Outcomes Associated with Choice of Prophylactic Antibiotics in Open Fractures

Abstract

Background:

The ideal antibiotic prophylaxis for open fractures is unknown. We evaluated outcomes following different antibiotic prophylaxis regimens for open fractures.

Methods:

This is a secondary analysis of data from PREP-IT. Prophylactic antibiotics were defined as any intravenous antibiotic given on the day of admission. The outcomes were surgical site infection (SSI) within 90 days and reoperation within 1 year. Logistic regression and an instrumental variable analysis that leveraged site-level variation accounted for confounding. Subgroup variation was evaluated by stratifying by Gustilo-Anderson classification (Types I and II versus III).

Results:

Of the 3,331 included participants, the mean age was 45 ± 18 years, 63% were male, 73% were White, 21% were Black, 2% were Asian, and 10% were Hispanic. Cefazolin monotherapy (58% of patients), ceftriaxone monotherapy (10%), and cefazolin plus gentamicin (6%) were the most common regimens. In the instrumental variable analysis, the odds of infection did not significantly differ with ceftriaxone use (odds ratio [OR], 1.24; 95% confidence interval [CI], 0.70 to 2.20; p = 0.45) or cefazolin plus gentamicin use (OR, 0.25; 95% CI, 0.03 to 2.04; p = 0.20) compared with cefazolin monotherapy. There were no significant differences between the regimens with respect to infection when stratified by Gustilo-Anderson type. However, we did observe a nearly 3-fold increase in the odds of infection with ceftriaxone use compared with cefazolin monotherapy (OR, 2.73; 95% CI, 0.96 to 7.79; p = 0.06) in Type-I and II fractures, and a 75% decrease in the odds of infection with cefazolin plus gentamicin use (OR, 0.25; 95% CI, 0.03 to 2.02; p = 0.19) compared with cefazolin monotherapy in Type-III fractures.

Conclusions:

Among patients with open fractures, antibiotic prophylaxis with ceftriaxone monotherapy did not provide significant benefits compared with cefazolin monotherapy in preventing infection in Type-I and II fractures. The findings suggest that cefazolin plus gentamicin might reduce the odds of infection in Type-III fractures compared with cefazolin monotherapy, but this difference was not statistically significant.

Level of Evidence:

Therapeutic Level II. See Instructions for Authors for a complete description of levels of evidence.

While the need for prophylactic antibiotics for open fractures is widely accepted^1–3, the optimal regimen and duration remain unclear due to side effects, protocol complexity, variation in antibiograms⁴, and antibiotic-resistant organisms^5–13. A subanalysis of the Program of Randomized Trials to Evaluate Preoperative Antiseptic Skin Solutions in Orthopaedic Trauma (PREP-IT) that was performed midway through study enrollment showed low adherence to the recommended guidelines for prophylaxis, with 60% compliance with first-generation cephalosporin monotherapy for Type-I and II fractures and <20% compliance with the cephalosporin plus an aminoglycoside for Type-III fractures¹⁴.

With the conclusion of PREP-IT^15,16, we evaluated the association between different antibiotic regimens and the development of surgical site infection (SSI).

Materials and Methods

Study Design and Participants

This is a secondary analysis of data from PREP-IT, the details of which have been previously described¹⁷. Patients were included if they were ≥18 years old and underwent surgical fixation of an open extremity fracture. Patients were excluded if they were initially managed at a non-participating hospital, had an active infection, or did not complete follow-up. Fractures were classified with use of both the Gustilo-Anderson classification^18,19 and the AO/OTA classification^20,21 by the attending surgeon at the initial debridement. Race and ethnicity were self-reported by the study participants. Ethics approval for this secondary analysis was obtained from the Hamilton Integrated Research Ethics Board (HiREB, #17587).

Study Exposure

A prophylactic antibiotic was defined as any intravenous antibiotic that was started the same calendar day as admission. The duration of antibiotic use was calculated by noting each contiguous calendar day from admission in which the patient received at least 1 dose of that antibiotic.

Study Outcome

The primary outcome was SSI as defined by the 2017 Centers for Disease Control and Prevention guideline²² and included superficialincisionalinfectionsoftheskinorsubcutaneoustissuewithin 30 days and subfascial infections involving the fractured bone or joint within 90 days postoperatively. The secondary outcome was unplanned fracture-related reoperation within 1 year for infection or for wound- or fracture-healing complications.

Statistical Analysis

Patient characteristics are described using medians with interquartile ranges (IQRs) for continuous data and counts with percentages for categorical data. We assessed whether the distribution of these characteristics varied by antibiotic regimen with use of Kruskal-Wallis tests for continuous data and chi-square tests for categorical data. The top 3 antibiotic regimens were analyzed because they constituted 74% of the cohort, whereas the remaining 26% was distributed among 77 different regimens, each representing <5% of the population studied, which precluded meaningful analysis of the event outcomes.

