| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Department of Radiology, Harvard Medical School, Massachusetts General
Hospital, 32 Fruit St., Boston, MA 02114.
2 Department of Radiology, Center for Molecular Imaging Research, Massachusetts
General Hospital/Harvard Medical School, CNY 149, 13th St., R.5406,
Charlestown, MA 02129.
Received October 11, 2001;
accepted after revision December 27, 2001.
Presented at the annual meeting of the American Roentgen Ray Society,
Seattle, April—May 2001.
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. Between March 1997 and September 1999, 312 patients with suspected diverticulitis were examined on helical CT using rectally administered colonic contrast material. CT scans that were positive for diverticulitis or indeterminate were rereviewed by two radiologists; CT interpretations were correlated with patients' clinical courses and surgical findings.
RESULTS. One hundred fourteen (37%) of the 312 CT scans were interpreted as positive for diverticulitis; 192 scans (61%), as negative; six scans (2%), as indeterminate. Of the 114 scans that were positive for diverticulitis, 109 (96%; sensitivity 96%, specificity 91%) showed bowel wall thickening; 108 (95%; sensitivity 96%, specificity 90%), fat stranding; 104 (91%; sensitivity 91%, specificity 67%), diverticula; 57 (50%; sensitivity 50%, specificity 100%), fascial thickening; 51 (45%; sensitivity 45%, specificity 97%), free fluid; 49 (43%; sensitivity 43%, specificity 100%), inflamed diverticula; 34 (30%; sensitivity 30%, specificity 100%), free air; 18 (16%; sensitivity 16%, specificity 100%), "arrowhead" signs; nine (8%; sensitivity 8%, specificity 99%), abscesses; four (4%; sensitivity 4%, specificity 100%,), phlegmons; five (4%; sensitivity 4%, specificity 99%), intramural air; two (2%; sensitivity 2%, specificity 100%), intramural sinus tracts. Overall CT interpretation had a sensitivity of 99%, a specificity of 99%, a positive predictive value of 99%, a negative predictive value of 99%, and an overall accuracy of 99%.
CONCLUSION. The two most frequent signs of diverticulitis were bowel wall thickening (96%) and fat stranding (95%). Less frequent but highly specific signs were fascial thickening (50%), free fluid (45%), and inflamed diverticula (43%).
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The classic clinical symptoms of acute, left-sided diverticulitis are pain and tenderness in the left lower abdominal quadrant, fever, and elevated WBC count [5]. However, on the basis of which segment of the colon is involved, the pain can be manifest in the right lower as well as in both upper abdominal quadrants. Because of a variety of unspecific symptoms such as nausea, vomiting, intermittent constipation, diarrhea, and even urinary symptoms due to bladder fistulas, the clinical diagnosis of diverticulitis is often uncertain, leading to misdiagnosis rates of up to 34% [6]. To improve diagnostic certainty and thus avoid unnecessary surgical interventions, many authors recommend early routine imaging of all patients with suspected diverticulitis. Currently, the imaging modality of choice to diagnose acute colonic diverticulitis is usually CT [7,8,9]. The efficacy of CT versus barium enema examination has been subject to lengthy debates in the past [10,11,12,13,14,15]. It is now widely acknowledged that the barium enema examination is inherently limited because diverticulitis is primarily an extramucosal process, and with barium enema, only secondary effects of the extramucosal inflammation on the barium column can be seen. CT, in contrast, can directly depict intramural and extramural inflammatory disease components and, thus, not only facilitate the diagnosis but also help in staging the disease and in establishing alternative diagnoses [5, 16, 17].
Optimal results on CT for suspected diverticulitis require the use of bowel contrast material [8]. Colonic contrast material administered through the rectum offers the advantage of evaluation of colonic wall thickness, intraluminal versus extraluminal air, and the "arrowhead" sign [18]. Oral contrast material may not allow evaluation of colonic wall thickness when the colon is partially collapsed [4]. In addition, it takes up to 2 hr for oral contrast material to opacify the large bowel. This extended time delays scanning and may not be feasible in an emergency setting. IV contrast material increases costs and risks without adding significantly new information.
