Clinical risk factors associated with cerebrospinal fluid leak in facial trauma: A retrospective analysis.

OBJECTIVE

Cerebrospinal fluid (CSF) leak occurs most commonly following skull fracture, with a CSF leakage complicating up to 2% of all head traumas. This study aims to identify demographic and injury characteristics correlated with the highest risk of CSF leak in patients with known facial fractures.

METHODS

Retrospective data was collected from a previously described trauma registry from 2010 to 2019. Patients over 18 years old with any type of facial fracture, known CSF leak status, available neuroimaging, and hospital admission were included. Chi-Square analysis for demographic and injury characteristic data were utilized.

RESULTS

A total of 79 patients with CSF leak and 4907 patients without CSF leak were included in the database. Patients with CSF leak tended to be younger than those without CSF leak (38.45 +/- 0.28 vs 44.08 +/- 0.28, M +/- SE, p = 0.0197). CSF leak depended on the mechanism of injury (MOI; X =27.02, df=2, p = 0.0000013), with CSF leak rates highest in penetrating injuries (4.87%) and motor vehicle accidents (1.78%) compared to blunt injuries (0.95%); age did not significantly differ between the MOI groups (p = 0.11). CSF leak was also more common in patients with a lower Glasgow coma scale (GCS; 7.95 +/- 0.58 vs 12.21 +/- 0.10, p = 10), LeFort type 2&3 and pan-facial fractures compared to all other facial fracture types (8.9% vs 1.2%, p = 10), and radiographic midline shift (29.4% vs 9.1%, p = 10). There was a trend towards a higher proportion of males in those with CSF leak compared to those without (83.3% vs 73.7% males, p = 0.073), and in patients with prolonged loss of consciousness (LOC; 9.43% with LOC > 1 h vs 2.69% LOC < 1 h, p = 0.056).

CONCLUSION

Facial fractures often present with CSF leak, and certain demographic and injury risk factors including younger age, worse GCS score, evidence of midline shift, and certain mechanisms of injury (penetrating and motor vehicle) are correlated with increased risk and warrant close screening and follow-up for CSF leak detection. LeFort type 2&3 and pan-facial fractures are at high risk of CSF leak.