| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 All authors: Department of Abdominal Radiology, New York University Medical Center, Tisch Hospital, Rm. HW202, 560 First Ave., New York, NY 10016.
Received June 12, 2001;
accepted after revision August 7, 2001.
Address correspondence to G. L. Bennett.
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. Four observers retrospectively reviewed CT scans in 75 patients (23 with acute gangrenous cholecystitis, 25 with acute nongangrenous cholecystitis, and 27 without cholecystitis). The following findings were evaluated: distention, mural thickening, wall enhancement, irregular wall, wall striation, intraluminal membranes, pericholecystic inflammation, gallstones, pericholecystic fluid, enhancement of liver parenchyma, pericholecystic abscess, and gas in the wall or lumen. Sensitivity and specificity of CT for gangrenous cholecystitis and for each finding were calculated. Two reviewers in consensus measured gallbladder dimension and wall thickness. Logistic regression models were used to predict gangrenous versus nongangrenous cholecystitis.
RESULTS. Sensitivity, specificity, and accuracy of CT for acute cholecystitis were 91.7%, 99.1%, and 94.3%, respectively, and for acute gangrenous cholecystitis were 29.3%, 96.0%, and 64.1%, respectively. Findings with the highest specificity for gangrenous cholecystitis were gas in the wall or lumen (100%), intraluminal membranes (99.5%), irregular or absent wall (97.6%), and abscess (96.6%). The difference between the mean gallbladder wall thickness and the short-axis dimension for the two groups with cholecystitis was statistically significant. In three patients with gangrenous cholecystitis, no mural enhancement was seen. Pericholecystic fluid also achieved statistical significance for the diagnosis of gangrene. Multivariate logistic regression analysis showed that the overall accuracy of CT for gangrenous cholecystitis was 86.7%.
CONCLUSION. CT findings most specific for acute gangrenous cholecystitis are gas in the wall or lumen, intraluminal membranes, irregular wall, and pericholecystic abscess. Gangrenous cholecystitis is associated with a lack of mural enhancement, pericholecystic fluid, and a greater degree of gallbladder distention and wall thickening.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Both sonographic and CT findings of acute gangrenous cholecystitis have been previously described [4,5,6,7,8,9,10]. CT findings include intraluminal membranes, hemorrhage into the lumen, and irregular or absent wall, a finding best observed with contrast-enhanced CT [11]. However, the CT findings in large series of patients with gangrenous cholecystitis compared with those of uncomplicated acute cholecystitis and normal patients have not been described, and, to our knowledge, the accuracy of CT for diagnosing this complication is unknown.
The purpose of this study was to describe the CT findings in a large series of patients with acute gangrenous cholecystitis and to compare CT findings in these patients with those of patients with acute nongangrenous cholecystitis. Our goal was to identify those CT findings that would allow for an accurate preoperative diagnosis of gangrenous cholecystitis, so that the surgeon may be informed of this potential complication.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CT Technique
CT scans were obtained using helical technique with either a HiSpeed or CTI
scanner (General Electric Medical Systems, Milwaukee, WI). All patients
included in the study received IV contrast material, with injection rates of 2
or 3 mL/sec. Slice thickness varied from 3 to 10 mm. In the gangrenous
cholecystitis group, slice thickness in 16 patients was 7 mm, in three
patients was 3 mm, and in two patients each was 5 and 10 mm. In the
nongangrenous acute cholecystitis group, slice thickness in 11 patients was 3
mm, in seven patients was 7 mm, in four patients was 5 mm, and in three
patients was 10 mm. In the group without cholecystitis, slice thickness was 7
mm in 24 patients, 5 mm in two patients, and 3 mm in one patient.
Radiologic Interpretation
The studies were randomized and retrospectively reviewed individually by
four experienced abdominal radiologists unaware of the pathologic diagnosis.
