| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Research |
1 All authors: Department of Radiology, St. Luke's-Roosevelt Hospital Center, 1000 Tenth Ave., Rm. 4C-12, New York, NY 10019.
Received September 21, 2007;
accepted after revision November 5, 2007.
Address correspondence to C. W. Koo
(ck2182{at}columbia.edu).
Abstract
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
SUBJECTS AND METHODS. Two hundred fifteen consecutive outpatients were randomly assigned to receive either whole milk (n = 115) or VoLumen (n = 100). Results were independently reviewed by two radiologists who were blinded to the oral contrast agent used. Degree of bowel distention was qualitatively scored on a 4-point scale, and bowel wall visibility was graded qualitatively on a yes-or-no basis. A questionnaire regarding oral contrast tolerability was provided to each patient. Cost comparison of the two agents was performed.
RESULTS. No statistically significant differences were seen between whole milk and VoLumen with respect to degree of bowel distention and mural visualization for all segments of bowel studied (p > 0.05 for both reviewers). Significantly more patients ranked milk as pleasant in taste compared with VoLumen (p < 0.0001). More patients preferred milk compared with VoLumen (p < 0.0001). Milk was better tolerated than VoLumen, with fewer abdominal side effects, including abdominal discomfort (p = 0.019), cramping (p = 0.019), nausea (p = 0.016), and diarrhea (p = 0.0002). The cost per patient for VoLumen is $18 compared with $1.48 for milk.
CONCLUSION. Whole milk is comparable to VoLumen with respect to bowel distention and bowel wall visualization and has a lower cost, better patient acceptance, and fewer adverse symptoms. Milk is a cost-effective alternative to VoLumen as a low-attenuation oral contrast agent.
Keywords: cost-effectiveness • low-attenuation oral contrast material • milk • patient tolerance • VoLumen
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Multiple low-attenuation oral contrast agents have been studied [5–11]. Although air had been advocated as an alternative to high-attenuation oral contrast material for gastric and large-bowel imaging [12], the production of artifacts and the requirement of wide window settings for viewing render it a suboptimal agent for soft-tissue contrast resolution [13]. Taking advantage of the inherent contrast between water attenuation and contrast material–enhanced bowel wall, Winter et al. [6] used water as an oral contrast agent, resulting in excellent 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. Previous investigations of fat-containing oral contrast agents such as one comprising 12.5% corn oil showed excellent gastrointestinal tract discrimination and mural visualization without a significant difference in patient tolerance as compared with barium suspension and iodine solutions [7]. However, such agents either are not commercially available or are fat-containing food supplements that are expensive. In addition, metoclopramide hydrochloride was required to promote gastrointestinal peristalsis inhibited by the high fat content. Moreover, such high fat content is often unacceptable to patients. Although polyethylene glycol (PEG) provides better small-bowel distention and reaches the colon more readily than a full-strength fiber mixture and water [14], the high cost per examination and patients' dissatisfaction with this unpalatable mixture and its abdominal side effects such as diarrhea prevented the widespread adoption of PEG preparations.
Whole milk (4%) has been shown to be cost-effective [5]. The fat contained in whole milk effectively slows upper gastrointestinal tract motility, obviating the use of either smooth muscle relaxants or metoclopramide hydrochloride. Whole milk provides superior mural visualization and discrimination of the pancreas and duodenum compared with barium-based contrast material and water [5]. In addition, gastrointestinal tract distention and small-bowel discrimination were comparable to those in studies using a barium suspension and superior to imaging performed with water [5]. More recently, a newly developed low-attenuation 0.1% barium suspension (VoLumen, E-Z-EM Inc.) has been shown to provide excellent gastrointestinal tract distention and superb visualization of mural features compared with barium suspension and a methylcellulose–water mixture [12].
To date, no studies to our knowledge have compared VoLumen and milk in terms of cost, effectiveness, and patient tolerance. Our prospective, single-blinded study seeks to examine these three issues. We hypothesize that milk will be as well tolerated as VoLumen while being more cost-effective and providing similar degrees of gastrointestinal tract distention and mural visualization.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
From July 2005 to April 2007, 215 consecutive outpatients referred for oral and IV contrast-enhanced abdominopelvic CT were randomly assigned to receive either whole milk (4%) or VoLumen. No flavoring was added to the whole milk. Exclusion criteria included history of lactose intolerance, documented allergy to iodinated contrast material, serum creatinine greater than 1.5 mg/dL, a history of gastric surgery, and individuals undergoing CT primarily for evalu ation of liver disorders because the pelvis and thus the distal small bowel are not routinely imaged in these patients.
