Volume 20, Issue 4 , Pages 455-460, April 2009
The Corona Mortis, a Frequent Vascular Variant Susceptible to Blunt Pelvic Trauma: Identification at Routine Multidetector CT
Article Outline
Purpose
To determine if contrast-enhanced multidetector computed tomography (CT) can demonstrate the “corona mortis” (translated as “crown of death”), a common variant obturator artery originating from the external iliac artery system that is susceptible to pelvic trauma.
Methods and Materials
A representative case study is described. The authors evaluated 50 consecutive patients undergoing routine clinically indicated, standard protocol, venous phase contrast-enhanced multidetector CT of the pelvis, 25 with a 16-slice scanner and 25 with a 64-slice scanner. Three data sets were created for each study, as follows: (a) 5 × 5-mm axial, (b) 1.25 × 1.0-mm axial, and (c) 1.25 × 1.0-mm coronal. Three radiologists independently reviewed the images for the presence or absence of corona morti.
Results
One hundred hemipelves in 50 patients were evaluated. In total, 29 corona morti were identified, including 10 on the 5-mm axial images, 25 on the 1.25-mm coronal images, and 29 on the 1.25-mm axial images. By consensus agreement, interpretation of challenging cases was easiest with the 64-slice images.
Conclusions
The corona mortis variant can be identified on routine contrast-enhanced multidetector CT scans in about one-third of patients. Thin (1.25-mm) CT reconstructions demonstrate this variation much more frequently than 5-mm-thick images. This suggests that the corona mortis may be prospectively identified at contrast-enhanced multidetector CT in pelvic trauma patients and help guide subsequent endovascular embolization. However, further study in the trauma population is necessary to confirm this.
Interventional radiology often plays an integral role in the treatment of acute trauma patients. Blunt pelvic injuries are common in level 1 trauma centers. Pelvic fractures and their associated bleeding are usually diagnosed with contrast-enhanced multidetector computed tomography (CT), which is readily available in most emergency departments today (1, 2). If such trauma leads to hemodynamic instability, interventional radiology has the unique capacity to rapidly and effectively confirm and then embolize hemorrhaging sites, with less morbidity and greater efficacy than traditional operative treatment (2–4).
It is important that interventional radiologists be familiar with the typical sites of pelvic bleeding from trauma as well as the potential anatomic variations of the origins and courses of bleeding arteries. A frequent pelvic normal variant artery susceptible to traumatic injury is the variant obturator artery, also known as the “corona mortis,” which is translated as “crown of death” (5, 6, 7). This entity has been described as either an obturator artery originating from the external iliac system or, more broadly, as an anastomotic branch between the external iliac system (either arterial or venous) and obturator vessels in the obturator canal (8–10). Herein, we first describe a representative case report of traumatic injury to a corona mortis artery successfully treated with catheter-guided embolization. This led to a subsequent prospective evaluation of 50 consecutive pelvic CT examinations to determine if variant obturator arteries can be detected at routine contrast-enhanced multidetector CT.
Representative Case Report
Our institutional review board does not require approval for single case reports. A 19-year-old Hispanic male was the restrained driver of a high-speed motor vehicle accident. At presentation to the emergency department, his Glasgow coma scale was 8, blood pressure 112/61 mm Hg, and heart rate 128 bpm. During primary survey, the patient was intubated and fluid resuscitation initiated. A chest tube was subsequently placed for the treatment of a left pneumothorax. The initial hemoglobin level was 13.1 g/dL (131 g/L). He was then taken to CT for further evaluation.
Contrast-enhanced multidetector CT of the abdomen and pelvis demonstrated a normal abdomen but acute fractures of the first sacral segment, left superior pubic ramus, and left ischiopubic ramus. In addition, a large pelvic hematoma was seen medial to the left iliac vessels with associated active extravasation (Fig 1). Interval hemoglobin and hematocrit levels were 11.3 g/dL (113 g/L) and 35%, respectively. His blood pressure remained stable, but persistent tachycardia was noted.
Figure 1.
Contrast-enhanced CT scan demonstrates a large left-sided pelvic hematoma with active extravasation (*). Note the decompressed bladder deviated to the right with opacified distal ureters posteriorly (arrows). Pelvic fractures are not shown.
