Piperacillin/Tazobactam versus Tobramycin-Based Antibiotic Prophylaxis for Type III Open Fractures

Abstract

Background:

Type III open fractures are associated with an infection rate as high as 50%. The optimal antibiotic for open fracture prophylaxis remains unclear, and the literature comparing the safety and efficacy of different antibiotic regimens is limited. The aim of this study was to compare the composite adverse events (AEs) in patients before and after a change in prophylactic antibiotic management for these injuries from a tobramycin- to a piperacillin/tazobactam-based regimen.

Methods:

This was a retrospective single-center cohort study of patients with Type III open fractures admitted from January 2010 to December 2016. Patients were included if they received either tobramycin plus cefazolin or clindamycin or piperacillin/tazobactam for fracture prophylaxis. The primary outcome was the rate of composite AEs, which included nephrotoxicity, surgical site infection (SSI), and hospital re-admission with surgical intervention. Secondary outcomes included the rate of SSI within 30 and 60 days after injury. Data were analyzed using the Student t-, Mann-Whitney U, and Fisher exact tests.

Results:

Eighty-five patients were included. There were 29 events in the tobramycin group compared with three in the piperacillin/tazobactam group. At 30 days, SSI had occurred in 17 patients (27.5%) in the tobramycin group and 1 patient (4.3%) in the piperacillin/tazobactam group (p = 0.033). At 60 days, SSI had occurred in three additional patients in the tobramycin group (p = 0.009).

Conclusion:

There was no difference in the composite AEs in the piperacillin/tazobactam compared with the tobramycin group. However, SSI within 30 and 60 days was significantly more common with tobramycin.

Open fractures are associated with significant morbidity. Infection is the most common serious complication and can result in osteomyelitis, loss of function, or amputation. Acute infection rates in Gustilo's Type III open fractures range from 20%–50% as reported in the literature depending on the regimen used for treatment [1,2]. Surgical debridement and stabilization continue to be the mainstay for open fracture management, and administration of prophylactic antibiotics has become part of the standard of care, decreasing the frequency of infection six-fold if given within three hours of injury [2 –5].

In 1984, Gustilo and colleagues demonstrated that 77% of Type III open-fracture infections were caused by gram-negative bacteria with the most common being Enterobacter spp. (33%) followed by Escherichia coli (21%) and Pseudomonas spp. (8%) [6,7]. The authors concluded that the addition of an aminoglycoside to a first-generation cephalosporin was needed to provide better coverage. A third-generation cephalosporin was recommended as an alternative to avoid aminoglycoside toxicity.

Despite these findings, the controversy continues over the ideal antibiotic prophylaxis regimen and the role of aminoglycosides for Type III open fractures. Although aminoglycosides provide excellent gram-negative coverage, there are challenges with their use. Specifically, aminoglycosides penetrate bone poorly and have high rates of nephrotoxicity [8 –12]. Also, achievement of therapeutic concentrations may be delayed in the trauma setting because of errors in the estimation of patient weight. Moreover, lower antibiotic peak concentrations may result from alterations in the volume of distribution after fluid resuscitation or blood product administration [13]. Finally, it remains unclear whether aminoglycosides achieve optimal concentrations for bactericidal activity because of the enhanced renal clearance in younger trauma patients, alterations in volume of distribution, and intra-patient variability.

Given the challenges and toxicities associated with aminoglycoside use in the trauma setting, many institutions have replaced these drugs with broad-spectrum beta-lactam agents such as piperacillin/tazobactam [14 –17]. Piperacillin/tazobactam provides a spectrum of activity similar to that of a cephalosporin plus an aminoglycoside, and it adds coverage against Clostridium. Piperacillin/tazobactam has a better safety profile and distribution into bone and damaged tissue than the aminoglycosides [11,12]. However, clinical data on the efficacy and safety of piperacillin/tazobactam compared with aminoglycosides in the setting of open fractures are lacking. Therefore, the authors undertook this study to compare the rate of adverse events (AEs) between high-dose tobramycin plus cefazolin or clindamycin and piperacillin/tazobactam for antibiotic prophylaxis after Type III open fractures.

