Prescribing Trends of Fixed-Dose Combination Antibiotics Not Recommended by the WHO (FNRs) for ICU Patients in Six Major Areas of China During a Seven-Year Period Authors: Liujun Zhou, Yuhua Zhao, Jianping Zhu, Jieqiong Liu, Gang Liang, Yi Yang, Gang Han, Zhenwei Yu Published: 2024 ABSTRACT AbstractObjectiveTo evaluate the prescribing trends of fixed-dose combination antibiotics not recommended by the WHO (FNRs) for intensive care unit (ICU) patients in six major areas of China from 2013 to 2019.MethodsA descriptive analysis was conducted using the pharmacy prescription data. Prescription data for patients admitted to the ICU were extracted from the Hospital Prescription Analysis Cooperative Project. Trends in FNR use were analyzed over a seven-year period, and the trends were further analyzed at the specific drug and hospital levels.ResultsA total of 15,596,620 prescriptions were eligible for analysis, and 1,492,793 patients were included. Among these patients, 91,515 (6.13%) received FNRs. The annual number of ICU patients who received FNR showed an increasing trend (P=0.007), but the percentage per year did not (P=0.764). The FNR use was usually higher in male patients than in female patients (P<0.001). Patients aged > 60 years had the highest percentage of patients who received FNRs (P<0.001). Among the eight FNRs identified in this study, cefoperazone/sulbactam was the most commonly used FNR in both patient numbers and prescribed hospitals, followed by piperacillin/sulbactam. The use of cefotaxime/sulbactam was less common but showed an increasing trend. There were significant differences among the regions.ConclusionThis study investigated the national landscape of FNR use among ICU patients. Attention should be given to the frequent use of FNRs in these patients. Data on the real-world effectiveness and safety of FNRs are urgently required.Keywords: antibiotic use, fixed-dose combination, irrational, prescription, cefoperazone, sulbactam Introduction Antibiotics play a pivotal role in combating infectious diseases, necessitating their rational use. Severe infections are prevalent in patients admitted to intensive care units (ICUs), and antibiotics are commonly administered to these patients.1 The rational use of antibiotics is paramount for preserving drug efficacy, minimizing adverse effects, conserving medical resources, and safeguarding the health of patients.2–5 Conversely, inappropriate use of antibiotics has led to decreased effectiveness, emergence of drug-resistant bacteria, and increased costs.6,7 Therefore, rational use of antibiotics is crucial for ICU infection management.2,8,9 Among all antibiotics, fixed-dose combination antibiotics (FDCAs) have become increasingly widespread.10–12 These drugs combine multiple active ingredients in a fixed ratio to improve treatment convenience, compliance, and potential synergistic effects.13,14 However, pharmacokinetic mismatches, insufficient evidence-based support, individual differences and regulatory and compliance issues may render certain FDCAs inappropriate for use in specific clinical scenarios.3,15 The WHO listed 103 combinations of FDCAs not recommended (FNR) in the 2021 ‎AWaRe classification. These FNRs are not evidence-based or recommended by international guidelines, and concerns are raised about their efficacy, safety and emergence of antimicrobial resistance (AMR).10,16–18 Despite these potential disadvantages, multiple FNRs have been approved and widely used in various countries, including China.16,19 There is little data regarding the use of FNR in ICU patients, which is important for antimicrobial stewardship. This study aimed to provide national data on FNR use in ICU patients for a seven-year period utilizing a comprehensive prescription database, and the results will be helpful for clinical practice, pharmacovigilance, and regulation. Methods Study Design This study was designed as a retrospective descriptive study based on prescription data. Ethical approval was obtained from the Ethics Committee of Sir Run Run Shaw Hospital, School of Medicine, Zhejiang University (Reference Number KEYAN2023-0209). The requirement for informed consent was waived as part of the study. The patients’ personal information was confidential to the researchers, and the study was conducted in compliance with the Declaration of Helsinki. Data Source Prescription data were extracted from the database of the Hospital Prescription Analysis Cooperative Project in China. The database contains prescription information of participating hospitals on 40 randomized days per year, and has been widely used in Chinese pharmacoepidemiology studies.20–25 In this study, prescription data from 66 hospitals in Beijing, Hangzhou, Chengdu, Guangzhou, Shanghai, and