Breaking Antimicrobial Resistance: High-Dose Amoxicillin with Clavulanic Acid for Urinary Tract Infections Due to Extended-Spectrum Beta-Lactamase (ESBL)-Producing

Background

Infections with extended-spectrum beta-lactamase (ESBL)-producing Enterobacteriaceae strains are a growing therapeutic problem worldwide [1,2]. The spread of these bacteria makes it a significant etiological factor for urinary tract infections (UTIs) among patients after kidney transplantation [3]. UTIs caused by ESBL-producing Enterobacteriaceae are diagnosed in approximately 10% of kidney graft recipients within the first year after transplantation [3]. They are among the most important risk factors for poor kidney graft survival [4,5]. ESBL-producing Klebsiella pneumoniae is one of the most often isolated strains causing UTIs in immunodeficient patients [1]. Currently, treatment for such cases consists only of carbapenems combined with an aminoglycoside, colistin, or fosfomycin [6,7]. Such treatment requires the long-term use of intravenous antibiotics, frequently combined with prolonged hospitalization, high costs, and infection recurrence [8]. The World Health Organization has placed ESBL-producing Enterobacteriaceae on the priority list for developing new effective drugs [9]. ESBL-producing K. pneumoniae strains are usually resistant to aminopenicillins combined with beta-lactamase inhibitors. However, findings show that using larger-than-recommended doses of aminopenicillin can be effective in certain clinical settings [10,11]. Aminopenicillins have an excellent safety profile and can be used at doses many times higher than those recommended for UTIs [12]. Amoxicillin is well tolerated; the highest recommended dosage is 6.0 g/day for treating late-stage Lyme disease.

This report presents the treatment results with maximal therapeutic oral doses of amoxicillin combined with average doses of clavulanic acid in a cohort of patients with UTIs caused by ESBL-producing K. pneumoniae strains.

Material and Methods

This observational cohort study included patients with UTIs caused by ESBL-producing K. pneumoniae, all of whom had pyuria and positive urine cultures. The study was approved by the Medical University of Warsaw Ethics Committee (consent number KB/208/2017). We excluded patients with severe UTIs who, in the investigator’s opinion, required hospitalization (eg, patients with kidney pain, fever, C-reactive protein levels >20 mg/dl, or decreased renal function (estimated glomerular filtration rate [eGFR] <15 ml/min/1.73 m2 or >50% increase in eGFR compared with the previous visit). The treatment protocol consisted of a large oral dose of amoxicillin combined with a standard daily dose of clavulanic acid down-titrated over a 90-day therapy period. The starting dosage was 2875 g of amoxicillin twice daily and 125 mg of clavulanic acid twice daily (amoxicillin/clavulanic acid tablets 875 mg/125 mg twice daily with two 1000-mg amoxicillin tablets twice daily) for the patients with an eGFR >30 ml/min/1.73 m2. The total daily dose in subjects with body weight >40 kg was 5750 mg. Doses exceeding 3000 mg/day are not recommended in children with body weight <40 kg. The patients with an eGFR of 15–30 ml/min/1.73 m2 received a starting dosage of 1875 g of amoxicillin twice daily and 125 mg of clavulanic acid twice daily. All patients were adults with a body weight of >40 kg; therefore, we did not adjust the dosage for kilograms of body weight. We down-titrated the doses every 7–14 days (Table 1). We checked blood tests, urinalysis, and urine cultures 7 and 14 days after the start of therapy, then every 3–4 weeks during the therapy, and 2 and 4 weeks after the end of the therapy. Midstream urine samples were collected, transported to the laboratory within 2 h, and cultured on MacConkey agar and CPS agar (bioMerieux, Marcy l’Etoile, France). The plates were incubated at 37°C for 24 h under aerobic conditions. Bacterial identification and antimicrobial susceptibility were performed using the Vitek2 system (bioMerieux). ESBL strains were identified by means of the DDST test, in Müeller-Hinton medium, plating the strain with a density of 0.5 McFarland, it uses ceftazidime (30 μg) and cefotaxime (30 μg) discs, placed 2 cm from the center of the disc containing amoxicillin/clavulanic acid (30 μg). To increase the sensitivity of the test and to determine the AmpC mechanism, an additional cefepime disc (30 μg) was placed. A positive result is evidenced by the appearance of the enlargement of the growth inhibition zone around the ceftazidime and/or cefotaxime disc from the side of the disc containing clavulanic acid (beta-lactamase inhibitor). We defined therapeutic failure as the presence of ESBL-positive K. pneumoniae in urine culture during therapy and recurrence as a positive urine culture with the same strain 1 month after treatment completion.

