Evaluating the Duration of Prophylactic Post-Operative Antibiotic Agents after Open Reduction Internal Fixation for Closed Fractures

Abstract

Background:

The importance of timely pre-operative antibiotic agents for effective surgical prophylaxis has been established but the optimal duration of antimicrobial coverage post-operatively has not yet been defined clearly. The purpose of this study was to determine if prophylactic post- operative cefazolin for 23 hours decreases the risk of surgical site infection (SSI) after open reduction internal fixation (ORIF) of closed extremity fractures.

Patients and Methods:

After Institutional Reviews Board approval, patients undergoing ORIF of closed extremity fractures who had a planned post-operative stay of at least 23 hours were randomly assigned to either receive 23 hours of cefazolin or a placebo. Both groups received weight-based pre-operative cefazolin and intra-operative re-dosing at three-hour intervals until surgery completion. The primary end point was infection. Patients were followed clinically until bony union. Published risk factors were accumulated as a risk score to help determine risk of SSI.

Results:

A total of 227 patients were randomized to either receive post-operative cefazolin or placebo and 160 patients completed clinical follow-up to bony union. There were 83 patients in the cefazolin group and 77 in the placebo group. Surgical site infections occurred in a total of 15 patients (9.4%) in this trial without any differences between the cefazolin and placebo groups. Patients with diabetes mellitus were 4.33 times more likely to develop an SSI (95% confidence interval [CI], 1.30–14.38; p = 0.02). Patients with a risk score of two or more were 3.14 times more likely to develop an infection (95% CI, 1.02–9.68; p < 0.05).

Conclusions:

Although not statistically significant, in a randomized double-blinded placebo-controlled trial, patients who were treated with a 23-hour post-operative regimen of antibiotics after ORIF were less likely to develop SSIs. Patients with diabetes mellitus and those with a risk score of two or greater were more likely to develop an SSI.

Implanted biomaterials are essential to the practice of orthopedic surgery. More than two million fracture–fixation devices are implanted in the United States each year and that number is likely to increase [1]. Implanted foreign bodies are highly susceptible to infection because of compromised host defense systems. The incidence of infection after open reduction and internal fixation (ORIF) is between 0.4% and 16.1% depending on the severity, location, and characteristics of the fracture [2,3]. There is a complex interaction between the implant, the host, and the micro-organisms that underlie the pathogenesis of implant-associated infection. Micro-organisms typically form a biofilm and enter a slow- or non-growing state around the implant leading to a higher resistance to most antimicrobial agents [4,5].

Surgical site infections (SSIs) represent 20% to 31% of health-care–related infections in hospitalized patients and result in increased morbidity, mortality, and hospital costs [6]. Surgical site infections result in an additional $3.45 billion to $10.07 billion in direct medical expenditures in the United States each year [7]. Less than 100 colony-forming units (CFUs) of Staphylococcus aureus can result in an implant-associated infection, therefore, SSIs may occur despite appropriate aseptic surgical technique [8]. Previous studies have established that antibiotic prophylaxis decreases the incidence of infection in orthopedic procedures with implants as well as in closed extremity fracture surgery [9 –12]. Although the importance of timely pre-operative antibiotic administration for effective surgical prophylaxis has been established, the optimal duration of antimicrobial coverage post-operatively has not yet been defined clearly. Many of the studies in the orthopedic literature comparing single-dose prophylaxis with multiple-dose prophylaxis have not shown beneficial effects of additional doses [13 –15]. Furthermore, those investigating fracture ORIF have had methodological flaws, such as use of multiple types of antibiotic agents or use of currently outdated antibiotic agents [12].

With the implementation of the Surgical Care Improvement Project [16] and bundled fracture care, the burden of post-operative complications will shift onto healthcare facilities and potentially providers, therefore it is important to elucidate further the optimal period of post-operative antibiotic prophylaxis. The purpose of this study was to determine prospectively the effectiveness of post-operative prophylactic cefazolin—the most common antibiotic used for surgical prophylaxis—at preventing SSIs in ORIF of closed limb fractures. Our null hypothesis was that there would be no difference in the incidence of infection between a single pre-operative dose versus 23 hours of post-operative antibiotic prophylaxis.

