- Email: [email protected]
Dynamic Computed Tomography Angiography in Suspected Brain Death: A Noninvasive Biomarker
Canadian Association of Radiologists Journal 65 (2014) 352e359 www.carjonline.org
Neuroradiology / Neuroradiologie
Dynamic Computed Tomography Angiography in Suspected Brain Death: A Noninvasive Biomarker Santanu Chakraborty, MD, FRCR*, Reem A. Adas, MD Division of Neuro-imaging, Department of Medical Imaging, University of Ottawa, The Ottawa Hospital, Ottawa, Ontario, Canada
Abstract Purpose: Neurologic determination of death or brain death is primarily a clinical diagnosis. This must respect all guarantees required by law and should be determined early to avoid unnecessary treatment and allow organ harvesting for transplantation. Ancillary testing is used in situations in which clinical assessment is impossible or confounded by other factors. Our purpose is to determine the utility of dynamic computed tomographic angiography (dCTA) as an ancillary test for diagnosis of brain death. Materials and Methods: We retrospectively reviewed 13 consecutive patients with suspected brain death in the intensive care unit who had dCTA. Contrast appearance timings recorded from the dCTA data were compared to findings from 15 controls selected from patients who presented with symptoms of acute stroke but showed no stroke in follow-up imaging. Results: The dCTA allows us to reliably assess cerebral blood flow and to record time of individual cerebral vessels opacification. It also helps us to assess the intracranial flow qualitatively against the flow in extracranial vessels as a reference. We compared the time difference between enhancement of the external and internal carotid arteries and branches. In all patients who were brain dead, internal carotid artery enhancement was delayed, which occurred after external carotid artery branches were opacified. Conclusion: In patients with suspected brain death, dCTA reliably demonstrated the lack of cerebral blood flow, with extracranial circulation as an internal reference. Our initial results suggest that inversion of time of contrast appearance between internal carotid artery and external carotid artery branches at the skull base could predict a lack of distal intracranial flow. Resume Objectif : La determination du diagnostic de deces neurologique ou de mort cerebrale est d’abord fondee sur un diagnostic clinique. Le processus doit satisfaire a toutes les exigences de la loi et ^etre effectue rapidement pour eviter les traitements inutiles et assurer le prelevement d’organes a des fins de transplantation. Dans les cas o u une evaluation clinique n’est pas possible, notamment en raison de facteurs confondants, on realise alors des tests auxiliaires. Notre objectif consiste a evaluer si l’angiographie par tomodensitometrie dynamique (ATDM dynamique) constitue un test auxiliaire efficace pour etablir un diagnostic de deces neurologique. Materiel et methodes : De fac¸on retrospective, nous avons examine le dossier de 13 patients consecutifs qui ont semble presenter une mort cerebrale a l’unite de soins intensifs et chez lesquels une ATDM dynamique a ete realisee. Les occurrences d’opacification relevees dans les donnees de l’ATDM dynamique ont ete comparees a celles observees chez 15 patients de reference, selectionnes en raison des sympt^omes d’AVC aigu qu’ils presentaient, mais dont les resultats aux examens d’imagerie de suivi n’ont revele aucun AVC. Resultats : L’ATDM dynamique permet d’etudier la circulation sanguine cerebrale en toute fiabilite et d’enregistrer l’occurrence d’opacification de chaque vaisseau cerebral examine. Elle permet egalement les comparaisons qualitatives entre la circulation intracr^anienne et celle des vaisseaux extra-cr^aniens. Nous avons donc compare l’occurrence d’opacification des branches de l’artere carotide externe a celle des branches de l’artere carotide interne. Dans tous les cas de mort cerebrale, l’opacification de l’artere carotide interne a accuse un retard par rapport a l’opacification des branches de l’artere carotide externe. Conclusion : L’ATDM dynamique peut demontrer en toute fiabilite l’absence de circulation cerebrale chez les patients chez lesquels on soupc¸onne une mort cerebrale lorsqu’on utilise la circulation extra-cr^anienne en guise de reference interne. Les resultats initiaux revelent
* Address for correspondence: Santanu Chakraborty, MD, FRCR, Division of Neuro-imaging, Department of Medical Imaging, University of Ottawa, The Ottawa Hospital, 1053 Carling Avenue, Ottawa, Ontario K1Y 4E9, Canada.
E-mail address: [email protected] (S. Chakraborty).
0846-5371/$ - see front matter Ó 2014 Canadian Association of Radiologists. All rights reserved. http://dx.doi.org/10.1016/j.carj.2014.02.002
Dynamic CTA in suspected brain death / Canadian Association of Radiologists Journal 65 (2014) 352e359
353
qu’une inversion dans la sequence d’opacification des branches de l’artere carotide interne et des branches de l’artere carotide externe situees a la base du cr^ane permettrait de predire l’absence de circulation dans les vaisseaux intracr^aniens distaux. Ó 2014 Canadian Association of Radiologists. All rights reserved. Key Words: Computed tomography; Brain death; Angiogram; Ancillary test; Dynamic computed tomography angiogram
Neurologic determination of death, or brain death (BD), is a clinical diagnosis. The diagnosis of BD must respect all guarantees required by law and should be determined as early as possible to avoid unnecessary treatment and to allow organ harvesting for transplantation [1e3]. Ancillary testing is used when clinical assessment is impossible or confounded by other factors [3,4]. Demonstration of the global absence of cerebral blood flow is the standard finding for confirmation of neurologic determination of death by ancillary testing [5]. In this article, we present the findings of 13 consecutive patients with clinically indeterminate BD who have had a dynamic computed tomographic angiogram (dCTA) to confirm and/or support the clinical diagnosis and to describe a dCTA sign referred to as the ‘‘inversion sign,’’ which is seen in these patients. Material and Methods Patient Selection Our study was approved by the hospital research ethics board. We retrospectively reviewed the charts and radiologic studies of 13 patients in the intensive care unit (age range, 18-81 years; mean age, 46 years; 11 men and 2 women) with severe brain damage and a working diagnosis of BD. In all these cases, a computed tomography angiogram (CTA) was requested by the intensive care unit staff or treating physicians as an ancillary tool for the diagnosis of BD. The decision to switch conventional single-phase CTA to dCTAwas made by consensus, based on our experience in using dCTA in stroke imaging and our knowledge of its utility to provide dynamic blood flow information. Patient demographics, presentation, protocol used, and known confounding factors are listed in Table 1. The final declaration of death was made by repeated clinical testing at a later time point. No catheter cerebral angiogram or nuclear medicine testing was performed in this clinical cohort. Contrast appearance timings recorded from the dCTA data were recorded and compared with findings from 15 control patients (age range, 46-91 years; mean age, 65 years; 7 men and 8 women). The control group consisted of patients who presented with acute stroke and were scanned as per stroke protocol but during follow-up imaging demonstrated no stroke. They were all free from significant carotid artery disease by the North American Symptomatic Carotid Endarterectomy Trial (NASCET) criteria. dCTA protocol By using a 320-row volume CT scanner (Aquilion ONE; Toshiba, Otawara, Japan), dCTA of the whole brain was performed. This system uses 320 ultra-higheresolution detector
rows (0.5 mm in width) to image the entire brain in a single gantry rotation. By using this CT scanner, we are able to get whole-brain CT perfusion data. From the same data set, dCTA images are reconstructed [6], which enables us to analyse the blood flow in the entire cranial circulation in a noninvasive way with high spatial and temporal resolution. Compared with oldergeneration CT scanners that are able to generate a CTA image of the brain in a predetermined time point (snapshot views, standard CTA), dCTA can monitor the contrast flow of the intra- and extracranial circulation during the total scanning time (80 seconds in the BD protocol). The predetermined time point used in an older CT scanner is often unreliable in these patients due to the abnormal and/or delayed flow. A dCTA eliminates the presumption of optimal timing used in conventional CTA because it monitors the flow of contrast in the cerebral vasculature during the whole scanning period (Figure 1). A dCTA permits us to acquire a time series of bone subtracted or nonsubtracted CTA images of the whole head in a noninvasive way, which provides excellent temporal flow information. In our BD imaging protocol, whole-brain imaging is performed in 23 volume sets. It begins at 7 seconds from the start of injection of contrast (the first volume is used as a mask for subtraction), then a single volume set is acquired every 2 seconds from 10-35 seconds (total 13 volumes), followed by every 5 seconds from 40-80 seconds (9 volumes) after the initial contrast bolus (Figure 1). Only 40 mL of intravenous contrast is used. The radiation dose for this scan is 5.3 mSv, equivalent to 2 years of background radiation. The stroke protocol is performed with similar time intervals to the BD protocol but has fewer volume sets (19 volume sets), and the total scanning period is only 60 seconds. The generation of CT perfusion maps are dependent on significant intracranial flow and a detectable arterial input function. In our BD cases, the processing software was unable to generate perfusion maps. Data Analysis Time-resolved dCTA images were generated by using a Vitrea workstation (Toshiba, Canada). The imaging studies of both groups, patients with BD and control patients, were reviewed by a single observer (S.C.) and the timings of contrast arrival in the intra- and extracranial vessels were recorded from each case. We recorded time of contrast arrival in the distal cervical (at skull base) and supraclinoid segments of both internal carotid arteries (ICA) and the superficial temporal branch of external carotid arteries (ECA) at the skull base and vertex. In addition, bilateral anterior cerebral arteries (A1 and A3 segments), middle cerebral arteries (M1 and cortical segments), and posterior cerebral arteries (P2 segments) also were
354
S. Chakraborty, R. A. Adas / Canadian Association of Radiologists Journal 65 (2014) 352e359
Table 1 Patients demographics, presentation, dCTA protocol used and known confounding factors are shown Patient no.