We evaluated the associations between the antibiotic regimens and our study outcomes with use of logistic regression and instrumental variable analyses. For the instrumental variable analysis, we classified each participating hospital based on its propensity to use each antibiotic regimen. We assessed the association between antibiotic duration and SSI with use of logistic regression, adjusting for antibiotic regimen, age, fracture location, Gustilo-Anderson classification, embedded wound contamination, and Injury Severity Score. The associations between the study exposures and outcomes are reported as odds ratios with 95% confidence intervals. The threshold for significance was set at an alpha of 0.05, and we did not adjust for multiple comparisons. The statistical analyses were performed using R (version 4.2.2; R Foundation for Statistical Computing).

Results

Patient Characteristics

A total of 3,338 patients met the eligibility criteria. We excluded 7 (0.2%) of the patients because intravenous antibiotics were not initiated on the day of hospital admission. Of the study participants who were included, the mean age (and standard deviation) was 45 ±18 years, 63% were male, 73% were White, 21% were Black, 2% were Asian, and 10% were Hispanic. The most common antibiotic regimens that were administered were cefazolin monotherapy (58% of patients), ceftriaxone monotherapy (10%), and cefazolin plus gentamicin (6%) (Table I). Regimens that each constituted <1% of the study cohort were collectively categorized as “other combinations.” Figure 1 demonstrates the frequency of antibiotic usage across the different study sites, stratified by fracture type.

TABLE I -.

Most Common Intravenous Antibiotic Regimens Initiated on the Day of Hospital Admission^*

Regimen	Overall (N = 3331)	Gustilo-Anderson I or II (N = 1879)	Gustilo-Anderson IIIA (N = 1282)	Gustilo-Anderson IIIB or IIIC (N = 170)
Cefazolin only	1943 (58%)	1273 (68%)	613 (48%)	57 (34%)
Ceftriaxone only	344 (10%)	181 (10%)	147 (11%)	16 (9%)
Cefazolin and gentamicin	207 (6%)	51 (3%)	134 (10%)	22 (13%)
Cefazolin and vancomycin	123 (4%)	65 (3%)	55 (4%)	3 (2%)
Cefazolin and another antibiotic	77 (2%)	50 (3%)	26 (2%)	1 (1%)
Cefazolin and tobramycin	58 (2%)	29 (2%)	24 (2%)	5 (3%)
Clindamycin only	44 (1%)	28 (1%)	14 (1%)	2 (1%)
Vancomycin only	16 (0.5%)	10 (0.5%)	3 (0.2%)	3 (2%)
Piperacillin + tazobactam only	9 (0.3%)	0 (0%)	9 (0.7%)	0 (0%)
Other combinations	510 (15%)	192 (10%)	257 (20%)	61 (36%)

^*Values are given as the number of patients, with the percentage in parentheses.

Fig. 1.

Antibiotic regimens at the 34 participating sites, stratified by Gustilo-Anderson type.

There were significant differences in patient characteristics between the different antibiotic groups (Table II). Patients with a lower-extremity fracture, more wound contamination, or a more severe Gustilo-Anderson type were more likely to receive cefazolin plus gentamicin.

TABLE II -.

Patient Characteristics by Intravenous Antibiotic Regimen^*

Characteristic	Cefazolin (N = 1943)	Ceftriaxone (N = 344)	Cefazolin and Gentamicin (N = 207)	Other (N = 837)	P Value
Age (yr)	42 (29, 59)	44 (29, 59)	40 (29, 53)	43 (29, 58)	0.27
Sex†					0.18
Female	713 (37%)	144 (42%)	69 (33%)	318 (38%)
Male	1228 (63%)	200 (58%)	138 (67%)	519 (62%)
Race					<0.001
White	1,363 (70%)	268 (78%)	139 (67%)	676 (81%)
Black	465 (24%)	66 (19%)	62 (30%)	109 (13%)
Asian	49 (3%)	2 (1%)	2 (1%)	13 (2%)
Other	66 (3%)	8 (2%)	4 (2%)	39 (5%)
Hispanic	209 (11%)	22 (6%)	25 (12%)	70 (8%)	0.02
BMI (kg/m2)	28 (24, 32)	28 (24, 33)	28 (24, 34)	28 (24, 33)	0.20
No. of open fractures					<0.001
1	1829 (94%)	311 (90%)	182 (88%)	752 (90%)
2	102 (5%)	30 (9%)	20 (10%)	73 (9%)
3	12 (1%)	3 (1%)	5 (2%)	12 (1%)
Lower-extremity fracture	1391 (72%)	259 (75%)	178 (86%)	658 (79%)	<0.001
Gustilo-Anderson classification					<0.001
Type I	565 (29%)	76 (22%)	13 (6%)	136 (16%)
Type II	708 (36%)	105 (31%)	38 (18%)	238 (28%)
Type IIIA	613 (32%)	147 (43%)	134 (65%)	388 (46%)
Type IIIB	37 (2%)	11 (3%)	20 (10%)	54 (6%)
Type IIIC	20 (1%)	5 (2%)	2 (1%)	21 (3%)
Embedded wound contamination	122 (6%)	29 (8%)	40 (19%)	126 (15%)	<0.001
Workplace injury	170 (9%)	35 (10%)	24 (12%)	96 (11%)	0.12

^*Values are given as the median, with the IQR in parentheses, or as the number of patients, with the percentage in parentheses.

