| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Gastrointestinal Imaging |
1 From the Departments of Radiology (I.P., D.L., B.S., N.M.R.), Obstetrics and Gynecology (D.L., A.D.E.), and General Medicine and Primary Care (L.N.), Beth Israel Deaconess Medical Center, One Deaconess Rd, Boston, MA 02215; and Harvard Medical School, Boston, Mass (I.P., D.L., B.S., N.M.R.). Received January 28, 2005; revision requested March 21; revision received April 15; accepted June 1; final version accepted July 5. Address correspondence to I.P. (e-mail: ipedrosa{at}bidmc.harvard.edu).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Materials and Methods: The study was approved by the committee on clinical investigations and was HIPAA compliant. The informed consent requirement was waived. MR images were obtained in 51 consecutive pregnant patients (mean age, 28.3 years) who were clinically suspected of having acute appendicitis. In this protocol for pregnant patients, MR imaging is performed when findings at ultrasonography (US) are inconclusive or additional information is needed. Four patients had appendicitis, which was confirmed at surgery in three patients and at follow-up computed tomography in one patient. Initial interpretations were used for patient care and to calculate diagnostic accuracy. The appendix was considered normal at MR imaging if its diameter was less than or equal to 6 mm or if it was filled with air, oral contrast material, or both. An enlarged fluid-filled appendix (>7 mm in diameter) was considered an abnormal finding. An appendix with a diameter of 6–7 mm was considered an inconclusive finding; in those cases, the presence of periappendiceal inflammation was used for the final diagnosis. Three radiologists retrospectively assessed the visualization of the appendix by using a 5-point scale. Statistical analysis was performed by using the median and Fisher exact tests and the Spearman correlation coefficient.
Results: MR images were positive for appendicitis in four patients and inconclusive in three. In the three patients with inconclusive results, the appendix was not seen in two patients and was borderline enlarged (7 mm in diameter) in the third. The overall sensitivity, specificity, prevalence-adjusted positive and negative predictive values, and accuracy for MR imaging was 100%, 93.6%, 1.4%, 100%, and 94.0%, respectively.
Conclusion: MR imaging is an excellent modality for use in excluding acute appendicitis in pregnant women who present with acute abdominal pain and in whom a normal appendix is not visualized at US.
© RSNA, 2006
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Difficulty in rendering a definitive diagnosis of appendicitis may result in delayed treatment and complications from delayed diagnosis of appendicitis. For example, appendiceal perforation is much more common in pregnant women than in the general population, with reported rates up to 55% (1,5).
The lack of specificity in signs and symptoms has also resulted in a higher acceptable false-negative laparotomy rate in pregnant women compared with that of the general population. Indeed, the frequency of appendicitis among pregnant patients who undergo appendectomy varies from 48% to 80% (6). Imaging studies can facilitate a presurgical diagnosis and thus decrease the number of unnecessary surgeries without compromising the outcome of patients with acute appendicitis (7–9).
Ultrasonography (US) with a graded compression technique is the imaging modality of choice in pregnant patients who present with right lower quadrant pain because of its availability and lack of ionizing radiation (10,11). This technique, however, has the following limitations: (a) Graded compression US may not be feasible owing to the size of the enlarged gravid uterus, particularly in the 3rd trimester (11); (b) a normal appendix is visualized in 13%–50% of patients who are not pregnant; and (c) the negative predictive value of a nonvisualized appendix is, at best, 90% (11–16).
Computed tomography (CT), which is often the modality of choice in the evaluation of acute appendicitis in patients who are not pregnant, has an estimated radiation dose as high as 30 mGy (3 rad) to the uterus with use of conventional protocols (17). Thus, particularly in pregnant patients, there is a need for an accurate noninvasive imaging technique that avoids ionizing radiation in patients who present with right-sided pain. In this regard, magnetic resonance (MR) imaging offers an alternative.
To our knowledge, the value of MR imaging in the assessment of pregnant women suspected of having acute appendicitis has been reported in only a small series of patients (18,19), with no reported data about the sensitivity, specificity, and positive and negative predictive values. Thus, the aim of our study was to retrospectively assess the diagnostic performance of MR imaging in pregnant patients suspected of having acute appendicitis.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Imaging Protocol and Initial Image Interpretation
In our typical clinical protocol, US is used as the first imaging assessment of pregnant patients with right lower quadrant pain. MR imaging has typically been restricted to those patients in whom US findings are inconclusive or in whom additional questions must be answered. US was not performed in three of the 51 patients, however, because the clinical staff decided to proceed directly to MR imaging. Two patients underwent CT after MR imaging. One of these patients underwent CT at our institution to help confirm "tip" appendicitis. In the second patient, CT was performed at an outside institution for a repeat episode of right lower quadrant pain.