Patients and Methods

Study design

This was a retrospective cohort study at a single academic Level 1 trauma center with more than 2,500 injuries per year. Patients with Gustilo-Anderson Type III open fractures who presented to the study institution between January 2010 and December 2016 were identified by a query of the hospital trauma database. Patients were included if they were 18 years of age or older and if they received high-dose tobramycin plus either cefazolin or clindamycin (for penicillin-allergic patients) or piperacillin/tazobactam. The exclusion criteria were death within 24 hours of hospital admission, injuries that resulted in amputation during the initial admission, incarcerated patients, patients who received both study antibiotics (tobramycin and piperacillin/tazobactam) for prophylaxis, or those receiving renal replacement therapy prior to admission. Patients who were given traditional-dose tobramycin (<5 mg/kg) and those who received study antibiotics for other indications during the initial 48 hours of hospitalization also were excluded. Patients who received either study drug for other indications were included only if the study antibiotics were started after the initial open fracture management (48 hours after injury or 24 hours after wound closure). The study was approved by the local Biomedical Institutional Review Board.

All patients included in the study received “standard of care treatment,” which included early evaluation of the fracture(s) by an orthopaedic on-call resident; irrigation with normal saline, Betadine or both; closed reduction; and splinting while in the trauma bay. Antibiotic selection was based on initial evaluation and typing of fractures based on the Gustilo-Anderson classification in the trauma bay. Fractures ungraded in the medical records were graded retrospectively according to the Gustilo-Anderson fracture classification by an attending orthopaedic trauma surgeon following review of imaging, the operative report(s), and the complete medical records. Patients were included if the orthopaedic trauma surgeon graded the extremity fracture Type III. For patients with normal renal function, antibiotics were dosed as follows: Piperacillin/tazobactam 4.5 g every 8 hours (first dose administered over 30 minutes and later doses over 4 hours), cefazolin 2g every 8 hours, and tobramycin 7 mg/kg dosed according to hospital standard practice.

Operational definitions

Adverse events included surgical site infection (SSI), nephrotoxicity, and hospital re-admission with surgical intervention because of infection.

An SSI was defined as chart documentation of superficial or deep SSI, positive site culture, need for additional surgical intervention after wound closure, or documentation of starting antibiotic therapy for confirmed or suspected infection related to the open fracture.

Nephrotoxic agents were defined using the study by Gerlach and colleagues [8] and included vancomycin, iodinated contrast medium, vasopressors, diuretics, angiotensin-converting enzyme inhibitors, angiotensin II receptor blockers, and non-steroidal anti-inflammatory drugs.

Immunosuppressed patients were those with documented human immunodeficiency virus infection plus acquired immunodeficiency syndrome defining illness or CD4 < 200 cells/mcl, and those who received immunosuppressive therapy (cytotoxic agents within 30 days, corticosteroids at a dose of 15 mg/day or more of prednisone or equivalent for >2 weeks, and anti-rejection medications within two weeks before admission with open fracture).

Nephrotoxicity was defined as acute kidney injury (AKI) with a 50% or more increase in serum creatinine concentration from baseline per the Kidney Disease Improving Global Outcomes (KDIGO) criteria [18].

Data and outcomes

Demographic data including age, gender, admission weight, mechanism of injury, and hospital and intensive care unit length of stay were collected. Fracture location, chronological data (time from injury to Emergency Department [ED] arrival, to first dose of antibiotics, to surgical intervention, and to wound closure), and Injury Severity Score were collected. Times were rounded to the nearest 15 minutes. Data related to antibiotic regimen administered including the antibiotics given, total number of doses, and duration of therapy were collected. Baseline and daily serum creatinine concentration were collected up to 48 hours after the last dose of the study antibiotic for the open fracture, as well as use of the nephrotoxic medications identified by the research team. Co-morbidities including a history of diabetes mellitus, pre-injury immunosuppression, and smoking status also were recorded. In-hospital admission and progress notes, discharge summaries, operative reports, laboratory data, and post-discharge follow-up clinic notes were reviewed for AEs. All data were extracted from the electronic medical record by a single research team member, with 50% of the charts reviewed by a second team member to confirm accuracy.

