| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Technical Developments |
1 From the Department of Radiology, Beth Israel Deaconess Medical Center and Harvard Medical School, 330 Brookline Ave, Boston, MA 02215. From the 1997 RSNA scientific assembly. Received June 5, 1998; revision requested July 30; revision received October 8; accepted November 19. Address reprint requests to V.R.
 |
Abstract |
|---|
 TOP Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Index terms: Barium, 70.12112, 70.12115, 70.12116, 70.12119 • Computed tomography (CT), contrast enhancement, 70.12112, 70.12115, 70.12116, 70.12119 • Contrast media, comparative studies, 70.12112, 70.12115, 70.12116, 70.12119 • Gastrointestinal tract, CT, 70.12112, 70.12115, 70.12116, 70.12119 • Milk, 70.12119
|
Introduction |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In modern image processing, segmentation of bone and high-attenuation intestinal contrast agents is one of the most time-consuming aspects of various rendering methods, which often discourages operators. Thus, in many centers, oral contrast materials are not used for abdominal CT angiography. However, this has two drawbacks. Collapsed bowel may be confused for soft-tissue masses. We found this particularly bothersome when we integrated CT and CT angiography to stage pancreatic, renal, liver, or other tumors. Even for purely angiographic indications, difficulties in bowel discrimination produce problems in interpretation of the nonangiographic portion of the cross-sectional images. In addition, use of oral contrast agents helps keep patients well hydrated, which is important when large doses of intravenous contrast material are used.
Owing to the inherent contrast between water attenuation adjacent to contrast material–enhanced bowel wall, Winter et al (16) used water as a low-attenuation oral contrast agent for abdominal helical CT with exquisite luminal and mucosal depiction of the upper gastrointestinal tract. However, the use of water requires the additional use of an upper gastrointestinal smooth muscle relaxant and does not provide adequate evaluation of the distal small bowel. In previous investigations of a negative-attenuation oral contrast agent comprising 12.5% corn oil emulsion in comparison with high-attenuation agents such as 2% iodine solution or barium suspension, results demonstrated significant improvement in gastrointestinal tract discrimination and mural visualization without a significant difference in patient tolerance (5,11,17–19). However, such agents are either not commercially available or require in-house preparation or are fat-containing food supplements that are expensive. In addition, the high fat content may be unacceptable to patients.
Agents that contain 12.5% fat were found to be adequate oral contrast media with or without the use of intravenous contrast material, but emulsions with a lower concentration of fat were as effective when used with intravenous contrast agents (19). For intravenous contrast-enhanced CT of the abdomen, we hypothesized that milk containing 4% fat used as a low-attenuation oral contrast agent would provide results comparable to those achieved with corn oil emulsion but without the undesirable high fat content. Furthermore, the fat content in milk would be sufficient to slow gastrointestinal tract peristalsis. This would allow distention and mural discrimination of the upper gastrointestinal tract that were superior to those with water, without the need for additional smooth muscle relaxants. Although 2% milk (2% fat) would be preferable to whole milk (4%) because of lower fat and caloric content, we doubted that it would be substantially more useful than water (19). The purpose of this study was to test the utility of milk as an oral contrast agent for intravenous contrast-enhanced abdominal CT by comparing results with 4% or 2% milk to those with water, no oral contrast agent, and conventional barium suspension.
|
Materials and Methods |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
A volume of 400–600 mL of oral contrast agent was ingested from 1 hour to 40 minutes before imaging and an additional volume of 200–400 mL was ingested 20 minutes before. The amount ingested was dependent on patient tolerance, whereas the type of contrast agent was predetermined and changed on a weekly basis. Thus, it was predetermined that in the first week of the study all patients included in the study would receive 4% milk; in the second week, 2% milk; in the third week, water; and in the fourth week, no contrast agent. This weekly sequence was repeated three times until enough cases were collected for statistical analysis and the study ended.
In addition, 12 consecutive patients (seven men and five women; age range, 30–68 years; mean age, 56 years) who received 800 mL to 1 L of 2% barium suspension in the second week of the study were included for comparison (5). We did not test iodine-containing oral contrast agents because we do not use them routinely in our clinical practice, and we did not think results with them would be noticeably different from those with the barium-based products. All patients underwent scanning with a high-speed helical CT scanner (Advantage; GE Medical Systems, Milwaukee, Wis) and received a bolus of 100–150 mL of nonionic intravenous contrast material (ioversol injection 68%, Optiray 320; Mallinckrodt, St Louis, Mo) delivered with a power injector at a rate of 2–3 mL/sec. Only one patient experienced nausea and vomiting during the injection.
