DOI: 10.1148/radiol.2371040066
(Radiology 2005;237:114-122.)
Nontraumatic Acute Abdominal Pain: Unenhanced Helical CT Compared with Three-View Acute Abdominal Series1
Andrew B. MacKersie, MD,
Michael J. Lane, MD,
Robert T. Gerhardt, MD, MPH,
Harry A. Claypool, MD,
Sean Keenan, MD,
Douglas S. Katz, MD and
Jonathan E. Tucker, PhD
1 From the Department of Radiology, Wilford Hall Medical Center, Lackland Air Force Base, Tex (A.B.M.H.A.C., J.E.T.); South Texas Radiology Group PA, San Antonio, Tex (M.J.L.); Departments of Emergency Medicine (R.T.G.) and Radiology (H.A.C., J.E.T.), Brooke Army Medical Center, Fort Sam Houston, Tex; Department of Emergency Medicine, Womack Army Medical Center, Fort Bragg, NC (S.K.); and Department of Radiology, Winthrop-University Hospital, Mineola, NY (D.S.K.). Received February 29, 2004; revision requested May 5; revision received October 6; accepted November 4.
Address correspondence to A.B.M., 822 Coldwater Creek Circle, Niceville, FL 32578 (e-mail: amackersie{at}hotmail.com).
 |
ABSTRACT
|
|---|
PURPOSE: To prospectively evaluate and compare the diagnostic accuracy of unenhanced helical computed tomography (CT) for patients with nontraumatic acute abdominal pain with that of traditional abdominal radiography.
MATERIALS AND METHODS: Institutional review board approval and informed consent were obtained; this study was completed before implementation of the HIPAA. Ninety-one patients (44 men; 47 women; age range, 18–84 years; mean age, 48.5 years) with acute nontraumatic abdominal pain over a 7-month period were referred by the emergency department of one institution. These patients underwent a three-view acute abdominal series (AAS) and unenhanced helical CT. AAS included an upright chest radiograph and upright and supine abdominal radiographs. Unenhanced helical CT images with 5-mm collimation were obtained from the lung bases to the pubic symphysis, without intravenous, oral, or rectal contrast material. AAS and unenhanced helical CT images were each separately and prospectively interpreted by a different experienced radiologist who was blinded to patient history and the images and interpretation of the other examination for each patient. Final diagnosis was established with surgical, pathologic, and clinical follow-up. The sensitivity, specificity, accuracy, positive predictive value, negative predictive value, and positive and negative likelihood ratios were calculated for AAS and unenhanced helical CT. Confidence intervals of 95% were calculated for each value with the standard equation for population proportions. Results of AAS and unenhanced helical CT examinations were compared with 
2 analysis.
RESULTS: Among the 91 patients examined, unenhanced helical CT yielded an overall sensitivity, specificity, and accuracy of 96.0%, 95.1%, and 95.6%, respectively. The AAS interpretations yielded an overall sensitivity, specificity, and accuracy of 30.0%, 87.8%, and 56.0%, respectively. The accuracy of unenhanced helical CT was significantly greater than the accuracy of AAS (P < .05).
CONCLUSION: AAS is an insensitive technique in the evaluation of nontraumatic acute abdominal pain in adults. Unenhanced helical CT is an accurate technique in the evaluation of adult patients with nontraumatic acute abdominal pain and should be considered as an alternative to radiography as the initial imaging modality.
|
INTRODUCTION
|
|---|
The broad spectrum of diseases and disorders that are indicated by acute abdominal pain, each with a myriad of presentations, continues to confound emergency physicians and other acute care providers. Many of these patients undergo some form of radiographic examination; traditionally, a single view of the abdomen is obtained, or a three-view acute abdominal series (AAS), which includes an upright chest radiograph and upright and supine radiographs of the abdomen, is performed. Often, imaging is ordered prior to the completion of the physical examination and laboratory analysis. The reason for this is likely multifactorial, including demands on the acute care physician to quickly diagnose the cause of the symptoms and provide a disposition in a busy emergency department. Cope (1) commented, "all who have had much experience of the group of cases known generally as the acute abdomen will probably agree that in that condition early diagnosis is exceptional." Thus, the appropriate imaging examination for these complex cases must be selected expeditiously and with certainty.
The increased availability of helical computed tomography (CT) in the late 1990s led to a series of articles published in Radiology in 1997 and 1998 in which Mindelzun and Jeffrey (2), Baker (3), and Roebuck and Metreweli (4) discussed the potential benefits and risks of using unenhanced helical CT compared with abdominal radiography in the evaluation of acute abdominal pain. The strongest criticism of unenhanced helical CT was directed at the increased radiation burden, as compared with abdominal radiography, while the lack of diagnostic accuracy was considered the biggest limitation of abdominal radiography.