The patient was taken from the CT scanner directly to the angiography suite for emergent pelvic angiography. A two-unit transfusion of packed red blood cells was initiated. Access was gained into the right common femoral artery, and a 5-F vascular sheath was placed. A 5-F Cobra catheter was advanced into the left internal iliac artery and selective angiograms obtained in various projections as well as with coaxial subselective injections via a microcatheter into several internal iliac artery branches, but without definitive demonstration of active extravasation. Further review of the images suggested the source to be from an atypical obturator artery originating from the left inferior epigastric artery (noted by reflux from an internal iliac artery injection); this was confirmed with selective external iliac angiography (Fig 2). The anomalous branch was then catheterized with the microcatheter and embolized with 510–700-μm polyvinyl alcohol particles (Boston Scientific, Natick, Massachusetts), followed by a Gelfoam (Pharmacia & Upjohn, New York, New York) slurry until hemostasis was achieved. Selective right internal iliac angiography also demonstrated smaller foci of extravasation from the obturator and internal pudendal arteries; these were embolized in a standard fashion. The patient remained hemodynamically stable throughout the procedure.
Figure 2.
Left external iliac artery injection demonstrates the source of bleeding to be from a variant obturator artery (*) originating from the inferior epigastric artery (IEA). Subselective coaxial catheterization was performed, and the corona mortis was subsequently embolized to stasis (not shown). EIA = external iliac artery.
He was then taken to the surgical intensive care unit, where his postprocedural hemoglobin level was 9 g/dL (90 g/L). Another 2 units of packed red blood cells was administered. The patient's hemoglobin level normalized, and he remained stable throughout the remainder of his hospital stay. He was extubated 2 days later and subsequently transferred to the basic care unit, where his chest tube was removed and physical therapy begun. He was discharged home 8 days after the trauma in stable condition.
Materials and Methods
Institutional review board approval was obtained for this study. The CT scans of 50 consecutive patients (≥18 years old) undergoing clinically indicated and ordered routine contrast-enhanced multidetector CT of the pelvis for all indications that included venous phase imaging were evaluated. Twenty-five patients were scanned with a 16-slice scanner (Lightspeed; GE Healthcare, Milwaukee, Wisconsin), and 25 patients were scanned concurrently with a 64-slice scanner (Lightspeed VCT; GE Healthcare). Sixteen-slice contrast-enhanced multidetector CT was performed per our standard pelvic CT protocol, as follows: 16 × 1.25-mm detector configuration, pitch of 1.375, table speed of 27.5 mm/rotation, 0.7-second gantry rotation, 140 kV, and automatic milliampere. Sixty-four-slice contrast-enhanced multidetector CT was also performed per our standard pelvic CT protocol, as follows: 64 × 0.625-mm detector configuration, pitch of 0.98, table speed of 39 mm/rotation, 0.5-second gantry rotation, 120 kV, and automatic milliampere. Omnipaque 300 (GE Healthcare, Princeton, New Jersey) was generally administered on a weight-based sliding scale at a rate of about 2.0 mL/sec with a scan delay of approximately 60–80 seconds. Contiguous 5-mm-thick axial images were constructed as per standard protocol and reported in a timely fashion. The technologists then reformatted the data into both axial (1.25/1.0 mm) and coronal (1.25/1.0 mm) data sets. Two radiologists with Certificates of Added Qualification in interventional radiology and expertise in abdominal-pelvic imaging (J.C.S., G.E.W.) and one interventional radiology fellow (B.H.B.) evaluated the images on an AGFA workstation (Ridgefield Park, New Jersey) for the presence or absence of arterial corona morti. The obturator arteries were identified in the obturator canal and followed back to their origins; they were classified as having typical anatomy if they originated from the anterior division of the internal iliac artery or a “corona mortis” variant if they originated from the external iliac artery system. It was important to differentiate between variant arterial and venous obturator vessels. Where we had disconcordant individual interpretations, we subsequently met to achieve consensus on all cases.
