The Global Alliance for Infections in Surgery: defining a model for antimicrobial stewardship—results from an international cross-sectional survey

Background Antimicrobial Stewardship Programs (ASPs) have been promoted to optimize antimicrobial usage and patient outcomes, and to reduce the emergence of antimicrobial-resistant organisms. However, the best strategies for an ASP are not definitively established and are likely to vary based on local culture, policy, and routine clinical practice, and probably limited resources in middle-income countries. The aim of this study is to evaluate structures and resources of antimicrobial stewardship teams (ASTs) in surgical departments from different regions of the world. Methods A cross-sectional web-based survey was conducted in 2016 on 173 physicians who participated in the AGORA (Antimicrobials: A Global Alliance for Optimizing their Rational Use in Intra-Abdominal Infections) project and on 658 international experts in the fields of ASPs, infection control, and infections in surgery. Results The response rate was 19.4%. One hundred fifty-six (98.7%) participants stated their hospital had a multidisciplinary AST. The median number of physicians working inside the team was five [interquartile range 4–6]. An infectious disease specialist, a microbiologist and an infection control specialist were, respectively, present in 80.1, 76.3, and 67.9% of the ASTs. A surgeon was a component in 59.0% of cases and was significantly more likely to be present in university hospitals (89.5%, p < 0.05) compared to community teaching (83.3%) and community hospitals (66.7%). Protocols for pre-operative prophylaxis and for antimicrobial treatment of surgical infections were respectively implemented in 96.2 and 82.3% of the hospitals. The majority of the surgical departments implemented both persuasive and restrictive interventions (72.8%). The most common types of interventions in surgical departments were dissemination of educational materials (62.5%), expert approval (61.0%), audit and feedback (55.1%), educational outreach (53.7%), and compulsory order forms (51.5%). Conclusion The survey showed a heterogeneous organization of ASPs worldwide, demonstrating the necessity of a multidisciplinary and collaborative approach in the battle against antimicrobial resistance in surgical infections, and the importance of educational efforts towards this goal. Electronic supplementary material The online version of this article (doi:10.1186/s13017-017-0145-2) contains supplementary material, which is available to authorized users.


Background
Antimicrobial Stewardship Programs (ASPs) have been promoted to optimize antimicrobial usage and patient outcomes and reduce the emergence of antimicrobial-resistant organisms. However, the best strategies for an ASP are not definitively established and are likely to vary based on local culture, policy and routine clinical practice, and probably limited resources in middle-income countries [1,2]. Many hospitals remain without formal programs and those that do continue to struggle with gaining acceptance across service lines [3]. Moreover, identifying optimal efforts to impact system change has been challenging [4].
Restriction strategies may be effective at controlling use but raise issues of prescriber autonomy and require a large personnel commitment. Encouraging multidisciplinary collaboration within health systems to ensure that prophylactic, empirical, and targeted use of antimicrobial agents results in optimal patient outcomes is mandatory in the current era of antimicrobial resistance.
A panel of experts from the Surgical Infection Society (SIS) and World Society of Emergency Surgery (WSES) has recently published a review with the aim of defining the role of surgeons within the ASPs. The panel proposed that the best means of improving antimicrobial stewardship in surgical units worldwide should involve collaboration among various specialties within institutions including prescribing clinicians and pharmacists [5].
In 2016, a multidisciplinary task force from 79 different countries joined a global project to develop a consensus on the rational use of antimicrobials for patients with intra-abdominal infections (IAIs). The project has been termed AGORA (Antimicrobials: A Global Alliance for Optimizing their Rational Use in Intra-Abdominal Infections) [1].
Recently the Global Alliance for Infections in Surgery was founded and experts from 87 countries worldwide joined the highly diverse and skilled International Advisory Board. This alliance, promoted by the WSES, includes an interdisciplinary group of hospital administrators, epidemiologists, infection control specialists, infectious disease specialists, microbiologists, clinical pharmacologists and hospital pharmacists, surgeons, and intensivists. The mission of this alliance is to educate healthcare providers promoting the standards of care in managing infections in surgery worldwide [6]. Therefore, this study was conducted to evaluate the structure and resources of antimicrobial stewardship teams (ASTs) in surgical departments from different regions of the world.

