11% of wounded military personnel with penetrating cranial injuries developed a central nervous system infection. These patients were more critically injured and required more invasive neurosurgical procedures, had more retained fragment, and more cerebrospinal fluid leaks.
KEY WORDS: Penetrating brain injury, cranial infection, meningitis, traumatic brain injury, military
BACKGROUND
Penetrating brain injuries are a potentially lethal injury associated with substantial morbidity and mortality. We examined characteristics and outcomes among military personnel who sustained battlefield-related open and penetrating cranial injuries during military conflicts in Iraq and Afghanistan.
METHODS
Military personnel wounded during deployment (2009–2014) were included if they sustained an open or penetrating cranial injury and were admitted to participating hospitals in the United States. Injury characteristics, treatment course, neurosurgical interventions, antibiotic use, and infection profiles were examined.
RESULTS
The study population included 106 wounded personnel, of whom 12 (11.3%) had an intracranial infection. Posttrauma prophylactic antibiotics were prescribed in more than 98% of patients. Patients who developed central nervous system (CNS) infections were more likely to have undergone a ventriculostomy (p = 0.003), had a ventriculostomy in place for a longer period (17 vs. 11 days; p = 0.007), had more neurosurgical procedures (p < 0.001), and have lower presenting Glasgow Coma Scale (p = 0.01) and higher Sequential Organ Failure Assessment scores (p = 0.018). Time to diagnosis of CNS infection was a median of 12 days postinjury (interquartile range, 7–22 days) with differences in timing by injury severity (critical head injury had median of 6 days, while maximal [currently untreatable] head injury had a median of 13.5 days), presence of other injury profiles in addition to head/face/neck (median, 22 days), and the presence of other infections in addition to CNS infections (median, 13.5 days). The overall length of hospitalization was a median of 50 days, and two patients died.
CONCLUSION
Approximately 11% of wounded military personnel with open and penetrating cranial injuries developed CNS infections. These patients were more critically injured (e.g., lower Glasgow Coma Scale and higher Sequential Organ Failure Assessment scores) and required more invasive neurosurgical procedures.
LEVEL OF EVIDENCE
Prognostic and Epidemiological; Level IV.
Penetrating traumatic brain injury (pTBI) is a devastating and often lethal injury associated with high morbidity and mortality. For penetrating injuries secondary to gunshot wounds in a civilian setting, 75% die on the scene, 15% die en route to the hospital or during initial acute neurosurgical management, and 9.4% survive for eventual discharge to rehabilitation.1 Between 2000 and 2014, 32,996 US military personnel sustained combat-related moderate to severe traumatic brain injuries, of which 78.6% were secondary to blast injuries and 42.9% were penetrating injuries.2 Not only are these injuries devastating to the patients and their families but their aftercare is time intensive, extensive, and expensive.3
One major cause of additional morbidity is the high rate of infections associated with penetrating and open brain injuries. Recent infections rate estimates following pTBI range from 7% in the largest civilian multicenter review to as high as 25% in military populations.4–8 In the military setting, wounds are thought to be at an elevated risk of infection because of the high occurrence of penetrating injuries, contamination, and prolonged transport to the final treatment destination. However, advances in the military health care system and deployed clinical practice guidelines (CPGs) have led to rapid evacuation times and administration of early treatments, including prophylactic antibiotics. It is important to identify injury patterns, complications, or treatment courses that may contribute to or are more often seen with central nervous system (CNS) infections to help better define optimal care for patients injured in future conflicts. We describe the epidemiology of battlefield-related open and penetrating cranial injuries and related CNS infections sustained in military conflict in Iraq and Afghanistan.
PATIENTS AND METHODS
Study Design
Criteria for inclusion in this retrospective observational study was being an active-duty service member or Department of Defense (DoD) beneficiary 18 years or older who sustained an open skull fracture or penetrating CNS injury while deployed (June 1, 2009, through December 31, 2014) and required medical evacuation to a regional medical center in Germany before transition to a participating military hospital in the United States. All patients who met the inclusion criteria were included in the analysis. Patients were excluded if they sustained only a closed head injury, were not transferred to a participating US hospital, or if records of their hospitalization were not available for review. This study (protocol 351767) was approved by the institutional review board. This study followed the Strengthening the Reporting of Observational Studies in Epidemiology reporting guidelines (Supplemental Digital Content, Supplementary Data 1, http://links.lww.com/TA/D22).
Patient demographics, injury characteristics, trauma history, and early trauma care data were obtained from the DoD Trauma Registry.9 Infection-related data (e.g., syndromes, antimicrobial use, and microbiology) were collected through the DoD Trauma Registry Infectious Disease module.10 Supplemental data abstraction from medical records was used to collect comprehensive data related to open skull fractures and penetrating CNS injury characteristics, treatment, and outcomes, such as occurrence of retained fragments, cerebrospinal fluid (CSF) leak, timing of neurosurgical interventions, and use/duration of intracranial monitors (see Supplemental Digital Content, Supplementary Data 2, http://links.lww.com/TA/D23, for definitions of variables utilized in the study). In addition, time between injury and CNS infection (days) was also investigated, with the date of CNS infection measured as either the (1) date of collection of a positive culture from a CNS source or in the absence of a positive culture or (2) physician-recorded date of suspected infection. Additional outcomes analyzed were mortality, development of non-CNS infections, and total inpatient hospital days.
