Abstract
Objectives
We assessed osteomyelitis risk factors in U.S. military personnel with combat-related open tibia fractures (2003–2009).
Methods
Patients with open tibia fractures who met diagnostic criteria of osteomyelitis were identified as cases using Military Health System data and verified through medical record review. Controls were patients with open tibia fractures who did not meet osteomyelitis criteria. Gustilo-Andersen (GA) fracture classification scheme was modified to include transtibial amputations (TTAs) as the most severe level. Logistic regression multivariable odds ratios (OR; 95% confidence interval [CI]) were assessed.
Results
A total of 130 tibia osteomyelitis cases and 85 controls were identified. Excluding patients with TTAs, osteomyelitis cases had significantly longer time to radiographic union compared to controls (median: 210 versus 165 days). Blast injuries, antibiotic bead utilization, ≥GA-IIIb fractures (highest risk with TTA [OR: 15.10; CI: 3.22–71.07]) and foreign body at fracture site were significantly associated with developing osteomyelitis. In a separate model, the Orthopaedic Trauma Association Open Fracture Classification muscle variable was significant with increasing risk from muscle loss (OR: 5.62; CI: 2.21–14.25) to dead muscle (OR: 8.46; CI: 3.31–21.64). When TTAs were excluded, significant risk factors were similar and included sustaining an injury between 2003 and 2006.
Conclusions
Patients with severe blast trauma resulting in significant muscle damage are at the highest risk for osteomyelitis. The time period association coincides with a timeframe when several trauma system practice changes were initiated (e.g., increased negative pressure wound therapy, decreased high pressure irrigation, and reduced crystalloid use).
Level of Evidence
Prognostic Level III
INTRODUCTION
Orthopaedic injuries accounted for 82% of wounded personnel in Iraq and Afghanistan between 2001 and 2005, of which 53% were penetrating soft-tissue injuries and 26% were fractures.1 Civilian and military trauma care are similar in including prompt antibiotic administration, wound debridement and irrigation, fracture and hemodynamic stabilization, timely wound closure or coverage, and rehabilitation. Nevertheless, the injury mechanism and environment are different for combat casualties. Blast injuries from high-energy improvised explosive devices (IEDs), which are uncommon in civilian trauma, result in extensive wound contamination and tissue loss.2–5 Furthermore, the combat casualty care environment involves medical evacuation through multiple levels of care (i.e., combat zone facilities, regional hospitals outside combat zone, and United States hospitals for definitive treatment and rehabilitation).6
Although combat casualty care has improved during the last decade, infectious complications, including osteomyelitis, remain a frequent occurrence.7–13 An evaluation of 664 combat casualties admitted to the orthopaedic service at Brooke Army Medical Center (2003–2006) found that 15% were diagnosed with osteomyelitis.12 Moreover, among wounded United Kingdom (U.K.) military personnel with open tibia fractures, 23% developed deep infections.14 In civilian cases, patients with open fractures frequently develop infections,15–19 with severe tibia fractures (Type III) having rates as high as 35%.20–25 Among combat casualties with Type III open tibia fractures, deep infections, including osteomyelitis, are associated with increased resource utilization stemming from revision surgeries and late amputations (≥12 weeks post-injury).26
There are a variety of factors that may contribute to a patient’s outcome after an orthopaedic injury, and more specifically whether an infection develops.26,27 Using a comprehensive review of data over a 6-year period and from multiple clinical sites, we assessed risk factors for development of osteomyelitis in combat-wounded individuals with open tibia fractures, including transtibial amputations (TTAs).
METHODS
Study Population
Data were collected from U.S. service members with deployment-related orthopaedic trauma sustained from March 19, 2003 through December 31, 2009. Patients were medically evacuated from the combat theater to Landstuhl Regional Medical Center (Germany), and subsequently transferred to a U.S. military hospital: Walter Reed Army Medical Center (Washington, DC), National Naval Medical Center (Bethesda, MD), or Brooke Army Medical Center (San Antonio, TX). The study was approved by the Infectious Disease Institutional Review Board of the Uniformed Services University of the Health Sciences.
Case-Control Classification
The Military Health System Data Repository was surveyed for patients with International Classification of Diseases (9th revision, Clinical Modification) codes related to lower extremity injuries. The Department of Defense Trauma Registry and Military Health System pharmacy, laboratory, and radiology records were also reviewed. Medical records from all potential subjects were independently reviewed by an infectious disease and orthopaedic surgery physician, using standardized criteria to verify case and control classifications. The study population was restricted to patients with trauma-related open tibia fractures. Using a modified version of the Gustilo-Anderson (GA) classification system, open fractures were categorized (at the discretion of the orthopaedic surgeon) into four grades: GA-I, GA-II, GA-III, as well as an additional category for traumatic and early surgical TTAs.28 Fractures were classified in accordance with the Orthopaedic Trauma Association’s (OTA) Open Fracture Classification (OFC) system.29,30
All cases were classified in accordance with the Centers for Disease Control and Prevention, National Healthcare Safety Network (NHSN) published grading of osteomyelitis diagnoses as definite, probable, or possible.31 The NHSN definitions are utilized by state registries and surveillance systems,32–34 as well as the European Centre for Disease Prevention and Control.35 Definite osteomyelitis required either a positive bone culture or evidence of bone infection on direct examination during a surgical procedure or histopathological examination. A probable osteomyelitis was defined by occurrence of ≥2 of the following signs/symptoms (>38°C temperature, localized swelling, localized heat, localized tenderness, drainage at site) in addition to either organisms cultured from blood samples or radiographic evidence of infection. A possible osteomyelitis required environmental contamination at time of injury, deep wound tissue growing any organism, and evidence of either local (purulence or necrotic tissue at affected site) or systemic inflammation (>38°C temperature, ≥12,000 μL leukocytes, or elevated C-reactive protein or erythrocyte sedimentation rate). Patients with a fracture delayed union (or established nonunion) on follow-up examination with evidence of systemic inflammation were also considered to have a possible osteomyelitis. Inclusion in the control population required an open tibia fracture without meeting diagnostic criteria for osteomyelitis.
Statistical Analysis
Characteristics of patients with and without osteomyelitis were compared using Chi-square testing (or Fisher’s exact test) for categorical variables and Wilcoxon rank-sum test for continuous variables. Potential risk factors for development of osteomyelitis were examined through univariable and multivariable logistic regression models. Covariates with a p-value ≤0.2 in the univariable model were considered for inclusion in the multivariable model. Stepwise, backward, and forward model selections were conducted to choose the multivariable risk factor model. Risk factors with an overall p-value <0.05 were retained in the model.