Patients and Methods

After receiving Institutional Review Board (IRB) approval, patients undergoing closed limb fracture ORIF with an anticipated 23-hour stay between January 2008 and May 2011 were assigned randomly one of two treatment groups. In this study, cefazolin was used because it is the recommend prophylactic antibiotic for non-penicillin allergic patients without a history of methicillin-resistant Staphylococcus aureus (MRSA) [17]. Inclusion criteria included patients 18 years or older who were scheduled for ORIF or placement of a prosthetic device in the treatment of closed limb fractures with an anticipated 23-hour stay. Patients were excluded if they were less than 18 years old, presented with a known hypersensitivity to cephalosporins, history of MRSA infection, pregnant, immunosuppressed, unable to give informed consent, or had antimicrobial use or symptoms of infection within one week of surgery.

Both groups received weight-based intravenous cefazolin within one hour of the surgical incision, per standard protocol at our institution. If a tourniquet was used, the cefazolin was administered at least 10 minutes prior to inflation. A two-gram dose of cefazolin was administered to patients weighing more than 80 kg. A second one-gram dose of cefazolin was given at three-hour intervals until surgical completion [17]. Upon completion of the surgical procedure, patients were then randomized to either receive one-gram doses of cefazolin every eight hours for the next 23 hours (group 1). In total, patients in group 1 received two post-operative doses of cefazolin whereas patients randomly assigned to group 2 received a placebo of normal saline in the same research packaging and at the same time points as group 1. Group assignments were not disclosed to the evaluators responsible for clinical examination or to the patients until the end of the study. The institutional investigational pharmacy was responsible for following the established randomization protocol, preparing study medications, and handling delivery of the medications to the appropriate areas.

Patient demographics (gender, race, age, body mass index [BMI], diabetes mellitus, and tobacco use), clinical course, and surgical characteristics were recorded. Standard clinical and radiographic post-operative visits were used for follow-up and occurred at 10 to 21 days, 6 weeks, 12 weeks and every 6 to 8 weeks thereafter until bony union occurred. Site infection was defined as purulent drainage at the operative site with the presence of one or more of the classic signs and symptoms of inflammation (rubor, calor, tumor, dolor) [18,19].

Superficial infections were defined as those patients managed with antibiotic agents and local wound care whereas deep infections were defined as those patients requiring operative management. Bacteriologic cultures were obtained intra-operatively for all patients with a suspected deep infection, but superficial site infections were diagnosed clinically.

Previous work has demonstrated that there are numerous patient-related and treatment-related factors that predispose to infectious complications [13,18,20 –23]. Relevant factors include: smoking, age older than 65, diabetes mellitus, obesity (BMI >35), duration of surgery more than three hours, and urinary catheterization. Based on these factors patients were assigned one point for each factor resulting in a risk score ranging from zero (lowest risk) to seven (highest risk). For urinary catheterization, scores varied as followed: 0 = no catheter; 1 = catheter less than 48 hours, 2 = catheter longer than 48 hours.

Statistical evaluation was performed with descriptive statistics for patient characteristics. Fisher exact test was used for patient characteristics, infection rates, and the relation between site infection and the risk factors listed in the previous paragraph. Odds ratios were also determined for each risk factor variable. Statistical significance was defined as p < 0.05.

Results

A total of 227 patients with closed limb fracture were identified, provided consent, and randomly assigned to a treatment group at our American College of Surgeons (ACS) Level 1 hospital prior to the ORIF procedure. One hundred sixty patients (70.5%) continued to meet inclusion criteria and completed clinical follow-up to fracture union. Eighty-three patients were randomly assigned to treatment with cefazolin whereas there were 77 patients in the placebo group. The cefazolin cohort comprised 34 males and 49 females with an average age of 50 (range, 19–95) whereas there were 46 males and 31 females with an average age of 47 years (range, 18–88) in the placebo cohort. Other than gender, there were no statistically significant differences between groups with respect to patient demographics, clinical, or surgical characteristics (Table 1). A fall (59.4%) was the most common mechanism of injury for patients in both cohorts (Table 2). Finally, lower extremity fracture was the most frequent presentation in both cohorts (Table 3).

