Publication
Article
Peer Reviewed
Pharmacy Practice in Focus: Health Systems
Implementation of a guidance document for periprocedural antibiotic prophylaxis resulted in improved appropriateness of antibiotic dose, selection, and duration for interventional radiology procedures.
Implementation of a guidance document for periprocedural antibiotic prophylaxis resulted in improved appropriateness of antibiotic dose, selection, and duration for interventional radiology procedures.
Image credit: Arush | stock.adobe.com
Minimal guidance is available regarding antibiotic prophylaxis in interventional radiology (IR) procedures. This retrospective, multicenter, observational pre/post study aimed to assess adherence with periprocedural antibiotic prophylaxis recommendations for IR procedures before and after implementation by a multidisciplinary team at Cone Health in Greensboro, North Carolina. Study results revealed that implementation of a guidance document for periprocedural prophylaxis improved appropriateness of antibiotic dose, selection, and duration.
Interventional radiology (IR) procedures are typically minimally invasive and tend to have fewer infectious complications than open surgical procedures, as small incisions made during IR procedures seldom become sites of clinically significant infection.1 As a result, antibiotic prophylaxis is currently recommended for IR procedures not classified as clean procedures, with clean procedures being defined as performed without active inflammation, not breaking sterile technique, and not involving genitourinary, gastrointestinal, or respiratory tract entry.2 Infectious complications in IR procedures are most often the result of bacteria being introduced into the bloodstream.2 Antibiotic prophylaxis surrounding IR procedures aims to clear bacterial contamination from the bloodstream to prevent infection, seeding of foreign material such as a stent, or necrotic tissue resulting from embolization or ablation.2
Data to support periprocedural prophylaxis for IR procedures are limited. The most recent guidance for antibiotic prophylaxis during IR procedures was provided by the Society of Interventional Radiology and endorsed by the Cardiovascular and Interventional Radiological Society of Europe and the Canadian Association for Interventional Radiology in 2018.2 The risk for bacterial translocation depends on procedure type, access site inflammation, and involved organ system. Certain comorbidities, such as obesity, diabetes, and immunosuppression, increase the risk for infection.2
With limited guidance available regarding the use of antibiotic prophylaxis in IR procedures, IR providers sought assistance for antibiotic recommendations from the Cone Health Antimicrobial Management Program (CHAMP), the local antimicrobial stewardship team at Cone Health. In February 2022, CHAMP partnered with IR providers to author and implement a guidance document for periprocedural antibiotic prophylaxis based on available literature while also considering the hospital formulary. The purpose of this study was to compare adherence to recommended antibiotic prophylaxis before and after implementing an institutional guidance document.
This is a multicenter, prospective study with a historical cohort categorized as nonhuman subjects research by a local institutional review board. Postintervention data of patients undergoing IR procedures from April 2022 to December 2022 were compared with a historical cohort from April 2021 to December 2021 at 3 centers within Cone Health. Patients who underwent IR procedures were identified by procedure type from the electronic medical record. Data collected included procedure type, methicillin-resistant Staphylococcus aureus colonization status, ordering provider, patient demographics, drug allergies, and antibiotics used. The following procedures were included for analysis: port placement, transjugular intrahepatic portosystemic shunt (TIPS), percutaneous nephrostomy tube placement, percutaneous cholecystostomy drain placement, percutaneous cholangiogram/biliary drain placement, kyphoplasty, and transarterial radioembolization (TARE).
The primary objective of this study was to compare the composite of appropriateness of antibiotic selection (Table 1), dosage (Table 2), and duration for periprocedural prophylaxis as indicated in the guidance document (Figure) before and after guidance implementation. This was assessed for a composite of all procedures and each individual procedure. Secondary objectives, evaluated for each individual procedure both before and after guidance implementation, included whether an antibiotic was indicated and given, correct antibiotic selection, correct antibiotic dose based on weight, correct duration, and procedural-related infection for port placement. Data were defined using descriptive statistics (frequencies and percentages). Primary and secondary objectives were assessed using Fisher exact test using Stata version 15.1 (StataCorp LLC).