We had previously obtained institutional approval for use of no contrast agent or low-attenuation oral contrast agents, including water and emulsions containing up to 12.5% fat for abdominal CT angiography, abdominal integrated CT and CT angiography, and CT urography. By the time the present study was performed, these procedures were already accepted as routine clinical practice (12–15), and we had stopped obtaining informed consent for use of low-attenuation oral contrast agents in abdominal scanning.
The studies were reviewed by two board-certified radiologists (S.E.T., V.R.) independently. They were not aware whether or not oral contrast material was given or of the type of low-attenuation agent used. They graded gastrointestinal tract distention and mural visualization on CT scans obtained in the stomach, duodenum, jejunum, and ileum. They used a five-point scale from 1 to 3 with half points: 1, none; 1.5, poor; 2, partial; 2.5, good; and 3, full distention and mural visualization. Similarly, discrimination at four anatomic regions of the small- bowel loops from the jejunum and ileum and of the pancreas from the duodenum were also ranked on a five-point scale from 1 to 3 with half points. Discrimination of the bowel was defined on the basis of the reviewer's ability to identify bowel loops as such, as opposed to cysts, lymph nodes, soft-tissue nodules, or tumors. Discrimination of the pancreas and duodenum was defined on the basis of the reviewer's confidence (11). The four anatomic regions included the following: region 1, head of the pancreas from the gastric antrum and duodenal bulb; region 2, head of the pancreas from the descending, second portion, of the duodenum; region 3, head and uncinate process of the pancreas from the junction of the descending, second portion, and the transverse, third portion, of the duodenum; and region 4, the uncinate process of the head of the pancreas from the transverse, third portion, of the duodenum. The grades for each reviewer were averaged in each case, and the mean of the grades for the two reviewers in the four regions was used as a single score.
Statistical analysis of the differences in median scores for distention, mural visualization, and discrimination was performed by means of the Kruskal-Wallis test. A P value less than or equal to .05 was considered statistically significant.
|
Results |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
|
|
|
|
|
|
|
Gastrointestinal Tract Distention
For gastrointestinal tract distention by location, grades were significantly superior for 4% milk compared with all other contrast agents for gastric and duodenal distention (median grades, 2.75 and 2.50, respectively) and compared with no contrast agent, water, and 2% milk for jejunal and ileal distention (median grades, 2.25 and 2.50, respectively). Grades with 4% milk and barium suspension, however, were not significantly different for jejunal distention, but grades with barium were superior to those with 4% milk for ileal distention (median grades, 2.65 and 2.50, respectively). Grades with 2% milk and barium suspension were not statistically different for gastric distention, but grades with 2% milk were superior to those with barium for duodenal distention and inferior to those with barium for jejunal and ileal distention. In addition, grades with 2% milk were superior to those with water and no contrast agent for distention of the stomach, duodenum, jejunum, and ileum. Grades with water were significantly superior to those with no contrast agent for gastric distention, but they were not superior for distention of the duodenum, jejunum, and ileum.
Gastrointestinal Tract Mural Visualization
Grades for visualization of the wall of the stomach, duodenum, jejunum, and ileum were significantly superior with both 4% and 2% milk compared with no contrast material, water, and barium suspension. Grades with 4% milk were significantly better than those with 2% milk for visualization of the wall of the stomach, duodenum, jejunum (median grades, 2.75, 2.50, and 2.25, respectively) but not the ileum. Grades with water and barium suspension were not significantly different for gastric wall visualization, but they were both superior to grades with no contrast material. Grades with no contrast material, water, and barium were not significantly different for mural visualization in the duodenum and jejunum, but grades with barium were inferior to those with the other two contrast agents for mural visualization in the ileum.
Gastrointestinal Tract Discrimination
In pancreas-duodenum discrimination, grades with 4% milk were significantly superior to those with all other oral contrast agents in all four regions (median scores, 3.00, 2.75, 2.63, and 2.75 for regions 1, 2, 3, and 4, respectively). Grades with 2% milk were significantly superior to those with barium suspension in regions 2 and 3 but not in regions 1 and 4. Overall grades for pancreas-duodenum discrimination, however, were not clearly superior with 2% milk, barium suspension, water, or no contrast material.
Grades for discrimination of small- bowel loops from the jejunum and ileum were not significantly different with 4% milk or barium suspension (median score for each, 2.50) Grades for discrimination of loops from the jejunum were significantly superior for water compared with no contrast agent and for 2% milk compared with water, and grades with 4% milk or barium were superior to those with 2% milk. Grades for ileal discrimination were not significantly different with no contrast agent or water, and with 2% milk, 4% milk, or barium suspension, but grades with each of the latter were significantly superior to those with no contrast agent or water.
