Abstract
We examined risk factors for combat-related extremity wound infections (CEWI) among U.S. military patients injured in Iraq and Afghanistan (2009–2012). Patients with ≥1 combat-related, open extremity wound admitted to a participating U.S. hospital (≤7 days postinjury) were retrospectively assessed. The population was classified based upon most severe injury (amputation, open fracture without amputation, or open soft-tissue injury defined as non-fracture/non-amputation wounds). Among 1271 eligible patients, 395 (31%) patients had ≥1 amputation, 457 (36%) had open fractures, and 419 (33%) had open soft-tissue wounds as their most severe injury, respectively. Among patients with traumatic amputations, 100 (47%) developed a CEWI compared to 66 (14%) and 12 (3%) patients with open fractures and open soft-tissue wounds, respectively. In a Cox proportional hazard analysis restricted to CEWIs ≤30 days postinjury among the traumatic amputation and open fracture groups, sustaining an amputation (hazard ratio: 1.79; 95% confidence interval: 1.25–2.56), blood transfusion ≤24 hours postinjury, improvised explosive device blast, first documented shock index ≥0.80, and >4 injury sites were independently associated with CEWI risk. The presence of a non-extremity infection at least 4 days prior to a CEWI diagnosis was associated with lower CEWI risk, suggesting impact of recent exposure to directed antimicrobial therapy. Further assessment of early clinical management will help to elucidate risk factor contribution. The wound classification system provides a comprehensive approach in assessment of injury and clinical factors for the risk and outcomes of an extremity wound infection.
Keywords: extremity wounds, combat-related, trauma infections, polytrauma, wound infections
INTRODUCTION
It is well recognized that extremity trauma is the predominant injury pattern among combat casualties.1–9 In particular, approximately 82% of military personnel wounded between 2001 and 2005 during the wars in Iraq and Afghanistan sustained trauma to at least one extremity.1 Furthermore, following transition of service members into Afghanistan, a high proportion of severe polytrauma was reported among combat casualties who sustained blast injuries while on foot patrol. This injury pattern was characterized by an amputation of a lower extremity (at or above knee) with serious injury or amputation of the opposite extremity as well as abdominal, pelvic or genitourinary injury. Injuries of the upper extremities were also common with many combat casualties incurring triple or quadruple amputations.5,10 Despite the high injury severity, mortality declined during the recent wars as a result of the implementation of a Joint Trauma System, improvements in combat casualty care, and advancements in personal protective equipment.11–13
As the survivability of combat casualties with grievous injuries increased, there was a corresponding rise in post-traumatic infections,14–18 particularly associated with the extremities.9,19–23 In both civilian and military populations, post-traumatic extremity wound infections are often associated with high resource utilization and significant morbidity, including reoperations and late amputations.24–28 As a result, the identification of predictors for extremity wound infections is a critical element for improving clinical outcomes.
Prior studies have identified injury severity, blast injury mechanism, volume of blood transfusions, sustaining an open or soft-tissue injury, use of prophylactic antibiotics within 48 hours postinjury, having an external fixator, and admission to the intensive care unit with risk of infectious complications following combat-related trauma.15,29 Use of antibiotics immediately after injury within the combat theater or during medical evacuation, application of tourniquets in the field, and fasciotomy have also been associated with risk of extremity wound infection.19 Bone loss had also been associated with infections following combat-related open fractures.23
With the high resource utilization and morbidity associated with trauma-related infections, further information on predictors is needed to improve clinical outcomes. As a result, we developed refined classifications and methodology to improve characterization of extremity polytrauma. Using this classification system, we examined risk factors for the development of combat-related extremity wound infections (CEWIs) among wounded U.S. military personnel.
METHODS
Study Population and Data Sources
The overarching study is the Department of Defense (DoD) – Veterans Affairs Trauma Infectious Disease Outcomes Study (TIDOS), which is a longitudinal, observational, multicenter cohort study designed to examine short- and long-term infectious complications among military personnel injured during deployment in Iraq and Afghanistan.14 Patients were eligible for inclusion in TIDOS if they were active-duty personnel or DoD beneficiaries, 18 years of age or older who were injured during deployment and medically evacuated to Landstuhl Regional Medical Center (LRMC; Germany) before being transferred to a participating military hospital in the USA. The participating U.S hospitals were Walter Reed National Military Medical Center in the National Capital Region (Walter Reed Army Medical Center and National Naval Medical Center prior to September 2011) and Brooke Army Medical Center in Texas. The study was approved by the Infectious Disease Institutional Review Board of the Uniformed Services University of the Health Sciences in Bethesda, Maryland.
Inclusion in this retrospective cohort analysis required that patients sustained at least one combat-related open extremity wound excluding shoulder and pelvic injuries between June 2009 and May 2012 and were admitted to LRMC within 6 days of injury before being transferred to a participating U.S. hospital within 7 days postinjury. Wounds were excluded from the analysis if they were amputation or fractures of the digits, distal to a more proximal surgical amputation prior to admission at a participating U.S. hospital, or if the qualifying extremity wound was not detailed in the medical records until admission to a U.S. hospital (Fig. 1).
FIGURE 1.
Study population derivation of patients with combat-related open extremity wounds, by wound type and infectious outcome. aThe population for the risk factor model excluded early surgical amputations and only considered infections diagnosed within 30 days postinjury.