BMI = body mass index.

^†Sex was not reported by 2 participants in the cefazolin group.

Study Outcomes

The number of patients who completed follow-up was 3,216 (97%) for 90 days and 2,999 (90%) for the full year. A total of 231 patients (6.9%) developed the study’s primary outcome (Table III). We observed significant associations between the antibiotic groups and the study outcomes in the unadjusted analyses (Table III). However, in the instrumental variable analysis to adjust for confounding, there were no significant associations between the different antibiotic groups and the study outcomes (Table III). Although the differences were not statistically significant, the point estimates suggest higher odds of infection and reoperation with ceftriaxone use compared with cefazolin monotherapy and lower odds of infection and reoperation with cefazolin plus gentamicin use compared with cefazolin monotherapy (Table III). When patients were stratified by Gustilo-Anderson classification, ceftriaxone use was associated with a nearly threefold increase in the odds of SSI compared with cefazolin monotherapy among Type-I or II open fractures in the instrumental variable analysis, although this difference did not reach significance (OR, 2.73; 95% CI, 0.96 to 7.79; p = 0.06; Table IV). In Type-III fractures, the findings suggest a 75% reduction in the odds of infection with cefazolin plus gentamicin use compared with cefazolin monotherapy (OR, 0.25; 95% CI, 0.03 to 2.02; p = 0.19), but this difference did not reach statistical significance.

TABLE III -.

Association of Antibiotic Regimen with Study Outcomes^*

Outcome	No. (%) of Patients†	Unadjusted		Instrumental Variable‡
		OR (95% CI)	P Value	OR (95% CI)	P Value
SSI within 90 days
Cefazolin	107 (5.7%)	Ref.		Ref.
Ceftriaxone	27 (8.1%)	1.46 (0.93–2.24)	0.09	1.24 (0.70–2.20)	0.45
Cefazolin and gentamicin	22 (10.9%)	2.03 (1.22–3.23)	0.004	0.25 (0.03–2.04)	0.20
Other	75 (9.2%)	1.69 (1.24–2.30)	<0.001	-	-
Reoperation for infection within 12 months
Cefazolin	133 (6.9%)	Ref.		Ref.
Ceftriaxone	31 (9.0%)	1.35 (0.88–2.00)	0.15	1.47 (0.84–2.56)	0.17
Cefazolin and gentamicin	28 (13.5%)	2.13 (1.35–3.24)	<0.001	0.34 (0.05–2.49)	0.29
Other	85 (10.2%)	1.54 (1.15–2.04)	0.003	-	-
Reoperation for fracture-healing complications within 12 months
Cefazolin	121 (7.0%)	Ref.		Ref.
Ceftriaxone	35 (11.0%)	1.65 (1.10–2.43)	0.01	1.63 (0.92–2.89)	0.09
Cefazolin and gentamicin	27 (14.2%)	2.21 (1.39–3.41)	<0.001	0.85 (0.12–5.86)	0.87
Other	53 (6.9%)	1.00 (0.71–1.38)	0.98	-	-

^*OR = odds ratio, CI = confidence interval.

^†Values are given as the number and percentage of patients with that event in that antibiotic regimen subgroup and include only those patients who completed follow-up.

^‡Instrumental variable analyses were adjusted for Gustilo-Anderson classification.

TABLE IV -.

Subgroup Analysis of the Association of Antibiotic Regimen with SSI, by Gustilo-Anderson Classification^*

Subgroup	No. (%) of Patients with SSI	Unadjusted		Instrumental Variable†
		OR (95% CI)	P Value	OR (95% CI)	P Value
Gustilo-Anderson Type I or II (N = 1861)
Cefazolin	44 (3.6%)	Ref.		Ref.
Ceftriaxone	11 (6.3%)	1.79 (0.86–3.41)	0.09	2.73 (0.96–7.79)	0.06
Cefazolin and gentamicin	4 (8.0%)	2.32 (0.68–6.03)	0.12	0.94 (0.01–17602)	0.99
Other	17 (4.8%)	1.34 (0.74–2.33)	0.33	-	-
Gustilo-Anderson Type III (N = 1470)
Cefazolin	63 (9.6%)	Ref.		Ref.
Ceftriaxone	16 (10.2%)	1.07 (0.58–1.87)	0.81	0.91 (0.45–1.81)	0.78
Cefazolin and gentamicin	18 (11.9%)	1.28 (0.71–2.19)	0.39	0.25 (0.03–2.02)	0.19
Other	58 (12.8%)	1.39 (0.95–2.03)	0.09	-	-

^*Excludes patients with missing outcome data.