Table 1.

Demographic, Clinical, and Operative Characteristics

Parameter	Total (n = 160)	Control (n = 77)	Cefazolin (n = 83)	p
Gender	80 M, 80 F	46 M, 31 F	34 M, 49 F	0.03
Age (range)	48 (18–95)	47 (18–88)	50 (19–95)	0.40
BMI	29.7	29.9	29.5	0.77
Tobacco use (%)	58 (36.3)	25 (32.5)	33 (39.8)	0.41
Diabetes mellitus (%)	20 (12.5)	9 (11.7)	11 (13.3)	0.81
Ethanol dependence (%)	8 (5.0)	1 (1.3)	7 (8.4)	0.07
Renal disease (%)	4 (2.5)	3 (3.9)	1 (1.2)	0.35
Cardiac disease (%)	21 (13.1)	9 (11.7)	12 (14.5)	0.65
Neurologic disease (%)	14 (8.8)	9 (11.7)	5 (6.0)	0.27
Peripheral vascular disease (%)	3 (1.9)	0 (0)	3 (3.6)	0.25
Osteoporosis (%)	8 (5.0)	2 (2.6)	6 (7.2)	0.28
Urinary catheter (%)	59 (36.9)	28 (36.4)	31 (37.3)	1.00
Days to ORIF	3.7	4.35	3.1	0.09
Duration of surgery (min)	136	132	140	0.46
More than five implants (%)	99 (61.9)	44 (57.1)	55 (66.3)	0.26

BMI = body mass index; ORIF = open reduction internal fixation.

Table 2.

Mechanism of Injury

Parameter	Total	Control	Cefazolin
Motor vehicle (%)	37 (23.1)	19 (24.7)	18 (21.7)
Fall (%)	95 (59.4)	45 (58.4)	50 (60.2)
Other (%)	19 (11.9)	7 (9.1)	12 (14.5)
All-terrain vehicle (%)	9 (5.6)	6 (7.8)	3 (3.6)

Table 3.

Fracture Location

Bone	Control	Cefazolin
Ankle	10	8
Calcaneus	3	1
Clavicle	2	2
Femur	27	17
Foot	2	2
Humerus	6	12
Patella	0	3
Pelvis	2	5
Radius/ulna	7	8
Sacrum	0	1
Tibia/fibula	18	24

Fifteen patients (9.4%) were diagnosed with a SSI. Six infections were classified as superficial (3.8%) because they were treated with antibiotic agents and local wound care whereas there were nine (5.6%) infections classified as deep. There were five infections in patients receiving cefazolin—one (1.2%) superficial and four (4.8%) deep. Ten infections occurred in the control cohort: five (6.5%) superficial and five (6.5%) deep. Patients treated with cefazolin were 82% less likely to experience a superficial infection (odds ratio [OR]; 0.18; 95% CI, 0.02–1.54; p = 0.12) and 27% less likely to experience a deep infection (OR; 0.73; 95% CI, 0.19–2.82; p = 0.65; Table 4). Staphylococcus aureus was identified as the most common causative organism.

Table 4.

Surgical Site Infections

Parameter	Control (n = 77)	Cefazolin (n = 83)	Odds ratio (95% CI)	p
Total infections (%)	10 (13.0)	5 (6.0)	0.42 (0.14–1.32)	0.14
Superficial infections (%)	5 (6.5)	1 (1.2)	0.18 (0.02–1.54)	0.12
Deep infections (%)	5 (6.5)	4 (4.8)	0.73 (0.19–2.82)	0.65

CI = confidence interval.

