Affiliations: Department of Emergency Medicine, University of
Rochester School of Medicine, Rochester, NY, USA | Departments of Biostatistics, University of Rochester
School of Medicine, Rochester, NY, USA | Department of Clinical Chemistry, University Hospital
Goettingen, Goettingen, Germany | Institute of Neuropathology, Münster University
Hospital, Münster, Germany
Note: [] Corresponding author: Jeffrey J. Bazarian, MD, MPH, University
of Rochester Medical Center, 601 Elmwood Ave, Box 655, Rochester, NY 14642,
USA. Tel.: +1 585 273 4112; Fax: +1 585 473 3515; E-mail:
jeff_bazarian@urmc.rochester.edu
Abstract: Purpose: To validate a correction factor for the extracranial
release of the astroglial protein, S-100B, based on concomitant creatine kinase
(CK) levels. Methods: The CK- S-100B relationship in non-head injured marathon
runners was used to derive a correction factor for the extracranial release of
S-100B. This factor was then applied to a separate cohort of 96 mild traumatic
brain injury (TBI) patients in whom both CK and S-100B levels were measured.
Corrected S-100B was compared to uncorrected S-100B for the prediction of
initial head CT, three-month headache and three-month post concussive syndrome
(PCS). Results: Corrected S-100B resulted in a statistically significant
improvement in the prediction of 3-month headache (area under curve [AUC] 0.46
vs 0.52, p=0.02), but not PCS or initial head CT. Using a cutoff that maximizes
sensitivity (⩾90%), corrected S-100B improved the prediction of initial
head CT scan (negative predictive value from 75% [95% CI, 2.6%, 67.0%] to 96%
[95% CI: 83.5%, 99.8%]). Conclusions: Although S-100B is overall poorly predictive of
outcome, a correction factor using CK is a valid means of accounting for
extracranial release. By increasing the proportion of mild TBI patients
correctly categorized as low risk for abnormal head CT, CK-corrected S100-B can
further reduce the number of unnecessary brain CT scans performed after this
injury.