The primary outcome was the rate of composite AEs, which consisted of nephrotoxicity, SSI within 60 days, and SSI requiring hospital re-admission for surgical intervention within 60 days from injury. Aminoglycoside-based regimens have been associated with an incidence of AEs ranging from 10% to 58% depending on the patient population studied [8 –11]. We conservatively estimated the incidence of adverse events with a tobramycin-based regimen to be 10% and estimated a 5% difference in AEs between the groups. However, because of the recent implementation of the practice change to piperacillin/tazobactam, we designed our study in a 2:1 fashion (tobramycin:piperacillin/tazobactam) to achieve adequate sample size. To detect a 5% decrease in the incidence of AEs with 80% power, the necessary sample size was estimated to be 153 patients, with 102 patients in the high-dose tobramycin group and 51 in the piperacillin/tazobactam group.

Normally distributed continuous variables were compared using the Student t-test, and the Mann-Whitney U test was used for non-normally distributed data. Categorical variables were evaluated with the Fisher exact test. A two-tailed significance of <0.05 was considered statistically significant. All analyses were conducted in SAS version 9.3 (SAS Institute, Cary, NC).

Results

From January 2010 to December 2016, 432 patients were identified from the institution's trauma database with an open fracture, of whom 347 were excluded (Table 1). There thus were 85 patients who met the inclusion criteria. Sixty-two of these patients received high-dose tobramycin plus cefazolin or clindamycin, and 23 were given piperacillin/tazobactam. In the tobramycin group, eight patients had ungraded fractures that were identified as Type III by the research team orthopaedic trauma surgeon. The most common reasons for exclusion were a non-Type III open fracture (n = 217) and receipt of both high-dose tobramycin and piperacillin/tazobactam for antimicrobial prophylaxis (n = 34).

Table 1.

Patients Excluded from Study

Reason	No. (%)
Non-Type III open fracture	217 (62.5)
Received both study drugs	34 ( 9.8)
Other	26 ( 7.5)
Received gentamicin	23 ( 6.6)
Prisoner	22 ( 6.3)
Death <24 h of admission	16 ( 4.6)
Traumatic amputation	9 ( 2.5)

The mean age was similar in the tobramycin and piperacillin/tazobactam groups (42.5 ± 15.1 years versus 39.7 ± 17.5 years; p = 0.50) with the majority of patients in both groups being male. Blunt trauma caused the majority of injuries in both groups (98.4% in the tobramycin group and 87% in the piperacillin/tazobactam group; p = 0.059). The median hospital stay was longer in the piperacillin/tazobactam group (12 days vs. 7 days; p = 0.035). There were three immunocompromised patients in the piperacillin/tazobactam group and none in the tobramycin group (p = 0.02). There were no differences in any other baseline characteristics (Table 2).

Table 2.

Characteristics of Patients

	Tobramycin (n = 62)(%)	Piperacillin/Tazobactam (n = 23)(%)	p
Mean age (y) (SD)	42.5 (15.1)	39.7 (17.5)	0.50
Male	45 (72.6)	14 (60.9)	0.30
Blunt trauma	61 (98.4)	20 (87)	0.059
Median total ISS score (IQR)^a	10 [9, 14]	12.5 [9.3, 17]	0.44
Mechanism of injury
▪ Motor vehicle crash	20 (32.3)	15 (65.2)
▪ Motorcycle crash	23 (37.1)	1 (4.3)	-
▪ Fall	12 (19.4)	3 (13)
▪ Gunshot	1 (1.6)	3 (13)
▪ Unknown	6 (9.7)	1 (4.3)
Arrival from scene	51 (82.3)	20 (87)	0.75
Median hospital LOS (d) (IQR)	7 [4,16]	12 [7,19.5]	0.035
ICU admission	23 (37.1)	11 (47.8)	0.46
Median ICU LOS (d) (IQR)	7 [3.3, 11]	10 [4.5, 15]	0.53
Diabetes mellitus	3 (4.8)	2 (8.7)	0.61
History of renal impairment	0 (0)	1 (4.3)	0.27
Current tobacco use	18 (29)	8 (34.8)	0.61
Immunocompromised	0 (0)	3 (13)	0.02

