M. A. Brooke1, F. Castro-Moure2, A. K. Patel2, G. P. Victorino1 2Highland Hospital,Department Of Surgery,Oakland, CALIFORNIA, USA 1University Of California – San Francisco East Bay,San Francisco, CA, USA
Introduction: In recent years, the phenomenon of rapidly resolving acute subdural hematoma (RRASDH) has gained recognition, but it is still poorly understood. Previous studies have suggested the contribution of coagulopathy to rapid resolution, as well as the presence of a "low density band" between the ASDH and the skull, representing cerebrospinal fluid washout. Our goal was to investigate the significance of RRASDH, and examine any predictive factors.
Methods: A retrospective analysis was performed of all non-operatively managed ASDHs treated at our trauma center from 2011-2015. Inclusion criteria were ASDH on computed tomography (CT), admission Glasgow coma score (GCS) >7, and repeat CT to evaluate ASDH progression or regression over time. Rapid resolution was defined as decrease in hematoma thickness by 50% in 72 hours. Clinical data, CT findings, and trauma endpoints were collected and analyzed for resolving and non-resolving cases.
Results: There were 154 patients who met inclusion criteria. A change in hematoma thickness was associated with mortality, with patients who died showing a mean hematoma growth of 48% versus a 9% reduction in patients who survived (p=0.002). There were 29 cases of RRASDH, in which the average resolution rate was 0.23 mm/hr with a mean 78% size reduction. There were no predictive factors for rapid resolution with no differences between resolving and non-resolving groups in GCS, injury severity score, or initial CT findings (presence of cranial fractures, additional hemorrhage, cerebral edema, midline shift, or location of ASDH). These two groups also demonstrated no differences in endpoints such as mortality, ventilator days, ICU length of stay, or discharge destination. We found no difference between these two groups in either proportion of coagulopathic patients or gross PT/INR values. Finally, there was no difference in the prevalence of the "low density band" on CT. When compared with a group of patients who experienced rapid growth (>50% axial thickness in 72 hours), the RRASDH group did have a lower mortality (3.4% vs. 23.5%, p =0.04).
Conclusion: To our knowledge, this is the largest review of RRASDH. Our findings contradict some of the recent literature, as well as prevailing theories for the mechanism of rapid ASDH resolution. We found no significant relationship between coagulopathy or presence of a "low density band" and resolution of ASDH, shedding doubt on these theories of etiology. However, we also found no relationship between rapid resolution and better standard trauma outcomes, calling into question the basic significance of rapid resolution alone. What is far clearer is that rapid growth of ASDH is a poor clinical sign.