2025 年 48 巻 5 号 p. 571-576
Multidrug-resistant bacteria pose a major challenge in healthcare, and antimicrobial stewardship teams (ASTs) play a crucial role in optimizing antimicrobial use, particularly for last-resort drugs like meropenem (MEPM) and tazobactam/piperacillin (TAZ/PIPC). This study evaluated the impact of enhanced interventions, which included a hospital-wide policy restricting MEPM and TAZ/PIPC use to 5 d and mandating pre-treatment culture testing. A before-and-after study was conducted at a public hospital in Japan, comparing the pre- (June 2021–May 2022) and post-enhanced intervention (June 2022–May 2023) periods. The primary outcome was days of therapy (DOT) per 1000 patient days for MEPM and TAZ/PIPC. Secondary outcomes included antimicrobial use density (AUD), monthly number of patients receiving MEPM and TAZ/PIPC, 30-d mortality, and AST intervention proposals. Overall, 1896 patients received MEPM (pre: 591; post: 527) or TAZ/PIPC (pre: 411; post: 367). As a result, MEPM DOT decreased from 19.4 to 17.2 per 1000 patient days (p = 0.019), and AUD from 14.4 to 11.7 defined daily doses per 1000 patient days (p = 0.017). TAZ/PIPC DOT remained unchanged (p = 0.219), while AUD decreased from 8.7 to 7.6 (p = 0.046). Furthermore, the monthly number of patients receiving MEPM and TAZ/PIPC and their 30-d mortality showed no significant change. AST proposals increased from 209 to 359 for MEPM and from 116 to 238 for TAZ/PIPC (both p < 0.001). In conclusion, enhanced interventions effectively reduced MEPM use without increasing TAZ/PIPC use or worsening 30-d mortality, suggesting that structured guidelines may enhance antimicrobial stewardship in resource-limited settings.
The increase in the prevalence of multidrug-resistant (MDR) bacteria has made the development of effective antimicrobial therapy increasingly challenging.1) Carbapenems, a class of antibiotics with a broad spectrum of activity, have been demonstrated to be a safe and effective treatment option for severe infections, with a particular effect against Gram-negative bacteria and stability against resistance mechanisms commonly associated with many MDR bacteria, such as the production of extended-spectrum β-lactamases and AmpC β-lactamases.2,3) Tazobactam/piperacillin (TAZ/PIPC), a broad-spectrum antibiotic, is also an essential option for treating many MDR bacterial infections.4)
However, there is a positive correlation between the use of these drugs and the emergence of drug-resistant bacteria,5) underscoring the need for prudent use. Carbapenems, in particular, should be reserved as a last-resort treatment, and the effectiveness of carbapenem-sparing agents, such as TAZ/PIPC and cefmetazole, has been reported.6,7)
Guidelines for antimicrobial stewardship strongly recommend implementing restriction policies and prospective audit and feedback (PAF) to promote appropriate antimicrobial use.8) Restriction policies, in particular, have been reported as a powerful tool for reducing the use of specific antimicrobials. However, in Japan, the implementation of restriction policies is limited to only 19.2% of facilities, primarily because of insufficient human resources.9) Therefore, methods that can reduce antimicrobial use without the need for restrictive measures are required.
At our hospital, pharmacists have been conducting PAF-based interventions for all injectable antibiotics, contributing to a reduction in the length of hospital stay, the promotion of appropriate drug use, and a decrease in the detection rate of drug-resistant bacteria.10) Although the use of carbapenems and TAZ/PIPC initially decreased significantly following these interventions,11) a slight change occurred thereafter, with TAZ/PIPC use showing an increasing trend. In response, the antimicrobial stewardship team (AST) led the establishment of hospital compliance guidelines specifically targeting carbapenems (especially meropenem [MEPM]) and TAZ/PIPC, strengthening the intervention framework. This intervention aimed to further promote the appropriate use of antibiotics by limiting the duration of use to 5 d and mandating culture testing before initiating treatment. However, these new approaches have not been thoroughly examined in previous studies. Therefore, the present study aimed to evaluate the effectiveness of enhanced interventions led by the AST on MEPM and TAZ/PIPC use, in addition to the existing PAF interventions for all injectable antibiotic users.