OR = odds ratio, CI = confidence interval.

^†Instrumental variable analyses were adjusted for Gustilo-Anderson classification.

The median duration of antibiotics was 3 days (IQR, 2 to 3 days) for Type-I and II open fractures and 3 days (IQR, 2 to 4 days) for Type-III fractures. Among the 1,956 patients who received antibiotics for >2 days, 108 (5.5%) had Type-IIIB fractures, and 107 of the 108 were awaiting soft-tissue coverage. After adjusting for fracture severity, there was no difference in patient SSI rate based on the duration of therapy (Table V). However, an antibiotic duration of 4 days was associated with a nearly 48% increase in the risk of developing SSI, although the difference did not reach significance (OR, 1.48; 95% CI, 0.97 to 2.22; p = 0.06).

TABLE V -.

Association of Antibiotic Duration with Surgical Site Infection within 90 Days^*

Antibiotic Duration	No. (%) of Patients with SSI†	Unadjusted		Adjusted‡
		OR (95% CI)	P Value	OR (95% CI)	P Value
≤2 days	71 (5.4%)	Ref.		Ref.
3 days	56 (5.9%)	1.09 (0.76–1.57)	0.63	0.99 (0.68–1.43)	0.96
4 days	46 (10.6%)	2.09 (1.41–3.07)	<0.001	1.48 (0.97–2.22)	0.06
≥5 days	58 (11.4%)	2.25 (1.56–3.24)	<0.001	1.22 (0.81–1.82)	0.33

^*OR = odds ratio, CI = confidence interval.

^†Percentages are based on the patients who completed follow-up (1318, 955, 433, and 510 patients for ≤2, 3, 4, and ≥5 days of antibiotics, respectively).

^‡The adjusted analysis controlled for antibiotic regimen, age, fracture location, Gustilo-Anderson classification, embedded wound contamination, and Injury Severity Score. There was no evidence of subgroup effects based on embedded wound contamination or Gustilo-Anderson Type III versus Type I or II.

Discussion

We found that the antibiotic regimen that was used for open fracture prophylaxis did not significantly influence the odds of either short- or long-term infectious complications when stratified by Gustilo-Anderson type. Nearly all PREP-IT patients received antibiotics on the day of admission, with the most common regimens being cefazolin monotherapy, ceftriaxone monotherapy, and cefazolin with gentamicin. While there were no significant differences in the outcomes between these three choices, we did observe that ceftriaxone use was strongly associated with increased risks of infection and reoperation. A large portion of the study cohort did not receive any of the regimens recommended by traditional guidelines. We also found no benefit of >2 days of prophylactic antibiotic usage.

In this final cohort, <1% of patients received piperacillin + tazobactam, which is a notable difference from a prior interim study¹⁴. Although piperacillin + tazobactam has been proposed as a broad-spectrum single-infusion option for open fracture prophylaxis^5,23,24, its low usage in this cohort precluded any meaningful analysis.

Ceftriaxone was the second most common antibiotic used in this study, with some individual centers appearing to use it for nearly all open fractures. Its use in Type-I and II fractures was associated with a nearly threefold increase in infection, although this increase did not reach significance. There was also an association with increased reoperation risk within 1 year. Despite the increased spectrum of coverage with ceftriaxone, the trend toward decreased efficacy compared with cefazolin monotherapy for both prophylaxis and the treatment of common infections has been reported in non-orthopaedic studies, particularly with regard to methicillin-sensitive Staphylococcus aureus^25–27, which is the most common bacteria found in fracture-related infections^28–32. Similarly, ceftriaxone has been shown to be inferior to cefazolin monotherapy for antimicrobial prophylaxis, even in colorectal surgery^33,34. It has been proposed that ceftriaxone has slower tissue penetration and requires higher tissue concentrations to achieve the same bactericidal efficacy^25,35. Although ceftriaxone use has been proposed as an effective simplification of the recommended open fracture regimen^8,36, it may not be as effective as previously thought.

A striking finding from our analysis is the lack of a difference in outcomes among Type-III fractures, despite 46% of patients with Type-III fractures not receiving broad-spectrum gram-negative coverage. While typically considered a gram-positive agent, cefazolin has activity against some gram-negative organisms, including Escherichia coli and Klebsiella sp., which are among some of the most common gram-negative pathogens isolated from fracture-related infections^29,32. There is also growing evidence that broader gram-negative coverage beyond the spectrum of activity of cefazolin in Type-III fractures may not reduce infection. Bankhead-Kendall et al. performed a retrospective comparison of 126 Type-III fractures, with half of the patients receiving only a first-generation cephalosporin, and found no difference in SSI when compared with patients who received an additional aminoglycoside³⁷. Similarly, Suzuki et al. reviewed 1,041 Type-III open fractures from a multicenter fracture registry and found no difference in infection rates between the gram-positive coverage group (8%) and the broad-spectrum coverage group (9%)³⁸. Given the lack of clear evidence that broad gram-negative prophylaxis reduces infection risk in Type-III open fractures, as well as the concerns over antibiotic stewardship, some groups have recommended against routine expanded gram-negative coverage³⁹ or aminoglycoside use⁴⁰ in these fracture types.