Results
Fifty patients (23 men, 27 women; age range, 18–87 years; mean age, 26.5 years) underwent contrast-enhanced multidetector CT of the pelvis for abdominal pain (n = 21), malignancy or tumor (n = 15), abscess (n = 6), and other miscellaneous reasons (n = 8). One hundred hemipelves were evaluated for the identification of the obturator arteries, with specific note of vessel origin. Eighteen of the 50 patients (36%) had a least one corona mortis variant. Thirteen of the 50 obturator arteries (26%) on the 16-slice studies and 16 (32%) on the 64-slice studies demonstrated a corona mortis variant; this difference was not statistically different (P = .51). In total, 29 of the 100 hemipelves (29%) demonstrated an arterial corona mortis variant. All corona morti identified were identified on the 1.25-mm axial images, and all were noted to originate from the ipsilateral inferior epigastric artery (Figure 3, Figure 4). Of note, four corona morti arteries appeared to have additional communications with a co-dominant obturator artery emanating from the internal iliac artery. The prevalence of the corona morti was fairly evenly distributed between the right and left sides (15/50 [30%] vs 14/50 [28%]). Interestingly, 11 of the 50 patients had bilateral corona morti. In total, 10 corona morti were identified on the 5-mm images, 25 on the 1.25-mm coronal images, and 29 on the 1.25-mm axial images. While meeting to achieve consensus on discrepant individual interpretations, the reviewers all agreed that the 64-slice images were subjectively easier to read than the 16-slice images, particularly for challenging cases (eg, patients with suboptimally enhanced or diminutive caliber corona morti and those with regional distortions [eg, from lymphadenopathy]).
Figure 3.
Sequential 1.25-mm thin axial (a–d) and coronal reformatted (e–h) images from contrast-enhanced CT demonstrate variant left obturator artery (arrows) originating from the inferior epigastric artery. Note the general posteromedial and inferior course of this corona mortis across the superior pubic ramus and into the obturator canal. For comparison, the typical right obturator artery (arrows) is noted to originate from the internal iliac artery distribution.
Figure 4.
Cut-away three-dimensional volume-rendered left posterior oblique view of the right hemipelvis demonstrates the vulnerable course of the right corona mortis (CM). EIA = external iliac artery, IEA = inferior epigastric artery, * = symphysis pubis.
The average diameter of the corona morti was 2.5 mm (range, 1.6–3.5 mm). The average distance from where the arteries crossed over the superior pubic rami to the symphysis pubis was 56 mm (range, 41–72 mm).
Discussion
Traumatic pelvic fractures are associated with a substantial mortality rate, with reported mortality rates of 8%–50% (3). Pelvic trauma alone accounts for 6%–8% of all trauma deaths (2). For this reason, most patients with significant pelvic trauma who reach the emergency department alive must expeditiously be clinically assessed, resuscitated, and stabilized. Most patients with substantial pelvic trauma undergo CT examination because it is rapid and gives a global overview of the pelvis, including exquisite detail of fractures, dislocations, and very sensitive detection of associated bleeding sites. In patients with active bleeding at CT and/or hemodynamic instability, endovascular embolization is considered the first line of treatment. It has been shown that CT examination can often prospectively depict the bleeding arterial culprit (1). The obturator artery has been reported to be the third most commonly injured (11) and fifth most embolized hemorrhaging pelvic artery from trauma (4). However, “normal” anatomy of the pelvic arteries has many variations that must be considered in the acute setting of pelvic hemorrhage (12, 13). One of the most common variations of pelvic arteries is the obturator, with variable origin from different internal iliac artery branches 70% of the time and from the external iliac artery system 30% of the time in one study of 640 cadaver halves (14).
When the obturator artery originates from the external iliac artery system, it passes relatively vertically by or over the femoral ring on the superior pubic ramus and along the lacunar ligament before exiting the pelvis via the obturator canal. It has been given the name “corona mortis” because it forms a vascular “crown” that has potential for life-threatening hemorrhage with ilioinguinal or acetabular surgeries and pelvic trauma (5, 7, 15, 16). There is, however, no consensus that the corona mortis is limited to only a replaced obturator artery originating from the external iliac artery system. Anatomically, this variant has also been described as representing either venous or arterial communication between the external iliac system and the obturator vessels that originate from or drain into the internal iliac artery or vein, respectively (15, 17, 18, 19). In addition, the venous variation has been reported to be higher in both prevalence and prominence than its arterial correlate (8, 10, 18). This may be a confounding variable in the recognition of the arterial corona mortis at CT, particularly during venous phase imaging as is typically performed with routine pelvic imaging for trauma. Although it has been reported that the venous corona mortis is more likely to cause clinically significant hemorrhage in the surgical setting (17), we agree with others who believe that this is unlikely to be a significant source of hemorrhage in the blunt trauma setting (11). For purposes of the remainder of this discussion about pelvic trauma, we will consider the corona mortis to represent a variant dominant obturator artery originating from the external iliac artery system without a corresponding obturator artery from the internal iliac artery (25% in our study), or less commonly a co-dominant obturator artery originating from the external iliac artery system where a typical obturator artery also emanates from the internal iliac artery (4% in our study).
