Bowel dysfunction after stroke occurs frequently. It is widely known that there are many factors responsible for bowel dysfunction, such as neurogenic bowel, dysphagia, decreasing activities due to motor weakness, aging, inappropriate water or nutritional intake, depression, various drug intakes, and other factors [
14,
15]. However, relatively few studies have been performed on bowel dysfunction after stroke. Bowel dysfunction after stroke, a chronic complication due to neurogenic bowel, is difficult to manage. Additionally, bowel dysfunction complications tend to be less severe and less fatal compared to neurogenic bladder [
16]. However, if there is no appropriate management for bowel dysfunction after stroke, it will limit social activities, decrease the quality of life, cause psychological and social retreat, thus making it difficult for patients to perform daily duties. It is important to have a correct evaluation to properly manage bowel dysfunction. In order to achieve this, different methods have been developed. CTT is considered an objective method for the evaluation of colonic motility [
17]. CTT was used in this study as comparative criteria for different evaluation methods of bowel dysfunction. The easiest and most widely used method is clinical assessment of the bowel pattern usually conducted by taking history. Assessment on bowel pattern by history taking depends on subjective symptom and reports of patients. However, stroke patients' memories on their bowel habits are sometimes inaccurate [
4], resulting in patient's subjective interpretation. CTT, a proven objective evaluation method, has a significant relationship with the symptom of constipation. CTT not only can evaluate the total colon, but also can evaluate the segmental colon [
17]. However, this method is inapplicable to those who have difficulties in swallowing capsules that contain radio-opaque markers. CTT has the disadvantages of requiring at least 4 days to process, including two time plain abdominal radiography for the evaluation of results. Additionally, it has the limitation in outpatient clinics because patients have to intake radio-opaque Kolomark every day at the same time to have plain abdominal radiography on the 4th day at a hospital. Another objective evaluation method of constipation is plain abdominal radiography. Recently, some studies [
18,
19,
20] have reported that plain abdominal radiography has poor diagnostic accuracy for constipation. However, previous studies [
7,
8] have reported that plain abdominal radiography is significantly correlated with symptoms of constipation. Nonetheless, plain abdominal radiography does have some benefits. Compared to CTT, the plain abdominal radiography requires only onetime radiography. In addition, plain abdominal radiography has the advantages of less cost, simple to perform, and less exposure to radiation. In 2013, Park et al. [
21] found that plain abdominal radiography for spinal cord injured patients had a significant correlation with CTT on constipation scores. They also reported that the interinvestigator consistency was high. Thus, plain abdominal radiography is easy to use and applicable for clinics. With the constipation score obtained through the survey of the patients with stroke, this study evaluated the degree of stool retention using the plain abdominal radiography. We also found the bowel pattern using Bristol stool form scale. We examined the correlations by evaluating CTT to determine which method would be useful for bowel dysfunction in plain abdominal radiography. We further reported the convenience in clinical application. Our results revealed that there were significant differences in constipation score and Bristol stool form scale between the constipation and the non-constipation groups. Delay of total and left colon CTT were observed in constipation group, which was consistent with previous studies [
5,
6]. Regarding the rectosigmoid colon CTT, the delay was observed in the constipation group, which was different from previous studies. With respect to the stool retention score through the plain abdominal radiography, the Starreveld score showed a statistically significant high score in the ascending colon, transverse colon, and descending colon in comparison between the constipation and the non-constipation groups, while Leech score showed a high score in the right and left colon. These results showed that there was a significant difference between the two groups with respect to the constipation score, Bristol stool form scale, total CTT, and stool retention scores. The application of Roman II Diagnosis Criteria is considered applicable to the constipation diagnosis for stroke patients. Through total CTT and plain abdominal radiography, there was a statistically significant correlation between the Starreveld score and the Leech score. Similarly, there was a statistically significant correlation between segmental CTT and segmental stool retention score. Based on these results, not only the total CTT, but also each segmental CTT could be inferred through plain abdominal radiography. Thus, in addition to CTT, plain abdominal radiography was useful as an evaluation method for bowel dysfunction in stroke patients. Furthermore, there was a statistically significant correlation in stool retention score through the plain abdominal radiography with patients' subjective symptom and constipation score reflected by the bowel pattern. Recent studies [
18,
19,
20] have reported less diagnostic accuracy of plain abdominal radiography because it fails to reflect the serial bowel state of constipation. Although the amount of stool loading is subject to daily variation, our studies revealed significant correlation between CTT and plain abdominal radiography. Independently, plain abdominal radiography isn't a sufficient method to completely replace CTT. However, it can assist CTT in the diagnosis of constipation. With this correlation, we conclude that plain abdominal radiography can effectively evaluate the severity of constipation and determine the need for treatment. Because the daily performance abilities of stroke patients are significantly lower than those of normal people, examination convenience and simplicity must be considered in determining the method of choice. For patients who are treated at the outpatient clinics, it is difficult to evaluate CTT due to problems associated with its examination method and time. Therefore, to evaluate bowel dysfunction in outpatient clinics, it is necessary to prepare questionnaires to evaluate the management and conditions for the bowel dysfunction. If plain abdominal radiography is employed in replacement of the complicated CTT, the total and segmental colon activities can be known. Therefore, plain abdominal radiography is considered effective to obtain immediate results on the day of outpatient clinic visits, with the advantages of faster management and treatment as well as convenience for the patients.
Considering the characteristics of stool retention through colon segmentation on plain abdominal radiography, osmotic laxatives, a variety of enemas, suppositories, digital disimpaction, and other induction of bowel movement and management can be applied. Differences can arise among investigators when evaluating the degree of stool retention in plain abdominal radiography. This study employed two skilled and well-trained radiologists who understood the evaluation method well. They examined the plain abdominal radiography. The variations in results among those investigators were evaluated. There was a significant (p<0.01) consistency of inter-investigators regarding the degree of stool retention on using the Starreveld score and Leech score. If investigators are well trained in examining the degree of stool retention using the evaluation criteria, plain abdominal radiography can be effectively used as an objective evaluation method. However, regarding the Starreveld score which divides the total colon into four sections and evaluates each colon, there may be an overlap between the transverse and ascending colon or between the transverse and descending colon if there is excess stool retention. It may result in an error in evaluation. In such cases, it is useful to adopt the Leech score that divides the total colon into three different sections.
This study has some limitations. It did not consider patients' diets, underlying disease, or medications that might affect bowel dysfunction. Drugs that the patients previously used to control bowel dysfunction were not evaluated in this study. The correlations by comparing the evaluation methods for normal people and stroke patients were not determined. This study excluded patients who were unable to swallow Kolomark. Therefore, further studies are needed to address the above limitations.
In conclusion, this study examined the usefulness of plain abdominal radiography in evaluating bowel dysfunction after stroke. There were statistically significant correlations between the CTT and Starreveld and Leech scores evaluated by plain abdominal radiography, the most objective evaluation method for inspecting colon activities. Although plain abdominal radiography has poor diagnostic accuracy for constipation, it is useful for the evaluation of bowel dysfunction after stroke, therefore aiding the management of bowel status and implementation of appropriate treatment procedures. Due to its fast evaluation time and reduced radiation exposure, plain abdominal radiography could assist the diagnosis of constipation with applicability in clinics because of its convenience.