| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Clinical Observations |
1 Department of Radiology, University of Pittsburgh Medical Center, 200 Lothrop
St., Room 3950, Pittsburgh, PA 15213.
2 Present address: Department of Radiology, Chang Gung Memorial Hospital,
Taipei, Taiwan.
Received October 28, 2005;
accepted after revision May 16, 2006.
Address correspondence to M. P. Federle
(federlemp{at}upmc.edu).
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CONCLUSION. Attention to specific CT criteria allows accurate diagnosis of the specific cause of spontaneous abdominal hemorrhage even in patients who have both coagulopathy and an abdominal aortic aneurysm.
Keywords: abdomen • abdominal imaging • aorta • CT
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CT has long been acknowledged as the best means of diagnosing abdominal hemorrhage, but there has been little investigation into the frequency and accuracy of various CT findings in diagnosing and distinguishing hemorrhage resulting from a ruptured aneurysm versus coagulopathy [1-10].
The purpose of our study was to test the hypothesis that the use of existing and new CT criteria would allow confident diagnosis of the specific cause of abdominal hemorrhage in patients with coagulopathy or abdominal aortic aneurysm.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Institutional review board approval was granted for this retrospective review. Patient consent was neither required nor obtained.
Sixty-three patients met the entrance criteria of having a discharge diagnosis of abdominal hemorrhage attributed to coagulopathy or ruptured aortic aneurysm. These included 40 men and 23 women ranging in age from 26 to 97 years (mean age, 62 years).
Medical records were reviewed for clinical or laboratory evidence of coagulopathy including the use of anticoagulant or antiplatelet medications such as heparin, warfarin, or Plavix (clopidogrel bisulfate, Sanofi-Synthelabo) and documentation of a prolonged international normalized ratio (INR), prolonged partial thromboplastin time (PTT), or prolonged prothrombin time (PT).
There were 46 patients who had abdominal hemorrhage attributed to coagulopathy in the discharge summary of their medical records. Forty patients were receiving anticoagulant or antiplatelet medications. Two patients had hemophilia (factor VIII deficiency), and four patients had thrombocytopenia or prolonged PTT of uncertain cause.
Seventeen patients had hemorrhage attributed to a ruptured AAA. In 13 of these patients, the diagnosis was confirmed at surgery immediately after CT evaluation. Four patients died without surgery soon after CT evaluation because the patient or family members refused surgery.
The honest broker prepared and presented all CT images and medical information in a coded, unidentified manner, leaving the investigators aware only of the age and sex of the patients.
CT was performed on one of several MDCT scanners with contiguous 5-mm-thick axial images constructed and displayed from the diaphragm to the symphysis pubis. Contrast material ([ioversol] Optiray 350, Mallinckrodt) was administered IV in 20 of 63 patients at a volume of 125 mL and a rate of 3 mL/s. Our department's CT protocol for evaluation of a potential ruptured aneurysm or spontaneous abdominal hemorrhage calls for sections through the abdomen and pelvis without the use of oral or IV contrast material, whereas scans to evaluate the source of abdominal pain are acquired with both oral and IV contrast material. Scans were obtained either all with or all without the use of IV contrast material, except for two patients who underwent CT before and after IV contrast administration. Oral contrast medium was given to 18 patients.
CT scans were reviewed in consensus on a PACS workstation by two radiologists experienced in abdominal imaging. The reviewers were blinded to all clinical information except the age and sex of the patient.
One of the reviewers had not been the initial recorder of any of the clinical CT scans. The other reviewer was one of 11 staff radiologists who interpret abdominal CT scans and may have interpreted some of the original CT scans. (The exact number is uncertain because of the blinded nature of this review.) Each CT scan was evaluated on the basis of the reviewers' prior personal experience and published criteria for the following signs: location of the abdominal hemorrhage, "hematocrit effect" [7] (linear separation of cellular and liquid components of blood), and active extravasation of vascular contrast material [8]. Patients with aneurysms were evaluated for additional CT findings. Location of hemorrhage was specified as involving one or more of the following compartments: rectus abdominis sheath, iliopsoas muscle group, or retro-peritoneum (including the perirenal, anterior, and posterior pararenal spaces) individually or in combination. Iliopsoas hemorrhage was defined by enlargement of the muscle group, with or without adjacent stranding of retroperitoneal fat planes. Hemorrhage that appeared to be primarily in the retroperitoneum but only touching the iliopsoas compartment was classified as retroperitoneal rather than iliopsoas involvement.
The two CT reviewers attempted to assign a cause to the abdominal hemorrhage using the following CT criteria, again on the basis of a combination of prior experience and published criteria. Hemorrhage due to coagulopathy was diagnosed on CT based on the finding of hemorrhage with the hematocrit effect, bleeding isolated to or involving the rectus sheath, bleeding isolated to the iliopsoas compartment, or active extravasation at a site distant from the aorta. Diagnosis of coagulopathic hemorrhage required at least two of these findings to be present.
