| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Gastrointestinal Imaging |
1 From the Department of Radiology, Beth Israel Deaconess Medical Center and Harvard Medical School, 330 Brookline Ave, Boston, MA 02215. From the 2004 RSNA Annual Meeting. Received February 4, 2005; revision requested April 4; revision received May 3; accepted June 3; final version accepted January 22, 2006. Address correspondence to V.R. (e-mail: vraptopo{at}bidmc.harvard.edu).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Materials and Methods: Institutional review board approval was obtained for this HIPAA-compliant study, with waiver of informed consent. Records were retrospectively reviewed in patients who presented to the emergency department between April 29 and October 31, 2003, with right lower quadrant pain. Scanning was performed with the same eight–detector row CT scanner by using oral and (unless contraindicated) intravenous contrast agents, and transverse and coronal reformations were obtained. Two radiologists prospectively evaluated all scans at the time of the examination and rendered a consensus opinion. Clinical follow-up of at least 3 months' duration was performed retrospectively for patients whose appendix was not visualized to determine whether appendicitis had developed. Statistical analysis and calculation of percentages with confidence intervals (CIs) were performed.
Results: Of the 400 consecutive patients who underwent multidetector CT, 132 (33.0%) were male and 268 (67.0%) were female. Eighty patients (20.0%) had acute appendicitis and 79 (19.8%) had another cause for abdominal pain. A normal appendix with no other cause for pain was seen in 182 patients (45.5%). In 59 patients (14.8%), the appendix was not visualized. Of these 59 patients, 50 had adequate follow-up. Clinical follow-up was uneventful in 49 of these 50 patients. Thus, on otherwise normal multidetector CT scans in patients suspected of having acute appendicitis, nonvisualization of the appendix was negative for appendicitis in 98% (95% CI: 71%, 100%) of cases. Conversely, when the appendix was seen at multidetector CT and was abnormal, appendicitis was present in 95% (95% CI: 72%, 100%) of cases.
Conclusion: In patients with right lower quadrant pain, a nonvisualized appendix at multidetector CT reliably excludes acute appendicitis.
© RSNA, 2006
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
On a considerable number of CT scans obtained for right lower quadrant pain, the appendix is not visualized. In two recent reports, this occurred in 13% (11) and 14% (12) of studies. This finding, or "nonfinding," often presents radiologists and clinicians with a decision-making dilemma. The results of a recently published study by Nikolaidis et al (12) suggest that nonvisualization of the appendix on transverse CT scans may safely be used to exclude acute appendicitis if no secondary CT findings of appendicitis are present. Recent advances in multidetector CT allow the acquisition of high-quality multiplanar reformations, which have been reported to enhance the visualization of the appendix (13). Improvement in the visualization of the appendix could strengthen the confidence of a negative test result.
Independently and without knowledge of the study by Nikolaidis et al (12), we performed a similar investigation, the purpose of which was to retrospectively determine the value of the nonvisualized appendix at multidetector CT in patients with acute right lower quadrant pain in whom appendicitis was a consideration.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Imaging Technique Protocol
All multidetector CT scans were obtained by using the same eight–detector row CT scanner (Lightspeed Ultra; GE Medical Systems, Milwaukee, Wis). The imaging protocol included the administration of an oral contrast agent (600–900 mL of barium sulfate [Readi-Cat 2; E-Z-Em, Westbury, NY]) 60–90 minutes before scanning. The protocol did not call for the administration of a rectal contrast agent. Unless contraindicated (eg, because of allergy or impaired renal function), an intravenous contrast agent (100–150 mL of ioversol [Optiray 320; Mallinckrodt, St Louis, Mo]) was also administered at a rate of 2.5 mL/sec, and scanning was performed 60 seconds after the start of the injection.
