Hidden No More: Capturing the Full Picture of Prolonged Perioperative Antibiotic Prophylaxis.

Abstract

Whereas appropriate perioperative antibiotic prophylaxis (AP) has been shown to reduce the rates of surgical site infection (SSI) in adults, the concept of appropriate AP in the context of pediatric procedures is less well defined.1 Indeed, there is very few high-quality data to guide the multiple dimensions of perioperative AP use in children, including antibiotic choice, antibiotic dose, timing of antibiotic administration and redosing, and antibiotic duration. As a result, there is considerable variability in the use of AP among children undergoing surgical procedures.2,3 More recently, multiple clinical practice guidelines for the use of perioperative AP have been published,4–7 some of which include pediatric-specific recommendations on the basis of available evidence, expert opinion, and an extrapolation of data in adult patients. Several studies have documented noncompliance with these recommendations, including the use of perioperative AP when it is not indicated (eg, clean surgical cases without insertion of prosthetic material),3,8,9 failure to use perioperative AP when it is likely to be beneficial,10 and the use of inappropriately broad or narrow spectrum antibiotics.2,9,10 Given the scope and potential consequences of inappropriate perioperative AP in children, it has been identified as a research priority.11In the absence of data demonstrating a benefit to prolonged AP durations, most guidelines suggest limiting perioperative AP to no more than 24 hours,4–7 with recent recommendations from the Centers for Disease Control and Prevention (CDC) suggesting that a single preoperative dose of antibiotics is sufficient for clean and clean-contaminated procedures.6 Despite these recommendations, it is well-documented that children often receive perioperative AP in excess of 24 hours.2,9 Because unnecessary and prolonged courses of AP have been shown to increase the risk of acquisition of resistant bacterial infections, Clostridioides difficile infection (CDI),3 adverse drug events (ADEs),3 and prolonged hospitalizations,12 quantifying the extent of this problem across the spectrum of pediatric procedures is fundamental to improving the use of perioperative AP in children.In this issue of Hospital Pediatrics, Salas et al13 performed a retrospective review of all surgical procedures performed at their hospital between January and December 2017. During this time frame, existing strategies to ensure appropriate antibiotic use at the authors’ hospital included a well-established antimicrobial stewardship program (ASP) and institutional guidelines for the use of perioperative AP that recommended a duration of < 24 hours for all procedures. The authors calculated the total duration of AP in hours and categorized the duration as none, single dose, multiple doses within 24 hours, and >24 hours. The authors also employed a novel metric, potential excess days of therapy (DOT), defined as the number of hours of actual AP minus the expected duration of 24 hours (normalized to days), to quantify the extent of unnecessary perioperative AP exposure.Consistent with previous studies, the authors found that perioperative AP was frequently continued beyond 24 hours at their institution and that prolonged use occurred across all procedure and subspecialty categories. Interestingly, the authors were able to identify 2 specific patterns of AP utilization. Whereas the use of perioperative AP >24 hours appeared to be unintentional in some cases (eg, duration of 26 hours), there was also purposefully prolonged AP for specified periods of time (eg, duration of 7 days). The overall amount of potentially inappropriate antibiotic exposure was substantial, with 5733 potential excess DOT. Finally, the authors identified significant variability in the duration of antibiotic exposure within specific procedures, with the duration ranging from as much as 0 to 429 hours of AP for certain surgeries.Limitations of the study include the use of data from 2017, which precedes more recent guidance from the CDC6 and renewed focus on this topic, thus potentially overestimating the current scope of the problem. The data are also based on a single center experience and may not reflect practice elsewhere, particularly at hospitals that do not have similar ASPs or institutional guidelines. Whereas the authors were able to quantify potential excess antibiotic use, there was limited data on the impact of prolonged antibiotic exposure on either positive (eg, reduced incidence of SSIs) or negative (eg, increased incidence of ADEs) clinical outcomes. The authors did find that the incidence of CDI was more common among patients who received surgical AP versus those who did not; however, methodological issues, including the use of a multiplex gastrointestinal polymerase chain reaction panel to diagnose CDI at the authors’ center, limit the ability to draw any definite conclusions regarding this result.Still, this study is an important addition to the literature. The data are from one of the largest freestanding children’s hospitals in the country, and this study is the most comprehensive attempt at quantifying the duration of excessive perioperative AP across the full complement of pediatric procedures and specialties. Furthermore, the results of the study provide new and provocative information that can help inform future antimicrobial stewardship interventions. First, the authors’ data suggest that previous studies evaluating only the absolute percentage of patients who receive AP >24 hours are missing part of the story. By also looking at potential excess antibiotic DOTs, it becomes clear that the relative contribution to unnecessary antibiotic exposure can be significant, even if the total fraction of procedures with AP given >24 hours is relatively small. For example, although only 11% of ear, nose, and throat procedures received >24 hours of AP, they contributed a significant amount to the total excess antibiotic DOT and may be a high-yield target for perioperative stewardship interventions. Second, it is clear that significant variability exists in the duration of perioperative AP for specific procedures, whereas prolonged AP is common in others. Again, the granularity of the information provided in this study can help ensure that future stewardship efforts are focused on areas of greatest opportunity. Finally, the observation that potential excess antibiotic duration is the result of either purposeful or inadvertent antibiotic prolongation is important, because potential solutions to these issues may be quite different. As pointed out by the authors, automatic stop dates embedded in the electronic health record may help ensure that antibiotics are discontinued within 24 hours; however, if surgeons believe that a fixed duration of antibiotics (eg, 7 days) is key to SSI prevention, additional strategies will be required.Whereas the authors have elucidated several key elements relevant to the scope of prolonged perioperative AP among pediatric patients, critical questions remain. What are the consequences, both positive and negative, of prolonged AP exposure? As was discussed by the authors, in the absence of a well-defined, standard duration of surgical prophylaxis for many pediatric procedures, AP >24 hours can only be considered “potentially” excessive. The absence of clinical data, including the rates of SSI and adverse clinical events, in both this and previous studies, makes weighing the potential risks and benefits of prolonged AP challenging. Furthermore, the risk-benefit calculation may favor a more prolonged course of perioperative AP for select procedures or in certain populations.14 Although a randomized clinical trial comparing specific durations of perioperative AP may be helpful, the relative rarity of SSIs in children and difficulty disentangling the contribution of AP duration from the many other factors that likely contribute to SSIs makes the execution of such studies challenging.1 Instead, national databases, like the American College of Surgeons National Surgical Quality Improvement Pediatric Program (NSQIP-P),15 and well-executed quality improvement efforts are likely necessary to capture patient outcomes and evaluate the impact of specific AP stewardship interventions. In addition to clinical outcomes, an analysis of the financial impact of suboptimal perioperative AP, including the financial consequences of ADEs related to excessive AP, is needed. Although the authors’ experience is unlikely to be unique, it would be helpful to understand the extent to which the same issues exist at other institutions. If there are differences between hospitals, it would be interesting to know why they exist and how best practices can be adopted at other institutions. Finally, even if the problem and associated outcomes are defined, what are the strategies that will help move the needle toward improved perioperative AP use? A variety of quality improvement studies have shown that suboptimal perioperative AP can be improved via the deployment of institutional guidelines, electronic alerts, clinical decision support, and provider education1,16–19; however, qualitative studies that interrogate the factors that contribute to perioperative AP use by surgeons are also necessary.20The study by Salas et al13 sheds needed light on the problem of prolonged perioperative AP and provides a blueprint for future studies aimed at better understanding perioperative AP practices. We look forward to the results of national quality improvement efforts, like NSQIP-P, and encourage surgeons, hospitalists, infectious disease specialists, infection preventionists, and antimicrobial stewardship programs to collaborate on efforts to better define the optimal approach to perioperative AP for pediatric patients.