A Multidisciplinary Approach to Improvements in Perioperative Antibiotic Prophylaxis at a Military Treatment Facility

ABSTRACT

Introduction

Surgical site infections complicate 2%-5% of surgeries. According to the Centers for Disease Control and Prevention, half of all surgical site infections are preventable. Adherence to published recommendations regarding perioperative antibiotic administration decreases the incidence of surgical site infections. Members of the Department of Anesthesia noticed casual observations of inaccurate prescribing of antibiotics at our institution, Brooke Army Medical Center, and approached the Antimicrobial Stewardship Program to collaborate on this issue.

Materials and Methods

A team of anesthesiologists, clinical pharmacists, and infectious disease specialists collaborated with the Department of Surgery to improve this effort as part of a multiyear project from 2018 to 2021. We first assessed adherence to recommended perioperative antibiotic use to establish a baseline and next, noticing gaps, created a project with the goal to improve compliance to >90% across surveyed measures. Our key interventions included educational initiatives, creation of facility-specific guidelines, peer benchmarking, updating order sets, interdisciplinary collaboration, creation of intraoperative reminders and visual aids, and tailored presentations to selected services.

Results

Of 292 charts (2.3% of cases from January to October 2018) reviewed pre-intervention, compliance rates were 84% for antibiotic choice, 92% for dose, 65% for redosing, and 71% for postoperative administration. Of doses, 100% were timed correctly, and thus, this variable was not targeted. Post-intervention, our review of 387 charts (10% of cases from May to November 2020) showed no change in correct antibiotic choice (84%) and statistical improvement to 96% for correct dose, 95% for correct redosing, and 85% for correct postoperative administration (P < .05 for all).

Conclusions

Our multidisciplinary approach of collaboration with multiple departments, creating guidelines and providing feedback, improved compliance with perioperative antibiotic administration recommendations.

INTRODUCTION

Perioperative antibiotic prophylaxis has decreased the incidence of surgical site infections from 30%-60% to 2%-5%.¹ Despite this, the CDC estimates that 50% of all surgical site infections are preventable.¹ The American Society of Health-System Pharmacists, Infectious Disease Society of America, Surgical Infection Society, and Society for Healthcare Epidemiology of America provided recommendations in 2013 about timing, dosing, agent selection, and postoperative dosing if appropriate.^1,2 Incomplete adherence to these recommendations increases the risk of infection.³ However, overprescribing of antibiotics does not decrease surgical site infections and leads to increased multidrug-resistant organisms, end-organ damage, and Clostridioides difficile infection.³ Our institution, Brooke Army Medical Center, has an Antimicrobial Stewardship Program (ASP) that attempts to curb overprescribing of antibiotics and guide appropriate choice and dose of agents. Such programs have been used at Military Treatment Facilities (MTFs) dating back to at least 2006 and their use in combat theaters has been well described.^14,15 DoD MTFs have specifically been identified in the National Action Plan for Combating Antibiotic-Resistant Bacteria as a line of effort for improvement in antibiotic prescribing. The state of antimicrobial stewardship in the DoD is currently being evaluated by studying the structure and outcomes of ASPs at military institutions (L. Lynch, personal correspondence, 2022).

At our institution, members of the Department of Anesthesia recognized that intraoperative redosing of antibiotics was inconsistently performed during lengthy cases. They approached the institution’s ASP with this concern. Simultaneously, hospital leadership reached out to the ASP to raise concern over occasional inaccurate antibiotic dosage prescribing because of the use of outdated perioperative order sets in the electronic health record (at the time, Essentris). A team was formed to investigate the scope of the problem and consider interventions. We determined success at >90% adherence to guidelines.

Baseline data were analyzed from January to October of 2018, reviewing a representative subset of cases performed at our medical center for adherence to the 2013 interdisciplinary guidelines.² We reviewed agent choice per surgery, dosing, timing, intraoperative redosing, and postoperative dosing. While correct timing of antibiotics was performed consistently at our institution, we found that only 84% of surgeries were adherent to the correct antibiotic choice (or omission), 65% were correctly redosed intraoperatively for long procedures, and only 71% had correct postoperative administration and decided on these as the major goals of the project.

METHODS

A list of all surgeries at our institution was generated electronically. Of these surgeries, a subset involving cases from each department was analyzed. A chart review of these subset cases was performed by members of the team to assess adherence in each domain of prescribing as compared to 2013 guidelines. Findings from our baseline assessment informed our areas of concentration. Descriptive data were compiled and presented as part of peer benchmarking.

Interventions

Educational initiatives

First, we presented our project, rationale, and baseline data at Surgical Grand Rounds in July 2019. Following this initial presentation, team members presented at the Department of Surgery meetings and targeted individual services. The educational efforts were delivered by anesthesiologists, pharmacists, and infectious disease specialists.

