| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1
Department of Radiology and Nuclear Medicine, Hospital Barmherzige Brueder,
Kajetanerplatz 1, A-5010 Salzburg, Austria.
2
Department of Surgery, Hospital Barmherzige Brueder, A-5010 Salzburg,
Austria.
Received August 17, 1999;
accepted after revision March 17, 2000.
Address correspondence to A. Hollerweger.
Abstract
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
SUBJECTS AND METHODS. Eighty-six consecutive patients with clinically suspected acute sigmoid diverticulitis were referred for transabdominal sonography as the initial imaging method. In 46 patients, transrectal sonography was performed in addition to transabdominal sonography if pain was localized to the mid lower abdomen and if a disease process could not be visualized or could be only partially visualized on transabdominal examination. An end-firing 5-9-MHz endocavitary probe was used for transrectal sonography.
RESULTS. Thirty-four of 50 patients with a final diagnosis of sigmoid diverticulitis underwent both transabdominal and transrectal sonography. In 20 patients, transrectal sonography showed relevant additional information: six diagnoses of diverticulitis were established on transrectal sonography alone. Transrectal sonography revealed one perforation, five abscesses, and three fistulous complications that were not shown on transabdominal sonography. In the remaining five patients, correct diagnoses were supported on transabdominal examinations, but only transrectal sonography could show an inflamed diverticulum. In 10 patients, transrectal sonography revealed signs of diverticulitis but no relevant information in addition to the results from transabdominal sonography. Four false-negative and two false-positive results were revealed with transrectal sonography.
CONCLUSION. Transrectal sonography is accurate for confirming clinically suspected acute colonic diverticulitis when the lower sigmoid colon is affected. It helps avoid false-negative results and defines the severity of disease in the lower sigmoid colon better than transabdominal sonography alone. Transrectal sonography can increase the sensitivity of sonography for diagnosing sigmoid diverticulitis.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The effective role of sonography in the examination of patients with suspected acute diverticulitis has been shown in various studies [9,10,11]. The major reason for false-positive findings on sonograms is the nonspecificity of colonic mural thickening that may also occur in Crohn's disease, ulcerative colitis, ischemic colitis, infectious ileocolitis, carcinoma, and lymphoma [12]. Other false-positive findings on sonograms can result from pericolonic abscesses from perforated appendicitis or perforated colonic carcinoma. False-negative findings on sonograms result primarily from a failure to visualize a diseased colonic segment most frequently occurring in the lower sigmoid colon. In this colonic segment, overlying gas often interferes with establishing a correct diagnosis with transabdominal sonography [12].
Radiologists at our hospital routinely use sonography as the initial imaging technique in patients clinically suspected of having acute colonic diverticulitis. Recently, transrectal sonography with an end-firing endocavitary probe has been used in addition to transabdominal sonography for this purpose. Transrectal sonography aids in establishing an accurate diagnosis in the lower sigmoid colon, which is often poorly accessible on transabdominal sonography. Moreover, superior visualization of the sonographic features of diverticulitis in the lower sigmoid colon is possible because of the higher resolution of transrectal sonography.
The aim of this study was to prospectively evaluate the usefulness of complementary transrectal sonography for the diagnosis of sigmoid diverticulitis, especially diverticulitis of the lower sigmoid colon.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Forty-six patients underwent both transabdominal and transrectal sonography (27 women and 19 men; 40-89 years old; mean age, 63 years). The time from the onset of clinical symptoms until hospital admission ranged from 1 day to 4 weeks (mean time, 4.5 days). Clinical symptoms included abdominal pain and tenderness, fever, leukocytosis, change in bowel habits, and urinary tract symptoms.
Of these 46 patients undergoing transrectal sonography, 19 had pathologic correlation after surgery (n = 16) or colonoscopy with biopsy (n = 3). The remaining 27 patients had at least one other imaging study performed, including CT (n = 21), contrast enema (n = 11), or colonoscopy (n = 9). In all patients with conservative treatment (n = 30), clinical and sonographic follow-up was available. The final diagnosis was based on either histologic confirmation or a consensus opinion based on the results of all imaging methods performed and clinical follow-up. The consensus opinion was established by the referring clinician and the radiologists who performed the examinations.