Interobserver Agreement
In 110 studies, 1,493 observations were made by the two reviewers. These observations agreed within one grade in 1,284 (86%) instances. The 
value for exact agreement was 0.3, which indicates fair to moderate agreement (P < .001). The weighted value for agreement within one grade was 0.7, which indicates a substantial to almost perfect agreement between the two observers (P < .001).
|
Discussion |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The increasing popularity of abdominal CT angiography and other image processing techniques with maximum intensity projection or shaded surface display has renewed a need for a efficacious low-attenuation oral contrast agent, because traditional high-attenuation contrast agents interfere with image processing techniques. Corn oil emulsion has been found to work well as a low-attenuation contrast agent (5,11,17–19), but the high fat content and commercial unavailability make it impractical for day-to-day use. It is estimated that 500 mL of 12.5% corn oil contains 500 calories and 58 g of fat. In, addition, metoclopramide hydrochloride must be used, which increases the expense and is not totally without risk such as extrapyramidal symptoms, which usually manifest as acute dystonic reactions. Medical oil–based food supplements, such as Lipomul (Upjohn, Kalamazoo, Mich) could be used instead of corn oil, but a high cost of more than $25 per study and high caloric intake render these agents impractical.
Recent work has demonstrated water to be a feasible alternative (16). Without the use of smooth muscle relaxants, however, rapid passage through the upper gastrointestinal tract results in inferior distention of the duodenum with resultant difficulties in discrimination from the pancreatic head. The use of intravenous glucagon may increase costs by an additional $35 per study, and it may hinder gastric emptying. In a recent study by Winter et al (16), water with glucagon was associated with nausea and emesis in 14% and 7% of their patients, respectively. It is not clear whether this was due to the use of glucagon, the ionic intravenous contrast material, or a combination of the two. Whatever the cause, these adverse reactions increase patient discomfort and the risk of aspiration. In our study, we attribute the low frequency of nausea and vomiting (one patient) to the exclusive use of nonionic intravenous contrast material.
Whole milk contains 4% fat. This effectively slows passage through the upper gastrointestinal tract, which results in superior distention and mural discrimination of the stomach and duodenum compared to those with 2% milk, barium suspension, water, or no contrast agent. The decreased peristalsis results in prolonged distention of the duodenum, which is easily discriminated from the pancreatic head. With 4% milk as opposed to water as a low-attenuation oral contrast agent, there is no need to administer a smooth muscle relaxant, such as glucagon, to achieve adequate distention of the upper gastrointestinal tract, which eliminates potential undesirable side effects and additional expense. On the other hand, the decreased peristalsis was not as pronounced as that with corn oil emulsion, and metoclopramide hydrochloride was not needed to increase gastrointestinal tract peristalsis. Similarly, we did not observe gallbladder stimulation (Fig 2a).
Of the low-attenuation oral contrast agents we tested, grades with 4% milk were also superior for distention and mural discrimination in the jejunum and ileum. The loops were easily distinguished as such on the basis of their low-attenuation contents and visualization of the plicae circulares, which obviates a pseudotumor. However, use of milk as an abdominal oral contrast agent necessitates use of an intravenous contrast agent. With high-attenuation agents, bowel loop discrimination is based on high attenuation and the anatomic relationship of the loops to the mesentery. With corn oil or milk, the additional bowel loop architecture provided by the plicae circulares helps in jejunum versus bowel loop discrimination. In the ileum, however, this discrimination may be more difficult because of the relative lack of internal architecture of the loops. These difficulties may arise in distinguishing ileal small bowel from low-attenuation lymph nodes; small cysts around the pancreas, ovaries, and, rarely, kidneys; or small periappendiceal or diverticular abscesses. The lack of absolute discrimination based on attenuation alone is compensated for by the depiction of fine anatomic relationships and structural architecture with modern intravenous contrast-enhanced helical CT. Regardless, we withhold milk in cases of suspected abdominal abscess, pancreatitis, or appendicitis.
There is some evidence that a barium- or iodine-based oral contrast medium may increase the local absorbed dose of the intestinal mucosa by as much as five times as a result of photoelectric interactions at the contrast agent–mucosa interface (20,21). This may be important in modern multiphase scanning, and milk may be used as an alternative contrast agent.
Compared to results with whole milk, results with 2% milk were significantly inferior for gastrointestinal tract distention or pancreas-duodenum discrimination. In addition, results with barium suspension were superior to those with 2% milk for bowel loop discrimination and were not significantly different for distention or pancreas-duodenum discrimination. Whole milk should be the oral contrast agent preferred for techniques in which use of high-attenuation contrast agent is undesirable, such as CT angiography. CT angiography is often performed to evaluate pancreatic disease, and the superior duodenal discrimination afforded by whole milk would facilitate evaluation of pancreatic head masses. Similarly in CT urography, in addition to multiplanar reconstructions, a digital image of the abdomen (scout view) obtained after contrast agent injection provides a urographic view of the collecting systems that is unobscured by superimposed high-attenuation bowel contents.
