Predicting Outcome and Conservative Treatment Failure in Patients with Skull Fracture after Traumatic Brain Injury: A Retrospective Cohort Study.

ABSTRACT

OBJECTIVE: Traumatic brain injury (TBI) remains a major cause of morbidity and mortality worldwide. The prognostic value of skull fracture (SF) remains to be clearly defined. To evaluate the need for neurosurgical intervention and determine the risk factors of conservative treatment failure (CTF), we retrieved from the hospital database the records of patients with SF after TBI. METHODS: We analyzed 146 consecutive patients (mean age 49.8 ± 17.5 years) treated at the department of neurosurgery in a 5-year period. Clinical data, radiologic reports, and laboratory results were evaluated retrospectively. RESULTS: A total of 63% of patients were treated conservatively, 21.9% were operated on immediately, and 15.1% experienced CTF. Overall, 73.3% had a favorable outcome; the mortality rate was 13%. Intracranial bleeding occurred in 96.6% of cases, basilar SF in 61%, and cerebrospinal fluid (CSF) leak in 2.8%. The independent risk factors for outcome were Glasgow Coma Scale (GCS) score, age, and platelet count (PCT). The independent risk factors for CTF were epidural hematoma, subdural hematoma, mass effect, edema, international normalized ratio, PCT, mean platelet volume, and CSF leakage. The consensus decision tree algorithm used at the accident and emergency department indicated patients with no need for neurosurgical intervention with an accuracy of 91.7%, sensitivity of 88.9%, and featured the importance of mass effect, GCS, and epidural hematoma. CONCLUSIONS: Tests included in the complete blood count appeared useful for predicting the course in patients with SF, although the most important factors were age and neurologic status, as well as radiologic findings. Our decision tree requires further validation before it can be used in everyday practice.