Infections were defined based on clinical findings and laboratory test results and classified in accordance with standardized definitions from the Centers for Disease Control and Prevention, National Healthcare Safety Network.11 In addition, patients were deemed to have a CNS infection if there was a positive CSF culture or clinical suspicion of a physician with directed antimicrobial therapy (≥5 days).
Statistical Analysis
The primary outcome of interest was development of a CNS infection (yes/no). Differences in patient demographics, injury characteristics, and treatment details between patients who developed a CNS infection and those who did not were investigated using χ2 or Fisher's exact tests (for categorical explanatory variables) or Wilcoxon rank-sum tests (for continuous explanatory variables). Patients with missing data were excluded from comparisons of those specific variables. All statistical analyses were performed using SAS 9.4 (SAS, Cary, NC), and a significance threshold of p < 0.05 was used for all comparisons.
RESULTS
Study Population
A total of 6,087 wounded military personnel were admitted to a regional medical center in Germany during the study period with 2,699 patients subsequently transferred to a participating US military hospital. Of the 235 patients suspected to have either open skull fractures or penetrating CNS injuries, 106 were confirmed after review of the electronic medical record with 93 patients (88%) admitted to a hospital in the northeastern region and 13 patients (12%) admitted to a hospital in the southwestern region (Supplemental Digital Content, Supplementary Data 3, http://links.lww.com/TA/D24). The patients included in the analysis were young (median age, 25 years) males primarily injured in Afghanistan (95%) via a blast mechanism (70%; Table 1).
TABLE 1.
Characteristics Among Patients With pTBI or Open Skull Fracture, Stratified by Infection Status
| Characteristic, n (%) | Total Patients (N = 106) | Patients With CNS Infection (n = 12) | Patients Without CNS Infection (n = 94) | p |
|---|---|---|---|---|
| Age, median (IQR), y | 25 (22–29) | 24 (22–27) | 25 (22–29) | 0.64 |
| Male | 106 (100) | 12 (100) | 94 (100) | — |
| Branch of service | 0.43 | |||
| Air Force | 3 (2.8) | 1 (8.3) | 2 (2.1) | |
| Army | 70 (66.0) | 7 (58.3) | 63 (67.0) | |
| Marine | 26 (24.5) | 3 (25.0) | 23 (24.5) | |
| Navy | 3 (2.8) | 0 (0) | 3 (3.2) | |
| Other | 4 (3.8) | 1 (8.3) | 3 (3.2) | |
| Injured in: | 0.99 | |||
| Afghanistan | 101 (95.3) | 12 (100) | 89 (94.7) | |
| Iraq | 4 (3.8) | 0 (0) | 4 (4.3) | |
| Nontheater | 1 (0.9) | 0 (0) | 1 (1.1) | |
| Injury mechanism | 0.33 | |||
| Blast IED | 45 (42.5) | 3 (25.0) | 42 (44.7) | |
| Blast non-IED | 29 (27.4) | 3 (25.0) | 26 (27.7) | |
| Gunshot wound | 27 (25.5) | 5 (41.7) | 22 (23.4) | |
| Motor vehicle crash | 1 (0.9) | 0 (0) | 1 (1.1) | |
| Other | 4 (3.8) | 1 (8.3) | 3 (3.2) | |
| Injury Severity Score | 0.99 | |||
| 1–9 (Minor) | 0 (0) | 0 (0) | 0 (0) | |
| 10–15 (Moderate) | 0 (0) | 0 (0) | 0 (0) | |
| 16–25 (Severe) | 5 (4.7) | 0 (0) | 5 (5.3) | |
| ≥26 (Critical) | 101 (95.3) | 12 (100) | 89 (94.7) | |
| First documented shock index | 0.37 | |||
| <0.65 | 42 (39.6) | 5 (41.7) | 37 (39.4) | |
| 0.65 to <0.80 | 26 (24.5) | 1 (8.3) | 25 (26.6) | |
| >0.80 | 38 (35.9) | 6 (50.0) | 32 (34.0) | |
| Injury profile | ||||
| HFN | 106 (100) | 12 (100) | 94 (100) | — |
| Abdomen | 10 (9.4) | 1 (8.3) | 9 (9.6) | 0.99 |
| Pelvic/genitourinary | 4 (3.8) | 2 (16.7) | 2 (2.1) | 0.06 |
| Upper extremity | 22 (21.7) | 3 (25.0) | 19 (20.2) | 0.71 |
| Lower extremity | 26 (24.5) | 3 (25.0) | 23 (24.5) | 0.73 |
| Thorax | 26 (24.5) | 5 (41.7) | 21 (22.3) | 0.16 |
| Spine | 9 (8.5) | 0 (0) | 9 (9.6) | 0.59 |
| CNS injury severity* | 0.20 | |||
| Max HFN AIS = 4 | 6 (5.7) | 0 (0) | 6 (6.4) | |
| Max HFN AIS = 5 | 39 (36.8) | 2 (16.7) | 37 (39.4) | |
| Max HFN AIS = 6 | 61 (57.6) | 10 (83.3) | 51 (54.3) | |
| Isolated head injury (i.e., only HFN injury profile) | 0.57 | |||
| Yes | 52 (49.1) | 5 (41.7) | 47 (50.0) | |
| No | 54 (50.9) | 7 (58.3) | 47 (50.0) | |
| >Total no. non-HFN profiles | 0.24 | |||
| 0 | 52 (49.1) | 5 (41.7) | 47 (50.0) | |
| 1 | 29 (27.4) | 4 (33.3) | 25 (26.6) | |
| 2 | 13 (12.3) | 1 (8.3) | 12 (12.8) | |
| 3 | 6 (5.7) | 0 (0) | 6 (6.4) | |
| 4 | 6 (5.7) | 2 (16.7) | 4 (4.3) | |
*Abbreviated Injury Scale severity codes: 2, moderate; 3, serious; 4, severe; 5, critical; and 6, maximal (currently untreatable).