Age, y
Sex
Pathology and/or precipitating event
Confounding factors if known
Time to scan after precipitating event
dCTA protocol
1
52
M
After laminectomy, hyperkalemia, unwitnessed cardiac arrest Massive left MCA stroke and hemorrhage after tPA
No brainstem reflex but spinal reflex present Continuing deterioration, assessment difficult Hypothermia, prolonged sedation, difficult eye assessment due to injury Sedative overdose
14 h
Stroke
2
58
M
35 h
BD
3
26
M
Trauma, motorcycle collision, intracranial hemorrhage
10 d, 22 h
Stroke
Repeated clinical assessment and withdrawal of life support
4
18
M
5
58
M
Drug overdose, was found in VSA Found in a hotel room, intracranial hemorrhage
6h
BD
Opioid infusion
24 h
Stroke
Not documented
13 h
BD
Severe multiorgan failure and refractory shock Repeated clinical assessment and withdrawal of life support Discussion with family, withdrawal of life support
6
61
F
7
37
F
Coma
12 h
BD
Discussion with family, withdrawal of life support
8
49
M
Alcohol and sedatives
14 h
BD
Brainstem reflexes
4h
BD
Generalized seizurelike activity and sedatives
2 d, 8 h
BD
Discussion with family, organs donated Repeated clinical assessment and withdrawal of life support Preserved intracranial flow, patient stopped spontaneous breathing on d 4
9
56
M
10a
58
M
11
25
M
Spinal reflexes
13 h
BD
Declared dead after repeated clinical examination
12
81
M
Unreliable corneal reflexes and vasopressors Spinal reflexes
12 h
BD
2 d, 5 h
BD
Discussion with family, withdrawal of life support, organs donated Repeated clinical test to determine BD, organs donated
13
19
M
Cardiac arrest in restaurant, subarachnoid hemorrhage due to giant left MCA aneurysm, rebleeding Postoperative day 14 after hip replacement, cardiac arrest, pulmonary embolism, and seizures Fall from stairs, large subdural hematoma and GCS 3 Ruptured untreatable basilar aneurysm and SAH, found in VSA Cardiac arrest while running in a marathon, significant downtime, multiple coronary artery stenosis, coma, and generalized myoclonus Polytrauma, high-speed singlevehicle rollover, intracranial injury Large left subdural hematoma secondary to fall, on warfarin Self-inflicted injury to pharynx and trachea, followed by cardiac arrest
Follow-up Repeated clinical assessment and withdrawal of life support Discussion with family, organs donated
BD ¼ brain death; dCTA ¼ dynamic computed tomographic angiography; GCS ¼ Glasgow Coma Scale; MCA ¼ middle cerebral arteries ; SAH ¼ subarachnoid hemorhage; tPA ¼ tissue plasminogen activator; VSA ¼ vital signs absent. a This patient had normal intracranial flow.
recorded. Furthermore, the vertebral arteries (V3 and V4 segments) as well as the basilar artery, the internal cerebral veins, and the vein of Galen were analysed. Times recorded from the ICA and ECA branches at skull base were compared in both patient groups by using the type 3 t test. A P value of less than (P < .05) was considered statistically significant. Results The dCTA was technically successful in all the cases. Twelve of the 13 patients with suspected BD demonstrated a lack of normal intracranial flow, whereas 1 patient (no. 10) had preserved intracranial flow at the time of the scan. All 12
patients had delayed ICA enhancement (Figure 2B, C), which, when present, occurred after the time of enhancement of the ECA (measured at the level of the skull base). This is opposite to the time of appearance in the control group and is described as an inversion of the time of contrast appearance between ICA and ECA. The patient with persistent intracranial flow had severe brain oedema and was believed to be in status epilepticus (patient collapsed during a marathon and was resuscitated). The patient stopped spontaneous breathing 4 days after the scan, and life support was withdrawn at that time. The average enhancement time for both ECAs in 12 patients with BD was 16 seconds (mean [standard deviation (SD)], 16.08 3.85 seconds) when excluding the patient
Dynamic CTA in suspected brain death / Canadian Association of Radiologists Journal 65 (2014) 352e359
355
Figure 1. The timing diagram explains the acquisition of whole-brain volumes at each time point starting at 7 seconds from the time of the start of contrast injection. This volume uses 300 mA and is used as a mask for subtraction. Next, in the arterial phase (10-35 seconds) whole-brain volumes are acquired every 2 seconds. In the venous phase (40-80 seconds), volumes are acquired at 5-second intervals. By using each individual or all volumes, we could assess flow of contrast in intra- and extracranial circulation during the whole scanning duration (80 seconds). In contrast, older computed tomography scanners could only obtain a snapshot view of the brain circulation at a predetermined time point, which could miss the contrast, depending on the circulation time in an individual patient.
Figure 2. The upper panel (A-C) shows images from a 52-year-old man with diabetes who had postoperative cardiac arrest. (A) Noncontrast computed tomography (CT) of the head, showing diffuse loss of grey matter density lower than the white matter, so called ‘‘reversal sign,’’ a sign of diffuse ischemia. (B) A dynamic CTangiography (dCTA) at 22 seconds, showing good flow in the external carotid branches to the level of vertex but no evidence of flow in internal carotid arteries (ICA) and intracranial branches. (C) At 60 seconds, there is meager contrast flow noted in the right ICA, only to the level of the skull base; please note the scalp arteries have completely cleared of contrast. The lower panel (D, E) shows dCTA images from a 17-year-old man with a suicidal overdose (Video 1). (D) Frontal view dCTA at 24 seconds, showing delayed filling of the basilar artery and bilateral ICAs at the skull base but no filling of the distal branches. Complete filling of the superficial temporal and external carotid branches are noted, which is helpful as an internal reference. (E) Frontal view at 28 seconds, showing minimal filling of the right M1 segment and no filling of the bilateral distal middle cerebral arteries and posterior cerebral arteries branches. (F) A normal control with simultaneous filling of intra- and extracranial branches.
