Authors: Lin Li, Xue Zhan, Yuxia Chen, Jun Li, Yuting Wang
Categories: Original Articles: Gastroenterology, Meckel’s diverticulum, children, diagnosis, small bowel capsule endoscopy
Source: European Journal of Gastroenterology & Hepatology
Meckel diverticulum (MD) is an important cause of gastrointestinal bleeding in children. Small bowel capsule endoscopy (SBCE) is a first-line examination method applied to patients with obscure gastrointestinal bleeding, but there are few studies on its application in children with MD. This article aims to provide evidence in favor of the auxiliary diagnosis of MD in children by analyzing its characteristics using SBCE.
We retrospectively collected the clinical data of patients with suspected MD.
A total of 58 children were included in this study. All 58 children presented overt gastrointestinal bleeding (bloody stool or melena). Capsule endoscopy identified protruding lesions in 2 cases, double-lumen changes in 30 cases (all considered as MD), vascular lesions in 7 cases, intestinal mucosal inflammatory lesions in 3 cases, ulcers or erosion in 3 cases, and no obvious abnormalities in SBCE in 12 cases. Both SBCE and technetium-99 scans were performed for 24 cases, 22 of which were diagnosed MD by their combined results, giving a diagnostic coincidence rate of 91.7%. Eight cases were highly suspected as MD but were negative for the technetium-99 scan and positive for SBCE.
SBCE has high accuracy in the diagnosis of MD in children, especially when performed in combination with a technetium-99 scan, which can greatly improve the diagnostic rate of MD in children.
Keywords: children, diagnosis, Meckel’s diverticulum, small bowel capsule endoscopy
Meckel’s diverticulum (MD) is one of the most common congenital gastrointestinal malformations ^[1]^ and is characterized by a true diverticulum formed by incomplete occlusion of the yolk tube (umbilical mesentery) during embryonic development. Its prevalence rate in the general population is between 0.3 and 2.9% ^[2].^ Most cases of MD may be asymptomatic for life, sometimes only being found during abdominal surgery; alternatively, the condition may instead involve serious complications, such as massive gastrointestinal bleeding or perforation ^[2].^ Small bowel capsule endoscopy (SBCE) is a first-line examination method for obscure gastrointestinal bleeding ^[3]^. However, there exist few studies on its application in children with MD. This article aims to provide a reference for the auxiliary diagnosis of MD in children by analyzing the characteristics of MD in SBCE.
We retrospectively collected the clinical data, abdominal imaging results, radionuclide scanning results, surgical treatment plan, and pathological results of patients with obscure gastrointestinal bleeding who were hospitalized, examined by SBCE, and finally confirmed to have MD by surgery and pathology in the affiliated Children’s Hospital of Chongqing Medical University from September 2020 to June 2022. This study was approved by the Ethics Committee of the Children’s Hospital of Chongqing Medical University. The parents or legal guardians of all participants who agreed to undergo SBCE signed an informed consent form.
We used the PillCamTMSB3 capsule endoscope system to complete the examinations. The patients were fed a low residue and semi-liquid diet starting 24 h before the operation, which was then, followed by 3 h of fasting immediately before the operation. The patients drank compound polyethylene glycol electrolyte solution (25 ml/kg) to clear the intestines at both 12 h and 3 h before the operation, with each drink being consumed within a 1-hour window. A 30 ml volume of silicone oil was also consumed by each patient 3 h before the operation. After swallowing the SBCE capsule, each patient was placed in the right supine position with their upper body raised by 30 degrees. If the SBCE capsule did not enter the duodenum 2 h after being swallowed, a domperidone tablet (0.3 mg/kg, maximum dose of 10 mg) was administered. If the capsule still failed to enter the duodenum 4 h after being swallowed or if the patient had difficulty swallowing, the capsule was then sent to the distal duodenum by gastroscope. When the capsule entered the colon or after the examination was over, the patient’s normal diet was resumed. The time of the capsule passing through the small intestine, the retention rate of the capsule, and the pathological changes of the small intestine were recorded.
The software package SPSS 26 was used to conduct all statistical analyses. Metrological data have been expressed as the mean ± SD (x ± s), while classified data have been expressed by the number of cases and their proportion (n, %). Multiple groups were compared by Fisher’s exact probability test or the chi-square test. P < 0.05 was used to indicate a significant difference. We considered the gold standard of MD diagnosis to be that confirmed by both operation and pathological biopsy.
