Abdominal Imaging

, 34:642

CT colonography: same-day tagging regimen with iodixanol and reduced cathartic preparation

Authors

    • Diagnostic and Interventional RadiologyUniversity of Pisa
  • Francesca Turini
    • Diagnostic and Interventional RadiologyUniversity of Pisa
  • Francesca Cerri
    • Diagnostic and Interventional RadiologyUniversity of Pisa
  • Paola Vagli
    • Diagnostic and Interventional RadiologyUniversity of Pisa
  • Carlo Bartolozzi
    • Diagnostic and Interventional RadiologyUniversity of Pisa
Article

DOI: 10.1007/s00261-008-9453-z

Cite this article as:
Neri, E., Turini, F., Cerri, F. et al. Abdom Imaging (2009) 34: 642. doi:10.1007/s00261-008-9453-z

Abstract

The purpose of the study was to test the tagging performance and patient’s acceptance of a reduced cathartic preparation, based on iodixanol and PEG, administered to patients 3 h before the exam. One hundred and thirty-two asymptomatic patients were enrolled. As colonic cleansing we used PEG macrogol 3350. For fluid tagging iodixanol was orally administered 3 h before the exam, in a total dose of 50 mL mixed with 34.8 g of PEG in 750 mL of water. Image’s review showed 446 segments (56.4%) clean of feces and 346 segments (43.6%) with feces. Untagged fluid was observed in 74/706 (10.5%) segments; inhomogeneous tagging in 129/706 (16%); the average density of fluid was 1054.74 UH in the cecum–ascending colon and 905.14 UH in the descending–sigmoid colon; the average difference of density between right and left colonic segments was 149 UH, and it was statistically significant (P = 0.016). No side effects related to the consumption of Movicol were reported. Very few side effects related to the tagging solution were reported: mild nausea in 7 (0.05%) patients, mild diarrhea in 10 (0.07%). An average rank of 9 points (SD ± 1) on a 10-point scale (10 = no discomfort, 0 = severe discomfort) resulted from the self-administered questionnaire, showing an excellent acceptance of the preparation. Same day fluid tagging with iodixanol provides an optimal fluid tagging, it is completely tolerated by the patient, and it can be performed under medical control.

Keywords

Ct colonographyTaggingBowel cleansingTechniqueColonComputed tomography

CT colonography (CTC), although regarded as a non-invasive examination, is burdened by the need to get a good colonic cleansing. Often a cathartic cleansing similar to that used for examining traditional colonoscopy is used. However, such a procedure is inconvenient for the patient, it presents possible side effects, and it can cause poor quality of life in the days before the examination.

In a study by Weitzman et al. [1] it is shown how poor compliance with the bowel preparation contributes to reduce the patient acceptance of the screening procedure.

For the above reasons, the focus of research has in recent years turned to the study of milder preparations that allow reaching a greater patient’s acceptance, which is fundamental to introduce the CTC as a screening test [2, 3]. A step forward was achieved with the introduction of fecal/fluid tagging, alternatively with the oral administration of barium or ionic iodinated contrast material (sodium amidotrizoate and meglumine amidotrizoate; Gastrografin), or a combination of both. This made it possible to distinguish wall organic lesions from fecal residues thus reducing the need of a complete and cathartic cleansing [4]. Fletcher et al. [5], despited were using a single row CT, observed that the benefit of tagging was in the increase in specificity (from 90% to 97%) in the detection of advance adenomas (diameter >1 cm). The study of Iannaccone et al. [6] showed that by adding Gastrografin to regular meals (180 mL × 6 meals), the sensitivity for the identification of colorectal polyps with size greater or equal to 8 mm reached 95.5%. Pickhardt et al. [79] obtained excellent results in sensitivity and specificity in a large screening single center trial with respect to trials not using oral tagging. The oral administration of iodinated contrast media outside of the hospital is not however recommended, because it can cause, although rarely, some side effects that can range from mild reactions such as diarrhea, nausea, or vomiting to more severe reactions. Even if orally administered or introduced into body cavities, a small amount of contrast is resorbed by the mucosa and can be found in the blood stream. Such resorption causes an increase of eosinophilia, preparing the patient to potential allergic reactions [1012]. Epidermolysis for skin contact has been also described [13].