AIS, Abbreviated Injury Scale; HFN, head/face/neck; IQR, interquartile range.
CNS Infections
A total of 12 patients (11.3%) who had a CNS infection during hospitalization were identified. Blast and gunshot wound were the primary injury mechanisms among the CNS infection patients (50% and 41.7%, respectively). There was no significant difference in injury mechanism or severity between the patients with and without CNS infections (Table 1). Patients largely had a CNS infection attributed to coagulase-negative staphylococci (42%) with the majority being meningitis/ventriculitis (83%; Table 2).
TABLE 2.
Microbiology and Infection Location of the CNS Infections
| Microbiology and Infection Location, n (%)* | Patients With CNS Infections (n = 12) |
|---|---|
| Organisms present on culture | |
| Coagulase-negative Staphylococci | 5 (41.7) |
| Escherichia coli | 2 (16.7) |
| Candida albicans | 2 (16.7) |
| Klebsiella aerogenes | 1 (8.3) |
| Enterococcus faecium | 1 (8.3) |
| Enterococcus faecalis | 1 (8.3) |
| Streptococcus spp. | 1 (8.3) |
| CNS infection location | |
| Brain abscess | 3 (25.0) |
| Subdural infection | 0 (0) |
| Encephalitis | 2 (16.7) |
| Epidural infection | 3 (25.0) |
| Meningitis/ventriculitis | 10 (83.3) |
| Spinal abscess | 0 (0) |
*Patients may have more than one infection, so the data presented sum to more than the total number of patients.
A significantly greater proportion of patients with a CNS infection had retained fragments in their penetrating CNS injury compared with those without CNS infection (100% vs. 70%; p = 0.03). Nevertheless, among the 78 patients with a retained fragment, there was not a significant difference with regard to the specific fragment types (i.e., bone, shrapnel/bullet, or both) between the patients with and without CNS infections (p = 0.91; Table 3). Patients with a CNS infection had a significantly decreased presenting Glasgow Coma Scale (GCS) (median, 8 vs. 14; p = 0.01). More patients with a CNS infection also had a CSF leak (42% vs. 17% among patients without a CNS infection); however, it was not statistically significant (p = 0.06).
TABLE 3.
Injury Characteristics Among Patients With pTBI or Open Skull Fracture, Stratified by Infection Status
| Characteristic, n (%) | Total Patients (N = 106) | Patients With CNS Infection (n = 12) | Patients Without CNS Infection (n = 94) | p |
|---|---|---|---|---|
| Retained fragment | 0.03 | |||
| Yes | 78 (73.6) | 12 (100) | 66 (70.2) | |
| No | 28 (26.4) | 0 (0) | 28 (29.8) | |
| Fragment type* | 0.91 | |||
| Bone | 27 (25.5) | 5 (41.7) | 22 (33.9) | |
| Shrapnel/bullet | 10 (9.4) | 1 (8.3) | 9 (13.9) | |
| Both | 40 (37.7) | 6 (50.0) | 34 (52.3) | |
| Missing | 1 | 0 | 1 | |
| Presenting GCS, median (IQR) | 13.0 (7–15) | 8 (7–11) | 14 (8–15) | 0.01 |
| Missing | 19 | 3 | 16 | |
| CSF leak | 0.06 | |||
| Yes | 21 (19.8) | 5 (41.7) | 16 (17.0) | |
| No | 85 (80.2) | 7 (58.3) | 78 (83.0) | |
| Foreign object | 0.28 | |||
| Yes | 55 (51.9) | 8 (66.7) | 47 (50.0) | |
| No | 51 (48.1) | 4 (33.3) | 47 (50.0) | |
| Injury type | 0.37 | |||
| Missile | 40 (37.7) | 6 (50.0) | 34 (36.2) | |
| Blast | 60 (56.6) | 5 (41.7) | 55 (58.5) | |
| Other | 6 (5.7) | 1 (8.3) | 5 (5.3) |
*Differences in the type of retained fragment in patients with CNS infection versus those without were examined among the subgroup of 78 patients who had a retained fragment.
IQR, interquartile range.
Regarding neurosurgical procedures, patients who developed a CNS infection were more likely to have undergone ventriculostomy (91.7% vs. 45.7% among non-CNS infection patients; p = 0.003), had an external ventricular drain (EVD) in place for longer (median, 17 vs. 11 days; p = 0.007), and underwent a higher number of neurosurgical procedures (33% with ≥3 procedures vs. 2.1%; p < 0.001) (Table 4). Although not statistically significant, there was also an increase in placement of an intracranial pressure (ICP) monitor (33.3% vs. 19.2%) and lumbar drain placement (8.3% vs. 3.2%) in those who developed a CNS infection (Table 4).
TABLE 4.