The six clinical characteristics previously identified as predisposing to infectious complications were used to calculate a risk score. When considering the individual risk factors, we found that patients with diabetes mellitus were 4.33 times more likely to develop an SSI (95% CI, 1.30–14.38; p = 0.02). We were unable to identify any other independent risk factors that increased the incidence of SSI. On average, patients with SSIs had a higher risk score in comparison to their non-infected counterparts (2 vs. 1.4; p = 0.12). Finally, patients with a risk score of two or higher were 3.18 times more likely to become infected (95% CI, 1.03–9.78; p = 0.04; Table 5).

Table 5.

Risk Factors Associated with Surgical Site Infections

Parameter	Non-Infected	Infected	Odds ratio (95% CI)	p
Age >65 y (%)	31 (21.4)	2 (13.3)	0.57 (0.12–2.64)	0.47
BMI >35 (%)	28 (19.4)	4 (28.6)	1.66 (0.48–5.67)	0.42
Tobacco use (%)	53 (36.6)	5 (33.3)	0.87 (0.28–2.67)	0.80
Diabetes mellitus (%)	15 (10.3)	5 (33.3)	4.33 (1.30–14.38)	0.02
Surgery >180 min (%)	32 (22.1)	5 (33.3)	1.77 (0.56–5.54)	0.43
Urinary catheter
< 48 h (%)	22 (15.2)	2 (13.3)	0.86 (0.18–4.07)	0.85
> 48 h (%)	35 (24.1)	3 (20.0)	0.79 (0.21–2.94)	0.72
Risk score
Control cohort	1.4	1.7	NA	0.72
Cefazolin cohort	1.7	2.6	NA	0.07
Total risk score	1.4	2	NA	0.12
Risk score >2 (%)	56 (38.6)	10 (66.7)	3.18 (1.03–9.78)	0.04

NA = not applicable.

Discussion

Surgical site infections are one of the most common adverse events occurring in hospitalized patients and are problematic for a multitude of reasons including their propensity to contribute to surgical morbidity and mortality. Patients who experience SSIs are predisposed to a prolonged clinical course including multiple surgeries, multiple hospitalizations, longer healing times, and greater time away from work. Furthermore, this sequela places a strain on the healthcare system. Successful management of infected fractures status post-ORIF often requires numerous operative debridements and intravenous antibiotics [24]. As a result, when infections occur, healthcare costs may increase by 300%, hospital stays may increase by two weeks, and re-hospitalization rates may double [25].

The results of this study suggest that although there was not a statistically significant difference in the overall infection rate, patients who were treated with a 23-hour post-operative regimen of antibiotic agents after ORIF were less likely to develop SSI. In a recent meta-analysis, Slobogean et al. [14] examined single- versus multiple-dose regimens of prophylactic antibiotic agents in closed long bone fractures. In total, this group reviewed seven prospective randomized trials for long bone fractures. When the data were pooled, there was no difference when analyzing infection rates between single- and multiple-dose regimens. The authors stated that it would take approximately 25,000 patients to meet statistical significance. In comparing the results of this meta-analysis to the current study it is worth noting that only one of the seven trials included in the review used cefazolin as the sole antibiotic in comparison to a placebo [26]. This trial focused on hip fracture patients and used a two-gram loading dose with three post-operative doses at six-hour intervals. The findings of this present study are promising in that patients treated with cefazolin were less likely to experience superficial infections, and whereas cefazolin may ultimately confer some degree of protection with respect to all SSIs further work must be done to elucidate this. Based on our study's current rate of infection, 274 patients would be needed in each group to confer statistical significance.

