Detecting subarachnoid hemorrhage: Comparison of combined FLAIR/SWI versus CT

OBJECTIVES: Aim of this study was to compare the utility of susceptibility weighted imaging (SWI) with the established diagnostic techniques CT and fluid attenuated inversion recovery (FLAIR) in their detecting capacity of subarachnoid hemorrhage (SAH), and further to compare the combined SWI/FLAIR MRI data with CT to evaluate whether MRI is more accurate than CT. METHODS: Twenty-five patients with acute SAH underwent CT and MRI within 6 days after symptom onset. Underlying pathology for SAH was head trauma (n=9), ruptured aneurysm (n=6), ruptured arteriovenous malformation (n=2), and spontaneous bleeding (n=8). SWI, FLAIR, and CT data were analyzed. The anatomical distribution of SAH was subdivided into 8 subarachnoid regions with three peripheral cisterns (frontal-parietal, temporal-occipital, sylvian), two central cisterns and spaces (interhemispheric, intraventricular), and the perimesencephalic, posterior fossa, superior cerebellar cisterns. RESULTS: SAH was detected in a total of 146 subarachnoid regions. CT identified 110 (75.3%), FLAIR 127 (87%), and SWI 129 (88.4%) involved regions. Combined FLAIR and SWI identified all 146 detectable regions (100%). FLAIR was sensitive for frontal-parietal, temporal-occipital and Sylvian cistern SAH, while SWI was particularly sensitive for interhemispheric and intraventricular hemorrhage. CONCLUSIONS: By combining SWI and FLAIR, MRI yields a distinctly higher detection rate for SAH than CT alone, particularly due to their complementary detection characteristics in different anatomical regions. Detection strength of SWI is high in central areas, whereas FLAIR shows a better detection rate in peripheral areas.

Assessing intravascular volume by difference in pulse pressure in pigs submitted to graded hemorrhage

We assessed changes in intravascular volume monitored by difference in pulse pressure (dPP%) after stepwise hemorrhage in an experimental pig model. Six pigs (23-25 kg) were anesthetized (isoflurane 1.5 vol%) and mechanically ventilated to keep end-tidal CO2 (etCO2) at 35 mmHg. A PA-catheter and an arterial catheter were placed via femoral access. During and after surgery, animals received lactated Ringer's solution as long as they were considered volume responders (dPP>13%). Then animals were allowed to stabilize from the induction of anesthesia and insertion of catheters for 30 min. After stabilization, baseline measurements were taken. Five percent of blood volume was withdrawn, followed by another 5%, and then in 10%-increments until death from exsanguination occurred. After withdrawal of 5% of blood volume, all pigs were considered volume responders (dPP>13%); dPP rose significantly from 6.1+/-3.3% to 19.4+/-4.2%. The regression analysis of stepwise hemorrhage revealed a linear relation between blood loss (hemorrhage in %) and dPP (y=0.99*x+14; R2=0.7764; P<.0001). In addition, dPP was the only parameter that changed significantly between baseline and a blood loss of 5% (P<0.01), whereas cardiac output, stroke volume, heart rate, MAP, central venous pressure, pulmonary artery occlusion pressure, and systemic vascular resistance, respectively, remained unchanged. We conclude that in an experimental hypovolemic pig model, dPP correlates well with blood loss.

Postmortem radiology of fatal hemorrhage: measurements of cross-sectional areas of major blood vessels and volumes of aorta and spleen on MDCT and volumes of heart chambers on MRI

OBJECTIVE: Autopsy determination of fatal hemorrhage as the cause of death is often a difficult diagnosis in forensic medicine. No quantitative system for accurately measuring the blood volume in a corpse has been developed. MATERIALS AND METHODS: This article describes the measurement and evaluation of the cross-sectional areas of major blood vessels, of the diameter of the right pulmonary artery, of the volumes of thoracic aorta and spleen on MDCT, and of the volumes of heart chambers on MRI in 65 autopsy-verified cases of fatal hemorrhage or no fatal hemorrhage. RESULTS: Most cases with a cause of death of "fatal hemorrhage" had collapsed vessels. The finding of a collapsed superior vena cava, main pulmonary artery, or right pulmonary artery was 100% specific for fatal hemorrhage. The mean volumes of the thoracic aorta and of each of the heart chambers and the mean cross-sectional areas of all vessels except the inferior vena cava and abdominal aorta were significantly smaller in fatal hemorrhage than in no fatal hemorrhage. CONCLUSION: For the quantitative differentiation of fatal hemorrhage from other causes of death, we propose a three-step algorithm with measurements of the diameter of the right pulmonary artery, the cross-sectional area of the main pulmonary artery, and the volume of the right atrium (specificity, 100%; sensitivity, 95%). However, this algorithm must be corroborated in a prospective study, which would eliminate the limitations of this study. Quantitative postmortem cross-sectional imaging might become a reliable objective method to assess the question of fatal hemorrhage in forensic medicine.

