Evaluation of intravenous amoxicillin-clavulanate use in two Canadian hospitals

We describe our experience with intravenous amoxicillin-clavulanate, which is new to the Canadian market. The majority of patients were successfully de-escalated from piperacillin-tazobactam or a carbapenem for respiratory infections or skin and soft tissue infections. Intravenous amoxicillin-clavulanate provides a good alternative in an era of rising Pseudomonas aeruginosa resistance.

and feedback, recommendations for de-escalation to IV amoxicillin-clavulanate were incorporated into daily practice following formulary addition.
All admitted adult patients on IV amoxicillin-clavulanate were included.We used AMS software (Lumed, Sherbrooke, Canada) to identify patients who received IV amoxicillin-clavulanate, and collected demographic and clinical data via the electronic medical record.Primary outcomes included appropriateness of indication and duration of IV amoxicillin-clavulanate and clinical success.Appropriate indications for IV amoxicillin-clavulanate were based on provincial guidelines (Supplemental Table 1).Transition to oral therapy was assessed as part of appropriateness of indication on day 3 of IV amoxicillin-clavulanate. Appropriate duration of therapy was based on locally approved guidelines.Clinical success was defined as not requiring re-escalation to piperacillintazobactam or a carbapenem during hospitalization within 30 days of IV amoxicillin-clavulanate. Secondary outcomes included readmission rate, reason(s) for readmission, and mortality rate within 30 days of stopping IV amoxicillin-clavulanate. Patients who received less than 24 hours of IV amoxicillinclavulanate were excluded from evaluation of clinical success and secondary outcomes.Only the first episode of IV amoxicillinclavulanate use was evaluated for primary and secondary outcomes.An AMS pharmacist reviewed all cases, with verification by an ID/AMS physician for cases with ambiguous outcomes.Descriptive statistics were used for analysis.A letter of exemption from the institutional research ethics board was obtained.

Results
One hundred and thirteen charts were identified.Two patients were excluded as they did not receive IV amoxicillin-clavulanate per the medication administration record.Patient characteristics, indications, type of prescriber, duration, and reasons for choosing IV amoxicillin-clavulanate are summarized in Table 1.
Respiratory tract (complex parapneumonic effusions and empyema), and skin or soft tissue infections (diabetic foot infections and post-operative infections) were the most common indications.IV amoxicillin-clavulanate was used for de-escalation from piperacillin-tazobactam or a carbapenem in 59% of cases.
Primary and secondary outcomes are summarized in Table 2. IV amoxicillin-clavulanate was prescribed appropriately in 95% of cases, of which 12% were based on AMS recommendations.IV amoxicillin-clavulanate was the only option in one-third of the cases (eg culture grew Enterococcus species), whereas ceftriaxone with or without metronidazole could have been used in the remaining cases.The main reason for choosing IV amoxicillinclavulanate in those cases was to facilitate subsequent direct oral transition.ID physicians and hospitalists were the main prescribers.Reasons why transition to oral amoxicillin-clavulanate was not feasible are shown in Table 2.
Clinical success was reviewed in 97 patients, 13 patients received less than 1 day of therapy and were excluded.One patient was lost to follow-up.Clinical success was achieved in 72% of patients, and IV amoxicillin-clavulanate was well tolerated.Clinical deterioration while on treatment was the main reason for re-escalation to other broad-spectrum antibiotics.

Discussion
There is a scarcity of literature to describe the use of IV amoxicillinclavulanate in Canada.A pilot study in Alberta evaluated IV amoxicillin-clavulanate as an alternative to piperacillin- tazobactam for general surgery patients.However, the uptake by surgeons was low, which the authors attributed to the unfamiliarity of this new product. 7In contrast, we educated a broad range of prescribers (eg, hospitalists, respirologists, general surgeons) when IV amoxicillin-clavulanate was added to formulary, plus ongoing education via prospective audit and feedback.IV amoxicillin-clavulanate was used for de-escalation from piperacillin-tazobactam in 59% of patients.One can argue that ceftriaxone with or without metronidazole may also be used, but IV amoxicillin-clavulanate was the only choice in one-third of these patients due to culture susceptibility, drug allergy, etc.A number of these patients were also thought to have failed ceftriaxone prior to escalation to piperacillin-tazobactam.
Amoxicillin-clavulanate may have some preferable characteristics compared to ceftriaxone.Third-generation cephalosporins are more strongly associated with healthcare-associated Clostridioides difficile infection than beta-lactam-β-lactamase inhibitor combinations. 8Cephalosporins also have a stronger association than beta-lactam-β-lactamase inhibitor combinations with acquiring colonization with extended-spectrum β-lactamaseproducing Gram-negative bacilli. 9nother advantage of IV amoxicillin-clavulanate is that it allows for direct oral transition.The median duration of IV amoxicillinclavulanate was 5 days.Most patients requiring prolonged therapy (> 7 days) were reviewed by specialists (e.g., ID physicians, respirologists).A number of patients had empyema or complex parapneumonic effusions; these findings were similar to a study by Artoisenet et al. 10 ID specialists prescribed IV amoxicillin-clavulanate in 40% of cases.Their patients typically had complex skin and soft tissue infections, diabetic foot infections, or bone and joint infections; once deemed not to have risk factors for P. aeruginosa or other resistant organisms, they were de-escalated to IV amoxicillinclavulanate.
Clinical success was achieved in 72% of our patients.The reasons for broadening therapy were multifactorial including inadequate source control, patient risk factors, and prolonged hospitalization leading to growth of multidrug resistant organisms.Fifteen percent of patients were readmitted within 30 days of IV amoxicillin-clavulanate use; the majority were for reasons unrelated to the original infection.Mortality rate was 18%, mainly from cancer-related complications.
One strength of this study is inclusion of both a tertiary hospital and a smaller community hospital to compare their prescribing patterns of IV amoxicillin-clavulanate.We noted that all the inappropriate duration happened at the community hospital; this is possibly due to fewer ID physician and AMS pharmacist present at that site.A limitation is missing information for some patients due to the retrospective nature of this study.We did not evaluate the total days of piperacillin-tazobactam potentially saved by switching to IV amoxicillin-clavulanate. Since the cost of both IV antibiotics is similar, we do not expect substantial drug cost savings with the switch.
In an era of rising antibiotic resistance, IV amoxicillin-clavulanate provides a good alternative for patients who do not require piperacillin-tazobactam.Our study describes those who would benefit from this regimen: empirically for patients with communityacquired respiratory infections (eg empyema), complicated skin and soft tissue infections (eg diabetic foot infections), or when polymicrobial coverage is needed (eg intra-abdominal infections).

Table 1 .
Summary of patient demographics and prescribing pattern of intravenous (IV) amoxicillin-clavulanate

Table 2 .
Summary of primary and secondary outcomes a Does not add up to 55, as patient can have > 1 reason