RESULTS
Study Population
A total of 130 osteomyelitis cases and 85 controls were identified (Table 1). Patients were predominantly male (97%), injured in Iraq (86%), and sustained injuries between 2003 and 2006 (59%). A significantly greater proportion of osteomyelitis cases incurred injuries via a blast mechanism (82% versus 62%; p=0.001); however, there was no significant difference related to injury severity scores and blood transfusions within 24 hours post-injury between cases and controls. A higher proportion of patients injured in Iraq were from the earlier time period compared to Afghanistan (62% versus 40%; p=0.022), due to casualty rates from the respective operational theaters.
Table 1.
Demographics and Injury Characteristics, No. (%), of Wounded Military Personnel with and without Tibia Osteomyelitis
| Total (N=215) | Osteomyelitis Cases (N=130) | Controls (N=85) | P-value | |
|---|---|---|---|---|
| Male | 209 (97.2) | 125 (96.2) | 84 (98.8) | 0.407 |
| Age at time of injury, median (IQR) | 24 (21–30) | 24 (21–31) | 23 (21–29) | 0.962 |
| History of tobacco use1 | 89 (47.6) | 59 (51.8) | 30 (41.1) | 0.178 |
| Operational theater | 0.045 | |||
| Iraq | 185 (86.0) | 117 (90.0) | 68 (80.0) | |
| Afghanistan | 30 (14.0) | 13 (10.0) | 17 (20.0) | |
| Time period | 0.258 | |||
| 2003–2006 | 127 (59.1) | 81 (62.3) | 46 (54.1) | |
| 2007–2009 | 88 (40.9) | 49 (37.7) | 39 (45.9) | |
| Branch of service | 0.345 | |||
| Army | 177 (82.3) | 110 (84.6) | 67 (78.8) | |
| Marine | 31 (14.4) | 18 (13.8) | 13 (15.3) | |
| Air Force | 3 (1.4) | 1 (0.8) | 2 (2.4) | |
| Navy | 4 (1.9) | 1 (0.8) | 3 (3.5) | |
| Blast mechanism of injury | 160 (74.4) | 107 (82.3) | 53 (62.4) | 0.001 |
| Blood product transfusions within first 24 hours | 0.069 | |||
| None or missing units | 105 (48.8) | 55 (42.3) | 50 (58.8) | |
| 1–9 units | 65 (30.2) | 44 (33.8) | 21 (24.7) | |
| ≥10 units | 45 (20.9) | 31 (23.8) | 14 (16.5) | |
| Injury severity score2 | 0.523 | |||
| 0–9 (mild) | 75 (34.9) | 44 (33.8) | 31 (36.5) | |
| 10–15 (moderate) | 83 (38.6) | 54 (41.5) | 29 (34.1) | |
| ≥16 (severe to life-threatening) | 57 (26.5) | 32 (24.6) | 25 (29.4) |
IQR - interquartile range
Tobacco history was not known for 28 patients (16 cases and 12 controls). Percentages are based on total minus unknown.
The Injury Severity Score is an overall measure calculated for each patient based on the top three maximum Abbreviated Injury Scale anatomical region values48
Clinical and Open Fracture Characteristics
Among the osteomyelitis cases, 25 (19%) met the definition for either definite or probable, while the remaining 105 (81%) met criteria for a possible osteomyelitis. Except for a higher proportion of localized site swelling among definite/probable cases compared to possible, there was no significant difference between the classification groups regarding clinical presentation for local signs and symptoms (data not shown). Regarding systemic inflammation, the median maximum temperature was 38.7°C (interquartile range [IQR]: 37.8–39.4°C) for definite/probable and 38.9°C (IQR: 38.3–39.4) for possible cases. The median maximum white blood cell count was 17.8 (interquartile range [IQR]: 12.3–19.6) for definite cases, 17.4 (IQR: 13.7–24.5) for probable cases, and 18.8 (IQR: 13.9–25.1) for possible cases. For definite cases, the median maximum C-reactive protein level was 13.3 (IQR: 5.4–23.8), while it was 15.2 (IQR: 8.5–17.7 for probable cases and 16.1 (IQR: 8.6–22.5) for possible cases. Lastly, erythrocyte sedimentation rate was a maximum median value of 90.5 (IQR: 78–115.5), 97 (IQR: 76–120), and 113 (IQR: 93–120) for definite, probable, and possible cases, respectively.
Information on timing of first operating room visit in theater was available for 74 patients, of which 85% were seen within the first 6 hours post-injury (95% seen within 24 hours). The majority of subjects (81%) recovered Gram-negative bacteria with the predominant organism across the definite/probable and possible cases being Acinetobacter calcoaceticus baumannii complex (44% and 56%, respectively).
Using the modified GA fracture classification, significantly more osteomyelitis cases had GA-IIIb fractures and TTAs (p<0.001; Table 2). Regarding the OTA OFC for skin, muscle, bone loss, and contamination variables, osteomyelitis cases had more extensive degloving, dead muscle, loss of muscle function, or partial/complete muscle excision, segmental bone loss, and contamination embedded in bone or deep soft-tissue (p≤0.001).
Table 2.
Characteristics, No. (%), of Open Tibia Fractures Sustained by Military Personnel
| Total (N=215) | Osteomyelitis Cases (N=130) | Controls (N=85) | P-value | |
|---|---|---|---|---|
| Fracture Class1 | <0.001 | |||
| GA-II | 22 (10.2) | 5 (3.8) | 17 (20.0) | |
| GA-IIIa | 62 (28.8) | 32 (24.6) | 30 (35.3) | |
| GA-IIIb | 51 (23.7) | 37 (28.5) | 14 (16.5) | |
| GA-IIIc | 22 (10.2) | 13 (10.0) | 9 (10.6) | |
| TTA2 | 54 (25.1) | 41 (31.5) | 13 (15.3) | |
| Open fracture not otherwise specified | 4 (1.9) | 2 (1.5) | 2 (2.4) | |
| OTA OFC: Skin3 | <0.001 | |||
| Can be approximated | 59 (28.8) | 22 (17.5) | 37 (46.8) | |
| Cannot be approximated | 65 (31.7) | 44 (34.9) | 21 (26.6) | |
| Extensive degloving | 81 (39.5) | 60 (47.6) | 21 (26.6) | |
| OTA OFC: Muscle4,5 | <0.001 | |||
| Grade I | 40 (20.9) | 9 (7.7) | 31 (41.9) | |
| Grade II | 75 (39.3) | 52 (44.4) | 23 (31.1) | |
| Grade III | 76 (39.8) | 56 (47.9) | 20 (27.0) | |
| OTA OFC: Arterial6 | 0.065 | |||
| No injury | 135 (64.9) | 74 (58.7) | 61 (74.4) | |
| Injury without ischemia | 10 (4.8) | 7 (5.6) | 3 (3.7) | |
| Injury with distal ischemia | 63 (30.3) | 45 (35.7) | 18 (22.0) | |
| OTA OFC: Bone loss7 | 0.001 | |||
| None | 55 (26.4) | 23 (18.1) | 32 (39.5) | |
| Bone missing/devascularized with contact by proximal/distal fragments | 69 (33.2) | 43 (33.9) | 26 (32.1) | |
| Segmental bone loss | 84 (40.4) | 61 (48.0) | 23 (28.4) | |
| OTA OFA: Contamination8 | 0.001 | |||
| None or minimal | 6 (2.9) | 0 | 6 (7.4) | |
| Surface contamination | 37 (18.0) | 16 (12.8) | 21 (25.9) | |
| Embedded in bone / deep soft-tissue | 96 (46.6) | 67 (53.6) | 29 (35.8) | |
| High-risk environmental conditions | 67 (32.5) | 42 (33.6) | 25 (30.9) |
TTA - transtibial amputation; OTA OFC - Orthopaedic Trauma Association’s Open Fracture Classification
Based on a modified Gustilo-Anderson (GA) classification.28 There were no patients with GA-I fractures in the analysis.