We previously reported a sensitivity of 97%, a specificity of 100%, and an overall accuracy of 99% using CT with 5-mm collimation and administering bowel contrast material only through the colon [8]. The purpose of this investigation was to retrospectively review—using the reported technique—the frequency and accuracy of the individual signs of diverticulitis in a large number of patients examined on CT using rectally administered colonic contrast material. Knowing the diagnostic impact of the CT signs of diverticulitis enhances diagnostic accuracy and assists in differentiating diverticulitis from other conditions.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Patient Preparation and CT Technique
Patients were placed on the CT scanner table, and 400-1000 mL (one patient
received 1350 mL) of a 4% iothalamate meglumine solution (Conray;
Mallinckrodt, St. Louis, MO) was infused via gravity drip into the colon
through IV tubing and a soft rubber rectal catheter (Flexi-Tip Junior; E-Z-M,
Montreal, Canada) without taping the catheter or using a balloon. The
concentration of iodine in the contrast solution was 11.28 mg/mL. Contrast
material administration typically required approximately 5 min; the exact
volume of contrast material depended on each patient's size and subjective
sensation of fullness. Administration of contrast material was stopped if
there was leaking of contrast material around the catheter. The contrast
material bag was lifted approximately 3 ft (90 cm) above the CT table during
administration.
All CT was performed on a commercially available scanner (HiSpeed Advantage; General Electric Medical Systems, Milwaukee, WI). A digital abdominal radiograph was obtained on the CT scanner, followed by helical scanning from the diaphragm to the public symphysis. Helical parameters included 5-mm collimation, 7.5 mm/sec table speed (1.5 pitch), and 7.5-mm image spacing. Images were viewed in soft-tissue window settings (window width, 340 H; window level, 40 H) or on a picture archiving and communication system (PACS) and laser-printed on a 20-on-one film format.
Fifty-five patients were not initially suspected of having diverticulitis, and their first CT examination was performed with oral and/or IV contrast material. These 55 patients were then examined on repeated CT with rectally administered colonic contrast material. For oral contrast material, one of three separate 300-mL cups of water, each mixed with 7.5 mL of meglumine diatrizoate (Gastrografin; Bristol-Meyers Squibb, Wallingford, CT) was administered immediately, at 20 min, and at 40 min, with scanning beginning 60-90 min after administration of contrast material. For IV contrast material, 135 mL of 60% Conray (Mallinckrodt) was administered at a rate of 2 mL/sec; the infusion began 70 sec before scanning.
Analysis of CT Findings
The CT scans were interpreted prospectively by staff radiologists in the
emergency radiology division. The reports of this initial interpretation were
analyzed and tallied as described for the retrospective review that follows.
If a radiographic sign was not listed in the initial report, the finding was
assumed to be negative. If a sign was described as "possible" or
"cannot exclude," the finding was given the attribute
"equivocal." For the retrospective review, each CT scan
interpreted as consistent with diverticulitis or indeterminate in the initial
report was rereviewed by two emergency radiology division staff members who
achieved consensus. The findings on CT were tallied, including overall
diagnosis, possible alternative diagnosis, and the presence of individual
signs of diverticulitis: bowel wall thickening, fat stranding, diverticula,
fascial thickening, free fluid, inflamed diverticulum, muscular wall
hypertrophy, free air, arrowhead sign, phlegmon, abscess, intramural air, and
intramural sinus tract.
Findings on CT were considered positive, negative, or equivocal for diverticulitis on the basis of the reinterpretation. In cases in which no definite interpretation could be made, results were assessed as indeterminate.
Definition of Signs of Diverticulitis
The examined signs of diverticulitis were defined as follows: bowel wall
thickening, the small-bowel wall greater than 3 mm and the large-bowel wall
greater than 4 mm; fat stranding, a linear or inhomogeneous soft-tissue
density interspersed in the fat; diverticulum, an outpouching from the bowel
wall; fascial thickening, peritoneal lining or septa in the fat thicker than
usual; free fluid, water-density fluid found in the peritoneal space; inflamed
diverticulum, a diverticulum located in the center of fat stranding; free air,
air found in the peritoneal cavity outside bowel; arrowhead sign, an
arrowhead-shaped configuration of contrast material found at the orifice of a
diverticulum; abscess, a collection of fluid or soft-tissue density that may
or may not contain free air; phlegmon, an ill-defined mass of soft-tissue
density without air; intramural air, air present in the bowel wall; intramural
sinus tract, a linear streak of contrast material found in the bowel wall.
Criteria for the Clinical Diagnosis of Diverticulitis
Positive CT findings in 114 patients were confirmed at surgery in 16 and at
gross and microscopic pathologic examination in 12 patients with surgery
occurring during the same hospitalization. In the remaining 86 patients, the
diagnosis of diverticulitis was based on the clinical course and diagnosis at
the time of discharge. Patients were diagnosed with findings positive for
diverticulitis only if the following criteria were fulfilled: presence of
fever, elevated WBC, and two or more clinical signs and symptoms associated
with diverticulitis as described elsewhere
[8]; response to antibiotic
treatment; exclusion of alternative diagnoses; and location of abdominal pain
at the same side as the location of inflammation on CT.