Patient demographics, injury characteristics, early clinical indicators, and hospitalization were prospectively collected through the DoD Trauma Registry (DoDTR).30 Information on infectious outcomes was obtained from the DoDTR TIDOS infectious disease module through the TIDOS project.14
Injury Characterization
Injury patterns were classified based on the injury narrative and Abbreviated Injury Scale (AIS) codes31 generated through Tri-Code (Digital Innovation, Inc., Forest Hill, MD). All wounds were mapped to pre-defined anatomic TIDOS injury sites (upper arm, elbow, lower arm, wrist, hand, thigh, knee, lower leg, ankle, and foot) and side (left and right) using AIS codes and injury descriptions. Wounds were categorized as either amputation (traumatic or early surgical if within seven days postinjury), open fracture or open soft-tissue wounds (e.g., degloving injuries, lacerations, and penetrating wounds). Patients were classified based on the most severe wound sustained with amputation as the top category, followed by open fracture and open soft-tissue wounds. Amputations or fractures of the digits, toes or interphalangeals; closed wounds; non-combat related wounds; and wounds with an unknown AIS severity score were excluded from the study (Fig. 1).
Wounds with missing side information were processed based on a series of assumptions. First, if an amputation was missing a side, it was excluded if there were two other amputations with known sides reported in the same body region. In the case of two wounds with same injury type (e.g., amputation, fracture, or soft-tissue wound) being reported at the same site and side was only identified with one of the wounds, the wound without a known side was excluded. If a fracture was missing a side and was reported at the same TIDOS site as another fracture that had a side identified, it was assumed to be on that same side if it contained more than one bone. If injuries sustained by one patient were all related to one side, then a wound with missing side information from that same patient was considered to be on the same side as the other injuries. Amputations missing site-specific details (listed as not otherwise specified; NOS) were excluded if the patient had another amputation on the same side and within the same body region that did include site-specific details.
Infection Classification
Skin and soft-tissue infections (SSTIs) and osteomyelitis were identified based upon clinical findings and laboratory test results via medical chart review and classified using the standardized definitions of the National Healthcare Safety Network.14,32 Infections that did not meet the a priori definition were included in the analysis if there was a recorded clinical diagnosis along with directed antimicrobial treatment (duration of ≥5 days for SSTIs and ≥21 days for osteomyelitis). If there was a record of an alternate diagnosis and discontinuation of antimicrobial treatment, the infection was excluded.
Statistical Analysis
Categorical and continuous variables were compared using Fisher’s exact tests and non-parametric tests, respectively. Variables were examined for their association with time-to-CEWI within the first 30 days postinjury using a Cox proportional hazard model. Due to the low prevalence of infections among open soft-tissue wounds alone, the population utilized in the risk factor analysis was restricted to patients with traumatic amputations and open fractures without amputations (patients who only experienced open soft-tissue wounds or had early surgical amputations for traumatic distal wounds were excluded). The risk factor analysis further restricted the population to infections diagnosed within 30 days postinjury. Statistical analysis was conducted with SAS version 9.4 (SAS, Cary, NC, USA). Data are expressed as hazard ratios with 95% confidence intervals. Statistical significance was defined as p < 0.05.
RESULTS
Study Population
A total of 1858 patients were admitted to participating U.S. hospitals for a combat-related injury, of which 1271 (68%) met criteria for inclusion in the analysis (Fig. 1). The patients were predominantly young men (median age of 24 years) who sustained blast-related trauma (81%) while serving in support of operations in Afghanistan (94%; Table I). Among the trauma patients, 395 (31%) had an amputation (traumatic or early surgical) as their most severe injury, 457 (36%) had an open fracture, and 419 (33%) had an open soft-tissue wound.
TABLE I.
Characteristics and Injury Circumstances of Military Personnel With Combat-Related Extremity Wounds
| Total Patients | Patients With Amputations | Patients With Fractures | Patients With Open Soft-tissue Woundsa | p-Value | |
|---|---|---|---|---|---|
| (N = 1271) | (N = 395) | (N = 457) | (N = 419) | ||
| Maleb | 1223 (98.9) | 376 (98.7) | 441 (98.9) | 406 (99.3) | 0.739 |
| Age, median years (IQR) | 24 (22–28) | 24 (22–27) | 24 (22–28) | 24 (22–28) | 0.019 |
| Operational theater | 0.003 | ||||
| Afghanistan | 1190 (93.6) | 384 (97.2) | 415 (90.8) | 391 (93.3) | |
| Iraq | 64 (5.0) | 9 (2.3) | 34 (7.4) | 21 (5.0) | |
| Other | 17 (1.3) | 2 (0.5) | 8 (1.8) | 7 (1.7) | |
| Branch of servicec | <0.001 | ||||
| Air Force | 30 (2.5) | 5 (1.3) | 17 (3.9) | 8 (2.0) | |
| Army | 711 (58.5) | 180 (47.7) | 281 (63.7) | 250 (62.8) | |
| Marine | 429 (35.3) | 182 (48.3) | 122 (27.7) | 125 (31.4) | |
| Navy | 38 (3.1) | 9 (2.4) | 15 (3.4) | 14 (3.5) | |
| Other | 8 (0.7) | 1 (0.3) | 6 (1.4) | 1 (0.3) | |
| Mechanism of Injury | |||||
| Blast | 1032 (81.2) | 391 (99.0) | 327 (71.6) | 314 (74.9) | <0.001 |
| IED | 823 (64.8) | 358 (90.6) | 249 (54.5) | 216 (51.6) | <0.001 |
| Non-IED | 209 (16.4) | 33 (8.4) | 78 (17.1) | 98 (23.4) | |