One hundred adult patients received 1,350 mL of VoLumen divided into four doses (450 mL 45 and 30 minutes before scanning, 300 mL 15 minutes and 150 mL immediately before scanning), and 115 patients received 600–1,000 mL of whole milk (4%) divided into two doses (400–600 mL 1 hour and 200–400 mL 20 minutes before CT acquisition). These protocols were devised on the basis of previously published protocols [5, 11].
Nine CT examinations in the VoLumen group and 16 examinations in the milk group were performed using a single-slice helical detector unit (Picker, Picker International) with a pitch of 1.75 mm and a 5 x 5 mm detector configuration, creating 5-mm sections. The remainder of the CT examinations in each study group were performed using a 4-MDCT scanner (Aquilion, Toshiba America Medical Systems) with a pitch of 4–5 mm and a 3 x 3 mm detector configuration, creating 3-mm sections. All examinations were performed using IV contrast material (Ultravist 300 [iopromide], Bayer HealthCare [formerly Berlex Laboratories]) that was administered by means of a power injector (Envision CT, MedRad) at rates of 2–3 mL/s. A dual acquisition protocol was used. Phase 1 (portal venous phase) CT images were obtained from the xiphoid to the pubic symphysis 75 seconds after the initiation of the IV contrast bolus. Phase 2 (excretory phase) images were obtained 3–4 minutes after phase 1 scanning was completed from the superior aspect of the liver to the pubic symphysis. This phase was obtained for evaluation of the excretory system. Neither IV glucagon nor metoclopramide hydrochloride was administered.
The studies were qualitatively evaluated indepen dently by two board-certified radiologists for luminal distention of the duodenum, jejunum, and ileum, which was graded on a scale of 0–3, with 0 being no distention, 1 being minimal (1 cm) distention, 2 being good partial (1–2 cm) distention, and 3 being excellent (> 2 cm) dis tention. Evaluation of the antrum was based on a similar qualitative grading scale of 0–3 without specific measurements. Both reviewers had 23 years of experience in interpreting CT of the abdomen and pelvis. Nonuniform duodenal dis tention was frequently noted, and thus the duo denum was radiographically divided into two portions for analysis, with duodenum 1 including the first two portions of the duodenum and duodenum 2 comprising the third and fourth portions of the duodenum. Nonuniform jejunal and ileal distention were less frequently en countered. In cases in which nonuniform jejunal and ileal distention were noted, segmental evalu ation was performed and the best score was recorded. Visualization of bowel wall enhancement of the antrum, duodenum, jejunum, and ileum was evaluated on a yes-or-no basis. Both radiologists were blinded to the type of oral contrast material administered.
A questionnaire regarding oral contrast tolerability was provided to each patient after completion of CT. The questionnaire asked each patient to rate the taste of the oral contrast material on a scale of 1–3, with 1 being intolerable, 2 being unpleasant but tolerable, and 3 being pleasant. In addition, the questionnaire asked whether the patient had difficulty ingesting the entire volume of contrast material and whether the patient disliked drinking whole milk or oral contrast material. Side effects, including ab dominal dis comfort or cramping, nausea, vomiting, flatulence, and diarrhea, were recorded.
A cost comparison of the two agents was also performed.
Statistical Analysis
Mean age differences between the milk and VoLumen groups were evaluated
using the Student's t test. Sex distribution was calculated and the
differences were evaluated with the chi-square test. Degree of bowel
distention in the two study groups using the aforementioned 4-point scale was
assessed by each reviewer, and the percentage of cases receiving each score
was calculated. The Jonckheere-Terpstra trend test was used to evaluate for
statistically significant differences. Percentages of cases with adequate
bowel wall visualization in each study group were calculated for each
reviewer, and the chi-square test was used to examine the differences between
the two groups. The Fisher's exact test was performed when the results from
the chi-square test may not be valid. For interobserver agreement, the
weighted kappa coefficient was calculated for distention and the simple kappa
coefficient was determined for visualization. Contrast agent palatability was
rated by each patient in the two study groups using the aforementioned 3-point
scale, and the percentage of cases receiving each score was calculated. The
Jonckheere-Terpstra trend test was used to evaluate for statistically
significant differences. Oral contrast preference, the magnitude of adverse
gastrointestinal symptoms, and the percentage of patients who experienced
difficulty ingesting the requisite volume of oral contrast material were
calculated, and the chi-square test was used to examine the differences
between the two groups. The Fisher's exact test was performed when the results
from the chisquare test were not valid. A p value less than or equal
to 0.05 was considered significant for all of these statistical analyses.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Qualitative Analysis
No statistically significant differences were seen between whole milk and
VoLumen with respect to the degree of bowel distention among all segments of
the bowel studied (Table 1,
Figs. 1A,
1B,
1C,
2A,
2B,
2C,
3,
4). For example, the absolute
percentage differences of cases scoring 0, 1, 2, and 3 for antral distention
as interpreted by reviewer 1 were 1%, 2.3%, 3.5%, and 6.7%, respectively, and
were determined to be statistically insignificant (p = 0.198).