Tianjin were selected because these hospitals participated in the program continuously from 2013 to 2019 and were located in the north, west, south, and east, thus covering a wide area of China. Brief hospital information is shown in Table S1. Prescription Inclusion and Data Extraction Prescriptions meeting the following criteria were included in the analysis: (1) prescriptions written during 2013 and 2019, (2) prescriptions from the aforementioned hospitals, and (3) prescriptions written for patients admitted to the ICU. The following information about prescriptions was extracted: patient code, sex, age, date, location, diagnosis, and the generic name and price of antibiotics. Prescriptions with missing data were excluded. Patient codes were reorganized in the dataset so that individual participants could not be identified. The study was conducted between October 2023 and January 2024. Analysis The main results were the yearly proportion of patients who received FNRs and their trends. Any drug within the FDCAs not recommended by the WHO was defined as an FNRs. The yearly number of ICU patients was calculated using the patient code in the extracted prescriptions, and the yearly number of patients who received FNR was calculated by counting the number of patients who had at least one FNR prescription. The characteristics of the patients who received FNRs are descriptively presented. The proportion of patients receiving FNRs each year was calculated using the following equation, and the overall trends of FNR use were described for the 7-year observation period: trends in FNR use according to sex, region, and specific FNR were evaluated. Subgroup analyses by age, sex, specific drugs and geographical religion were conducted. Four age groups were set to determine whether the trends in FNR use were driven by a particular age group. The four age groups were children and adolescents (2–17 years), young adults (18–45 years), middle-aged adults (46–59 years), and older adults (60 years and older). The data were processed using Access software (Microsoft, Redmond, WA, United States). The chi-square test was used to compare patients in males vs females in each year, and the Mann–Kendall trend test was used to assess trends in prescribed drugs. A log-linear test was used to assess trends in proportions. All the statistical analyses were carried out using R V4.0.5 software. Statistical significance was set at a P value < 0.05. Results Characteristics and Overall Trends of FNR of Included Prescriptions A total of 15596620 prescriptions were eligible for analysis, and 1492793 patients were included. Among these patients, 91515 (6.13%) received FNRs. As shown in Figure 1, the yearly number of ICU patients who received FNRs showed an increasing trend (P=0.007), while the percentage of these patients did not show a significant trend over 7 years (P=0.764). The demographic characteristics of the patients and those who received FNR treatment are shown in Table 1. The number of male patients was usually greater than that of female patients (chi-square test, P<0.001). Patients aged > 60 years had the highest percentage of patients who received FNRs (chi-square test, P<0.001). However, the proportion of FNR users among patients aged 18–45 years increased during the study period (P= 0.035). Trends in the yearly number and proportion of patients using FNRs. Table 1Demographic Characteristics of Included Patients from 2013 to 2019Total2013201420152016201720182019PICU patientsAge2–1758,08863056629856984529718944589700.07218–45227,59826,99431,09536,32936,81635,50931,26729,588>0.99946–59301,70434,09736,87444,32147,38050,04347,88441,1050.133>60905,40389,725114,930147,076144,077154,264140,907114,4240.764SexMale93252596,234116,817146,752149,677156,735146,790119,5200.230Female56026860,88772,71189,54387,04892,79982,71374,5670.548Total1492793157,121189,528236,295236,725249,534229,503194,0870.368ICU patients using FNRsAge2–1725852502294124013984644310.13318–4510,99112301334163016931653163918120.03546–5915,05816311842202121902309231227530.003>6062,88173788169935293039999919494860.133SexMale5929267727536866087329212889394870.007Female3222337174038475548555147471649950.072Total9151510,48911,57413,41513,58714,35913,60914,4820.006ProportionAge2–174.45%3.97%3.45%4.81%4.74%4.10%4.91%4.80%0.23018–454.83%4.56%4.29%4.49%4.60%4.66%5.24%6.12%0.01646–594.99%4.78%5.00%4.56%4.62%4.61%4.83%6.70%0.548>606.95%8.22%7.11%6.36%6.46%6.48%6.52%8.29%0.764SexMale6.36%7.04%6.45%5.90%5.83%5.88%6.06%7.94%>0.999Female5.75%6.10%5.55%5.31%5.58%5.55%5.70%6.70%0.44Overall6.13%6.68%6.11%5.68%5.74%5.75%5.93%7.46%0.764Notes: P-value for trend in number and proportion of patients were assessed by Mann–Kendall trend test and