Amputation prior to infection. Among the osteomyelitis case TTAs, 31 were traumatic and 10 were from surgery prior to admission in the United States. All control TTAs were traumatic.
There are 10 patients with unknown/missing classifications (4 cases and 6 controls). Percentages based on total minus unknown.
Grade I - no muscle in area, no appreciable muscle necrosis, some muscle injury with intact function; Grade II - loss of muscle but function remains, some localized necrosis requiring excision, intact muscle-tendon unit; Grade III - dead muscle, loss of function, partial/complete excision, complete disruption of unit, defect does not approximated.
There are 24 patients with unknown/missing classifications (13 cases and 11 controls). Percentages based on total minus unknown.
There are 7 patients with unknown/missing classification (4 case and 3 controls). Percentages based on total minus unknown.
There are 7 patients with unknown classification (3 cases and 4 controls). Percentages based on total minus unknown.
There are 6 patients with unknown classification (2 cases and 4 controls). Percentages based on total minus unknown.
Management and Outcomes
Excluding TTAs, the first type of orthopaedic hardware for the majority of cases and controls involved external fixation of the tibia/fibula (91% and 90%, respectively; Table 3). In contrast, there was a significant difference in the last type of orthopaedic hardware with osteomyelitis cases having a higher proportion of external fixation (68%) compared to controls (33%; p<0.001). A greater amount of osteomyelitis cases were treated with antibiotic beads (approximately 82% vancomycin; 18% tobramycin or composition not stated) compared to controls (p<0.001). Information on use of prophylactic antibiotics within 48 hours post-injury in the combat theater was lacking for 77% of subjects and, thus, not included in the analysis.
Table 3.
Management and Outcomes of Military Personnel with and without Tibia Osteomyelitis, No. (%)
| Total (N=215) | Osteomyelitis Cases (N=130) | Controls (N=85) | P-value | |
|---|---|---|---|---|
| First orthopaedic hardware1 | 1.000 | |||
| External fixation of tibia/fibula | 129 (90.8) | 74 (91.4) | 55 (90.2) | |
| Internal fixation of tibia/fibula | 13 (9.2) | 7 (8.6) | 6 (9.8) | |
| Last orthopaedic hardware1 | <0.001 | |||
| External fixation of tibia/fibula | 75 (52.8) | 55 (67.9) | 20 (32.8) | |
| Internal fixation of tibia/fibula | 67 (47.2) | 26 (32.1) | 41 (67.2) | |
| Any external fixation of tibia/fibula2 | 132 (82.0) | 76 (85.4) | 56 (77.8) | 0.223 |
| Any internal fixation of tibia/fibula2 | 73 (45.3) | 31 (34.8) | 42 (58.3) | 0.004 |
| Time to definitive orthopaedic surgery1 | 0.002 | |||
| Final stabilization <30 days | 66 (46.5) | 27 (33.3) | 39 (63.9) | |
| Final stabilization 30 to <60 days | 8 (5.6) | 6 (7.4) | 2 (3.3) | |
| Final stabilization 2 – 6 months | 14 (9.9) | 12 (14.8) | 2 (3.3) | |
| Final stabilization >6 months | 54 (38.0) | 36 (44.4) | 18 (29.5) | |
| Median time to radiographic union (IQR)3 | 191 (121–281) | 210 (145–318) | 165 (108–229) | 0.014 |
| Time to radiographic union3 | 0.282 | |||
| <6 months | 51 (46.8) | 22 (38.6) | 29 (55.8) | |
| 6 to <9 months | 29 (26.6) | 16 (28.1) | 13 (25.0) | |
| 9 to <12 months | 14 (12.8) | 9 (15.8) | 5 (9.6) | |
| ≥12 months | 15 (13.8) | 10 (17.5) | 5 (9.6) | |
| Type of foreign body at bone site4 | 0.107 | |||
| Fragment with hardware | 74 (34.6) | 50 (38.8) | 24 (28.2) | |
| Fragment only | 31 (14.5) | 22 (17.1) | 9 (10.6) | |
| Hardware only | 76 (35.5) | 39 (30.2) | 37 (43.5) | |
| No foreign body | 33 (15.4) | 18 (14.0) | 15 (17.6) | |
| Bone graft5 | 0.144 | |||
| Allograft only | 6 (2.8) | 4 (3.1) | 2 (2.4) | |
| BMP only | 19 (8.9) | 12 (9.3) | 7 (8.3) | |
| Autograft | 2 (0.9) | 0 | 2 (2.4) | |
| Combination | 9 (4.2) | 4 (3.1) | 5 (6.0) | |
| Unspecified graft | 6 (2.8) | 6 (4.7) | 0 | |
| Antibiotic beads used6 | 101 (48.1) | 77 (59.2) | 24 (30.0) | <0.001 |
| Vancomycin | 83 (39.5) | 63 (48.5) | 20 (25.0) | |
| Death | 1 (0.5) | 0 | 1 (1.2) | 0.395 |
BMP - bone morphogenetic proteins; IQR - interquartile range
Excludes patients with transtibial amputations and no type of fixation/hardware (total = 142; osteomyelitis cases = 81; controls = 61)
Excludes patients with transtibial amputations (total = 161; osteomyelitis cases = 89; controls = 72).
Excludes patients with transtibial amputations and missing time to radiographic union (total = 109; osteomyelitis cases = 57; controls = 52)
Foreign body information missing for 1 osteomyelitis case
Bone graft information unknown for 1 osteomyelitis case and 1 control. Percentages based on total minus unknown.
Antibiotic bead use information unknown for 5 controls. Percentages based on total minus unknown. Antibiotic beads were prior to infection diagnosis.
Osteomyelitis cases had a significantly longer duration to final stabilization (>6 months post-injury for 44% versus 30%; p=0.002; Table 3). Median time to radiographic union was 210 days for osteomyelitis cases and 165 days for controls (p=0.014).