Multiple other factors may have more influence on SSI following an open fracture than the spectrum of prophylactic antibiotic coverage. The time to antibiotic administration, time to definitive coverage, time to arrival at the definitive treatment center, and use of topical antibiotics have all been shown to affect infection risk and may have played a larger role than the prophylactic antibiotic choice in the present study^1,3,41–48.

Our finding that there was little protective benefit with an extended duration of antibiotics agrees with previous studies. While most of these studies were retrospective, 1 prospective randomized trial compared 24 hours versus 5 days of a second-generation cephalosporin and found that 24 hours of antibiotic prophylaxis was noninferior⁴⁹. This finding has been supported by subsequent systematic reviews and meta-analyses^2,50. In contrast, we found that prophylactic antibiotic delivery beyond 2 days did not provide any benefit; however, it is possible that this is because we used calendar days, rather than hours, to define time.

A significant strength of this study is its use of a dataset that was collected for a large, multicenter, international, prospective, high-quality group of trials focused on fracture-related infections. All study sites had to demonstrate robust research infrastructure and patient volumes prior to participation in the study. Thus, there was a wealth of detailed and complete data collected regarding wounds, fracture classifications, patient risk factors, and clinical interventions, as well as <5% loss to follow-up for the primary outcome. Furthermore, there was no difference observed in either of the 2 studies under PREP-IT with respect to different skin preparations for open fractures, so it is unlikely that the actual trial intervention had an effect on the infection rates^15,16.

The weaknesses of this study are also related to the fact that it is a secondary analysis of a group of large randomized controlled trials. The antibiotic regimens were not a priori interventions, which resulted in a large number of different protocols and limited our ability to analyze outcomes. We did not have detailed data regarding time to antibiotic delivery, dosage, or other indications for antibiotics to confirm the use of the antibiotic as an agent for prophylaxis. Finally, the fact that all data were collected following screening for a group of randomized controlled trials limits the generalizability of our findings. Although >8,000 patients with open fractures were screened for the 2 studies under PREP-IT, less than half were enrolled. Additionally, all sites were major trauma centers, so the patient outcomes may not reflect community practice. Nonetheless, this study most likely reflects the most current practice patterns across North America.

Despite wide variation in the choice of antibiotic regimens for prophylaxis of infection in open fractures, cefazolin monotherapy was the most common antibiotic prophylaxis used in patients who were enrolled in the PREP-IT trials, across all levels of fracture severity. Multiple other factors, such as the time to antibiotic administration^1,3, time to definitive coverage^41–43, time to arrival at the definitive treatment center⁴⁵, and use of topical antibiotics^46–48, may have a larger influence on the development of SSI following an open fracture. ■

Members of the PREP-IT Investigators include:

Executive Committee: Gerard P. Slobogean (Principal Investigator, University of Maryland School of Medicine, Baltimore, MD); Sheila Sprague (Principal Investigator, McMaster University, Hamilton, ON); Jeffrey Wells (Patient Representative, Trauma Survivors Network, Falls Church, VA); Mohit Bhandari (Principal Investigator, McMaster University, Hamilton, ON)

Steering Committee: Gerard P. Slobogean (Co-Chair, University of Maryland School of Medicine, Baltimore, MD); Mohit Bhandari (Co-Chair, McMaster University, Hamilton, ON); Sheila Sprague (Principal Investigator, McMaster University, Hamilton, ON); Anthony D. Harris (University of Maryland School of Medicine, Baltimore, MD); C. Daniel Mullins (University of Maryland School of Pharmacy, Baltimore, MD); Lehana Thabane (McMaster University, Hamilton, ON); Jeffrey Wells (Trauma Survivors Network, Falls Church, VA); Amber Wood (Association of periOperative Registered Nurses, Denver, CO)

Adjudication Committee: Gregory J. Della Rocca (Chair, University of Missouri, Columbia, MO); Anthony D. Harris (University of Maryland School of Medicine, Baltimore, MD); Joan Hebden (University of Maryland School of Medicine, Baltimore, MD); Kyle J. Jeray (Greenville Health System, Greenville, SC); Lucas S. Marchand (University of Utah, Salt Lake City, UT); Lyndsay M. O’Hara (University of Maryland School of Medicine, Baltimore, MD); Robert Zura (LSU Health, New Orleans, LA); Christopher Lee (University of California, Los Angeles, Los Angeles, CA); Joseph Patterson (University of Southern California, Los Angeles, CA)