Informed consent was obtained from all patients before the MR examination.
Patients received an oral contrast material preparation starting 1–1.5 hours before the MR examination. A combination of 300 mL of ferumoxsil (Gastromark; Mallinckrodt Medical, St Louis, Mo) and 300 mL of barium sulfate (Readi-Cat 2; E-Z-Em Canada, Westbury, NY) was used. This solution provides negative contrast on T1- and T2-weighted images within the bowel lumen while eliminating problematic susceptibility artifacts (20). Intravenous contrast material was not used.
MR examinations were performed with 1.5-T units (Vision, Siemens, Iselin, NJ; or Excite TwinSpeed, GE Medical Systems, Waukesha, Wis) and a surface phased-array coil. The patient was imaged in the supine position. Examinations were monitored by a radiologist to ensure adequate coverage of the appendix and area of interest. Delayed repeat imaging was deemed necessary by the monitoring radiologist in two patients to visualize contrast material in the cecum. All sequences were performed during a breath hold of 20–24 seconds, and the total examination time was approximately 30 minutes.
Half-Fourier single-shot fast spin-echo (SE) images were obtained in the transverse, coronal, and sagittal planes through the lower abdomen and pelvis with the following parameters: repetition time msec/echo time msec, 800–1100/60 (in this case, the repetition time is the time between successive single-shot excitations); 4-mm-thick sections; 1-mm gap; 192 x 256 matrix; 130°–155° flip angle; 62-kHz bandwidth; and 35–40-cm field of view (FOV). Transverse single-shot fast SE images were obtained by using the same imaging parameters and frequency-selective fat saturation.
Transverse time-of-flight T2*-weighted gradient-echo images were obtained with the following parameters: repetition time of 30 msec, minimum full echo time, 45° flip angle, 3-mm-thick sections, 1-mm gap, 256 x 128 matrix, 31-kHz bandwidth, and 35-cm FOV. Transverse T1-weighted in-phase and opposed-phase images were obtained with the following parameters: repetition time of 205 msec, echo times of 2.2 and 4.5 msec, 80° flip angle, 5-mm-thick sections, 2-mm gap, 160 x 256 matrix, and 35-cm FOV.
Initial interpretations were provided by one of four attending radiologists who covered the abdominal MR section on a given day, two of whom (I.P., N.M.R.) were authors. The MR experience of the four radiologists at the time of their first interpretation in this study was 1 year (I.P.), 3 years, 9 years (N.M.R.), and 23 years. The appendix was considered normal when its diameter was equal to or less than 6 mm and/or it was filled with oral contrast material, air, or both. Low signal intensity on T2-weighted images and blooming effect caused by magnetic susceptibility on T2*-weighted images within the appendix was used as a sign that there was air or oral contrast material within (Fig 1).
|
|
Initial diagnoses were recorded as positive, negative, or inconclusive for appendicitis. Inconclusive diagnoses were rendered early in the series in patients in whom the appendix was not visualized but no inflammatory changes were noted. Later in the series, this constellation was interpreted as a negative finding.
Data Collection
Data regarding the results of the initial interpretation of the US, CT, and MR findings; clinical examination (eg, symptoms, physical examination, and white blood cell count); and surgical, pathologic, and clinical follow-up were collected by two authors (I.P., A.D.E.). In the four patients with appendicitis, the reference standards were findings at surgery and pathologic examination (n = 3) and a confirmatory CT scan (n = 1) (tip appendicitis that improved after antibiotic therapy). The reference standards in the 47 patients without appendicitis were clinical follow-up (n = 45) and surgery (n = 2) (one patient had an ectopic pregnancy, and one had ovarian torsion).
Retrospective Review of MR Images
Three radiologists (D.L., I.P., and B.S., with 11, 3, and 2 years of MR imaging experience, respectively) retrospectively reviewed the MR images in consensus. Differences of opinion were resolved by consensus. The time between the last initial interpretation and the retrospective review was 3 months. The MR image review was conducted without information about the US findings or outcome; however, two reviewers were involved in patient care at the initial evaluation and, therefore, it is possible that they were aware of the initial interpretation and clinical outcome at the retrospective review. To minimize potential recall bias, data regarding the initial interpretation of US, CT, and MR images were collected by an author (I.P.) after the retrospective review was performed.