Total ISS score for tobramycin (sample size n = 60), piperacillin/tazobactam (sample size n = 18).

ICU = intensive care unit; IQR = interquartile range; ISS = Injury Severity Score; LOS = length of stay; SD = standard deviation.

Overall, 85% of the patients in both groups had only one open fracture, with a total of 72 fractures in the tobramycin group and 26 in the piperacillin/tazobactam group. The majority of fractures were in the lower extremity (83.6% in the tobramycin group and 85% in the piperacillin/tazobactam group; p = 0.9). The median time from injury to ED arrival was shorter in the piperacillin/tazobactam group, 43.5 minutes versus 62 minutes in the tobramycin group (p = 0.029). The median time from injury to initial antibiotic therapy likewise was longer in the tobramycin group (91.5 minutes versus 77 minutes; p = 0.11). However, when comparing median time to administration of the piperacillin/tazobactam and the combination of cefazolin/clindamycin and tobramycin, there was a statistically significant difference (77 vs. 179 minutes; p < 0.05). The median time from injury to operating room (OR) was not significantly different in the two groups (six hours in the tobramycin group versus five hours in the piperacillin/tazobactam group). However, more patients in the tobramycin group (88.8%) than in the piperacillin/tazobactam group (69.6%) were taken to the OR within the first 12 hours after injury (p = 0.017). The time from injury to wound closure was similar in the groups. On the basis of the total doses received, the median duration of antibiotic therapy was three days in the piperacillin/tazobactam group and two days in the tobramycin-based regimen group.

Of the 62 patients in the tobramycin group, 20 (32.3%) experienced at least one AE compared with 3 (13%) in the piperacillin/tazobactam group (p = 0.10). The composite AEs in the tobramycin-based group were 29 compared with three in the piperacillin/tazobactam group (Table 3). Of the 20 patients in the tobramycin group who experienced AEs, 9 (45%) experienced two AEs compared with none in the piperacillin/tazobactam group. At 30 days, SSI occurred in 17 patients (27.5%) in the tobramycin group and 1 (4.3%) in the piperacillin/tazobactam group (p = 0.033). At 60 days, three additional SSI occurred in the tobramycin group compared with none in the piperacillin/tazobactam group (p = 0.009). Hospital re-admission with surgical intervention occurred in 9 patients (14.5%) in the tobramycin group and none in the piperacillin/tazobactam group (p = 0.10). No patients in the tobramycin group experienced nephrotoxicity versus 2 (8.7%) in the piperacillin/tazobactam group (p = 0.07).

Table 3.

Number (%) of Adverse Events

Number	Tobramycin (n = 62)	Piperacillin/Tazobactam (n = 23)	p
0	42 (67.7)	20 (87)	0.10
≥1	20 (32.3)	3 (13)	0.10

At our institution, wound cultures are not collected routinely during the initial presentation of injury. Rather, wound cultures were collected when SSI was suspected and antibiotic treatment was initiated. For patients who required surgical debridement because of infection, there were eight positive wound cultures, all of which were in the tobramycin group (Table 4). One polymicrobial wound culture grew two drug-resistant organisms (methicillin-resistant Staphylococcus aureus and vancomycin-resistant Enterococcus). Using regression analysis of patients who experienced AEs, site of injury, body mass index (BMI) ≥35 kg/m², co-morbidities, and time to first dose of antibiotics were not identified as risk factors for the observed outcomes. Because only one SSI occurred in the piperacillin/tazobactam group, a multivariable regression analysis was not feasible. For those with documented SSIs, the median time to antibiotic commencement was 97 minutes in the tobramycin group compared with 79 minutes in the single piperacillin/tazobactam event.