Imaging Techniques
Transabdominal sonograms were obtained with 2-4-MHz convex and 4-7-MHz or
5-10-MHz linear multifrequency transducers on commercially available equipment
(HDI 3000; Advanced Technology Laboratories, Bothell, WA). Patients were
scanned without any preparation with the graded compression technique
[13].
Transrectal sonography was performed with a multifrequency 5-9-MHz end-firing endocavitary probe (HDI 3000, Advanced Technology Laboratories) having an outer diameter of 19 mm and a maximal imaging sector of 150°. The examination was performed without administration of cleansing enemas, and the transducer was covered by a latex sheath. Patients were scanned in a left lateral decubitus position. The end-firing probe allows visualization of the sigmoid colon, located deeply within the true pelvis and poorly accessible on transabdominal sonography (Fig. 1).
|
CT was performed on a helical scanner (X-Vision; Toshiba Medical Systems, Tokyo, Japan) within 24 hr of sonography. Helical parameters for the region of interest were slice thickness of 5 mm and table feed of 7 mm. Immediately before examination, the rectum and distal colon were insufflated with diluted water-soluble contrast enema (Gastrografin; Schering, Berlin, Germany) or air. Oral contrast solution and IV contrast material were not regularly administered. CT was considered diagnostic when colonic wall thickening was associated with the presence of diverticula, with inflammation of pericolic fat, or with pericolic abscesses in complicated diverticulitis.
Contrast enemas were performed with a single contrast technique with a water-soluble contrast material (Gastrografin). Delay from initial sonography was 1-6 days. Diagnostic criteria for diverticulitis were luminal narrowing of the bowel with irregularity of the mucosa or a leakage of contrast material outside the lumen combined with diverticulosis. CT scans and barium enemas were interpreted by a radiologist who was unaware of the results of the preceding sonography. The dictated reports of these CT and contrast enema examinations were used for the results.
Endoscopic findings were considered indicative of diverticulitis in the presence of luminal narrowing of a diverticular bowel segment and signs of inflammatory mucosal changes. Endoscopic examination was performed 1-14 days after cross-sectional imaging. The endoscopist was not blinded to the preceding imaging results.
Interpretation of Sonographic Findings
Sonographic diagnosis of diverticulitis was made if the following three
signs were present: segmental bowel wall thickening (
5 mm), inflammatory
changes of pericolic fat (increased echogenicity of pericolic fat and loss of
compressibility), and inflamed diverticula (variable echogenic outpouchings of
the colonic wall centered in the pericolic inflammation)
[10]. If no inflamed
diverticulum was shown in a diverticular bowel segment, evidence of abscesses
(mural or pericolic fluid collections), fistulous complications (linear
echogenic foci), and perforations (extraluminal gas) were also interpreted as
consistent with diverticulitis. The presence of diverticulosis was not used as
a diagnostic criterion. The radiologist performing the sonographic
examinations prospectively scored both transabdominal and transrectal
examinations for the three criteria indicated.
Immediately after sonography, results of transabdominal and transrectal sonography were compared and divided into two categories.
Category A.—This category included patients in whom transrectal sonography provided relevant additional information as follows: sonographic signs of diverticulitis were shown only with transrectal sonography (false-negative transabdominal sonography); complications (abscesses, fistulas, and perforations) were detected in addition to the results of transabdominal sonography; and transabdominal sonographic findings were suggestive of diverticulitis, but all three sonographic criteria of diverticulitis could be shown only on transrectal sonography.
Category B.—This category included those patients with transrectal sonography providing no relevant additional information. Either transrectal sonography showed signs of diverticulitis with no relevant additional findings compared with those of transabdominal sonography, or the inflamed colonic segment was not visible on transrectal sonography (false-negative for technical reasons).
Statistical parameters were calculated for those 46 patients who underwent both transabdominal and transrectal sonography for suspected diverticulitis of the lower sigmoid colon. In our study, sonograms were interpreted as having positive findings for diverticulitis on the basis of visualization of all three criteria mentioned previously. In contrast, most sonographic studies have required only two of these signs for a positive diagnosis [5, 9, 10]. Statistical parameters changed whether we applied only two or all three sonographic signs. Therefore, parameters were calculated for both applications to obtain results comparable with those of earlier studies.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Category A
Relevant additional information was shown in 20 patients
(Table 1).