Bleeding from a ruptured AAA was diagnosed on CT by the presence of an aneurysm greater than 4 cm in diameter [4] with hemorrhage contiguous to the aorta for a length of at least 3 cm in addition to at least one of these additional criteria: the crescent sign [10] (hyperattenuating foci within luminal thrombus or plaque), an eccentric or draped aorta [9], discontinuity of a calcified wall, or contrast extravasation adjacent to the aorta.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Without knowledge of the clinical data or outcome, the CT reviewers correctly identified 46 of 46 patients who had abdominal hemorrhage due to coagulopathy and 15 of 17 patients who had hemorrhage due to a ruptured AAA. There were two patients who had CT evidence of an AAA and abdominal hemorrhage with findings that the reviewers considered equivocal for a coagulopathic source of hemorrhage.
Of the 46 patients with coagulopathic hemorrhage, 42 patients had no AAA (or other apparent cause of bleeding). The patients were analyzed as a separate group—group 1 (Table 1); 36 patients (86%) had CT evidence of the hematocrit effect (Fig. 1A, 1B, 1C). Fourteen patients (33%) had bleeding isolated to (n =4) or involving (n = 10) the rectus sheath. Fifteen patients (36%) had bleeding isolated to (n =1) or involving (n = 14) the iliopsoas compartment. Thirteen patients (31%) had bleeding into one or more retroperitoneal compartments, whereas no coagulopathic patients in group 1 had bleeding contiguous with a long (> 3 cm) segment of the abdominal aorta. Two patients also had intraperitoneal hemorrhage in addition to other sites. Seven patients in group 1 had an IV bolus injection of contrast material, and active extravasation distant from the aorta was identified in three of these patients (43%) (Figs. 2A, 2B and 3A, 3B). All patients in group 1 were correctly diagnosed as having coagulopathic hemorrhage on the basis of CT findings.
|
|
|
|
|
|
|
|
Of the 17 patients who had hemorrhage attributed to a ruptured AAA, 12 patients had only an AAA and no clinical evidence of coagulopathy and are designated as group 2 (Table 1). Only one of these patients had a small (2.5 x 1.5 cm) hematoma with a hematocrit effect, whereas all had blood contiguous with the aorta (> 3 cm) and none had bleeding into the rectus sheath (Fig. 4A, 4B). Other signs of aneurysmal rupture in group 2 included the crescent sign in nine of the 12 (75%), a draped or eccentric aorta in 12 (100%), and discontinuity of a calcified aortic wall in 11 (92%). Eight of these patients had an IV contrast injection, and active extravasation near the aorta was identified in three of these patients (38%). None of the patients with a ruptured AAA had bleeding isolated to the iliopsoas compartment, whereas all had bleeding into one (17%) or more (83%) retroperitoneal spaces. Ten patients had retroperitoneal bleeding abutting the iliopsoas compartment but were judged to be bleeding primarily into the periaortic and retroperitoneal spaces rather than the iliopsoas.
|
|
Eleven of the 12 patients in group 2 were correctly diagnosed as having hemorrhage due to a ruptured AAA. The one patient who had the small hematoma with a hematocrit sign was considered equivocal.
Nine patients had an AAA and clinical evidence of coagulopathy and are designated as group 3 (Table 1). The hemorrhage in five of these patients was proven at surgery to be due to a ruptured AAA, whereas four patients had hemorrhage attributed by their physicians to coagulopathy. Of the five patients in group 3 with a ruptured AAA, all had hemorrhage contiguous with the aorta and bleeding into one or more retroperitoneal spaces.
|
|
Of the four patients in group 3 who had a coagulopathic bleed, all four patients had evidence of hematocrit effect and only one had blood contiguous with a long segment of the aorta. Three of these four patients (75%) had bleeding into the iliopsoas compartment. One patient in this group had active extravasation distant from the aorta. One patient in group 3 who was diagnosed with a coagulopathic hemorrhage had had a prior rupture of an AAA treated by stent-graft. In the postoperative period, he had a coagulopathic hemorrhage correctly identified on CT because of the hematocrit effect sign despite the presence of blood contiguous with the aorta (Fig. 5A, 5B).