Negative CT findings in 192 patients were confirmed at surgery in 19 and at gross and microscopic pathologic examination in four patients. In the remaining 169 patients, the diagnosis of diverticulitis was based on the hospital course and diagnosis at the time of discharge.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
One hundred fourteen CT scans (37%) of the 312 patients examined for suspected diverticulitis were interpreted as positive for diverticulitis; 192 scans (62%), as negative for diverticulitis; six (2%), as indeterminate. Taking into account the results from clinical follow-up and surgical findings, we report the following: 113 true-positive, one false-positive, 191 true-negative, one false-negative, and six indeterminate CT findings.
Of the six indeterminate cases, one patient was found to have ischemic colitis at biopsy, four patients were discharged with a clinically established diagnosis of diverticulitis that was treated conservatively, and one patient was discharged with the diagnosis of a syncope.
The overall true-positive finding rate on CT was 37%, the true-negative finding rate was 62%, the sensitivity was 99%, the specificity was 99%, the positive predictive value was 99%, the negative predictive value was 99%, and the overall accuracy was 99%.
Tables 1 and 2 list the signs considered suggestive of diverticulitis according to previous communications in the literature [4, 8, 9, 18, 19, 20] and the frequency with which the signs were interpreted as positive, negative, or indeterminate for diverticulitis on CT; the number of true-positive, true-negative, false-positive, and false-negative interpretations; and the sensitivity, specificity, positive predictive value, and negative predictive value of each sign.
|
|
The two most frequent and sensitive signs of diverticulitis on CT with colonic contrast material were bowel wall thickening (96%) and fat stranding (95%) (Figs. 1A,1B,2A,2B,3A,3B,4,5). Less frequent signs were fascial thickening (50%; Figs. 1A,1B,2A,2B,3A,3B), free fluid (45%; Fig. 2A,2B), and inflamed diverticulum (43%; Figs. 3A,3B and 4), all of which were noted to have a high specificity in the examined patient population. Notably, diverticula (Figs. 1A,1B and 5) were identified in the colon in only 104 patients (91%).
|
|
|
|
|
|
|
|
Alternative conditions were diagnosed on CT in 134 (70%) of 192 patients with a final clinical diagnosis other than diverticulitis, including infectious or ischemic colitis (18 patients, 13% of alternative conditions), gallstones (13, 10%), small-bowel obstruction (13, 10%), and urethral stones (eight, 6%). The identification of alternative diagnoses avoided unnecessary surgery and accelerated appropriate treatment in these patients.
Diverticulitis was located in the sigmoid colon in 63 (55%), at the junction of descending and sigmoid colon in 19 (17%), in the descending colon in 18 (16%), in the ascending colon in 13 (11%), in the cecum in two (2%), and in the transverse colon in one (1%) patient. In two patients, diverticulitis affected both the ascending colon and the descending or sigmoid colon.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
All these steps in the disease process can be depicted on CT. However, adequate filling and distention of the colon must be achieved by a contrast agent because in a partially collapsed colon, colonic wall thickness, intraluminal versus extraluminal air, and the arrowhead sign may not be detected. We have previously reported high accuracy in the diagnosis of diverticulitis using only contrast material via the rectum and thin-section (5-mm-collimation) helical CT [8]. To our knowledge, our current investigation, which uses this technique, represents the largest number of patients in whom individual CT signs diagnostic for diverticulitis have been thoroughly analyzed. Chintapalli et al. [9] concentrated on CT signs that are diagnostic of both diverticulitis and colon cancer. Thus, findings like inflamed diverticulum, arrowhead sign, muscular wall hypertrophy, fascial thickening, and abscess were not evaluated. Also, collimations between 8 and 10 mm and different CT scanners were used. All patients received oral contrast material, but not all received rectal contrast material. In the study by Hulnick et al. [21], contrast material was administered IV and orally, and 10-mm-collimation was used. The findings not evaluated were inflamed diverticulum, free air, free fluid, muscular wall hypertrophy, and arrowhead sign. Pradel et al. [3] used inconsistent CT protocols and only four diagnostic criteria for diverticulitis: bowel wall thickening, diverticula, fat stranding, and pericolic abscess. Their overall CT accuracy was 84%, and thus significantly lower than our current and previous accuracy of 99% [8]. Similarly, other investigations included few signs or concentrated on specific locations of diverticulitis in the colon [16, 17, 19, 22].
