Evaluation of the performance of two neutral oral contrast agents in computed tomography enterography: A randomized controlled trial

Objective To compare the performances, tolerability and acceptability of mannitol and polyethylene glycol (PEG) as oral contrast agents in patients undergoing computed tomography enterography (CTE). Methods Patients aged 18‐75 years indicated for CTE were randomized to receive either mannitol or PEG as contrast agents. The coronal reconstructed images of each abdominal quadrant were assessed for maximum distention, proportion of distended bowel loops, presence of inhomogeneous contents and visibility of the small bowel wall. Overall subjective imaging quality assessment and patients’ tolerability and acceptability were recorded. Results Seventy patients were enrolled and randomized into two groups. In the per‐protocol analysis, no significant differences in imaging quality was found in bowel distention maximum diameter, wall visibility and intestinal homogeneity (all P > 0.05). The mean nausea score was lower in the mannitol group (0 [0‐0] vs 1.0 [0‐3.0], P < 0.001). Mannitol was superior to PEG in taste (9.0 [8.0‐10.0] vs 7.0 [5.0‐8.0], P < 0.001), patients’ willingness to reuse the drug (9.0 [8.0‐10.0] vs 8.0 [7.0‐9.0], P = 0.036), satisfaction (9.0 [8.0‐10.0] vs 8.0 [7.0‐9.0], P = 0.022) and ease of completion (9.0 [8.0‐9.3] vs 8.0 [6.5‐9.0], P = 0.030). Conclusions Both mannitol and PEG provided good bowel distention and visualization of the bowel wall. However, mannitol was significantly superior to PEG in patients’ tolerability and acceptability.


| INTRODUCTION
Thanks to the development of imaging technology, computed tomography enterography (CTE) has become one of the first-line modalities for detecting diseases of the small bowel, particularly inflammatory bowel diseases (IBD). [1][2][3] Unlike conventional computed tomography (CT) examination, CTE allows the visualization of both small intestinal wall and lumen after ingested a large volume of contrast agents. 4 Moreover, CTE shows clearly the pathological changes of the small bowel by presenting mural stratification, segmental mural hyperenhancement, increased density of mesenteric fat and engorged vasa recta. [5][6][7] Quality of a CTE examination depends mainly on adequate intestinal distention and wall visibility, which optimize the resolution of the bowel wall and lumen. 8 Oral contrast agents used in CTE examinations are considered the key to achieving a satisfactory outcome. In IBD patients the form of mural enhancement is crucial to the diagnosis, especially for Crohn's disease. 9,10 Neutral oral contrast agents, which are isodense, are preferred to positive oral contrast materials in CTE as these agents improve the conspicuity of bowel mucosal and mural hyperenhancement. 11,12 Several studies have compared the performances of various neutral contrast agents used in CTE, including water, whole milk, methylcellulose, polyethylene glycol (PEG), mannitol, lactulose and lowconcentration barium. [13][14][15] Excellent imaging quality combined with a good taste makes mannitol one of the most widely used neutral contrast agents in CTE. 14,16 However, risks of dehydration and loss of electrolytes, as well as that of producing explosive gas after mannitol ingestion have significantly limited its clinical application. 17,18 Although no serious adverse reactions have been reported, it remains to be confirmed whether the use of isotonic mannitol as an oral contrast agent in CTE is accompanied by other side effects. PEG has also been recommended as an oral contrast agent for CTE because of its safety; however, certain inherent attributes of PEG may have negative effects on the imaging quality of CTE. Moreover, its poor taste and drainage effect as a volumetric laxative may also affect the imaging quality and patients' tolerance to the agent. 13,19,20 As few relevant prospective studies are available, 15 whether the quality of CTE imaging using PEG as an oral contrast agent was inferior to that using mannitol, despite its safety advantages, remains to be investigated.
The aim of this study was to evaluate prospectively the performances of two neutral oral contrast agents, mannitol and PEG, in CTE, and to assess the image quality and patient's tolerability and acceptability.

| Randomization
The allocation sequences were generated by a computer and encapsulated in sealed, opaque envelopes by an independent investigator.
Patients were randomized to receive either of the two neutral oral contrast agents by opening one of these envelopes. The nurses, image technicians, radiologists and analysts were all blinded to the patients' randomization.

| Regimens for preparation
The patients were instructed to take PEG (2 L) before the CTE for bowel preparation. They were then randomized to receive 1500 mL oral contrast agent, either mannitol (isotonic mannitol solution diluted with hypertonic mannitol; Baxter, Shanghai, China) or PEG (pineapple-flavored PEG electrolyte power dissolved to 1500-mL volume; Wanhe Pharmaceutical, Shenzhen, Guangdong Province, China), in 50 minutes: 1000 mL within the first 30 minutes, 250 mL in the next 10 minutes and the remaining 250 mL in the last 10 minutes.
After taking the contrast agent, 20 mg anisodamine hydrochloride was given intramuscularly. Patients were then transferred immediately to the CT room, and CT scan was performed within 10-20 minutes after the injection.

| CTE procedure
Plain and contrast-enhanced triphasic CT scans were performed by using a dual-source CT (Siemens, Munich, Germany

| Assessment of CTE images
The primary outcome was the imaging quality of CTE by using the two neutral oral contrast agents. All the CT findings and coronal reconstruction CT images were obtained and evaluated by a single experienced radiologist (QSZ) in a blinded manner on a computer workstation using a picture archiving and communication system (Impax, AGFA, Mortsel, Belgium). The abdomen was divided into four quadrants using a vertical line through the xiphoid process and a horizontal line through the iliac spine. 13 Based on prior studies and radiologists' recommendations, 21,22 intestinal segments of no less than 20 mm in diameter were defined as adequately distended segments. Each quadrant was assessed for the overall distention based on the proportion of adequately distended bowel segments, which was graded on a scale of 1-4: 1, poor distention (0%-25%); 2, fair distention (26%-50%); 3, good distention at each of the four quadrants were recorded by the observer, which was recorded as "yes" or "no". The overall evaluation included a judgment on whether the contrast agent had reached the ileocecal part, as well as an overall assessment with a score of 1 (poor) to 4 (excellent), based on the abovementioned criteria.