|
|
|
|
|
|
|
|
|
Both whole milk and VoLumen showed excellent antral distention, with 81.7% of milk and 75% of VoLumen cases receiving the maximal score of 3 by reviewer 1, and 79.1% of milk and 75% of VoLumen studies receiving a score of 3 by reviewer 2. On the other hand, the two oral contrast materials showed only minimal to good duodenal distention, with the first two segments of the duodenum (duodenum 1) being better distended than the latter two segments of the duodenum (duodenum 2). For example, reviewer 1 gave a distention score of 2 or greater to 46.1% of duodenum 1 cases in the milk group and to 53% of cases in the VoLumen group versus a distention score of 2 or greater to only 35.6% of duodenum 2 cases in the milk group and to 34% of cases in the VoLumen group. Reviewer 2 rated the duodenum similarly, showing better distention in duodenum 1 than in duodenum 2. Distention seems to improve as one progresses to the mid and distal parts of the small bowel, with ileal distention slightly better than jejunal distention, as shown by a higher percentage of cases receiving a score of 2 or higher for ileal distention. Although not statistically significant, this suggests a trend toward improved distal small-bowel distention. Specifically, reviewer 1 gave a score of 2 or higher to 74.8% of cases in the milk group and to 73% in the VoLumen group for ileal distention, while giving the same scores to only 67.8% of cases in the milk group and to 76% of cases in the VoLumen group for jejunal distention. Similar to the distention results, no statistically significant differences were seen between the two groups for visualization of bowel wall enhancement (Table 2). Moreover, both groups showed excellent visualization of the antrum, jejunum, and ileum, while showing only good visualization of the duodenum, with minimally better visualization of the first two segments than of the distal two segments.
|
Interobserver Agreement
Interobserver agreement was excellent for all segments of the bowel, with a
coefficient of 1.0 being perfect. Perfect agreement was achieved for
visualization of ileal wall enhancement for the milk group (simple kappa
coefficient, 1.000). Minimally less agreement occurred between the two
reviewers for bowel distention in the milk group compared with the VoLumen
group, with the weighted kappa coefficient for each bowel segment ranging from
0.893 to 0.990 for the milk group and from 0.955 to 0.985 for the VoLumen
group. Contrarily, greater agreement occurred between the two reviewers for
bowel visualization in the milk group compared with the VoLumen group (simple
kappa coefficient, 0.756–1.0 for the milk group and 0–0.954 for
the VoLumen group). The simple kappa coefficient for visualization of the
antral wall cannot be calculated because none of the reviewers rated any case
as "no visualization," but the degree of disagreement was small.
Specifically, disagreement occurred in only three of 115 cases in the milk
group and in no cases in the VoLumen group. Slightly less agreement was also
seen for visualization of bowel wall enhancement in the milk group compared
with the VoLumen group (simple kappa coefficient, 0.756–1.000 for the
milk group and 0.864–0.954 for the VoLumen group). The p values
for all significance levels were less than 0.0001 and represent two-sided
significance levels.
Patient Tolerability and Contrast Preference
One hundred twelve of the 115 patients from the milk group and 83 of the
100 patients from the VoLumen group completed the tolerability questionnaire.
Patients seemed to tolerate milk much better than VoLumen
(Table 3). Specifically, more
patients thought milk had a more pleasant taste than VoLumen, with 77.7% of
patients giving milk a score of 3 versus only 43.4% of patients in the VoLumen
group. A significantly smaller percentage of patients disliked ingesting milk
compared with VoLumen (3.6% vs 66.3%, p < 0.0001). Moreover,
almost twice as many patients had difficulty consuming the entire volume of
VoLumen compared with milk (9.6% and 5.3% for the VoLumen and milk groups,
respectively). With the exception of flatulence, milk also caused few er
gastrointestinal symptoms than VoLumen
(Table 4).
|
|
Cost Comparison
The cost for milk is significantly less than the cost for VoLumen per
study. Specifically, milk costs $1.48 per study versus $18 per study for
VoLumen.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
We found no statistically significant differences between whole milk and VoLumen with respect to degree of bowel distention and mural visualization among all segments of bowel studied. However, whole milk had better patient acceptance, fewer abdominal symptoms, and lower cost than VoLumen. With rising medical costs and an increasing volume of CT examinations performed for the evaluation of gastrointestinal abnormalities, an oral contrast agent that enables effective gastrointestinal tract evaluation while minimizing cost is ideal, and whole milk seems to possess these characteristics. Moreover, patient satisfaction and willingness to ingest milk may lead to higher patient compliance in undergoing CT examinations, which may lead to earlier detection and treatment of diseases.