356
S. Chakraborty, R. A. Adas / Canadian Association of Radiologists Journal 65 (2014) 352e359
Table 2 Contrast appearance times in ICA and ECA at the level of the skull base in patient with brain death Patient no. (n ¼ 13) 1 2 3 4 5 6 7 8 9 10b 11 12 13
Right ICA, s 40 80a 25 17 20 25 23 23 15 11 33 40 50
Right ECA, s 17 19 17 15 9 20 15 19 13 13 13 23 13
Left ICA, s a
80 40 80a 17 25 31 23 23 15 11 21 40 33
Left ECA, s 17 19 17 15 9 20 15 19 13 13 13 23 13
ECA ¼ external carotid artery; ICA ¼ internal carotid artery. a Values recorded at 80 s in patients who, at the end of the study, showed no enhancement in their ICA. b The patient with preserved intracranial flow.
with preserved intracranial flow. The mean right ICA enhancement time was 33 seconds (mean [SD], 32.58 18.3 seconds). The mean left ICA enhancement time was 36 seconds (mean [SD], 35.67 22.2 seconds). Both right and left ICAs and ECAs values for patients with BD are detailed in Table 2. All control patients showed bilateral ICAs enhancement, which occurred either simultaneously or before the ECAs. Bilateral ICA and ECA enhancement times are shown in Table 3. When using the unequal variance ‘‘type 3’’ t test to compare the enhancement times of each vessel in both groups, there were no statistically significant differences (P > .05) in the enhancement times of both ECAs (P ¼ .29 for the right, and P ¼ .22 for the left). For the ICAs, there was a statistical difference in the enhancement times between the 2 groups; P < .004 for the right and P ¼ .005 for Table 3 Contrast appearance times in ICA and ECA at the level of the skull base in the control group Patient no. (n ¼ 15)
Right ICA, s
Right ECA, s
Left ICA, s
Left ECA, s
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
13 17 13 13 19 19 11 15 17 15 11 11 11 7 7
13 19 13 13 21 21 11 17 19 17 13 19 11 13 7
13 17 13 13 19 19 11 15 17 15 11 11 11 7 7
13 17 13 13 21 21 11 17 19 17 13 11 11 7 7
ECA ¼ external carotid artery; ICA ¼ internal carotid artery.
the left. The graphs in Figure 3 (upper panel) show the relationship of the timing between the left ECA and ICA in both patients with BD and the control group. The lower panel bar chart shows the mean contrast arrival time between the right ICA and ECA in both groups. Note the inversed relationship of ECA and ICA enhancement timings in the BD group compared with controls. The average delay time of the ICA enhancement compared with the ECA in patients with BD was 16.5 seconds on the right and 19.6 seconds on the left. For our control group, the mean delay time was e1.07 seconds on the right and e0.8 seconds on the left. Trickling of contrast (described as delayed, slow flow) (Figure 2D, E) was noted in the supraclinoid ICAs bilaterally in 7 patients, 6 patients also had similar findings in either M1 or A1 segments, and 4 patients demonstrated faint contrast enhancement in their intracranial vertebralebasilar arteries. Minimal contrast was noted in the venous system of 3 patients on delayed images. The presence of contrast in intracranial arteries can be confusing, especially when using a static CTA technique. In all of these patients, there was absent qualitative intracranial flow (best appreciated in dCTA images in a movie format) (Video 1). Discussion In this study, we describe the use of dCTA as an ancillary test in patients with suspected BD to show the lack of intracranial flow. This noninvasive technique provides a means to assess qualitative intracranial flow compared with the extracranial arteries as an internal reference. The ability for continuous flow assessment gives us extra diagnostic confidence in the presence of ‘‘stasis filling’’ [7] or trickling of contrast. We also describe the phenomenon of inversion of ICA and ECA enhancement times at the level of the skull base in patients with BD as an ‘‘inversion sign.’’ In our series, this correlates with a lack of global intracranial flow. This finding may be helpful in the assessment of dCTA generated from the limited coverage (at the skull base) perfusion data by using other standard multislice CT scanners. The BD [8] definition has evolved over the past few decades from being a coma depasse in 1959 [9] to the modernday definition developed by the American Academy of Neurology [10] in 1994. Simply, deep coma, lack of brain stem reflexes, and spontaneous breathing composes the triad for diagnosis. Many confounding factors may exist, which limit proper assessment, examples of which are mentioned in Table 1. Ancillary tests are used to provide necessary support to the clinical diagnosis of BD. Cerebral 4-vessel catheter angiogram remains the criterion standard ancillary test of choice in confirming the diagnosis of BD, followed by hexamethylpropylene amine oximeesingle photon emission computed tomography radionuclide angiography. Cerebral circulatory arrest occurs when the intracranial pressure exceeds arterial inflow pressure. The lack of cerebral vessel enhancement beyond the foramen magnum and carotid siphons is usually diagnostic. Due to the limitations and
Dynamic CTA in suspected brain death / Canadian Association of Radiologists Journal 65 (2014) 352e359
357
Figure 3. Upper panel graphs, showing the relationship of the timing of the left external carotid arteries (ECA) and internal carotid arteries (ICA) contrast appearance in both subjects who are brain dead and the control subjects. There is inversion of ECA and ICA timings in the brain death group compared with the control group. The lower panel bar charts show mean contrast appearances in the right ECAs and the ICAs in both groups.
invasiveness of these techniques, attempts to validate the use of cross-sectional imaging have been made. The use of CT to diagnose BD was proposed as early as 1978 [2,11] and then again in 1985 as an alternative to cerebral angiography [12]. The Medical and Scientific Council of the French Transplant Establishment recognizes CTA as a reliable technique for the diagnosis of BD [13], after accepting the method of Dupas et al [1] of the 7-point CTA score for BD diagnosis in 1998. Newer studies [14e17], however, showed less sensitivity and reported that intracranial opacification on CTA was present in a substantial number of patients (11%48%). In 2006, Combes et al [15] questioned the reliability of CTA in BD. They found that CTA did not confirm the clinical diagnosis of BD in 13 of the 43 patients who were clinically BD, whereas findings of the catheter cerebral angiography confirmed the clinical diagnosis of BD in all 43 patients. A more recent study, in 2009, by Frampas et al [16] proposed the use of a simplified 4-point CTA score, based on the lack of opacification of the middle cerebral arteries and internal cerebral veins, which they concluded has a sensitivity of 85.7% and specificity of 100%. A recent evidence-based guideline update by Wijdicks et al (4) concluded that, although crosssectional vascular imaging (CTA, magnetic resonance angiography) is currently used in clinical practice, further studies are needed to prove their reliability. Limitations with Standard or 2-Phase CTA In all the studies mentioned above, a 2-phase contrast CTA (early arterial and delayed venous) was used. This
technique provides a static volume of brain and vessels images performed during 2 specified points in time (snapshot views). The most common limitation encountered in patients who are BD is called ‘‘stasis filling’’ or contrast visualization in delayed images in the proximal intracranial arteries and veins. This is well documented in the BD literature [7,18] and also is noted in the catheter angiographic studies [19,20]. This finding can be more pronounced in CTA due to inherent higher contrast resolution of CTs compared with subtraction radiography. In a more recent study [21], opacification of the proximal middle cerebral arteries segment was demonstrated in arterial series in 19 cases and in venous (55 seconds after contrast injection) series in 37 of a total 63 cases. Opacification of other intracranial arteries was visible in only 3 patients in the arterial series but 22 patients in venous series. Even minimal contrast filling in deep cerebral veins was noted in 3 patients in our series. We believe that this ‘‘stasis flow’’ does not indicate true cerebral blood flow and does not contradict the BD diagnosis. But this phenomenon causes significant diagnostic confusion and is likely a reason why CTA is not widely accepted for BD diagnosis. Benefits of dCTA A dCTA, in contrast to traditional CTA, has the advantage of providing visualization of dynamic blood flow information, which makes it possible to analyse the pattern of flow qualitatively (Video 1). So, the presence of a small amount of contrast in the proximal intracranial arteries (as