Data and Safety Monitoring Committee: Michael J. Gardner (Chair, Stanford University School of Medicine, Palo Alto, CA); Jenna Blasman (Patient Representative, Kitchener, ON); Jonah Davies (University of Washington, Seattle, WA); Stephen Liang (Washington University, St. Louis, MO); Monica Taljaard (Ottawa Hospital Research Institute, Ottawa, ON)

Research Methodology Core: P.J. Devereaux (McMaster University, Hamilton, ON); Gordon H. Guyatt (McMaster University, Hamilton, ON); Lehana Thabane (McMaster University, Hamilton, ON); Diane Heels-Ansdell (McMaster University, Hamilton, ON)

Patient Centered Outcomes Core: Debra Marvel (Patient Representative, Baltimore, MD); Jana Palmer (Patient Representative, Baltimore, MD); Jeffrey Wells (Patient, Trauma Survivors Network, Falls Church, VA); Jeff Friedrich (Editor, Slate Magazine, Washington, DC); C. Daniel Mullins (University of Maryland School of Medicine, Baltimore, MD); Nathan N. O’Hara (University of Maryland School of Medicine, Baltimore, MD); Frances Grissom (Trauma Survivor Network, Baltimore, MD)

Orthopaedic Surgery Core: Gregory J. Della Rocca (University of Missouri, Columbia, MO); I. Leah Gitajn (Dartmouth University, Hanover, NH); Kyle J. Jeray (Greenville Health System, Greenville, SC); Saam Morshed (San Francisco General Hospital, San Francisco, CA); Robert V. O’Toole (University of Maryland School of Medicine, Baltimore, MD); Bradley A. Petrisor (Hamilton Health Sciences, Hamilton, ON)

Operating Room Core: Franca Mossuto (Hamilton Health Sciences, Hamilton, ON)

Infectious Disease Core: Anthony D. Harris (University of Maryland School of Medicine, Baltimore, MD); Manjari G. Joshi (University of Maryland School of Medicine, Baltimore, MD)

Military Core: Jean-Claude D’Alleyrand (Landstuhl Regional Medical Center, Landstuhl, Germany); Justin Fowler (Brooke Army Medical Center, Fort Sam Houston, TX); Jessica Rivera (Brooke Army Medical Center, Fort Sam Houston, TX); Max Talbot (Canadian Armed Forces, Montreal, QC)

McMaster University Methods Center (Hamilton, ON): Sheila Sprague (Principal Investigator); Mohit Bhandari (Principal Investigator); David Pogorzelski (Research Coordinator); Shannon Dodds (Research Coordinator); Jordan Leonard (Research Coordinator); Silvia Li (Research Coordinator); Alejandra Rojas (Research Coordinator); Gina Del Fabbro (Research Coordinator); Olivia Paige Szasz (Research Coordinator); Diane Heels-Ansdell (Statistician); Sofia Bzovsky (Statistical Analyst), Paula McKay (Manager); Alexandra Minea (Research Coordinator); Jodi Gallant (Research Coordinator); Kevin Murphy (Research Coordinator)

University of Maryland School of Medicine Administrative Center (Baltimore, MD): Gerard P. Slobogean (Principal Investigator); Nathan N. O’Hara (Manager); Andrea Howe (Project Manager); Yasmin Degani (Project Manager); Haley Demyanovich (Project Manager)

University of Maryland School of Pharmacy, The PATIENTS Program (Baltimore, MD): C. Daniel Mullins (Executive Director); Michelle Medeiros (Director of Research); Genevieve Polk (Assistant Director, Dissemination and Research); Eric Kettering (Senior Instructional Technology and Dissemination Specialist); Nirmen Mahal (Program Specialist)

PREP-IT Clinical Sites:

Lead Clinical Site (Aqueous-PREP and PREPARE):

R Adams Cowley Shock Trauma Center, University of Maryland School of Medicine, Baltimore, MD: Gerard P. Slobogean, Robert V. O’Toole, Nathan O’Hara, Gregory Schrank, Andrew Eglseder, Aaron Johnson, Christopher Langhammer, Jason Nascone, Raymond Pensy, Andrew Pollak, Marcus Sciadini, Eric Hempen, Yasmin Degani, Andrea Howe, Megan Camara, Heather Phipps, Joshua Rudnicki, Christine Holler, Katherine Aumann, Jennifer Spelta, Jed Nieves, Bryce Smetzer, Julia Dunning, Kimberly Krudener, Erin Lock, Lea Marineau

Aqueous-PREP and PREPARE:

Hamilton Health Sciences – General Site, Hamilton, ON: Brad A. Petrisor, Herman Johal, Bill Ristevski, Dale Williams, Matthew Denkers, Krishan Rajaratnam, Jamal Al-Asiri, Jodi Gallant, Kaitlyn Pusztai, Sarah MacRae, Sara Renaud, Marko Gjorgjievski