Ahn et al (5) concluded "abdominal radiographs are not sensitive in the evaluation of adult patients presenting to the emergency department with nontraumatic abdominal pain." In this retrospective study, in which the diagnostic accuracy of abdominal radiography was examined in 871 patients, only a minority of patients underwent both abdominal radiography and CT. To our knowledge, with the exception of a study by Levine et al (6) that compared abdominal radiographs and unenhanced helical CT images for renal colic, there have been no prospective studies in which abdominal radiography has been directly compared with CT in patients with abdominal pain.
With the increasing availability of helical CT, including both helical CT and multi–detector row CT, unenhanced helical CT is emerging as an option for acute care physicians to use in the evaluation of patients who present with acute abdominal pain. The purpose of this study was to prospectively evaluate the diagnostic accuracy of unenhanced helical CT in patients with nontraumatic acute abdominal pain as compared with traditional abdominal radiography.
|
MATERIALS AND METHODS
|
|---|
Patient Selection
After approval by our institutional review board, 103 patients with nontraumatic acute abdominal pain were referred by the emergency department at our institution for unenhanced helical CT between April and October 2001. Informed consent was obtained by the emergency department physicians. Our study was completed prior to implementation of the Health Insurance Portability Accountability Act. Patients with symptoms of less than 7 days duration were referred. There were 12 patients who underwent unenhanced helical CT but were excluded because they did not undergo AAS. The 91 patients who underwent both unenhanced helical CT and AAS ranged in age from 18 to 84 years (average age, 48.5 years ± 18.7). There were 44 men who ranged in age from 19 to 79 years (average age, 46.8 years ± 19.5) and 47 women who ranged in age from 18 to 84 years (average age, 50.2 years ± 18.1). Patients who were clinically intoxicated or judged by the clinician not to be mentally competent to make medical decisions regarding their care were excluded. Pregnant women and women whose primary symptom was vaginal bleeding or vaginal discharge were also excluded. Men whose primary symptom was nonhemorrhagic penile discharge were excluded, as were any patients presenting with the primary symptom of dysuria or hematuria without flank pain. Finally, patients who declined to participate or withdrew voluntarily prior to completion of the study or for whom follow-up information could not be obtained were excluded.
Imaging Techniques
The AAS performed in each patient included upright chest radiography and upright and supine abdominal radiography. Radiographs were obtained by using standard imaging plates (ST-Vn; FujiFilm Medical Systems, Tokyo, Japan) and an image processor (FCR 5000; FujiFilm Medical Systems). In the abdomen, tube voltage was 75–85 kVp, and tube current was 50 mAs; in the chest, tube voltage was 110 kV and tube current was 20 mAs. The entire system is equivalent to a 200-speed system. Unenhanced helical CT was performed with either a four–detector row helical CT scanner and 5-mm collimation (pitch, 0.875; 120 kVp; 300 mAs) (MX8000; Marconi, Cleveland, Ohio) or a single–detector row helical CT scanner and 5-mm collimation (pitch, 1.6; 120 kVp; 240 mAs) (PQ 6000; Picker International, Highland Heights, Ohio). Transverse images were obtained from the lung bases to the pubic symphysis without intravenous, oral, or rectal administration of contrast material. Both examinations were performed within a 2-hour window for each patient. Radiation doses for AAS and unenhanced helical CT were estimated by a medical physicist (J.E.T.) using the data of Jones and Shrimpton (7) and the International Commission on Radiological Protection (8), respectively.
Image Interpretation and Final Diagnosis
AAS and unenhanced helical CT findings were prospectively interpreted by two radiologists (H.A.C. and M.J.L., respectively) who were blinded to the specific patient history and findings of a physical examination, laboratory analysis, and other imaging examinations for each patient. The images were interpreted with only the knowledge that patients presented with abdominal pain. The radiologist interpreting the unenhanced helical CT images was fellowship trained in abdominal imaging and had 7 years of postfellowship experience; the radiologist interpreting AAS findings had 49 years of experience interpreting abdominal radiographs. Final diagnosis was established with surgical, gross pathologic, and clinical follow-up. Final diagnosis was assigned after a senior emergency medicine resident (S.K.) and an emergency medicine physician who had 7 years of experience in emergency medicine after residency (R.T.G.) reviewed all the data. Patients were followed up for as many as 6 months to establish the final diagnosis. All available information was used to establish the final diagnosis, including radiologic and gross pathologic reports, follow-up interviews, and follow-up radiologic studies (if needed). For example, appendicitis was the final diagnosis after review of the final gross pathologic report. Nonspecific abdominal pain was diagnosed in patients whose symptoms resolved without identifiable cause and who continued to be asymptomatic at 6-month follow-up.
The participants in this study underwent all other examinations and laboratory studies that were deemed clinically appropriate by the referring physicians, including additional radiologic examinations that were indicated.