To help confirm that a suspected structure was the appendix, the cross-reference tool in our picture archiving and communication system (Centricity; GE Medical Systems, Milwaukee, Wis) was used; an appearance consistent with that of the appendix had to be present in each of the three standard planes to confirm that the structure was the appendix. Time-of-flight images were used to ensure that the region thought to be the appendix was not a vein. A 5-point scale was used to rate the identification of the appendix, as follows: 1 = not identified, 2 = less than half identified, 3 = approximately half identified, 4 = more than half identified, and 5 = well visualized in its entirety.
The transverse diameter of the appendix was measured in millimeters. The presence or absence of periappendiceal fat stranding, phlegmon, or abscess was noted. The location of most of the appendix relative to the vertebral bodies and disk spaces of the lumbosacral spine was determined by using the lowest non–rib-bearing vertebral body as L1.
The presence or absence of oral contrast material in the cecum was noted. Additional findings on the MR images that could help explain the patients' pain were noted, including abnormalities of the right adnexa, degenerated fibroids, right-sided hydronephrosis (graded subjectively as mild, moderate, or severe), right-sided gonadal vein enlargement of more than 1 cm, venous thrombosis, and subchorionic hemorrhage.
Statistical Analysis
The initial interpretations of the MR images were used to calculate the accuracy of MR imaging in the diagnosis of acute appendicitis. For the two patients who underwent more than one MR examination, only the first examination was used to determine the accuracy of MR imaging findings. Inconclusive MR interpretations were considered false-positive findings in the calculation of sensitivity and specificity.
To assess the prevalence of appendicitis in our pregnant population, a Health Insurance Portability and Accountability Act–compliant retrospective review of pregnant women in our hospital database between January 1999 and September 2004 was performed, with approval of our institutional review board and waiver of informed consent. These findings were used to calculate sensitivity, specificity, positive and negative predictive values, and accuracy. These data are reported with 95% confidence intervals.
Nonparametric statistical tests were used to evaluate continuous patient characteristic variables (median test) and categorical variables (Fisher exact test).
The relationship between the location of the appendix and gestational age was examined graphically by using the cubic smoothing spline, and the Spearman correlation coefficient was computed on a scale based on vertebral body and disk space level from L2 to S2.
All statistical analyses were performed with statistical software (SAS version 9.1; SAS Institute, Cary, NC). A P value of less than .05 was considered to indicate a significant difference.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
US showed acute appendicitis in two of the 48 patients, and appendicitis was confirmed at surgery. In the other 46 patients (44 without appendicitis, two with appendicitis), the appendix was not visualized at US. CT helped confirm tip appendicitis in one patient who improved clinically after antibiotic therapy.
MR images were initially interpreted as positive for acute appendicitis in the four patients with appendicitis (Figs 2–4; Table 1). MR images were initially interpreted as negative in the 44 patients who did not have appendicitis.
|
|
|
|
|
|
|
|
|
|
|
|
Retrospective Review of MR Images
Oral contrast material was visible on MR images in the cecum in 37 of the 51 patients (72%) and was not visible in the cecum in 14 (27%). Visualization of the appendix was achieved more often in patients with contrast material in the cecum (n = 37, 86.5%) than in those without contrast material in the cecum (n = 14, 78.6%); however, these differences were not statistically significant (P = .67). The appendix was not identified in eight of the 51 patients. In the remaining patients, approximately half of the appendix was identified in eight patients and more than half of the appendix was identified in six. In 29 patients, the appendix was well visualized in its entirety. There was a trend toward better visualization of the appendix with an earlier gestational age; however, this trend was not statistically significant (P = .1). Of the 47 patients without acute appendicitis, MR imaging showed a normal appendix in 39 (83%). The appendix could not be seen in eight of those 47 patients (17%).
The mean appendiceal diameter was 4.9 mm (range, 3–7 mm) in 47 patients with no appendicitis versus 11.6 mm (range, 7–17 mm) in patients with appendicitis (P < .01). Periappendiceal fat stranding was present in three of the four patients with appendicitis. MR imaging showed an appendiceal phlegmon in one patient with appendicitis.
There was moderate correlation between gestational age and the anatomic level of the appendix (Spearman correlation coefficient, 0.44; P < .01; Fig 6).
|
|
Six patients had degenerated fibroids at the site of maximum tenderness in the abdomen. Findings at MR imaging in these patients included diffuse central high signal intensity on T1-weighted images, diffuse increased signal intensity or focal areas of high signal intensity on T2-weighted images, and high signal intensity surrounding the fibroid on T2-weighted images.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
MR imaging during pregnancy has no known deleterious effects to the fetus, and its use in patients who need additional imaging in pregnancy has been advocated by the safety committee of the Society of Magnetic Resonance Imaging (21,22).