Adding the four patients in group 3 who had the hematocrit sign of coagulopathy to the 36 patients in group 1, 40 of 46 patients (87%) had this CT sign of coagulopathic hemorrhage. Eight of the nine patients in group 3 (89%) were correctly diagnosed by CT findings as having hemorrhage due to coagulopathy or AAA, with the one case considered equivocal.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We focused our attention on identifying CT criteria that would allow confident diagnosis of abdominal hemorrhage due to coagulopathy versus that due to a ruptured AAA. In most cases, the combination of clinical information and CT findings has allowed confident diagnosis, with diagnosis of retroperitoneal hemorrhage being among the earliest applications of body CT in 1977 [5]. With the aging population and the increased use of medications with variable anticoagulation effects (including aspirin, nonsteroidal antiinflammatory drugs, Plavix, heparin, and warfarin), it is inevitable that radiologists will be confronted with CT evidence of abdominal hemorrhage in patients who have an AAA and who may also be coagulopathic. Determining whether the patient is bleeding from a ruptured AAA or from coagulopathy is critically important in deciding on surgical or medical intervention.
Relying on clinical or laboratory data alone in this decision is problematic because the radiologist frequently lacks crucial information about the hematologic status of the patient including clotting factors (such as INR) and serum hematocrit. Moreover, it is well established that patients may have spontaneous abdominal hemorrhage even when the standard measures of clotting function are within normal or therapeutic range [2]. If, therefore, CT can offer reliable evidence of not only the presence but also the cause of abdominal hemorrhage, there would be important clinical value.
Our results indicate that CT is accurate in this setting. We retrospectively, but blindly, studied the CT scans of a large group of predominantly elderly patients, many of whom had aortic aneurysms, and were able to make an accurate judgment as to whether the abdominal hemorrhage present in these patients was due to a ruptured aneurysm or coagulopathy.
CT findings of a coagulopathic hemorrhage consist of hematoma with the hematocrit effect (present in 87%) and bleeding involving the rectus sheath or limited to the iliopsoas compartment. The hematocrit effect or sign (cellular-fluid level) was first noted by investigators from the Mayo Clinic [7] 23 years ago but has received little attention or investigation in the interim. This sign indicates settling of cellular elements in the dependent position of a hematoma and is distinguished from the more heterogeneous clotted blood found in patients with normal coagulation. We found this to be a highly sensitive and specific sign of a coagulopathic hemorrhage. This finding is rarely associated with a ruptured AAA (only one of our subjects). Bleeding that involves the retroperitoneal compartments is seen with either source of hemorrhage, and the iliopsoas compartment may be involved along with the retroperitoneum in either source. Because the various retroperitoneal compartments communicate to some extent, this is not surprising [13].
Active extravasation of contrast material can be seen with either source also, but its exact site is important because coagulopathic patients usually have bleeding distant from the abdominal aorta. Only 20 of the patients in our study had IV contrast-enhanced CT, and active extravasation was noted in eight patients. Because the site of extravasation may be an important finding, this may suggest a more important role for IV administration of contrast material in the setting of spontaneous abdominal hemorrhage.
CT can routinely depict aortic aneurysms, and it is well known that the risk of aneurysmal rupture is related to diameter, with those less than 5 cm in diameter very unlikely to bleed. Signs of impending or limited rupture are also familiar to most radiologists and include the draped aorta [9], crescent sign [10], and interruption of a continuous ring of aortic wall calcification. These signs were present in more than 80% of our subjects with a ruptured AAA. In our experience (not published), we have observed that a ruptured AAA results in hemorrhage that is contiguous with the aorta over a substantial length of the vessel. Having chosen a length of > 3 cm of contiguity as a criterion, we found that this sign was present in all 17 patients who had a ruptured AAA and was not present in any of the patients with a coagulopathic hemorrhage, except for one patient who had surgical repair of a ruptured AAA before a coagulopathic hemorrhage.
Bleeding from a ruptured AAA never involves the rectus sheath, uncommonly involves the iliopsoas compartment, and almost always involves multiple retroperitoneal spaces. Because the iliopsoas and retroperitoneal compartments are contiguous, it may be difficult to exclude some iliopsoas involvement in a patient with a ruptured AAA, but the retroperitoneal involvement is the main finding. Active extravasation, when present, is from the site of the aneurysmal leak. Only one patient with a ruptured AAA had a hematoma with a hematocrit effect, although his hematoma was small (2.5 x 1.5 cm). This patient also had renal and cardiac failure, which are known to predispose to clotting disorders, and may have had a small coagulopathic bleed in addition to the ruptured AAA.
As with any retrospective study, ours has limitations. Surgical proof was impossible to obtain in many of our subjects and proof of coagulopathic hemorrhage involved inclusion of some of the CT findings that we were attempting to assess. We eliminated from consideration those patients who had abdominal hemorrhage from trauma or ruptured neoplasms. Although this could theoretically introduce some selection bias, we believe that hemorrhage in these settings is easily attributable to its source on the basis of CT and clinical criteria. Nevertheless, we believe that there was compelling clinical proof of diagnosis in both the coagulopathic and aneurysm patients.
In summary, we have described CT criteria that allow confident diagnosis of hemorrhage due to coagulopathy, even in patients with an abdominal aortic aneurysm.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