As a matter of fact, the contrast medium more frequently used for fluid tagging is Gastrografin, that is an ionic agent and consequently hyperosmolar. Because of this characteristic, it draws liquids from the bowel into the lumen directly, and it quite frequently causes disturbances of water and electrolyte balance [14, 15]. For the above reasons some authors recommended the use of non-ionic contrast for the opacification of the intestine during CT exams [16]. Non-ionic iodinated agents are less hypertonic than their ionic counterparts, and have an improved safety profile when compared with the safety profile of ionic agents [17]. Non-ionic iodinated agents such as iopromide and iohexol have been already used in the study by Zalis et al. [18] with both successful technical results and a good acceptance by the patients. In their study, the contrast media were administered in the 2 days before virtual colonoscopy examination.

Recently the European Society of Gastrointestinal and Abdominal Radiology has promoted a survey among European experts in CTC [19]. The majority of respondents were in favor of a reduced bowel purgation and fecal tagging in colorectal cancer screening programs with CTC, and pointed out that the choice of oral tagging agent should be based on local experience with due regard given to any history of allergy.

In the prospect of being able to make CTC a less demanding test and possibly the first candidate in the screening of colorectal cancer, we aimed to test the tagging performance and patients’ acceptance of a non-cathartic preparation, associated with the use of a solution containing iodixanol, administered to patients before the exam in the hospital under physician’s control.

Materials and methods

Patients and indications

A total of 132 asymptomatic patients were enrolled in the study. The age range was 42–76, with a mean of 58, a standard deviation of ±6. Exclusion criteria included history of familial adenomatous polyposis or hereditary non-polyposis cancer syndromes, prior colorectal surgery, a suspected diagnosis of inflammatory bowel disease, bowel obstruction, acute diverticulitis, contraindications to the ingestion of iodine-containing contrast agents. Moreover, for the purpose of the study, on the basis of the C-RADS (CT colonography reporting and data system), C-RADS 0 cases were excluded [20].

Written informed consent was obtained from all patients, after the purpose and protocol of the study had been fully explained. These consent procedures were reviewed and approved by our institutional review board.

Preparation

Diet

In the 3 days before the examination, patients were asked to avoid products containing fibers and to restrict the diet to meat, eggs, fish, and milk. They were asked to assume a light supper as vegetable or meat broth the day before the examination. Fasting was required for the day of the examination.

Colonic cleansing

For colonic cleansing we used PEG macrogol 3350 (Movicol, NORGINE ITALIA Srl), which is a mild laxative usually used to treat chronic constipation. Specifically, patients were asked to assume 13.8 g of PEG diluted in a glass of water at each of the six principal meals beginning 48 h before CTC. The reason for using Movicol was to soften the feces to improve the efficacy of tagging.

Fluid tagging

For fluid tagging, an oral non-ionic iodinated contrast agent (Iodixanol; Visipaque, GE Healthcare) with an iodine concentration of 320 mg/mL was administered in a total dose of 50 mL 3 h before the exam. Specifically, patients were asked to drink 50 mL of the contrast agent diluted in 750 mL of water; and 34.8 g of PEG were added to the solution to speed transit time. The adjunction of PEG was motivated by a previous experience carried out with oral consumption of 50 mL of iodixanol alone. In such a case we observed that with PEG a more homogeneous fluid tagging could be obtained (Fig. 1).
https://static-content.springer.com/image/art%3A10.1007%2Fs00261-008-9453-z/MediaObjects/261_2008_9453_Fig1_HTML.jpg
Fig. 1

(A) CTC in supine decubitus obtained 3 h after the oral intake of 50 mL of iodixanol. Untagged fluid is well evident in the descending colon (D) and non-homogeneous tagging in the ascending colon (A). T, transverse colon. (B) CTC in supine decubitus obtained 3 h after the oral intake of 50 mL of iodixanol and 750 mL of a PEG solution. The homogeneity of fluid tagging is almost equal between ascending (A) and descending (D) colon. T, transverse colon. (C) In the same case the fluid appears homogeneously tagged in the sigmoid colon (S) and in the rectum (R).

CTC study technique

Image acquisition

All examinations were performed by means of a multirow CT scanner (LightSpeed Plus; GE/Ealthcare). Images were obtained by using 5 mm collimation and 1.25 mm row thickness, with pitch 6 (7.5 mm/s), and low dose protocol, with 120 kVp and 30–80 mAs. Data were reconstructed at 0.6 mm intervals.