Table 4.

Wound Culture Results

Patient	Time from injury (D)	Organism(s) isolated
1	4	Serratia marcescens
2	5	Escherichia coli, Klebsiella oxytoca, Enterococcus faecium, Clostridium perfringens
3	6	Enterobacter cloacae
4	11	Serratia marcescens, Enterobacter cloacae, Klebsiella oxytoca, Enterococcus durans, Clostridium perfringens
5	13	Enterobacter cloacae
6	13	Streptococcus pyogenes, Enterococcus faecalis, Citrobacter freundii, Enterobacter cloacae, Stenotrophomonas maltophilia
7	29	Streptococcus pyogenes, methicillin-resistant Staphylococcus aureus, Enterococcus faecalis, vancomycin-resistant Enterococcus faecium
8	34	Serratia marcescens

Discussion

The authors found no statistically significant difference in the rate of composite AEs between piperacillin/tazobactam and tobramycin-based prophylaxis regimens for Type III open fractures (p = 0.10). However, the incidence of SSI within 30 and 60 days from the time of injury was higher in patients receiving tobramycin-based therapy (p < 0.05).

Similar to our study, Redfern et al. showed that patients who received piperacillin/tazobactam for Type III open fracture prophylaxis had fewer SSIs than those receiving the regimen containing gentamicin. Those investigators found a rate of SSI in the gentamicin group at 30 days of 21.6% versus 11.4% with piperacillin/tazobactam (p = 0.246) [15]. This is consistent with our study, where the 30-day incidence of SSIs was 4.3% in the piperacillin/tazobactam group and 27.4% in the tobramycin group (p = 0.033).

Although the decrease in SSIs with piperacillin/tazobactam may be attributable to efficacy differences between beta-lactam- and aminoglycoside-based regimens, it is important to consider other factors affecting the rate of infection when comparing the two antibiotic classes. In both studies, the duration of therapy in the piperacillin/tazobactam group was slightly longer than that of the aminoglycoside group. In our study, patients receiving piperacillin/tazobactam had a median duration of therapy of three days compared with two days in those receiving tobramycin. Of the patients who received tobramycin, 34 (55%), received treatment for 48 hours or less compared with 56% of patients in the piperacillin/tazobactam group receiving antibiotic prophylaxis for at least 72 hours. Similarly, the median duration of therapy reported in the study of Redfern et al. was four days in the piperacillin/tazobactam group and three days in the gentamicin group [15]. Because the duration of therapy in the piperacillin/tazobactam arm was longer in both studies, we are unable to conclude that the difference in the SSI incidence is attributable to the agent alone.

Another explanation for the difference in SSI rates between the two groups is the time to administration of the first dose of antibiotics. Earlier administration improves outcomes. Although no statistically significant difference between the two groups in time to first dose was found, there was a trend toward earlier initiation of antibiotic prophylaxis therapy in patients receiving piperacillin/tazobactam (92 minutes vs. 77 minutes; p = 0.11). When time to administration of the full regimen was evaluated (tobramycin plus cefazolin or clindamycin versus piperacillin/tazobactam), we found a significant difference (179 minutes versus 77 minutes; p < 0.05). However, these times are still within the three-hour recommendations for antimicrobial prophylaxis initiation for Type III open fractures [5]. At our institution, the weight-based dose of tobramycin is prepared in a centralized pharmacy located away from the ED whereas piperacillin/tazobactam is readily available in the trauma bay. The simplified pre-made dose of and access to piperacillin/tazobactam in the ED offers potential efficacy advantages over a weight-based aminoglycoside regimen.