|
Correct diagnosis could be established in six patients in whom transabdominal sonography did not reveal any sign of diverticulitis because of overlying intestinal gas (Fig. 2A,2B). In one patient, marked extraluminal gas collections, indicative of mesocolic perforation, were clearly revealed on transrectal sonography but not on transabdominal sonography. The precise extent was then determined on CT (Fig. 3A,3B). We found five additional diverticular abscesses (one intramural, four perisigmoidal) having diameters ranging from 2.2 to 4.5 cm (Figs. 4,5A,5B,5C,6). Three of the abscesses resolved after medical treatment and two were treated surgically. Moreover, three fistulous complications (one sigmoidovaginal fistula, one perisigmoidal fistula, and one intramural fistulous tract) were seen only on transrectal sonography. The patient with the sigmoidovaginal fistula underwent surgery. In contrast to transabdominal sonography, transrectal sonography was effective in revealing all signs of diverticulitis in the remaining five patients. Additional information mostly concerned the visualization of inflamed diverticula. Inflamed diverticula displayed variable echogenicity from hypoechoic to hyperechoic, with or without acoustic shadowing. In contrast to normal diverticula, inflamed diverticula were always centered in the pericolic inflammation.
|
|
|
|
|
|
|
|
|
Category B
Complementary transrectal sonography in 12 patients revealed no relevant
additional information. In two of these patients, transabdominal sonography
revealed signs of diverticulitis, but the diseased segment was not shown on
transrectal sonography (false-negative transrectal sonography). Two further
sonographic results (transabdominal and transrectal in the same two patients)
were false-negative. At presentation (1 and 2 weeks after clinical onset)
symptoms had already markedly improved, and sonography showed diverticula and
muscle wall hypertrophy, but no signs of acute diverticulitis. Altogether, we
noted four false-negative transrectal and eight false-negative transabdominal
findings on sonography.
Two false-positive results were found in our study with both transabdominal and transrectal sonography in the same two patients. The first patient had marked hypoechoic wall thickening, pericolic inflammation, and a fistulous tract to the pericolic fat. Bowel wall layers were not distinguishable. The second patient had moderate thickening of the rectum and sigmoid colon, slight pericolic inflammation, and numerous hypoechoic diverticula. Enlarged pericolic lymph nodes were seen in both patients. Final diagnoses were perforated carcinoma and ulcerative colitis, respectively.
By considering all three sonographic signs of diverticulitis for transabdominal investigation of the lower sigmoid colon, we found a sensitivity of 35% and a specificity of 83%. If only two signs were required for positive diagnosis, transabdominal sonography yielded a sensitivity of 76% and a specificity of 67%. Combined transabdominal and transrectal sonography with all three criteria yielded the best results for a sensitivity of 94% and a specificity of 83%. More details regarding other statistical parameters are listed in Table 2.
|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Category A
Transabdominal sonography without transrectal sonography would have yielded
false-negative results in six patients. Moreover, complications would have
been overlooked in nine patients. In this group, transrectal sonography
revealed five instances of abnormal features requiring surgical treatment.
Fistulas from the sigmoid colon extending to adjacent organs mostly affect the
urinary bladder and the vagina. This region is more readily accessible on
transrectal than on transabdominal sonography, and therefore, transrectal
sonography should also prove more sensitive in showing fistulous
complications.
Other additional information pertained to the visualization of inflamed diverticula. In cases of diverticulitis, paying attention to inflamed diverticula helps radiologists avoid false-positive sonographic findings. An inflamed diverticulum could be visualized in 24 (71%) of 34 patients who underwent both transabdominal and transrectal sonography. The inflamed diverticulum was seen in 11 patients on transrectal sonography alone, in three patients on transabdominal sonography alone, and in 10 patients on both kinds of sonography. This sign could be seen on CT in many cases of right-sided diverticulitis but less frequently in sigmoid diverticulitis [8, 15]. We did not observe inflamed diverticula in most cases of perisigmoidal abscesses, fistulas, and perforations in which the diverticulum was likely involved in the inflammatory process. This finding is in accordance with the findings of an earlier study [10].
Category B
In all patients, correct diagnosis was established on transabdominal
sonography. Transrectal sonography also revealed typical signs of
diverticulitis in most patients, but no relevant additional information was
obtained. Nevertheless, when clinical data suggest a more serious stage of
disease, transrectal sonography can help rule out extension of the
inflammatory process into a region not visible on transabdominal sonography.