The reviewers were informed that the cases included a mix of patients with
normal gallbladders, acute uncomplicated cholecystitis, and gangrenous
cholecystitis. The reviewers were asked to identify the presence of the
following 11 findings: gallbladder distention (qualitative assessment),
gallbladder wall thickening (>3 mm), gallstones, pericholecystic fluid,
pericholecystic fat stranding (increased soft-tissue density in the
pericholecystic fat), striated wall (alternating areas of low and high
attenuation), focally increased enhancement of adjacent liver parenchyma,
pericholecystic abscess (encapsulated fluid collection adjacent to the
gallbladder), intraluminal membranes (irregular intraluminal linear and
soft-tissue densities), irregular or absent wall (irregular contour or
enhancement), and gas in the wall or lumen. Reviewers were also asked to
assess for the presence or absence of contrast enhancement of the gallbladder
wall, to decide if acute cholecystitis was present, and to rank the likelihood
of gangrenous cholecystitis on a 5-point scale as follows: 1, definitely not
gangrenous cholecystitis; 2, probably not gangrenous cholecystitis; 3, not
sure; 4, probably gangrenous cholecystitis; 5, definitely gangrenous
cholecystitis. These rankings were based on assessment of the overall severity
of inflammatory changes and pericholecystic abscess or fluid collection. To
calculate overall sensitivity and specificity of CT for gangrenous
cholecystitis, a score of 4 or 5 was considered positive for gangrenous
cholecystitis.
Two additional observers (not included in the previously mentioned four and also unaware of the pathologic diagnosis) reviewed the images in consensus to determine the gallbladder dimensions for all three groups (short axis and long axis) and the gallbladder wall thickness. The long-axis and short-axis dimensions of the gallbladder and wall thickness were measured from outer wall to outer wall using hand held calipers. These two reviewers also performed a consensus evaluation of the presence of each of the CT findings described previously to determine the frequency of each finding in the patients with gangrenous cholecystitis versus those with uncomplicated acute cholecystitis and the normal controls.
Statistical Methods
Data from the four reviewers were used to assess the overall sensitivity,
specificity, and accuracy of a CT diagnosis of acute cholecystitis and for
acute gangrenous cholecystitis. Values were calculated separately for each
reviewer and also using combined observations. Sensitivity and specificity for
each of the individual findings described previously were also calculated on
the basis of combined reviews. The association between each of the 11 CT
findings and the presence of gangrenous cholecystitis was evaluated using
chisquare tests. Bonferroni correction was used to compensate for multiple
comparisons. The short-axis and long-axis dimensions and the gallbladder wall
thickness for the three groups were compared using the t test, with a
p value of 0.05 deemed to be significant. Logistic regression models
were used to predict acute cholecystitis versus normal gallbladder and
gangrenous versus nongangrenous cholecystitis on the basis of data from the
consensus observations of two reviewers.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
|
|
|
|
|
|
|
|
|
|
|
|
In three patients with gangrenous cholecystitis, there was no observable IV contrast enhancement of the gallbladder wall (Fig. 2). However, scans of all patients with uncomplicated acute cholecystitis showed mural enhancement. Additionally, in a total of 14 patients (three patients with gangrenous cholecystitis and 11 patients with acute nongangrenous cholecystitis), an enhancing vessel in the wall of the gallbladder was incidentally noted (Fig. 5). These patients were studied during rapid infusion of IV contrast material (3 mL/sec) with 3 mm collimation. This finding was not observed in any of the patients without cholecystitis.
The mean gallbladder size in short-axis and long-axis dimension and the mean gallbladder wall thickness determined from the consensus observation are given in Table 3. A statistically significant difference was seen in the size of the gallbladder in short axis for the gangrenous gallbladders versus the nongangrenous gallbladders (p = 0.01) and between both groups and the normal gallbladders (p < 0.001). No statistical difference was found between the two groups with cholecystitis with respect to long-axis dimension (p = 0.12), although a statistically significant difference was found between the two groups with cholecystitis with respect to wall thickness (p = 0.02).