Results

The study included 9 patients (4 women, 5 men), including 7 kidney graft recipients, 1 liver graft recipient, and 1 patient with chronic kidney disease. Seven patients had a previous history of UTIs caused by ESBL-producing K. pneumoniae treated with carbapenems in the study group. One patient had a history of a short 4-week treatment with high-dose amoxycillin/clavulanate and she was retreated with the full 3-month protocol. Still, only 1 patient had the first episode of UTI caused by these bacteria. The intestinal carriage of ESBL-producing K. pneumoniae was positive in 4 patients, and 4 patients had a history of urinary tract abnormalities (Table 2).

Table 3 shows the urine culture results. The UTI symptoms of all patients resolved within the first week of therapy. The urine cultures became negative after 7 days and remained negative for all patients throughout the treatment period and 1-month follow-up period, during which we observed no recurrence of the infection. All patients were followed-up for 6 months after the antibiotic therapy. In 1 case, the patient experienced a subsequent UTI caused by ESBL-producing K. pneumoniae 4 months after the end of therapy; the patient was hospitalized with fever and impaired graft function and underwent 3 weeks of meropenem therapy. There was no significant change in the GFR during the 6-month follow-up period (mean eGFR of 38.5 vs 35.4 ml/min/1.73 m2 before and at the end of follow-up, respectively).

Discussion

Our study showed that prolonged therapy with large starting doses of oral amoxicillin with clavulanic acid is a highly effective and safe alternative therapy for cases of UTIs caused by ESBL-producing K. pneumoniae. This therapy demonstrated 100% effectiveness, despite bacterial strains resistant to amoxicillin/clavulanic acid in vitro. The main reason we attempted such therapy was our patients’ unsatisfactory results of previous treatments. To date, patients in our center with UTIs caused by ESBL-producing K. pneumoniae, regardless of their clinical condition, required hospitalization due to the need for intravenous carbapenems, which have been considered front-line therapy, both in the general population and in kidney graft recipients [6,7]. This therapy has been highly effective and has led to clinical improvement and sterile urine cultures after only a few days of therapy. Initially, we used a 14-day therapy; however, we observed a very high infection recurrence rate, especially in the kidney transplant recipients, who required subsequent hospitalizations. Extending the therapy to 21 days and adding aminoglycosides reduced the frequency of infection recurrence; however, several patients still required subsequent hospitalizations. Managing these infections is challenging, with a limited number of available antibiotics and limited evidence supporting the therapies’ effectiveness. The optimal regimen and duration of therapy in the case of UTIs due to ESBL-producing K. pneumoniae remain undefined. Carbapenems and aminoglycosides are usually the only option but require prolonged hospitalization. One of its serious clinical problems is the high recurrence rate, reaching over 50% despite the therapy’s effectiveness [13]. Pouch et al reported treatment efficacy with different antibiotics (polymyxin B, gentamycin, amikacin, cefepime, and tobramycin), with a median therapy duration of 15 days. The authors observed a higher rate of infection clearance with aminoglycoside-containing regimens (100%), but within 3 months 40% of the patients experienced UTI recurrence. However, due to the nephrotoxicity and ototoxicity, the aminoglycoside regimen is disqualified from routine use, particularly in elderly patients and patients with chronic kidney disease, including post-transplantation patients. The effectiveness of other therapeutic regimens was lower [14].

Oral fosfomycin might also be a successful treatment for UTIs caused by multidrug-resistant bacteria in kidney transplant recipients. In a multicenter study from Spain, therapy with oral fosfomycin resulted in microbiological cure in 70% of cases 1 month after the therapy. However, only 14% of the isolates were ESBL-producing bacteria [15]. Moreover, oral fosfomycin is not recommended by the European Committee on Antimicrobial Susceptibility Testing to treat complicated UTIs. Most of our patients were kidney transplant recipients with urinary tract abnormalities.