Statistical Analysis
Any AAS or unenhanced helical CT examination in which findings were suggestive of acute disease was considered a positive result. If the findings suggested diagnosis that was confirmed with surgical, gross pathologic, and/or clinical follow-up, the diagnosis was considered true-positive. For example, the finding of dilated gas-filled small-bowel loops with multiple air-fluid levels was considered a true-positive result at AAS in a patient with the final diagnosis of incarcerated inguinal hernia. Likewise, identification of findings believed to be clinically important that could not be corroborated with other imaging studies or with clinical or surgical follow-up was considered a false-positive result. Findings were considered true-negative if the AAS or unenhanced helical CT findings were not suggestive of an acute process and nonspecific abdominal pain was diagnosed in the patient. Finally, findings were considered false-negative when AAS or unenhanced helical CT findings did not indicate the acute process that was ultimately found in the patient.
In this study, the independent variable is the diagnostic test (ie, AAS or unenhanced helical CT). The dependent variable is the test result (ie, negative or positive). The null hypothesis is that there is no difference in accuracy between diagnostic tests. The alternative hypothesis is that unenhanced helical CT is more accurate than AAS. Sensitivity, specificity, positive predictive value, negative predictive value, accuracy, and positive and negative likelihood ratio were calculated. Confidence intervals of 95% were calculated for both AAS and unenhanced helical CT by using the standard equation for population proportions (9). The effects of the positive and negative likelihood ratios on the pretest and posttest probabilities were calculated along with the 95% confidence intervals and displayed in a graphic format. The results of AAS and unenhanced helical CT were also compared with the McNemar test. Proportions of men and women were compared by using the standard equation for population proportions to calculate 95% confidence intervals.
|
RESULTS
|
|---|
Unenhanced Helical CT versus AAS
Among the 91 patients examined with both unenhanced helical CT and AAS, unenhanced helical CT interpretations included 48 true-positive, 39 true-negative, two false-positive, and two false-negative diagnoses. The sensitivity, specificity, and accuracy of unenhanced helical CT are 96.0%, 95.1%, and 95.6% (95% confidence interval: 0.914, 0.998), respectively. The AAS interpretations included 15 true-positive, 36 true-negative, five false-positive, and 35 false-negative diagnoses, which yielded an overall sensitivity, specificity, and accuracy of 30.0%, 87.8%, and 56.0% (95% confidence interval: 0.458, 0.662), respectively. There was no overlap between the 95% confidence interval estimations for accuracy, which supports the alternative hypothesis. Additionally, the results of the AAS and unenhanced helical CT were compared with the McNemar test, which shows a statistically significant difference (two-tailed P < .001; 2 = 23.361, with one degree of freedom). The proportion of men is not significantly different from the proportion of women (P > .05). The positive and negative likelihood ratios for AAS are less than 10 and greater than 0.1, respectively. The positive and negative likelihood ratios for unenhanced helical CT are greater than 10 and less than 0.1, respectively (Table 1, Fig 1).
In our study, effective whole-body radiation doses for unenhanced multi–detector row CT, which were estimated by using the data of Jones and Shrimpton (7), were 1.2 rem (12 mSv) for men and 1.7 rem (17 mSv) for women. There was a difference in the radiation dose between men and women because the female reproductive organs are located in the scanning field, whereas the male reproductive organs are located outside of the scanning field. The effective whole-body radiation dose for AAS, which was estimated by using the guidelines set forth by the International Commission on Radiological Protection (8), was 0.244 rem (2.44 mSv).
Final Diagnoses
The most common final diagnoses (Table 2) were acute appendicitis (n = 10) (Fig 2), renal colic (n = 6), and acute diverticulitis (n = 6) (Fig 3). Unenhanced helical CT was used to make the correct diagnosis in all 22 patients with the most commonly diagnosed abnormalities, and AAS was used to detect abnormalities associated with these diagnoses in six of 22 patients. Malignancy was diagnosed in seven patients and consisted of nonmetastatic colon cancer (n = 2), gastric cancer (n = 1), and metastatic cancer (n = 4); of the patients with metastatic cancer, two had colon adenocarcinoma (n = 2) (Fig 4), one had small cell carcinoma, and one had pancreatic adenocarcinoma. Unenhanced helical CT was used to correctly diagnose all seven cases of malignancy, and AAS was used to correctly diagnose two of these seven cases.
View larger version (127K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2a. Images show acute appendicitis in a 40-year-old man. (a) Upright abdominal radiograph without evidence of acute abdominal process. AAS diagnosis, negative acute; result, false-negative. (b) Transverse unenhanced helical CT image demonstrates an appendicolith (black arrowhead), thickened appendix (white arrowhead), and periappendiceal fat stranding (arrow). Unenhanced helical CT diagnosis, acute appendicits; result, true-positive.
|
|
View larger version (133K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2b. Images show acute appendicitis in a 40-year-old man. (a) Upright abdominal radiograph without evidence of acute abdominal process. AAS diagnosis, negative acute; result, false-negative. (b) Transverse unenhanced helical CT image demonstrates an appendicolith (black arrowhead), thickened appendix (white arrowhead), and periappendiceal fat stranding (arrow). Unenhanced helical CT diagnosis, acute appendicits; result, true-positive.