In an attempt to minimize bowel peristalsis and reduce colonic spasm, a spasmolytic agent was administered intravenously before CTC. Hyoscine-N-butylbromide (20 mg, Buscopan; Boehringer Ingelheim, Ingelheim, Germany) was routinely prescribed; if contraindicated (e.g., for prostate hyperplasia, tachycardia, or glaucoma), no spasmolytic agent was used. With the patient in the left lateral decubitus position, the colon was insufflated with CO2 by means of a dedicated rectal enema tube with balloon. The tube was equipped with an anti-reflux valve to avoid reflux of fecal material. The CO2 delivery was controlled by a dedicated electronic device (PROTOCO2L Colon Insufflator System; E-Z-EM, NY, USA).

A single-breath-hold acquisition was used to encompass the entire colon, and repeated twice, with the patient firstly in supine and secondly in prone position or, if not possible, in left decubitus position.

CTC image analysis

Evaluation of the colon

CTC images were transferred to an independent workstation (CADCOLON; IM3D, Torino, Italy) that allowed the simultaneous 2D and 3D viewing of prone and supine images. However, a primary 2D view was used for image evaluation of both decubitus. The colon was evaluated per patient and on a segmental basis, considering six segments: cecum, ascending, transverse, descending, sigmoid colon, and rectum. For all assessments a bone window was used (Windows 3000 UH, Level 300 UH).

Colonic cleansing

The residual feces were divided in two groups according to their size as measured on the axial slices: less than 6 mm and 6 mm or greater. If both sizes were present in one segment, the segment was considered to have residual feces of 6 mm or greater. The reason for distinguishing between less than 6 mm and 6 mm or greater ones was that it is generally accepted that polypoid lesions of size less than 6 mm can be ignored; therefore, feces of size less than 6 mm were not considered.

Fluid tagging

The amount of residual fluid was evaluated on supine axial slices by considering its proportion with respect to the maximal antero-posterior diameter of the segment of the colon where it was detected. In this way, we pointed out four groups: 0%, less than 25%, 25–50%, and greater than 50%. In segments with different fluid levels, only the largest fluid level was considered.

The tagged appearance was first assessed on a visual basis: untagged, non-homogeneously tagged, or homogeneously tagged.

The homogeneity of fluid tagging was also evaluated collecting the attenuation values (UH) of tagged fluid (untagged cases were excluded). Mean density and standard deviation (SD) of tagged fluid were determined in four colonic segments (ascending, transverse, descending, and sigmoid-rectum).

Patient’s acceptance

A self-administered questionnaire was used for the assessment of the discomfort related to the preparation. After bowel preparation and before the exam, and then after the exam, the patients were administered a questionnaire which was about the discomfort related to the 3 days of diet, to the bowel cleansing, and to the fluid tagging. The patients rated their discomfort on a visual ten-point scale ranging from 0 (severe discomfort) to 10 (no discomfort).

The questionnaire given to the patients also included space for comments and for reporting side effects.

To evaluate the difference of acceptance with respect to a cathartic preparation, a control group of 10 patients (group B), who underwent cathartic cleansing with 3–4L of PEG solution the day before the exam, was introduced. They received the same questionnaire as the group that underwent tagging (group A).

Lesions detection

Colorectal lesions were measured on supine and prone decubitus in 2D and with multiplanar reconstructions, and maximum diameter was considered for classification. Consequently, they were categorized on the basis of C-RADS [20]. Accordingly, patients with 3 or less then 3 medium size polyps (C2) were suggested to undergo a 3-year follow-up, patients with more than 3 medium size polyp or at least 1 advanced adenoma (C3) were referred to colonoscopy, and patients with colorectal masses (C4) were referred to surgical consultation. Concerning the follow-up of C2 cases, an agreement was reached with the patient and in some cases the follow-up was shortened to 1 year.

Statistical analysis

All statistical calculations were performed by using the software packages StatView (SAS Institute, Campus Drive, NC) and Excel (Microsoft, Redmond, Wash).

The Wilcoxon signed rank test was used to compare the score of patient acceptance between groups A and B.

The average density of fluid in cecum-ascending and descending-sigmoid colon were compared with the Student t-test. The average differences in density and standard deviation were also calculated.

