Background

Penetrating traumatic brain injury (pTBI) is the most lethal form of TBI, with mortality rates as high as 90%. This high mortality rate leads many providers to feel that the treatment of pTBI is futile. Contrary to this point of view, several studies have shown that victims of pTBI who present with a Glasgow Coma Scale (GCS) ≥6 have a reasonable chance of a meaningful outcome. This study sought to investigate outcomes of pTBI patients based on GCS score who underwent neurosurgical intervention (craniotomy or craniectomy) and compare them with patients who did not undergo surgical intervention. Materials and methods

The study represents a secondary analysis of the data that were collected from 2006 to 2016 from 17 institutions as part of a multi-center study, investigating clinical outcomes for adult patients sustaining pTBI and surviving >72 h. Patients were divided into those with GCS 3-5 and those with GCS ≥6. Within these groups, patients were stratified by whether they received surgical intervention, compared with standard non-surgical care. Patient level data (age and gender), clinical data (Injury Severity Score and Abbreviated Injury Score), GCS on admission, post-op infection rates, and outcomes data (mortality, length of stay [LOS], intensive care unit LOS) were collected. Both groups were compared using independent sample t-test or chi-squared test. Results

Seven hundred twenty patients with pTBI were identified over 11 y, out of which 336 (46.7%) underwent surgery. The mean Injury Severity Score and Abbreviated Injury Score on admission were higher in the surgical intervention group than their non-surgical counterpart in patients with a GCS ≥6 (P < 0.0001). Patients with GCS of 3-5 with surgical intervention demonstrated a higher survival rate than non-surgical patients (P < 0.0001). In the GCS ≥6 group, surgical intervention did not impact near-term mortality. Intensive care unit LOS was significantly longer in the surgical intervention group in patients with GCS ≥ 6 (P < 0.0001) and GCS of 3-5 (P < 0.0001), as was total hospital LOS (P < 0.0001). Patients with a GCS 3-5 and ≥6 who underwent surgical intervention were more likely to develop a central nervous system infection (P = 0.016; P = 0.017). Conclusions

Surgical intervention in pTBI patients with GCS 3-5 results in improved mortality but comes at a cost of increased resource utilization in the form of longer LOS and higher infection rate. On the other hand, in patients with GCS ≥6, surgery does not provide significant benefits in patient survival. Future prospective studies providing insight as to the impact of surgery on the resource utilization and quality of survival would be beneficial in determining the need for surgical intervention in this population.