Methods
We conducted a cross-sectional electronic survey evaluating the structure and resources of ASTs in surgical departments. The survey was designed by a multidisciplinary team of investigators including an epidemiologist, a surgeon, an infectious diseases physician, a pharmacologist, and a microbiologist. The questionnaire was piloted among five physicians for face and content validity.
The 24-item self-administered questionnaire collected information from multidisciplinary experts-mostly physicians-about characteristics and composition of the hospital team, implementation of local procedures, availability of antimicrobial use monitoring and surveillance systems, presence of an ASP, and related interventions (Additional file 1). An electronic invitation with a link to the survey was sent to 831 physicians: 173 physicians who participated in the AGORA project [1], and a large number (658) of international experts in the fields of antimicrobial stewardship, infection control, and infections in surgery identified after a thorough investigation using the PubMed database. The survey was Internetbased (using http://www.docs.google.com). Participation was voluntary but not anonymous; however, the confidentiality of respondents and their choices was ensured. No incentives were provided to the respondents. The study was open for 6 weeks between September 30 and November 11, 2016. Reminders were sent to all those who had not replied after 1 and 3 weeks. Due to the characteristics of the survey, a response rate ranging between 15 and 25% was expected.
Data were entered in an Excel database (Microsoft Corporation, Redmond, Washington, USA) and analyzed using Stata 11.0 software package (StataCorp, College Station, TX). Descriptive analyses included medians and interquartile ranges (IQR) for continuous variables or frequency (%) for categorical variables The two-sided chi-square or Fisher's exact test was used for categorical variables, as appropriate. All tests were two-sided, and p values of 0.05 or lower were considered statistically significant.

Results
Baseline data: coverage, response rate, working setting, and professional profile A total of 161 (19.4%) of the 831 experts who were contacted by email completed the survey after two reminders. One incomplete survey was excluded from the study. In two cases the participants were from the same institution and only one survey was considered. One hundred fifty-eight responses were included in our analysis. Participants work settings and professional profiles are summarized in Table 1.
The response rate was similar to that of previous studies promoted by WSES [1,7,8].
As in the other WSES studies [1,7,8], participants were not homogeneously distributed across all geographic regions of the world due to the difficulty in recruiting participants in some areas of the world. However all geographic regions were represented in the survey.

Characteristics of the team
One hundred fifty-six (98.7%) participants stated their hospital had a multidisciplinary AST. Ninety participants (90/156, 57.7%) declared they were currently members of the team, with no difference in frequency between different WHO regions. The median number of physicians working inside the team was five [IQR [4][5][6]. Characteristics of the team are in Table 2.
One hundred thirty-five (135/158, 85.4%) participants had at least one surgeon with an interest or skills in surgical infections within the surgical department of their hospital; a surgeon was significantly more likely to be present in university hospitals (89.5%, two-sided chisquare test p < 0.05) compared to community teaching hospitals (83.3%) and community hospitals (66.7%).

Implementation of protocols and monitoring systems
Implementation of protocols and monitoring systems in 158 hospitals are reported in Table 3.
The vast majority of respondents (152/158, 96.2%) stated that their hospitals have a protocol for pre-operative prophylaxis. The protocol covered all surgical wards in 124 (78.5%) cases. A protocol for antimicrobial treatment of surgical infections was available in 130 (82.3%) hospitals; however, only 70 (44.3%) had it available in every surgical ward. One hundred twenty-eight (81.0%) hospitals had both a protocol for peri-operative prophylaxis and for antimicrobial treatment of surgical infections available, while four (4/158, 2.5%) hospitals lacked both.
Among 130 surgical wards implementing a protocol for antimicrobial treatment of surgical infections, 97 (74.6%) participants stated it included interventions to reduce the duration of therapy, 88 (67.7%) interventions to switch selected antimicrobials from intravenous-to-oral therapy, 78 (60.0%) interventions for alternative dosing strategies based on pharmacokinetics and pharmacodynamics, with significant difference between community hospitals (11.1%, two-sided Fischer's exact test p < 0.05) compared to university (57.0%) and community teaching (60.0%) hospitals. Thirty-five (26.9%) participants reported the use of biological markers -such as procalcitonin to decrease antimicrobial use in critically ill patients.

Implementations of ASPs and related interventions
One hundred fifty-five (155/158, 98.1%) participants declared their hospital had an ASP running.
Types of ASPs and related interventions in surgical departments and in all hospital wards are described in detail in Table 4.
Six (6/41, 14.6%) surgical departments implementing a formulary restriction do not perform any monitoring system of used antimicrobials, and 4 (4/41, 9.8%) do not carry out any systematic reports about resistance data. Furthermore, 6 (7/70, 10.0%) surgical departments using a compulsory order form do not perform any monitoring system of used antimicrobials, and 11 (11/70, 15.7%) do not carry out any systematic reports about resistance data.
Characteristics of the implementation of protocols, monitoring systems, and ASPs interventions in surgical departments are detailed in Table 5.