Case-Control Risk Analysis
Blast injury, GA-IIIa and higher fracture class, TTAs (traumatic and early surgical), OTA OFC variables, use of antibiotic beads, and presence of foreign body at bone site (fragment and/or hardware) were significantly associated with risk of osteomyelitis in the univariable model (Table 4). Due to high correlation between GA fracture classification and OTA OFC variables, the two classification schemes warranted separate multivariable assessments (Table 5). In the multivariable model including GA fracture classification, sustaining a blast injury, receiving a classification of GA-IIIb or higher, use of antibiotic beads, and presence of a foreign body fragment at tibia site (with/without hardware) were independently associated with osteomyelitis risk. Traumatic or early surgical TTA prior to infection incurred the highest risk. In the OTA OFC model, use of antibiotic beads remained a significant risk factors and the OTA OFC variable related to muscle loss showed an increasing risk of osteomyelitis.
Table 4.
Unadjusted Logistic Regression Analysis of Risk Factors Associated with Osteomyelitis in Wounded Military Personnel with Open Tibia Fractures
| Full Population | Restricted Population:Excludes TTAs1 | |||
|---|---|---|---|---|
|
| ||||
| OR (95% CI) | P-value | OR (95% CI) | P-value | |
| History of tobacco use | 1.52 (0.87–2.68) | 0.143 | 1.32 (0.70–2.50) | 0.396 |
| Time period | ||||
| 2003–2006 | 1.40 (0.80–2.44) | 0.233 | 1.84 (0.96–3.51) | 0.066 |
| 2007–2009 | Reference | Reference | ||
| Blast mechanism of injury | 2.81 (1.50–5.27) | 0.001 | 2.63 (1.32–5.25) | 0.006 |
| Blood product transfusions within first 24 hours | ||||
| None or missing units | Reference | Reference | ||
| 1–9 units | 1.90 (1.00–3.63) | 0.051 | 1.46 (0.72–2.97) | 0.297 |
| ≥10 units | 2.01 (0.96–4.21) | 0.063 | 1.49 (0.58–3.79) | 0.404 |
| Injury severity score | ||||
| 0–9 (mild) | Reference | Reference | ||
| 10–15 (moderate) | 1.31 (0.69–2.50) | 0.409 | 1.19 (0.59–2.41) | 0.625 |
| ≥16 (severe to life-threatening) | 0.90 (0.45–1.81) | 0.771 | 0.72 (0.31–1.65) | 0.432 |
| Fracture Class2 | ||||
| GA-II | Reference | Reference | ||
| GA-IIIa | 3.63 (1.19–11.06) | 0.024 | 3.63 (1.19–11.06) | 0.024 |
| GA-IIIb | 8.99 (1.19–11.06) | <0.001 | 8.99 (2.78–29.00) | <0.001 |
| GA-IIIc | 4.91 (1.32–18.20) | 0.017 | 4.91 (1.32–18.20) | 0.017 |
| TTA | 10.72 (3.31–34.76) | <0.001 | NA | |
| OTA OFC: Skin | ||||
| Can be approximated | Reference | Reference | ||
| Cannot be approximated | 3.52 (1.68–7.39) | 0.001 | 3.52 (1.68–7.39) | <0.001 |
| Extensive degloving | 4.48 (2.11–9.55) | <0.001 | 4.62 (1.76–12.15) | 0.002 |
| OTA OFC: Muscle3 | ||||
| Grade I | Reference | Reference | ||
| Grade II | 7.79 (3.20–18.96) | <0.001 | 7.79 (3.20–18.96) | <0.001 |
| Grade III | 9.64 (3.92–23.74) | <0.001 | 9.35 (2.99–29.26) | <0.001 |
| OTA OFC: Arterial | ||||
| No injury | Reference | Reference | ||
| Injury without ischemia | 1.92 (0.48–7.75) | 0.358 | 2.00 (0.50–8.09) | 0.329 |
| Injury with distal ischemia | 1.95 (1.04–3.68) | 0.039 | 1.43 (0.49–4.17) | 0.510 |
| OTA OFC: Bone loss | ||||
| None | Reference | Reference | ||
| Bone missing/devascularized with contact by proximal/distal fragments | 2.30 (1.12–4.75) | 0.024 | 2.46 (1.18–5.14) | 0.017 |
| Segmental bone loss | 3.69 (1.80–7.57) | <0.001 | 3.81 (1.52–9.53) | 0.004 |
| OTA OFA: Contamination | ||||
| None, minimal, or surface | Reference | Reference | ||
| Embedded in bone / deep soft-tissue | 3.90 (1.83–8.31) | <0.001 | 3.87 (1.70–8.83) | 0.001 |
| High-risk environmental conditions | 2.83 (1.28–6.26) | 0.010 | 2.49 (1.07–5.79) | 0.035 |
| Time to definitive orthopaedic surgery4 | 1.00 (1.00–1.00) | 0.588 | 1.00 (1.00–1.00) | 0.413 |
| Use of antibiotic beads | 3.39 (1.87–6.13) | <0.001 | 3.54 (1.81–6.93) | <0.001 |
| Type of foreign body at tibia site | ||||
| No foreign body | Reference | Reference | ||
| Fragment with hardware | 1.74 (0.75–4.02) | 0.198 | 4.00 (0.92–17.39) | 0.065 |
| Fragment only | 2.04 (0.72–5.73) | 0.178 | 2.00 (0.31–12.84) | 0.465 |
| Hardware only | 0.88 (0.39–1.99) | 0.757 | 1.78 (0.41–7.71) | 0.438 |
| Bone graft | ||||
| None | Reference | Reference | ||
| Allograft only | 1.32 (0.24–7.41) | 0.752 | 1.75 (0.31–9.90) | 0.529 |
| BMP only | 1.13 (0.42–3.02) | 0.805 | 1.50 (0.55–4.07) | 0.429 |
| Autograft | <0.01 (<0.01 - >999.99) | 0.987 | <0.01 (<0.01 - >999.99) | 0.987 |
| Combination | 0.53 (0.14–2.04) | 0.354 | 0.70 (0.18–2.73) | 0.606 |
| Unspecified graft | >999.99 (<0.01 - >999.99) | 0.978 | >999.99 (<0.01 - >999.99) | 0.978 |
CI – confidence interval; GA – Gustilo-Anderson; OR – odds ratio; OTA OFC – Orthopaedic Trauma Association’s Open Fracture Classification TTA – transtibial amputation
Analysis excludes TTAs (population: 89 osteomyelitis cases and 72 controls)
Based on a modified GA classification.28 There were no patients with GA-I fractures in the analysis. TTA were prior to infection.