In addition to the use of whole milk for CT angiography, it should be considered an alternative to high-attenuation contrast agent in cases of disease suspected in the upper abdomen, including the pancreas. The superior distention and mural discrimination of the stomach and duodenum should encourage its use as the preferred contrast agent in specific evaluations of these structures. The routine use of milk as an oral contrast agent for abdominal CT would also represent a cost savings: The retail cost for the 2% barium solution currently used in our department is approximately $7.20 per examination compared with less than $1.00 for an equal amount of milk. Compared with 12.5% corn oil emulsion, milk affords both monetary savings and a decreased dietary load. A dose of 500 mL of whole milk contains 300 calories, 8 g of fat, and 35 mg of cholesterol.
Presently in our practice, we use whole milk as an oral contrast agent in evaluation of the urinary tract; in cases of suspected pancreatic or other upper abdominal tumors; and in special examinations including multiphasic abdominal CT, especially of the liver, abdominal CT angiography, CT cholangiopancreatography, and CT urography. For routine abdominal or pelvic survey and suspected inflammatory conditions, we use high-attenuation contrast agents.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
3 Current address: Department of Radiology, University of Crete, Heraklion, Greece.
Author contributions: Guarantors of integrity of entire study, S.E.T., V.R.; study concepts, S.E.T., V.R.; study design, S.E.T., V.R.; definition of intellectual content, V.R.; literature research, S.E.T., V.R., P.P; clinical studies, all authors; data acquisition, all authors; data analysis, S.E.T., V.R.; statistical analysis, V.R.; manuscript preparation, S.E.T., V.R.; manuscript editing, S.E.T., V.R.; manuscript review, S.E.T., V.R., R.L.S., M.M.J.M.
|
References |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  C. W. Koo, L. R. Shah-Patel, J. W. Baer, and D. H. Frager Cost-Effectiveness and Patient Tolerance of Low-Attenuation Oral Contrast Material: Milk Versus VoLumen Am. J. Roentgenol., May 1, 2008; 190(5): 1307 - 1313. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Siewert, J. Sosna, A. McNamara, V. Raptopoulos, and J. B. Kruskal Quality Initiatives: Missed Lesions at Abdominal Oncologic CT: Lessons Learned from Quality Assurance RadioGraphics, May 1, 2008; 28(3): 623 - 638. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. D. T. Maglinte and J. E. Huprich Invited Commentary * Author's Response. RadioGraphics, May 1, 2006; 26(3): 657 - 662. [Full Text] [PDF]
|
|
|
|
|
|
N. Erbay, V. Raptopoulos, E. A. Pomfret, I. R. Kamel, and J. B. Kruskal Living Donor Liver Transplantation in Adults: Vascular Variants Important in Surgical Planning for Donors and Recipients Am. J. Roentgenol., July 1, 2003; 181(1): 109 - 114. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. M. Horton and E. K. Fishman Current Role of CT in Imaging of the Stomach RadioGraphics, January 1, 2003; 23(1): 75 - 87. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. V. Jayaraman, W. W. Mayo-Smith, J. S. Movson, D. E. Dupuy, and M. T. Wallach CT of the Duodenum: An Overlooked Segment Gets Its Due RadioGraphics, October 1, 2001; 21(90001): S147 - 160. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 I. R. Kamel, S. N. Goldberg, M. T. Keogan, M. P. Rosen, and V. Raptopoulos Right Lower Quadrant Pain and Suspected Appendicitis: Nonfocused Appendiceal CT-Review of 100 Cases Radiology, October 1, 2000; 217(1): 159 - 163. [Abstract] [Full Text]
|
|
|
|
|
|
I. R. Kamel, V. Raptopoulos, E. A. Pomfret, J. B. Kruskal, R. A. Kane, C.-S. Yam, and R. L. Jenkins Living Adult Right Lobe Liver Transplantation: Imaging Before Surgery with Multidetector Multiphase CT Am. J. Roentgenol., October 1, 2000; 175(4): 1141 - 1143. [Full Text] [PDF]
|
|
|
|
|
|
R. Pochaczevsky and V. Raptopoulos Carbon Dioxide as a Low-Attenuation Oral Contrast Agent Dr Raptopoulos responds: Radiology, March 1, 2000; 214(3): 918 - 918. [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