|
|
View larger version (132K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3a. Images show acute diverticulitis in a 63-year-old-man. (a) Supine abdominal radiograph without evidence of acute abdominal process and with indeterminate pelvic opacity (arrowhead) AAS diagnosis, negative acute; result, false-negative. (b) Transverse unenhanced helical CT image shows multiple sigmoid diverticuli (arrowheads) and pericolonic fat stranding (arrow). Unenhanced helical CT diagnosis, diverticulitis; result, true-positive.
|
|
View larger version (161K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3b. Images show acute diverticulitis in a 63-year-old-man. (a) Supine abdominal radiograph without evidence of acute abdominal process and with indeterminate pelvic opacity (arrowhead) AAS diagnosis, negative acute; result, false-negative. (b) Transverse unenhanced helical CT image shows multiple sigmoid diverticuli (arrowheads) and pericolonic fat stranding (arrow). Unenhanced helical CT diagnosis, diverticulitis; result, true-positive.
|
|
View larger version (107K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4a. Images show metastatic colon cancer in a 45-year-old man. (a) Upright chest radiograph demonstrates a nodule in the cardiophrenic sulcus (arrow). AAS diagnosis, multiple pulmonary nodules; result, true-positive. (b) Transverse unenhanced helical CT image demonstrates multiple hypoattenuating liver nodules consistent with metastatic disease. (c) Transverse unenhanced helical CT image demonstrates cecal mass (m), with pericecal fat stranding (arrow) and pericolonic lymphadenopathy (arrowhead). Unenhanced helical CT diagnosis, metastatic colon cancer; result, true-positive.
|
|
View larger version (132K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4b. Images show metastatic colon cancer in a 45-year-old man. (a) Upright chest radiograph demonstrates a nodule in the cardiophrenic sulcus (arrow). AAS diagnosis, multiple pulmonary nodules; result, true-positive. (b) Transverse unenhanced helical CT image demonstrates multiple hypoattenuating liver nodules consistent with metastatic disease. (c) Transverse unenhanced helical CT image demonstrates cecal mass (m), with pericecal fat stranding (arrow) and pericolonic lymphadenopathy (arrowhead). Unenhanced helical CT diagnosis, metastatic colon cancer; result, true-positive.
|
|
View larger version (114K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4c. Images show metastatic colon cancer in a 45-year-old man. (a) Upright chest radiograph demonstrates a nodule in the cardiophrenic sulcus (arrow). AAS diagnosis, multiple pulmonary nodules; result, true-positive. (b) Transverse unenhanced helical CT image demonstrates multiple hypoattenuating liver nodules consistent with metastatic disease. (c) Transverse unenhanced helical CT image demonstrates cecal mass (m), with pericecal fat stranding (arrow) and pericolonic lymphadenopathy (arrowhead). Unenhanced helical CT diagnosis, metastatic colon cancer; result, true-positive.
|
|
False-positive diagnoses were made in two of the 91 patients with unenhanced helical CT; nonspecific abdominal pain was ultimately diagnosed in both patients. An initial diagnosis of small-bowel obstruction was made in one of these patients on the basis of unenhanced helical CT images, and a right upper quadrant inflammatory process was diagnosed in the other patient on the basis of fat stranding that appeared adjacent to the inferior margin of the liver. False-negative diagnoses were made with unenhanced helical CT in two of 91 patients. In one patient, choledocholithiasis was diagnosed at endoscopic retrograde cholangiopancreatography following negative unenhanced helical CT and negative right upper quadrant ultrasonography (US) findings. In the other patient, symptomatic cholelithiasis was diagnosed with right upper quadrant US.
There was one case in this study in which free intraperitoneal air was correctly diagnosed with both modalities (Fig 5). Small-bowel obstruction was ultimately diagnosed in three patients, and it was correctly diagnosed with both modalities (Fig 6). Gallstone pancreatitis was identified in one patient and correctly diagnosed with unenhanced helical CT after a false-negative AAS result (Fig 7).
View larger version (84K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 5a. Images show incarcerated ventral hernia with small-bowel obstruction and perforation in a 63-year-old woman. (a) Upright chest radiograph shows gas underneath the central portion of the diaphragm (cupola sign) (arrow). AAS diagnosis, pneumoperitoneum; result, true-positive. (b) Transverse unenhanced helical CT image shows free air in nondependent portion of the abdomen (arrow). (c) Transverse unenhanced helical CT image shows incarcerated ventral hernia and pneumatosis intestinalis (arrow). Unenhanced helical CT diagnosis, incarcerated ventral hernia with perforation and ischemic small bowel; result, true-positive.
|
|
View larger version (152K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 5b. Images show incarcerated ventral hernia with small-bowel obstruction and perforation in a 63-year-old woman. (a) Upright chest radiograph shows gas underneath the central portion of the diaphragm (cupola sign) (arrow). AAS diagnosis, pneumoperitoneum; result, true-positive. (b) Transverse unenhanced helical CT image shows free air in nondependent portion of the abdomen (arrow). (c) Transverse unenhanced helical CT image shows incarcerated ventral hernia and pneumatosis intestinalis (arrow). Unenhanced helical CT diagnosis, incarcerated ventral hernia with perforation and ischemic small bowel; result, true-positive.