A before-and-after comparative study was conducted at a public hospital in Japan (Ogaki Municipal Hospital) between June 2021 and May 2023. Outcomes before and after enhancement, defined as the implementation of institutional guidelines and strengthened interventions to optimize antimicrobial use, were compared. The pre-enhancement period was defined as June 1, 2021, to May 31, 2022, and the post-enhancement period as June 1, 2022, to May 31, 2023. The patient selection flowchart is presented in Fig. 1. A total of 1943 hospitalized patients who received MEPM or TAZ/PIPC for the first time were included. After excluding 47 early deaths (<48 h), 1896 patients were analyzed. Data were collected retrospectively from electronic medical records and included information on sex, age, length of hospital stay, 30-d mortality from the start of the target antibiotic, antibiotics used, duration of therapy, and major infections. Additionally, data on AST round recommendations and acceptance rates, recommendations made by pharmacists, and the number of days until recommendations were made were also collected. Information on whether the target antibiotics were used as the initial treatment, the blood culture positivity rate, and the number of patients who were switched from the target antibiotics to other agents was included. Our hospital is an 817-bed facility that provides advanced acute-care services. The hospital does not have a dedicated infectious disease department. An opt-out option was made available to all patients, allowing them to decline participation in the study. This study was approved by the hospital’s ethics committee (Approval No.: 20230518-9).
MEPM, meropenem; TAZ/PIPC, tazobactam/piperacillin.
Before June 2022, antimicrobial stewardship relied on pharmacist-led PAF for all injectable antibiotics, which was maintained throughout the study. Pharmacists spent 7.2 h/d using a customized infection management system (BACT Web®; Eiken Chemical Co., Ltd., Tokyo, Japan). Pharmacists were initiated on the day after the start of injectable antibiotic therapy (or the next business day if the following day was a holiday).11) These reviews assessed antibiotic appropriateness based on pathogenic microorganisms, infection sites, renal function, and pharmacokinetics/pharmacodynamics. Recommendations for de-escalation, switching to oral antibiotics, or therapy adjustments were communicated to physicians via phone or electronic medical records. Problematic cases identified in PAF reviews were discussed twice a week during AST rounds on Tuesdays and Fridays with a multidisciplinary team consisting of physicians, pharmacists (Board-Certified Infection Control Pharmacy Specialists), microbiology technicians, and nurses. The discussions focused on chart reviews and electronic recommendations. The increasing use of MEPM and TAZ/PIPC became problematic due to the lack of institutional policies on pre-treatment culture testing or therapy duration. In June 2022, AST guidelines mandated pre-treatment cultures and restricted therapy to 5 d unless extended by AST (Table 1). Compliance was monitored during AST rounds, and deviations—such as failure to conduct pre-treatment cultures with MEPM or TAZ/PIPC, or extending therapy beyond 5 d without AST approval—were documented. Compliance rates were shared in monthly hospital meetings to enhance staff awareness and adherence.
| Please implement the following items when using empirically: |
|
|
| In principle, the period of administration should not exceed 5 d from the date of administration, and the following should be performed after the start of administration. |
|
|
MEPM, meropenem; TAZ/PIPC, tazobactam/piperacillin; AST, antimicrobial stewardship team.
The primary outcome was the monthly days of therapy (DOT) per 1000 patient days for MEPM and TAZ/PIPC, comparing the pre- and post-enhancement periods. The secondary outcome, monthly antimicrobial use density (AUD) per 1000 patient days, was calculated using defined daily doses (DDD) based on Japan Surveillance for Infection Prevention and Healthcare Epidemiology standards (https://amrcrc.ncgm.go.jp/surveillance/030/20181128172757.html, accessed February 1, 2025). Additionally, evaluations included the monthly number of patients receiving these antibiotics, 30-d mortality rates, and AST intervention proposals.
