Orally Administered Amoxicillin/Clavulanate: Current Role in Outpatient Therapy

This article is published with digital features, including a summary slide, to facilitate understanding of the article. To view digital features for this article go to https://​doi.​org/​10.​6084/​m9.​figshare.​13259639

Both amoxicillin and clavulanate show good oral absorption (about 60% oral bioavailability) [4]. Moreover, there is no pharmacokinetic interaction between amoxicillin and clavulanate. The fasted or fed state has minimal effect on the absorption and pharmacokinetics of amoxicillin. However, absorption of clavulanate in the fed state is greater relative to the fasted state.

Mean amoxicillin and clavulanate potassium pharmacokinetic parameters are shown in the Table 1. The half-life of amoxicillin and clavulanate after oral administration is 1.3 and 1 h, respectively. Both amoxicillin and clavulanate are low serum protein-bound drugs; 18% for amoxicillin and 25% for clavulanate. In general, amoxicillin and clavulanate are well distributed in body tissues. The mean concentrations in tracheal mucosa were 200% and 118% of the corresponding serum levels for amoxicillin and clavulanate respectively [5]. Unlike macrolides, fluoroquinolones and tetracyclines, amoxicillin and clavulanate do not attain high exposures in epithelial lining fluid (ELF) (ELF: unbound plasma exposure 0.35 for amoxicillin) [6]. Amoxicillin does not undergo appreciable metabolism and 50–85% of the drug is excreted in the urine as intact, 6 h after an oral dose. However, clavulanate is metabolized to a significant extent and approximately 25–40% of intact drug is excreted in urine after an oral dose [2, 7].

Being a β-lactam, the %f T > minimum inhibitory concentration (MIC) (proportion of time during which plasma unbound concentration exceeds MIC) is the PK/PD index driving the efficacy of amoxicillin. However, the PK/PD index of clavulanate in the presence of amoxicillin has not been yet studied. Since the PK/PD of β-lactamase inhibitors has been a subject of investigation in recent times (only after such investigations were undertaken for avibactam), it is therefore not surprising that no PK/PD information is available for clavulanate in the presence of amoxicillin. However, the PK/PD driver of clavulanate in the presence of another partner, ceftibuten, has been described. A 20.59% free T > 0.5 mg/L was found to be linked with a static effect in a neutropenic mice thigh infection model [8].

The PK/PD targets of stand-alone amoxicillin have been described but in limited studies. In an in-vitro kinetic model, a T > 50% was required for amoxicillin to exert maximal killing of penicillin-susceptible and penicillin-intermediate S. pneumoniae [9]. The EUCAST rationale document provides f T > MIC of 30–35% for Enterobacterales, 25–35% for S. pneumoniae and H. influenzae as PK/PD targets for these pathogens [10]. However, this rationale document for f T > MIC attainments is based on PK of intravenously administered (IV) amoxicillin. Employing the PK/PD target for Enterobacterales (f T > MIC of 30%), the Monte Carlo simulation of amoxicillin showed greater than 90% probability of target attainment (PTA) in plasma for MICs up to 2 mg/L in the 500 mg, q8h, IV dose regimen which worryingly drops to a mere 8% at the CLSI and the US Committee on Antimicrobial Susceptibility Testing (USCAST) susceptibility breakpoint of 8/4 (2:1 ratio MIC) mg/L (Table 2). Since the bioavailability of orally administered amoxicillin is approximately 60%, the PTA is expected to be even lower for orally administered amoxicillin. Cattrall et al. showed that even an oral dose of 1000 mg, q8h did not attain 90% PTA (target 32.5% f T > MIC) at 8 mg/L [11]. Considering these observations, the clinical utility of orally administered amoxicillin/clavulanate in treating serious infections such as pyelonephritis caused by Enterobacterales with MICs around 8 mg/L is questionable.

The European Committee on Antimicrobial Susceptibility Testing (EUCAST) Enterobacterales susceptibility breakpoint of 32 mg/L for orally administered amoxicillin/clavulanate (standard dose 500/125 mg, q8h) is applicable only for uncomplicated urinary tract infections (UTIs). It should be noted that EUCAST MIC breakpoints are based on amoxicillin MICs determined in the presence of fixed 2 mg/L.

In the case of lower respiratory infections, where S. pneumoniae is the primary causative pathogen, the efficacy is driven by the antibiotic concentrations in the epithelial lining fluid. Therefore, the PTA based on plasma exposures is not an appropriate method to judge the clinical potential. Nevertheless, multiple prospective or retrospectively collected clinical data established the clinical utility of amoxicillin/clavulanate for mild to moderate respiratory tract infections [12]. Moreover, unlike in the case of Enterobacterales, resistance to penicillins and other β-lactams remained very low in S. pneumoniae. Since β-lactamase-mediated resistance is not found with S. pneumoniae, clavulanate does not play a role in efficacy in pneumococcal infections.

The rationality behind the dose selection for a β-lactam/β-lactamase inhibitor depends on the tolerability vs the PK/PD requirement to achieve efficacy. In the case of amoxicillin/clavulanate, rather than PK/PD, clinical experience and dosing convenience guided the selection of dose regimens. As mentioned earlier, initially, the combination of amoxicillin/clavulanate was introduced as a 250/125 mg, q8h dosing regimen. To align with the standard amoxicillin dosage, the amoxicillin/clavulanate regimen of 500/125 mg, q8h was registered in Europe (in 1982) and the USA (in 1986) [2]. Over the years, the ratio of amoxicillin to clavulanate has varied to reflect prescribing needs, to improve convenience and to treat more severe infections or those caused by resistant organisms. Rather than the PK/PD, the reason to prescribe a twice-a-day regimen (q12h) instead of a thrice-a-day regimen (q8h) is given by the outcomes of the clinical studies which show satisfactory efficacy and safety with a twice-a-day regimen. The highest recommended dose, 875/125 mg, q12h is well tolerated albeit with diarrhoea being the common adverse reaction.