| Non-blastd | 239 (18.8) | 4 (1.0) | 130 (28.4) | 105 (25.1) | |
| Single injury mechanism | 0.055 | ||||
| Yes | 1213 (95.4) | 385 (97.5) | 432 (94.5) | 396 (94.5) | |
| No | 58 (4.6) | 10 (2.5) | 25 (5.5) | 23 (5.5) | |
| Injured on foot patrole | 478 (41.2) | 254 (64.5) | 97 (24.9) | 127 (33.7) | <0.001 |
| Number of limbs injured | <0.001 | ||||
| 1 | 451 (35.5) | 32 (8.1) | 203 (44.4) | 216 (51.6) | |
| 2 | 369 (29.0) | 104 (26.3) | 148 (32.4) | 117 (27.9) | |
| 3 | 294 (23.1) | 154 (39.0) | 86 (18.8) | 54 (12.9) | |
| 4 | 157 (12.4) | 105 (26.6) | 20 (4.4) | 32 (7.6) | |
| Injury location | <0.001 | ||||
| Both upper and lower extremity | 577 (45.4) | 278 (70.4) | 164 (35.9) | 135 (32.2) | |
| Lower extremity only | 469 (36.9) | 105 (26.6) | 203 (44.4) | 161 (38.4) | |
| Upper extremity only | 225 (17.7) | 12 (3.0) | 90 (19.7) | 123 (29.4) | |
| Injury severity score, median (IQR) | 21 (10–33) | 30 (21–38) | 17 (10–27) | 12 (6–27) | <0.001 |
| 0–9 (minor) | 261 (20.5) | 9 (2.3) | 73 (16.0) | 179 (42.7) | <0.001 |
| 10–15 (moderate) | 192 (15.1) | 4 (1.0) | 135 (29.5) | 53 (12.6) | |
| 16–24 (severe) | 285 (22.4) | 103 (26.1) | 120 (26.3) | 62 (14.8) | |
| ≥25 (life-threatening) | 533 (41.9) | 279 (70.6) | 129 (28.2) | 125 (29.8) | |
| RBC units within 24 hours postinjury, median (IQR) | 9 (4–18) | 15 (9–26) | 5 (2–12) | 4 (2–8) | <0.001 |
| None/missing | 577 | 44 | 218 | 315 | <0.001 |
| 1–9 | 3340 (49.0) | 94 (26.8) | 161 (67.4) | 85 (81.7) | |
| 10–20 | 200 (28.8) | 129 (36.8) | 58 (24.3) | 13 (12.5) | |
| ≥21 | 154 (22.2) | 128 (36.5) | 20 (8.4) | 6 (5.8) | |
| First documented shock index, median (IQR) | 0.8 (0.6–1.0) | 1.0 (0.7–1.3) | 0.7 (0.6–0.9) | 0.7 (0.6–0.8) | <0.001 |
| <0.65 | 416 (32.7) | 62 (15.7) | 180 (39.4) | 174 (41.5) | <0.001 |
| 0.65 to <0.80 | 294 (23.1) | 70 (17.7) | 104 (22.8) | 120 (28.6) | |
| ≥0.80 | 561 (44.1) | 263 (66.6) | 173 (37.9) | 125 (29.8) |
IED, improvised explosive blast; IQR, interquartile range; RBC, red blood cell.
aOpen soft-tissue wounds includes degloving injuries, lacerations, and penetrating wounds.
bGender information is missing for 35 patients. Percentages and p-value based on total minus missing.
cBranch of service is missing for 55 patients. Percentages and p-value based on total minus missing.
dNon-blast includes gunshot wound, motor vehicle crash, and other mechanism.
eMounted versus dismounted status is missing for 110 patients. Percentages and p-value based on total minus missing.
A higher proportion of patients with amputations were injured in Afghanistan (97% versus 91% and 93%) via a blast mechanism (99% versus 72% and 75%) compared with patients with open fractures or soft-tissue wounds, respectively (Table I). In addition, 71% of patients with amputations incurred life threatening injuries (injury severity score ≥25) and 66% sustained injuries to at least three extremities compared to 28% and 23% of patients with open fractures and 30% and 20% with open soft-tissue wounds, respectively. Lastly, patients with amputations had a higher first documented shock index (median 1.0 versus 0.7 with other injury groups) and required a greater number of units of blood within 24 hours postinjury (median 15 versus 4-5 units).
Hospitalization Findings
A higher proportion of patients with amputations were admitted to the intensive care unit at LRMC (77%) and the U.S. hospitals (63%) compared to patients with open fractures (43% and 31%, respectively) and open soft-tissue wounds (38% and 29%, respectively; Table II). Sequential Organ Failure Assessment (SOFA) scores at admission were also higher in the amputation group versus the open soft-tissue wound groups (median of 4 versus 2 and 1 with open fractures and open soft-tissue wounds, respectively). Patients with amputations and open fractures primarily required ≥3 visits to the operating room (67% for both), while the majority of patients with open soft-tissue wounds had ≤2 visits (66%).
TABLE II.
Hospitalization Characteristics Among Military Personnel With Combat-Related Extremity Wounds
| Total Patients | Patients With Amputations | Patients With Fractures | Patients With Open Soft-tissue Woundsa | p-Value | |
|---|---|---|---|---|---|
| (N = 1271) | (N = 395) | (N = 457) | (N = 419) | ||
| Admission unit at LRMCb | <0.001 | ||||
| Intensive care unit | 659 (51.9) | 304 (77.0) | 195 (42.9) | 160 (38.2) | |
| Non-critical care | 610 (48.1) | 91 (23.0) | 260 (57.1) | 259 (61.8) | |
| Admission unit at U.S. hospitalsc | <0.001 | ||||
| Intensive care unit | 511 (40.3) | 247 (62.5) | 142 (31.1) | 122 (29.3) | |
| Non-critical care | 757 (59.7) | 148 (37.5) | 314 (68.9) | 295 (70.7) | |
| Admission SOFA score, median (IQR) | 2.0 (0–5.0) | 4.0 (1.0–7.0) | 2.0 (0–5.0) | 1.0 (0–3.0) | <0.001 |
| Mechanical ventilationd | <0.001 | ||||
| None at LRMC or 1st U.S. hospital | 828 (65.1) | 178 (45.1) | 326 (71.3) | 324 (77.3) | |
| LRMC ± U.S. hospital ≤7 days | 203 (16.0) | 94 (23.8) | 59 (12.9) | 50 (11.9) | |
| LRMC ± U.S. hospital >7 days | 240 (18.9) | 123 (31.1) | 72 (15.8) | 45 (10.7) | |
| OR visit prior to infection diagnosis or hospital discharge | <0.001 | ||||
| None | 153 (12.0) | 45 (11.4) | 24 (5.3) | 84 (20.0) | |
| 1–2 | 405 (31.9) | 84 (21.3) | 127 (27.8) | 194 (46.3) | |
| 3–4 | 364 (28.6) | 131 (33.2) | 142 (31.1) | 91 (21.7) | |
| ≥5 | 349 (27.5) | 135 (34.2) | 164 (35.9) | 50 (11.9) | |
| Surgical amputation prior to U.S. hospital admission | 183 (14.4) | 183 (46.3) | 0 | 0 | <0.001 |
| Duration of total hospitalization, median days (IQR) | 24 (14–44) | 42 (27–59) | 24 (16–43) | 13 (9–23) | <0.001 |
IQR, interquartile range; LRMC, Landstuhl Regional Medical Center; OR, operating room; SOFA, sequential organ failure assessment.
