Trends in Management of Cerebrospinal Fluid Leaks Among Pediatric Head Trauma Patients With Skull Base Fractures

Abstract

Abstract INTRODUCTION Cerebrospinal fluid (CSF) leaks occur in association with skull base fractures (SBF) among head-trauma patients. Using a national database, management trends of CSF-leaks associated with SBF in pediatric patients were examined. METHODS Weighted data were analyzed from the Nationwide Inpatient Sample 2000-2014 and all pediatric head-trauma patients having a diagnosis of SBF identified. Injury severity was mapped using the Abbreviated Injury Severity (AIS) scale and categorized as mild: AIS < 3, moderate: AIS = 3, and severe: AIS >3. Trends over time and across 5-yr age groups were examined at a significance level of P < .05. RESULTS A total of 699 173 SBF cases were diagnosed, representing 23.25% of all pediatric head-trauma cases. The majority were closed-SBF (93.14%) and involved male patients (68.60%), and mean patient age was 11.25 yr (SD = 6.49). The median AIS-injury score was 3.0. Severe TBI was more common among 15 to 19 yr olds (P < .001). Overall, CSF-leaks occurred in 2.10%, were more common in older children versus under-5 yr olds (P < .05), and were mostly managed conservatively with fewer patients requiring duraplasty (8.48%) and/or lumbar-drain placement (9.54%). The requirement for nonconservative management, including duraplasty (OR = 5.53; 95% CI = 1.58–19.37; P < .001) and lumbar-drain placement (OR = 2.64; 95% CI = 1.06–6.60; P = .03), was higher among 15 to 19 versus under-5 yr olds. The risk of bacterial meningitis in patients with CSF-leaks was slightly increased among under-5 yr olds compared to older children (P < .05). In-patient mortality associated with SBF was 4.68%, was higher among 15 to 19 versus under-5 yr olds (6.55% vs 3.44%, P < .001) and in open versus closed SBF (16.07% vs 3.84%, P < .001), and decreased significantly across time from 5.09% in 2000 to 3.75% in 2014 (P < .001). The decline in mortality seen across all age groups was significant only in under-5 and in 15 to 19 yr olds (P = .02 and .03, respectively). There was no increased mortality associated with CSF-leaks (P = .65). CONCLUSION CSF-leaks occurring among pediatric head-trauma patients are mostly managed conservatively. Younger children may be at an increased risk of meningitis following CSF-leaks.