|
Excellent prediction of acute cholecystitis is possible using a multivariate logistic regression model involving CT characteristics. The overall accuracy of such a model was 95.8%; sensitivity, 95.6%; and specificity, 96.3%. The forward selection based on conditional parameter removal revealed that virtually identical prediction is possible using the combination of only four key characteristics: wall thickness (p < 0.0001), the short-axis dimension (p < 0.0001), the presence of stones (p = 0.006), and the presence of fluid (p = 0.05). Prediction of gangrenous versus nongangrenous acute cholecystitis using a multivariable logistic regression model was also possible but slightly less accurate. The best model achieved an overall accuracy of 86.7%, sensitivity of 85.7%, and specificity of 87.5%. The key independent variables were the presence of an irregular wall (p = 0.0006), lack of wall enhancement (p = 0.0008), measured wall thickness (p = 0.008), and the short-axis dimension of the gallbladder (p = 0.03). If the measured dimensions were not available, then a model based on remaining characteristics showed a sensitivity decrease to the 72.7% level.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Gangrenous cholecystitis is a severe form of acute cholecystitis, which results from marked distention of the gallbladder with increased tension in the wall. Associated inflammation leads to ischemic necrosis of the gallbladder wall, with or without associated cystic artery thrombosis [17]. The incidence of gangrenous cholecystitis ranges from 2% to 29.6% in various surgical series [1,2,3, 18] and generally occurs in older patients. There is an increased incidence in men and in patients with coexisting cardiovascular disease and leukocytosis greater than 17,000 WBC/mL [18, 19]. Once the disease is suspected, patients with gangrenous cholecystitis generally undergo emergency cholecystectomy or cholecystostomy to avoid life-threatening complications. These patients frequently require an open surgical procedure rather than laparoscopic cholecystectomy, with conversion rates from laparoscopic to open cholecystectomy ranging from 8% to 75% [19,20,21,22].
The hallmark on sonography of gangrenous cholecystitis is the presence of heterogeneous or striated thickening of the gallbladder wall, which is often irregular with projections into the lumen and pericholecystic fluid collections [4,5,6]. The presence of intraluminal membranes representing desquamative gallbladder mucosa is a specific finding but is less common. Sonographic findings typical of uncomplicated acute cholecystitis may be absent in this subset of patients. In a recent surgical series of patients with gangrenous cholecystitis, 28% of patients had a sonogram that was not diagnostic of gallbladder inflammation, mainly because of an absent sonographic Murphy sign and gallbladder wall thickness of less than 3 mm [18]. Another report found the sonographic Murphy sign positive in only 33% of these patients [23]. It is postulated that this finding is related to denervation of the gallbladder wall.
CT findings in gangrenous cholecystitis have also been described, including intraluminal membranes, hemorrhage into the lumen, and irregular or absent wall [7,8,9,10,11]. However, the overall accuracy of CT for the diagnosis of acute gangrenous cholecystitis and the sensitivity and specificity of these findings have not been well evaluated previously.
In our series, CT had high sensitivity and specificity for the diagnosis of acute cholecystitis with an overall accuracy of 94.3%. These findings support the conclusion that CT is a reliable imaging modality to establish a confident diagnosis of acute cholecystitis. CT is also highly specific (96.0%) for identifying patients with acute gangrenous cholecystitis but is insensitive (29.3%). The findings most specific for the diagnosis of gangrenous cholecystitis are gas in the wall or lumen, intraluminal membranes, irregular or absent wall, and pericholecystic abscess. Therefore, when these findings are present, the possibility of gangrenous cholecystitis should be strongly considered, particularly for an irregular gallbladder wall, which achieved high specificity and statistical significance. Pericholecystic fluid also achieved statistical significance for the diagnosis of gangrenous cholecystitis, although the specificity was lower, likely caused by the small number of observations. However, these results indicate that the presence of pericholecystic fluid should also be considered predictive of the severity of acute cholecystitis. Mural striation and adjacent hepatic enhancement also achieved relatively high specificity. Unfortunately, the absence of these findings is not as helpful given the low sensitivities achieved.
Our results also show that the greater the degree of gallbladder distention in the short axis and wall thickening, the greater the likelihood of gangrenous cholecystitis. The average dimension of the long axis of the gangrenous gallbladder was 0.9 cm larger than the nongangrenous one, whereas the short axis was 0.7 cm larger. However, only the short axis proved to be statistically significant in discriminating gangrenous from acute nongangrenous cholecystitis. This finding is due to poorer precision of estimating the long axis, with a standard deviation more than twice as large as that for the short axis, and may be related, in part, to limitations related to variable slice thickness.
Logistic regression models show that CT characteristics provide almost perfect (only three of 75 cases misclassified) prediction of acute cholecystitis. Moreover, 96% accuracy was possible using a reduced model involving the knowledge of wall thickness and the short-axis dimension of the gallbladder, the presence of stones, and fluid. Many other variables correlated strongly with the presence of cholecystitis and could be used to construct other predictive models at 95% accuracy level. Prediction of gangrenous cholecystitis is also possible. Good sensitivity (86%) can be achieved using all CT-derived characteristics. Using this model, we identified the key independent variables: irregular wall, lack of wall enhancement, wall thickness, and short-axis dimension.