Most commonly, the corona mortis originates from the inferior epigastric artery (14, 20, 21) and uncommonly directly from the external iliac artery (7, 9, 22). It characteristically crosses the superior pubic ramus 33–62 mm lateral to the symphysis pubis (5, 7, 8) and into the obturator foramen and is easily visualized at angiographic examination (5). In anatomic studies, the prevalence of the corona mortis (per hemipelvic side) has been reported as 1%–43% in cadaver studies (7, 9, 10, 12, 14, 15, 19, 20, 22), 29% in a reported angiographic study (5), and 10%–22% in several endoscopic studies (6, 17). Of note, the wide range of reported prevalences of the corona mortis may be related to the discretion of the authors at inclusion/exclusion diameter criteria, with ranges including “macroscopically seen;” less than 1 mm, more than 1 mm, and more than 2 mm in diameter; and as no reported size criteria (5, 7, 17, 19, 22). Significantly, most of the identified corona morti in our study were between 2.0 and 3.0 mm. Interestingly, it has been noted that the prevalence of identifying a clinically significant corona mortis may be much lower in clinical practice, that is, at the time of surgery. Darmanis et al (15) only found five vascular structures over the symphysis pubis in 492 anterior surgical approaches to the pelvis, and Letournel and Judet (23) encountered only one corona mortis large enough to necessitate clamping and ligation in their experience. It is possible that spasm, inadvertent laceration or cauterization, and retraction of the vessels may account for the discrepancy between the various methodologies for studying the true prevalence of the corona mortis.
Due to its relatively diminutive caliber and intimate association with bony structures, postprocessed CT images with three-dimensional rendering techniques are not likely to be routinely helpful in the demonstration of the corona mortis. In addition, 5-mm-thick axial images are probably, in general, too thick to reliably depict the corona mortis due to volume averaging with adjacent structures. With thick images, it is likely to be misinterpreted for a variant obturator vein. Although coronal-planar CT reformations have been shown to enhance reader confidence in the depiction of various abdominal-pelvic anatomy (24, 25), the overlapping 1.25-mm thin axial imaging set was most useful in our ability to diagnose the obturator arteries' origins from the mostly vertical (ie, perpendicular to the imaging plane) course of this vessel.
There have been few prior reports of embolization of a hemorrhaging corona mortis in the English literature (26, 27, 28, 29). None of these reports commented on the potential ability to prospectively identify a variant obturator artery at CT examination. This may save valuable time, contrast medium, radiation, and operator stress in the trauma scenario. In our presented case, initial CT examination demonstrated pelvic fractures and suggested obturator hemorrhage. At subsequent angiography, initial internal iliac artery angiograms were negative. In this case, delayed recognition of the normal variant origin of the bleeding obturator artery on external iliac angiograms fortunately did not have any demonstrable permanent detrimental effect. Even in retrospect, the corona mortis could not be confidently diagnosed on the routine 5-mm-thick axial images from admission CT. Unfortunately, the raw data from the CT study had been deleted at the time of the initiation of this study, thus preventing subsequent postprocessing to create thinner sections. For the interventionalist, it has been suggested to routinely perform external iliac angiography when hemorrhage in the obturator region is detected at preprocedure CT (3, 27). As demonstrated in this study, a small percentage of patients have co-dominant obturator arteries originating from both the internal and external iliac arterial systems, and “back door” bleeding may occur if the corona mortis is not properly identified and embolized.
Limitations of our study include the fact that no acute trauma patients were included in our study. In these patients, regional distortion due to fractures and hematomas, as well as a hypovolemic state that may lead to diminutive caliber of vascular structures, may render a corona mortis more difficult to confidently identify at CT. In addition, we had no surgical or pathologic confirmation of the presence or absence of the corona mortis variant. However, the relatively high prevalence in our study is within the range of that previously reported in surgical and pathologic studies. Our study population was fairly young, which likely limited the number of patients with significant atherosclerotic disease, which can limit the ability to detect vascular opacification within small calcified vessels. In addition, we did not specifically evaluate other CT scanners, scanning techniques, or reconstruction parameters. Finally, we did not specifically evaluate for the ability of CT to help diagnose the venous corona mortis variant, which may theoretically also have some clinical implications if it is traumatized.