Grade I - no muscle in area, no appreciable muscle necrosis, some muscle injury with intact function; Grade II - loss of muscle but function remains, some localized necrosis requiring excision, intact muscle-tendon unit; Grade III - dead muscle, loss of function, partial/complete excision, complete disruption of unit, defect does not approximated.
Excludes patients with transtibial amputations and no type of fixation/hardware
Table 5.
Adjusted Logistic Regression Analysis of Risk Factors Associated with Osteomyelitis in Wounded Military Personnel with Open Tibia Fractures (Odd Ratios; 95% Confidence Intervals)
| Factors | Full Population | Restricted Population:Excludes TTAs1 | ||
|---|---|---|---|---|
| GA Class Model2 | OTA OFC Model2 | GA Class Model2 | OTA OFC Model2 | |
| Time Period | ||||
| 2003–2006 | – | – | 4.09 (1.77–9.46) | 2.58 (1.05–6.32) |
| 2007–2009 | – | – | Reference | Reference |
| Blast injury mechanism | 2.38 (1.14–4.99) | – | 3.21 (1.37–7.52) | 2.68 (1.06–6.79) |
| Fracture classification3 | ||||
| GA-II | Reference | – | Reference | – |
| GA-IIIa | 3.06 (0.89–10.54)4 | – | 2.66 (0.75–9.39)4 | – |
| GA-IIIb | 4.81 (1.33–17.40) | – | 5.19 (1.39–19.38) | – |
| GA-IIIc | 5.68 (1.26–25.60) | – | 4.08 (0.90–18.50)4 | – |
| TTA | 15.10 (3.22–71.07) | – | NA | – |
| Use of antibiotic beads | 3.07 (1.59–5.94) | 2.86 (1.44–5.69) | 4.12 (1.86–9.11) | 3.19 (1.33–7.66) |
| OTA OFC: Muscle,5 | ||||
| Grade I | – | Reference | – | Reference |
| Grade II | – | 5.62 (2.21–14.25) | – | 3.99 (1.50–10.59) |
| Grade III | – | 8.46 (3.31–21.64) | – | 8.08 (2.16–30.20) |
| Type of foreign body at tibia site | ||||
| No foreign body | Reference | – | – | – |
| Fragment with hardware | 5.65 (1.36–23.38) | – | – | – |
| Fragment only | 4.04 (1.15–14.21) | – | – | – |
| Hardware only | 3.14 (0.83–11.90)4 | – | – | – |
TTA – transtibial amputation; GA – Gustilo-Anderson; OTA OFC – Orthopaedic Trauma Association’s Open Fracture Classification
Analysis excludes TTAs (population: 89 osteomyelitis cases and 72 controls)
GA fracture classification and the OTA OFC variables were highly correlated. The multivariable analysis was run separately after selecting for either the GA fracture classification (GA Class Model) or OTA OFC (OTA OFC Model) to be assessed in the model. Stepwise, backward, and forward model selections were conducted to choose the final multivariable risk factor model.
Using a modified GA classification of open fractures.28 There were no GA-I fractures. TTAs were prior to infection.
P-values for variables that were not significant in the model of the full population are: 0.077 for GA-IIIa and 0.092 for hardware. P-values for variables that were not significant in the model of the restricted population are: 0.129 for GA-IIIa and 0.069 for GA-IIIc.
Grade I - no muscle in area, no appreciable muscle necrosis, some muscle injury with intact function; Grade II - loss of muscle but function remains, some localized necrosis requiring excision, intact muscle-tendon unit; Grade III - dead muscle, loss of function, partial/complete excision, complete disruption of unit, defect does not approximated.
Restricted analyses excluded patients with TTA, leaving 89 osteomyelitis cases and 72 controls (Tables 4 and 5). In both the GA classification and OTA OFC variable models, being injured between 2003 and 2006, sustaining a blast injury, and use of antibiotic beads were independent risk factors for developing osteomyelitis (Table 5). In the GA classification model, the level of GA-IIIb was significantly associated with osteomyelitis risk; however, GA-IIIc was not. Similar to the prior model, the OTA OFC variable related to muscle loss was an independent risk factor.
DISCUSSION
Trauma-associated osteomyelitis in this severely injured population is a significant complication given the high frequency of blast trauma resulting in severe open fractures and/or amputations, as well as the frequent contamination of wounds. While our analysis does have inherent limitations, it provides a systematic investigation of factors associated with osteomyelitis risk in a unique patient population. Our analysis found that blast injuries, severity related to open tibia fractures, sustaining an injury between 2003 and 2006, as well as presence of a foreign body at the tibia fracture site, were also risk factors.
The highest risk of developing osteomyelitis was traumatic or early surgical TTAs (Table 5), most frequently associated with IED blast injuries. Having a foreign body fragment (with/without hardware) at the tibia site was also a risk factor and likely results from IED shrapnel. Although risk factors were not assessed, a recent analysis of open tibia fractures among wounded U.K. military personnel found degree of bone loss was significantly associated with infections.14 In our analysis, a significantly higher proportion of segmental bone loss was observed among osteomyelitis cases and while it was significant in the univariable model, the variable was not retained in the multivariable model.
By excluding patients with TTA, the multivariable model assessed characteristics of a population of open fracture patients more comparable to civilian trauma patients. Similar to the full model, OTA OFC for muscle loss remained an independent predictor, indicating high importance of soft-tissue injury severity as related to risk of osteomyelitis. This finding corresponds to a recent analysis of open tibia fractures among civilian patients that identified soft-tissue injury (suggestive of muscle damage/loss) as the most important infection risk factor.25 In our analysis, GA-IIIb fracture classification was also associated with osteomyelitis risk; however, GA-IIIc was not. This is possibly due to limited statistical power as there were only 22 patients with a GA-IIIc classification in our population.
Among civilian trauma patients, fracture classification of GA-IIIa or higher, occurrence of at least one immune comprising condition, and injury location have been identified as independent predictors of osteomyelitis.36 In addition, a recent analysis of 486 patients with open tibia fractures confirmed significance of GA-III fracture classification with infection risk. Specifically, there were a 9.3, 25.4, and 108.9 times more likely risk to become infected with GA-IIIa, GA-IIIb, and GA-IIIc fractures, respectively.25 These data corroborate our findings that infection risk increases with severity of fracture/traumatic TTA. Specifically, in our analysis, the largest proportion of infection was observed with patients classified as GA-IIIb (37 out of 51; 73%), which carried the highest risk after excluding TTAs (Table 5). Nevertheless, the OTA OFC may be a more accurate representation given that it more precisely accounts for soft-tissue injury severity and degree of contamination, with the highest proportion (56 out of 76; 74%) and risk associated with dead muscle (Table 5). One argument to reduce the high infection rate following TTAs would be to perform more proximal early amputations. Unfortunately, zones of injury following IED injuries are frequently massive and most patients have (at least) soft-tissue injuries proximal to the level of the tibial injury, making a “clean” amputation proximal to the injury impossible or requiring dramatic measures, such as hip disarticulations. Furthermore, despite the high early infection rate, all TTA in our study were successfully salvaged at that level, resulting in maintenance of the native knee joint and ostensibly better lifelong function than might be afforded by knee disarticulations or transfemoral amputations.