A total of 58 children were included in this study, including 41 males and 17 females. All 58 children presented overt gastrointestinal bleeding (blood stool and melena), 46 cases presented blood in the stool, and 12 cases had black stool. Among them, 29 children were examined and treated by operation, including 25 males and 4 females. Twenty-three cases had bloody stool and six cases had melena. The shortest time for the occurrence or persistence/recurrence of gastrointestinal bleeding symptoms was 1 day and the longest was half a year. There were five cases with dizziness or syncope, nine cases with abdominal pain, two cases with diarrhea, and one case with vomiting (Table 1).
Among the 58 cases of SBCE, protruding lesions were found in two cases, double-lumen changes in 30 cases (all considered as MD), vascular lesions in seven cases, intestinal mucosal inflammatory lesions in three cases, ulcer or erosion in three cases, and no obvious abnormality in SBCE was found in 13 cases (Table 2).
Among the 29 cases who underwent the operation, the small bowel transit time of the SBCE capsule was 297.1 ± 128.3 min (64, 612). The average time taken for the SBCE capsule to locate lesions in the small intestine was 239.2 ± 98.9 min (78, 481). The average distance between the MD and ileocecum was 45.2 ± 16.8 cm (20 cm, 100 cm). All the capsules ingested by the 29 children were naturally excreted from the body, and no capsule retention occurred (Table 3).
Among the 29 cases, three cases had no obvious abnormality found by SBCE, while 26 cases were identified to have an abnormality. Among the 26 abnormal cases, SBCE identified multiple ulcers of the small intestine in one case, protruding lesions in one case (Fig. 1b), and double-lumen changes in 24 cases (Fig. 1a). The coincidence rate with the surgical and pathological diagnoses was 82.8% (Table 4).
Fig. 1. The images of MD patients diagnosed by VCE, operation. (a) Double lumens in patient 7. (b) Protruding lesions in patient 21. (c) Surgically resected tissue of MD in patient 7. MD, Meckel diverticulum.
The SBCE results of all four female children were consistent with the diagnosis, while 20 of the 25 male children had SBCE results in accordance with the diagnosis. The Fisher’s exact probability test results were compared between the two groups to find that P = 1.0 > 0.05, suggesting that gender did not affect the accuracy of SBCE diagnosis (Table 5).
Among the nine cases of children who were ≤6 years old, the SBCE results of six cases were in accordance with the diagnosis, while those of three cases were not in accordance with the diagnosis. Among 13 cases of children who were 7–11 years old, the SBCE results of 11 cases were in accordance with the diagnosis, while two cases were not. Of the seven cases of children ≥12 years old, the SBCE results of all were consistent with the diagnosis. The chi-square test was used to compare the three χ^2^ = 3.12 < 5.99 (P = 0.05; ν = 2) and P > 0.05, indicating that age did not affect the diagnostic accuracy of SBCE (Table 5).
The capsule endoscope was swallowed in 18 cases, of which, the SBCE results of 15 cases were in accordance with the diagnosis, while those of three cases were not. In 11 cases, the capsule was sent to the duodenum by gastroscopy, of which, the SBCE results of nine cases were consistent with the diagnosis while those of two cases were not. Fisher’s exact probability test was used to compare the two P = 1.0 > 0.05, suggesting that the method of capsule intake did not affect the diagnostic accuracy of SBCE (Table 5).
For bowel preparation, we defined good preparation of the intestinal tract as the the mucous membrane was in perfect condition for more than 75% of the recording time, the tract contained only some liquids or fragments, and the contained liquids or fragments did not seem to interfere with the overall quality of the examination. We defined moderate preparation as the the mucous membrane was under perfect observation conditions for 50–75% of the recording time, and there were enough fluids, bubbles, or fragments to rule out a completely reliable examination. If the mucous membrane could only be observed for less than 50% of the recording time and there was a large number of liquids, bubbles, or fragments that affected the interpretation of the examination, the bowel preparation was considered poor. Four children had no recorded effect of bowel preparation, so they were not included in the comparison. A total of 25 children were included in the comparison, among which 14 cases met the criteria for good bowel preparation, all were in accordance with the diagnosis, and 10 cases were considered moderately prepared, of which six cases were in accordance with the diagnosis, while four cases were not. One case was classified as having poor preparation, though the results were in accordance with the diagnosis. The chi-square test was used to compare the three groups, where χ^2^ = 7.14 > 5.99 (P = 0.05, ν = 2) and P < 0.05, which means that the different bowel preparation outcomes are likely to have had an impact on the coincidence rate of SBCE diagnosis (Table 5).
The age, sex, height, weight, effect of bowel preparation, small bowel transmitting time, capsule intake mode, and SBCE final results were analyzed by multiple regression analysis. First, by the collinear analysis of each factor, it was concluded that the variance inflation factor values of age, height, and weight all exceeded 10, indicating them to be collinear, so these three factors were excluded. Multiple regression analysis was conducted according to sex, the effect of bowel preparation, small bowel transmitting time, capsule intake mode, and the SBCE final results; the adjusted R^2^ was 0.1908, indicating that gender and other four factors could explain 19.08% of SBCE results. F = 2.346 and P = 0.099 > 0.05, suggesting the regression model had no significant statistical significance (Table 6).