Neurosurgical Procedure and Hospitalization Characteristics Among Patients With pTBI or Open Skull Fracture, Stratified by Infection Status
| Characteristic, n (%) | Total Patients (N = 106) | Patients With CNS Infection (n = 12) | Patients Without CNS Infection (n = 94) | p |
|---|---|---|---|---|
| Hardware inserted | 58 (54.7) | 9 (75.0) | 49 (52.1) | 0.13 |
| No. hardware insertion procedures per patient | 0.31 | |||
| 0 | 48 (45.3) | 3 (25.0) | 45 (47.9) | |
| 1 | 57 (53.8) | 9 (75.0) | 48 (51.1) | |
| 2 | 1 (0.9) | 0 (0) | 1 (1.1) | |
| Underwent ventriculostomy | 54 (50.9) | 11 (91.7) | 43 (45.7) | 0.003 |
| No. ventriculostomies per patient | <0.001 | |||
| 0 | 52 (59.1) | 1 (8.3) | 51 (54.3) | |
| 1 | 46 (43.4) | 7 (58.3) | 39 (41.5) | |
| 2 | 7 (6.6) | 3 (25.0) | 4 (4.3) | |
| 3 | 1 (0.9) | 1 (8.3) | 0 (0) | |
| Ventricular drain in place, median (IQR),* d | 12 (7–16) | 17 (10–21) | 11 (7–14) | 0.007 |
| Shunt inserted | 1 (0.9) | 1 (8.3) | 0 (0) | 0.11 |
| ICP monitor inserted | 22 (20.8) | 4 (33.3) | 18 (19.2) | 0.27 |
| No. ICP procedures per patient | 0.34 | |||
| 0 | 84 (79.3) | 8 (66.7) | 76 (80.9) | |
| 1 | 21 (19.8) | 4 (33.3) | 17 (18.1) | |
| 2 | 1 (0.9) | 0 (0) | 1 (1.1) | |
| ICP in place, median (IQR),** d | 4 (3–6) | 5 (4–12) | 4 (3–6) | 0.20 |
| Lumbar drain inserted | 4 (3.8) | 1 (8.3) | 3 (3.2) | 0.39 |
| Lumbar drain in place, median (IQR), d | 7 (5–8) | 7 (7–7) | 6 (3–8) | — |
| No. any neurosurgical procedures per patient | <0.001 | |||
| 0 | 23 (21.7) | 0 (0) | 23 (24.5) | |
| 1 | 61 (57.6) | 3 (25.0) | 58 (61.7) | |
| 2 | 16 (15.1) | 5 (41.7) | 11 (11.7) | |
| 3 | 4 (3.8) | 2 (16.7) | 2 (2.1) | |
| 4 | 1 (0.9) | 1 (8.3) | 0 (0) | |
| 5 | 1 (0.9) | 1 (8.3) | 0 (0) | |
| Type of neurosurgeries administered | 0.002 | |||
| Craniectomy only | 45 (42.9) | 4 (33.3) | 41 (43.6) | |
| Craniotomy only | 25 (23.8) | 2 (16.7) | 23 (24.5) | |
| Other only† | 1 (0.9) | 0 (0) | 1 (1.1) | |
| Craniectomy and craniotomy | 6 (5.7) | 3 (25.0) | 3 (3.2) | |
| Craniectomy and other† | 5 (4.8) | 3 (25.0) | 2 (2.1) | |
| Craniotomy and other† | 0 (0) | 0 (0) | 0 (0) | |
| Craniectomy, craniotomy, and other† | 1 (1.0) | 0 (0) | 1 (1.1) | |
| None | 23 (21.9) | 0 (0) | 23 (24.5) | |
| ICU admission | 0.99 | |||
| None | 4 (3.8) | 0 (0) | 4 (4.3) | |
| Germany hospital ICU only | 20 (18.9) | 2 (16.7) | 18 (19.2) | |
| US hospital ICU ± Germany hospital | 82 (77.4) | 10 (83.3) | 72 (76.6) | |
| Mechanical ventilation | 0.002 | |||
| None | 32 (30.2) | 0 (0) | 32 (34.0) | |
| Germany hospital ± US hospital ≤1 wk | 58 (54.7) | 12 (100) | 46 (48.9) | |
| Germany hospital only | 16 (15.1) | 0 (0) | 16 (17.0) | |
| OR visits ≤2 wk of injury,‡ median (IQR) | 2 (0–3) | 3 (2–5) | 1 (0–2) | 0.040 |
| Total hospitalization, median (IQR), d | 26.5 (18.0–39.0) | 50.5 (36.5–55.0) | 24.5 (17.0–35.0) | <0.001 |
| Germany hospital admission SOFA, median (IQR) | 5.0 (2.0–8.0) | 8.0 (5.5–9.0) | 5.0 (1.0–8.0) | 0.018 |
| US hospital admission SOFA, median (IQR) | 5.0 (1.5–7.0) | 5.5 (4.0–7.0) | 5.0 (1.0–7.0) | 0.27 |
*Duration missing for three patients without a CNS infection who had a ventricular drain placed.
**Duration missing for one patient with a CNS infection and three patients without a CNS infection who had ICP monitoring.
†Other includes subgaleal collection washout, middle fossa repair (autograft), repair of CSF leak, lumbar drain placement, anterior fossa recon, subdural/epidural evacuation, wound exploration and duraplasty, flap coverage of open cranial defect, and irrigation and debridement over complex open skull fracture.
‡Total OR visits for any reason within 2 weeks of injury, not specific to CNS trauma.
ICU, intensive care unit; IQR, interquartile range; OR, operating room; SOFA, Sequential Organ Failure Assessment.