|
|
View larger version (177K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 5c. Images show incarcerated ventral hernia with small-bowel obstruction and perforation in a 63-year-old woman. (a) Upright chest radiograph shows gas underneath the central portion of the diaphragm (cupola sign) (arrow). AAS diagnosis, pneumoperitoneum; result, true-positive. (b) Transverse unenhanced helical CT image shows free air in nondependent portion of the abdomen (arrow). (c) Transverse unenhanced helical CT image shows incarcerated ventral hernia and pneumatosis intestinalis (arrow). Unenhanced helical CT diagnosis, incarcerated ventral hernia with perforation and ischemic small bowel; result, true-positive.
|
|
View larger version (127K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 6a. Images show small-bowel obstruction in a 76-year-old man. (a) Upright radiograph of the abdomen demonstrates multiple dilated gas-filled small-bowel loops (arrowheads) with multiple multilevel air-fluid levels (arrows). AAS diagnosis, small-bowel obstruction; result, true-positive. (b) Transverse unenhanced helical CT image shows multiple dilated gas-filled small-bowel loops (arrowheads) with multiple multilevel air-fluid levels (arrows). Unenhanced helical CT diagnosis, small-bowel obstruction; result, true-positive.
|
|
View larger version (104K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 6b. Images show small-bowel obstruction in a 76-year-old man. (a) Upright radiograph of the abdomen demonstrates multiple dilated gas-filled small-bowel loops (arrowheads) with multiple multilevel air-fluid levels (arrows). AAS diagnosis, small-bowel obstruction; result, true-positive. (b) Transverse unenhanced helical CT image shows multiple dilated gas-filled small-bowel loops (arrowheads) with multiple multilevel air-fluid levels (arrows). Unenhanced helical CT diagnosis, small-bowel obstruction; result, true-positive.
|
|
View larger version (139K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 7a. Images show gallstone pancreatitis in an 84-year-old man. (a) Supine abominal radiograph without evidence of acute abdominal process. AAS diagnosis, negative acute; result, false-negative. (b) Transverse unenhanced helical CT image shows peripancreatic fat stranding and fluid in the anterior pararenal space (arrow) consistent with pancreatitis. (c) Transverse unenhanced helical CT image shows radiopaque gallstones (arrow) in gallbladder neck. Unenhanced helical CT diagnosis, gallstone pancreatitis; result, true-positive.
|
|
View larger version (111K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 7b. Images show gallstone pancreatitis in an 84-year-old man. (a) Supine abominal radiograph without evidence of acute abdominal process. AAS diagnosis, negative acute; result, false-negative. (b) Transverse unenhanced helical CT image shows peripancreatic fat stranding and fluid in the anterior pararenal space (arrow) consistent with pancreatitis. (c) Transverse unenhanced helical CT image shows radiopaque gallstones (arrow) in gallbladder neck. Unenhanced helical CT diagnosis, gallstone pancreatitis; result, true-positive.
|
|
View larger version (110K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 7c. Images show gallstone pancreatitis in an 84-year-old man. (a) Supine abominal radiograph without evidence of acute abdominal process. AAS diagnosis, negative acute; result, false-negative. (b) Transverse unenhanced helical CT image shows peripancreatic fat stranding and fluid in the anterior pararenal space (arrow) consistent with pancreatitis. (c) Transverse unenhanced helical CT image shows radiopaque gallstones (arrow) in gallbladder neck. Unenhanced helical CT diagnosis, gallstone pancreatitis; result, true-positive.
|
|
Of the 91 patients evaluated, only 10 underwent follow-up contrast material–enhanced CT with intravenous and oral contrast material. Among these 10 patients, there were no instances in which contrast-enhanced CT images were suggestive of an acute process that was not already identified on unenhanced helical CT images. Contrast-enhanced CT images were obtained before patients were released from the emergency department or discharged from the hospital.
|
DISCUSSION
|
|---|
The accurate clinical assessment of acute abdominal pain remains one of the more challenging areas of medicine. The variety of conditions that require emergent medical management, and often surgical management, vary widely in clinical presentation and physical examination and laboratory analysis findings. In patients with acute abdominal pain, CT has been shown to increase the referring physician's level of certainty in the diagnosis, reduce hospital admission rates, and help guide the therapeutic strategy, including surgical intervention (10,11). Several authors who have used unenhanced helical CT in the imaging of patients suspected of having appendicitis, diverticulitis, and renal colic have described a broad range of disorders identified on CT images, which mimic these very common conditions (12–15). This is a direct challenge to the use of the time-honored and inexpensive technique of abdominal radiography, which radiologists have taught to countless clinicians and surgeons, as the initial imaging modality of choice in patients with acute abdominal pain. Balthazar (16) once paraphrased the aphorism of renowned French physiologist Claude Bernard that the greatest hindrance to our learning and accepting new facts is the baggage of knowledge we already possess and with which we live comfortably. Balthazar goes on to say that "changing our way of practice and abandoning cherished procedures for a newer unfamiliar and uncertain world is always difficult" (16).