At MR imaging, the normal collapsed appendix is seen as a tubular structure with a diameter equal to or smaller than 6 mm and an intermediate signal intensity similar to that of muscle on T1- and T2-weighted images. The presence of air or oral contrast material within the lumen of the appendix yields low signal intensity in the central region of the appendix. The diagnosis of acute appendicitis can be excluded in those patients with complete filling of the appendiceal lumen by air or oral contrast material. Oto et al (18) found a normal appendix on MR images in 17 of 19 (89%) pregnant patients with right lower quadrant pain. We had a similar experience, with visualization of the normal appendix in 39 of 47 patients (83%). The high rate of visualization of the appendix was likely due to the use of single-shot fast SE sequences that limit artifacts from bowel and breathing motion and yield strong contrast between the appendix and the periappendiceal fat. In addition, the blooming effect caused by air and/or oral contrast material within the appendiceal lumen on T2*-weighted time-of-flight images can help differentiate small blood vessels (with high signal intensity) from the appendix (with low signal intensity). The obstructed fluid-filled appendix demonstrates high signal intensity within its lumen on T2-weighted images and can be readily recognized when the adjacent bowel is adequately opacified with negative oral contrast material.
The relatively poor positive predictive value in our series deserves comment. We had four true-positive and three false-positive findings; in many cases, the positive predictive value would have been reported as 57%. The correct value, however, must incorporate the prevalence of disease as we have done; this approach yields a dramatically reduced value of 1.4%. We anticipate that this value will improve with larger studies, more experience with the technique, and refined criteria.
Nonetheless, the 100% negative predictive value is of great merit, providing an effective means for ruling out a surgical diagnosis in this patient population with a low prevalence of the disorder. With use of MR imaging in this manner, many patients can be spared radiation exposure from CT that may have been otherwise needed to determine patient treatment.
The natural history in patients with acute appendicitis is usually perforation or abscess formation if appendectomy is not performed (23). Spontaneous resolution of appendicitis is seen in 8% of cases (23). Approximately 38% of patients with spontaneously resolving appendicitis have recurrent appendicitis, with 70% occurring within the year after the first event (23).
Alternative Diagnoses
Of importance is that MR imaging can provide an alternative diagnosis in pregnant patients with abdominopelvic pain. Findings in patients in their 1st trimester without appendicitis included ruptured hemorrhagic corpus luteum cysts (n = 2) and ectopic pregnancy (n = 1).
Disorders related to the ovarian vein are important considerations in this patient population. Although time-of-flight images were initially included in our MR imaging protocol because of the potential to detect ovarian vein thrombosis, no such cases were demonstrated. These sequences, however, were helpful for differentiating veins from a contrast material–filled appendix.
Eleven patients had right gonadal vein enlargement, which was discovered in the late 2nd trimester and beyond. Similar results have been described at laparotomy in pregnant patients with right-sided abdominal pain (24).
Smooth muscle relaxation in the ureters secondary to hormonal changes and extrinsic compression by the gravid uterus are responsible for the common ureteral dilatation seen in pregnancy, which is more pronounced on the right side and is a common cause of right-sided pain. Obstructing stones are also seen in pregnancy as a cause of right-sided pain. In our series, moderate to severe hydronephrosis, changes consistent with recent stone passage, and gas in the urinary bladder were all findings suggestive of a urologic cause of pain.
Degenerating fibroids can cause abdominopelvic pain in pregnancy (25). We found degenerated fibroids in six of our patients. Point tenderness elicited when pressure is applied to a fibroid with a US probe is an excellent ancillary feature supportive of a degenerative fibroid as a source of pain. MR imaging can show features of red degeneration and is useful in those circumstances in which direct compression of the fibroid is not possible (26,27).
Limitations
Our study was limited by selection bias because not all pregnant patients with abdominopelvic pain underwent imaging. In addition, many of these patients at our institution undergo US as the only imaging modality.
A second limitation is that the initial interpretations of the MR images were performed by the attending radiologist who covered the service and was aware of the US results in most cases. Therefore, the sensitivity of MR imaging could be falsely elevated on the basis of the nonblinded nature of these interpretations. This reflects the general practice of MR imaging, however, and the way we expect it to be used in the future.
Third, the number of patients with acute appendicitis in our series is small. The statistical calculations most dependent on the small number of positive cases yielded a broad range for their 95% confidence interval. More studies are needed with a larger sample size to help verify the sensitivity and positive predictive value of MR imaging. The relatively tight confidence interval for the negative predictive value of MR imaging, however, shows that this technique is a reliable means for excluding appendicitis in pregnancy.