P values of less than or equal to 0.05 (95% confidence interval) indicated a statistically significant difference.

Proportions of C-RADS categories were calculated.

Results

Fecal residues

The results of the assessment of the amount of residual feces by colonic segment are listed in Table 1.
Table 1

Number of cases with residual feces per segmental location

Total segments 792

Cecum

Ascending

Transverse

Descending

Sigmoid

Rectum

Overall (%)

0

69

86

86

61

69

75

564

<6 mm

34

23

29

54

40

23

256

>6 mm

29

23

17

17

23

34

18

There were 446 segments (56.4%) clean of feces and 346 segments (43.6%) with feces. Most feces were located in the descending-rectum: 191 segments representing 24% of the total number of segments; in the ceco-ascending colon we found 13.7% of fecal residues and 14% in the transverse.

There were 203 segments with feces smaller then 6 mm and 143 segments with feces of 6 mm or greater, 25.6% and 18% of the segments, respectively.

Amount of residual fluid

The results of the assessment of the amount of residual fluid are listed in Table 2.
Table 2

Results of the assessment of the amount residual fluid

Total segments 792 (%)

Cecum

Ascending

Transverse

Descending

Sigmoid

Rectum

Overall (%)

0

0

2

2

15

32

35

108

<25

16

35

36

45

53

58

307

25–50

72

78

71

59

38

35

448

>50

44

17

23

13

9

4

138

The vast majority of colonic segments were filled by fluid for less than 50% (overall 86.2% of colonic segments). The largest amount of fluid was in the cecum and conversely the smallest in the rectum.

Fluid tagging

The results of the assessment of the fecal/fluid tagging are listed in Table 3.
Table 3

Homogeneity of fluid tagging by colonic segment

Total segments 706

Cecum

Ascending

Transverse

Descending

Sigmoid

Rectum

Overall (%)

Untagged

0

6

9

15

15

29

105

Unhomogeneously tagged

6

23

21

21

17

25

16

Homogeneously tagged

126

101

98

79

68

43

723

Untagged fluid was observed in 74/706 (10.5%) segments, which were essentially located in the left colon, as expected due to the longer transit time needed for an appropriate dilution of iodine. For the same reasons, inhomogeneous tagging was observed in 129/706 (16%) segments, mostly in the left colon.

The average density of fluid was 1054.74 UH (min.309, max 1944; std. dev. ± 391.28) in the cecum-ascending colon, and 905.14 UH in the descending-sigmoid colon (min.234, max 1809; std. dev. ± 443.83); the average difference of density between right and left colonic segments was 149 UH, and it was statistically significant (P = 0.016).

Patient’s acceptance

No side effects related to the consumption of Movicol were reported. Very few side effects related to the tagging solution were reported: mild nausea in 7 (0.05%) patients, mild diarrhea in 10 (0.07%). An average rank of 9 points (std. dev. ± 1) resulted from the self-administered questionnaire, showing an excellent acceptance of the preparation. All patients reported that the tagging solution was well tolerated and the diet restriction combined with the consumption of Movicol did not influence their lifestyle the days prior to the exam.

In the control group B the average rank was 5.3 (std. dev. ± 1.4). The difference between group A and B was statistically significant (P = 0.0051). These patients reported nausea in 4/10 (40%) and vomiting in 2/10 (20%), and all, as expected, reported diarrhea.

Colorectal lesions

Colorectal lesions were suspected at CTC in 18/132 (13%) cases; therefore, the patients were categorized with C-RADS as follows: C1 (n = 114), C2 (n = 14), C3 (n = 3), and C4 (n = 1). Six out of 14 patients classified as C2 underwent 1–3 years follow-up. The remaining eight patients underwent colonoscopy that confirmed CTC findings in seven out eight (88%) cases (one false positive). All the lesions classified as C3 and C4 were confirmed by colonoscopy. Among C3 lesions one low grade dysplasia polyp was found by means of histopathology. The patient classified as C4 underwent surgery that confirmed an annular stenosing carcinoma of the transverse colon (4 cm extension along the colonic segment).