In this study, we sought to prospectively evaluate the accuracy of interpretation of AAS and unenhanced helical CT findings in adult patients who present with acute abdominal pain. Unenhanced helical CT was shown to be more accurate than AAS, as one might expect (P < .05). Two questions remain: (a) did AAS add any information and (b) was AAS effective in the evaluation of adults with nontraumatic acute abdominal pain? In this series, there were no patients in whom AAS was used to make diagnoses that were missed at unenhanced helical CT. Considering all patients in this study, the sensitivity of AAS was 30%, with a negative predictive value of 51%. Additionally, the likelihood ratios calculated for AAS suggest that the test is unlikely to substantially affect the pretest probability. Given the added cost of $136–$184 (2), the increased radiation dose of 0.244 rem (2.44 mSv) (8), and the relatively poor performance, the results of this series would not support the generalized use of AAS in patients with acute abdominal pain.
Proponents of AAS believe that abdominal radiographs are more accurate than CT images in the identification of free intraperitoneal air and bowel obstruction (3). There was one patient in this study in whom free intraperitoneal air was correctly diagnosed with both modalities. In this patient, only unenhanced helical CT revealed the cause of the intraperitoneal air, which was an incarcerated ventral hernia with pneumatosis intestinalis and perforation. In this study, small-bowel obstruction was ultimately diagnosed in three patients; in these patients, the correct diagnosis was made with both modalities. Small-bowel obstruction was incorrectly diagnosed on the basis of AAS findings in two patients; however, unenhanced helical CT images were negative for an acute process, and the correct diagnosis was made. In another patient, unenhanced helical CT findings were interpreted as indicative of small-bowel obstruction, AAS findings were interpreted as indicative of a "possible right paraumbilical mass," and nonspecific abdominal pain was ultimately diagnosed in this patient.
In this series, AAS was shown to have a sensitivity of 100% in the detection of small-bowel obstruction; however, unenhanced helical CT was used to correctly identify the cause and/or location of the obstruction in all three patients, thus directing the appropriate course of action in these patients. We believe, as do our surgical colleagues, that prior to commencing an invasive surgical procedure, it is incumbent on the surgeon to gather as much information as possible to minimize complications.
One should not conclude from these data that all patients with acute abdominal pain should undergo unenhanced helical CT. Likewise, one should not assume that these data suggest that patients should never undergo AAS. The data do suggest that interpretation of AAS findings provides no additional information beyond that obtained with unenhanced helical CT. Additionally, the data suggest that using AAS to check for common causes of adult nontraumatic acute abdominal pain in all such patients is ineffective.
The fact that unenhanced helical CT exposes the patient to more radiation than does AAS is not in dispute. Given the data obtained in this study, however, there is no statistically significant relationship between the results of AAS and the final diagnosis (P < .001). From these data, one might conclude that any radiation from AAS was unnecessary, given its poor diagnostic capabilities in patients with acute abdominal pain. Again, given the poor sensitivity, negative predictive value, and likelihood ratios of AAS, the time and money spent pursuing this examination in all patients with acute abdominal pain could likewise be considered unnecessary. The care of each patient should, of course, be planned on a case-by-case basis. Although our emergency medicine coinvestigators are currently attempting to elucidate clinical guidelines for the evaluation of patients who present with acute abdominal pain, to our knowledge, there is no well-accepted algorithm in place at the present time.
One might question the value of unenhanced helical CT if the patient ultimately will undergo contrast-enhanced CT with oral, intravenous, and possibly rectal contrast material. In this series, among the 10 patients that underwent follow-up contrast-enhanced CT, there were no cases in which contrast-enhanced CT findings changed the diagnosis that was made on the basis of unenhanced helical CT findings. Findings in some studies have shown that only inexperienced readers derive significantly increased accuracy from contrast-enhanced CT as opposed to unenhanced CT (17). However, specific clinical diagnoses, such as pyelonephritis, pancreatitis, aortic dissection, and ischemic bowel, may be best evaluated with the use of intravenous contrast material and a tailored examination (18). Additionally, patients suspected of having biliary disease or female patients with the primary symptom of pelvic pain may be best evaluated with US as the initial imaging study. Again, the care and work-up of each patient must be performed on a case-by-case basis, given the experience and comfort level of the interpreting radiologist.
Limitations of this study include possible selection bias, as the patients were referred by the emergency medicine physicians at a rate of two or three patients per day after obtaining informed consent. Selection bias may explain the difference in proportions of diagnoses in this series compared with what might be expected for all patients presenting with nontraumatic acute abdominal pain. For example, in this series, a final diagnosis of nonspecific abdominal pain was made in 41 (45%) of the 91 participants, whereas this diagnosis was expected in 34% of the pateints; only 10 (11%) of patients had appendicitis, whereas appendicitis was expected in 28% of patients by referencing the de Dombal study (19). Alternatively, our patient population may differ from that of de Dombal (19), which may make these results less applicable to the general population.