A single scan series was obtained from the lung base to the symphysis pubis by using a 1.25-mm collimation (reconstructed at 0.8-mm intervals), 0.5-second gantry rotation, and 13.5-mm table speed per gantry rotation. This allowed scanning of the abdomen and pelvis in 10–15 seconds (ie, during a single breath hold). A soft-tissue reconstruction algorithm was used. From the 1.25-mm collimation, routine 5-mm-thick continuous transverse sections and 5-mm-thick continuous coronal sections were produced, with no overlap at the scanning console. These sections were sent to a picture archiving and communication system (Centricity workstation; GE Medical Systems) for interpretation and storage. Additional thin sections or interactive multiplanar images were not obtained with our acute abdominal pain protocol.
Image Analysis
The transverse and coronal reformations were displayed on our hospital's picture archiving and communication system for interpretation. At the time of acquisition, all images were interpreted by two radiologists: an abdominal imaging attending physician and a radiology resident who had passed institutional requirements to cover emergency department studies without direct supervision. The final report was recorded with agreement between these two radiologists. In all, seven attending radiologists (with a minimum of 3 years of experience in abdominal imaging) and 15 radiology residents (end of 1st year [postgraduate year 2] to end of 3rd year [postgraduate year 4]) were involved in interpreting the images.
The two interpreting radiologists agreed on the diagnosis and followed several general guidelines that were agreed on by our institution's abdominal imaging section. The resident usually reviews and renders an interpretation first, followed by a review and discussion of the images with the attending physician. If there is a disagreement, the attending physician's opinion prevails, after considerable discussion. It is extremely rare that the opinion of another attending physician is sought to arbitrate a disagreement. Interpretation included evaluation of the appendix on transverse and coronal multiplanar reformations. An appendix was considered normal when it was less than 6 mm in diameter and lacked a thickened enhancing wall and intraluminal calcifications (appendicoliths) or periappendiceal stranding. A diagnosis of appendicitis was made when the appendix had a thickened enhancing wall and a diameter of 8 mm or more. If the diameter of the appendix was between 6 and 8 mm, a diagnosis of appendicitis was made if there was evidence of periappendiceal stranding (5).
Clinical Follow-up and Surgical Pathologic Correlation
For patients in whom the appendix was not visualized, a retrospective chart review was performed by two authors (S.G. and F.K.) to determine these patients' clinical course and whether acute appendicitis had developed. Scans were selected for analysis before October 31, 2003, so that adequate time had passed (at least 9 months at the time the analysis was performed) for the episode of abdominal pain to have resolved or been treated. This also allowed adequate time for follow-up within the medical system.
By using the hospital electronic medical records, we performed a review of the emergency department notes, clinic notes, further imaging studies, procedure reports (eg, for colonoscopy), surgical reports, and pathology reports in this subset of patients. Correlation with laboratory findings, such as white blood cell count, was not performed because these results were considered nonspecific for acute appendicitis.
For the small group of patients who underwent no follow-up after discharge from the emergency department, an institutional review board–approved letter was sent to their homes describing the study and all the elements of informed consent. The letter requested that these patients contact the investigators by telephone for an interview. After verbal informed consent was obtained from these patients, one of two authors (S.G. or V.R.) asked the patients a short set of institutional review board–approved questions to determine whether they had appendicitis at the time of the multidetector CT examination. Review of hospital electronic medical records, including surgical and pathology reports, was also performed by one author (F.K.) for patients in whom appendicitis was positively diagnosed at multidetector CT to determine whether these patients were actually found to have acute appendicitis.
Statistical Analysis
Descriptive statistics of the age and sex of the patients in the four groups (appendicitis, alternative acute diagnosis other than appendicitis, normal appendix, and nonvisualized appendix) and the significance of the difference between the group that had a nonvisualized appendix and the other three groups were evaluated by using the 
2 test for sex and a t test for age (S.G. and F.K.). A P value of less than .05 was considered to indicate a statistically significant difference, and statistical analysis was performed by using statistical software (SPSS 11.0 for Windows; SPSS, Chicago, Ill).