aOpen soft-tissue wounds includes degloving injuries, lacerations, and penetrating wounds.
bAdmitting unit information is missing for two patients. Percentages and p-value based on total minus missing.
cAdmitting unit information is missing for three patients. Percentages and p-value based on total minus missing.
dMechanical ventilation data are limited to LRMC and first participating U.S. hospital.
A total of 183 (46%) patients with an amputation had an early surgical amputation prior to transfer to a participating U.S. hospital, resulting in a population of 212 (54%) patients with traumatic amputations (Fig. 1). The majority of traumatic amputations involved the lower extremities (196 patients; 92%); however, 39 (18%) patients had upper extremity amputations (patients may have multiple amputations). Overall, patients with amputations had a longer period of hospitalization (median of 42 days) compared to the open fracture and open soft-tissue wound groups (median of 24 and 13 days, respectively).
Infectious Outcomes
Among the 395 patients with amputations (traumatic or early surgical), 207 (52%) were diagnosed with a CEWI. After restricting to the 212 patients with traumatic amputations (183 with early surgical excluded), 100 (47%) had a CEWI diagnosed during their initial hospitalization (no time restriction) with 43% of patients with a lower extremity amputation diagnosed with a CEWI (Table III). In particular, 68% of patients with above knee amputations developed a CEWI. Regarding upper extremity amputations, 28% of patients were also diagnosed with a CEWI. Among the group of patients with >1 amputation, 94 patients also had an open fracture, of whom 20% developed CEWI related to the open fracture. Furthermore, 157 patients with >1 amputation also had an open soft-tissue wound with 10% being diagnosed with a CEWI of the soft-tissue wound.
TABLE III.
Distribution of Infections Among Wounded Military Personnel by Injury Pattern
| Patient Groups by Most Severe Injury | ||||||||
|---|---|---|---|---|---|---|---|---|
| Traumatic Amputations | Open Fractures | Open Soft-tissue Woundsa | Total | |||||
| (N = 212) | (N = 457) | (N = 419) | (N = 1088) | |||||
| N | ≥1 CEWI | N | ≥1 CEWI | N | ≥1 CEWI | N | ≥1 CEWI | |
| N (row %) | N (row %) | N (row %) | N (row %) | |||||
| Amputations | ||||||||
| Overall | 212 | 87 (41.0) | NA | NA | 212 | 87 (41.0) | ||
| Lower extremity | 196 | 84 (42.9) | NA | NA | 196 | 84 (42.9) | ||
| AKA | 93 | 63 (67.7) | NA | NA | 93 | 63 (67.7) | ||
| BKA | 106 | 24 (22.6) | NA | NA | 106 | 24 (22.6) | ||
| Foot | 15 | 0 | NA | NA | 15 | 0 | ||
| Leg, NOS | 1 | 0 | NA | NA | 1 | 0 | ||
| Upper extremity | 39 | 11 (28.2) | NA | NA | 39 | 11 (28.2) | ||
| AEA | 12 | 4 (33.3) | NA | NA | 12 | 4 (33.3) | ||
| Arm, NOS | 3 | 1 (33.3) | NA | NA | 3 | 1 (33.3) | ||
| BEA | 9 | 3 (33.3) | NA | NA | 9 | 3 (33.3) | ||
| Hand | 16 | 3 (18.8) | NA | NA | 16 | 3 (18.8) | ||
| Open fractures | ||||||||
| Overall | 94 | 19 (20.2) | 457 | 59 (12.9) | NA | 551 | 78 (14.2) | |
| Lower extremity | 43 | 5 (11.6) | 313 | 48 (15.3) | NA | 356 | 53 (14.9) | |
| Ankle | 7 | 0 | 35 | 2 (5.7) | NA | 42 | 2 (4.8) | |
| Foot | 7 | 0 | 101 | 11 (10.9) | NA | 108 | 11 (10.2) | |
| Lower leg | 25 | 2 (8.0) | 213 | 25 (11.7) | NA | 238 | 27 (11.3) | |
| Patella | 3 | 1 (33.3) | 18 | 2 (11.1) | NA | 21 | 3 (14.3) | |
| Upper leg | 14 | 2 (14.3) | 71 | 13 (18.3) | NA | 85 | 15 (17.7) | |
| Upper extremity | 61 | 14 (23.0) | 193 | 11 (5.7) | NA | 254 | 25 (9.8) | |
| Hand | 30 | 6 (20.0) | 50 | 0 | NA | 80 | 6 (7.5) | |
| Lower arm | 33 | 8 (24.2) | 100 | 7 (7.0) | NA | 133 | 15 (11.3) | |
| Upper arm | 10 | 1 (10.0) | 83 | 4 (4.8) | NA | 93 | 5 (5.4) | |
| Open soft-tissue wounds | ||||||||
| Overall | 157 | 16 (10.2) | 303 | 11 (3.6) | 419 | 12 (2.9) | 879 | 39 (4.4) |
| Lower extremity | 99 | 13 (13.1) | 244 | 10 (4.1) | 296 | 9 (3.0) | 639 | 32 (5.0) |
| Ankle | 1 | 0 | 7 | 1 (14.3) | 0 | 0 | 8 | 1 (12.5) |
| Foot | 3 | 0 | 16 | 1 (6.3) | 10 | 0 | 29 | 1 (3.5) |
| Leg, NOS | 68 | 2 (2.9) | 181 | 2 (1.1) | 196 | 2 (1.0) | 445 | 6 (1.4) |
| Lower leg | 3 | 0 | 6 | 0 | 8 | 0 | 17 | 0 |
| Patella | 13 | 0 (0.00) | 25 | 2 (8.0) | 24 | 1 (4.2) | 62 | 3 (4.8) |
| Upper leg | 45 | 11 (24.4) | 103 | 4 (3.9) | 163 | 6 (3.7) | 311 | 21 (6.8) |
| Upper extremity | 125 | 3 (2.4) | 159 | 1 (0.6) | 258 | 3 (1.2) | 542 | 7 (1.3) |
| Arm, NOS | 88 | 0 | 120 | 0 | 203 | 2 (1.0) | 411 | 2 (0.5) |
| Elbow | 4 | 0 | 12 | 1 (8.3) | 4 | 0 | 20 | 1 (5.0) |