Discussion

A homogeneous fluid tagging was obtained in 71.5% of the colonic segments, and, as reported above, the untagged or non-homogeneous fluid was in the left colon. As a matter of fact, this shows that the transit time is a key factor in improving fluid tagging when using a same day regimen. Our patients were admitted in the Radiology Department very early in the morning, and soon started the consumption of the iodine solution. We decided to scan the patients after 3 h as it was considered a reasonable time frame to allow that the solution reaches the left colon with an adequate mixing with the residual fluid. However the presence of 10.5% of colonic segments with untagged fluid suggests that the time interval from the oral consumption of iodine and CT scanning should be increased. Presumably, patients with chronic constipation would need to wait at least 4 h.

One drawback of this tagging regimen was the presence of untagged solid fecal residues that occurred in 43.6% of the segments and in a minor percentage (18.6%) the diameter of stool was equal or greater than 6 mm, mostly located in the left colon. However, even if the feces were untagged, they were easily recognizable while moving to the non-dependent part of the colon after the patient moved into the supine and prone position, and often they were seen floating in a pool of iodinated agent, thus allowing distinction from wall lesion (Fig. 2). Therefore, even if stools greater than 6 mm should be considered in differential analysis from polyps, the advantage of positional changes according to patient decubitus could be exploited in image interpretation.
https://static-content.springer.com/image/art%3A10.1007%2Fs00261-008-9453-z/MediaObjects/261_2008_9453_Fig2_HTML.jpg
Fig. 2

(A) Primary CTC acquisition in supine in a patient with same day iodine fluid tagging. In the descending colon a 7 mm solid stool (arrow) floats within the residual fluid. (B) In prone decubitus the stool (arrow) changes its position accordingly following gravity and fluid redistribution along the colonic lumen.

In our tagging scheme there are two essential steps which impacted on the patient’s acceptance: the colonic cleansing (performed at home) and the tagging (performed in the Hospital).

The consumption of Movicol, which is a mild laxative, did not cause any cathartic effect; therefore, the patients did not experience any adverse event as it could be in case of using the traditional colonoscopy cleansing solutions (like oral sodium phosphate or polyethylene glycol) [21, 22]. In most of the cases patients reported a softened consistency of the feces, especially in those with chronic constipation, and this was considered a successful result of the study. In fact the softening of the feces reduced the amount of solid residues that were difficult to tag by a same day preparation. Most important was the fact that the consumption of Movicol did not change the patient’s habits or lifestyle the days before the exam.

The idea of a same day preparation stems from the need of monitoring patients just after the consumption of iodated contrast media. As discussed above, the oral consumption of iodixanol + PEG solution was done 3 h before the CT scan, and mild nausea or diarrhea occurred in very few cases. Zalis et al. [18] in their recent study comparing barium sulfate suspension, iopromide, and iohexol plus magnesium citrate, report the occurrence of mild diarrhea in very few cases in all the three tagging regimens, but there were no other adverse symptoms reported by the subjects; moreover, the majority of patients indicated that they had “tolerated well” the contrast material regimen. We believe that the side effects occur in our study were most probably due to PEG. Iodixanol is iso-osmolar and should not cause diarrhea; on the other hand, iodinated contrast agents can cause allergic reaction, but very rarely in case of oral consumption. It has been shown also that the administration of the dimeric iodixanol presents a lower frequency of adverse events/discomfort than the administration of monomeric contrast media [23]. In any case, we did not observe any adverse effect due to allergy and we could easily control mild side effects (PEG’s related), in the patients being under observation in the radiological waiting room. The patient’s tolerance was high, as showed by the excellent results of the questionnaire. In general each patient reported that the unique mild discomfort was related to pneumocolon.

It is important to point out that the protocol tested in this experience appears ideal for a screening program. The studies reporting patient’s opinion emphasize that the worst part of a CTC or a colonoscopy exam is the preparation; therefore, any non-cathartic colonic preparation is greatly accepted [1]. On the other hand, tagging could be obtained under medical control and with good technical quality on the same day of the CTC, giving the patient the feeling of a fast and safe procedure that can be performed without limiting the daily life.

We believe that from the technical point of view and patient’s tolerance the preparation proposed in this preliminary experience could be tested in a screening program, since it provides an optimal fluid tagging, it is completely tolerated by the patient (no side effects, no cathartic effects, no altered lifestyle and habits), and mostly important the iodine consumption is under medical control. Further assessments are required to determine its efficacy in improving the detection of colorectal lesions.

Copyright information

© Springer Science+Business Media, LLC 2008