For patients with a nonvisualized appendix and otherwise normal findings at multidetector CT, the percentage and 95% confidence interval (CI) of patients without appendicitis were calculated. The 95% CIs were calculated to demonstrate the randomness of the sample size (14,15). Similarly, two authors (S.G. and V.R.) calculated the percentage and 95% CI of patients with signs of appendicitis at multidetector CT who were found to have appendicitis.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
A normal appendix and a normal multidetector CT scan that did not show another cause for abdominal pain were seen in 182 (45.5%; 95% CI: 39%, 52%) of 400 patients. In 79 patients (19.8%; 95% CI: 15%, 24%), other causes for pain were found at multidetector CT. These causes included inflammatory bowel disease or colitis (n = 26), diverticulitis (n = 15), obstructing nephrolithiasis (n = 11), acute cholecystitis (n = 5), typhilitis or mesenteric adenitis (n = 4), cancer (n = 4), pyelonephritis or renal infarction (n = 3), small-bowel obstruction (n = 3), volvulus (n = 2), ruptured ovarian cyst or dermoid (n = 2), tubo-ovarian abscess (n = 2), bladder outlet obstruction (n = 1), and pelvic arteriovenous malformation (n = 1).
A diagnosis of acute appendicitis at multidetector CT was made in 80 (20.0%; 95% CI: 16%, 24%) of 400 patients at the time of presentation. The appendix was not visualized and no other abnormalities were seen in 59 patients (14.8%; 95% CI: 11%, 19%). There was no statistically significant difference between groups regarding mean age or sex.
Nonvisualized Appendix
Fifty of the 59 patients whose appendix was not visualized underwent adequate follow-up according to physician notes (46 patients) in the electronic medical records or according to a response to the mailed questionnaire (four patients). Follow-up showed that 49 patients did not exhibit symptoms that would have raised suspicion or indicated a diagnosis of acute appendicitis. These patients had an uneventful course, with resolution of abdominal pain and/or identification of another cause of abdominal pain. Nine patients had no further follow-up noted in their medical records and did not respond to the mailed questionnaire. At the time of their visit, all nine patients left the emergency department in good condition, with reduced or resolved pain and instructions to call or return to the hospital if their pain recurred or worsened.
When only the 50 patients with definite follow-up were used for calculations, the percentage of patients with a nonvisualized appendix who were negative for appendicitis was 98% (95% CI: 71%, 100%; 49 of 50 patients). When the nine patients who did not respond to the follow-up questionnaire were also considered to be negative for appendicitis, the percentage of patients with a nonvisualized appendix who were negative for appendicitis did not change considerably (58 [98%; 95% CI: 68%, 100%] of 59 patients).
Follow-up revealed that one patient whose appendix was not visualized returned 15 weeks after the initial visit to our emergency department; at that time, acute appendicitis was identified at repeat multidetector CT. This patient was a 29-year-old woman who initially presented to the emergency department 11 days after an uncomplicated cesarean delivery and who underwent multidetector CT for right lower quadrant pain and fever (Figure). The patient was admitted to the hospital at that time and was treated with intravenous antibiotics for 24 hours. Her condition improved during admission, and she was discharged without oral antibiotics. The diagnosis was indicated as "postoperative infection of unclear etiology." The patient's symptoms resolved with an uneventful course until she presented 15 weeks later with right lower quadrant pain and fever. Acute appendicitis was recognized at repeat multidetector CT (Figure), and she underwent laparoscopic appendectomy. Surgical pathology reports showed acute gangrenous appendicitis. The patient had an uncomplicated postoperative course and was discharged on postoperative day 2.
|
|
|
|
Of the 80 patients in whom acute appendicitis was diagnosed at multidetector CT, six did not undergo surgery. These patients received intravenous antibiotics and their clinical conditions improved. Of the 74 patients who underwent appendectomy, 70 were found to have appendicitis. In three of the 74 patients, an alternative diagnosis was found at surgical pathologic examination (carcinoma of the appendix, endometriosis, and ruptured ovarian cyst), and one patient had no abnormal findings. According to these results, the percentage of patients with multidetector CT findings that were positive for appendicitis (ie, those who had pathologically proved appendicitis) was 95% (95% CI: 72%, 100%; 70 of 74 patients).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
CT is also beneficial in the evaluation of right lower quadrant pain because it facilitates diagnosis of a variety of conditions other than appendicitis that cause pain, as was seen in our study. The sensitivity, specificity, positive predictive value, and negative predictive value of the visualized appendix have been reported to be as high as 98%, 99%, 97%, and 98%, respectively (6,18,19). In our study, the percentage of scans that were positive for appendicitis at both multidetector CT and surgical pathologic examination was 95%.