log-linear test, respectively. The proportion is defined as the number of patients using FNRs divided by the total number of ICU patients. Demographic Characteristics of Included Patients from 2013 to 2019 Notes: P-value for trend in number and proportion of patients were assessed by Mann–Kendall trend test and log-linear test, respectively. The proportion is defined as the number of patients using FNRs divided by the total number of ICU patients. Trends in Specific Drugs Eight FNRs were identified in the present study, seven of which were compound preparations of β-lactams and enzyme inhibitors. The prescription trends for each FNR were analyzed. As shown in Figure 2, the use of cefoperazone/sulbactam ranked first, followed by piperacillin/sulbactam, cefoperazone/tazobactam, mezlocillin/sulbactam, cefotaxime/sulbactam, ceftriaxone/tazobactam, and amoxicillin/sulbactam. Detailed data regarding the total amount of each medication used and the proportion of ICU patients using the medication are shown in Table 2. The use of cefoperazone/sulbactam and mezlocillin/sulbactam significantly increased prescription volume (both P <0.05). Concurrently, a notable decrease was observed in the proportion of piperacillin/sulbactam users (P=0.07), whereas a substantial increase in the proportion of cefotaxime/sulbactam users was detected (P=0.048). Trends in use of each FNR. (A) The number of patients. (B) The usage proportion trends by drug. Table 2Number of ICU Patients with a Specific FNR from 2013 to 2019Total2013201420152016201720182019P1P2Cefoperazone/sulbactam56229(3.77%)6512(4.14%)7050(3.72%)7553(3.20%)8041(3.40%)8730(3.50%)8393(3.66%)9950(5.13%)0.0070.764Piperacillin/sulbactam17077(1.44%)2597(1.65%)2490(1.31%)2866(1.21%)2777(1.17%)2481(0.99%)1994(0.87%)1872(0.96%)0.0720.007Cefoperazone/tazobactam8754(0.59%)767(0.49%)1044(0.55%)1388(0.59%)1222(0.52%)1532(0.61%)1698(0.74%)1103(0.57%)0.1330.133Mezlocillin/sulbactam4410(0.30%)399(0.25%)621(0.33%)684(0.29%)486(0.21%)703(0.28%)693(0.30%)824(0.42%)0.0350.368Cefotaxime/sulbactam3586(0.24%)93(0.06%)97(0.05%)527(0.22%)752(0.32%)801(0.32%)667(0.29%)649(0.33%)0.1330.048Ceftriaxone/tazobactam1740(0.12%)149(0.09%)271(0.14%)430(0.18%)418(0.18%)151(0.06%)178(0.08%)143(0.07%)0.5480.45Amoxicillin/sulbactam395(0.026%)57(0.036%)64(0.034%)125(0.053%)37(0.016%)50(0.020%)44(0.019%)18(0.009%)0.1330.071Amoxicillin/flucloxacillin48(0.003%)10(0.006%)29(0.015%)9(0.004%)——————Notes: Data were presented by number (percentage) of patients. P1-value for trend in number of patients which were assessed by Mann–Kendall trend test; P2- value for trend in proportions, which is defined as the number of patients using FNRs divided by the total number of ICU patients. Number of ICU Patients with a Specific FNR from 2013 to 2019 Notes: Data were presented by number (percentage) of patients. P1-value for trend in number of patients which were assessed by Mann–Kendall trend test; P2- value for trend in proportions, which is defined as the number of patients using FNRs divided by the total number of ICU patients. Table 3 shows the number and proportion of hospitals providing each FNR. Cefoperazone/sulbactam is the most widely used treatment, followed by piperacillin/sulbactam. Other FNRs were prescribed in fewer than 30% of the hospitals, while cefotaxime/sulbactam showed an increasing trend (P=0.011). Table 3Number and Proportion of Hospitals Providing Specific FNR from 2013 to 2019Total2013201420152016201720182019P1P2Cefoperazone/Sulbactam64(100%)56(87.5%)61(95.3%)58(87.9%)58(90.6%)62(96.9%)63(98.4%)62(96.9%)0.0650.048Piperacillin/Sulbactam32(50.0%)26(40.6%)27(42.2%)28(42.4%)29(45.3%)24(37.5%)24(37.5%)23(35.9%)0.2870.288Cefoperazone/Tazobactam13(20.3%)10(15.6%)11(17.2%)12(18.20%)9(14.10%)11(17.2%)10(15.6%)10(15.6%)0.7510.751Mezlocillin/Sulbactam20(31.3%)13(20.3%)14(21.9%)12(18.2%)11(17.2%)15(23.4%)13(20.3%)11(17.2%)0.5380.539Cefotaxime/Sulbactam11(17.2%)2(3.1%)5(7.8%)8(12.1%)8(12.5%)8(12.5%)10(15.6%)10(15.6%)0.0110.006Ceftriaxone/Tazobactam6(9.38%)3(4.69%)6(9.38%)5(7.58%)5(7.81%)5(7.81%)4(6.25%)3(4.69%)0.3410.539Amoxicillin/Sulbactam5(7.81%)2(3.13%)4(6.25%)4(6.06%)2(3.13%)2(3.13%)1(1.56%)1(1.56%)0.0760.057Amoxicillin/Flucloxacillin3(4.69%)3(4.69%)2(3.13%)1(1.52%)——————Notes: P1-value for trend in number of hospitals which FNR were prescribed, which were assessed by Mann–Kendall trend test; P2- value for trend in proportions, which is defined as the number of hospitals using FNRs divided by the total number of hospitals in specific year. Number and Proportion of Hospitals Providing Specific FNR from 2013 to 2019 Notes: P1-value for trend in number of hospitals which FNR were prescribed, which were assessed by Mann–Kendall trend test; P2- value for trend in proportions, which