Early antibiotics (prescribed within 48 hours posttrauma) were received by 98.1% of patients with no significant differences between patients with and without CNS infections (Table 5). First-generation cephalosporins, which are recommended as first-line posttrauma antibiotic prophylaxis per the Joint Trauma System (JTS) CPG12,13 accounted for the majority (received by 87% of patients). Among the patients with a CNS infection, antibiotics most commonly received at any time before the CNS infection diagnosis, excluding first-generation cephalosporins (mentioned previously) and doxycycline (used for antimalarial prophylaxis), were vancomycin (75%), fluoroquinolone (67%), and carbapenem (58%; Supplemental Digital Content, Supplementary Data 4, http://links.lww.com/TA/D25). In the 2 weeks after CNS infection diagnosis, antibiotics received were primarily vancomycin (67%) and carbapenem (42%).
TABLE 5.
Early Antibiotic Utilization Among Patients With pTBI or Open Skull Fracture (Within 48 Hours Postinjury), Stratified by Infection Status
| Characteristic, n (%) | Total Patients (N = 106) | Patients With CNS Infection (n = 12) | Patients Without CNS Infection (n = 94) | p |
|---|---|---|---|---|
| Received antibiotics 48 h postinjury | 104 (98.1) | 12 (100) | 92 (97.9) | 0.99 |
| Antibiotic administered* | ||||
| JTS CPG recommended posttrauma prophylactic antibiotics** | ||||
| Cephalosporin, first generation | 92 (86.8) | 11 (91.7) | 81 (86.2) | 0.99 |
| Clindamycin | 19 (17.9) | 1 (8.3) | 18 (19.2) | 0.69 |
| Aminopenicillin | 16 (15.1) | 2 (16.7) | 14 (14.9) | 0.99 |
| Metronidazole | 7 (6.6) | 1 (8.3) | 6 (6.4) | 0.58 |
| Other antimicrobials prescribed | ||||
| Fluoroquinolone | 31 (29.3) | 5 (41.7) | 26 (27.7) | 0.33 |
| Cephalosporin, third generation | 10 (9.4) | 2 (16.7) | 8 (8.5) | 0.32 |
| Antipseudomonal penicillin | 8 (7.6) | 2 (16.7) | 6 (6.4) | 0.22 |
| Vancomycin | 6 (5.7) | 1 (8.3) | 5 (5.3) | 0.52 |
| Aminoglycoside | 3 (2.8) | 0 (0) | 3 (3.2) | 0.99 |
| Carbapenem | 2 (1.9) | 0 (0) | 2 (2.1) | 0.99 |
| Macrolide | 1 (0.9) | 0 (0) | 1 (1.1) | 0.99 |
| Antimalarial prophylaxis | ||||
| Doxycycline | 22 (20.8) | 3 (25.0) | 19 (20.2) | 0.71 |
| Total unique antibiotics administered per patient | 0.55 | |||
| 0 | 2 (1.9) | 0 (0) | 2 (2.1) | |
| 1 | 30 (28.3) | 2 (16.7) | 28 (29.8) | |
| 2 | 43 (40.6) | 6 (50.0) | 37 (39.4) | |
| 3 | 23 (21.7) | 2 (16.7) | 21 (22.3) | |
| 4 | 8 (7.5) | 2 (16.7) | 6 (6.4) | |
Patients with a CNS infection were more likely to require mechanical ventilation at both a hospital in Germany and at the US military hospitals (p = 0.002), had a greater number of operating room visits within the first 2 weeks (all procedures, including noncranial) (median, 3 vs. 1; p = 0.04) and longer total hospitalization (median, 50.5 days vs. 24.5 days; p < 0.001), and had higher assessment of acute morbidity of critical illness at a hospital in Germany (median Sequential Organ Failure Assessment score, 8 vs. 5; p = 0.018; Table 4). Although not statistically significant, there was a trend toward an increased time from injury to CNS infection diagnosis among patients who also had non-CNS infections (13.5 days vs. 9 days for those with only CNS infections), those with more severe CNS injuries (13.5 days vs. 6 for those with less severe CNS injuries), and those with additional injuries other than head/face/neck (e.g., extremities) (22 days vs. 7 days for those with only head/face/neck injuries; Table 6). There was no significant difference in mortality between the patients with and without CNS infections (16.7% vs. 6.4%, respectively; p = 0.22).
TABLE 6.
Timing of Infection Diagnosis, Length of Hospitalization, and Mortality Among Patients With a CNS Infection
| Characteristic | n | Time to CNS Infection Diagnosis, Median (IQR), d | Inpatient Hospital Stay, Median (IQR), d | Mortality, n (%) |
|---|---|---|---|---|
| Patients with CNS Infection | 12 | 12 (7–22) | 50 (37–55) | 2 (16.7) |
| US hospital | ||||
| Northeastern Region | 9 | 13 (7–22) | 54 (44–56) | 1 (11.1) |
| Southwestern Region | 3 | 11 (7–23) | 37 (15–52) | 1 (33.3) |
| Non-CNS infections present* | ||||
| Yes | 10 | 13.5 (7–22) | 53 (44–56) | 2 (20.0) |
| No | 2 | 9 (7–11) | 31 (25–37) | 0 (0) |
| Non-HFN injury profiles | ||||
| Yes | 7 | 22 (8–23) | 54 (36–56) | 2 (28.6) |
| No | 5 | 7 (7–11) | 44 (37–49) | 0 (0) |
| CNS injury severity** | ||||
| Max head AIS = 6 | 10 | 13.5 (8–22) | 53 (37–56) | 2 (20.0) |
| Max head AIS = 5 | 2 | 6 (5–7) | 35 (25–44) | 0 (0) |
*Non-CNS infections largely included pneumonia, bloodstream infections, and skin and soft-tissue infections.