358
S. Chakraborty, R. A. Adas / Canadian Association of Radiologists Journal 65 (2014) 352e359
in 7 patients of 12 in this series) does not sway us to confidently report the lack of global intracranial flow. This is the most beneficial aspect of dCTA in these cases. It also images the entire cranial vasculature simultaneously and allows comparison with the extracranial arterial flow, which can be used as an internal reference. Aside from the intravenous contrast injection, it is a noninvasive and safe technique. A dCTA is considerably less time consuming and technically less demanding than cerebral angiography for both patients and radiologists. This technique does not exclusively need a whole-brain CT scanner for data acquisition. Other techniques, such as volume helical shuttle mode [22] or the table toggle technique for extended coverage of brain perfusion, can generate similar dCTA. It requires a significantly low volume of contrast (40 mL), which reduces the risk of renal dysfunction and which might compromise organ harvesting. In all but 1 of the patients with suspected BD, enhancement of the ECAs occurred before the ICAs. This particular patient had a preserved normal intracranial flow at the time of scanning, although he was suspected clinically to have had severe brain damage (due to anoxic injury) or status epilepticus. Spontaneous breathing stopped 4 days after the scan, and life support was withdrawn at that time. One of the characteristic signs of BD in noncontrast CT is the ‘‘reversal sign’’ (Figure 2A), and this is well described in the literature [23], in which a reversal of density between grey and white matter is seen in cases of patients with diffuse anoxic cerebral injury. It is caused by the loss of grey matter density due to cytotoxic oedema after cell death, which results in a relative low density of grey matter compared with the adjacent white matter. In this study, we describe the ‘‘inversion sign’’ in dCTA related to inversion of time of appearance of contrast in ICAs and ECAs in cases of BD (Figure 2B-E). The delayed abnormal filling of the ICAs observed in the BD group did not occur in any of the control group patients. This finding led us to conclude that inversion of comparative flow between the ICAs and ECAs at the level of the skull base may predict the lack of qualitative intracranial flow as seen in our study group. Also, the patient in the BD group with preserved intracranial flow had no inversion of contrast timing at the skull base, thus confirming the prediction for distal global intracranial flow. Application of this concept could be used by limited coverage perfusion acquisition at the skull base in standard multislice CT scanners to predict a lack of global intracranial flow. Limitations of This Study This was a clinical observational study and has the limitations of selection bias, a small number of subjects, and a variety of initiating causes as well as a variable time range from the clinical suspicion of BD to the time that the scans were performed. Also, catheter angiography or radionuclide studies were not performed in this clinical cohort. Prospective larger studies, with comparison with
catheter angiography or radionuclide angiography as criterion standards, are required in the future for validation of this technique. Conclusion In contrast to traditional CTA, dCTA has the advantage of providing visualization of dynamic blood flow, which makes it possible to analyse the pattern of intracranial flow qualitatively, which allows us to confidently report the lack of global intracranial flow and not be swayed by the presence of a small amount of contrast in the proximal intracranial arteries. This advantage overcomes the limitation of ‘‘stasis filling’’ seen in standard 2-phase CTA. In addition, the simultaneous imaging of the entire cranial vasculature provided by this technique allows comparison with the extracranial arterial flow. This advantage led us to observe and describe an ‘‘inversion sign,’’ which refers to the inversion of flow pattern between the ICAs and the ECAs in patients with brain death due to delayed ICA flow. This finding strengthens the utility of noninvasive dCTA as an ancillary test for diagnosing brain death. References [1] Dupas B, Gayet-Delacroix M, Villers D, et al. Diagnosis of brain death using two-phase spiral ct. AJNR Am J Neuroradiol 1998;19:641e7. [2] Rappaport ZH, Epstein F. Computerized axial tomography in the preoperative evaluation of posterior fossa tumors in children. Childs Brain 1978;4:170e9. [3] Shemie SD. Brain arrest to neurological determination of death to organ utilization: the evolution of hospital-based organ donation strategies in Canada. Can J Anaesth 2006;53:747e52. [4] Wijdicks EF, Varelas PN, Gronseth GS, et al. Evidence-based guideline update: determining brain death in adults: report of the Quality Standards Subcommittee of the American Academy of Neurology. Neurology 2010;74:1911e8. [5] Shemie SD, Lee D, Sharpe M, et al. Brain blood flow in the neurological determination of death: Canadian expert report. Can J Neurol Sci 2008;35:140e5. [6] Salomon EJ, Barfett J, Willems PW, et al. Dynamic CT angiography and CT perfusion employing a 320-detector row CT: protocol and current clinical applications. Klin Neuroradiol 2009;19:187e96. [7] van der Lugt A. Imaging tests in determination of brain death. Neuroradiology 2010;52:945e7. [8] A definition of irreversible coma. Report of the Ad Hoc Committee of the Harvard Medical School to examine the definition of brain death. JAMA 1968;205:337e40. [9] Mollaret P, Goulon M. The depassed coma (preliminary memoir) [in French]. Rev Neurol (Paris) 1959;101:3e15. [10] Practice parameters for determining brain death in adults (summary statement). The Quality Standards Subcommittee of the American Academy of Neurology. Neurology 1995;45:1012e4. [11] Rangel RA. Computerized axial tomography in brain death. Stroke 1978;9:597e8. [12] Planitzer J, Zschenderlein R, Schulze HA, et al. Computer tomography studies in irreversible cerebral function loss (brain death) [in German with English abstract]. Psychiatr Neurol Med Psychol (Leipz) 1985;37: 509e17. [13] Recommendations du conseil medical et scientifique de l’etablissement franc¸ais des greffes. Les techniques d’angiographie dans le diagnostic de la mort encephalique. Deliberation n 2000.12 en seance du 21 novembre 2000, dossier s3 d2.
Dynamic CTA in suspected brain death / Canadian Association of Radiologists Journal 65 (2014) 352e359 [14] Leclerc X, Taschner CA, Vidal A, et al. The role of spiral CT for the assessment of the intracranial circulation in suspected brain-death. J Neuroradiol 2006;33:90e5. [15] Combes JC, Chomel A, Ricolfi F, et al. Reliability of computed tomographic angiography in the diagnosis of brain death. Transplant Proc 2007;39:16e20. [16] Frampas E, Videcoq M, de Kerviler E, et al. Ct angiography for brain death diagnosis. AJNR Am J Neuroradiol 2009;30:1566e70. [17] Escudero D, Otero J, Marques L, et al. Diagnosing brain death by CT perfusion and multislice CT angiography. Neurocrit Care 2009;11: 261e71. [18] Hassler W, Steinmetz H, Pirschel J. Transcranial Doppler study of intracranial circulatory arrest. J Neurosurg 1989;71:195e201.
359
[19] Kricheff II , Pinto RS, George AE, et al. Angiographic findings in brain death. Ann N Y Acad Sci 1978;315:168e83. [20] Bradac GB, Simon RS. Angiography in brain death. Neuroradiology 1974;7:25e8. [21] Welschehold S, Kerz T, Boor S, et al. Detection of intracranial circulatory arrest in brain death using cranial CT-angiography. Eur J Neurol 2013;20:173e9. [22] Volume helical shuttle CT perfusion identifies stroke patients. Available at http://www.auntminnie.com/index.aspx?sec¼ser&sub¼def&pag¼dis& ItemID¼90309. Accessed March 12, 2014. [23] Han BK, Towbin RB, De Courten-Myers G, et al. Reversal sign on CT: effect of anoxic/ischemic cerebral injury in children. AJNR Am J Neuroradiol 1989;10:1191e8.