Clinical practice guideline generation

During the initial phases of the intervention, we discovered that many surgical services were unaware of the 2013 guidelines and/or of modern dosing strategies. A common concern we heard was that the 2013 guidelines were not endorsed by their specific subspecialty society, although all subspecialty-specific guidelines had similar recommendations. As such, we created a document that cross-referenced to subspecialty guidelines as appropriate (available in the Supplementary material). Some voiced concern that the recommendation of short-duration postoperative antibiotics was made via inappropriate extrapolation. In these cases, we collaborated with them to perform a comprehensive literature review on the cases of concern. This was overseen and edited by infectious disease specialists.

Peer benchmarking

Each specialty’s data were presented in comparison to the other surgical services. These were presented at the Surgery Grand Rounds in July 2019. Additionally, each service’s “report card” was sent to the service chief for dissemination.

Updating order sets

Although we had a high baseline level of adherence to weight-based dosing (92%), we recognized early that antibiotics were selected from pre-populated order sets in the electronic health record. Individual services were contacted to update the dosing of cefazolin from 1 g for all to 2 g for patients >80 kg and 3 g for patients >120 kg.² Vancomycin was updated to be dosed at 15 mg/kg.²

Interdisciplinary collaboration

Subspecialty investment was generated by framing the problem in the context of a recent surgical site infection outbreak. Individual services were recruited to the effort through literature review to better inform the clinical practice guidelines, to recommend which cases they would like reviewed (often chosen for complexity or risk), and to help collect data for benchmarking. Additionally, each service was contacted directly about updating weight-based antibiotic dosing in their order sets and interfaced with one-on-one to review the literature and provide education as requested.

Intraoperative reminders

The anesthesiology department generated intraoperative reminder cards reflecting redosing guidelines and posted them in key locations in the operating room.

Statistical Analysis

Data were presented as pre- (January to October 2018) and post-interventions (May to November 2020) for each variable assessed, and the results were grouped as a whole and divided based on surgical specialty (Tables I–IV). Data were interpreted using either Pearson’s chi-squared test or Fisher’s exact test to check whether five or fewer cases were reviewed. The results were interpreted using SPSS.

RESULTS

A total of 679 cases were reviewed for perioperative antibiotic management during both our pre- (n = 292) and post-intervention (n = 387) data collection, reflecting a representative percentage of thousands of cases performed at our tertiary medical facility. We were able to examine a higher percentage of patients in the post-intervention period because of decreased volume during the COVID-19 pandemic.

Our first variable assessed, choice of appropriate preoperative prophylactic antibiotic, demonstrated no change across all queried surgical services between our pre- and post-intervention periods (Table I). In fact, some services (General Surgery and Urology) demonstrated significantly diminished adherence with agent choice in our post-intervention data (General Surgery 94% vs. 82% adherence, P = .046; Urology 96% vs. 75%, P = .034).

Assessment of correct dose of prophylactic agents (Table II) showed statistically significant post-intervention improvement in adherence across all services (92% vs. 96% adherence, P = .025), with some services showing vast increases in the proportion of cases with appropriate dose choices.

Our next measure, appropriate intraoperative redosing of antibiotics (Table III), demonstrated statistically significant improvement when considering all services (65% vs. 95% adherence, P ≤ .001). Additionally, several services also showed individual statistically significant improvement in adherence to appropriate redosing strategies.

Finally, the results of our fourth variable reviewing the appropriate continuation of agents into the postoperative period (which varies widely based on surgical specialty and type of procedure) are represented in Table IV. We found statistically significant improvement in compliance across all services queried (71% vs. 85% compliance, P ≤ .001). In reviewing individual services, we found statistically significant improvement in not only the Ear, Nose, and Throat and Plastic Surgery services but also in the Ophthalmology service with a nonsignificant decline in adherence.

DISCUSSION

During the planning phase of our project, we determined a goal compliance rate across all measured variables of >90%. This accounts for the unrealistic expectation that compliance would ever approach or achieve 100%, which is unachievable when patient-specific factors and other variables are considered. We were able to reach or maintain >90% for both correct dose and intraoperative redosing. Although we did not meet >90% compliance for antibiotic choice or postoperative administration, we noticed improvements in postoperative antibiotic utilization. All measures barring antibiotic choice demonstrated statistically significant pre- and post-intervention improvement. In doing this project, we noticed some idiosyncrasies among adherence to specific services. We provided peer benchmarking data to each service and have engaged with several services to further explore reasons for nonadherence or apparent decreases in metrics reported.

Our project was received positively at our medical facility. The large improvement noted in intraoperative redosing of agents (from 65% to 95% adherence) reflects acceptance and implementation of our improvements by involved services, specifically the anesthesia department, whose providers are responsible for the administration of intraoperative redoses. Members of various surgical services assisted our team in collecting service-specific post-intervention data, which further reflect buy-in to our project by the surgical departments.