**Abbreviated Injury Scale severity codes: 5 indicates critical and 6, maximal (currently untreatable).
AIS, Abbreviated Injury Scale; HFN, head/face/neck.
DISCUSSION
Of the wounded military personnel with penetrating brain injuries in our study, 11.3% went on to develop a CNS infection during their initial hospitalization. These patients tended to have severe injuries with high head Abbreviated Injury Scale and low presenting GCS and required ICP monitoring for longer periods. In addition, the patients with a CNS infection required extensive care throughout their hospitalizations, undergoing more neurosurgical and nonneurosurgical procedures, remaining in the hospital for longer, and having a greater requirement of mechanical ventilation compared with patients with penetrating brain injuries who did not develop CNS infections. In general, these were severely injured and complicated patients with an overall mortality for those with a CNS infection of 16.7%; mortality of those without a CNS infection was 6.4%. Injury severity, treatment course characteristics, infection rates, and eventual mortality are greatly influenced by the injury mechanisms seen in a combat setting and the advanced evacuation system in place to bring these patients to definitive care.
In a prior 8-year analysis of 137 wounded military personnel who sustained a pTBI in Iraq, 24.8% developed meningitis/ventriculitis.5 Although Weisbrod and colleagues5 did not compare characteristics of patients who did and did not develop meningitis/ventriculitis, there are differences between their study population and the patients included in our study, which may explain our lower proportion of infections, particularly with regard to the frequency of neurosurgical monitoring and interventions. While use of EVD was similar (50.9% in our study and 60.6% in Weisbrod et al.5), the use of ICP monitoring was lower in our analysis (21%) compared with the Weisbrod study (37.2%). A notable difference is that 21.7% of patients in our study did not require surgical invention compared with 13% in the Weisbrod paper. In particular, 78.8% of patients in the Weisbrod population underwent a craniectomy compared with 54% in our study (craniectomy only or in combination with other procedures). Weisbrod and colleagues5 postulated that the high use of neurosurgical intervention may have been a contributing factor to the increased rate of infections observed in their study, which is consistent with our finding of more neurosurgical interventions among patients with CNS infections.
Penetrating brain injuries in military conflicts differ from those seen in a civilian population where injuries tend to be lower velocity missiles compared with high-velocity missiles and improvised explosive devices (IEDs) seen on the battlefield.14 Higher velocity missiles and secondary projectiles transfer more energy to the surrounding tissue, have increased cavitation effect, and cause increased bone fragmentation. Specifically, IEDs have been the cause of most combat casualties in recent military conflicts.15,16 This remains true for our study population where 70% sustained a blast injury. Blast trauma resulting from IED detonation is pathophysiologically different from other penetrating trauma. These devices create additional complex injury patterns, which can consist of primary (the blast pressure wave), secondary (projectile objects), tertiary (physical displacement of the body), and quaternary injuries (related to other explosive effects). The transmission of blast waves and projectiles entering the skull led to the implementation of large decompressive hemicraniectomies, which allowed for subsequent swelling of the neural tissue and mitigation of the secondary injury from ischemia secondary to elevated ICPs, permitting transport to more definitive care.17 This leads to multiple required surgical procedures, as the patient will eventually require cranioplasty. In our study, 75.5% of patients underwent at least one neurosurgical procedure, and 50.9% had implantation of an EVD.
Among the 106 individuals in our study population who had an open skull fracture or penetrating cranial injury, 11.3% were diagnosed with a CNS infection during their initial hospitalization. These patients were characterized by decreased presenting GCS, retained fragments, placement of an EVD, increased days with a ventriculostomy in place, and requiring more neurosurgical procedures. Other factors that tended to be seen in patients with a CNS infection included presence of a CSF leak, any intracranial monitoring device, and a lumbar drain. In prior military conflicts, similar characteristics (e.g., GCS score and retained bone fragments) as well as mode of injury, number of lobes involved, and place of surgical intervention were significantly related to the development of a CNS infection.18 Factors in civilian pTBI that are related to increased risk of infection include surgical interventions and ICP monitoring.7
The current military JTS CPG have acknowledged many of these factors and made recommendations to neurosurgeons in the prehospital setting that can help to mitigate the risk of infection.12 Untreated CSF fistulas have long been recognized as a high risk of infection, dating back to the Korean war.19 Dural closure/expansileduraplasty, scalp closure, and use of grafts to facilitate closure if primary closure is not feasible are encouraged. Although retained fragments are related to an increased infectious risk, aggressive pursuit of these fragments is not advised and can worsen neurological outcomes.20 Instead, the JTS CPG advocates for washout of frank contamination and removal of devitalized brain tissue.12
There is variability in the literature and recommendations regarding use of prophylactic antibiotics. The current JTS CPG recommends antibiotic prophylaxis for pTBI with cefazolin (first line; clindamycin as an alternative) and suggests considering the addition of metronidazole if the injury appears contaminated with soil for anaerobic coverage.12,13 Increased rates of multidrug-resistant organisms were recovered following treatment of pTBI with broad-spectrum antibiotics in Iraq and Afghanistan.21 In our study, 98.1% of pTBI patients received antibiotics within 48 hours postinjury (Table 5). Posttrauma prophylactic antibiotics were near universally received with 87% of patients receiving first-generation cephalosporins. Overall, no differences were seen between the specific classes or number of early antibiotics (recommended prophylactic antibiotics and other antimicrobials) administered between the patients with and without CNS infections, indicating the importance of other factors (e.g., injury severity, critical illness, and invasive neurosurgical interventions) in the development of CNS infections. To the best of our knowledge, there has not been a randomized controlled trial evaluating efficacy of posttrauma antimicrobial prophylaxis in preventing infections; however, civilian observational data have showed that prophylactic antibiotics did not change the overall rate of CNS infections, but intracranial monitors and surgical procedures were statistically significant.7,8 In particular, a recent study analyzing prophylactic antibiotic use in pTBI in civilian trauma reported that 25% of patients developed an infection with 33.9% among those receiving antibiotics and 19.8% among those who did not receive antibiotics.6 Because use of prophylactic antibiotics was widespread in our study population, in accordance with the JTS CPG, no conclusions can be drawn regarding their effectiveness. Aside from doxycycline for malaria prophylaxis, the most commonly prescribed antibiotic regardless of infection status was cephalosporins, which is similar to practice patterns seen in civilian literature.8 After the CNS infection diagnosis, aside from doxycycline, the most commonly received antibiotics were vancomycin and carbapenem (Supplemental Digital Content, Supplementary Data 4, http://links.lww.com/TA/D25).