The rate of nephrotoxicity with aminoglycosides differs depending on the patient population and clinical setting. Reported nephrotoxicity incidence ranges from 10% to more than 50% in critically ill patients [8 –10]. We found no significant difference in nephrotoxicity between the piperacillin/tazobactam and tobramycin groups. Two patients in the piperacillin/tazobactam group and none in the tobramycin group experienced a more than 50% serum creatinine increase from baseline after receiving the study antibiotics. Although not statistically significant, more patients in the piperacillin/tazobactam group than in tobramycin group (78.3% versus 58.1%) received at least one nephrotoxic agent (p = 0.13) (Table 5). Of the two patients who experienced nephrotoxicity, one received vancomycin, vasopressors, and iodinated contrast medium; and the second received vancomycin, iodinated contrast medium, and blood products. Our study validated the low incidence of nephrotoxicity reported by others [19,20] who evaluated the effect of nephrotoxic agents on renal function in patients with Type III open fractures. Pannell et al. found that in patients with normal renal function, gentamicin plus cefazolin did not increase rates of renal dysfunction compared with cefazolin alone [19]. Similarly, Tessier et al. did not find any association between gentamicin and acute kidney injury in patients with Type III open fractures [20]. This lower rate of nephrotoxicity differs from other published studies reporting higher rates with prolonged use of aminoglycosides (four or more days) [7 –9,21]. In our study, the potential for nephrotoxicity may have contributed to the shorter duration of tobramycin therapy, leading to the low rate of nephrotoxicity as well as the higher rate of SSIs in this group.

Table 5.

Number (%) of Concurrent Nephrotoxic Agents

	Tobramycin (n = 62)	Piperacillin/Tazobactam (n = 23)
Iodinated contrast medium	34 (54.8)	18 (78.3)
Vancomycin	7 (11.3)	8 (34.8)
Vasopressors	0	2 (8.7)
Others (NSAIDs, diuretics)	4 (6.5)	5 (21.7)

NSAIDs = non-steroidal anti-inflammatory drugs.

Our study is limited by its retrospective nature and the accuracy with which the data were recorded in the electronic medical record. Because this was a retrospective study, we were neither able to control for duration of therapy nor evaluate the impact of confounding factors such as the use of antibiotic beads or tobramycin-impregnated bone cement. This study did not reach power, and we were unable to determine whether there was a significant difference between the study arms in the composite rate of observed AEs. However, we did identify a statistically significant advantage for piperacillin/tazobactam in SSI rates at 30 and 60 days. Finally, the study site is part of a large academic health system, and the study time period was six years. Therefore, there is the potential that changes of practice in the management of such a patient population in the pre-, intra-, and post-operative settings affected our findings. However, the orthopaedic surgical technique for timing, debridement, and fixation did not change over the study period.

The authors were unable to conclude that there is a statistically significant difference in composite AEs when piperacillin/tazobactam is used rather than tobramycin for Type III open fracture prophylaxis. However, a statistically significant efficacy advantage was seen as decreased SSI within 60 days from injury with the use of piperacillin/tazobactam was identified. Until a randomized study comparing the long-term efficacy and AEs associated with piperacillin-tazobactam versus an aminoglycoside-based regimen for Type III open fracture prophylaxis is completed, the study institution will continue to use piperacillin/tazobactam as the primary agent for prophylaxis.

Footnotes

Author Disclosure Statement

The authors have no conflicts of interest with regard to this manuscript.

There was no financial support for this work.

The study was approved by the local Biomedical Institutional Review Board.

References

Weitz-Marshall

, Bosse

. Timing of closure of open fractures. J Am Acad Orthop Surg, 2002; 10:379–384.

Luchette

, Bone

, Born

, et al. EAST Practice Management Guidelines Workgroup: Practice management guidelines for prophylactic antibiotic use in open fractures. J Trauma, 2011; 70:751–754.

