| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Experimental Studies |
1 From the Departments of Radiology (H.J.C., T.A.A., W.F.M., R.J.R., M.E.J., J.E.D., D.F.K.), Neurosurgery (G.A.H., S.B.H.), and Pathology (J.W.M.), University of Virginia Health Sciences Center. Received April 13, 1998; revision requested July 1; final revision received December 16; accepted April 8, 1999. T.A.A. is supported by the RSNA Research and Education Foundation as a 1997 Research Resident. D.F.K. is supported by the RSNA Research and Education Foundation as a 1997 Bracco/RSNA Scholar. Address reprint requests to H.J.C., Department of Radiology, Emory University Hospital, 1364 Clifton Rd NE, Atlanta, GA 30322 (e-mail: harry_cloft@emory.org).
 |
Abstract |
|---|
 TOP Abstract IntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
MATERIALS AND METHODS: An experimental aneurysm model was created in rabbits by means of endovascular balloon occlusion of the left common carotid artery, which created an aneurysm at the bifurcation formed by the aortic arch and the brachiocephalic trunk. A total of 18 aneurysms were created. In eight rabbits, the aneurysms were incubated with intraluminal elastase to induce degeneration of the elastic laminae. The animals were followed up with angiography for as long as 3 months. The animals were sacrificed at various times, and histologic evaluation of the aneurysm was performed.
RESULTS: Ten aneurysms created without elastase infusion were all very small or completely closed at 1–3 months. Six aneurysms created with elastase infusion had long-term patency (two were patent at 1 month and four, at 3 months). The elastase aneurysms had a mean width of 3 mm (range, 2–3.5 mm) and a mean length of 5 mm (range, 3–7 mm). Histologic evaluation revealed destruction of the normal elastin layers, which allowed the artery to become aneurysmal.
CONCLUSION: This aneurysm model re-created the hemodynamic forces and size of human cerebral bifurcation aneurysms and maintained the integrity of the endothelium. The creation of the aneurysms was rapid, reliable, and reproducible.
Index terms: Aneurysm, carotid, 904.73 • Aneurysm, cerebral, 17.73 • Aneurysm, therapy, 17.1264, 904.1264, 904.73 • Animals • Carotid arteries, therapeutic blockade, 17.1264, 904.1264
|
Introduction |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
|
MATERIALS AND METHODS |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Creation of Aneurysms without Elastase
The right inguinal region of each rabbit was shaved and prepared in a sterile manner. The right superficial femoral artery was exposed at surgery and was accessed with a 20-gauge Teflon sheath, and an 0.018-inch–diameter guide wire was placed. Serial dilations were performed, and a 6-F sheath was placed.
A 6-F Envoy guiding catheter (Cordis Endovascular, Miami Lakes, Fla) was placed into the left common carotid artery. A hand-tied detachable latex balloon (no. 15 Debrun; Nycomed, Paris, France) was mounted on a microcatheter (Tracker-18; Boston Scientific Target). The balloon and microcatheter system was placed coaxially through the guiding catheter into the left common carotid artery. The balloon was inflated in a position approximately 2 cm distal to the origin of the vessel. The balloon was detached in the usual fashion. The guiding catheter and sheath were removed, and the right superficial femoral artery was ligated with 2.0 silk suture. The skin incision was closed with Polysorb 3-0 absorbable suture (Ethicon, Somerville, NJ).
Creation of Aneurysms with Elastase
Because the aneurysms created without elastase infusion did not remain patent, elastase infusion was added to the procedure to destroy the elastic laminae and thereby prevent the arterial contraction that occurred after balloon occlusion. In eight rabbits, the experimental aneurysms were incubated with intraluminal elastase to induce degeneration of the elastic lamina in the wall of the aneurysm.