| Hand | 45 | 1 (2.2) | 28 | 0 | 49 | 0 | 122 | 1 (0.8) |
| Lower arm | 36 | 1 (2.8) | 24 | 0 | 45 | 1 (2.2) | 105 | 2 (1.9) |
| Upper arm | 12 | 0 | 16 | 0 | 40 | 0 | 68 | 0 |
| Wrist | 7 | 1 (14.3) | 4 | 0 | 4 | 0 | 15 | 1 (6.7) |
| Total | 212 | 100 (47.2) | 457 | 66 (14.4) | 419 | 12 (2.9) | 1,088 | 178 (16.4) |
AKA, above knee amputation; AEA, above elbow amputation; BEA, below elbow amputation; BKA, below knee amputation; CEWI, combat-related extremity wound infection; LE, lower extremity; NA, not applicable; NOS, not otherwise specified; UE, upper extremity.
aOpen soft-tissue wounds includes degloving injuries, lacerations, and penetrating wounds.
A CEWI was diagnosed in 66 (14%) patients with an open fracture as their most severe injury (Table III). Approximately 15% of patients with lower extremity fractures developed a CEWI compared to 6% of upper extremity fractures. A total of 303 patients with open fractures as the most severe wound also had open soft-tissue wounds, of whom nearly 4% were diagnosed with a CEWI related to the soft-tissue wound. Among the 419 patients with open soft-tissue wounds as their most severe level, 12 (3%) had a CEWI.
Predictors for CEWIs Based on Initial Injury
Risk factors for the development of CEWIs were assessed among the 669 patients with traumatic amputations and open fractures (Fig. 1). Infections included in the risk factor analysis were restricted to those that were diagnosed within 30 days postinjury (99 and 64 in the amputation and open fracture groups, respectively). In a Cox proportional univariable analysis, having an amputation, blast mechanism of injury (most commonly improvised explosive device [IED]), injury severity score, first documented shock index, volume of blood transfused within 24 hours of injury, use of mechanical ventilation, admission to the intensive care unit, injuries to >4 injury sites, and having both lower and upper extremity injuries were significantly associated with risk of CEWI (data not shown). Sustaining an injury while in a vehicle (versus foot patrol) and having a non-extremity wound infection at least 4 days prior to CEWI diagnosis were associated with a decreased risk of CEWI (data not shown). Having at least one operating room visit (surrogate for wound debridements) prior to infection diagnosis (or hospital discharge) at LRMC or the U.S. hospitals was also associated with a reduced CEWI risk with the protective effect increasing from a hazard ratio of 0.25 (95% confidence interval: 0.16–0.41; p < 0.0001) with 1–2 operating room visits to 0.07 (95% confidence interval: 0.04–0.11; p < 0.0001) with ≥5 operating room visits.
While number of operating room visits was statistically significant in the univariable model, it is a time-varying variable and could not be included into the Cox proportional hazards model as this would violate the key assumption of constant hazard over time. The variable was assessed in a Kaplan–Meier survival plot and a lower number of operating room visits was associated with a reduced time to infection (Fig. 2).
FIGURE 2.
Kaplan–Meier survival plot of time to CEWI diagnosis, stratified by number of OR visits. Plots are censored to time of first CEWI or 30 days postinjury. Log-rank Chi-square: 169.17 (p < 0.0001); Wilcoxon Chi-square: 209.72 (p < 0.0001).
All other variables statistically significant in the univariable analysis were assessed for inclusion in the multivariable model (Table IV). Having a traumatic amputation was independently associated with CEWI risk as was sustaining an IED blast, first document shock index ≥0.8, ≥10 units of red blood cells transfused within 24 hours of injury, and >4 injury sites. Specifically, patients who had ≥1 traumatic amputation were 79% more likely to have a CEWI (Hazard ratio: 1.79; 95% confidence interval 1.25–2.56) compared to patients that had open fractures as their most severe injury. Furthermore, IED blast patients were twice as likely to have a CEWI compared to non-IED blast injuries, while patients with a first documented shock index of ≥0.8 had a 48% increased risk of CEWI, compared to patients in which the first documented shock index was <0.8. For patients that received at least 10 units of blood within the first 24 hours postinjury, the increased risk was more than two times that than for patients who did not receive any blood transfusion. Lastly, patients that had at least 4 injured sites had a 80% increased risk of CEWI compared to patients with 4 or less injured sites.