Statistical AnalysisCategorical and continuous variables were analyzed using the chi-square test and Mann–Whitney U test, respectively, with the significance level set at <5%. Statistical analyses were performed using EZR software (v. 1.55).12)
The number of patients treated with MEPM decreased from 591 to 527 after the intervention enhancement (Table 2). Significant differences in sex and age were noted between the 2 periods, with no changes in 30-d mortality or hospital stay length. The proportion of patients switched from MEPM to other antibiotics rose significantly from 50.8 to 56.9% (p = 0.045). AST recommendations increased notably from 35.4 to 68.1% (p < 0.001), alongside a decrease in the median time to AST rounds from 5 to 4 d (p < 0.001). The acceptance rate of AST recommendations fell significantly from 87.1 to 75.5% after enhancement (p = 0.001), especially for discontinuation recommendations, which dropped from 78.0 to 54.2% (p = 0.003). No significant change was found in the acceptance rate of pharmacist recommendations before and after the enhancement.
| Before enhancement (N = 591) |
After enhancement (N = 527) |
p-Value | |||
|---|---|---|---|---|---|
| Male (%) | 382 | (64.6) | 302 | (57.3) | 0.014 |
| Age [IQR] | 75 | [67–82] | 73 | [62–82] | 0.014 |
| Number of deaths within 30 d of MEPM initiation (%) | 104 | (17.6) | 91 | (17.3) | 0.947 |
| Length of hospital stay [IQR] | 25 | [14–41] | 25 | [13–46] | 0.451 |
| Number of patients initially treated with MEPM (%) | 257 | (43.5) | 237 | (45.0) | 0.661 |
| Number of patients with positive blood cultures (%) | 195 | (33.0) | 183 | (34.7) | 0.584 |
| Patients switched from MEPM to other antibiotics (%) | 300 | (50.8) | 300 | (56.9) | 0.045 |
| Days until AST round [IQR] | 5 | [3–6] | 4 | [3–5] | < 0.001 |
| AST rounds, total number (%) | 209 | (35.4) | 359 | (68.1) | < 0.001 |
| Continue (%) | 72 | (12.2) | 115 | (21.8) | < 0.001 |
| Change (%) | 78 | (13.2) | 150 | (28.5) | < 0.001 |
| Discontinuation (%) | 58 | (9.8) | 94 | (17.8) | < 0.001 |
| Days until pharmacist intervention [IQR] | 3 | [2–5] | 2.5 | [2–4.75] | 0.004 |
| Pharmacist intervention, total number (%) | 118 | (20.0) | 138 | (26.2) | 0.016 |
| Change (%) | 68 | (11.5) | 78 | (14.8) | 0.123 |
| Dosage (%) | 34 | (5.8) | 48 | (9.1) | 0.042 |
| Discontinuation (%) | 14 | (2.4) | 11 | (2.1) | 0.908 |
| Culture request (%) | 2 | (0.3) | 1 | (0.2) | 1.000 |
| Days of administration of MEPM [IQR] | 6.0 | [4.0–8.0] | 5.0 | [4.0–8.0] | 0.761 |
| Hospital compliance policies | |||||
| Full adherence (%) | — | 356 | (67.5) | — | |
| Pre-antibiotic culturing (%) | 521 | (88.2) | 469 | (89.0) | 0.730 |
| Blood culture | 491 | (83.1) | 426 | (80.8) | 0.369 |
| Sputum culture | 161 | (27.2) | 120 | (22.8) | 0.099 |
| Urine culture | 265 | (44.8) | 209 | (39.7) | 0.091 |
| Other cultures | 144 | (24.4) | 145 | (27.5) | 0.258 |
| Antibiotic treatment duration within 5 d (%) | — | 397 | (75.4) | — | |
| Main infection (%) | 0.003 | ||||
| Sepsis (%) | 185 | (31.3) | 170 | (32.3) | |
| Pneumonia (%) | 98 | (16.6) | 65 | (12.3) | |
| Peritonitis/intra-abdominal abscess (%) | 63 | (10.7) | 94 | (17.8) | |
| Febrile neutropenia (%) | 63 | (10.7) | 40 | (6.7) | |
| Urinary tract infection (%) | 59 | (10.0) | 44 | (8.3) | |
| Hepatobiliary infection (%) | 30 | (5.1) | 43 | (8.2) | |
| Skin and soft tissue infection (%) | 27 | (4.6) | 20 | (3.8) | |
| Bacterial meningitis (%) | 12 | (2.0) | 6 | (1.1) | |
| Osteomyelitis/arthritis (%) | 12 | (2.0) | 4 | (0.8) | |
| Surgical site infection (%) | 11 | (1.9) | 10 | (1.9) | |
| Pleurisy/empyema (%) | 8 | (1.4) | 3 | (0.6) | |
| Catheter-related bloodstream infection (%) | 7 | (1.2) | 11 | (2.1) | |
| Other (%) | 16 | (2.7) | 17 | (3.2) | |
MEPM, meropenem; IQR, interquartile range; AST, antimicrobial stewardship team.