Virtopsy hemorrhage of the posterior cricoarytenoid muscle by blunt force to the neck in postmortem multislice computed tomography and magnetic resonance imaging

In forensic autopsies, one of the most important and common signs of violence to the neck is hemorrhages of the soft tissues. The Institute of Forensic Medicine in Bern evaluates the usefulness of postmortem multislice computed tomography (MSCT) and magnetic resonance imaging (MRI) of forensic cases prior to autopsy. The aim of this study was to prove the sensitivity of postmortem MSCT and MRI in the detection of hemorrhages of the neck muscles. A full body scan prior to and a detailed scan of the explanted larynx after autopsy were performed. MSCT detected multiple fractures of the larynx. Detailed MRI was able to demonstrate the hemorrhage of the left posterior cricoarytenoid muscle. The minor hemorrhage of the right posterior cricoarytenoid muscle could not be detected with certainty. Although more experience is required, we conclude that combined MRI and MSCT examination is a useful tool for documentation and examination of neck muscle hemorrhages in forensic cases.

Postmortem noninvasive virtual autopsy: extrapleural hemorrhage after blunt thoracic trauma

A 19-year-old man speeding recklessly along a highway caused a left-frontal crash with another car. After his vehicle came to a standstill, he climbed out of the wreck and crawled across the tarmac to the other side of the road, where he died several minutes after the accident and before the arrival of an ambulance. Postmortem multislice computed tomography (MSCT) demonstrated fractures of the first, second, and third ribs and scapula on the left, an extrapleural hemorrhage in the apical region of the left thorax, as well as a large amount of blood in the left thoracic cavity. These radiologic findings were indicative of a delayed rupture of a traumatic extrapleural hematoma into the pleural space. A traditional autopsy confirmed the very rare diagnosis of a traumatic extrapleural hemorrhage with a delayed rupture.

Hemorrhage in vitamin K deficiency--a preventable entity

Der diaplazentare Transport von Vitamin K ist kaum messbar, die Muttermilch ist arm an Vitamin K und die intestinale Flora von Neugeborenen produziert praktisch kein Vitamin K. Deshalb weisen gesunde Neugeborene «physiologischerweise» tiefe Vitamin-K-Spiegel auf, was durch Verminderung der Vitamin-K-abhängigen Gerinnungsfaktoren zu schweren Mangelblutungen führen kann. Die klassische Form der Vitamin-K-Mangelblutung tritt mit einer Inzidenz von bis zu 1.5% bis zum 7. Lebenstag auf, die Spätform wird bis zur 12. Lebenswoche bei bis zu 10 von 100000 Neugeborenen festgestellt. Mit einer adäquaten Vitamin-K-Prophylaxe lassen sich Vitamin-K-Mangelblutungen grösstenteils verhindern. Die heute in der Schweiz empfohlene Prophylaxe von 3 oralen Dosen à 2 mg Konakion® MM in der 4. Lebensstunde, am 4. Lebenstag und in der 4. Lebenswoche ist bezüglich unerwünschter Nebenwirkung äusserst sicher, insbesondere besteht kein erhöhtes Krebsrisiko. Angesichts der dramatischen Folgen der häufig intrakraniell lokalisierten Vitamin-K-Mangelblutungen ist sowohl bei medizinischen Fachpersonen als auch bei den Eltern eine möglichst gute Compliance für diese einfache und sichere Prophylaxe wie auch eine entsprechende Aufklärung anzustreben.

rFVIIa in the treatment of persistent hemorrhage in pediatric patients on ECMO following surgery for congenital heart disease