The significant association with the earlier time period (2003–2006) correlates with changing U.S. military practice patterns and development of clinical practice guidelines for infection prevention in combat casualties.37,38 At the start of the study period, use of high-pressure irrigation was standard practice for treatment of open wounds/fractures and a recent meta-analysis identified pulsatile lavage as an infection risk factor with open fractures (risk ratio: 2.70; 95% confidence interval: 1.03–7.05).39 Around 2006–2007, use of pulsatile lavage decreased and negative pressure wound therapy became more frequent due to data demonstrating promotion of wound healing and reduction in infectious complications. Specifically, studies have shown that negative pressure wound therapy is associated with improved bacterial clearance, blood flow, and granulation tissue in the wounds, along with a reduction in edema, infections, and wound dehiscence.37,40–42 Moreover, important changes in resuscitation occurred around the time period with more restricted use of crystalloid products and optimization of blood product transfusion (1:1 ratio of packed red blood cells and plasma).43 Although not directly assessed, these changes lessening third space fluid may have impacted risk for late infection events. Regarding the association of osteomyelitis risk with use of antibiotic beads prior to infection diagnosis, the surgeon’s decision to use antibiotic beads is based on appearance and severity of the wound, along with extent of defect. During the wars in Iraq and Afghanistan, IEDs were the predominant injury mechanism, resulting in grievous extremity injuries characterized by extensive soft-tissue and muscle damage.44 Due to the nature of the injuries, risk of infection was high; therefore, it is probable that antibiotic beads were placed preferentially in more severe injuries which, in turn, were more prone to develop osteomyelitis. While use of antibiotics was not identified as risk factors in the recent meta-analysis, factors related to polytrauma and fracture severity were associated with infections,39 which is consistent with our findings.
The recent meta-analysis examined a wide range of infection risk factors associated with open fractures.39 Smoking was found to be associated with increased risk of infections;39 however, it was not identified as an independent predictor in our analysis nor in the analysis of U.K. military personnel.14 Prior analyses have examined the timing of antibiotic prophylaxis with regards to infections; however, Kortram et al.39 found that osteomyelitis risk was not elevated when antibiotics were received >3 hours after injury (risk ratio: 1.29; 95% confidence interval: 0.59–2.79). Furthermore, blood transfusions have been shown to modulate the immune system and are associated with infection risk, including osteomyelitis.6,39 Blood volume was not associated with osteomyelitis risk in this military population with battlefield injuries. Timing of surgical debridement was also examined as a risk factor. Guidelines recommend that surgical debridement occurs <6 hours for the management of open fractures. Precise timing of first operative procedure was available for 74 patients in our analysis with 85% having their first operating room visit <6 hours post-injury. In the recent meta-analysis, timing of debridement (>6 hours versus <6 hours) was not significantly associated with infection risk,39 indicating delayed surgical debridement may not negatively impact morbidity. These findings confirmed the results of a separate meta-analysis, which compared data between patients with open tibia fractures who received early (<6 hours) and late (>6 hours) surgical debridements and found no association between timing of debridement and rate of overall infections, deep infections, and nonunion.45 While not significant as a risk factor in the meta-analysis,39 external fixation has been associated with higher infection rates with GA-I and GA-II fractures, while it decreased likelihood of infections with GA-III fractures.46 GA-III fractures comprised 84% of our population (excluding TTA) and while use of any external fixation was not significantly different between cases and controls (85% versus 78%, respectively), more cases did have external fixation as their last orthopedic hardware (68% versus 33%; p<0.001). The timing of wound closure has also been examined with occurrence of closure >5 days post-injury being identified as an infection risk factor with open tibia fractures.47
Our analysis does have several important limitations. In particular, the majority of our cases are classified as possible based on NHSN criteria, which could have allowed the inclusion of cases with deep soft-tissue infections (with or without osteomyelitis) in our analysis. This limitation stems from the inherent complexity of combat trauma resulting in open tibia fractures or amputations. Due to the retrospective nature of this analysis, we were limited to reviewing available medical records in which operative descriptions may lack the specificity required to classify an osteomyelitis as definite coupled with infrequent bone cultures and histopathology. In addition, due to the low number of patients with open tibia fractures who did not meet osteomyelitis diagnostic criteria, we were not able to match the case and control populations. Detailed information on use of early post-trauma antibiotics in the combat theater is lacking. Precise timing between injury and initial surgical care was unavailable for the majority of patients. Another limitation is the lack of available information on the timing of wound closure in our population. Furthermore, data in our analysis may not be applicable to general civilian trauma as the majority of the population sustained blast-related trauma. Lastly, patient-level data related to practice patterns (e.g., use of negative pressure wound therapy) is lacking.
Despite these limitations, our analysis demonstrates risk factors for osteomyelitis among combat casualties is highest with severe blast trauma resulting in substantial muscle loss/damage with the most extreme circumstances being trauma/early surgical TTAs. Time-dependent variables related to early care were not included in this analysis, but could represent important risk factors and should be evaluated in future research. Future research includes an evaluation of the risk of recurrent osteomyelitis based on initial classification, wound severity, microbiology, antibiotic therapy, and early surgical care.
Acknowledgments
Sources of Funding: Support for this work was provided by the Infectious Disease Clinical Research Program, a DoD program executed through the Uniformed Services University of the Health Sciences, Department of Preventive Medicine and Biostatistics. This project is funded by the National Institute of Allergy and Infectious Diseases, National Institutes of Health, under Inter-Agency Agreement Y1-AI-5072, and Department of the Navy under the Wounded, Ill, and Injured Program.
Footnotes
Presented in part at the 2017 Military Health System Research Symposium (27–30 August 2017, Kissimmee, FL), 2016 AAOS/OTA/SOMOS/ORS Extremity War Injuries XI: Maintaining Force Readiness During an Era of Military Transition (18–20 January 2016, Washington, DC), and 2015 ID Week, A Joint Meeting of IDSA, SHEA, HIVMA, and PIDS (7–11 October 2015, San Diego, CA).
Conflicts of Interest: The authors have no conflicts of interest to declare.
Disclaimer: The views expressed are those of the authors do not necessarily reflect the official views of the Uniformed Services University of the Health Sciences, Henry M. Jackson Foundation for the Advancement of Military Medicine, Inc., the National Institutes of Health or the Department of Health and Human Services, Brooke Army Medical Center, Walter Reed National Military Medical Center, Landstuhl Regional Medical Center, the U.S. Army Medical Department, the U.S. Army Office of the Surgeon General, the Department of Defense, or the Departments of the Army, Navy or Air Force. Mention of trade names, commercial products, or organization does not imply endorsement by the U.S. Government.