In the study of TAZ/PIPC, patient numbers dropped from 411 to 367 following the intervention enhancement (Table 3). Despite a significant difference in sex between the periods, mortality and hospital stay duration remained consistent. The proportion of patients undergoing AST intervention rose sharply from 28.2 to 64.9% (p < 0.001), and the median time to AST rounds shortened from 5 to 4 d (p = 0.011). Nevertheless, the acceptance rate of AST recommendations fell from 86.2 to 73.1% (p = 0.009), with discontinuation recommendations dropping from 75.0 to 52.8% (p = 0.019). As with MEPM, there were no significant changes in the acceptance rate of pharmacist recommendations post-enhancement.
| Before enhancement (N = 411) |
After enhancement (N = 367) |
p-Value | |||
|---|---|---|---|---|---|
| Male (%) | 298 | (72.5) | 240 | (65.4) | 0.039 |
| Age [IQR] | 75 | [68–82] | 76 | [70–83] | 0.657 |
| Number of deaths within 30 d of TAZ/PIPC initiation (%) | 74 | (18.0) | 82 | (22.3) | 0.156 |
| Length of hospital stay [IQR] | 22 | [13–34] | 22 | [14–35] | 0.794 |
| Number of patients initially treated with TAZ/PIPC (%) | 195 | (47.4) | 156 | (42.5) | 0.190 |
| Number of patients with positive blood cultures (%) | 57 | (13.9) | 57 | (15.5) | 0.580 |
| Patients switched from TAZ/PIPC to other antibiotics (%) | 147 | (35.8) | 140 | (38.1) | 0.540 |
| Days until AST round [IQR] | 5 | [3–7] | 4 | [3–5] | 0.011 |
| AST rounds, total number (%) | 116 | (28.2) | 238 | (64.9) | < 0.001 |
| Continue (%) | 42 | (10.2) | 78 | (21.3) | < 0.001 |
| Change (%) | 26 | (6.3) | 71 | (19.3) | < 0.001 |
| Discontinuation (%) | 48 | (11.7) | 89 | (24.3) | < 0.001 |
| Days until pharmacist intervention [IQR] | 3 | [2–4] | 3 | [2–4] | 0.346 |
| Pharmacist intervention number (%) | 69 | (16.8) | 64 | (17.4) | 0.885 |
| Change (%) | 38 | (9.2) | 37 | (10.1) | 0.785 |
| Dosage (%) | 14 | (3.4) | 16 | (4.4) | 0.615 |
| Discontinuation (%) | 8 | (1.9) | 6 | (1.6) | 0.955 |
| Culture request (%) | 9 | (2.2) | 5 | (1.4) | 0.551 |
| Days of administration of TAZ/PIPC [IQR] | 6.0 | [4.0–8.0] | 6.0 | [4.0–8.0] | 0.100 |
| Hospital compliance policies | |||||
| Full adherence (%) | — | 218 | (59.4) | — | |
| Pre-antibiotic culturing (%) | 338 | (82.2) | 302 | (82.3) | 1.000 |
| Blood culture | 278 | (67.6) | 247 | (67.3) | 0.981 |
| Sputum culture | 148 | (36.0) | 169 | (46.0) | 0.006 |
| Urine culture | 134 | (32.6) | 110 | (30.0) | 0.476 |
| Other cultures | 97 | (23.6) | 51 | (13.9) | < 0.001 |