Neuroradiology / Neuroradiologie
Dynamic Computed Tomography Angiography in Suspected Brain Death: A Noninvasive Biomarker Santanu Chakraborty, MD, FRCR*, Reem A. Adas, MD Division of Neuro-imaging, Department of Medical Imaging, University of Ottawa, The Ottawa Hospital, Ottawa, Ontario, Canada
Abstract Purpose: Neurologic determination of death or brain death is primarily a clinical diagnosis. This must respect all guarantees required by law and should be determined early to avoid unnecessary treatment and allow organ harvesting for transplantation. Ancillary testing is used in situations in which clinical assessment is impossible or confounded by other factors. Our purpose is to determine the utility of dynamic computed tomographic angiography (dCTA) as an ancillary test for diagnosis of brain death. Materials and Methods: We retrospectively reviewed 13 consecutive patients with suspected brain death in the intensive care unit who had dCTA. Contrast appearance timings recorded from the dCTA data were compared to findings from 15 controls selected from patients who presented with symptoms of acute stroke but showed no stroke in follow-up imaging. Results: The dCTA allows us to reliably assess cerebral blood flow and to record time of individual cerebral vessels opacification. It also helps us to assess the intracranial flow qualitatively against the flow in extracranial vessels as a reference. We compared the time difference between enhancement of the external and internal carotid arteries and branches. In all patients who were brain dead, internal carotid artery enhancement was delayed, which occurred after external carotid artery branches were opacified. Conclusion: In patients with suspected brain death, dCTA reliably demonstrated the lack of cerebral blood flow, with extracranial circulation as an internal reference. Our initial results suggest that inversion of time of contrast appearance between internal carotid artery and external carotid artery branches at the skull base could predict a lack of distal intracranial flow. Resume Objectif : La determination du diagnostic de deces neurologique ou de mort cerebrale est d’abord fondee sur un diagnostic clinique. Le processus doit satisfaire a toutes les exigences de la loi et ^etre effectue rapidement pour eviter les traitements inutiles et assurer le prelevement d’organes a des fins de transplantation. Dans les cas o u une evaluation clinique n’est pas possible, notamment en raison de facteurs confondants, on realise alors des tests auxiliaires. Notre objectif consiste a evaluer si l’angiographie par tomodensitometrie dynamique (ATDM dynamique) constitue un test auxiliaire efficace pour etablir un diagnostic de deces neurologique. Materiel et methodes : De fac¸on retrospective, nous avons examine le dossier de 13 patients consecutifs qui ont semble presenter une mort cerebrale a l’unite de soins intensifs et chez lesquels une ATDM dynamique a ete realisee. Les occurrences d’opacification relevees dans les donnees de l’ATDM dynamique ont ete comparees a celles observees chez 15 patients de reference, selectionnes en raison des sympt^omes d’AVC aigu qu’ils presentaient, mais dont les resultats aux examens d’imagerie de suivi n’ont revele aucun AVC. Resultats : L’ATDM dynamique permet d’etudier la circulation sanguine cerebrale en toute fiabilite et d’enregistrer l’occurrence d’opacification de chaque vaisseau cerebral examine. Elle permet egalement les comparaisons qualitatives entre la circulation intracr^anienne et celle des vaisseaux extra-cr^aniens. Nous avons donc compare l’occurrence d’opacification des branches de l’artere carotide externe a celle des branches de l’artere carotide interne. Dans tous les cas de mort cerebrale, l’opacification de l’artere carotide interne a accuse un retard par rapport a l’opacification des branches de l’artere carotide externe. Conclusion : L’ATDM dynamique peut demontrer en toute fiabilite l’absence de circulation cerebrale chez les patients chez lesquels on soupc¸onne une mort cerebrale lorsqu’on utilise la circulation extra-cr^anienne en guise de reference interne. Les resultats initiaux revelent
* Address for correspondence: Santanu Chakraborty, MD, FRCR, Division of Neuro-imaging, Department of Medical Imaging, University of Ottawa, The Ottawa Hospital, 1053 Carling Avenue, Ottawa, Ontario K1Y 4E9, Canada.
E-mail address: [email protected] (S. Chakraborty).
0846-5371/$ - see front matter Ó 2014 Canadian Association of Radiologists. All rights reserved. http://dx.doi.org/10.1016/j.carj.2014.02.002
Dynamic CTA in suspected brain death / Canadian Association of Radiologists Journal 65 (2014) 352e359
353
qu’une inversion dans la sequence d’opacification des branches de l’artere carotide interne et des branches de l’artere carotide externe situees a la base du cr^ane permettrait de predire l’absence de circulation dans les vaisseaux intracr^aniens distaux. Ó 2014 Canadian Association of Radiologists. All rights reserved. Key Words: Computed tomography; Brain death; Angiogram; Ancillary test; Dynamic computed tomography angiogram
Neurologic determination of death, or brain death (BD), is a clinical diagnosis. The diagnosis of BD must respect all guarantees required by law and should be determined as early as possible to avoid unnecessary treatment and to allow organ harvesting for transplantation [1e3]. Ancillary testing is used when clinical assessment is impossible or confounded by other factors [3,4]. Demonstration of the global absence of cerebral blood flow is the standard finding for confirmation of neurologic determination of death by ancillary testing [5]. In this article, we present the findings of 13 consecutive patients with clinically indeterminate BD who have had a dynamic computed tomographic angiogram (dCTA) to confirm and/or support the clinical diagnosis and to describe a dCTA sign referred to as the ‘‘inversion sign,’’ which is seen in these patients. Material and Methods Patient Selection Our study was approved by the hospital research ethics board. We retrospectively reviewed the charts and radiologic studies of 13 patients in the intensive care unit (age range, 18-81 years; mean age, 46 years; 11 men and 2 women) with severe brain damage and a working diagnosis of BD. In all these cases, a computed tomography angiogram (CTA) was requested by the intensive care unit staff or treating physicians as an ancillary tool for the diagnosis of BD. The decision to switch conventional single-phase CTA to dCTAwas made by consensus, based on our experience in using dCTA in stroke imaging and our knowledge of its utility to provide dynamic blood flow information. Patient demographics, presentation, protocol used, and known confounding factors are listed in Table 1. The final declaration of death was made by repeated clinical testing at a later time point. No catheter cerebral angiogram or nuclear medicine testing was performed in this clinical cohort. Contrast appearance timings recorded from the dCTA data were recorded and compared with findings from 15 control patients (age range, 46-91 years; mean age, 65 years; 7 men and 8 women). The control group consisted of patients who presented with acute stroke and were scanned as per stroke protocol but during follow-up imaging demonstrated no stroke. They were all free from significant carotid artery disease by the North American Symptomatic Carotid Endarterectomy Trial (NASCET) criteria. dCTA protocol By using a 320-row volume CT scanner (Aquilion ONE; Toshiba, Otawara, Japan), dCTA of the whole brain was performed. This system uses 320 ultra-higheresolution detector
rows (0.5 mm in width) to image the entire brain in a single gantry rotation. By using this CT scanner, we are able to get whole-brain CT perfusion data. From the same data set, dCTA images are reconstructed [6], which enables us to analyse the blood flow in the entire cranial circulation in a noninvasive way with high spatial and temporal resolution. Compared with oldergeneration CT scanners that are able to generate a CTA image of the brain in a predetermined time point (snapshot views, standard CTA), dCTA can monitor the contrast flow of the intra- and extracranial circulation during the total scanning time (80 seconds in the BD protocol). The predetermined time point used in an older CT scanner is often unreliable in these patients due to the abnormal and/or delayed flow. A dCTA eliminates the presumption of optimal timing used in conventional CTA because it monitors the flow of contrast in the cerebral vasculature during the whole scanning period (Figure 1). A dCTA permits us to acquire a time series of bone subtracted or nonsubtracted CTA images of the whole head in a noninvasive way, which provides excellent temporal flow information. In our BD imaging protocol, whole-brain imaging is performed in 23 volume sets. It begins at 7 seconds from the start of injection of contrast (the first volume is used as a mask for subtraction), then a single volume set is acquired every 2 seconds from 10-35 seconds (total 13 volumes), followed by every 5 seconds from 40-80 seconds (9 volumes) after the initial contrast bolus (Figure 1). Only 40 mL of intravenous contrast is used. The radiation dose for this scan is 5.3 mSv, equivalent to 2 years of background radiation. The stroke protocol is performed with similar time intervals to the BD protocol but has fewer volume sets (19 volume sets), and the total scanning period is only 60 seconds. The generation of CT perfusion maps are dependent on significant intracranial flow and a detectable arterial input function. In our BD cases, the processing software was unable to generate perfusion maps. Data Analysis Time-resolved dCTA images were generated by using a Vitrea workstation (Toshiba, Canada). The imaging studies of both groups, patients with BD and control patients, were reviewed by a single observer (S.C.) and the timings of contrast arrival in the intra- and extracranial vessels were recorded from each case. We recorded time of contrast arrival in the distal cervical (at skull base) and supraclinoid segments of both internal carotid arteries (ICA) and the superficial temporal branch of external carotid arteries (ECA) at the skull base and vertex. In addition, bilateral anterior cerebral arteries (A1 and A3 segments), middle cerebral arteries (M1 and cortical segments), and posterior cerebral arteries (P2 segments) also were
354
S. Chakraborty, R. A. Adas / Canadian Association of Radiologists Journal 65 (2014) 352e359
Table 1 Patients demographics, presentation, dCTA protocol used and known confounding factors are shown Patient no.