A project like this is useful at any facility that performs surgical procedures. Tracking of surgical site infections is a national requirement, and all facilities are nationally mandated to have ASPs that perform key functions.^14,16 Tracking and improving antibiotic utilization is one key function, as well as generating and disseminating clinical practice guidelines. This also has the potential to improve care for thousands of patients a year, depending on the volume of surgical cases. Additionally, our interventions are relatively simple and inexpensive (presentations and education, posting of laminated Clinical Practice Guidelines (CPGs) in operating rooms, and modifying order sets). The total effort required to collect and interpret data, create CPGs and disseminate them, and generate materials for and give presentations was completed by our team of five individuals on top of full-time work requirements. It is reasonable to expect that these interventions are feasible and translatable to other health facilities, military, or otherwise, with operating rooms and surgical services.

Currently, there is a transition across all MTFs to a Defense Health Agency (DHA)–wide charting system, MHS Genesis. Within this system, order sets (termed power plans) are standardized at the DHA level, rather than for each institution. This offers a benefit because order sets could be made with the appropriate antibiotic selection and dose and then implemented within all MTFs. An important caveat to such DHA-wide order sets would be allowing for nuances between facilities regarding certain cases and/or patients, which could be addressed by a facility’s ASP.

Other key transferrable lessons are enlisting the support of specialties that are impacted by the interventions. Unrelated to the project, there was an uptick in surgical site infections in 2019 when the main efforts began, which was used to frame the need to revise our approach and build a shared vision among the involved departments. There was also only one instance of reconsideration of a recommendation after a targeted literature review with a surgical service. The process of a systematic literature review helped build a team.

We believe that our success despite multiple challenges is evidence of sustainability at our institution. Our project, originally planned to span 18 months, required 4 years to date and is still ongoing. Despite our challenges, we have demonstrated success. Our challenges highlight the need to make high-priority projects an ongoing institutional priority, which has allowed for success through deployments, leaves of absence, personnel turnover, and pandemics.

Our interventions were implemented across multiple areas of care and within numerous departments and required collaboration between different services (pharmacy, infectious disease, surgery, and anesthesia) to effectively manage a patient’s perioperative antibiotics. Partnering with these groups ensured that the best available evidence was used to prepare our CPGs, which were then disseminated to all targeted services.

Limitations of our project include its implementation at a single facility, meaning its translatability to other institutions remains unproven. Additionally, the personnel turnover in an MTF related to changes of station and deployments means the educational interventions undertaken at the beginning of our project require reemphasis with new staff to ensure continued buy-in. Our CPGs offer broad recommendations for prophylactic agents and lack the nuance a unique circumstance may require. They were not intended to supplant clinician judgment, but rather to serve as an additional resource. The physical copies of the CPGs that were placed in operating environments require surveillance to ensure they remain usable, in addition to possible updating as new evidence becomes available.

CONCLUSION

With appropriate antibiotic prophylaxis, surgical site infections will complicate 2%-5% of surgeries, although up to half of these infections are preventable.¹ Proper management of perioperative antibiotics requires the participation and compliance of multiple services and is an important step in preventing these infections. Through our multidisciplinary quality improvement project, we were able to collaborate among multiple departments, create guidelines, and provide feedback to improve compliance with perioperative antibiotic administration recommendations at our institution.

ACKNOWLEDGMENTS

We would like to thank COL Katharine Markell, MD, FACS, FASCRS (Ret.) for her support in this project.

SUPPLEMENTARY MATERIAL

SUPPLEMENTARY MATERIAL is available at Military Medicine online.

FUNDING

No funding was provided for this project.

CONFLICT OF INTEREST STATEMENT

No conflicts of interest to note for any of the authors of this project.

DATA AVAILABILITY

The data that support the findings of this study are available on request from the corresponding author. All data are freely accessible.

CLINICAL TRIAL REGISTRATION

None declared.

INSTITUTIONAL REVIEW BOARD (HUMAN SUBJECTS)

This project was completed as part of the usual clinical duties of the ASP and did not require IRB approval as it was granted a non-research determination.

INSTITUTIONAL ANIMAL CARE AND USE COMMITTEE (IACUC)

Not applicable.

INSTITUTIONAL CLEARANCE

Institutional clearance does not apply.

INDIVIDUAL AUTHOR CONTRIBUTION STATEMENT

A.L.B. and S.E.G. collected and analyzed the data. A.L.B. and A.E.B. drafted the original manuscript. A.M. and A.E.B. designed this project, reviewed, and edited the manuscript. All authors (A.E.B, A.M., A.L.B, S.E.G, and A.L.S) read and approved the final manuscript.

REFERENCES

Author notes