Patients in our study who had a CNS infection were more likely to need mechanical ventilation at both before and after admission to the US hospitals, had higher acute morbidity of critical illness scores, and had a longer hospitalization. This may be due to the complications caused by the infection or may point to a patient with an overall more critical clinical picture that would be more susceptible to infection. Other studies have similarly shown longer hospitalizations for those with an intracranial infection.6 However, the hospital length of stay for patients in our study was overall much longer regardless of infection status (median, 26.5 days), which may be a reflection of the severity of CNS and non-CNS injuries. The presence of non-CNS infections or severe injuries in other body areas led to an increased length of time on diagnosing a CNS infection. This could be due to a delay in diagnosis (masking from other infections or presence of multiple injured organ systems with probable infectious culprits) or a delay in presentation, or could again indicate critically ill patients being susceptible to CNS infections.
This analysis has limitations inherent with retrospective analyses. Medical records of 31 patients were no longer available (paper charts not available for review and electronic medical record no longer used secondary to closing of the hospital). and injury characteristics and infection status could not be verified. These patients could not ultimately be included in this analysis. Patients of interest were identified using injury and procedure codes. Although these were designed to be more inclusive, some patients with penetrating injuries could have been incorrectly coded and missed for inclusion in this study.
CONCLUSION
The study identified characteristics of patients who developed CNS infection after penetrating or open cranial injuries. The majority of patients received posttrauma prophylactic antibiotics within 48 hours postinjury and after infection diagnosis; patients largely were prescribed vancomycin and carbapenem. Patients who developed CNS infections were critically ill (i.e., elevated Sequential Organ Failure Assessment score, more days on the ventilator, longer hospitalizations, lower presenting GCS, and higher head Abbreviated Injury Scale), underwent more neurosurgical procedures (i.e., ventriculostomy placement and surgical interventions), and had retained fragments and more CSF leaks. These findings are similar to those seen from previous military conflicts and civilian populations and can be used to help inform care of personnel with these injuries in future conflicts with more studies needed to determine the need for prophylactic antibiotics.
Supplementary Material
AUTHORSHIP
M.R.M., J.H.B., L.S., D.R.T., and B.A.D. contributed in the study design. M.R.M., J.H.B., J.M.Y., and B.A.D. contributed in the data collection. E.S. and F.S. contributed in the data analysis. All authors contributed in the data interpretation. M.R.M. and B.A.D. contributed in the manuscript writing. All authors contributed in the critical revision and approval of manuscript.
ACKNOWLEDGMENTS
We thank 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.
Support for this work (IDCRP-024) 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 through a cooperative agreement with The Henry M. Jackson Foundation for the Advancement of Military Medicine, Inc. 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 Defense Health Program, US DoD, under award HU0001190002; and the Department of the Navy under the Wounded, Ill, and Injured Program (HU0001-10-1-0014).
DISCLOSURE
The authors declare no conflicts of interest.
The views expressed are those of the authors and do not 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 Institute of Health or the Department of Health and Human Services; Brooke Army Medical Center; Walter Reed National Military Medical Center; the US Army Medical Department; the US Army Office of the Surgeon General; the DoD; Defense Health Agency; the Departments of the Army, Navy or Air Force; or the US Government. Mention of trade names, commercial products, or organizations does not imply endorsement by the US Government.
Footnotes
Supplemental digital content is available for this article. Direct URL citations appear in the printed text, and links to the digital files are provided in the HTML text of this article on the journal’s Web site (www.jtrauma.com).
Contributor Information
Melissa R. Meister, Email: melissarmeister@gmail.com.
Jason H. Boulter, Email: jhboulter@gmail.com.
Joseph M. Yabes, Email: joseph.m.yabes2.mil@health.mil.
Erica Sercy, Email: esercy@idcrp.org.
Faraz Shaikh, Email: binqasim@gmail.com.
Hana Yokoi, Email: hanayokoi95@gmail.com.
Laveta Stewart, Email: lstewart@idcrp.org.
Michaela M. Scanlon, Email: michaela.scanlon@usuhs.edu.
Margaret M. Shields, Email: margaret.shields@usuhs.edu.
Alexander Kim, Email: ajk186@georgetown.edu.
David R. Tribble, Email: dtribble@idcrp.org.
Viktor Bartanusz, Email: viktor.bartanusz.ctr@usuhs.edu.
Bradley A. Dengler, Email: bdengler86@gmail.com.
