Risk Analysis of Hemorrhages in Stereo-Electroencephalography Procedures

Abstract

Listen

Translate article

Abstract INTRODUCTION Despite the variations in the implantation techniques used, stereo-electroencephalography (SEEG) has been reported to be quite safe with a major complication incidence of less than 1%. While some studies have attempted to classify complications, there is currently no standardized or evidence-based method for either the detection or reporting of complications in SEEG, in particular, hemorrhages. As a result, clinicians examining the current literature rely on centers reporting “clinically significant,” or “major” complications which can be arbitrarily and vaguely defined in each specific instance. METHODS This was a retrospective, single center observational study examining every SEEG implantation performed at our center from 2009 to 2017. This consisted of 549 consecutive SEEG implantations using a variety of stereotactic and imaging techniques. A hemorrhage grading system was applied by a blinded neuroradiologist to every postimplant and postexplant CT scan. Hemorrhages were classified as asymptomatic or symptomatic based on neurological deficit seen on exam. Statistical analysis included multivariate regression using relevant preoperative variables to predict the presence of hemorrhage. RESULTS A toal of 105 implantations (19.1%) had any type of hemorrhage seen on postimplant CT. Of these, 93 (16.9%) were asymptomatic and 12 (2.2%) were symptomatic with 3 implantations (0.6%) resulting in either a permanent deficit (2, 0.4%) or death (1, 0.2%). Male sex, increased number of electrodes and increasing age were associated with increased risk of postimplant hemorrhage on multivariate analysis. Increasing score in the grading system was related to a statistically significant increase in the likelihood of a symptomatic hemorrhage. CONCLUSION Detailed examination of every postimplantation CT reveals that the total hemorrhage rate appears higher than previously reported. Most of these hemorrhages are small and asymptomatic. Our grading system may be useful to risk stratify these hemorrhages and awaits prospective validation.