Before SBCE, to identify the cause of gastrointestinal bleeding, 24 children underwent the MD scan (technetium-99 scan), of which 14 cases had results consistent with the diagnosis, and the diagnostic coincidence rate was 58.3%. Gastroscopy was performed in 20 cases, no obvious lesion was found in six cases, gastritis was found in 13 cases, and ectopic gastric mucosa in the upper segment of the esophagus was found in one case. Abdominal ultrasonography was conducted in 29 cases, of which 13 were in accordance with the diagnosis, with a diagnostic coincidence rate of 44.8%. The diagnostic coincidence rate of SBCE with the MD scan and abdominal ultrasound determined by the chi-square test was χ^2^ = 9.1 > 5.99 (P = 0.05, v = 2); P < 0.05. This suggests that the diagnostic coincidence rate of SBCE was higher than that of MD scan and abdominal ultrasound (Table 7). Twenty-four cases underwent both SBCE and technetium-99 scans, of which 22 cases were diagnosed with the combination of the examinations, giving a diagnostic coincidence rate of 91.7%. MD was highly suspected, but the technetium-99 scan was negative and SBCE was positive in eight cases.
MD is one of the most common congenital gastrointestinal malformations ^[3]^, which is characterized by a true diverticulum formed by the incomplete occlusion of the yolk tube (umbilical mesentery) during embryonic development. Its prevalence rate in the general population is between 0.3% and 2.9% ^[2]^. Most cases of MD may be asymptomatic for life, sometimes only being found during abdominal surgery, while it may also involve serious complications, such as massive gastrointestinal bleeding or perforation ^[4]^.
SBCE plays an important role in the evaluation of small bowel disease. At present, SBCE has become a first-line examination method used for obscure gastrointestinal bleeding ^[5]^. In this study, SBCE, abdominal ultrasound, gastroduodenoscopy, and the radionuclide technetium-99 scan were used to diagnose MD. The results obtained by both operation and pathology were taken as the gold standard. Twenty-four children were diagnosed by SBCE examination, and the diagnostic coincidence rate was 82.8%, which is similar to a study by Slobodan et al. ^[6]^ and greatly exceeded that of either the radionuclide technetium-99 scan (58.3%) or abdominal ultrasound (44.8%). Combined examination by the use of SBCE and the technetium-99 scan resulted in a diagnosis rate of 91.7% (22/24). In eight cases, MD was highly suspected, but the technetium-99 scan was negative and SBCE was positive. The above results reveal that SBCE demonstrates great utility in improving the correct diagnosis rate of MD. Gastroduodenoscopy has no advantage in the diagnosis of MD, but it plays an auxiliary role in excluding upper gastrointestinal bleeding.
SBCE can be used to detect MD, though its positive results may be modulated by the patient’s age, intestinal cleanliness, and capsule retention in the diverticulum ^[7]^. In our study, sex, age, and capsule intake mode had no significant effect on the diagnostic coincidence rate of SBCE, but the effect of bowel preparation on the diagnostic coincidence rate of SBCE was statistically significant, which means that different bowel preparation effects have an impact on SBCE diagnosis.
Our results showed that gastrointestinal bleeding; including bloody stool and melena; was the most common manifestation in children with MD, which was also found to be accompanied by abdominal pain, dizziness or syncope, and vomiting. Among the 29 children who underwent operation, one case had protruding lesions and 24 cases had double-lumen changes, which was similar to the findings of several other existing case reports ^[8–11]^. Thus, double-lumen signs are the main form of MD change identified by SBCE and are highly specific for the diagnosis of MD.
In conclusion, SBCE has high accuracy in the diagnosis of MD in children, especially when combined with the technetium-99 scan. These findings provide supporting evidence in favor of its use in clinical applications.
We thank the medical staff of the Department of Gastroenterology of the Children’s Hospital of Chongqing Medical University for their contributions to the informed consent of the children’s families, the implementation of capsule endoscopy and the collection of medical history. Thank you for your support and help in our collecting case data. We thank all the patients who volunteered to participate.
L.L.: data analysis, manuscript drafting, editing. Y.W.: research conception and design; analyze the endoscopic image of SBCE and report. A manuscript editor of important content. X.Z.: collection of clinical cases. Y.C.: collection of clinical cases. J.L.: analysis of endoscopic images of SBCE and report.
There are no conflicts of interest.