A group from Spain recently published a multicenter study showing that therapy with beta-lactam/beta-lactamase inhibitors is as effective as carbapenem-based therapy in bacteremia secondary to UTI in kidney transplant recipients. However, these results are only applicable to piperacillin-tazobactam because only 2 patients in the study group underwent therapy with amoxicillin-clavulanic acid [16]. Amoxicillin, like most penicillins, can be used in large doses without significantly increasing the risk of adverse effects. The use of large doses of amoxicillin in treating otitis media was not associated with more frequent adverse effects, including diarrhea, compared with standard doses [17]. The results of a prospective study with 51 pediatric patients at a poison control center suggest that doses smaller than 250 mg/kg of amoxicillin are not associated with significant clinical symptoms [10]. Large doses of amoxicillin are effective in treating drug-resistant strains of pneumococcal carriage and community-acquired pneumonia and even as dual therapy with esomeprazole to eradicate Helicobacter pylori [8].

There has also been concern that conventional doses of beta-lactams might not always achieve adequate pharmacokinetic-pharmacodynamic indices [18]. Dose adjustment is necessary for chronic kidney disease with an eGFR <30 mL/min. Patients should maintain adequate fluid intake and urinary output to reduce the possibility of amoxicillin crystalluria. Cases of hepatoxicity with amoxicillin/clavulanic acid have been reported, but these were attributed to the adverse effects of the beta-lactamase inhibitor and not to the antibiotic itself [19]. Amoxicillin is well tolerated; the maximal recommended therapeutic dose is 6.0 g/day in adults and children with body weight ≥40 kg and 3.0 g/day in subjects with body weight <40 kg. The maximal recommended therapeutic dose in this subgroup is 100 mg/kg body weight/day.

Overuse of carbapenems has led to a growing problem of increasing numbers of carbapenem-resistant K. pneumoniae strains such as outer membrane protein mutations and the selection of beta-lactamases able to hydrolyze carbapenems [20,21]. A recent study using a transmission model based on antibiotic consumption and demographic data from 11 European countries showed that reducing nosocomial transmission and hospital antibiotic consumption had the strongest impact on the prevalence of ESBL/carbapenem-resistant bacterial strains in the population [22]. The outpatient use of amoxicillin/clavulanic acid instead of the inpatient use of carbapenems for treating UTIs caused by ESBL-producing K. pneumoniae might have an additional benefit in reducing the number of carbapenem-resistant bacteria.

The long-term use of amoxicillin/clavulanic acid in prophylactic doses in our study might be controversial due to the previously observed recurrence of infection despite 2–3 weeks of therapy with intravenous antibiotics, which might indicate this bacterium’s extensive ability to survive despite the use of potentially effective therapy. The high recurrence rate in kidney transplant patients might be caused by immunosuppressive therapy and concomitant urinary outflow disorders resulting from the kidney graft itself [23]. However, it appears that certain properties of K. pneumoniae bacteria might enable the bacterium to survive in the urinary tract despite correct and initially effective therapy and then cause reinfection shortly after the end of therapy.

In recent years, an increase has been observed in the frequency of infections caused by hypervirulent strains of K. pneumoniae especially in immunocompromised patients. The cases are related to genetic mutations that produce a hypercapsule consisting of a mucoviscous polysaccharide coating generating bacterial hypermucoviscosity [24]. Such strains are more resistant to opsonization and phagocytosis by immune cells, allowing them to survive longer in tissues. Hypervirulent strains of K. pneumoniae can produce fimbriae, which are important factors for bacterial adhesion to epithelial cells and biofilm formation [25,26]. Colonization of the gastrointestinal tract by K. pneumoniae strains is another significant cofactor of UTIs recurrence [27].

Our study has certain limitations, given that we studied a small group of patients and that a control group was not included. An additional limitation is the lack of data on the sensitivity of K. pneumoniae strains to fosfomycin. This is due to the lack of EUCAST recommendations regarding the MIC values for K. pneumoniae strains causing UTI at the time of the study. Nevertheless, the resistance in K. pneumoniae strains was overcome with large doses of amoxicillin with clavulanic acid in all studied cases, despite the apparent in vitro insensitivity of the bacteria to this drug. This means that the high efficacy of this therapy observed in vivo differs significantly from that expected, based solely on the culture result. Importantly, most of our study patients experienced a recurrence of the infection despite previous therapy for several sessions. It is also worth mentioning that the success of the therapy reported in this study was the basis for starting the use of large doses of amoxicillin in kidney transplant patients with recurrent ESBL-producing K. pneumoniae at our center.

Conclusions

It is possible to break the resistance of ESBL-producing K. pneumoniae strains by using the maximal therapeutic doses of oral amoxicillin with clavulanic acid. Such treatment could be a safe and inexpensive alternative to carbapenems in selected cases of UTI.