Both whole milk and VoLumen showed excellent antral distention compared with the small bowel. This phenomenon is probably because a portion of the oral contrast material was consumed immediately before scanning and it did not have sufficient time to exit the stomach. This situation is ideal for and is often used in pancreatic and biliary studies [6, 20]. The slightly decreased distention in the distal duodenum is likely secondary to peristalsis propelling the contrast material that was administrated 45–60 minutes before scanning into the distal small bowel, whereas the contrast material administrated immediately before scanning had not yet reached the distal duodenum. Distention of the duodenum, especially the third portion, is further compromised by its anatomic situation between solid organs and vascular structures. Similar to the distention results, visualization of the antral, jejunal, and ileal walls was slightly better than visualization of the duodenal wall. The visualization results paralleling the distention data support the general observation that better bowel distention imparts better bowel wall visualization.
IV glucagon was not used in either the milk or the VoLumen group, yet good to excellent gastrointestinal tract distention was achieved. Such a phenomenon is expected for milk because the 4% fat contained in milk effectively slows peristalsis, resulting in superb distention. However, adequate bowel distention may not have been expected for VoLumen without the administration of IV glucagon. Although the specific mechanism is unknown, the sorbitol component of the 0.1% barium mixture may be sufficiently effective to obviate using glucagon. In addition, the continuous consumption of the oral contrast material over 45 minutes may have contributed to the relatively uniform bowel distention without the need for glucagon. The elimination of pharmacologic effects of IV glucagon, such as inhibition of gastric emptying, is beneficial to patients in many respects. In the study performed by Winter et al. [6], the combination of water and glucagon was associated with nausea and emesis in 14% and 7% of patients, respectively [6]. Although it is uncertain whether these symptoms were due to the use of glucagon, the ionic IV contrast material, or a combination of the two, such adverse reactions increase patient discomfort and the risk of aspiration. In addition, the ability to forgo IV glucagon administration will reduce costs by up to $35 per study.
Our study and the contrast agents examined have several limitations. Only qualitative evaluation of bowel distention was performed because quantitative analysis performed by measuring the widest wall-to-wall diameter is, at best, a surrogate measure of the overall distention in a segment of the gastrointestinal tract, as previously shown [11]. In addition, to our knowledge, no readily available method exists to date that allows volumetric assessment of the entire segment of interest. Moreover, a previous study has shown that quantitative measures parallel those of qualitative assessment [11]. Optimal visualization is afforded by the concomitant administration of IV contrast material for both low-attenuation oral contrast agents used in this study because visualization of the contrast-enhanced bowel wall is necessary for recognition of a bowel loop. Therefore, cases performed without IV contrast material were excluded from our study, so our data may not be applicable to CT studies performed without IV contrast material. In addition, the need for continuous ingestion and timely access to a CT scanner once the contrast agent has been consumed requires close supervision by radiology department personnel for the entire duration of contrast intake. Such a practice may not be feasible in emergency departments, where there may be no available supervising personnel, thus potentially limiting the use of such contrast materials to an outpatient population. Lactose intolerance limited our study population to those who can tolerate lactose-containing milk. However, with the advent of lactose-free milk products, such a limitation may be overcome by administering lactose-free milk, but such a practice will require further investigation. Finally, our study did not compare whole milk or VoLumen with traditional positive oral contrast material because previous studies have already shown that both whole milk and VoLumen compare favorably with traditional positive oral contrast materials in gastrointestinal tract distention and mural visualization [5, 11].
Despite these drawbacks, the results of our study show that whole milk is comparable for bowel distention and mural visualization to VoLumen. In the setting of better patient acceptance, fewer gastrointestinal symptoms after consumption, and lower cost, whole milk proves to be a cost-effective alternative to VoLumen as a low-attenuation oral contrast agent. The ability of this inexpensive medium to produce adequate bowel wall visualization provides considerable potential for further cost-effective applications of volume imaging with MDCT.
Acknowledgments
We thank the St. Luke's radiology technologists and residents for data
acquisition and Theresa E. Perlis and Marvin Friedman for statistical
consultation, guidance, and data analysis.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