BACKGROUND: Patients who require extracorporeal membrane oxygenation (ECMO) postsurgery for congenital heart disease (CHD) frequently experience severe bleeding episodes. Whereas recombinant-activated factor VII (rFVIIa) has proven efficacy in counteracting intractable hemorrhage in various scenarios, its use in patients on ECMO is limited by the increased risk for thrombotic events. METHODS: Between December 2004 and January 2006, ECMO was used in 10 pediatric patients following cardiac surgery, of whom seven were treated with rFVIIa because of intractable hemorrhage. Their medical records were reviewed with respect to variations in chest tube output and transfusion requirements, occlusion of or thrombus formation in the ECMO circuit and the occurrence of thromboembolic events. Outcome and rate of ECMO circuit occlusion were compared with historic controls. RESULTS: Three patients died, and four survived (none of the deaths was attributable to thrombus formation or bleeding). All patients were treated with aprotinin prior to and during rFVIIa therapy. Two patients developed an occlusion of the oxygenator, one after receiving co-medication with a FXIII concentrate, another after RBC transfusion in the ECMO system. In two patients, thrombus formation was observed in the ECMO system on inspection after discontinuation. Thromboembolic events were not observed. CONCLUSIONS: Recombinant-activated factor VII in a median dosage of 90 microg.kg(-1) was used in seven pediatric patients on ECMO. Rates of ECMO system occlusions and mortality did not differ from historic controls. Neither the reduction of chest tube output nor the blood product transfusion requirements did reach statistical significance.

[Perspectives in the treatment of subarachnoid-hemorrhage-induced cerebral vasospasm]

Cerebral vasospasm is still the most important cause of death and disability after rupture of intracranial aneurysms. The therapeutic strategies in the treatment of subarachnoid hemorrhage induced vasospasm vasospasm include four groups: 1) prevention of vasospasm; 2) reversion of vasospasm; 3) improvement of cerebral perfusion; and 4) neuroprotection and rescue therapies. Recent experimental studies allowed the design of phase II clinical studies which demonstrated positive results with medications and compounds such as statins (simvastatin and pravastatin) and endothelin-1 receptor antagonists (clasozentan). Moreover, experimental and clinical evidences showed the advantages of early cerebrospinal fluid drainage, intrathecal administration of NO-donors, effects of Ca2+ protein kinase inhibitor (Fasudil) and catecholamines on the cerebral vessels. This review article summarizes the stage of investigation of these medications and therapeutic strategies which will be relevant in the treatment of cerebral vasospasm.

Traumatic extra-axial hemorrhage: correlation of postmortem MSCT, MRI, and forensic-pathological findings

PURPOSE: To evaluate the diagnostic accuracy of in situ postmortem multislice computed tomography (MSCT) and magnetic resonance imaging (MRI) in the detection of primary traumatic extra-axial hemorrhage. MATERIALS AND METHODS: Thirty forensic neurotrauma cases and 10 nontraumatic controls who underwent both in situ postmortem cranial MSCT and MR imaging before autopsy were retrospectively reviewed. Both imaging modalities were analyzed in view of their accuracy, sensitivity, and specificity concerning the detection of extra-axial hemorrhage. Statistical significance was calculated using the McNemar test. kappa values for interobserver agreement were calculated for extra-axial hemorrhage types and to quantify the agreement between both modalities as well as MRI, CT, and forensics, respectively. RESULTS: Analysis of the detection of hemorrhagic localizations showed an accuracy, sensitivity, and specificity of 89%, 82%, and 92% using CT, and 90%, 83%, and 94% using MRI, respectively. MRI was more sensitive than CT in the detection of subarachnoid hemorrhagic localizations (P = 0.001), whereas no significant difference resulted from the detection of epidural and subdural hemorrhagic findings (P = 0.248 and P = 0.104, respectively). Interobserver agreement for all extra-axial hemorrhage types was substantial (CT kappa = 0.76; MRI kappa = 0.77). The agreement of both modalitites was almost perfect (readers 1 and 2 kappa = 0.88). CONCLUSION: CT and MRI are of comparable potential as forensic diagnostic tools for traumatic extra-axial hemorrhage. Not only of forensic, but also of clinical interest is the observation that most thin blood layers escape the radiological evaluation.

Prior statin use, intracranial hemorrhage, and outcome after intra-arterial thrombolysis for acute ischemic stroke