is defined as the number of hospitals using FNRs divided by the total number of hospitals in specific year. Trends by Geographical Religions The prescription trends of FNR were further analyzed according to different geographical regions. As shown in Table 4 and Table S2, significant differences exist among the regions. The proportion of patients receiving FNR was greater in Guangzhou (17.2%, chi-square test, P<0.001), while the proportions in Hangzhou and Chengdu were lower (3.82% and 3.83%, chi-square test, P<0.001). The number of patients receiving FNRs increased in three regions (Hangzhou, Tianjin, and Guangzhou) (all P<0.05), which calls for special attention. Table 4Trends in FNR Use by Geographical Regions from 2013 to 2019CitiesTotal2013201420152016201720182019P1P2Shanghai7183(5.74%)925(5.33%)1066(6.35%)1078(6.13%)901(5.74%)1013(6.40%)990(5.98%)1210(4.80%)0.5480.764Hangzhou11533(3.82%)1305(3.88%)1469(3.77%)1653(3.97%)1716(4.22%)1765(4.00%)1720(3.39%)1905(3.66%)0.0070.764Beijing14984(8.08%)2147(7.50%)1934(7.44%)2029(7.22%)2241(8.45%)2321(9.47%)2278(8.90%)2034(7.78%)0.3680.368Tianjin9303(6.48%)1031(7.72%)1001(5.37%)1127(5.01%)1160(5.29%)1788(7.62%)1500(7.00%)1696(7.63%)0.0350.764Chengdu22445(3.83%)2572(5.28%)3336(4.63%)3663(3.41%)3328(3.00%)3198(2.73%)3012(3.38%)3336(8.31%)>0.9990.550Guangzhou26067(17.2%)2509(16.3%)2768(16.3%)3865(20.3%)4241(20.2%)4274(17.5%)4109(15.7%)4301(15.2%)0.0160.288Notes: Data were presented by number (percentage) of patients who received FNRs. P1-value for trend in number of patients, which were assessed by Mann–Kendall trend test. P2- value for trend in proportions, which was assess using log-linear test. Trends in FNR Use by Geographical Regions from 2013 to 2019 Notes: Data were presented by number (percentage) of patients who received FNRs. P1-value for trend in number of patients, which were assessed by Mann–Kendall trend test. P2- value for trend in proportions, which was assess using log-linear test. Discussion To our knowledge, this study is the first to analyze trends in the use of FNR over time at the national level using surveillance data. As the sample was large and covered six areas of China, the results are nationally representative. More than 6% of ICU patients have used FNR, and this proportion has been increasing in general (7.46% in 2019). The proportion of elderly patients using FNR was always greater than the average. Eight FNRs were identified in this study. Cefoperazone/sulbactam and piperacillin/sulbactam are frequently and widely used. The use of cefotaxime/sulbactam has also increased. Currently, there are limited data available on the use of FNRs at the national level. The available information primarily pertains to the consumption of FDC antibiotics, and only a few literature reports mention the use of specific FNR.10,26–30 According to a previous report that analyzed antibiotic usage in 76 countries from 2000 to 2015, FNR antibiotics were employed in 20 (26.3%) countries.29 The contribution of FNR use to overall antibiotic consumption was less than 3%. However, the usage of specific countries, such as Egypt, India, and Pakistan, were significantly higher, at 9.6%, 7.5%, and 4.0%, respectively.29 Another study had reviewed global antibiotic sales data and identified 119 FDC antibiotics. The majority (92%) of these FDCs were not approved by the FDA and comprised 5.1% of total sales. China had the second highest FDC count (25/119). Although only 8 FNRs were identified in this study, more than 6% of the ICU patients had received FNR. This percentage was high compared with existed reports. Thus, special attention should be given to the better management of FNRs. Cefoperazone/sulbactam was the most frequently used FNR in this study, as it has shown good clinical effectiveness and is recommended as a first-line empirical antibiotic in multiple anti-infection guidelines in China.31 A meta-analysis of 110 studies on the effectiveness and cure rate of cefoperazone/sulbactam showed that the effectiveness rate of cefoperazone/sulbactam was 80.3%, and the cure rate was 50.1% for the treatment of clinical infections.32 An in vitro study showed that the addition of sulbactam to cefoperazone could enhance the antibacterial activity of cefoperazone against carbapenem-resistant Acinetobacter baumannii but had no effect on carbapenem-resistant Pseudomonas aeruginosa.33 A literature review showed that if the combination of cefoperazone/sulbactam could achieve pharmacokinetic optimization in human serum, its clinical application could be expanded.34 The major concern regarding cefoperazone/sulbactam is its potential relationship with coagulopathy and bleeding events, but recent studies have shown that the risk is acceptable.35,36 