Two-year Mortality and Functional Outcomes in Combat-related Penetrating Brain Injury: Battlefield Through Rehabilitation

M. Benjamin Larkin, MD, PharmD; Erin K. M. Graves, MD, MA; Jason H. Boulter, MD; Nicholas S. Szuflita, MD, MPH; R. Michael Meyer, MD; Michael E. Porambo, MD; John J. Delaney, MD; Randy S. Bell, MD


Neurosurg Focus. 2019;45(6):e4 

In This Article


Penetrating brain injury represents one of the most challenging clinical entities for both civilian and military neurosurgeons. Since 2003, the conflicts in Iraq and Afghanistan have resulted in the highest number of PBIs for US servicemen and -women since the Vietnam War. Previous studies analyzing outcomes for these patients have suggested that mortality is much lower than in civilian cohorts, and good functional outcomes can be obtained even for those with low admission GCS scores (< 5).[1,4,17,34,35] These studies were hampered by the absence of analysis of data from the point of injury (immortal time bias), and considered outcomes only for those patients who survived to reach definitive care after long air evacuation transport to the continental US. This study was therefore undertaken for 2 reasons: first, to assess whether functional outcomes and mortality differed from what has been reported in prior military studies, and second, to assess whether or not outcomes for PBI differ descriptively between military and civilian populations. To our knowledge, this is the first paper analyzing long-term outcomes for patients with PBI in a US military cohort from the point of injury forward.

This study revealed that good functional outcomes were achieved for those who sustained combat-related PBI and survived their initial resuscitation (GOS scores at 6, 12, and 24 months), confirming conclusions from prior studies.[4,35] The mortality rate in this study, 21%, was higher than that reported in previous military cohorts, but lower than in published civilian studies.[1,4,17,34,35] This finding is not surprising, and our rate of 21% is likely a truer estimation of mortality from PBI in these conflicts than previously estimates. Commensurate with prior reports, in our study cohort, explosive devices were the leading cause of PBI.[4,22,35] This study also revealed an association between the following factors and overall worse outcome at all time points: admission GCS score ≤ 5, admission ISS ≥ 26, brain herniation on admission CT scan of the head, and gunshot wound as the mechanism of injury.

The association between low admission GCS score and worse early outcome has been well described for civilian cohorts.[2,16,24,27] In these civilian cohorts, where gunshot wound was the predominant mechanism of PBI, over 90% of patients who presented with an admission GCS score of 3–5 died, with almost none achieving a good functional outcome. In contrast, in the current study, half of the patients presented with an admission GCS score of 3–5, and roughly one-third of these patients died and two-thirds survived to achieve good functional recovery. Explanations for the improved survivability may include the use of protective gear (helmets, body armor) in the military population and the predominant mechanism of injury in the military population (blast; see below).[6,7,19,21,31–33] It is also conceivable that the reported admission GCS score in this military population may be artificially lower than the actual admission GCS score as a result of ongoing sedation/neuromuscular blockade from field resuscitation and/or a poor understanding of the GCS by reporting field personnel.[18,29] In our view, this also complicates estimation of field GCS scores in civilian settings. The fact that those patients in our cohort who did survive showed an upward mobility in their GOS scores is consistent with previous studies that show only a weak correlation between admission GCS scores and long-term outcome and may be explained by the above considerations.[10,11,15,20,26] All that said, the fact that so many patients with admission GCS scores ≤ 5 had good functional outcomes in this cohort should caution forward deployed neurosurgeons to consider intervention, rather than expectant management, in this population. Further study is necessary to elucidate the granular differences in this unique population.

It is interesting to note that those patients in this cohort who presented with PBI secondary to gunshot wound had worse functional outcomes at all time points than those who suffered a blast PBI. The mortality rate for those who presented with GSWH was 41% (9/22 patients) compared to 14% for those who suffered blast PBI; this GSWH mortality rate more closely mirrors the civilian experience. This observation is interesting when considering the complexity (blast overpressure, polytrauma) of blast-related injuries, especially in this setting where those with admission ISS ≥ 26 had worse functional outcomes.[8,9,14,23,25,31]

Nearly half of the patients in this study required some form of surgical intervention. As expected, those who required surgery presented with lower admission GCS scores in comparison to those who did not require surgery. Interestingly, analysis revealed that long-term functional outcomes did not differ between those requiring surgery and those who did not. While this finding was surprising and may be a reflection of the predominant mechanism of injury (blast vs GSWH) and/or the tendency for early and aggressive intervention by military surgeons, the small size of the population precludes solid relational conclusions.[5,12,30,35]


The results of this study, while suggestive of trends for all US military personnel, apply only to those patients included in this cohort. As with many previous studies concerning trauma, all of the patients in this cohort were male. The results may therefore not apply to the active-duty female population.

Despite the intention of limiting immortal time bias, there remains some element in this study. Only those patients who were evacuated to the Multinational Medical Unit at KAF and were evaluated by military medical personnel were included in this study. Those individuals killed in action in the applicable area of operations during this timeframe were not included in this study. As such, the mortality rate associated with PBI in US military personnel may be higher than that reflected in this study.

The authors acknowledge that the efficacy of GOS score determination is known to decline from structured patient evaluations to telephone and further correspondence interviews. However, because of the retrospective nature of this study, and to avoid recall bias, the authors felt that GOS score determination from review of the electronic health record was more reliable than the patient's recollection of historical functional status.

Lastly, CT scans obtained in theater follow the patient to the medical treatment facilities that later care for them, which may or may not have been WRNMMC in the 80 cases included in this study. As a result, at the time of data collection, only the radiological reports were available for review.