The two false-negative results in this category can be explained by the
limited field of view of transrectal sonography.
It remains unclear in the two instances of false-negative results on both transabdominal and transrectal sonography whether this segment was simply not accessible or if the inflammation was too mild to be revealed on sonography. Endoscopic examination showed only mild inflammatory mucosal changes in one patient, and findings were normal except for moderate luminal narrowing in the other patient. In the latter patient, surgery was performed for treatment of chronic recurrent diverticulitis. The histologic specimen showed localized peridiverticular inflammation. Laboratory parameters in both patients were within the normal ranges. In our study, all clinically relevant acute inflammatory processes were discovered on sonography.
Referring to the false-positive results, we are convinced that even more criteria than the three cited here must be applied to differentiate diverticulitis from other diseases. In the patient with sigmoid carcinoma, different layers of the bowel wall were not revealed, and enlarged pericolic lymph nodes were present. In contrast, bowel wall layers are preserved in most cases of diverticulitis, and enlarged lymph nodes are usually not present. Therefore, diagnostic interpretation also included adenocarcinoma in the differential diagnosis. In patients with marked thickening of the colon on CT, the detection of pericolic lymph nodes is also used to differentiate colonic cancer from diverticulitis [16, 17]. In the patient with ulcerative colitis, a segment of 40 cm long was affected, and numerous inflamed diverticula were visible. In our experience, however, patients with diverticulitis have a short diseased segment and usually one, or occasionally more, inflamed diverticula. Considering these additional criteria, we believe that transrectal sonography can successfully differentiate diverticulitis from other diseases in most cases.
When diverticulitis is diagnosed on the basis of colonic mural thickening and pericolic inflammation alone, transabdominal sonography is a sensitive imaging method. However, colonic wall thickening is a nonspecific sign of diverticulitis [12]. Pericolic inflammation may also occur in other inflammatory and neoplastic conditions of the colon [12, 18]. In our opinion, the visualization of an inflamed diverticulum seems to be the most specific diagnostic sign and should be included to obtain an accurate diagnosis. When more sonographic criteria are applied for diagnosis, specificity improves. On the other hand, the sensitivity of transabdominal sonography will decrease in regions that are not easily accessible, like the lower sigmoid colon. This observation was true in some cases of diverticulitis in our study resulting in poor sensitivity. In these instances, additional transrectal sonography showed inflamed diverticula or complications, and therefore, correct diagnoses could be established. Combined transabdominal and transrectal sonography resulted in excellent sensitivity and good specificity even for the lower sigmoid colon.
One shortcoming of transrectal sonography is the limited field of view. A diseased colonic segment extending beyond this region is not accessible or is only partly accessible on transrectal sonography. Extensive pericolic inflammatory changes extending beyond this field of view cannot be completely realized, and extensive pericolic gas hinders confident assessment. Therefore, transrectal sonography should be only a complementary imaging technique to transabdominal sonography in patients suspected of having sigmoid diverticulitis. Nevertheless, we recommend liberal use of this method to exclude relevant disease in this segment, which is not particularly accessible on transabdominal sonography. Transabdominal sonography and transrectal sonography are more operator-dependant than CT. Orientation with transrectal sonography seems to be more difficult than it is for transabdominal sonography but may improve with increasing experience.
A general limitation of this study was that the results of the preceding transabdominal sonographic procedures were known to the investigator and, therefore, could have influenced the interpretation of transrectal sonography. Furthermore, most patients undergoing transrectal sonography suffered from diverticulitis. To evaluate the relevance of transrectal sonography in differentiating diverticulitis from carcinoma, Crohn's disease, and ischemic or ulcerative colitis, further investigation is necessary.
In conclusion, transrectal sonography has two distinct advantages: the lower sigmoid colon, which is normally poorly visible on transabdominal sonography, becomes easily accessible with transrectal sonography; and the high-frequency transducers enable excellent visualization of normal and abnormal features of this segment. Transrectal sonography is an important supplement to transabdominal sonography for immediate and accurate sonographic determination of acute sigmoid diverticulitis.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  J. J. Peterson, L. W. Bancroft, and M. J. Kransdorf Wooden Foreign Bodies: Imaging Appearance Am. J. Roentgenol., March 1, 2002; 178(3): 557 - 562. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