Our data support the conclusion that if contrast-enhanced CT shows irregular wall enhancement, this finding is specific for gangrenous cholecystitis, as noted by other authors [11]. In our series, complete lack of gallbladder wall enhancement was observed only in those patients with gangrenous cholecystitis, but the total number of patients was too small for statistical analysis. It is apparent that the ability of CT to discriminate between gangrenous and nongangrenous cholecystitis is improved when patients are studied after IV administration of contrast material. An interesting observation was the presence of an enhancing vessel in the wall of the gallbladder in 14 patients with acute cholecystitis and in no patients without cholecystitis. This difference was observed on studies performed with rapid bolus infusion of IV contrast material with 3-mm slice thickness. This finding may be analogous to observations of the cystic artery made on Doppler sonography in patients with acute cholecystitis [24] and may be an interesting subject for further investigation.
Our study has recognized limitations, largely because of the retrospective nature, including bias of the reviewers to cases with positive findings. However, the reviewers knew that normal patients were also included in the mix of cases they were reviewing. There is also likely selection bias because the patients included in both study groups represented only a small percentage of the total number of patients with cholecystitis treated during the study period. The indications for imaging with CT are not known, and possibly only the most ill patients were imaged. Furthermore, there was a variation in time interval between CT scanning and the time of cholecystectomy. However, no statistically significant difference was found between the two study populations, and the mean time to surgery was shorter for the patients with gangrenous cholecystitis. This finding would impact on overall sensitivity and sensitivity of each finding, but not the specificity. Also, there were variations in the CT techniques used with respect to contrast-injection rates and slice thickness, which likely impact our ability to identify findings such as mural striation, intraluminal membranes, and adjacent hepatic enhancement. Last, the total number of observations for many of the findings evaluated in this study was small, limiting assessment of statistical significance.
In conclusion, air in the gallbladder wall or lumen, irregular or absent gallbladder wall, intraluminal membranes, pericholecystic abscess, and lack of gallbladder wall enhancement are specific CT findings of acute cholecystitis complicated by gangrene. Marked gallbladder wall thickening and distention in the short axis, particularly when combined with the previously mentioned findings, are also highly suggestive of gangrenous cholecystitis. Increased contrast enhancement in the adjacent liver parenchyma, wall striation, and pericholecystic fluid are less specific but also useful findings to predict the severity of cholecystitis. When these findings are observed on CT, the possibility of gallbladder necrosis can be suggested. The surgeon can then be informed that more immediate surgical intervention may be necessary with higher possibility of conversion to open cholecystectomy.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  J. Stoker, A. van Randen, W. Lameris, and M. A. Boermeester Imaging Patients with Acute Abdominal Pain Radiology, October 1, 2009; 253(1): 31 - 46. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. A. Smith, J. R. Dillman, K. M. Elsayes, C. O. Menias, and R. O. Bude Cross-Sectional Imaging of Acute and Chronic Gallbladder Inflammatory Disease Am. J. Roentgenol., January 1, 2009; 192(1): 188 - 196. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Altun, R. C. Semelka, J. Elias Jr, L. Braga, V. Voultsinos, J. Patel, N. C. Balci, and J. T. Woosley Acute Cholecystitis: MR Findings and Differentiation from Chronic Cholecystitis Radiology, July 1, 2007; 244(1): 174 - 183. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Watanabe, M. Nagayama, A. Okumura, Y. Amoh, T. Katsube, T. Suga, S. Koyama, K. Nakatani, and Y. Dodo MR Imaging of Acute Biliary Disorders RadioGraphics, March 1, 2007; 27(2): 477 - 495. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. M. Rucker, C. O. Menias, and S. Bhalla Mimics of Renal Colic: Alternative Diagnoses at Unenhanced Helical CT RadioGraphics, October 1, 2004; 24(suppl_1): S11 - S28. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. E. Hanbidge, P. M. Buckler, M. E. O'Malley, and S. R. Wilson From the RSNA Refresher Courses: Imaging Evaluation for Acute Pain in the Right Upper Quadrant RadioGraphics, July 1, 2004; 24(4): 1117 - 1135. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R Zissin, A Osadchy, M Shapiro-Feinberg, and G Gayer CT of a thickened-wall gall bladder Br. J. Radiol., February 1, 2003; 76(902): 137 - 143. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