BACKGROUND AND PURPOSE: There are only limited data on whether prior statin use and/or cholesterol levels are associated with intracranial hemorrhage (ICH) and outcome after intra-arterial thrombolysis. The purpose of this study was to evaluate the association of statin pretreatment and cholesterol levels with the overall frequency of ICH, the frequency of symptomatic ICH, and clinical outcome at 3 months. METHODS: We analyzed 311 consecutive patients (mean age, 63 years; 43% women) who received intra-arterial thrombolysis. RESULTS: Statin pretreatment was present in 18%. The frequency of any ICH was 20.6% and of symptomatic ICH 4.8%. Patients with any ICH were more often taking statins (30% versus 15%, P=0.005), more often had atrial fibrillation (45% versus 30%, P=0.016), had more severe strokes (mean National Institute of Health Stroke Scale score 16.5 versus 14.7, P=0.022), and less often good collaterals (16% versus 24%, P=0.001). Patients with symptomatic ICH were more often taking statins (40% versus 15%, P=0.009) and had less often good collaterals (0% versus 24%, P<0.001). Any ICH or symptomatic ICH were not associated with cholesterol levels. After multivariate analysis, the frequency of any ICH remained independently associated with previous statin use (OR, 3.1; 95% CI, 1.53 to 6.39; P=0.004), atrial fibrillation (OR, 2.5; CI, 1.35 to 4.75; P=0.004), National Institutes of Health Stroke Scale score (OR, 1.1; CI, 1.00 to 1.10; P=0.037), and worse collaterals (OR, 1.7; CI, 1.19 to 2.42; P=0.004). There was no association of outcome with prior statin use, total cholesterol level, or low-density lipoprotein cholesterol level. CONCLUSIONS: Prior statin use, but not cholesterol levels on admission, is associated with a higher frequency of any ICH after intra-arterial thrombolysis without impact on outcome.

Transdermal nitroglycerin in patients with subarachnoid hemorrhage

Delayed ischemic neurological deficit (DIND) following cerebral vasospasm remains a cause for high morbidity and mortality in patients with subarachnoid hemorrhage (SAH). There is experimental and clinical evidence of positive effects of nitric oxide (NO) donors on cerebral vasospasm. We therefore analysed the effect of transdermal nitroglycerin in patients with SAH measuring transcranial Doppler velocities (TCD), cerebral blood flow (CBF) and DIND. Nitroglycerin was used in a target dose of 14 microg/kg/h. TCD assessment was performed daily. CBF measurements were done using the perfusion CT-technique. Blood pressure, volume intake and vasopressor administration, were registered. Nine patients were randomly assigned either to the nitroglycerin group (N-group) and eight patients in the control group (C-group). Mean TCD values in the extracranial portion of the internal carotid artery (ICA) were lower in the N-group (p<0.005). Mean TCD in the middle cerebral arteries (MCA) showed no difference. The Lindegaard ratio was higher in the N-group (p<0.04). CBF in the N-group was higher than in the C-group (p<0.03). Even though nitroglycerin reduces blood pressure and lowers ICA TCD-values and increases the Lindegaard ratio, a higher CBF was measured in the N-group. Thus, nitroglycerin influences the cerebral vascular tone and increases CBF. SAH therapy with nitroglycerin is possible without increasing the risk of DIND. The exact timing of onset, duration and reduction of nitroglycerin administration in respect to the appearance of vasospasm may have a strong impact on the success of such a therapy.

Measurement of nitric oxide and brain tissue oxygen tension in patients after severe subarachnoid hemorrhage

OBJECTIVE: Nitric oxide (NO), one of the most powerful endogenous vasodilators, is thought to play a major role in the development of delayed vasospasm in patients with subarachnoid hemorrhage (SAH). However, the role of the production of cerebral NO in patients with SAH is not known. In other SAH studies, NO metabolites such as nitrite and nitrate have been demonstrated to be decreased in cerebrospinal fluid and in plasma. METHODS: In this study, a microdialysis probe was used, along with a multiparameter sensor, to measure NO metabolites, brain tissue oxygen tension, brain tissue carbon dioxide tension, and pH in the cortex of patients with severe SAH who were at risk for developing secondary brain damage and vasospasm. NO metabolites, glucose, and lactate were analyzed in the dialysates to determine the time course of NO metabolite changes and to test the interrelationship between the analytes and clinical variables. RESULTS: Brain tissue oxygen tension was strongly correlated to dialysate nitrate and nitrite (r2 = 0.326; P < 0.001); however, no correlation was noted between brain tissue oxygen tension and NO metabolites in cerebrospinal fluid (r2 = 0.018; P = 0.734). No significant correlation between NO production, brain tissue carbon dioxide tension, and dialysate glucose and lactate was observed. CONCLUSION: Cerebral ischemia and compromised substrate delivery are often responsible for high morbidity rates and poor outcomes after SAH. The relationship between brain tissue oxygen and cerebral NO metabolites that we demonstrate suggests that substrate delivery and NO are linked in the pathophysiology of vasospasm after SAH.