Factors associated with posttraumatic meningitis among traumatic head injury patients: a nationwide study in Japan

This was a retrospective observational study that used data of emergency trauma patients registered in the JTDB. The study period was the 12-year period from January 2004 to December 2015. In this study, we included patients with a head Abbreviated Injury Scale (AIS) score of 3 or more who were transported to a JTDB-participating hospital and were registered in the JTDB registry [16]. We excluded those patients who were in cardiopulmonary arrest on hospital arrival, whose mandatory data were missing. Cardiopulmonary arrest on hospital arrival was defined as a patient, whose systolic blood pressure was 0 mmHg and/or heart rate was 0 bpm on hospital arrival. From the JTDB database, we extracted factors such as age, sex, past medical history, type of trauma, CSF leakage, type of skull fracture, burr hole surgery in the emergency department, operative procedure, and procedures performed during the neurosurgery, and repeat surgery within 48 h of the first surgery. In this study, we defined multiple trauma as head trauma patients with AIS score of 3 or more in body areas other than the head. In addition, we defined CSF leakage as a patient for whom AIS code 150204.3 was recorded. In addition, those with AIS codes 150408.4, 150406.4, 150404.3, 150402.2, and 150400.2 were defined as having a skull fracture, and those with AIS codes 150206.4, 150204.3, 150202.3, and 150200.3 were defined as having a basilar skull fracture. The primary operative method of neurosurgery was classified as burr hole surgery or craniotomy, and the procedures performed during the neurosurgery were classified as evacuation of hematoma, decompressive craniectomy, lobectomy, duraplasty, cranioplasty, and external ventricular drainage. This study was approved by the ethics committee of Osaka University Graduate School of Medicine (No. 16260). Personnel identifiers were removed beforehand from the JTDB database, and thus, the patients’ right to informed consent was waived.

The JTDB was launched in 2003 by the Japanese Association for the Surgery of Trauma (Trauma Surgery Committee) and the Japanese Association for Acute Medicine (Committee for Clinical Care Evaluation) [1, 17], similar to trauma databases in North America, Europe, and Oceania [18]. By 2016, 256 major emergency medical institutions across Japan had been registered in the JTDB database [1]. These hospitals have equal ability to that of Level I trauma centers in the United States. Data were collected via the Internet from participating institutions. The physicians and medical assistants who attended the AIS-coding course were the main registers of the data [19].

The JTDB captures trauma patient data on age, sex, mechanism of injury, AIS code (version 1998), Injury Severity Score (ISS), vital signs on hospital arrival, date and time series from hospital arrival to discharge, medical treatments such as interventional radiology, surgical operation, and CT scanning, complications, and mortality at discharge [19]. ISS was calculated from the top three scores of AIS in nine sites classified by AIS codes.

The endpoint was the occurrence of meningitis during hospitalization, and we defined this meningitis as posttraumatic meningitis. We extracted data on the occurrence of meningitis from the JTDB registry.

In this study, we assessed factors associated with the occurrence of posttraumatic meningitis with multivariable logistic regression analysis and calculated the adjusted odds ratio (AOR) and 95% confidential interval (CI). Multivariable logistic regression analysis was performed by forced enter-method. The independent parameters were age group (0–9 years, 10–19 years, 20–29 years, 30–39 years, 40–49 years, 50–59 years, 60–69 years, 70–79 years, 80 years, or older), sex, diabetes, implementation of hemodialysis, multiple trauma or single head trauma, presence or absence of CSF leakage, type of skull fracture, implementation of burr hole surgery in the emergency department, first neurosurgical operation, procedures performed during the first neurosurgical operation, and repeat surgery within 48 h of the first surgery. These independent parameters that were biologically essential and considered to be associated with outcomes were included in the multivariable regression analysis-based previous studies [3‐9, 20, 21]. As a further sub-analysis, we divided these patients into the single head trauma group and the multiple trauma group [22], and these independent parameters associated with posttraumatic meningitis were also assessed with multivariable logistic regression analysis. All tests were two-tailed, and a P value of < 0.05 was considered statistically significant. Statistical analysis was performed by SPSS version 23.0J (IBM Crop., Armonk, NY, USA). This manuscript was written based on the STROBE statement to assess the reporting of cohort and cross-sectional studies.

There are some limitations in this study. First, because we extracted data on meningitis from a trauma registry, we did not obtain information on how the diagnosis of meningitis was made. However, because the morbidity rate of the patients with posttraumatic meningitis was 0.4 to 1.4% in the previous studies on posttraumatic meningitis [4, 35], and the rate in the present study was equivalent to the rates in these studies, we believe that the diagnostic accuracy would be appropriate. Second, information on the administration of antibiotics in each patient was unknown in this study, but it would be important, as the administration of antibiotics before the occurrence of posttraumatic meningitis might lead to its prevention. However, antibiotic treatments based on the Surviving Sepsis Campaign Guidelines were performed in many intensive care units in Japan [36]. Therefore, similar administration of antibiotic treatments would have been performed on the present study patients. Third, CSF leakage in patients with basilar skull fracture was assessed in this study, but CSF leakage due to frontal sinus injury could not be assessed, because there were no data on this condition. Finally, this study was an observational study, and there may be some unknown confounding factors.