Age, y
Sex
Pathology and/or precipitating event
Confounding factors if known
Time to scan after precipitating event
dCTA protocol
1
52
M
After laminectomy, hyperkalemia, unwitnessed cardiac arrest Massive left MCA stroke and hemorrhage after tPA
No brainstem reflex but spinal reflex present Continuing deterioration, assessment difficult Hypothermia, prolonged sedation, difficult eye assessment due to injury Sedative overdose
14 h
Stroke
2
58
M
35 h
BD
3
26
M
Trauma, motorcycle collision, intracranial hemorrhage
10 d, 22 h
Stroke
Repeated clinical assessment and withdrawal of life support
4
18
M
5
58
M
Drug overdose, was found in VSA Found in a hotel room, intracranial hemorrhage
6h
BD
Opioid infusion
24 h
Stroke
Not documented
13 h
BD
Severe multiorgan failure and refractory shock Repeated clinical assessment and withdrawal of life support Discussion with family, withdrawal of life support
6
61
F
7
37
F
Coma
12 h
BD
Discussion with family, withdrawal of life support
8
49
M
Alcohol and sedatives
14 h
BD
Brainstem reflexes
4h
BD
Generalized seizurelike activity and sedatives
2 d, 8 h
BD
Discussion with family, organs donated Repeated clinical assessment and withdrawal of life support Preserved intracranial flow, patient stopped spontaneous breathing on d 4
9
56
M
10a
58
M
11
25
M
Spinal reflexes
13 h
BD
Declared dead after repeated clinical examination
12
81
M
Unreliable corneal reflexes and vasopressors Spinal reflexes
12 h
BD
2 d, 5 h
BD
Discussion with family, withdrawal of life support, organs donated Repeated clinical test to determine BD, organs donated
13
19
M
Cardiac arrest in restaurant, subarachnoid hemorrhage due to giant left MCA aneurysm, rebleeding Postoperative day 14 after hip replacement, cardiac arrest, pulmonary embolism, and seizures Fall from stairs, large subdural hematoma and GCS 3 Ruptured untreatable basilar aneurysm and SAH, found in VSA Cardiac arrest while running in a marathon, significant downtime, multiple coronary artery stenosis, coma, and generalized myoclonus Polytrauma, high-speed singlevehicle rollover, intracranial injury Large left subdural hematoma secondary to fall, on warfarin Self-inflicted injury to pharynx and trachea, followed by cardiac arrest
Follow-up Repeated clinical assessment and withdrawal of life support Discussion with family, organs donated
BD ¼ brain death; dCTA ¼ dynamic computed tomographic angiography; GCS ¼ Glasgow Coma Scale; MCA ¼ middle cerebral arteries ; SAH ¼ subarachnoid hemorhage; tPA ¼ tissue plasminogen activator; VSA ¼ vital signs absent. a This patient had normal intracranial flow.
recorded. Furthermore, the vertebral arteries (V3 and V4 segments) as well as the basilar artery, the internal cerebral veins, and the vein of Galen were analysed. Times recorded from the ICA and ECA branches at skull base were compared in both patient groups by using the type 3 t test. A P value of less than (P < .05) was considered statistically significant. Results The dCTA was technically successful in all the cases. Twelve of the 13 patients with suspected BD demonstrated a lack of normal intracranial flow, whereas 1 patient (no. 10) had preserved intracranial flow at the time of the scan. All 12
patients had delayed ICA enhancement (Figure 2B, C), which, when present, occurred after the time of enhancement of the ECA (measured at the level of the skull base). This is opposite to the time of appearance in the control group and is described as an inversion of the time of contrast appearance between ICA and ECA. The patient with persistent intracranial flow had severe brain oedema and was believed to be in status epilepticus (patient collapsed during a marathon and was resuscitated). The patient stopped spontaneous breathing 4 days after the scan, and life support was withdrawn at that time. The average enhancement time for both ECAs in 12 patients with BD was 16 seconds (mean [standard deviation (SD)], 16.08 3.85 seconds) when excluding the patient
Dynamic CTA in suspected brain death / Canadian Association of Radiologists Journal 65 (2014) 352e359
355
Figure 1. The timing diagram explains the acquisition of whole-brain volumes at each time point starting at 7 seconds from the time of the start of contrast injection. This volume uses 300 mA and is used as a mask for subtraction. Next, in the arterial phase (10-35 seconds) whole-brain volumes are acquired every 2 seconds. In the venous phase (40-80 seconds), volumes are acquired at 5-second intervals. By using each individual or all volumes, we could assess flow of contrast in intra- and extracranial circulation during the whole scanning duration (80 seconds). In contrast, older computed tomography scanners could only obtain a snapshot view of the brain circulation at a predetermined time point, which could miss the contrast, depending on the circulation time in an individual patient.
Figure 2. The upper panel (A-C) shows images from a 52-year-old man with diabetes who had postoperative cardiac arrest. (A) Noncontrast computed tomography (CT) of the head, showing diffuse loss of grey matter density lower than the white matter, so called ‘‘reversal sign,’’ a sign of diffuse ischemia. (B) A dynamic CTangiography (dCTA) at 22 seconds, showing good flow in the external carotid branches to the level of vertex but no evidence of flow in internal carotid arteries (ICA) and intracranial branches. (C) At 60 seconds, there is meager contrast flow noted in the right ICA, only to the level of the skull base; please note the scalp arteries have completely cleared of contrast. The lower panel (D, E) shows dCTA images from a 17-year-old man with a suicidal overdose (Video 1). (D) Frontal view dCTA at 24 seconds, showing delayed filling of the basilar artery and bilateral ICAs at the skull base but no filling of the distal branches. Complete filling of the superficial temporal and external carotid branches are noted, which is helpful as an internal reference. (E) Frontal view at 28 seconds, showing minimal filling of the right M1 segment and no filling of the bilateral distal middle cerebral arteries and posterior cerebral arteries branches. (F) A normal control with simultaneous filling of intra- and extracranial branches.
