BACKGROUND: Traumatic brain injuries (TBI) in subjects with craniofacial fractures are usually diagnosed by emergency room physicians. We investigated how often TBI remains unrecorded in these subjects, and whether diagnostic accuracy has improved after the implementation of new TBI guidelines. METHODS: All subjects with craniofacial fractures admitted to Oulu University Hospital in 1999 and in 2007 were retrospectively identified. New guidelines for improving the diagnostic accuracy of TBI were implemented between 2000 and 2006. Clinical symptoms of TBI were gathered from notes on hospital charts and compared to the recorded diagnoses at discharge. Logistic regression was used to identify independent predictors for TBI to remain unrecorded. RESULTS: Of 194 subjects with craniofacial fracture, 111(57%) had TBI, 40 in 1999 and 71 in 2007. Fifty-one TBIs (46%) remained unrecorded at discharge, 48 being mild and 3 moderate-to-severe. Subjects with unrecorded TBI were significantly less frequently referred to follow-up visits. Failures to record the TBI diagnosis were less frequent (29/71, 41%) in 2007 than in 1999 (22/40, 55%), but the difference was not statistically significant. The most significant independent predictor for this failure was the clinical specialty (other than neurology/neurosurgery) of the examining physician (p<0.001). The subject's alcohol intoxication did not hamper the diagnosis of TBI. CONCLUSIONS: TBIs remain frequently unrecorded in subjects with craniofacial fractures. Recording of mild TBI slightly but insignificantly improved after the implementation of new guidelines.
BACKGROUND:Traumatic brain injuries (TBI) in subjects with craniofacial fractures are usually diagnosed by emergency room physicians. We investigated how often TBI remains unrecorded in these subjects, and whether diagnostic accuracy has improved after the implementation of new TBI guidelines. METHODS: All subjects with craniofacial fractures admitted to Oulu University Hospital in 1999 and in 2007 were retrospectively identified. New guidelines for improving the diagnostic accuracy of TBI were implemented between 2000 and 2006. Clinical symptoms of TBI were gathered from notes on hospital charts and compared to the recorded diagnoses at discharge. Logistic regression was used to identify independent predictors for TBI to remain unrecorded. RESULTS: Of 194 subjects with craniofacial fracture, 111(57%) had TBI, 40 in 1999 and 71 in 2007. Fifty-one TBIs (46%) remained unrecorded at discharge, 48 being mild and 3 moderate-to-severe. Subjects with unrecorded TBI were significantly less frequently referred to follow-up visits. Failures to record the TBI diagnosis were less frequent (29/71, 41%) in 2007 than in 1999 (22/40, 55%), but the difference was not statistically significant. The most significant independent predictor for this failure was the clinical specialty (other than neurology/neurosurgery) of the examining physician (p<0.001). The subject's alcohol intoxication did not hamper the diagnosis of TBI. CONCLUSIONS: TBIs remain frequently unrecorded in subjects with craniofacial fractures. Recording of mild TBI slightly but insignificantly improved after the implementation of new guidelines.