The following additions were made to the procedure described previously. After balloon detachment, the guiding catheter was maintained in place at the origin of the left common carotid artery. A 6-F guiding catheter (Cordis Endovascular) approximated the size of the vessel and occluded the origin, so a closed system was achieved. A Tracker-18 catheter (Boston Scientific Target) was advanced through the guiding catheter into the arterial stump. Fifty units of bovine type I pancreatic elastase (Sigma Chemical, St Louis, Mo) was infused into the arterial stump and was left in place for 30 minutes. The microcatheter was then removed, and the guiding catheter was advanced into the ascending aorta. The microcatheter, guiding catheter, and sheath were removed, and the right superficial femoral artery was ligated with 2.0 silk suture. The skin incision was closed with Polysorb 3-0 absorbable suture (Ethicon).
Follow-up for the Aneurysms Created without Elastase
Angiography was performed immediately before sacrificing the animals. Anesthesia was induced with an intramuscular injection of a mixture of ketamine and xylazine. Maintenance anesthesia was achieved by means of an intravenous infusion of pentobarbital via the ear vein. The left femoral artery was exposed at surgery, and a 5-F catheter was placed in the ascending aorta. Nonionic contrast material (Omnipaque [iohexol]; Nycomed Amersham, Princeton, NJ) was injected for cut-film angiography. The rabbits were sacrificed with a lethal dose of intravenous pentobarbital. Five rabbits were sacrificed at 1 month after the creation of the aneurysm, and five rabbits were sacrificed at 2–3 months.
Follow-up for the Aneurysms Created with Elastase
After we conducted the experiment with creating aneurysms without elastase as described previously, we learned that evaluation of the model aneurysms could be performed with intravenous digital subtraction angiography instead of invasive catheter angiography. Intravenous digital subtraction angiography was performed by injecting 10 mL of the nonionic contrast material into an ear vein during digital imaging while the animal was sedated with an intramuscular injection of ketamine and xylazine. Intravenous digital subtraction angiograms were obtained at 2 weeks and at 1 month after the procedure in all animals except the two that were sacrificed immediately after the aneurysm was created.
Rabbits were sacrificed with a lethal dose of intravenous pentobarbital at various intervals after occlusion of the left common carotid artery, as follows: immediately after occlusion (n = 2), at 1 month (n = 2), and at 3 months (n = 3). One rabbit was allowed to survive so we could continue to monitor the long-term patency of the aneurysm. Repeat intravenous digital subtraction angiography was performed immediately prior to sacrificing those animals that were sacrificed at 3 months to facilitate correlation of angiographic and histologic findings.
Histologic Evaluation
After the animals were sacrificed, the aortic arch and brachiocephalic vessels were excised and fixed with formalin. The specimens were embedded in paraffin and sectioned for histologic evaluation with hematoxylin-eosin and Verhoeff-van Giesen staining. The samples were evaluated by a pathologist (J.W.M.) for evidence of inflammation and thrombus formation and to determine the integrity of the elastic laminae.
|
RESULTS |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
|
|
|
|
|
|
|
|
Clinical Sequelae
We noted no neurologic deficits secondary to the procedure in any of the rabbits. We also noted no systemic reaction to the elastase infusion.
|
DISCUSSION |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Occlusion of the common carotid artery in rabbits could be performed without creating a neurologic insult because of the collateral blood flow to the brain via the circle of Willis. The jet of blood flow within the ascending aorta was directed at the orifice of the experimental aneurysm. This high-flow condition was expected to reduce thrombosis within the aneurysm and provide hemodynamic stresses similar to those in intracranial bifurcation aneurysms.
The aorta of the rabbit measured 4-5 mm, which was similar to the size of human carotid and basilar arteries. Therefore, it had a size that was appropriate for a parent vessel of an experimental aneurysm. The average size of the aneurysm created was 3 x 5 mm, which was well within the typical size range of human cerebral aneurysms and which was an appropriate size for endovascular therapy. On the angiograms obtained at 1 and 3 months, the size and shape of the aneurysms were unchanged.