Discussion
Antimicrobial stewardship programs (ASP) are a key strategy to curb the spread of antibiotic resistance [3,9]. The best strategies for an ASP are not definitively established and are likely to vary based on local routine clinical practice [7], despite several guidelines on the topic [9,10]. Successful ASPs should focus on collaboration between healthcare professionals in order to share knowledge and best practices. It is essential for an ASP to have at least one member who is an infectious diseases specialist. Pharmacists with advanced training or longstanding clinical experience in infectious diseases are also key actors for the design and implementation of the stewardship program interventions [11]. Infection control specialists and hospital epidemiologists should coordinate efforts on monitoring and preventing healthcare-associated infections and in analyzing and reporting "real-time" data to prevent infections, improve antimicrobial use, and minimize secondary spread of resistance. Microbiologists should actively guide the proper use of tests and the flow of laboratory results. Being involved in providing surveillance data on antimicrobial resistance, they should provide periodic reports on antimicrobial resistance data allowing the multidisciplinary team to determine the ongoing burden of antimicrobial resistance in the hospital. Moreover, timely and accurate reporting of microbiology susceptibility test results allows selection of more appropriate targeted therapy, and may help reduce broad-spectrum antimicrobial use.
Surgeons with adequate knowledge in surgical infections and surgical anatomy when involved in ASPs may audit antibiotic prescriptions, provide feedback to the prescribers and integrate best practices of antimicrobial use among surgeons, and act as champions among colleagues. Although many surgeons are aware of the problem of antimicrobial resistance, most underestimate it in their own hospital [1]. Very few studies have been published on the role of ASPs in general surgical departments. In 2015, Cakmakci [12] suggested that the engagement of surgeons in ASPs might be crucial to their success. In 2013, however, Duane et al. showed poor compliance of surgical services with ASP recommendations [13]. Surgeons need to take part in addressing the global issue of antimicrobial resistance. Failure to do so will be catastrophic to patients and programs [3].
Infections are the main factors contributing to mortality in intensive care units (ICU) [14].
Intensivists have a critical role in treating multidrug resistant organisms in ICUs in critically ill patients. They have a crucial role in prescribing antimicrobial agents for our most challenging patients and are at the forefront of a successful ASP [15].
Finally, without adequate support from hospital administration, the ASP will be inadequate or inconsistent since the programs do not generate revenue [16]. Engagement of hospital administration has been confirmed as a key factor for both developing and sustaining an ASP [17].
In most cases, our survey demonstrated that ASPs do not involve a true multi-disciplinary approach.
An infectious diseases specialist and a hospital pharmacist were part of the team in 125 (80.1%) and in 95 (60.9%) cases, respectively. Only 87 (55.8%) teams included both an infectious diseases specialist and a hospital pharmacist. An infection control specialist and a hospital epidemiologist were part of the team in 106 (67.9%) and in 64 (41.0%) cases, respectively. It is possible that in some hospitals,  Strategies of ASPs should be tailored based on individual hospital characteristics and personnel and resources available. The Infectious Diseases Society of America/ Society for Healthcare Epidemiology of America (IDSA/ SHEA) guidelines identified two core proactive evidencebased strategies and several supplemental strategies for promoting antimicrobial stewardship [7,8]: first, a restrictive strategy based on a proactive strategy of either formulary restriction or a requirement for pre-approval for specific drugs or both, and second, a persuasive strategy of performing prospective audit with intervention and feedback to the prescriber.
Our survey showed that 23 (16.9%) surgical departments have developed persuasive interventions, 14 (10.3%) restrictive interventions and 99 (72.8%) both of them. ASP policies should be based on both international/national antibiotic guidelines, and tailored to local microbiology and resistance patterns. Local clinical practice guidelines and algorithms can be an effective way to standardize prescribing practices based on the country's epidemiology. Standardizing a shared protocol of antimicrobial prophylaxis should represent the first step of any Antimicrobial Stewardship program.
One hundred fifty-two (96.2%) participants stated their hospitals have a protocol for surgical antibiotic prophylaxis. Among the 158 hospitals, a protocol for antibiotic prophylaxis is present in all surgical wards in 124 (78.5%) of hospitals while only in some surgical wards in 28 (17.7%) hospitals.
A protocol for antibiotic treatment was present in all surgical wards in 70 (44.3%) hospitals, while only in some surgical wards in 60 (38.0%) hospitals. Among 130 hospitals implementing a protocol for antimicrobial treatment of surgical infections, 97 (74.6%) participants stated that it included interventions to reduce the duration of therapy, 88 (67.7%) interventions to switch select antimicrobials from intravenous-to-oral therapy, 78 (60.0%) interventions for alternative dosing strategies based on pharmacokinetic and pharmacodynamic principles, with substantial difference between community hospitals (11.1%, two-sided Fischer's exact test p < 0.05), university (57.0%) and community teaching (60.0%) ones. Thirty-five (26.9%) participants admitted to the use of biological markerssuch as procalcitonin -to decrease antimicrobial use in critically ill patients.