TABLE IV.
Multivariable Cox Proportional Hazard Model of Risk Factors for the Development of a CEWI
| Characteristics, No. (%) | Total | Patients With CEWIa | Patients Without CEWI | Hazard Ratio (95% Confidence Interval) |
|---|---|---|---|---|
| (N = 669) | (N = 163) | (N = 506) | ||
| Patient classification by most severe injury | ||||
| Amputationsb | 212 (31.7) | 99 (60.7) | 113 (22.3) | 1.79 (1.25-2.56) |
| Open fractures | 457 (68.3) | 64 (39.3) | 393 (77.7) | 1.00 |
| Mechanism sub-type | ||||
| IED | 438 (65.5) | 144 (88.3) | 294 (58.1) | 1.99 (1.19–3.35) |
| Non-IED | 231 (34.5) | 19 (11.7) | 212 (41.9) | 1.00 |
| Had confirmed non-extremity infection at least -4 days before CEWIc | ||||
| Yes | 106 (15.8) | 18 (11.0) | 88 (17.4) | 0.28 (0.17–0.46) |
| No | 563 (84.2) | 145 (89.0) | 418 (82.6) | 1.00 |
| First documented shock index | ||||
| <0.80 | 360 (53.8) | 52 (31.9) | 308 (60.9) | 1.00 |
| ≥0.80 | 309 (46.2) | 111 (68.1) | 198 (39.1) | 1.48 (1.04–2.12) |
| First 24 hour blood transfusion | ||||
| Missing units or none | 248 (37.1) | 21 (12.9) | 227 (44.9) | 1.00 |
| <10 units | 214 (32.0) | 39 (23.9) | 175 (34.6) | 1.51 (0.87–2.61) |
| 10–20 units | 24 (18.5) | 50 (30.7) | 74 (14.6) | 2.86 (1.63–5.03) |
| >20 units | 83 (12.4) | 53 (32.5) | 30 (5.9) | 4.83 (2.66–8.76) |
| Number of extremity injury groups/injury sites | ||||
| ≤4 | 493 (73.7) | 86 (30.7) | 407 (80.4) | 1.00 |
| >4 | 176 (26.3) | 77 (47.2) | 99 (19.6) | 1.80 (1.31–2.47) |
aRestricted to infections diagnosed within 30 days postinjury.
bRestricted to traumatic amputations.
cNon-extremity infections largely included bloodstream (27%) and pneumonia (32%).
Having a confirmed infection other than a CEWI (e.g., bloodstream and pneumonia) at least 4 days prior to CEWI diagnosis was significantly associated with a reduced risk of CEWI (Hazard ratio: 0.28; 95% confidence interval: 0.17–0.46).
DISCUSSION
Whether incurred during combat or civilian activities, extremity trauma may be complicated by infections with potential significant long-term morbidity and healthcare costs. The identification of predictive factors related to the development of CEWIs may support improved outcomes by promoting earlier diagnosis and surgical and medical treatment. Our analysis found that traumatic amputations, IED blast injuries, polytrauma (>4 injury sites), and indicators of injury severity (shock index and blood transfusions) were independently associated with an increased risk of developing a CEWI. Unlike the factors related to injury severity which are not modifiable, a reduced risk of CEWI was associated with the occurrence of a non-extremity infection (i.e., pneumonia or bloodstream infection) prior to CEWI diagnosis. This finding suggests that there are factors related to medical management (e.g., antibiotic therapy) which may reduce the likelihood of a trauma patient developing a CEWI. An improved understanding of antimicrobial exposure (type and duration) around the time of combat-related infectious diagnoses will assist in identification of optimal treatment strategies. While our preliminary risk factor model does have limitations, further assessment is planned to refine the model with increased precision related to antimicrobial therapy, as well as the type and timing of surgical procedures, in order to better understand the impact of medical and surgical management on the risk of CEWIs.
While other studies have examined risk factors for the development of trauma-related infections, they have primarily been conducted on a patient level. In our analysis, we utilized wound classification methods to categorize patients based on their most severe wound experienced, providing more granular wound data. This methodology, along with matching infections at the wound level, allowed our risk factor analysis to go beyond the patient level and assess the impact of different injury patterns for infection risk. Modifiable factors related to the surgical and medical management of extremity trauma are complex. In particular, wound debridements are an integral aspect of wound care and play a key role in prevention of infection. Due to limitations inherent with retrospective analyses, precision of data related to surgical care was lacking. A crude surrogate for debridement was assessed using number of operating room visits. Interestingly, this indicator variable was strongly associated in an inverse direction with CEWI risk in the unadjusted univariable model and reduced time-to-infection in the Kaplan–Meier plot. Operating room visits, a time-varying variable, could not be included in the final Cox proportional hazards model due to violating the constant hazards assumption. Further assessment of wound-specific surgical care is necessary to delineate the relationship with infection risk.
In our preliminary model, we also found that having a non-extremity infection at least 4 days prior to a CEWI diagnosis had a 70% reduced risk of developing a CEWI. This finding suggests that receipt of antibiotics for more than 4 days may be precluding the development of a CEWI. As a result, further analysis is warranted to assess the type of antibiotic prescribed and duration of use for incorporation into a refined CEWI risk factor model. Furthermore, to fully understand the role of antibiotic use for non-extremity infections have on the risk of CEWI development, further analyses must consider other embedded factors potentially affecting the outcome.
