Neuroimaging studies are needed to make the diagnosis and elucidate the etiology of ICH. Computed tomography (CT) is considered the gold standard (Fig. 3.1); though, magnetic resonance imaging (MRI) is more useful in elucidating the etiology of ICH (Fig. 3.2). Blood on CT appears hyperdense or isodense in patients with severe anemia, whereas blood cell degradation with time leads to altered MRI sequences. Small cerebellar hemorrhages in the posterior fossa can be missed on cranial CT unless thin cuts of the posterior fossa are obtained. CT and MRI equally perform in identifying ICH. CT may be superior in demonstrating associated subarachnoid hemorrhage, whereas MRI is certainly better in identifying an underlying structural pathology such as in those with hemorrhagic transformation of a mass lesion of arteriovenous malformation. Gradient-echo T2-weighted MRI is very helpful in demonstrating silent deposition of blood products that can be a risk factor for spontaneous ICH (Fig. 3.3). The risk of seizure varies considerably according to the size, location, and underlying etiology of the hemorrhage as well as the clinical characteristics that are discussed in detail .

It is important to understand the natural history of seizures in the setting of ICH, as it helps deciding the type of diagnostic and therapeutic approaches. Seizures tend to present early in the course with 50–70 % occurring within 24 h from onset and 90 % by the 3rd day [4, 7, 10, 14, 15]. Earlier in the chapter, it was mentioned that early seizures occur in 14–41 % [6–9, 15], while the 30-day risk of seizures is 8 % [15] with a reported frequency of SE of 0.3 %, which represents about 20 % of those patients with ICH who develop seizures [11–13].

Some studies have identified factors associated with higher risk of developing seizures with ICH. These factors include imaging features like the volume of the accumulated blood within the cranial cavity, the presence of hydrocephalus, involvement of the cortex and/or lobar hemorrhage, midline shift, and the presence of associated subarachnoid and subdural blood. Clinical factors predicting seizures include age, history of alcohol abuse, the presence of sepsis, severity of the neurological deficit measured by the National Institutes of Health Stroke Scale (NIHSS) or Glasgow Coma Score, in addition to the presence of significant disability as measured by modified Rankin score (Table 3.2) [2–5, 7–9, 16–25]. In a large administrative dataset from the National Inpatient Sample (NIS), Bateman et al. found that renal disease, coagulopathy, history of alcohol abuse, sodium abnormalities, and the presence of a tumor as cause of the hemorrhage were associated with SE [11].

There have been some efforts to stratify the risk of late seizures in ICH. Haapaniemi et al. created the CAVE score, “C” for cortical involvement, “A” for age < 65 years, “V” for volume > 10 cc, and “E” for early seizures < 7 days from onset. The authors utilized a cohort of 764 ICH patients who survived to 7th day from the Helsinki ICH study to derive the score and a cohort of 325 subjects from the Lillie Prognosis of Intracerebral Hemorrhage study to validate the score. The cumulative risk of seizures was 7.1, 10, 10.2, 11, and 11.8 % at 1, 2, 3, 4, and 5 years, respectively. Each item of the score has a value of 1 and the cumulative risk of seizures is 0.6, 3.6, 9.8, 34.8, and 46.7 % with 0, 1, 2, 3, and 4 points, respectively. The CAVE score was validated and found reliable in identifying patients with a higher risk for the development of seizures [8] .

Reports on the risk of epilepsy provide a variety of numbers that reflect the different definitions used in the studies and the study design. Arntz et al. prospectively studied a cohort of 697 patients with first stroke age 18–50 years, 66 of those had ICH. Authors used the definition by the International League Against Epilepsy (ILAE) of one seizure in the presence of an enduring condition that can cause epilepsy [6]. The overall risk of epilepsy in that cohort was 11.3 % including patients with ischemic strokes and transient ischemic attacks. The overall risk of epilepsy in patients with ICH was 16.7 %. When the risk projected at 5 years in Kaplan–Meier survival analysis, the risk of epilepsy was about 20 % when patients did not have recurrent seizures as compared to a risk of about 13 % for epilepsy with recurrent seizures. Importantly, the risk of seizure recurrence among patients with ICH and seizures was higher in patients with late-onset seizures (> 7 days) as compared to those with early onset of seizures (< 7 days), 6/7 patients (85 %) versus 1/4 (25 %) [6]. In contrast, Sung and colleagues reported a retrospective review of 1402 patients with ICH. A total of 64 patients (4.6 %) had seizures, and 35 (2.5 %) developed epilepsy that was defined as the presence of 2 or more seizures [9]. Several demographic, clinical, and imaging features have been evaluated as possible predictors of seizures (Table 3.2). Immediate seizures were exclusively correlated with characteristics of ICH (lobar location and small volume). Lobar location has been widely recognized as the most potent predictor of immediate seizures [9, 14, 17, 25]. Similarly, hematoma enlargement is usually associated with seizure and poor outcome. The occurrence of SE seems to be influenced by nonlesional factors such as alcohol abuse. The relations between alcohol and seizures are complex; however, both alcohol abuse and alcohol withdrawal may precipitate seizures [7]. A study by Passero et al. reported that recurrent seizures are associated with clinical events including brain infarct, hematoma enlargement, and sudden suspension of AEDs [15] .