Previous work has established that the following patient-related and treatment-related factors predispose patients to infectious complications: smoking; age older than 65; diabetes mellitus; BMI >35; duration of surgery longer than three hours; and urinary catheterization (0 = no catheter, 1 = catheter less than 48 hours, 2 = catheter longer than 48 hours [13,18,20 –23,27]. Given these factors, patents were assigned a risk score ranging from zero (lowest risk) to seven (highest risk) where each risk factor contributed one point toward their total score. In this study, patients with diabetes mellitus presented with infections at a higher incidence. We were unable to establish an association between infection and any of the other of the individual risk factors comprising the risk score but we did discover that patients with a risk score greater than or equal to two were 3.14 times more likely to become infected. Furthermore, it is worth noting that patients treated with cefazolin who did develop an SSI presented with a higher risk score than their counterparts. This would seem to indicate that the administration of cefazolin conferred some degree of benefit.

Prophylactic post-operative administration of antibiotic agents may provide a measured benefit in preventing SSIs in patients undergoing operative intervention for closed fractures. Surgical site infections are problematic for the patient and the healthcare system given the resulting prolonged hospital course, rate of re-hospitalization, and anticipated increased healthcare costs [25]. Slobogean et al. [28] evaluated the cost effectiveness of single-dose pre-operative antibiotic prophylaxis versus the multiple-dose strategy and reported the results in 2007 United States dollars and quality of life days (QALD) gained based on the infection rates reported in the literature. Both strategies resulted in a similar average cost of approximately $2,600 and had the same QALD gained (28). Ultimately, the authors concluded that there was no real financial benefit to the multiple-dose regimen. Although this may be true, patients with diabetes mellitus or those with multiple risk factors pose a unique challenge given their increased vulnerability. Ultimately, it may not be prudent to dose all patients with a post-operative course of antibiotic agents yet it may be beneficial to consider prescribing post-operative antibiotics in these high-risk patients.

The authors recognize that there are limitations to the present study. The infection rate observed in this study is much higher than the average rate of 0.5% to 2% observed after internal fixation of closed fractures [8,29,30]. This variation may be attributed to the fact that our institution is a tertiary care referral center; many of the patients transferred to our institution are underserved and present with multiple comorbidities and complex pathophysiology. Moving forward, of the 227 patients with closed limb fractures who were identified, provided consent, and randomly assigned to a treatment group, 30% of patients were subsequently excluded from data analysis because of an incomplete data set or failure to follow up to bony union. Again, this finding is because of the dynamics of our institutional referral patterns with patients living up to six hours away. Although our findings indicate patients treated with cefazolin were less likely to experience superficial infections we failed to achieve statistical significance when evaluating the total infection rate in those treated with cefazolin. Given the current infection rate, we would need approximately 274 patients in each cohort to display statistical significance and therefore further work must be done to examine these findings in a larger study population. Finally, all patients with limb fractures undergoing ORIF and an anticipated length of stay of at least 23 hours were included in this study. Restricting enrollment to patients with a similar fracture pattern may result in more robust results, however the majority (5/7) of prospective trials that were used in the Slobogean et al. [14] meta-analysis used a similar recruitment strategy.

Although there was no significant difference in the infection rates in this study, there was a strong trend of a 116% increase in the single-dose cohort. This might suggest that the single dose is inferior to the 23-hour prophylaxis, but a study with a larger sample size is recommended. Perhaps the most significant finding in our study is that patients with diabetes mellitus and those with multiple risk factors were more likely to develop SSIs. Although the data may not support prescribing 23 hours of post-operative antibiotic agents in all cases, it may be prudent to consider this algorithm in patients with diabetes mellitus and those with a higher risk score.

Footnotes

Acknowledgments

This project was completed without the use of external funding and the authors declare no conflict of interest with this study. This research was given as part of a podium presentation at the Orthopaedic Trauma Association's Annual Meeting, held in Tampa, Florida, in October 2014.