REFERENCES
- 1.Aarabi B Tofighi B Kufera JA Hadley J Ahn ES Cooper C, et al. Predictors of outcome in civilian gunshot wounds to the head. J Neurosurg. 2014;120(5):1138–1146. [DOI] [PubMed] [Google Scholar]
- 2.Fang R, Markandaya M, DuBose JJ, Cancio LC, Shackelford S, Blackbourne LH. Early in-theater management of combat-related traumatic brain injury: a prospective, observational study to identify opportunities for performance improvement. J Trauma Acute Care Surg. 2015;79(4 Suppl 2):S181–S187. [DOI] [PubMed] [Google Scholar]
- 3.van Dijck JTJM, Dijkman MD, Ophuis RH, de Ruiter GCW, Peul WC, Polinder S. In-hospital costs after severe traumatic brain injury: a systematic review and quality assessment. PloS One. 2019;14(5):e0216743. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Kazim SF, Shamim MS, Tahir MZ, Enam SA, Waheed S. Management of penetrating brain injury. J Emerg Trauma Shock. 2011;4(3):395–402. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Weisbrod AB Rodriguez C Bell R Neal C Armonda R Dorlac W, et al. Long-term outcomes of combat casualties sustaining penetrating traumatic brain injury. J Trauma Acute Care Surg. 2012;73(6):1525–1530. [DOI] [PubMed] [Google Scholar]
- 6.Jimenez CM, Polo J, España JA. Risk factors for intracranial infection secondary to penetrating craniocerebral gunshot wounds in civilian practice. World Neurosurg. 2013;79(5–6):749–755. [DOI] [PubMed] [Google Scholar]
- 7.Harmon LA Haase DJ Kufera JA Adnan S Cabral D Lottenberg L, et al. Infection after penetrating brain injury—an Eastern Association for the Surgery of Trauma multicenter study oral presentation at the 32nd annual meeting of the Eastern Association for the Surgery of Trauma, January 15–19, 2019, in Austin, Texas. J Trauma Acute Care Surg. 2019;87(1):61–67. [DOI] [PubMed] [Google Scholar]
- 8.Johnson M Carroll CP Cass D Andaluz N Foreman B Goodman MD, et al. Single-center experience with antibiotic prophylaxis and infectious complications in civilian cranial gunshot wounds. Neurosurgery Open. 2021;2(1):okaa013. [Google Scholar]
- 9.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):1366–1372. [DOI] [PubMed] [Google Scholar]
- 10.Tribble DR, Spott MA, Shackleford SA, Gurney JM, Murray CK. Department of Defense Trauma Registry Infectious Disease Module Impact on Clinical Practice. Mil Med. 2022;187(Suppl 2):7–16. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Centers for Disease Control and Prevention . CDC/NHSN Surveillance Definitions for Specific Types of Infections. Updated January 2023. Available at: http://www.cdc.gov/nhsn/pdfs/pscmanual/17pscnosinfdef_current.pdf. Accessed January 26, 2023.
- 12.McCafferty R Neal C Marshall S Pamplin J Bell R Rivet D, et al. Neurosurgery and severe head injury (CPG ID: 30): Joint Trama System. Updated March 2, 2017. Available at: https://jts.amedd.army.mil/assets/docs/cpgs/Neurosurgery_and_Severe_Head_Injury_02_Mar_2017_ID30.pdf. Accessed January 26, 2023.
- 13.Barsoumian A Solberg SL Maves RC Markelz AE Crouch H Yun H, et al. Infection prevention in combat-related injuries (CPG ID: 24). Updated January 27, 2021. Available at: https://jts.amedd.army.mil/assets/docs/cpgs/Infection_Prevention_in_Combat-related_Injuries_27_Jan_2021_ID24.pdf. Accessed January 26, 2023.
- 14.Alvis-Miranda HR, Adie Villafane R, Rojas A, Alcala-Cerra G, Moscote-Salazar LR. Management of craniocerebral gunshot injuries: a review. Korean J Neurotrauma. 2015;11(2):35–43. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.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]
- 16.Mazurek MT, Ficke JR. The scope of wounds encountered in casualties from the global war on terrorism: from the battlefield to the tertiary treatment facility. J Am Acad Orthop Surg. 2006;14 (10 Spec No.):S18–S23. [DOI] [PubMed] [Google Scholar]
- 17.Bell RS Mossop CM Dirks MS Stephens FL Mulligan L Ecker R, et al. Early decompressive craniectomy for severe penetrating and closed head injury during wartime. Neurosurg Focus. 2010;28(5):E1. [DOI] [PubMed] [Google Scholar]
- 18.Aarabi B, Taghipour M, Alibaii E, Kamgarpour A. Central nervous system infections after military missile head wounds. Neurosurgery. 1998;42(3):500–507. [DOI] [PubMed] [Google Scholar]
- 19.Wallace PB, Meirowsky AM. The repair of dural defects by graft. An analysis of 540 penetrating wounds of the brain incurred in the Korean war. Ann Surg. 1960;151(2):174–180. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Brandvold B, Levi L, Feinsod M, George ED. Penetrating craniocerebral injuries in the Israeli involvement in the Lebanese conflict, 1982–1985. Analysis of a less aggressive surgical approach. J Neurosurg. 1990;72(1):15–21. [DOI] [PubMed] [Google Scholar]
- 21.Rosenfeld JV, Bell RS, Armonda R. Current concepts in penetrating and blast injury to the central nervous system. World J Surg. 2015;39(6):1352–1362. [DOI] [PMC free article] [PubMed] [Google Scholar]