Nonvisualization of the appendix is relatively common and was found on 15% of the scans in our study, despite the use of an eight–detector row CT scanner with 1.25-mm collimation and the routine use of coronal multiplanar reformations. Clinicians are often uncomfortable discharging patients with a nonvisualized appendix from the emergency department, which leads to increased observation time and/or admittance to the hospital. Moreover, three patients in our study who had a nonvisualized appendix underwent an unnecessary appendectomy. A test (nonvisualized appendix at multidetector CT) that has a high negative predictive value, when satisfied, is excellent at excluding that disease (appendicitis). In our study, the high percentage (98%) of cases that were negative for disease will allow clinicians to exclude appendicitis as part of the differential diagnosis when the appendix is not visualized at multidetector CT.
In our study, the high percentage of multidetector CT scans that did not demonstrate the appendix and that were negative for appendicitis agrees with the recently published conclusion of Nikolaidis et al (12)—that the absence of a distinctly visualized appendix in the right lower quadrant and the absence of secondary findings can be used to exclude appendicitis. The two studies have similarities: they are both retrospective, and both used multidetector CT. However, there are technical and methodologic differences. The number of scans that demonstrated a nonvisualized appendix with documented follow-up was more than doubled (50 vs 20) in our study. Also, we used the imaging diagnosis of a nonvisualized appendix at the time of the examination rather than a retrospective assessment by radiologists who were not directly involved in the patients' care. This corresponds to our standard practice and decreases the bias from retrospective use of consensus by two radiologists. We used hospital records and patient interviews for our follow-up assessment. Thus, in our study, nine (15%) of 59 patients were lost to follow-up compared with 14 (41%) of 34 patients in the study by Nikolaidis et al (12) (P = .011).
In contrast to the study by Nikolaidis et al (12) in which patients with various CT abnormalities were included in the nonvisualized appendix group, the patients in our study with a nonvisualized appendix had an otherwise normal scan. This was done to limit any confounding pathologic abnormalities that could obscure the appendix, such as inflammation from adjacent diverticulitis or colitis. The percentage of nonvisualized appendices might have been larger had we retrospectively assessed the studies with an alternative diagnosis for the appendix. However, because abdominal pain that was not appendicitis had already been diagnosed in these patients, we believed that including them in the calculations would have falsely increased the value of the nonvisualized appendix. Also, we used an eight–detector row CT scanner with 5-mm transverse and coronal sections reconstructed from a 1.25-mm collimation compared with a four–detector row CT scanner and 5-mm collimation in the previous study. The use of multiplanar reformations has been reported to help localization of the appendix (13), although this hypothesis was not specifically tested in our study because each multidetector CT examination had both transverse and coronal reformatted images interpreted simultaneously.
Much discussion and work have gone into determining optimal imaging parameters for identifying the appendix and its associated pathologic abnormalities, including the use of intravenous, oral, and rectal contrast agents and focused or nonfocused scans (6–8,16,22,23). The protocol used in this study is optimal for depicting alternative acute intraabdominal conditions that can mimic appendicitis (16). In addition, a survey of academic institutions regarding imaging protocols used in patients suspected of having appendicitis showed that intravenous, oral, and rectal contrast agents are routinely given at 79%, 82%, and 32% of institutions, respectively (24). CT with oral and intravenous contrast agents continues to be the most commonly used technique in practice (24).
Three patients with a nonvisualized appendix underwent surgery for exploration and negative appendectomy on the basis of the clinical picture, including history of presentation, laboratory values, and physical examination findings. It is possible that, given these results, these patients would have undergone conservative treatment. Because a high percentage of multidetector CT scans that demonstrate a nonvisualized appendix are negative for appendicitis, clinicians should be more comfortable with conservative care in this group of patients.