Patients with amputations had the highest proportion of infections among the wound groups (47% versus 14% with open fractures and 3% with open soft-tissue wounds), which is likely the result of increased injury severity with this pattern (71% with ISS ≥25 vs 28–30% with other injury patterns). A recent analysis examined trauma registry data collected from combat casualties between 2009 and 2012. Among the personnel identified with at least one amputation, 17% had a wound infection.33 While this proportion is lower than our finding, it was observed that the proportion of wound infections significantly increased in patients with more than one amputation. A lower proportion of lower extremity open fracture infections in patients with amputations compared to lower extremity open fractures in patients in which the most severe injury was an open fracture was also observed and likely results from the management of the amputation on the same limb. In a separate 3-year TIDOS analysis of combat casualties, 53% of patients with a distal amputation and 79% of patients with proximal amputations developed at least one combat trauma-related infection. The occurrence of either a distal or proximal amputation was identified as an infection risk factor.34 However, unlike the analysis herein, these infections were not directly linked to the injury site via AIS. Our findings further support the value of matching infectious outcomes to the site of polytraumatic injuries.
Along with sustaining an amputation, factors related to injury severity were associated with a higher likelihood of developing a CEWI. These findings corroborate prior studies which found that higher injury severity, amputations, extensive soft-tissue injuries, muscle damage, fracture severity, large-volume blood transfusions, and injuries sustained via an IED mechanism were associated with infections.15,29,35
CONCLUSIONS
Due to the difficult nature of evaluating the risk of infection at the wound level among patients with polytrauma, prior studies have been frequently limited to person-level summaries of traumatic wounds. As a result, there was a need to evaluate infection risk while taking into consideration the wound type and location. Using our wound level classification system, we found that the likelihood of a subsequent CEWI increases based on wound severity (amputation versus open fracture). Our use of a newly defined wound classification method to evaluate predictive factors of CEWI corroborates other risk factor analyses and provides support for our wound classification methodology.
Based on these findings, further evaluation of the role of surgical and medical wound management to lessen the likelihood of developing a CEWI is warranted. In particular, timing and type of surgical procedures, antimicrobial regimens, and duration of antimicrobial use will be assessed. The CEWI risk factor model will be examined to determine the best approach to utilize these variables and refine the model. Furthermore, effectiveness of different antibiotic regimens should be assessed with regards to treatment of CEWIs to elucidate risk factor contribution and support best practices. The wound classification system provides a comprehensive approach in assessment of injury and clinical factors for the risk and outcomes of an extremity wound infection.
Acknowledgments
We are indebted to the Infectious Disease Clinical Research Program Trauma Infectious Disease Outcomes Study team of clinical coordinators, microbiology technicians, data managers, clinical site managers, and administrative support personnel for their tireless hours to ensure the success of this project.
Previous Presentation
Presented as an oral talk at the 2017 Military Health System Research Symposium, August 2017, Kissimmee, FL; abstract # MHSRS-17-1582.
Funding
Support for this work (IDCRP-024) was provided by the Infectious Disease Clinical Research Program (IDCRP), a Department of Defense program executed through the Uniformed Services University of the Health Sciences, Department of Preventive Medicine and Biostatistics. This project has been funded by the National Institute of Allergy and Infectious Diseases, National Institutes of Health, under Inter-Agency Agreement Y1-AI-5072, the Department of the Navy under the Wounded, Ill, and Injured Program, and the Defense Medical Research and Development Program. This supplement was sponsored by the Office of the Secretary of Defense for Health Affairs.
References
- 1. Owens BD, Kragh JF Jr, Macaitis J, Svoboda SJ, Wenke JC: Characterization of extremity wounds in Operation Iraqi Freedom and Operation Enduring Freedom. J Orthop Trauma 2007; 21(4): 254–7. [DOI] [PubMed] [Google Scholar]
- 2. Belmont PJ Jr., Goodman GP, Zacchilli M, Posner M, Evans C, Owens BD: Incidence and epidemiology of combat injuries sustained during “the surge” portion of operation Iraqi Freedom by a U.S. Army brigade combat team. J Trauma 2010; 68(1): 204–10. [DOI] [PubMed] [Google Scholar]
- 3. Belmont PJ, Owens BD, Schoenfeld AJ: Musculoskeletal injuries in Iraq and Afghanistan: epidemiology and outcomes following a decade of war. J Am Acad Orthop Surg 2016; 24(6): 341–8. [DOI] [PubMed] [Google Scholar]
- 4. Andersen CR, D’Alleyrand MP, Swiontkowski MF, Ficke JR: Extremity war injuries IVSM: extremity war injuries VIII: sequelae of combat injuries. J Am Acad Orthop Surg 2014; 22(1): 57–62. [DOI] [PubMed] [Google Scholar]
- 5. 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(6 Suppl 5): S520–34. [Google Scholar]
- 6. Krueger CA, Wenke JC, Ficke JR: Ten years at war: comprehensive analysis of amputation trends. J Trauma Acute Care Surg 2012; 73(6 Suppl 5): S438–44. [DOI] [PubMed] [Google Scholar]