In any healthcare setting, a significant amount of time and energy should be spent on infection control. Surveillance studies can help clinicians to identify trends in   All p values were calculated using two-sided chi-square test unless otherwise noted PAP pre-operative antimicrobial prophylaxis, TIS therapy for infections in surgery, ASP antimicrobial stewardship program a Calculated using two-sided Fisher's exact test. ASP antimicrobial stewardship program pathogens incidence and antimicrobial resistance, including identification of emerging pathogens at local level. The survey showed that 130 (83.3%) surgical departments had systematic reports about resistance data. Hospital pharmacists inside the multidisciplinary team should negotiate with hospital administration to obtain adequate and necessary infrastructure to measure antimicrobial use. Regular feedback about antimicrobial consumption can be an important determinant for change for healthcare professionals and policy makers to expedite progress towards prudent use of antimicrobials. The survey showed that 129 (81.6%) surgical departments had an antimicrobial monitoring system. Interestingly, 6 (6/41, 14.6%) surgical departments implementing a formulary restriction do not perform any monitoring system of used antimicrobials, and 4 (4/41, 9.8%) do not carry out any systematic reports about resistance data. Furthermore, 6 (7/70, 10.0%) surgical departments using a compulsory order form do not perform any monitoring system of used antimicrobials, and 11 (11/70, 15.7%) do not carry out any systematic reports about resistance data. In institutions that use restrictive interventions, monitoring overall trends in antimicrobial use and systematic reports about resistance data should be necessary to assess and respond to such shifts in use.
The ultimate goal of any stewardship program should be to stimulate a behavioral change in prescribing practices. In this context, education of prescribers is crucial to convince clinicians to use antibiotics judiciously. However, without concurrent interventions education alone is of little value. In this regard, various stewardship interventions have been implemented with the aim of improving adherence to guidelines. Where these interventions have been clinician focused, accumulating evidence suggests that educational interventions are mostly ineffective and result in insignificant changes to overall compliance [17]. It is possible that this might relate to cognitive dissonance, a process in which clinician-focused education fails to engage prescribers effectively, allowing them to ignore the evidence and to continue with their regular habits and practices. Alternative strategies of improving antibiotic management of surgical patients are needed and these may include guidance of clinicians in the institutional process of improvement, which has not as yet been addressed in guidelines [17]. The answer may lie within the principles and imperatives contained with the change of processes in hospitals.
It is highly important that faculty in academic medical centers and teaching hospitals focus on fundamental antibiotic stewardship principles in their preclinical and clinical curricula [18].
The survey found that dissemination of educational materials and educational outreach were developed respectively in 85 cases (62.5%) and 73 (53.7%) surgical departments.
This study has several limitations: with a response rate of just 19.4% we have to consider a response bias, and it is possible that non-participating physicians may have been less interested in ASPs than the participants and therefore it is possible that results are biased towards a better picture than it actually is. Furthermore, the study was conducted in a sample of physicians who participated in the AGORA project, and selecting international experts in the field again potentially resulting in an overrepresentation of hospitals with a considerably active ASP. No stratification or sampling according to medical specialty were pre-planned to ensure that all stakeholders were adequately represented, and finally our questionnaire was self-reported, has not been externally validated, and was evaluated in a single institution. The major strength of the study is its multinational (global) and multidisciplinary approach, to our best knowledge the first in this setting. Thus, our survey provides a benchmark to all interested stakeholders; it can be repeated over time to explore if better uniformity on a global platform of healthcare environments would develop in the future, and may be used to build consensus around the best practices in the field of prevention of surgical infections and rational use of antibiotics in a future project.

Conclusions
The results of the survey showed a heterogeneous organization of ASPs worldwide and demonstrated the need for a cohesive approach in order limit the emergence of antimicrobial resistance in surgical infections. Successful ASPs should focus on collaboration between all healthcare professionals in order to gain the widerpossible acceptance, share knowledge and spread best clinical practices. The main bias of the survey is the low response rate.

Additional file
Additional file 1: The international cross-sectional survey. (DOC 56 kb) Abbreviation ASP: Antimicrobial stewardship program