| Antibiotic treatment duration within 5 d (%) | — | 270 | (73.7) | — | |
| Main infection (%) | 0.017 | ||||
| Pneumonia (%) | 163 | (39.7) | 176 | (48.0) | |
| Peritonitis/intra-abdominal abscess (%) | 110 | (26.8) | 82 | (22.3) | |
| Sepsis (%) | 35 | (8.5) | 40 | (10.9) | |
| Hepatobiliary infection (%) | 30 | (7.3) | 17 | (4.6) | |
| Urinary tract infection (%) | 27 | (6.6) | 14 | (3.8) | |
| Febrile neutropenia (%) | 21 | (5.1) | 9 | (2.5) | |
| Skin and soft tissue infection (%) | 11 | (2.7) | 9 | (2.5) | |
| Pleurisy/empyema (%) | 4 | (1.0) | 12 | (3.3) | |
| Surgical site infection (%) | 4 | (1.0) | 1 | (0.3) | |
| Osteomyelitis/arthritis (%) | 3 | (0.7) | 2 | (0.5) | |
| Other (%) | 3 | (0.7) | 5 | (1.4) | |
TAZ/PIPC, tazobactam/piperacillin; IQR, interquartile range; AST, antimicrobial stewarddship team.
MEPM DOT significantly decreased from 19.4 to 17.2/1000 patient days (p = 0.019), while AUD decreased from 14.4 to 11.7 DDD/1000 patient days (p = 0.017). The monthly number of MEPM-treated patients showed no significant change (57.0–47.5, p = 0.125) (Table 4a).
| Before enhancement | After enhancement | p-Value | |||
|---|---|---|---|---|---|
| (a) Meropenem | |||||
| Days of therapy* [IQR] | 19.4 | [17.0–20.3] | 17.2 | [14.1–18.4] | 0.019 |
| Antimicrobial use density** [IQR] | 14.4 | [12.6–15.5] | 11.7 | [9.8–13.3] | 0.017 |
| Number of patients using antibiotics [IQR] | 57.0 | [52.3–61.3] | 47.5 | [45.8–58.0] | 0.125 |
| (b) Tazobactam/piperacillin | |||||
| Days of therapy* [IQR] | 12.2 | [11.0–13.1] | 10.5 | [9.5–11.9] | 0.219 |
| Antimicrobial use density** [IQR] | 8.7 | [8.2–9.9] | 7.6 | [6.5–8.5] | 0.046 |
| Number of patients using antibiotics [IQR] | 38.0 | [33.0–49.2] | 36.5 | [33.5–40.5] | 0.506 |
IQR, interquartile range. * Days of therapy indicate per 1000 patient days. ** Antimicrobial use density indicates defined daily doses per 1000 patient days.
TAZ/PIPC DOT decreased from 12.2 to 10.5/1000 patient days (p = 0.219). Furthermore, AUD significantly decreased from 8.7 to 7.6 DDD/1000 patient days (p = 0.046). The monthly number of TAZ/PIPC-treated patients showed no significant change (38.0–36.5, p = 0.506) (Table 4b).
Here, the hospital implemented additional interventions by the AST through the establishment of in-hospital guidelines and feedback mechanisms for the use of MEPM and TAZ/PIPC, in addition to the already ongoing pharmacist-led PAF for all injectable antibiotics. These interventions reduced MEPM use without worsening 30-d mortality or increasing TAZ/PIPC use.