Prisma Health, Greenville, SC: Kyle J. Jeray, John D. Adams, Michael L. Beckish, Christopher C. Bray, Timothy R. Brown, Andrew W. Cross, Timothy Dew, Gregory K. Faucher, Richard W. Gurich Jr., David E. Lazarus, S. John Millon, M. Christian Moody, M. Jason Palmer, Scott E. Porter, Thomas M. Schaller, Michael S. Sridhar, John L. Sanders, L. Edwin Rudisill Jr., Michael J. Garitty, Andrew S. Poole, Michael L. Sims, Clark M. Walker, Robert Carlisle, Erin A. Hofer, Brandon Huggins, Michael Hunter, William Marshall, Shea B. Ray, Cory Smith, Kyle M. Altman, Erin Pichiotino, Julia C. Quirion, Markus F. Loeffler, Austin A. Cole, Ethan J. Maltz, Wesley Parker, T. Bennett Ramsey, Alex Burnikel, Michael Colello, Russell Stewart, Jeremy Wise, Matthew Anderson, Joshua Eskew, Benjamin Judkins, James M. Miller, Stephanie L. Tanner, Rebecca G. Snider, Christine E. Townsend, Kayla H. Pham, Abigail Martin, Emily Robertson, Emily Bray, J. Wilson Sykes, Krystina Yoder, Kelsey Conner, Harper Abbott

IU Health Methodist Hospital, Indianapolis, IN: Roman M. Natoli, Todd O. McKinley, Walter W. Virkus, Anthony T. Sorkin, Jan P. Szatkowski, Brian H. Mullis, Yohan Jang, Luke A. Lopas, Lauren C. Hill, Maricela M. Diaz, Katelyn M. Garst, Emma W. Denari

San Antonio Military Medical Center, San Antonio, TX: Patrick Osborn, Justin Fowler, Sarah Pierrie

University of California, San Francisco, San Francisco, CA: Saam Morshed, Theodore Miclau, Meir Marmor, Amir Matityahu, R. Trigg McClellan, David Shearer, Paul Toogood, Anthony Ding, Jothi Murali, Ashraf El Naga, Jennifer Tangtiphaiboontana, Tigist Belaye, Eleni Berhaneselase, Dmitry Pokhvashchev

Aqueous-PREP:

Vanderbilt Medical Center, Nashville, TN: William T. Obremskey, Amir Alex Jahangir, Manish Sethi, Robert Boyce, Daniel J. Stinner, Phillip Mitchell, Karen Trochez, Elsa Rodriguez, Charles Pritchett, Natalie Hogan, A. Fidel Moreno

University of Florida, Gainesville, FL: Jennifer E. Hagen, Matthew Patrick, Richard Vlasak, Thomas Krupko, Michael Talerico, Marybeth Horodyski, Marissa Pazik, Elizabeth Lossada-Soto

McGovern Medical School at UTHealth Houston, Houston, TX: Joshua L. Gary, Stephen J. Warner, John W. Munz, Andrew M. Choo, Timothy S. Achor, Milton L. “Chip” Routt, Michael Kutzler, Sterling Boutte, Ryan J. Warth

Wright State University, Dayton, OH: Michael Prayson, Indresh Venkatarayappa, Brandon Horne, Jennifer Jerele, Linda Clark

Banner University Medical Center – Tucson, Tucson, AZ: Christina Boulton, Jason Lowe, John T. Ruth, Brad Askam, Andrea Seach, Alejandro Cruz, Breanna Featherston, Robin Carlson, Iliana Romero, Isaac Zarif

The CORE Institute, Phoenix, AZ: Niloofar Dehghan, Michael McKee, Clifford B. Jones, Debra L. Sietsema, Alyse Williams, Tayler Dykes

Vall d’Hebron University Hospital, Barcelona, Spain: Ernesto Guerra-Farfan, Jordi Tomas-Hernandez, Jordi Teixidor-Serra, Vicente Molero-Garcia, Jordi Selga-Marsa, Juan Antonio Porcel-Vazquez, Jose Vicente Andres-Peiro, Ignacio Esteban-Feliu, Nuria Vidal-Tarrason, Jordi Serracanta, Jorge Nuñez-Camarena, Maria del Mar Villar-Casares, Jaume Mestre-Torres, Pilar Lalueza-Broto, Felipe Moreira-Borim, Yaiza Garcia-Sanchez