Elastase was necessary to maintain the patency of the aneurysms. Without elastase, the occluded left common carotid artery contracted, which left no proximal stump to serve as an aneurysm model. Elastase administration presumably eliminated passive contraction of the artery after occlusion and thereby allowed it to dilate and form the aneurysm.
Other researchers have used the local application of elastase to induce or enlarge aneurysms. Anidjar et al (3) created fusiform aortic aneurysms in rats by using intraarterial incubation of porcine pancreatic elastase. Recently, this technique was extended to the creation of saccular aneurysms.
Cawley et al (4) created lateral aneurysms at surgery by ligating the external carotid artery near its origin and then by incubating the external carotid arterial stump with intraluminal porcine pancreatic elastase. The elastase disrupts the elastic lamina in the arterial wall. These aneurysms were followed for 2–12 weeks; 93% remained patent at angiography, and 40% were fully patent, without fibrin deposits or thrombus, at histologic evaluation. They speculated that the low patency rate was secondary to the low flow in the aneurysm. Their model has the advantages of a relatively simple surgical procedure, which is required for construction of the aneurysm, and an intact arterial wall. Recent data suggest that the morphology, cellular content, and appearance of the elastic lamina in these elastase-induced aneurysms are similar to those of human intracranial aneurysms (5).
Ideally, an experimental cerebral aneurysm should re-create the hemodynamic forces, physical dimensions, and radiographic appearance of human cerebral bifurcation aneurysms and should remain patent indefinitely, if untreated (6). In addition, the integrity of the endothelium should be maintained, and creation of the aneurysms should be rapid, reliable, and reproducible. The aneurysm model described in this report had all of these characteristics. This experimental bifurcation aneurysm was patent for at least 3 months after construction.
Although our technique was initially applied to the common carotid artery, it can be extended to other vascular trifurcations for the creation of bifurcation aneurysms with different sizes and flow characteristics. However, the size of the resultant aneurysm at other trifurcations in a rabbit would probably be small and would probably not be amenable to coil embolization. Trifurcations may exist in other animals, which would also allow formation of an aneurysm model amenable to coil embolization.
Surgical animal models of aneurysms have numerous shortcomings. Since bifurcation aneurysms are tedious to create at surgery (7), most surgical models rely on venous patch grafts that are sewn on straight segments. This procedure yields a lateral (or sidewall) aneurysm without the anatomic and hemodynamic characteristics typical of cerebral bifurcation aneurysms. Since the efficacy of an endovascular device such as the Guglielmi detachable coil is greatly influenced by hemodynamic forces at the aneurysm orifice (8), accurate modeling of hemodynamics is critical to creating a valid animal model.
In addition to the disadvantage of requiring tedious vascular surgery, use of the surgical method creates aneurysms with suture lines that encircle the neck of the aneurysm and close the vein pouch. This disruption of the endothelium causes the release of platelet-derived growth factor, among other factors, that leads to scarring and obliteration of the cavity of the aneurysm. Fibrotic scarring of vein-patch aneurysms is a considerable problem. Also, most investigators rely on more expensive swine or canine models because of the difficulty of performing surgical procedures in smaller animals.
Other methods of creating bifurcation aneurysms have been reported. A method conceptually similar to ours is that described by Roach (9) and by Boyce and Roach (10), in which a bifurcation aneurysm is created at the trifurcation of the abdominal aorta in dogs by surgically tying the canine tail (sacral) artery. The surgical creation of such an aneurysm requires access to the retroperitoneum, which is accomplished with a relatively simple vascular procedure. A limitation is that only the depth of the aneurysm, not the width of the orifice, can be varied experimentally. An advantage is that the wall of the aneurysm is arterial.
Like Roach (9), Boyce and Roach (10), and Cawley et al (4), we elected to create an aneurysm by occluding an arterial branch near its origin. This created an aneurysm with an arterial rather than a venous wall to better simulate the morphologic and histologic attributes of human intracranial aneurysms (5). Unlike the experimental aneurysms previously described, our experimental aneurysm was constructed primarily with endovascular techniques.