Therefore, it is necessary to obtain more convincing data to support or discourage the use of cefoperazone/sulbactam. Piperacillin/sulbactam, the second most frequently used FNR, has successfully addressed the issues of high chemical cost and frequent penicillin allergic reactions associated with piperacillin/tazobactam by replacing tazobactam with sulbactam. According to an in vitro experiment,37 the antibacterial activities of the two are similar for gram-positive bacteria. However, for gram-negative bacteria, especially Escherichia coli and Proteus vulgaris, the antibacterial activity of piperacillin/sulbactam was significantly higher than that of piperacillin/tazobactam. The antibacterial effects of other Enterobacteriaceae and Pseudomonas aeruginosa strains were comparable. Notably, for Stenotrophomonas maltophilia, the activity of piperacillin/sulbactam was greater. A randomized, single-blind, controlled clinical trial in China showed that piperacillin/sulbactam can be used as a suitable substitute for piperacillin/tazobactam for the treatment of community-acquired respiratory and urinary tract infections caused by β-lactamase-producing bacteria.38 In vitro experiments further confirmed that the addition of sulbactam to piperacillin improved the sensitivity of Pseudomonas aeruginosa and Acinetobacter baumannii to piperacillin.39 There are only some in vitro studies but little high-quality evidence-based medical evidence for cefoperazone/tazobactam, mezlocillin/sulbactam, cefotaxime/sulbactam, amoxicillin/sulbactam and amoxicillin/flucloxacillin.27,40,41 For drugs that are widely used in clinical settings and demonstrate clear therapeutic benefits, large-scale randomized controlled trials can be conducted to provide higher-level evidence of their efficacy and safety. This will enable a more comprehensive evaluation of drug effectiveness and safety, thereby providing doctors and patients with accurate information. For FNR, which has been proven to have no better clinical efficacy, standardized management systems should be established. These include developing guidelines for their use, limiting unnecessary utilization, promoting safer and more effective drug alternatives, and other measures. Additionally, healthcare facilities should strengthen supervision and management to ensure effective control of FNR. This study had several limitations. First, our analysis relied solely on prescription data, thereby rendering an evaluation of the adequacy of antimicrobial therapy as well as its outcomes infeasible. Additionally, there is a dearth of data pertaining to indications for prescribing antimicrobials, a subject that merits further exploration in subsequent research. Conclusion This study investigated the national landscape of FNR use among ICU patients in China. The overall FNR use in these patients was > 6%, which calls for special attention to rational use and better management. Among the eight FNRs identified in this study, cefoperazone/sulbactam and piperacillin/sulbactam were the most frequently used and widely distributed. Data on the real-world effectiveness and safety of these FNRs are urgently required to support or discourage their use. Acknowledgments The authors would like to thank the Hospital Prescription Analysis Cooperative Project of China for collecting and providing data. Funding Statement This work was supported by Zhejiang Medical Doctors Association (YS2022-2-002). Data Sharing Statement The original contributions presented in this study are included in the article/supplementary material, and further inquiries can be directed to the corresponding authors. Ethics Approval Studies involving human participants were reviewed and approved by the Ethics Committee of Sir Run Run Shaw Hospital, College of Medicine, Zhejiang University (Reference Number KEYAN2023-0209). Written informed consent for participation was not required for this study, in accordance with national legislation and institutional requirements. Author Contributions All authors made a significant contribution to the work reported, whether that is in the conception, study design, execution, acquisition of data, analysis and interpretation, or in all these areas; took part in drafting, revising or critically reviewing the article; gave final approval of the version to be published; have agreed on the journal to which the article has been submitted; and agree to be accountable for all aspects of the work. Disclosure No potential competing interest was reported by the authors. References Associated Data Data Availability Statement The original contributions presented in this study are included in the article/supplementary material, and further inquiries can be directed to the corresponding authors.