All authors meet authorship criteria for this article as described below. All authors have seen and approved the final manuscript as submitted. The senior authors (B.C.D., L.O.O., and D.D.G.) had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. B.C.D., D.D.G., and G.J.D.R. conceived and designed the study. B.D.C., L.O.O., G.J.D.R., D.A.V., J.P.S., and D.D.G. performed acquisition of data. B.D.C., L.O.O, G.J.D.R., D.A.V, J.P.S., and D.D.G. analyzed and interpreted the data. B.D.C., L.O.O., and D.D.G. drafted the manuscript. B.D.C., L.O.O., and D.D.G. performed critical revision of the manuscript. B.D.C., L.O.O., and D.D.G. provided statistical expertise. G.J.D.R., D.A.V., and J.P.S. provided administrative, technical, or material support. B.D.C. provided supervision and final approval of the submitted manuscript.

Author Disclosure Statement

Except as noted below the authors declare no conflicts of interest.

Brett D. Crist: Amedica Coporation: Stock or stock options; AO Trauma North America: Board or committee member; Arthrex, Inc: Other financial or material support; DePuy, A Johnson & Johnson Company: Paid consultant, paid presenter or speaker; Globus Medical: Other financial or material support; paid consultant; International Geriatric Fracture Society: Board or committee member; Journal of Hip Preservation: Editorial or governing board; Journal of Orthopaedic Trauma: Editorial or governing board; KCI: Paid consultant; Paid presenter or speaker; Research support; Mid-America Orthopaedic Association: Board or committee member; Orthopaedic Implant Company: Stock or stock options; Orthopaedic Trauma Association: Board or committee member; Synthes: Research support.

Gregory J. Della Rocca: American Academy of Orthopaedic Surgeons: Committee member; Amedica: Stock or stock options; American College of Surgeons: Board or committee member; American Orthopaedic Association: Board or committee member; Bioventus: Paid consultant; Geriatric Orthopaedic Surgery and Rehabilitation: Editorial or governing board; Journal of Orthopaedic Trauma: Editorial or governing board; Mergenet: Stock or stock options; Orthopaedic Trauma Association: Board or committee member; Synthes: Paid presenter or speaker; Research support; The Orthopaedic Implant Company: Stock or stock options; Wright Medical Technology, Inc.: IP royalties; Paid consultant.

James P. Stannard: Acelity: Paid consultant; AO North America: Board or committee member; Arthrex, Inc: Paid consultant; Research support; Coulter Foundation: Research support; DePuy, A Johnson & Johnson Company: Paid consultant; Journal of Knee Surgery: Editorial or governing board; Nuvasive: Paid consultant; Orthopaedic Trauma Association: Board or committee member; Smith & Nephew: Paid consultant; Thieme: Publishing royalties, financial or material support; U.S. Department of Defense: Research support.

Lasun O. Oladeji, David A. Volgas, and David D. Greenberg: Nothing to declare.

References

Darouiche

. Treatment of infections associated with surgical implants. N Engl J Med, 2004; 350:1422–1429.

Chen

, Schreiber

, Washington

, et al. What is the rate of methicillin-resistant Staphylococcus aureus and Gram-negative infections in open fractures?. Clin Orthop Relat Res, 2013; 471:3135–3140.

Zimmerli

. Clinical presentation and treatment of orthopaedic implant-associated infection. J Intern Med, 2014; 276:111–119.

de la Fuente-Nunez

, Reffuveille

, Fernandez

, et al. Bacterial biofilm development as a multicellular adaptation: Antibiotic resistance and new therapeutic strategies. Curr Opin Microbiol, 2013; 16:580–589.

Mauffrey

, Herbert

, Young

, et al. The role of biofilm on orthopaedic implants: The “Holy Grail” of post-traumatic infection management?. Eur J Trauma Emerg Surg, 2016; 42:411–416.

Owens

, Barrett

, Raetzman

, et al. Surgical site infections following ambulatory surgery procedures. JAMA, 2014; 311:709–716.

Shea

, Styhl

, King

, et al. Surgical site infection reduction program: Challenges and opportunities. J Pediatr Orthop, 2015; 35:S51–54.

Trampuz

, Zimmerli

. Antimicrobial agents in orthopaedic surgery: Prophylaxis and treatment. Drugs, 2006; 66:1089–1105.