The Guglielmi detachable coil recently approved by the U.S. Food and Drug Administration was initially tested and validated in a swine lateral aneurysm model (1). This study did not include untreated control aneurysms, and subsequent experience has shown that a high frequency of spontaneous thrombosis in the pig model can lead to misinterpretation of the results (11). Subsequently, the Guglielmi detachable coil was tested in canine lateral aneurysms (12), in which it demonstrated fibrotic changes within the aneurysm. Although this canine study also lacked a control group, lateral wall aneurysms in dogs appear to remain patent reliably for at least 12 weeks, if left untreated (13).
Aneurysms created at surgery in the lateral wall of the carotid artery in monkeys were treated with a Guglielmi detachable coil and were found to contain organized media–like tissue at 3 months; but again, no long-term untreated controls were included (14). A recent study of the Guglielmi detachable coil in surgically created bifurcation aneurysms in rabbits showed similar fibrotic changes within the aneurysm lumen (15).
Little is known about the histologic response to the Guglielmi detachable coil in human aneurysms. To our knowledge, the only published long-term histologic data appear in an article by Molyneux et al (16), who reported findings from two patients who died of causes unrelated to their aneurysms at 2 and 6 months after therapy. In both of these cases, unlike the cases in which the fibrotic scar was found in the animal models described previously, there was unorganized thrombus in the aneurysms, and there was no evidence of endothelialization in the neck of the aneurysm (16).
Possible explanations for the discrepancy between the animal and human data include the following: (a) Humans have an unusual histologic response to the platinum coil compared with that of animals, or (b) the animal model is invalid. The animal model could be invalid if the surgical wound at the neck of the aneurysm leads to a fibrotic reaction to the coil in animals that does not occur in humans, who have no such wound at the neck of the aneurysm. The endovascular model described here would be more valid than existing models if it did not elicit the fibrotic reaction previously described.
A valid animal model that reproduces the human response to embolic materials in aneurysms is essential for the further improvement of the Guglielmi detachable coil and other materials. The formation of soft, unorganized clot rather than a fibrotic scar in human aneurysms that are treated with a Guglielmi detachable coil may be a major factor in coil compaction and regrowth of the aneurysm after embolization.
Attempts have been made to improve the treatment of aneurysms with and to modify the biologic response to the Guglielmi detachable coil with ion implantation (17), collagen filaments (13,18), and collagen coatings (19). Because these modifications were tested in animal models that did not simulate the biologic response to the conventional Guglielmi detachable coil in human aneurysms, the results of these experiments reveal little about how these modifications might affect therapy for human aneurysms.
Our model is not a perfect simulation of intracranial aneurysms in humans. The tunica media of the model aneurysm was intact, unlike that of cerebral aneurysms, and cells from the tunica media could affect the biologic response to embolic materials. In addition, the model aneurysm is not surrounded with cerebrospinal fluid within the subarachnoid space, but rather, it is surrounded with mediastinal tissues, which might affect the response to embolic materials in the lumen. However, the intact endothelium may function as a barrier between the lumen and the vessel wall and mediastinum, which prevents interference from cells and biologically active molecules that are not normally present in human cerebral aneurysms.
The model aneurysm had no arterial wall in its dome, but rather, it had only a blood clot. We did not consider such clot formation to be detrimental to our model, since many human aneurysms, especially those that have ruptured, have thrombus within them. Also, thrombus forms in surgical models of aneurysms (4,20). It is doubtful that the presence of clot was detrimental to the model, since the luminal surface of the model aneurysm that was in contact with embolic materials tested was similar, if not identical, to the luminal surface of cerebral aneurysms. This clot formed a barrier between the lumen and the balloon (or other embolic materials that could have been used, such as coils) and prevented interaction between the embolic material used to make the model aneurysm and the embolic material tested in the lumen of the aneurysm.