- 7. Schoenfeld AJ, Dunn JC, Bader JO, Belmont PJ Jr: 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(2): 287–91. [DOI] [PubMed] [Google Scholar]
- 8. 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(1): 2–7. [PubMed] [Google Scholar]
- 9. Yun HC, Murray CK, Nelson KJ, Bosse MJ: Infection after orthopaedic trauma: prevention and treatment. J Orthop Trauma 2016; 30(Suppl 3): S21–6. [DOI] [PubMed] [Google Scholar]
- 10. Andersen RC, Fleming M, Forsberg JA, et al. : Dismounted complex blast injury. J Surg Orthop Adv 2012; 21(1): 2–7. [PubMed] [Google Scholar]
- 11. Aronson NE, Sanders JW, Moran KA: In harm’s way: infections in deployed American military forces. Clin Infect Dis 2006; 43(8): 1045–51. [DOI] [PubMed] [Google Scholar]
- 12. Penn-Barwell JG, Roberts SA, Midwinter MJ, Bishop JR: Improved survival in UK combat casualties from Iraq and Afghanistan: 2003-2012. J Trauma Acute Care Surg 2015; 78(5): 1014–20. [DOI] [PubMed] [Google Scholar]
- 13. Eastridge BJ, Costanzo G, Jenkins D, et al. : Impact of Joint Theater Trauma System initiatives on battlefield injury outcomes. Am J Surg 2009; 198(6): 852–7. [DOI] [PubMed] [Google Scholar]
- 14. 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(1 Suppl): S33–42. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Tribble DR, Li P, Warkentien TE, et al. : Impact of Operational Theater on combat and noncombat trauma-related infections. Mil Med 2016; 181(10): 1258–68. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Murray CK: Epidemiology of infections associated with combat-related injuries in Iraq and Afghanistan. J Trauma 2008; 64(3 Suppl): S232–8. [DOI] [PubMed] [Google Scholar]
- 17. Murray CK, Hinkle MK, Yun HC: History of infections associated with combat-related injuries. J Trauma 2008; 64(3 Suppl): S221–31. [DOI] [PubMed] [Google Scholar]
- 18. Murray CK, Obremskey WT, Hsu JR, et al. : Prevention of infections associated with combat-related extremity injuries. J Trauma 2011; 71(2 Suppl 2): S235–57. [DOI] [PubMed] [Google Scholar]
- 19. Brown KV, Murray CK, Clasper JC: Infectious complications of combat-related mangled extremity injuries in the British military. J Trauma 2010; 69(Suppl 1): S109–15. [DOI] [PubMed] [Google Scholar]
- 20. Murray CK, Hsu JR, Solomkin JS, et al. : Prevention and management of infections associated with combat-related extremity injuries. J Trauma 2008; 64(3 Suppl): S239–51. [DOI] [PubMed] [Google Scholar]
- 21. Johnson EN, Burns TC, Hayda RA, Hospenthal DR, Murray CK: Infectious complications of open type III tibial fractures among combat casualties. Clin Infect Dis 2007; 45(4): 409–15. [DOI] [PubMed] [Google Scholar]
- 22. Yun HC, Branstetter JG, Murray CK: Osteomyelitis in military personnel wounded in Iraq and Afghanistan. J Trauma 2008; 64(2 Suppl): S163–8. [DOI] [PubMed] [Google Scholar]
- 23. Penn-Barwell JG, Bennett PM, Mortiboy DE, Fries CA, Groom AF, Sargeant ID: Factors influencing infection in 10 years of battlefield open tibia fractures. Strategies Trauma Limb Reconstr 2016; 11(1): 13–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Huh J, Stinner DJ, Burns TC, Hsu JR,. Late Amputation Study Team : Infectious complications and soft tissue injury contribute to late amputation after severe lower extremity trauma. J Trauma 2011; 71(1 Suppl): S47–51. [DOI] [PubMed] [Google Scholar]
- 25. Tintle SM, Baechler MF, Nanos GP, Forsberg JA, Potter BK: Reoperations following combat-related upper-extremity amputations. J Bone Joint Surg Am 2012; 94(16): e1191–6. [DOI] [PubMed] [Google Scholar]
- 26. Tintle SM, Shawen SB, Forsberg JA, et al. : Reoperation after combat-related major lower extremity amputations. J Orthop Trauma 2014; 28(4): 232–7. [DOI] [PubMed] [Google Scholar]
- 27. Harris AM, Althausen PL, Kellam J, Bosse MJ, Castillo R, Lower Extremity Assessment Project Study Group : Complications following limb-threatening lower extremity trauma. J Orthop Trauma 2009; 23(1): 1–6. [DOI] [PubMed] [Google Scholar]
- 28. Zaraca F, Ponzoni A, Stringari C, Ebner JA, Giovannetti R, Ebner H: Lower extremity traumatic vascular injury at a level II trauma center: an analysis of limb loss risk factors and outcomes. Minerva Chir 2011; 66(5): 397–407. [PubMed] [Google Scholar]
- 29. Petersen K, Riddle MS, Danko JR, et al. : Trauma-related infections in battlefield casualties from Iraq. Ann Surg 2007; 245(5): 803–11. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Eastridge BJ, Jenkins D, Flaherty S, Schiller H, Holcomb JB: Trauma system development in a theater of war: Experiences from Operation Iraqi Freedom and Operation Enduring Freedom. J Trauma 2006; 61(6): 1366–72. [DOI] [PubMed] [Google Scholar]
- 31. Lawnick MM, Champion HR, Gennarelli T, et al. : Combat injury coding: A review and reconfiguration. J Trauma Acute Care Surg 2013; 75(4): 573–81. [DOI] [PubMed] [Google Scholar]
- 32. Centers for Disease Control and Prevention : CDC/NHSN Surveillance Definitions for Specific Types of Infections. 2018. Available at http://www.cdc.gov/nhsn/pdfs/pscmanual/17pscnosinfdef_current.pdf; accessed January 22, 2018.
- 33. Godfrey BW, Martin A, Chestovich PJ, Lee GH, Ingalls NK, Saldanha V: Patients with multiple traumatic amputations: An analysis of operation enduring freedom joint theatre trauma registry data. Injury 2017; 48(1): 75–9. [DOI] [PubMed] [Google Scholar]
- 34. Weintrob AC, Murray CK, Xu J, et al. : Early infections complicating the care of combat casualties from Iraq and Afghanistan. Surg Infect (Larchmt) 2018; 19(3): 286–97 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35. 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(1 Suppl): S62–73. [DOI] [PubMed] [Google Scholar]