References
- 1.Owens BD, Kragh JF, Jr, Macaitis J, et al. Characterization of extremity wounds in Operation Iraqi Freedom and Operation Enduring Freedom. J Orthop Trauma. 2007;21:254–257. doi: 10.1097/BOT.0b013e31802f78fb. [DOI] [PubMed] [Google Scholar]
- 2.Andersen CR, D’Alleyrand MP, Swiontkowski MF, et al. Extremity War Injuries VIII: Sequelae of Combat Injuries. J Am Acad Orthop Surg. 2014;22:57–62. doi: 10.5435/JAAOS-22-01-57. [DOI] [PubMed] [Google Scholar]
- 3.Ficke JR, Eastridge BJ, Butler F, et al. Dismounted complex blast injury report of the Army Dismounted Complex Blast Injury Task Force. J Trauma Acute Care Surg. 2012;73:S520–S534. [Google Scholar]
- 4.Schoenfeld AJ, Dunn JC, Bader JO, et al. The nature and extent of war injuries sustained by combat specialty personnel killed and wounded in Afghanistan and Iraq, 2003–2011. J Trauma Acute Care Surg. 2013;75:287–291. doi: 10.1097/TA.0b013e31829a0970. [DOI] [PubMed] [Google Scholar]
- 5.Krueger CA, Wenke JC, Ficke JR. Ten years at war: comprehensive analysis of amputation trends. J Trauma Acute Care Surg. 2012;73:S438–S444. doi: 10.1097/TA.0b013e318275469c. [DOI] [PubMed] [Google Scholar]
- 6.Tribble DR, Conger NG, Fraser S, et al. Infection-associated clinical outcomes in hospitalized medical evacuees after traumatic injury: Trauma Infectious Disease Outcome Study. J Trauma. 2011;71:S33–S42. doi: 10.1097/TA.0b013e318221162e. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Brown KV, Murray CK, Clasper JC. Infectious complications of combat-related mangled extremity injuries in the British military. J Trauma. 2010;69(Suppl 1):S109–S115. doi: 10.1097/TA.0b013e3181e4b33d. [DOI] [PubMed] [Google Scholar]
- 8.Johnson EN, Burns TC, Hayda RA, et al. Infectious complications of open type III tibial fractures among combat casualties. Clin Infect Dis. 2007;45:409–415. doi: 10.1086/520029. [DOI] [PubMed] [Google Scholar]
- 9.Murray CK, Obremskey WT, Hsu JR, et al. Prevention of infections associated with combat-related extremity injuries. J Trauma. 2011;71:S235–S257. doi: 10.1097/TA.0b013e318227ac5f. [DOI] [PubMed] [Google Scholar]
- 10.Murray CK, Wilkins K, Molter NC, et al. Infections complicating the care of combat casualties during Operations Iraqi Freedom and Enduring Freedom. J Trauma. 2011;71:S62–S73. doi: 10.1097/TA.0b013e3182218c99. [DOI] [PubMed] [Google Scholar]
- 11.Murray CK, Wilkins K, Molter NC, et al. Infections in combat casualties during Operations Iraqi and Enduring Freedom. J Trauma. 2009;66:S138–S144. doi: 10.1097/TA.0b013e31819d894c. [DOI] [PubMed] [Google Scholar]
- 12.Yun HC, Branstetter JG, Murray CK. Osteomyelitis in military personnel wounded in Iraq and Afghanistan. J Trauma. 2008;64:S163–168. doi: 10.1097/TA.0b013e318160868c. [DOI] [PubMed] [Google Scholar]
- 13.Burns TC, Stinner DJ, Mack AW, et al. Microbiology and injury characteristics in severe open tibia fractures from combat. J Trauma Acute Care Surg. 2012;72:1062–1067. doi: 10.1097/TA.0b013e318241f534. [DOI] [PubMed] [Google Scholar]
- 14.Penn-Barwell JG, Bennett PM, Mortiboy DE, et al. Factors influencing infection in 10 years of battlefield open tibia fractures. Strategies Trauma Limb Reconstr. 2016;11:13–18. doi: 10.1007/s11751-016-0250-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Gustilo RB, Mendoza RM, Williams DN. Problems in the management of type III (severe) open fractures: a new classification of type III open fractures. J Trauma. 1984;24:742–746. doi: 10.1097/00005373-198408000-00009. [DOI] [PubMed] [Google Scholar]
- 16.Patzakis MJ, Wilkins J. Factors influencing infection rate in open fracture wounds. Clin Orthop Relat Res. 1989:36–40. [PubMed] [Google Scholar]
- 17.Gustilo RB, Gruninger RP, Davis T. Classification of type III (severe) open fractures relative to treatment and results. Orthopedics. 1987;10:1781–1788. [PubMed] [Google Scholar]
- 18.Griffin M, Malahias M, Hindocha S, et al. Update on the management of compound lower limb fractures. Open Orthop J. 2012;6:518–524. doi: 10.2174/1874325001206010518. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Griffin M, Malahias M, Khan W, et al. Update on the management of open lower limb fractures. Open Orthop J. 2012;6:571–577. doi: 10.2174/1874325001206010571. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Parrett BM, Matros E, Pribaz JJ, et al. Lower extremity trauma: trends in the management of soft-tissue reconstruction of open tibia-fibula fractures. Plast Reconstr Surg. 2006;117:1315–1322. doi: 10.1097/01.prs.0000204959.18136.36. discussion 1323–1324. [DOI] [PubMed] [Google Scholar]
- 21.Rajasekaran S, Naresh Babu J, Dheenadhayalan J, et al. A score for predicting salvage and outcome in Gustilo type-IIIA and type-IIIB open tibial fractures. J Bone Joint Surg Br. 2006;88:1351–1360. doi: 10.1302/0301-620X.88B10.17631. [DOI] [PubMed] [Google Scholar]
- 22.Naique SB, Pearse M, Nanchahal J. Management of severe open tibial fractures: the need for combined orthopaedic and plastic surgical treatment in specialist centres. J Bone Joint Surg Br. 2006;88:351–357. doi: 10.1302/0301-620X.88B3.17120. [DOI] [PubMed] [Google Scholar]
- 23.Giannoudis PV, Papakostidis C, Roberts C. A review of the management of open fractures of the tibia and femur. J Bone Joint Surg Br. 2006;88:281–289. doi: 10.1302/0301-620X.88B3.16465. [DOI] [PubMed] [Google Scholar]
- 24.Yusof NM, Khalid KA, Zulkifly AH, et al. Factors associated with the outcome of open tibial fractures. Malays J Med Sci. 2013;20:47–53. [PMC free article] [PubMed] [Google Scholar]
- 25.Thakore RV, Francois EL, Nwosu SK, et al. The Gustilo-Anderson classification system as predictor of nonunion and infection in open tibia fractures. Eur J Trauma Emerg Surg. 2016 doi: 10.1007/s00068-016-0725-y. doi:10/1007/s00068-016-0725-y. [DOI] [PubMed] [Google Scholar]
- 26.Huh J, Stinner DJ, Burns TC, et al. Infectious complications and soft tissue injury contribute to late amputation after severe lower extremity trauma. J Trauma. 2011;71:S47–S51. doi: 10.1097/TA.0b013e318221181d. [DOI] [PubMed] [Google Scholar]
- 27.Napierala MA, Rivera JC, Burns TC, et al. Infection reduces return-to-duty rates for soldiers with Type III open tibia fractures. J Trauma Acute Care Surg. 2014;77:S194–S197. doi: 10.1097/TA.0000000000000364. [DOI] [PubMed] [Google Scholar]
- 28.Gustilo RB, Anderson JT. 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. [PubMed] [Google Scholar]
- 29.Marsh JL, Slongo TF, Agel J, et al. Fracture and dislocation classification compendium - 2007: Orthopaedic Trauma Association classification, database and outcomes committee. J Orthop Trauma. 2007;21:S1–S133. doi: 10.1097/00005131-200711101-00001. [DOI] [PubMed] [Google Scholar]
- 30.Orthopaedic Trauma Association: Open Fracture Study Group. A new classification scheme for open fractures. J Orthop Trauma. 2010;24:457–464. doi: 10.1097/BOT.0b013e3181c7cb6b. [DOI] [PubMed] [Google Scholar]
- 31.Centers for Disease Control and Prevention. [Accessed March 6, 2017];CDC/NHSN Surveillance Definitions for Specific Types of Infections. 2014 Available at: http://www.cdc.gov/nhsn/pdfs/pscmanual/17pscnosinfdef_current.pdf.