We report on an animal model of bifurcation aneurysms that accurately reproduced the hemodynamic forces and intact intima of human intracranial aneurysms and that was constructed with endovascular techniques. This model can facilitate the development and validation of new endovascular techniques for the treatment of human intracranial aneurysms. The long-term natural history of these model aneurysms beyond 3 months remains to be determined. It is doubtful that the aneurysms will rupture, since fusiform carotid aneurysms created with elastase perfusion in rabbits did not rupture during a 3-month follow-up (21). The blood pressures in rabbits are typically 90–130 mm Hg systolic and 80–90 mm Hg diastolic (22), which are within the normotensive ranges for adult humans. Blood pressure is important because it could affect blood clotting, clot maturation, and coil compaction.Practical application: We are currently treating experimental aneurysms, created with the technique described here, with the Guglielmi detachable coil to evaluate their response. These aneurysms will be followed for 6 months to assess evidence of coil compaction. At the end of the observational period, a histologic examination will be performed. Experiments can then be performed with a Guglielmi detachable coil that has been modified to alter the biologic response to the coil and to reduce the risk of recurrence of aneurysms in humans.
|
Footnotes |
|---|
|
References |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  M. J. DeLeo III, M. J. Gounis, B. Hong, J. C. Ford, A. K. Wakhloo, and A. A. Bogdanov Jr Carotid Artery Brain Aneurysm Model: In Vivo Molecular Enzyme-specific MR Imaging of Active Inflammation in a Pilot Study Radiology, June 22, 2009; (2009) 2523081426. [Abstract] [Full Text]
|
|
|
|
 |
|
 L. Gao, Y. Hoi, D. D. Swartz, J. Kolega, A. Siddiqui, and H. Meng Nascent Aneurysm Formation at the Basilar Terminus Induced by Hemodynamics Stroke, July 1, 2008; 39(7): 2085 - 2090. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Meng, Z. Wang, Y. Hoi, L. Gao, E. Metaxa, D. D. Swartz, and J. Kolega Complex Hemodynamics at the Apex of an Arterial Bifurcation Induces Vascular Remodeling Resembling Cerebral Aneurysm Initiation Stroke, June 1, 2007; 38(6): 1924 - 1931. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y.H. Ding, D. Dai, M.A. Danielson, R. Kadirvel, D.A. Lewis, H.J. Cloft, and D.F. Kallmes Control of Aneurysm Volume by Adjusting the Position of Ligation During Creation of Elastase-Induced Aneurysms: A Prospective Study AJNR Am. J. Neuroradiol., May 1, 2007; 28(5): 857 - 859. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Dai, Y. H. Ding, M. A. Danielson, R. Kadirvel, L. W. Hunter, W.-Z. Zhan, G. A. Helm, D. A. Lewis, H. J. Cloft, G. C. Sieck, et al. Endovascular Treatment of Experimental Aneurysms by Use of Fibroblast-Coated Platinum Coils: An Angiographic and Histopathologic Study Stroke, January 1, 2007; 38(1): 170 - 176. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Meng, Z. Wang, M. Kim, R.D. Ecker, and L.N. Hopkins Saccular aneurysms on straight and curved vessels are subject to different hemodynamics: implications of intravascular stenting. AJNR Am. J. Neuroradiol., October 1, 2006; 27(9): 1861 - 1865. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y.H. Ding, D. Dai, D.A. Lewis, M.A. Danielson, R. Kadirvel, J.N. Mandrekar, H.J. Cloft, and D.F. Kallmes Can neck size in elastase-induced aneurysms be controlled? A retrospective study. AJNR Am. J. Neuroradiol., September 1, 2006; 27(8): 1681 - 1684. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Onizuka, L. Miskolczi, M.J. Gounis, J. Seong, B.B. Lieber, and A.K. Wakhloo Elastase-induced aneurysms in rabbits: effect of postconstruction geometry on final size. AJNR Am. J. Neuroradiol., May 1, 2006; 27(5): 1129 - 1131. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y.H. Ding, D. Dai, D.A. Lewis, M.A. Danielson, R. Kadirvel, H.J. Cloft, and D.F. Kallmes Long-Term Patency of Elastase-Induced Aneurysm Model in Rabbits AJNR Am. J. Neuroradiol., January 1, 2006; 27(1): 139 - 141. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. H. Ding, D. Dai, D. A. Lewis, M. A. Danielson, R. Kadirvel, J. N. Mandrekar, H. J. Cloft, and D. F. Kallmes Can Neck Size in Elastase-Induced Aneurysms Be Controlled? A Prospective Study AJNR Am. J. Neuroradiol., October 1, 2005; 26(9): 2364 - 2367. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Miskolczi, B. Nemes, L. Cesar, O. Masanari, and M. J. Gounis Contrast Injection via the Central Artery of the Left Ear in Rabbits: A New Technique to Simplify Follow-Up Studies AJNR Am. J. Neuroradiol., September 1, 2005; 26(8): 1964 - 1966. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. S. Shin, Y. Niimi, Y. Yoshino, J. K. Song, M. Silane, and A. Berenstein Creation of Four Experimental Aneurysms with Different Hemodynamics in One Dog AJNR Am. J. Neuroradiol., August 1, 2005; 26(7): 1764 - 1767. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Raymond, I. Salazkin, A. Metcalfe, O. Robledo, G. Gevry, D. Roy, A. Weill, and F. Guilbert Lingual Artery Bifurcation Aneurysms for Training and Evaluation of Neurovascular Devices AJNR Am. J. Neuroradiol., September 1, 2004; 25(8): 1387 - 1390. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. H. Fujiwara and D. F. Kallmes Healing Response in Elastase-Induced Rabbit Aneurysms after Embolization with a New Platinum Coil System AJNR Am. J. Neuroradiol., August 1, 2002; 23(7): 1137 - 1144. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. F. Kallmes, N. H. Fujiwara, S. S. Berr, G. A. Helm, and H. J. Cloft Elastase-Induced Saccular Aneurysms in Rabbits: A Dose-Escalation Study AJNR Am. J. Neuroradiol., February 1, 2002; 23(2): 295 - 298. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. G. Short, N. H. Fujiwara, W. F. Marx, G. A. Helm, H. J. Cloft, and D. F. Kallmes Elastase-Induced Saccular Aneurysms in Rabbits: Comparison of Geometric Features with Those of Human Aneurysms AJNR Am. J. Neuroradiol., November 1, 2001; 22(10): 1833 - 1837. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. H. Fujiwara, H. J. Cloft, W. F. Marx, J. G. Short, M. E. Jensen, and D. F. Kallmes Serial Angiography in an Elastase-induced Aneurysm Model in Rabbits: Evidence for Progressive Aneurysm Enlargement after Creation AJNR Am. J. Neuroradiol., April 1, 2001; 22(4): 698 - 703. [Abstract] [Full Text]
|
|
|
|
|
|
W. F. Marx, H. J. Cloft, G. A. Helm, J. G. Short, H. M. Do, M. E. Jensen, and D. F. Kallmes Endovascular Treatment of Experimental Aneurysms by Use of Biologically Modified Embolic Devices: Coil-mediated Intraaneurysmal Delivery of Fibroblast Tissue Allografts AJNR Am. J. Neuroradiol., February 1, 2001; 22(2): 323 - 333. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Abruzzo, H. J. Cloft, and G. G. Shengelaia Reply AJNR Am. J. Neuroradiol., October 1, 2000; 21(9): 1771 - 1773. [Full Text]
|
|
|
|
 |
|
 T. A. Altes, H. J. Cloft, J. G. Short, A. DeGast, H. M. Do, G. A. Helm, and D. F. Kallmes Creation of Saccular Aneurysms in the Rabbit: A Model Suitable for Testing Endovascular Devices Am. J. Roentgenol., February 1, 2000; 174(2): 349 - 354. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
