenteral use. Clavulanic acid is a natural product, whereas sulbactam and tazobactam are semisynthetic.⁵ Their irreversible binding to the catalytic site of β-lactamases prevents inactivation of the accompanying β-lactam. Generally they have little intrinsic antimicrobial impact, but together with penicillins or cephalosporins they effect a wider spectrum of activity in association with lower minimum inhibitory concentrations (MICs).³

All 3 inhibitors are usually effective against pathogens producing plasmid-mediated β-lactamases, although the overall antibacterial spectrum of these drug combinations depends on the intrinsic activity of the accompanying β-lactam as well as the characteristics of the inhibitor. A significant proportion of plasmid-mediated ampicillin-resistant E coli strains exhibit resistance to β-lactam/β-lactamase inhibitor combinations, for instance, although piperacillin/tazobactam may still be efficacious.⁶ Furthermore, clavulanate weakly induces β-lactamase production, which may negate the antipseudomonal effect of ticarcillin if they are administered together.^{7 - 8} Similarly, other extended-spectrum cephalosporins (eg, ceftazidime, ceftriaxone, cefotaxime)⁸ may more strongly induce β-lactamase production than piperacillin, ticarcillin, or cefoperazone, and therefore have not been combined with β-lactamase inhibitors.

Although neither ampicillin nor sulbactam alone are well absorbed from the gut, the prodrug combination achieves a bioavailability of about 80%, which is comparable to the about 90% availability for oral amoxicillin/clavulanate.^{4 - 5 ,9 - 10} The β-lactam-to-inhibitor ratio ranges from 1:1 to 30:1 in terms of weight per dose. Intravenous amoxicillin/clavulanate contains a smaller proportion of clavulanate than do oral formulations because parenteral clavulanate is not subject to first-pass metabolism. Each component of all current β-lactam/β-lactamase inhibitor combinations have elimination half-lives of about 30 to 60 minutes. Tissue penetration is generally good but data for cerebrospinal fluid are lacking.^{1 ,4 - 5} However, because of differential pharmacokinetics of β-lactams and inhibitors, the intended tissue concentration ratios may not always be achieved, particularly for the currently available fixed-drug combinations. For instance, while tazobactam and sulbactam are excreted renally, clavulanate undergoes extensive hepatic metabolism.^{5 ,11} Consequently, in patients with renal failure (for whom dosing intervals are usually increased), a relative deficiency of clavulanate may occur.⁵

The most common side effects of β-lactam/β-lactamase inhibitor combinations include diarrhea, elevated liver enzyme levels, and rashes. These are usually mild and transient, although they may be somewhat more frequent than with the corresponding β-lactam antibiotic alone.^{4 ,7 ,10 - 11} Oral amoxicillin/clavulanate has been associated with diarrhea in as many as 10% of those exposed. On the other hand, overgrowth of colonic Clostridium difficile is less frequent with β-lactam/β-lactamase inhibitor combinations than with cephalosporins alone.^{9 - 10} Of course, other adverse effects that have been described for the β-lactam component may occur with combination therapy.⁴ Clavulanate has been associated with rare and usually reversible cholestatic hepatitis.⁹

While most upper respiratory tract infections are viral in origin, bacterial sinusitis, otitis media, and pharyngitis are usually caused by H influenzae, M catarrhalis, S pneumoniae, group A Streptococci, and S aureus. These bacteria are generally susceptible to oral amoxicillin/clavulanate or ampicillin/sulbactam.^{1 ,4 ,9 - 11}

In the community, many of the bacterial pathogens responsible for lower respiratory tract infections are similar to those that cause upper respiratory tract infections. However, pneumonia due to Mycoplasma, Chlamydia, and Legionella species are unresponsive to β-lactam/β-lactamase inhibitors. Appropriate oral doses of ampicillin/sulbactam or amoxicillin/clavulanate have achieved 80% clinical cure rates for community-acquired pneumonia and seem to be as efficacious and cost-effective as cephalosporins.^{1 ,9 - 10 ,12} For severe infections, therapy can be initiated parenterally.^{7 ,9} Infections due to PRSP pose a serious therapeutic challenge because resistance is due to PBP mutations, which are associated with multidrug resistance. Since amoxicillin has greater in vitro activity against S pneumoniae than other β-lactam antibiotics, strains with intermediate resistance to penicillin may be susceptible to amoxicillin/clavulanate, as long as dosages are commensurate with relevant MICs.^{1 ,13}

Because of their broader spectrum, ticarcillin/clavulanate or piperacillin/tazobactam can also be considered for initial treatment of severe pneumonia, as long as PRSP is unlikely.^{4 ,7} These agents are also useful against aspiration pneumonia because of their activity against anaerobes.^{4 ,7} In patients with nosocomial pneumonia, resistant gram-negative bacilli (including Pseudomonas aeruginosa) or MSSA are frequent pathogens. Piperacillin/tazobactam and cefoperazone/sulbactam provide effective empiric therapy in such cases, with clinical success rates approaching 80%, which is comparable to use of ceftazidime plus amikacin or imipenem/cilastatin.^{4 ,7 ,14 - 15}