AML, acute myeloid leukemia; ANC, absolute neutrophil count; BRTO, balloon-occluded retrograde transvenous obliteration; CRP, C-reactive protein; GI, gastrointestinal; HT, hematotoxicity; HD, hemodialysis; IVP, intravenous push; LBCL, large B-cell lymphoma; MCL, mantle cell lymphoma; MM, multiple myeloma; TIPS, transjugular intrahepatic portosystemic shunt
Greater adherence to the guidance document in the postintervention group, as demonstrated by the primary outcome results, occurred for the composite of all procedures (37.80% vs 68.97%, P < .001) (Table 3). Greater adherence was specifically demonstrated for port placement (0.00% vs 100.00%, P = .006), TIPS (15.38% vs 81.82%, P = .003), percutaneous cholangiogram (0.00% vs 50.00%, P = .040), and TARE (0.00% vs 42.86%, P = .002) procedures. No difference between groups was seen for percutaneous nephrostomy (0.00% vs 33.00%, P = .46), percutaneous cholecystostomy (0.00% vs 10.00%, P > .99), and kyphoplasty procedures (100.00% vs 100.00%, P > .990). Antibiotic selection and dosage were significantly improved for TIPS and TARE procedures in the postintervention group.
There was no difference in correct antibiotic duration between groups, and this outcome could not be accurately assessed for TARE procedures. No infections related to port placement were identified in the preintervention or postintervention groups. Additionally, there was no difference in the outcome of antibiotic given, if indicated, except for port placement procedures where antibiotics were given incorrectly in the preintervention group and were correctly not given in the postintervention group (0.00% correct vs 100.00% correct, P = .006).
N/A, not available.
Note: n/N is total correct/total number
The study findings indicate a significant improvement in adherence to guidance recommendations across multiple procedures as well as better standardization in periprocedural prophylaxis practices. For almost all procedures, prophylactic antibiotics were not extended beyond preprocedural doses for procedures only requiring prophylactic antibiotics. For most patients undergoing cholecystostomy, antibiotic orders rarely adhered to the guidance document. This is an expected finding, as many patients undergoing this procedure have acute infections such as cholecystitis.3 Appropriate prophylaxis for kyphoplasty procedures may be explained by ordering from a procedural antibiotic order set. Inappropriate antibiotic selection for TARE was likely due to the historical popularity of piperacillin/tazobactam among IR providers in the preintervention group, as piperacillin/tazobactam was the most commonly ordered inappropriate antibiotic. Port placements were selected as a procedure type of interest to evaluate periprocedural antibiotic use for a clean procedure for which antibiotics are not recommended.4,5 Implementation of the guidance document successfully prevented the inappropriate ordering of antibiotics for port placements. Although no procedural-related infections were identified, 1 patient in the preintervention group did have an extensive history of port malfunctions without infection.
The results of this study highlight several opportunities for improvement, including a potential solution for building antibiotic recommendations for IR procedures into an order set. However, communication with IR providers revealed that they are not presently interested in this intervention. Another area for improvement includes reinforcing provider education and potentially involving pharmacist review of the IR procedure schedule to proactively identify inappropriate antibiotic prophylaxis orders and address recommendations prior to the procedure.
Finally, although many antibiotic orders were identified as correct if patients’ records had penicillin allergy labels, it was discovered that many of these allergies were not true allergies and were actually intolerances, such as nausea. Improved assessment and reconciliation of patient allergies prior to ordering procedural antibiotic prophylaxis may permit more optimal antibiotic selection.
Courtney Jackson, PharmD, is a PGY-2 infectious diseases pharmacy resident at Atrium Health Wake Forest Baptist in North Carolina.
Jeremy Frens, PharmD, BCIDP, is a pharmacist at Moses Cone Hospital in Greensboro, North Carolina.
Austin Paytes, PharmD, is a pharmacist at Moses Cone Hospital in Greensboro, North Carolina.
Dylan Suttle, MD, is a physician at Cone Health in Greensboro, North Carolina.
There are several limitations of this study, including retrospective design and the small sample size of patients. Consequently, power was not calculated, and therefore, it is unknown whether the procedures with differences not detected in this study were underpowered to assess differences between groups effectively. Additionally, limitations with chart review did not permit the assessment of the timing of antibiotic administration being within 1 hour of incision time. Further, it was difficult to identify patients using procedure codes, which resulted in a small sample size and underestimation of the true number of IR procedures performed in the study period. As a result, a future direction of this project includes expanding upon this study population with patients from an available database where IR providers currently track procedure outcomes.
Additionally, with TARE, postprocedure antibiotics are recommended for patients without an intact sphincter of Oddi.6 However, documentation of this was not identified from reviewing procedure notes, making it difficult to assess appropriate duration of antibiotic therapy for this procedure type. Therefore, patients undergoing this procedure were determined to meet the primary outcome if their preprocedure antibiotic selection and dose were correct.
In conclusion, implementation of a guidance document for periprocedural antibiotic prophylaxis improved appropriateness of antibiotic dose, selection, and duration.
The authors have nothing to disclose.