Table 3 shows the prescribing information (indication and usage and dose regimens) of amoxicillin/clavulanate as per the US Food and Drug Administration (FDA) and European Medicines Agency summary of product characteristics (SmPC).

In the case of uncomplicated and moderate urinary tract infections in the community, the empirical oral antibiotics nitrofurantoin, trimethoprim/sulfamethoxazole, fosfomycin, pivmecillinam, fluoroquinolones (levofloxacin, ofloxacin and ciprofloxacin) as well as amoxicillin/clavulanate and oral cephalosporins are prescribed. A susceptibility study for 2017 SENTRY surveillance of E. coli isolates collected from US patients with UTI showed 77.9% susceptibility to amoxicillin/clavulanate [13]. More published studies also point towards contemporary UTI-causing Enterobacterales showing reduced susceptibility to amoxicillin/clavulanate (Table 4). This phenomenon is part of the larger trend of rising prevalence of extended spectrum beta-lactamases (ESBLs) coupled with OXA-1. Moreover, TEM-1 hyperproduction has also been implicated in resistance to amoxicillin/clavulanate [22]. It should be also noted that, as a result of systemic metabolism, urinary concentrations of clavulanate could be low at the recommended oral dose of 125 mg. Furthermore, no PK/PD data is available to support that the detected urinary levels of clavulanate are adequate to restore the amoxicillin activity against ESBL isolates. All these observations need to be considered regarding the current utility of orally administered amoxicillin/clavulanate for the treatment of community UTI infections that do not require hospitalization [23‐28].

With regards to community respiratory infections, since penicillin resistance in S. pneumoniae and β-lactamase-negative, ampicillin-resistant (BLNAR) in H. influenzae is very low, amoxicillin/clavulanate continues to be a promising choice. Moreover, the benefit of adding clavulanate to amoxicillin is only limited to BLPAR H. influenzae and β-lactamase-positive M. catarrhalis. There are also published (but limited) animal PK/PD data supporting the efficacy of amoxicillin/clavulanate for these pathogens. The American Thoracic Society and Infectious Diseases Society of America recommend use of amoxicillin/clavulanate in combination with a macrolide or doxycycline for outpatient treatment of community-acquired bacterial pneumonia in adults with comorbidities [29]. It should be noted that amoxicillin/clavulanate is not active against cell-wall-lacking atypical pathogens which are involved in community-acquired bacterial pneumonia.

Amoxicillin/clavulanate is active against Bacteroides spp. including B. fragilis that express β-lactamases. It is also active against β-lactamase-expressing Fusobacterium spp. Further, the standalone amoxicillin covers Peptostreptococcus spp. However, there is no formal approval for use of amoxicillin/clavulanate for intra-abdominal infections on the FDA label [7].

S. aureus is the leading causative pathogen implicated in skin and structure infections. The β-lactamase-producing S. aureus are susceptible to amoxicillin/clavulanate while methicillin-resistant S. aureus (MRSA) are resistant. Therefore, amoxicillin/clavulanate can be used for the treatment of skin and skin structure infections caused by β-lactamase-producing strains of S. aureus [7].

There is a difference between CLSI and EUCAST methods in determining the MICs for amoxicillin/clavulanate. The CLSI recommends a 2:1 ratio method while EUCAST recommends use of fixed 2 mg/L clavulanate. As a result, for a given bacterial strain, amoxicillin/clavulanate MICs could be discordant between these two methods. Furthermore, as shown in the Table 5, the interpretive criteria differ between CLSI and EUCAST. Studies showed that the EUCAST method of determining MICs had better correlation with clinical outcome [30, 31]. While appropriateness of these methods is still debatable, until harmonization is established, clinicians should interpret the susceptibility test results on the basis of the method used to determine the MICs.

Another important aspect that requires consideration is that resistance to amoxicillin/clavulanate depends not only on the presence of β-lactamases genes in the organism but even the expression levels of these β-lactamases could affect the amoxicillin/clavulanate MICs [32]. The clinical implication of this phenomenon is that the organism might underexpress the β-lactamases in susceptibility testing and thus turns out to be susceptible but in patients it could hyperproduce the β-lactamases (as a result of increase in bla_TEM or bla_AmpC copy numbers) during therapy leading to clinical failure [32]. Such discrepancies should be borne in mind for organisms that show MICs within ± 1 doubling dilution of the susceptible breakpoint.

Compared to amoxicillin, oral cephalosporins such as cefixime, cefpodoxime and ceftibuten are less vulnerable to ESBLs and stable to OXA-1; therefore, they are promising options in combination with clavulanate for the treatment of UTIs [33]. Such combinations are not approved in the USA and Europe, but are available in India [16]. In-vitro studies showed that MICs of these oral cephalosporins in the presence of clavulanate against ESBL isolates were below their respective susceptibility breakpoints [8, 15, 34]. Another potentially clinically beneficial approach is to combine amoxicillin/clavulanate with cefpodoxime or ceftibuten (both show high urinary concentrations) [34, 35]. However, the benefit of clavulanate plus oral cephalosporin is limited to E. coli and K. pneumoniae since clavulanate is an inducer of chromosomal AmpC enzyme present in pathogens such as Enterobacter spp. and Citrobacter spp. [16].