Appendicitis, peritonitis, intra-abdominal abscesses, and biliary tract infections are generally polymicrobial. In addition to surgical debridement and drainage, antibiotic cover against gram-negative bacilli, anaerobic pathogens, and enterococci is recommended. While standard regimens using various combinations of clindamycin, metronidazole, aminoglycosides, and second- or third-generation cephalosporins are complicated and require monitoring, β-lactam/β-lactamase inhibitor combinations afford simpler alternatives. The efficacy and cost-effectiveness of piperacillin/tazobactam are comparable to those of standard combination regimens or imipenem/cilastatin monotherapy, and clinical response rates approach 90%.^{4 ,7 ,15 - 16} Results with sulbactam combinations are similar.^{4 ,7 ,17} In addition, amoxicillin/clavulanate, ampicillin/sulbactam, and piperacillin/tazobactam (but not ticarcillin/clavulanate) provide clinically significant activity against enterococci.³

β-Lactam/β-lactamase inhibitor combinations have been successful in treating urinary tract infections (commonly due to β-lactamase–producing E coli).^{9 - 10} However, with the possible exception of piperacillin/tazobactam, their therapeutic role against these infections is being hampered by emergent inhibitor-resistant strains.⁶

Genotypic tests can differentiate true methicillin-resistant S aureus due to PBP mutations (against which β-lactam/β-lactamase inhibitor combinations are ineffective) from those that produce large amounts of β-lactamase against which high doses of β-lactam/β-lactamase inhibitor combinations and other antibiotics should be effective.¹⁸ In vitro and in vivo experimental data have demonstrated the synergistic activity of combination regimens.¹⁹

Although P aeruginosa is generally susceptible to piperacillin/tazobactam or cefoperazone/sulbactam, concomitant aminoglycoside therapy is recommended because of possible synergistic activity.^{4 ,9} Ticarcillin/clavulanate remains active against the multidrug (including carbapenem)–resistant nosocomial pathogen Stenotrophomonas maltophilia, and is an alternative to trimethoprim/sulfamethoxazole, or can be used synergistically with it in serious cases.⁴ Against most Acinetobacter species, which are usually resistant to multiple antibiotics, sulbactam possesses exceptional intrinsic activity and, like carbapenems, can be considered a drug of choice.^{1 ,4 ,20}

The emergence of extended-spectrum β-lactamase (ESBL)–producing bacteria is a global problem. The most common of these are E coli and Klebsiella species; 10% to 20% of strains produce ESBL, and the frequency appears to correlate with the extent of prior use of extended-spectrum cephalosporins.^{1 ,21} Extended-spectrum β-lactamases are believed to have mutated from a variety of plasmid-mediated penicillinases and can inactivate third-generation cephalosporins such as ceftazidime, cefotaxime, and ceftriaxone as well as monobactams such as aztreonam.^{1 ,21} Since such enzyme production often coexists with resistance to aminoglycosides and quinolones, carbapenems are the most reliable drug to eradicate these organisms.²² Sulbactam, tazobactam, or clavulanate may inhibit some of these mutated enzymes but are not generally recommended in cases of known ESBL infection because other resistance mechanisms may coexist.^{1 ,4 ,22} Some chromosomally encoded enzymes may also exhibit such extended-spectrum activity, and in this scenario too only carbapenems are clinically useful.^{1 ,3} Recently, however, mutated inhibitor-resistant enzymes have been identified.²³

Nevertheless, most β-lactam/β-lactamase inhibitor combinations have little propensity to select for strains producing specific chromosomally mediated β-lactamases or ESBLs, which could be of importance in controlling the emergence of resistant strains. Amoxicillin/clavulanate appears less likely to select PRSP than do cephalosporins, macrolides, or trimethoprim/sulfamethoxazole.²⁴ Moreover, higher rates of piperacillin/tazobactam use and lower rates of third-generation cephalosporin use can significantly decrease the prevalence of ESBL-producing organisms.²¹ The rates of C difficile diarrhea and rectal carriage of glycopeptide-resistant enterococci can be reduced with similar measures.^{25 - 26} In addition to overall antibiotic restrictions, antibiotic rotation policies involving β-lactam/β-lactamase inhibitor combinations have been proposed to reduce resistance rates in intensive care units.²⁷ Although β-lactam/β-lactamase inhibitor combinations may have less potential to select resistant bacteria, their profligate overuse should nonetheless be avoided so as to delay the inevitable emergence of bacterial resistance.