| Assessment of tolerability and acceptability of the patients
The secondary outcomes of the study were the patient's tolerability and acceptability of both neutral oral contrast agents. After completing the CT scan, each patient was interviewed using a questionnaire about their tolerability and acceptability. The patients were asked to rate the degree of nausea, vomiting, diarrhea, abdominal distention and pain during the ingestion of oral contrast regimens on a scale of 0-10, with 0 being no feeling and 10 being unbearable. They were also asked to rate the taste, ease of completion of the contrast agent ingestion, their own willingness to reuse the contrast agent and their satisfaction with the preparation on a scale of 0-10, with 0 being unacceptable and 10 being satisfactory. The presence or absence of dizziness, weakness and electrolyte disorders was also documented.

| Sample size calculation
The study was designed to be a non-inferiority trial, which is an experiment to test whether a new drug or therapy is not inferior to an existing positive control drug or therapy. and PP analyses were used to evaluate the primary outcome. P < 0.05 was considered significant for the statistical analyses.

| Patients' characteristics
From October 2017 to August 2018, 85 eligible inpatients were assessed for inclusion, of whom 15 were excluded due to intestinal obstruction (n = 7), a previous history of colorectal surgery (n = 2), severe constipation (n = 1) and declined to participate in the study (n = 5), respectively. Finally, a total of 70 patients were equally randomized into the mannitol group or the PEG group. Patients' characteristics are shown in Table 1. There were no significant differences in age, gender distribution, or body mass index between the two groups (all P > 0.05).
The imaging findings of 60 patients were analyzed in the PP analysis, while 10 patients were excluded due to their failure to finish the examination on time (n = 2), or to take the contrast agent as required (n = 4), intestinal dysplasia (n = 1), possible drug allergy (n = 1) and intestinal gas accumulation (n = 2). The flow diagram of patient enrollment and randomization is shown in Figure 1.

| CTE images
The diameters of the most distended bowel loops at the four abdomi-   In summary, we found that PEG was statistically not inferior to mannitol in CTE, based on the maximum mean diameter of the distended bowel loops and the imaging quality ( Figure 2).

| Patients' tolerability and acceptability of the contrast agents
Adverse events related to the use of contrast agents were analyzed (Table 3)

| Medical cost of the contrast agents
The cost of mannitol per patient was significantly less than that of PEG. Specifically, for each eligible exam, mannitol cost $1.28 per patient, while PEG cost $8.86 per patient.

| DISCUSSION
Our results showed that both substances were equally effective for imaging quality. However, the patient's tolerability and acceptability of mannitol were significantly superior to those of PEG, and with a lower cost.
This study suggested that the imaging quality using mannitol and PEG as oral contrast agent did not differ significantly with respect to small bowel distention, bowel wall visibility, and intestinal homogeneity.
Previous studies have shown that both mannitol and PEG can achieve relatively good imaging quality. 14,19 Because both mannitol and PEG are isodense that can improve the conspicuity of bowel lumen and wall.
Some studies have indicated that mannitol can be fermented by gut bacteria to produce hydrogen and methane gases, 17,18 which, as inhomogeneous components, may affect the image quality of CTE. In this study, no difference was found in intestinal homogenity between the two groups.
This may be because that hydrogen and methane are mainly produced during anaerobic bacterial activity in the human colorectum. 24 Moreover, the process may be too short to produce detectable gases.
Unexpectedly, in terms of the adverse events, the score for nausea in the PEG group was significantly higher than that of mannitol group, and one patient in the PEG group developed nausea. The score for taste of PEG in our study was significantly lower than that for mannitol, which is consistent with the results of Leduc et al. 13 Poor taste also raised other issues, as patients in the PEG group were significantly less satisfied with the preparation and showed a significantly lower willingness to reuse the contrast agent than those in the mannitol group.
In this study, the ease of completion score was significantly higher in the mannitol group and the results also showed that the completion rate of the pre-examination preparation in the PEG group was lower than that of the mannitol group, as three of the four patients who did not complete the prescribed preparation were those who ingested PEG. This may be due to the reason that all the patients used PEG as routine bowel preparation regime before CTE. In the mannitol group, the sequential use of the two agents made the taste of the contrast more distinct; while in the PEG group, repeated use of  both performance and patient's tolerability and acceptability of mannitol and PEG for CTE.
There were several limitations to our study. First, we involved inpatients only in our study. Thus, the efficacy of these two oral contrast agents cannot be generalized to outpatients. Second, patients were not followed up after they completed the questionnaire for subsequent adverse reactions such as diarrhea, leading to the inability to assess the effects on patients' quality of life after the examination.
Finally, the patients in our study were not blinded to the agents as the two drugs differed greatly in taste. Therefore, further investigation is needed to overcome these shortcomings.
In conclusion, both mannitol and PEG achieve good imaging quality with respect to the degree of bowel distention, wall visibility and intestinal homogeneity. In addition, the taste of mannitol was significantly superior to that of PEG, which was closely related to its acceptability and made it easier for patients to complete the prescribed preparation. Both mannitol and PEG are relatively safe oral neutral contrast agents. In general, with its good imaging effect and low cost, mannitol is a contrast agent that can be widely used. PEG formulations provide an alternative for some special groups because of its electrolyte composition.

CONFLICTS OF INTEREST
None.