Hospital Universitari Parc Tauli, Barcelona, Spain: Francesc Marcano-Fernández, Laia Martínez-Carreres, David Martí-Garín, Jorge Serrano-Sanz, Joel Sánchez-Fernández, Matsuyama Sanz-Molero, Alejandro Carballo, Xavier Pelfort, Francesc Acerboni-Flores, Anna Alavedra-Massana, Neus Anglada-Torres, Alexandre Berenguer, Jaume Cámara-Cabrera, Ariadna Caparros-García, Ferran Fillat-Gomà, Ruben Fuentes-López, Ramona Garcia-Rodriguez, Nuria Gimeno-Calavia, Marta Martínez-Álvarez, Patricia Martínez-Grau, Raúl Pellejero-García, Ona Ràfols-Perramon, Juan Manuel Peñalver, Mònica Salomó Domènech, Albert Soler-Cano, Aldo Velasco-Barrera, Christian Yela-Verdú, Mercedes Bueno-Ruiz, Estrella Sánchez-Palomino, Vito Andriola, Matilde Molina-Corbacho, Yeray Maldonado-Sotoca, Alfons Gasset-Teixidor, Jorge Blasco-Moreu, Núria Fernández-Poch, Josep Rodoreda-Puigdemasa, Arnau Verdaguer-Figuerola, Heber Enrique Cueva-Sevieri, Santiago Garcia-Gimenez

PREPARE:

FRASER HEALTH AUTHORITY/Royal Columbian Hospital, New Westminster, BC: Darius G. Viskontas, Kelly L. Apostle, Dory S. Boyer, Farhad O. Moola, Bertrand H. Perey, Trevor B. Stone, H. Michael Lemke, Ella Spicer, Kyrsten Payne

Inova Fairfax Medical Campus, Falls Church, VA: Robert A. Hymes, Cary C. Schwartzbach, Jeff E. Schulman, A. Stephen Malekzadeh, Michael A. Holzman, Greg E. Gaski, Lolita Ramsey

Wake Forest University School of Medicine, Winston-Salem, NC: Holly Pilson, Eben A. Carroll, Jason J. Halvorson, Sharon Babcock, J. Brett Goodman, Martha B. Holden, Wendy Williams, Taylor Hill, Ariel Brotherton, Janet Syme

MetroHealth Medical Center, Cleveland, OH: Nicholas M. Romeo, Heather A. Vallier, Anna Vergon, Ari Levine, Adam Hirschfeld

University of Utah, Salt Lake City, UT: Thomas F. Higgins, Justin M. Haller, David L. Rothberg, Lucas S. Marchand, Zachary M. Olsen, Abby V. McGowan, Sophia Hill, Morgan K. Dauk

University of Mississippi Medical Center, Jackson, MS: Patrick F. Bergin, George V. Russell, Matthew L. Graves, John Morellato, Sheketha L. McGee, Eldrin L. Bhanat, Ugur Yener, Rajinder Khanna, Priyanka Nehete, Sandra E. Powe

Sanford Health, Sioux Falls, SD: David Potter, Robert VanDemark III, Kyle Seabold, Nicholas Staudenmier

Dartmouth-Hitchcock Medical Center, Lebanon, NH: I. Leah Gitajn, Gerard Chang, Phillip Wolinsky, Devin S. Mullin, Theresa A. Chockbengboun, Jon Mikael Anderson, Bethany Malskis, Lillian Fisher

Atrium Health Musculoskeletal Institute, Carolinas Medical Center, Charlotte, NC: Kevin Phelps, Michael Bosse, Madhav Karunakar, Laurence Kempton, Stephen Sims, Joseph Hsu, Rachel Seymour, Christine Churchill, Ada Mayfield, Juliette Sweeney

Capital Region Health, University of Maryland, Largo, MD: Todd Jaeblon, Robert Beer, Haley K. Demyanovich, Brent Bauer, Sean Meredith, Sneh Talwar

University of Wisconsin Madison, Madison, WI: Christopher M. Domes, Gabrielle Kuhn

Duke University Hospital, Durham, NC: Mark J. Gage, Rachel M. Reilly, Ariana Paniagua, JaNell Dupree

Brigham and Women’s Hospital, Boston, MA: Michael J. Weaver, Arvind G. von Keudell, Abigail E. Sagona, Devon Brameier, Nishant Suneja

University of Pennsylvania, Philadelphia, PA: Samir Mehta, Derek Donegan, Annamarie Horan, Linda To

Massachusetts General Hospital, Boston, MA: Marilyn Heng, Mitchel B. Harris, David W. Lhowe, John G. Esposito, Ahmad Alnasser, Michael McTague

Bryan Medical Center, Lincoln, NE: Steven F. Shannon, Alesha N. Scott, Bobbi Clinch, Becky Weber

University of Cincinnati, Cincinnati, OH: Michael J. Beltran, Michael T. Archdeacon, Henry Claude Sagi, John D. Wyrick, Theodore Toan Le, Richard T. Laughlin, Cameron G. Thomson, Kimberly Hasselfeld

Cedars-Sinai Medical Center, Los Angeles, CA: Carol A. Lin, Mark S. Vrahas, Charles N. Moon, Milton T. Little, Geoffrey S. Marecek, Denice M. Dubuclet, John M. Garlich

University of California, Irvine, Orange, CA: John A. Scolaro, James R. Learned, Philip K. Lim, Susan Demas, Arya Amirhekmat, Yan Marco Dela Cruz, Hassan Abdallah