The arterial corona mortis is a common variant and potential significant source of hemorrhage in the setting of acute pelvic trauma. This variant was identified on thin, 1.25-mm-thick images in approximately one-third of the patients in this study but was missed on two-thirds of these cases when only 5-mm-thick images were evaluated. Its high prevalence suggests that when obturator arterial injury is initially suspected at CT due to acetabular or pubic rami fractures and adjacent hemorrhage, external iliac angiography is prudent. All identified corona morti in our study originated from the inferior epigastric artery. Thin (1.25-mm) CT reconstructions may be helpful in diagnosing a corona mortis prospectively; however, further study specifically in the trauma patient population would be necessary to confirm this.
References
- . Pelvic arterial hemorrhage in patients with pelvic fractures: detection with contrast-enhanced CT. Radiographics. 2004;24:1591–1606
- . Traumatic injuries: radiological hemostatic intervention at admission. Eur Radiol. 2002;12:979–993
- . Transcatheter embolization in pelvic trauma. Semin Intervent Radiol. 2004;21:23–35
- . Angiographic embolization for intraperitoneal and retroperitoneal injuries. World J Surg. 2000;24:539–545
- . Corona mortis: incidence and location. Arch Orthop Trauma Surg. 2002;122:163–164
- . A prospective endoscopic study of retropubic vascular anatomy in 121 patients undergoing endoscopic extraperitoneal inguinal hernioplasty. Surg Endosc. 2003;17:1376–1379
- . An anatomical study of corona mortis and its clinical significance. Chin J Traumatol. 2004;7:165–169
- . Corona mortis: incidence and location. Clin Orthop. 1996;329:97–101
- . Anastamotic vessels in the retropubic region: corona mortis. Folia Morphol. 2003;62:179–182
- . Variability of the obturator vessels. Clin Anat. 1997;10:328–332
- . Vascular injuries in blunt pelvic trauma. Radiol Clin North Am. 1981;19:171–186
- . Variations in origin of the parietal branches of the internal iliac artery. J Anat. 1952;86:423–430
- . Diagnostic angiography. In: 2nd ed.. Philadelphia, Pa: W.B. Saunders; 1991;p. 98–100
- . The origin of the obturator artery: a study of 640 body halves. Am J Anat. 1942;70:317–343
- . Corona mortis: an anatomical study with clinical implications in approaches to the pelvis and acetabulum. Clin Anat. 2007;20:433–439
- . The treatment of acetabular fractures through the ilioinguinal approach. Clin Orthop. 1993;292:62–76
- Surg Endosc. 2001;15:7275
- . The incidence and location of corona mortis: a study on 75 cadavers. Acta Orthop Scand. 2004;75:53–55
- . Vascular anatomy over the superior pubic rami in female cadavers. Am J Obstet Gynecol. 2005;193:2165–2168
- . Retropubic vascular hazards of the ilioinguinal exposure: a cadaveric and clinical study. J Orthop Trauma. 1996;10:156–159
- Gray's anatomy of the human body, Bartelby.com edition. Available at http://education.yahoo.com/reference/gray/subjects/subject/155. Accessed March 3, 2008
- . Variations of the pubic vascular anastamoses in black South Africans. Acta Anat. 1996;155:212–214
- . Surgical approaches to the acetabulum. In: Reginald AE editors. Fractures of the acetabulum. 2nd ed.. Berlin: Springer-Verlag; 1993;p. P381
- . MDCT of patients with acute abdominal pain: a new perspective using coronal reformations from submillimeter isotropic voxels. Am J Roentgenol. 2004;183:899–906
- . Experience with a continuous, single-pass whole-body multidetector CT protocol for trauma: the three-minute trauma CT scan. Emerg Radiol. 2001;8:250–256
- . Hemorrhage secondary to pelvic fracture: coil embolization of an aberrant obturator artery. J Endovasc Ther. 2003;10:676–680
- . Angiography and embolization of the corona mortis (aberrant obturator artery). Fortschr Rontgenstr. 1984;141:708–710
- . Avulsion of the pubic branch of the inferior epigastric artery: a cause of hemodynamic instability in minimally displaced fractures of the pubic rami. J Trauma. 2000;49:750–753
- . Hemodynamic instability following an avulsion of the corona mortis artery secondary to a benign pubic ramus fracture. J Trauma. 2007;62:E14–E17
From the 2007 SIR annual meeting.
None of the authors have identified a conflict of interest.
PII: S1051-0443(09)00009-8
doi:10.1016/j.jvir.2009.01.007
© 2009 SIR. Published by Elsevier Inc. All rights reserved.
Volume 20, Issue 4 , Pages 455-460, April 2009