Additional limitations include the fact that some patients did not undergo AAS; thus, they were excluded. The number of patients in this study is relatively small given the many varied conditions encompassed by this study. Also, our reference standard for establishing a final diagnosis varied on the basis of the diagnosis.
In a retrospective study of 1000 patients with nontraumatic acute abdominal pain, Ahn et al (5) concluded "abdominal radiographs are not sensitive in the evaluation of adult patients presenting to the emergency department with nontraumatic abdominal pain." In their study, 871 patients underwent abdominal radiography, and 188 patients underwent CT (predominantly contrast-enhanced CT); 14% (120 of 871 patients) of the patients underwent both radiography and CT. The findings of radiography were normal (official reports) in 24 (20%), nonspecific in 91 (76%), and abnormal in five (4%) of 120 patients who underwent both abdominal radiography and CT. Other researchers have concluded that abdominal radiography is of limited use in the assessment of patients with acute abdominal pain (20–23). Our study differs from previous studies in that our study, to our knowledge, is the first prospective direct comparison of abdominal radiography and unenhanced CT in the evaluation of patients with acute abdominal pain.
The decision of whether or not to obtain radiologic images of patients who present with acute abdominal pain remains a matter of clinical judgment. In patients in whom imaging is deemed necessary, should unenhanced helical CT be used as the initial imaging study in place of AAS? Any examination that would be substituted for AAS as the initial imaging study in patients with acute abdominal pain should have the following characteristics: It should be more accurate than AAS and equally as fast in the diagnosis of common abdominal processes. It should require no particular preparation. Ideally, it would involve equal or less radiation exposure to the patient. Unfortunately, an examination matching that exact description does not exist; however, unenhanced helical CT has been shown to be very accurate in the diagnosis of common disorders such as acute appendicitis, renal colic, and diverticulitis (12–15,24). Unenhanced helical CT requires no patient preparation, and the examination can be performed in less than 20 seconds.
The National Academy of Sciences noted "the possibility that there may be no risks from exposures comparable to external natural background radiation cannot be ruled out. At such low doses and dose rates, it must be acknowledged that the lower limit of the range of uncertainty in the risk estimates extends to zero." In a position statement of the Health Physics society, the following statement was made: "The Health Physics Society recommends that assessments of radiogenic health risks be limited to dose estimates near and above 10 rem. Below this level, only dose is credible and statements of associated risks are more speculative than credible." At this time, it may not be possible to accurately assess the exact risk incurred by the radiation exposure from a diagnostic procedure such as CT with dose levels of less than 10 rem (100 mSv). However, prudence dictates that we minimize radiation exposure. Certainly, removing the ineffective AAS from routine use in the evaluation of these patients would be a step in the right direction.
In conclusion, AAS is an insensitive technique in the evaluation of nontraumatic acute abdominal pain. Unenhanced helical CT is an effective technique in the evaluation of patients with nontraumatic acute abdominal pain, and it should be considered as an alternative to abdominal radiography as the initial imaging modality.
|
FOOTNOTES
|
|---|
Abbreviations: AAS = acute abdominal series
Authors stated no financial relationship to disclose.
The opinions or assertions contained herein are those of the authors and should not be construed as representing the views of the U.S. Department of Defense or the U.S. Government.
Author contributions: Guarantors of integrity of entire study, A.B.M., M.J.L., S.K.; study concepts and design, M.J.L., S.K., D.S.K.; literature research, A.B.M., S.K., J.E.T., D.S.K.; clinical studies, A.B.M., M.J.L., H.A.C., S.K., J.E.T.; data acquisition, all authors; data analysis/interpretation, A.B.M., M.J.L., R.T.G., S.K., D.S.K.; statistical analysis, A.B.M., R.T.G.; manuscript preparation, A.B.M., R.T.G., S.K., D.S.K.; manuscript definition of intellectual content, A.B.M., M.J.L., R.T.G.; manuscript revision/review, A.B.M., M.J.L., S.K., D.S.K.; manuscript editing and final version approval, A.B.M., M.J.L., D.S.K.
|
References
|
|---|
- Cope Z. Extract from the preface to the first edition. In: Silen W. Cope's early diagnosis of the acute abdomen. 20th ed. New York, NY: Oxford, 2000; ix–x.