Our study revealed a significant reduction in MEPM use following the enhancement of AST interventions. While the treatment duration with MEPM remained stable, the DOT decreased. As there was no significant change in the rate of culture submission before treatment initiation, this reduction in DOT was unlikely due to changes in culture results. Rather, it was likely attributable to new hospital guidelines, stringent AST monitoring, more frequent switches from MEPM to alternative agents, and a decline in the number of patients receiving MEPM. According to previous studies, implementing antibiotic restriction policies can reduce antibiotic use and improve antibiotic susceptibility rates.8) Additionally, setting specific criteria for antibiotic use and requiring pre-approval can decrease usage13); however, our study is novel in that it revealed a reduction in usage without requiring pre-approval by establishing specific in-hospital guidelines. Umemura et al. reported that increasing the duration of PAF led to a reduction in DOT for broad-spectrum antibiotics without affecting mortality rates.14) In our study, MEPM use decreased without changing the duration of PAF or affecting 30-d mortality rates. Notably, among the 600 patients who switched from MEPM to other agents, those who died within 30 d were 45 before enhancement and 34 after enhancement, showing no significant difference (p = 0.227). This suggests that the enhanced intervention successfully reduced MEPM use without adversely affecting mortality, making our approach potentially applicable in settings with limited human resources and serving as a useful model for other healthcare institutions.
However, despite the significant reduction in the AUD of TAZ/PIPC, no significant change in DOT was observed. This could be due to the already relatively low use of TAZ/PIPC in our hospital compared with the national average before the enhancement, which posed further challenges in reducing its usage.15) Additionally, approximately 48% of TAZ/PIPC usage at our hospital was for pneumonia treatment, and as recommended in the 2024 Adult Pneumonia Treatment Guidelines in Japan, the clinical necessity of TAZ/PIPC for treating severe pneumonia might have contributed to the difficulty in further reducing its usage. Meanwhile, the submission rate of sputum cultures increased after the intervention. Given that TAZ/PIPC was primarily used in patients with pneumonia, the increased sputum culture submission likely contributed to improved pathogen identification, thereby facilitating more appropriate treatment decisions.
This study has several limitations. First, it was a single-center, short-term, before-and-after study, which made it difficult to completely eliminate the effects of differences in observation periods owing to the limited number of cases. Therefore, a long-term, multicenter, prospective study is desirable. Second, differences in patient backgrounds were a limitation. As the intervention was conducted among patients using MEPM and TAZ/PIPC, it was impossible to strictly standardize the recommended treatment durations for different diseases before and after the intervention; thus, the patient background may have influenced the findings. Future studies should stratify patients based on specific characteristics, such as age, comorbidities, and infection severity, to better understand the impact of these variables on the effectiveness of AST interventions. Furthermore, while the blood culture positivity rate may have influenced the present study results, no significant changes were observed before and after the intervention. It is unlikely that pathogen identification substantially influenced the decision to switch from MEPM or TAZ/PIPC to other antimicrobial agents. Although this intervention successfully reduced MEPM use without increasing TAZ/PIPC use or worsening mortality rates, further optimization of TAZ/PIPC use remains challenging. Future interventions should include the formulation of clear, disease-specific clinical pathways and antibiotic selection criteria, particularly for pneumonia and intra-abdominal infections. Additionally, routine visualization and feedback regarding antibiotic usage and adherence to guidelines should be implemented.
The implementation of enhanced AST interventions, including strict guidelines for MEPM and TAZ/PIPC usage, was effective in reducing MEPM use without the need for additional human resources, worsening 30-d mortality, or increasing TAZ/PIPC use.
We are grateful to all of those who contributed to the antimicrobial stewardship program at our hospital.
This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
All the authors met the ICMJE authorship criteria. KO, TM, YS, KA, NH, TT, HS, MK, and TY conceived and designed the study. KO, TM, and YS analyzed the data and determined the structure and clinical significance of the manuscript. KO, YS, KA, and NH collected and collated the data. KO drafted the manuscript. TT, HS, MK, TY, and EU provided critical comments and revised the manuscript accordingly. All the authors approved the final version of the manuscript.
The authors declare no conflict of interest.