- 32.Louisiana Department of Health. [Accessed August 31, 2017];Louisiana heathcare-associated infections and antibiotic resistance. Available at: http://dhh.louisiana.gov/index.cfm/page/2783.
- 33.Oregon Health Authority. [Accessed August 31, 2017];Healthcare associated infections in Oregon. Available at: http://www.oregon.gov/oha/PH/DISEASESCONDITIONS/COMMUNICABLEDISEASE/HAI/Pages/index.aspx.
- 34.Texas Department of State Health Services. [Accessed August 31, 2017];Annual report on healthcare associated infections. 2014 Available at: http://txhsn.dshs.texas.gov/HCSReports/AnnualReports/2014%20Texas%20Annual%20Report_FINAL.pdf.
- 35.European Centre for Disease Control and Prevention. Technical document. Point prevalence survey of healthcare-associated infections and antimicrobial use in European acute care hospitals. [Accessed August 31, 2017];Protocol version 5.3. Available at: https://ecdc.europa.eu/sites/portal/files/media/en/publications/Publications/PPS-HAI-antimicrobial-use-EU-acute-care-hospitals-V5-3.pdf.
- 36.Bowen TR, Widmaier JC. Host classification predicts infection after open fracture. Clin Orthop Relat Res. 2005:205–211. doi: 10.1097/01.blo.0000150345.51508.74. [DOI] [PubMed] [Google Scholar]
- 37.Murray CK, Hsu JR, Solomkin JS, et al. Prevention and management of infections associated with combat-related extremity injuries. J Trauma. 2008;64:S239–S251. doi: 10.1097/TA.0b013e318163cd14. [DOI] [PubMed] [Google Scholar]
- 38.Hospenthal DR, Murray CK, Andersen RC, et al. Guidelines for the prevention of infections associated with combat-related injuries: 2011 update: endorsed by the Infectious Diseases Society of America and the Surgical Infection Society. J Trauma. 2011;71:S210–S234. doi: 10.1097/TA.0b013e318227ac4b. [DOI] [PubMed] [Google Scholar]
- 39.Kortram K, Bezstarosti H, Metsemakers WJ, et al. Risk factors for infectious complications after open fractures; a systematic review and meta-analysis. Int Orthop. 2017 doi: 10.1007/s00264-017-3556-5. [DOI] [PubMed] [Google Scholar]
- 40.Stannard JP, Volgas DA, Stewart R, et al. Negative pressure wound therapy after severe open fractures: a prospective randomized study. J Orthop Trauma. 2009;23:552–557. doi: 10.1097/BOT.0b013e3181a2e2b6. [DOI] [PubMed] [Google Scholar]
- 41.Leininger BE, Rasmussen TE, Smith DL, et al. Experience with wound VAC and delayed primary closure of contaminated soft tissue injuries in Iraq. J Trauma. 2006;61:1207–1211. doi: 10.1097/01.ta.0000241150.15342.da. [DOI] [PubMed] [Google Scholar]
- 42.Park CH, Shon OJ, Kim GB. Negative pressure wound therapy for Gustilo Anderson grade IIIb open tibial fractures. Indian J Orthop. 2016;50:536–542. doi: 10.4103/0019-5413.189604. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Langan NR, Eckert M, Martin MJ. Changing patterns of in-hospital deaths following implementation of damage control resuscitation practices in US forward military treatment facilities. JAMA Surg. 2014;149:904–912. doi: 10.1001/jamasurg.2014.940. [DOI] [PubMed] [Google Scholar]
- 44.Belmont PJ, Schoenfeld AJ, Goodman G. Epidemiology of combat wounds in Operation Iraqi Freedom and Operation Enduring Freedom: orthopaedic burden of disease. J Surg Orthop Adv. 2010;19:2–7. [PubMed] [Google Scholar]
- 45.Prodromidis AD, Charalambous CP. The 6-hour rule for surgical debridement of open tibial fractures. A systematic review and meta-analysis of infection and non-union rates. J Orthop Trauma. 2016;30:397–402. doi: 10.1097/BOT.0000000000000573. [DOI] [PubMed] [Google Scholar]
- 46.Duyos OA, Beaton-Comulada D, Davila-Parrilla A, et al. Management of open tibial shaft fractures: does the timing of surgery affect outcomes? J Am Acad Orthop Surg. 2017;25:230–238. doi: 10.5435/JAAOS-D-16-00127. [DOI] [PubMed] [Google Scholar]
- 47.Lack WD, Karunakar MA, Angerame MR, et al. Type III open tibia fractures: immediate antibiotic prophylaxis minimizes infection. J Orthop Trauma. 2015;29:1–6. doi: 10.1097/BOT.0000000000000262. [DOI] [PubMed] [Google Scholar]
- 48.Linn S. The injury severity score--Importance and uses. Ann Epidemiol. 1995;5:440–446. doi: 10.1016/1047-2797(95)00059-3. [DOI] [PubMed] [Google Scholar]