- Mindelzun RE, Jeffrey RB. Unenhanced helical CT for evaluating acute abdominal pain: a little more cost, a lot more information. Radiology 1997;205:43–45.[Free Full Text]
- Baker SR. Unenhanced helical CT versus plain abdominal radiography: a dissenting opinion [comment]. Radiology 1997;205:45–47.[Free Full Text]
- Roebuck DJ, Metreweli C. Radiation risk in CT for acute abdominal pain [letter]. Radiology 1998;209:287–288.[Medline]
- Ahn SH, Mayo-Smith WW, Murphy BL, Reinert SE, Cronan JJ. Acute nontraumatic abdominal pain in adult patients: abdominal radiograhy compared with CT evaluation. Radiology 2002;225:159–164.[Abstract/Free Full Text]
- Levine JA, Neitlich JD, Verga M, et al. Identification of ureteral calculi on plain radiographs in patients with flank pain: correlation with helical CT. Radiology 1997;204:27–31.[Abstract/Free Full Text]
- Jones DG, Shrimpton PC. Survey of CT practice in the UK. III. Normalised organ doses calculated using Monte Carlo techniques. Chilton, NRPB-R250. London, England: HMSO, 1991.
- International Commission on Radiological Protection (ICRP). Publication 62: radiological protection in biomedical research. Oxford, England: Pergamon, 1993.
- Duncan RC, Knapp RG, Miller MC 3rd. Introductory biostatistics for the health sciences. 2nd ed. New York, NY: Wiley, 1983; 161–205.
- Rosen MP, Sands DZ, Longmaid HE 3rd, Reynolds KF, Wagner M, Raptopoulos V. Impact of abdominal CT on the management of patients presenting to the emergency department with acute abdominal pain. AJR Am J Roentgenol 2000;174:1391–1396.[Abstract/Free Full Text]
- Siewert B, Raptopoulos V, Mueller MF, Rosen MP, Steer M. Impact of CT on diagnosis and management of acute abdomen in patients initially treated without surgery. AJR Am J Roentgenol 1997;168:173–178.[Abstract/Free Full Text]
- Lane MJ, Liu DM, Huynh MD, Jeffrey RB, Mindelzun RE, Katz DS. Suspected acute appendicitis: nonenhanced helical CT in 300 consecutive patients. Radiology 1999;213:341–346.[Abstract/Free Full Text]
- Lane MJ, Mindelzun RE. Appendicitis and its mimickers. Semin Ultrasound CT MR 1999;20:77–85.[CrossRef][Medline]
- Smith RC, Levine J, Dalrymple NC, Barish M, Rosenfield AT. Acute flank pain: a modern approach to diagnosis and management. Semin Ultrasound CT MR 1999;20:108–135.[CrossRef][Medline]
- Rao PM. CT of diverticulitis and alternative conditions. Semin Ultrasound CT MR 1999;20:86–93.[CrossRef][Medline]
- Balthazar EJ. Preface. Radiol Clin North Am 1994; 32:xi–xii.
- Jacobs JE, Birnbaum BA, Macari M, et al. Acute appendicitis: comparison of helical CT diagnosis-focused technique with oral contrast material versus nonfocused technique with oral and intravenous contrast material. Radiology 2001;220:683–690.[Abstract/Free Full Text]
- Urban BA, Fishman EK. Tailored helical CT evaluation of acute abdomen. RadioGraphics 2000;20:725–749.[Abstract/Free Full Text]
- de Dombal FT. The OMGE acute abdominal pain survey: progress report—1986. Scand J Gastroenterol Suppl 1988; 144:35–42.[Medline]
- Eisenberg RL, Heinken P, Hedgcock MW, Federle M, Goldberg HI. Evaluation of plain abdominal radiographs in the diagnosis of abdominal pain. Ann Surg 1983;197:464–469.[Medline]
- Campbell JP, Gunn AA. Plain abdominal radiographs and acute abdominal pain. Br J Surg 1988;75:554–556.[Medline]
- Greene CS. Indications for plain abdominal radiography in the emergency department. Ann Emerg Med 1986;15:257–260.[CrossRef][Medline]
- Brewer BJ, Golden GT, Hitch DC, Rudolf LE, Wangensteen SL. Abdominal pain: an analysis of 1,000 consecutive cases in a university hospital emergency room. Am J Surg 1976;131:219–223.[CrossRef][Medline]
- Malone AJ. Unenhanced CT in the evaluation of the acute abdomen: the community hospital experience. Semin Ultrasound CT MR 1999;20:68–76.[CrossRef][Medline]
This article has been cited by other articles:
|
|
|
|
 
A. van Randen, S. Bipat, A. H. Zwinderman, D. T. Ubbink, J. Stoker, and M. A. Boermeester
Acute Appendicitis: Meta-Analysis of Diagnostic Performance of CT and Graded Compression US Related to Prevalence of Disease
Radiology,
August 5, 2008;
(2008)
2483071652.
[Abstract]
[Full Text]
|
|
|
|
|
|
|
CT Beats Plain Films in Patients with Abdominal Pain
Journal Watch (General),
October 14, 2005;
2005(1014):
2 - 2.
[Full Text]
|
|
|
TOP
ABSTRACT
INTRODUCTION
= Positive test result, 
= negative test result. AAS has a slight effect on posttest probability, whereas unenhanced helical CT has a more pronounced effect. Postitive and negative likelihood ratios for both AAS and unenhanced helical CT, including the 95% confidence intervals, are included in 