Comparing the frequency of the findings in previous reports with our data, we agree that the most frequent findings are bowel wall thickening, fat stranding, and diverticula; these findings all occur in a range between 70% and 100% and are considered by many authors to be the most valuable in the diagnosis of diverticulitis [3, 8, 21, 22]. Abscess is also considered a valuable sign of diverticulitis, but its frequency has a greater variation in the literature ranging from 59% to only 7% [3, 8, 16, 17, 21, 22]. On the basis of the data of this investigation, however, we believe that there are at least four more signs: fascial thickening, muscular wall hypertrophy, inflamed diverticulum [20], and the arrowhead sign [20], all of which can add significant information and increase the certainty of or exclude a diagnosis of diverticulitis. Fascial thickening was present in 50% of the patients with a final diagnosis of diverticulitis but was present in none of the patients with a final diagnosis other than that of diverticulitis. This finding was also noted in a recent report that concluded that fascial thickening is frequent in diverticulitis but is infrequent in colon cancer (Thakrar J et al., presented at the American Society of Emergency Radiology meeting, March 2001).
We believe that the distinction of muscular wall hypertrophy from inflammatory bowel wall thickening as indicated in our study is important; muscular wall hypertrophy is a relatively specific sign (96% in our patient population) for diverticulitis because it represents the adaptation of the muscular layers of the colonic wall to long-standing elevated intraluminal pressure characteristic for the pathophysiology of diverticulosis as the underlying condition for diverticulitis. Both the arrowhead sign and the inflamed diverticulum have been previously reported to be specific signs that can improve the accuracy of the diagnosis and facilitate the exclusion of other colonic or paracolonic inflammatory conditions. The identification of both signs depends on optimal opacification and distention of the colonic lumen as achieved via rectal contrast administration [18]. Finally, we found the sensitivity of free fluid and free air (45% and 30%, respectively) to be higher than that reported by other authors. With their high specificity, they could also add to the diagnostic accuracy and should definitely be investigated if diverticulitis is suspected.
Our study had several limitations. The patients included in the investigation consisted solely of patients admitted to the emergency department who were clinically suspected of having diverticulitis. Certain signs, characterized as highly specific in this investigation, should be validated further in patients with other pathologic conditions not represented in our investigation. Also, most patients with a final diagnosis of diverticulitis had no surgical proof, a direct consequence of our principal aim in performing CT—that is, to reduce the percentage of unnecessary surgery.
In the group of patients examined in our investigation, colonic contrast material provided excellent opacification and distention of the bowel lumen. Colonic contrast material has also been shown to be safe [23]. In our patient population, the most sensitive signs of diverticulitis with colonic contrast-enhanced CT having a specificity of more than 90% were bowel wall thickening and fat stranding. The most specific signs of diverticulitis on colonic contrast-enhanced CT with a sensitivity of more than 40% were inflamed diverticulum, fascial thickening, and free fluid. Although less frequent, the arrowhead sign, muscular wall hypertrophy, and free air provided added specificity. The high overall accuracy of 99% in this and our previous studies is a combined result of the CT technique and consideration of a higher number of diagnostic signs than those in previous reports [8].
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  S. W. Anderson, J. A. Soto, B. C. Lucey, A. Ozonoff, J. D. Jordan, J. Ratevosian, A. S. Ulrich, N. K. Rathlev, P. M. Mitchell, C. Rebholz, et al. Abdominal 64-MDCT for Suspected Appendicitis: The Use of Oral and IV Contrast Material Versus IV Contrast Material Only Am. J. Roentgenol., November 1, 2009; 193(5): 1282 - 1288. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 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]
|
|
|
|
|
|
V. Goh, S. Halligan, S. A. Taylor, D. Burling, P. Bassett, and C. I. Bartram Differentiation between Diverticulitis and Colorectal Cancer: Quantitative CT Perfusion Measurements versus Morphologic Criteria--Initial Experience Radiology, February 1, 2007; 242(2): 456 - 462. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. F. Thoeni and J. P. Cello CT Imaging of Colitis Radiology, September 1, 2006; 240(3): 623 - 638. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. Siewert, G. Tye, J. Kruskal, J. Sosna, and F. Opelka Impact of CT-Guided Drainage in the Treatment of Diverticular Abscesses: Size Matters. Am. J. Roentgenol., March 1, 2006; 186(3): 680 - 686. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Tack, P. Bohy, I. Perlot, V. De Maertelaer, O. Alkeilani, S. Sourtzis, and P. A. Gevenois Suspected Acute Colon Diverticulitis: Imaging with Low-Dose Unenhanced Multi-Detector Row CT Radiology, October 1, 2005; 237(1): 189 - 196. [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]
|
|
|
|
|
|
P.-A. Poletti, A. Platon, O. Rutschmann, K. Kinkel, V. Nyikus, S. Ghiorghiu, P. Morel, F. Terrier, and C. D. Becker Acute Left Colonic Diverticulitis: Can CT Findings Be Used to Predict Recurrence? Am. J. Roentgenol., May 1, 2004; 182(5): 1159 - 1165. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