Gillespie

, Walenkamp

. Antibiotic prophylaxis for surgery for proximal femoral and other closed long bone fractures. Cochrane Database Syst Rev, 2010:CD000244.

10.

Jaeger

, Maier

, Kern

, et al. Antibiotics in trauma and orthopedic surgery—A primer of evidence-based recommendations. Injury, 2006; 37(Suppl 2):S74–80.

11.

Anderson

, Savage

, Vaccaro

, et al. Prevention of surgical site infection in spine surgery. Neurosurgery, 2017; 80:S114–S123.

12.

Chang

, Kennedy

, Bhandari

, et al. Effects of antibiotic prophylaxis in patients with open fracture of the extremities: A systematic review of randomized controlled trials. JBJS Rev, 2015 Jun 9; 3(6).

13.

Morita

, Smith

. Antimicrobial prophylaxis in orthopedic surgery. Orthopedics, 2005; 28:749–751.

14.

Slobogean

, Kennedy

, Davidson

, et al. Single- versus multiple-dose antibiotic prophylaxis in the surgical treatment of closed fractures: A meta-analysis. J Orthop Trauma, 2008; 22:264–269.

15.

Morrison

, White

, Asadollahi

, et al. Single versus multiple doses of antibiotic prophylaxis in limb fracture surgery. ANZ J Surg, 2012; 82:902–907.

16.

Lin

. Surgical Care Improvement Project: Improve Performance, Reduce Complications, and Comply with CMS. Marblehead, MA: HC Pro, Inc., 2007.

17.

Prokuski

. Prophylactic antibiotics in orthopaedic surgery. J Am Acad Orthop Surg, 2008; 16:283–293.

18.

Boxma

, Broekhuizen

, Patka

, et al. Randomised controlled trial of single-dose antibiotic prophylaxis in surgical treatment of closed fractures: The Dutch Trauma Trial. Lancet, 1996; 347:1133–1137.

19.

Chambers

, Gillespie

. Perioperative cefamandole prophylaxis against infections. J Bone Joint Surg Am, 1988; 70:950.

20.

Cushing

. Antibiotics in trauma. Surg Clin North Am, 1977; 57:165–177.

21.

Nichols

. Surgical wound infection. Am J Med, 1991; 91:54S–64S.

22.

Momaya

, Hlavacek

, Etier

, et al. Risk factors for infection after operative fixation of tibial plateau fractures. Injury, 2016; 47:1501–1505.

23.

Colman

, Wright

, Gruen

, et al. Prolonged operative time increases infection rate in tibial plateau fractures. Injury, 2013; 44:249–252.

24.

Rightmire

, Zurakowski

, Vrahas

. Acute infections after fracture repair: Management with hardware in place. Clin Orthop Relat Res, 2008; 466:466–472.

25.

Whitehouse

, Friedman

, Kirkland

, et al. The impact of surgical-site infections following orthopedic surgery at a community hospital and a university hospital: Adverse quality of life, excess length of stay, and extra cost. Infect Control Hosp Epidemiol, 2002; 23:183–189.

26.

Buckley

, Hughes

, Snodgrass

, et al. Perioperative cefazolin prophylaxis in hip fracture surgery. Can J Surg, 1990; 33:122–127.

27.

Reese

, Knepper

, Young

, et al. Development of a surgical site infection prediction model in orthopaedic trauma: The Denver Health Model. Injury, 2017; 48:2699–2704.

28.

Slobogean

, O'Brien

, Brauer

. Single-dose versus multiple-dose antibiotic prophylaxis for the surgical treatment of closed fractures. Acta Orthop, 2010; 81:256–262.

29.

Obremskey

, Bhandari

, Dirschl

, et al. Internal fixation versus arthroplasty of comminuted fractures of the distal humerus. J Orthop Trauma, 2003; 17:463–465.

30.

Perren

. Evolution of the internal fixation of long bone fractures. The scientific basis of biological internal fixation: Choosing a new balance between stability and biology. J Bone Joint Surg Br, 2002; 84:1093–1110.