A fourth limitation of our study is that we do not have pathologic confirmation in one of the cases in which MR imaging findings were positive for appendicitis. The findings of thickening limited to the appendiceal tip and minimal inflammatory changes in the periappendiceal fat, however, were confirmed with a correlative CT examination. The compelling, corroborative imaging features and the clinical improvement after antibiotic therapy were consistent with the diagnosis of acute appendicitis.
A fifth limitation is the change in interpretation of inconclusive studies during the course of this study. Early in the series, findings from MR imaging were interpreted as inconclusive if the appendix was not visualized. This was considered a false-positive result for the purpose of the statistical analysis and accounts for the low positive predictive value in our study. Later in the study, MR images that did not show inflammatory changes in the right lower quadrant were considered negative for acute appendicitis. This change reflects that learning curve during the course of the study.
In conclusion, MR imaging is an excellent modality for excluding acute appendicitis in pregnant women who present with acute abdominal pain and in whom the normal appendix is not visualized at US. On the basis of the low prevalence of acute appendicitis during pregnancy, MR imaging has the potential to eliminate unnecessary radiation from CT in a large number of patients by providing direct visualization of the normal appendix. Furthermore, MR imaging can offer an alternative diagnosis in a substantial number of pregnant women with right-sided abdominal pain. More studies with larger series of patients are needed to establish the accuracy of MR imaging in the diagnosis of acute appendicitis during pregnancy.
|
FOOTNOTES |
|---|
Abbreviations: FOV = field of view • SE = spin echo
Author contributions: Guarantors of integrity of entire study, I.P., D.L., N.M.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; approval of final version of submitted manuscript, all authors; literature research, I.P., D.L., A.D.E.; clinical studies, I.P., D.L., A.D.E., B.S., N.M.R.; statistical analysis, I.P., D.L., L.N.; and manuscript editing, I.P., D.L., N.M.R.
Authors stated no financial relationship to disclose.
|
References |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  A OTO, R ERNST, L GHULMIYYAH, D HUGHES, G SAADE, and G CHALJUB The role of MR cholangiopancreatography in the evaluation of pregnant patients with acute pancreaticobiliary disease Br. J. Radiol., April 1, 2009; 82(976): 279 - 285. [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]
|
|
|
|
|
|
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, October 1, 2008; 249(1): 97 - 106. [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]
|
|
|
|
|
|
M. Baldisserotto, S. G. Valduga, and C. F. J. S. da Cunha MR Imaging Evaluation of the Normal Appendix in Children and Adolescents Radiology, October 1, 2008; 249(1): 278 - 284. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. A. Jaffe, C. M. Miller, and E. M. Merkle Practice Patterns in Imaging of the Pregnant Patient with Abdominal Pain: A Survey of Academic Centers Am. J. Roentgenol., November 1, 2007; 189(5): 1128 - 1134. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. J. Patel, D. L. Reede, D. S. Katz, R. Subramaniam, and J. K. Amorosa Imaging the Pregnant Patient for Nonobstetric Conditions: Algorithms and Radiation Dose Considerations RadioGraphics, November 1, 2007; 27(6): 1705 - 1722. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Lazarus, W. W. Mayo-Smith, M. B. Mainiero, and P. K. Spencer CT in the Evaluation of Nontraumatic Abdominal Pain in Pregnant Women Radiology, September 1, 2007; 244(3): 784 - 790. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Singh, R. Danrad, P. F. Hahn, M. A. Blake, P. R. Mueller, and R. A. Novelline MR Imaging of the Acute Abdomen and Pelvis: Acute Appendicitis and Beyond RadioGraphics, September 1, 2007; 27(5): 1419 - 1431. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
I. Pedrosa, E. A. Zeikus, D. Levine, and N. M. Rofsky MR Imaging of Acute Right Lower Quadrant Pain in Pregnant and Nonpregnant Patients RadioGraphics, May 1, 2007; 27(3): 721 - 743. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. S. Katz, V. Merunka, J. J. Hines, E. M. Meiner, I. Pedrosa, D. Levine, and N. M. Rofsky Invited Commentary * Authors' Response RadioGraphics, May 1, 2007; 27(3): 743 - 753. [Full Text] [PDF]
|
|
|
|
 |
|
 E M Anderson and H K Bungay Imaging investigation of acute right iliac fossa pain Imaging, December 1, 2006; 18(4): 257 - 267. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
