Extradural hematomas: surgical and nonsurgical treatment.

We read with interest the paper, "Nonoperative Management of Acute Epidural Hematoma Diagnosed by CT: The Neuroradiologist's Role" (1). The paper focuses on the contribution of neuroradiology in providing information to neurosurgeons about which extracerebral hematomas can be treated nonsurgically. The conclusions are that favorable prognostic features can be identified (small lesion volume, minimal midline shift) as can unfavorable features (large volume, central lucent area, evidence of significant mass effect). Whereas we share these conclusions, the paper is followed by a comment from Sagher et al (2) which recommends restricting conservative management to those few cases of high convexity extradural hematomas (EDH) of an unprecise volume and to EDH detected 48 hours or more after trauma; furthermore, they consider routine conservative management of even asymptomatic EDH as "hard to justify." The comment is not surprising. A paper published in 1985 (3) reporting 12 cases of conservatively treated EDH was followed by the comment, "This is a thought provoking and, in some ways, courageous study. The patients are the ones with courage" (4). In 1986, a paper by our group (9) was followed by a comment by Cooper which stated, "Early operative evacuation of these lesions seems to be much safer, much less expensive and more truly conservative management" (6). The reluctance to accept the possibility of nonsurgical management of EDH seems to be the same in 1992 as it was in 1985. The availability of computed tomography (CT) scanners in the 80s and 90s produced a significant change in the population of patients harboring EDH who were admitted to neurosurgical wards. In our area, CT scanning of every minor head-injured adult patient with a skull fracture (7) allowed the detection of a large number of asymptomatic hematomas (8) . In the past these patients always existed but were submitted to a sort of "inadvertent conservative management" in peripheral hospitals. A few of them underwent clinical deterioration and were transferred (sometimes too late) to neurosurgery for treatment. Whereas EDH is the most frequent lesion if we examine a population of asymptomatic patients with skull fracture (7), they are present in only 12% to 17% (9 , 1 0) of patients who "talk and deteriorate." The goal is to identify the few cases at risk of deterioration and not to operate on the vast majority of EDHs that are not prone to enlarge. The pure identification of the EDH is a lifesaving procedure: the patient is transferred for observation to neurosurgery (in Europe, minor head-injured cases are kept for observation outside neurosurgery) and even if improper conservative management is instituted, clinical deterioration occurs where the patient can be