It is also not clear whether the one patient with a nonvisualized appendix who was later shown to have pathologically proved appendicitis had acute appendicitis at the time of initial presentation because there was a substantial time gap between presentations. In this patient, an undiagnosed case of acute appendicitis may have resolved with intravenous antibiotics, although she received antibiotics for only 24 hours and was discharged without oral antibiotics. Nevertheless, the outcome from the relatively conservative treatment of the nonvisualized appendix showed no adverse implications because the patient ultimately underwent an uncomplicated appendectomy.
One limitation of our study was the retrospective nature of follow-up. A second limitation was that 15% of patients were lost to follow-up. However, the chance that these patients developed acute appendicitis is low because they were discharged from the emergency department only after they were believed to be clinically stable, as is the practice in our hospital.
In conclusion, the results of this study show that, in patients with right lower quadrant pain that is unexplained at multidetector CT, a nonvisualized appendix is a reliable predictor of the absence of acute appendicitis.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Abbreviations: CI = confidence interval
Authors stated no financial relationship to disclose.
Author contributions: Guarantors of integrity of entire study, S.G., V.R.; study concepts/study design or data acquisition or data analysis/interpretation, all authors; manuscript drafting or manuscript revision for important intellectual content, all authors; manuscript final version approval, all authors; literature research, S.G., V.R., F.K., J.B.K.; clinical studies, all authors; statistical analysis, S.G., V.R.; and manuscript editing, S.G., V.R., B.S., J.B.K.
|
References |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  H. Seo, K. H. Lee, H. J. Kim, K. Kim, S.-B. Kang, S. Y. Kim, and Y. H. Kim Diagnosis of Acute Appendicitis With Sliding Slab Ray-Sum Interpretation of Low-Dose Unenhanced CT and Standard-Dose IV Contrast-Enhanced CT Scans Am. J. Roentgenol., July 1, 2009; 193(1): 96 - 105. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S.-M. Joo, K. H. Lee, Y. H. Kim, S. Y. Kim, K. Kim, K. J. Kim, and B. Kim Detection of the Normal Appendix with Low-Dose Unenhanced CT: Use of the Sliding Slab Averaging Technique Radiology, June 1, 2009; 251(3): 780 - 787. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. T. Johnson, K. M. Horton, S. Kawamoto, J. Eng, M. J. Bean, S. J. Shan, and E. K. Fishman MDCT for Suspected Appendicitis: Effect of Reconstruction Section Thickness on Diagnostic Accuracy, Rate of Appendiceal Visualization, and Reader Confidence Using Axial Images Am. J. Roentgenol., April 1, 2009; 192(4): 893 - 901. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
I. Pedrosa, M. Lafornara, P. V. Pandharipande, J. D. Goldsmith, and N. M. Rofsky Pregnant Patients Suspected of Having Acute Appendicitis: Effect of MR Imaging on Negative Laparotomy Rate and Appendiceal Perforation Rate Radiology, March 1, 2009; 250(3): 749 - 757. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Garcia, M. Hernanz-Schulman, D. L. Bennett, S. E. Morrow, C. Yu, and J. H. Kan Suspected Appendicitis in Children: Diagnostic Importance of Normal Abdominopelvic CT Findings with Nonvisualized Appendix Radiology, February 1, 2009; 250(2): 531 - 537. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. S. Lee, N. M. Rofsky, and I. Pedrosa Localization of the Appendix at MR Imaging during Pregnancy: Utility of the Cecal Tilt Angle Radiology, October 1, 2008; 249(1): 134 - 141. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Keyzer, S. Pargov, D. Tack, V. Creteur, P. Bohy, V. De Maertelaer, and P. A. Gevenois Normal Appendix in Adults: Reproducibility of Detection with Unenhanced and Contrast-Enhanced MDCT Am. J. Roentgenol., August 1, 2008; 191(2): 507 - 514. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