356
S. Chakraborty, R. A. Adas / Canadian Association of Radiologists Journal 65 (2014) 352e359
Table 2 Contrast appearance times in ICA and ECA at the level of the skull base in patient with brain death Patient no. (n ¼ 13) 1 2 3 4 5 6 7 8 9 10b 11 12 13
Right ICA, s 40 80a 25 17 20 25 23 23 15 11 33 40 50
Right ECA, s 17 19 17 15 9 20 15 19 13 13 13 23 13
Left ICA, s a
80 40 80a 17 25 31 23 23 15 11 21 40 33
Left ECA, s 17 19 17 15 9 20 15 19 13 13 13 23 13
ECA ¼ external carotid artery; ICA ¼ internal carotid artery. a Values recorded at 80 s in patients who, at the end of the study, showed no enhancement in their ICA. b The patient with preserved intracranial flow.
with preserved intracranial flow. The mean right ICA enhancement time was 33 seconds (mean [SD], 32.58 18.3 seconds). The mean left ICA enhancement time was 36 seconds (mean [SD], 35.67 22.2 seconds). Both right and left ICAs and ECAs values for patients with BD are detailed in Table 2. All control patients showed bilateral ICAs enhancement, which occurred either simultaneously or before the ECAs. Bilateral ICA and ECA enhancement times are shown in Table 3. When using the unequal variance ‘‘type 3’’ t test to compare the enhancement times of each vessel in both groups, there were no statistically significant differences (P > .05) in the enhancement times of both ECAs (P ¼ .29 for the right, and P ¼ .22 for the left). For the ICAs, there was a statistical difference in the enhancement times between the 2 groups; P < .004 for the right and P ¼ .005 for Table 3 Contrast appearance times in ICA and ECA at the level of the skull base in the control group Patient no. (n ¼ 15)
Right ICA, s
Right ECA, s
Left ICA, s
Left ECA, s
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
13 17 13 13 19 19 11 15 17 15 11 11 11 7 7
13 19 13 13 21 21 11 17 19 17 13 19 11 13 7
13 17 13 13 19 19 11 15 17 15 11 11 11 7 7
13 17 13 13 21 21 11 17 19 17 13 11 11 7 7
ECA ¼ external carotid artery; ICA ¼ internal carotid artery.
the left. The graphs in Figure 3 (upper panel) show the relationship of the timing between the left ECA and ICA in both patients with BD and the control group. The lower panel bar chart shows the mean contrast arrival time between the right ICA and ECA in both groups. Note the inversed relationship of ECA and ICA enhancement timings in the BD group compared with controls. The average delay time of the ICA enhancement compared with the ECA in patients with BD was 16.5 seconds on the right and 19.6 seconds on the left. For our control group, the mean delay time was e1.07 seconds on the right and e0.8 seconds on the left. Trickling of contrast (described as delayed, slow flow) (Figure 2D, E) was noted in the supraclinoid ICAs bilaterally in 7 patients, 6 patients also had similar findings in either M1 or A1 segments, and 4 patients demonstrated faint contrast enhancement in their intracranial vertebralebasilar arteries. Minimal contrast was noted in the venous system of 3 patients on delayed images. The presence of contrast in intracranial arteries can be confusing, especially when using a static CTA technique. In all of these patients, there was absent qualitative intracranial flow (best appreciated in dCTA images in a movie format) (Video 1). Discussion In this study, we describe the use of dCTA as an ancillary test in patients with suspected BD to show the lack of intracranial flow. This noninvasive technique provides a means to assess qualitative intracranial flow compared with the extracranial arteries as an internal reference. The ability for continuous flow assessment gives us extra diagnostic confidence in the presence of ‘‘stasis filling’’ [7] or trickling of contrast. We also describe the phenomenon of inversion of ICA and ECA enhancement times at the level of the skull base in patients with BD as an ‘‘inversion sign.’’ In our series, this correlates with a lack of global intracranial flow. This finding may be helpful in the assessment of dCTA generated from the limited coverage (at the skull base) perfusion data by using other standard multislice CT scanners. The BD [8] definition has evolved over the past few decades from being a coma depasse in 1959 [9] to the modernday definition developed by the American Academy of Neurology [10] in 1994. Simply, deep coma, lack of brain stem reflexes, and spontaneous breathing composes the triad for diagnosis. Many confounding factors may exist, which limit proper assessment, examples of which are mentioned in Table 1. Ancillary tests are used to provide necessary support to the clinical diagnosis of BD. Cerebral 4-vessel catheter angiogram remains the criterion standard ancillary test of choice in confirming the diagnosis of BD, followed by hexamethylpropylene amine oximeesingle photon emission computed tomography radionuclide angiography. Cerebral circulatory arrest occurs when the intracranial pressure exceeds arterial inflow pressure. The lack of cerebral vessel enhancement beyond the foramen magnum and carotid siphons is usually diagnostic. Due to the limitations and
Dynamic CTA in suspected brain death / Canadian Association of Radiologists Journal 65 (2014) 352e359
357
Figure 3. Upper panel graphs, showing the relationship of the timing of the left external carotid arteries (ECA) and internal carotid arteries (ICA) contrast appearance in both subjects who are brain dead and the control subjects. There is inversion of ECA and ICA timings in the brain death group compared with the control group. The lower panel bar charts show mean contrast appearances in the right ECAs and the ICAs in both groups.
invasiveness of these techniques, attempts to validate the use of cross-sectional imaging have been made. The use of CT to diagnose BD was proposed as early as 1978 [2,11] and then again in 1985 as an alternative to cerebral angiography [12]. The Medical and Scientific Council of the French Transplant Establishment recognizes CTA as a reliable technique for the diagnosis of BD [13], after accepting the method of Dupas et al [1] of the 7-point CTA score for BD diagnosis in 1998. Newer studies [14e17], however, showed less sensitivity and reported that intracranial opacification on CTA was present in a substantial number of patients (11%48%). In 2006, Combes et al [15] questioned the reliability of CTA in BD. They found that CTA did not confirm the clinical diagnosis of BD in 13 of the 43 patients who were clinically BD, whereas findings of the catheter cerebral angiography confirmed the clinical diagnosis of BD in all 43 patients. A more recent study, in 2009, by Frampas et al [16] proposed the use of a simplified 4-point CTA score, based on the lack of opacification of the middle cerebral arteries and internal cerebral veins, which they concluded has a sensitivity of 85.7% and specificity of 100%. A recent evidence-based guideline update by Wijdicks et al (4) concluded that, although crosssectional vascular imaging (CTA, magnetic resonance angiography) is currently used in clinical practice, further studies are needed to prove their reliability. Limitations with Standard or 2-Phase CTA In all the studies mentioned above, a 2-phase contrast CTA (early arterial and delayed venous) was used. This
technique provides a static volume of brain and vessels images performed during 2 specified points in time (snapshot views). The most common limitation encountered in patients who are BD is called ‘‘stasis filling’’ or contrast visualization in delayed images in the proximal intracranial arteries and veins. This is well documented in the BD literature [7,18] and also is noted in the catheter angiographic studies [19,20]. This finding can be more pronounced in CTA due to inherent higher contrast resolution of CTs compared with subtraction radiography. In a more recent study [21], opacification of the proximal middle cerebral arteries segment was demonstrated in arterial series in 19 cases and in venous (55 seconds after contrast injection) series in 37 of a total 63 cases. Opacification of other intracranial arteries was visible in only 3 patients in the arterial series but 22 patients in venous series. Even minimal contrast filling in deep cerebral veins was noted in 3 patients in our series. We believe that this ‘‘stasis flow’’ does not indicate true cerebral blood flow and does not contradict the BD diagnosis. But this phenomenon causes significant diagnostic confusion and is likely a reason why CTA is not widely accepted for BD diagnosis. Benefits of dCTA A dCTA, in contrast to traditional CTA, has the advantage of providing visualization of dynamic blood flow information, which makes it possible to analyse the pattern of flow qualitatively (Video 1). So, the presence of a small amount of contrast in the proximal intracranial arteries (as