Hoff

, Bonadies

, Cachecho

, Dorlac

. EAST Practice Management Guidelines Work Group: Update to practice management guidelines for prophylactic antibiotic use in open fractures. J Trauma, 2011; 70:751–754.

Hauser

, Charles

, Eachempati

. Prophylactic antibiotic use in open fractures: An evidence-based guideline. Surg Infect, 2006; 7:379–405.

Cross

, Swinotkowski

. Treatment principles in the management of open fractures. Indian J Orthop, 2008; 42:377–386.

Gustilo

, Anderson

. Prevention of infection in the treatment of one thousand and twenty-five open fractures of long bones: Retrospective and prospective analyses. J Bone Joint Surg Am, 1976; 58:453–458.

Gustilo

, Mendoza

, Williams

. Problems in management of Type III (severe) open fractures: A new classification of Type III open fractures. J Trauma, 1984; 24:742–746.

Gerlach

, Stawicki

, Cook

, Murphy

. Risk factors for aminoglycoside-associated nephrotoxicity in surgical intensive care unit patients. Int J Crit Illn Inj Soc, 2011; 1:17–21.

Bailey

, Torkington

, Anthony

, et al. Antibiotic-related acute kidney injury in patients undergoing elective joint replacement. Bone Joint J, 2014; 96-8:395–398.

10.

Ross

, Boscains

, Malhas

, Wigderowitz

. Perioperative renal morbidity secondary to gentamicin and flucloxacillin chemoprophylaxis for hip and knee arthroplasty. Scot Med J, 2013; 58:209–212.

11.

Incavo

, Ronchetti

, Choi

, et al. Penetration of piperacillin–tazobactam into cancellous and cortical bone tissues. Antimicrob Agents Chemother, 1994; 38:905–907.

12.

Zammit

, Fiorentino

, Cassar

, et al. Factors affecting gentamicin penetration in lower extremity ischemic tissues with ulcers. Int J Low Extrem Wounds, 2011; 10:130–137.

13.

Markantonis

, Kostopanagiotou

, Panidis

, et al. Effects of blood loss and fluid volume replacement on serum and tissue gentamicin concentrations during colorectal surgery. Clin Ther, 2004; 26:271–281.

14.

Rodriguez

, Jung

, Goulet

, et al. Evidence-based protocol for prophylaxis antibiotics in open fractures: Improved antibiotic stewardship with no increase in infection rates. J Trauma Acute Care Surg, 2014; 77:400–408.

15.

Redfern

, Wasilko

, Groth

, et al. Surgical site infections in patients with type 3 open fractures: Comparing antibiotic prophylaxis with cefazolin plus gentamicin versus piperacillin/tazobactam. J Orthop Trauma, 2016; 30:415–419.

16.

Janmohammadi

, Hasanjani Roshan

. Comparison of the efficacy of cefazolin plus gentamicin with cefazolin plus ciprofloxacin in management of type–IIIA open fractures. Iran Red Crescent Med J, 2011; 13:239–242.

17.

Patzakis

, Bains

, Lee

, et al. Prospective, randomized, double-blind study comparing single-agent antibiotic therapy, ciprofloxacin, to combination antibiotic therapy in open fracture wounds. J Orthop Trauma, 2000; 14:529–533.

18.

Kellum

, Lameire

. Diagnosis, evaluation, and management of acute kidney injury: A KDIGO summary (Part 1). Crit Care, 2013; 17:204.

19.

Pannell

, Banks

, Han

, et al. Antibiotic related acute kidney injury in patients treated for open fractures. Injury, 2016; 47:653–657.

20.

Tessier

, Moore

, Putty

, et al. Prophylactic gentamicin is not associated with acute kidney injury in patients with open fractures. Surg Infect, 2016; 17:720–723.

21.

Oliveira

, Silva

, Barbieri

, et al. Prevalence and risk factors for aminoglycoside nephrotoxicity in intensive care units. Antimicrob Agents Chemother, 2009; 53:2887–2891.