358
S. Chakraborty, R. A. Adas / Canadian Association of Radiologists Journal 65 (2014) 352e359
in 7 patients of 12 in this series) does not sway us to confidently report the lack of global intracranial flow. This is the most beneficial aspect of dCTA in these cases. It also images the entire cranial vasculature simultaneously and allows comparison with the extracranial arterial flow, which can be used as an internal reference. Aside from the intravenous contrast injection, it is a noninvasive and safe technique. A dCTA is considerably less time consuming and technically less demanding than cerebral angiography for both patients and radiologists. This technique does not exclusively need a whole-brain CT scanner for data acquisition. Other techniques, such as volume helical shuttle mode [22] or the table toggle technique for extended coverage of brain perfusion, can generate similar dCTA. It requires a significantly low volume of contrast (40 mL), which reduces the risk of renal dysfunction and which might compromise organ harvesting. In all but 1 of the patients with suspected BD, enhancement of the ECAs occurred before the ICAs. This particular patient had a preserved normal intracranial flow at the time of scanning, although he was suspected clinically to have had severe brain damage (due to anoxic injury) or status epilepticus. Spontaneous breathing stopped 4 days after the scan, and life support was withdrawn at that time. One of the characteristic signs of BD in noncontrast CT is the ‘‘reversal sign’’ (Figure 2A), and this is well described in the literature [23], in which a reversal of density between grey and white matter is seen in cases of patients with diffuse anoxic cerebral injury. It is caused by the loss of grey matter density due to cytotoxic oedema after cell death, which results in a relative low density of grey matter compared with the adjacent white matter. In this study, we describe the ‘‘inversion sign’’ in dCTA related to inversion of time of appearance of contrast in ICAs and ECAs in cases of BD (Figure 2B-E). The delayed abnormal filling of the ICAs observed in the BD group did not occur in any of the control group patients. This finding led us to conclude that inversion of comparative flow between the ICAs and ECAs at the level of the skull base may predict the lack of qualitative intracranial flow as seen in our study group. Also, the patient in the BD group with preserved intracranial flow had no inversion of contrast timing at the skull base, thus confirming the prediction for distal global intracranial flow. Application of this concept could be used by limited coverage perfusion acquisition at the skull base in standard multislice CT scanners to predict a lack of global intracranial flow. Limitations of This Study This was a clinical observational study and has the limitations of selection bias, a small number of subjects, and a variety of initiating causes as well as a variable time range from the clinical suspicion of BD to the time that the scans were performed. Also, catheter angiography or radionuclide studies were not performed in this clinical cohort. Prospective larger studies, with comparison with
catheter angiography or radionuclide angiography as criterion standards, are required in the future for validation of this technique. Conclusion In contrast to traditional CTA, dCTA has the advantage of providing visualization of dynamic blood flow, which makes it possible to analyse the pattern of intracranial flow qualitatively, which allows us to confidently report the lack of global intracranial flow and not be swayed by the presence of a small amount of contrast in the proximal intracranial arteries. This advantage overcomes the limitation of ‘‘stasis filling’’ seen in standard 2-phase CTA. In addition, the simultaneous imaging of the entire cranial vasculature provided by this technique allows comparison with the extracranial arterial flow. This advantage led us to observe and describe an ‘‘inversion sign,’’ which refers to the inversion of flow pattern between the ICAs and the ECAs in patients with brain death due to delayed ICA flow. This finding strengthens the utility of noninvasive dCTA as an ancillary test for diagnosing brain death. References [1] Dupas B, Gayet-Delacroix M, Villers D, et al. Diagnosis of brain death using two-phase spiral ct. AJNR Am J Neuroradiol 1998;19:641e7. [2] Rappaport ZH, Epstein F. Computerized axial tomography in the preoperative evaluation of posterior fossa tumors in children. Childs Brain 1978;4:170e9. [3] Shemie SD. Brain arrest to neurological determination of death to organ utilization: the evolution of hospital-based organ donation strategies in Canada. Can J Anaesth 2006;53:747e52. [4] Wijdicks EF, Varelas PN, Gronseth GS, et al. Evidence-based guideline update: determining brain death in adults: report of the Quality Standards Subcommittee of the American Academy of Neurology. Neurology 2010;74:1911e8. [5] Shemie SD, Lee D, Sharpe M, et al. Brain blood flow in the neurological determination of death: Canadian expert report. Can J Neurol Sci 2008;35:140e5. [6] Salomon EJ, Barfett J, Willems PW, et al. Dynamic CT angiography and CT perfusion employing a 320-detector row CT: protocol and current clinical applications. Klin Neuroradiol 2009;19:187e96. [7] van der Lugt A. Imaging tests in determination of brain death. Neuroradiology 2010;52:945e7. [8] A definition of irreversible coma. Report of the Ad Hoc Committee of the Harvard Medical School to examine the definition of brain death. JAMA 1968;205:337e40. [9] Mollaret P, Goulon M. The depassed coma (preliminary memoir) [in French]. Rev Neurol (Paris) 1959;101:3e15. [10] Practice parameters for determining brain death in adults (summary statement). The Quality Standards Subcommittee of the American Academy of Neurology. Neurology 1995;45:1012e4. [11] Rangel RA. Computerized axial tomography in brain death. Stroke 1978;9:597e8. [12] Planitzer J, Zschenderlein R, Schulze HA, et al. Computer tomography studies in irreversible cerebral function loss (brain death) [in German with English abstract]. Psychiatr Neurol Med Psychol (Leipz) 1985;37: 509e17. [13] Recommendations du conseil medical et scientifique de l’etablissement franc¸ais des greffes. Les techniques d’angiographie dans le diagnostic de la mort encephalique. Deliberation n 2000.12 en seance du 21 novembre 2000, dossier s3 d2.
Dynamic CTA in suspected brain death / Canadian Association of Radiologists Journal 65 (2014) 352e359 [14] Leclerc X, Taschner CA, Vidal A, et al. The role of spiral CT for the assessment of the intracranial circulation in suspected brain-death. J Neuroradiol 2006;33:90e5. [15] Combes JC, Chomel A, Ricolfi F, et al. Reliability of computed tomographic angiography in the diagnosis of brain death. Transplant Proc 2007;39:16e20. [16] Frampas E, Videcoq M, de Kerviler E, et al. Ct angiography for brain death diagnosis. AJNR Am J Neuroradiol 2009;30:1566e70. [17] Escudero D, Otero J, Marques L, et al. Diagnosing brain death by CT perfusion and multislice CT angiography. Neurocrit Care 2009;11: 261e71. [18] Hassler W, Steinmetz H, Pirschel J. Transcranial Doppler study of intracranial circulatory arrest. J Neurosurg 1989;71:195e201.
359
[19] Kricheff II , Pinto RS, George AE, et al. Angiographic findings in brain death. Ann N Y Acad Sci 1978;315:168e83. [20] Bradac GB, Simon RS. Angiography in brain death. Neuroradiology 1974;7:25e8. [21] Welschehold S, Kerz T, Boor S, et al. Detection of intracranial circulatory arrest in brain death using cranial CT-angiography. Eur J Neurol 2013;20:173e9. [22] Volume helical shuttle CT perfusion identifies stroke patients. Available at http://www.auntminnie.com/index.aspx?sec¼ser&sub¼def&pag¼dis& ItemID¼90309. Accessed March 12, 2014. [23] Han BK, Towbin RB, De Courten-Myers G, et al. Reversal sign on CT: effect of anoxic/ischemic cerebral injury in children. AJNR Am J Neuroradiol 1989;10:1191e8.