If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Reprint requests: Zhuan Liao, National Clinical Research Center for Digestive Diseases, Department of Gastroenterology, Changhai Hospital, Naval Medical University, 168 Changhai Rd, Shanghai 200433, China.
Affiliations
National Clinical Research Center for Digestive Diseases, Department of Gastroenterology, Changhai Hospital, Naval Medical University, Shanghai, China
Compared with conventional endoscopy, magnetically controlled capsule gastroscopy (MCCG) can be further optimized in gastric examination time and complete visualization of upper GI (UGI) mucosa. The second-generation MCCG (MCCG-2) was developed with higher image resolution and adaptive frame rate, and we aimed to evaluate its clinical availability for UGI examination in this study.
Methods
Consecutive patients undergoing MCCG examination between May to June 2019 were prospectively enrolled and randomized to swallow the first-generation MCCG (MCCG-1) or MCCG-2 in a 1:1 ratio. The main outcomes included visualization of the esophagus and duodenum, operation-related parameters, image quality, maneuverability, detection of lesions, and safety evaluation.
Results
Eighty patients were enrolled. In the MCCG-2 group, frames captured for esophageal mucosa and Z-line were 171.00 and 2.00, significantly increased from those in the MCCG-1 group (97.00 [P = .002] and .00 [P = .028], respectively). The gastric examination time was shortened from 7.78 ± .97 minutes to 5.27 ± .74 minutes (P < .001), with the total running time of the capsule extended from 702.83 minutes to 1001.99 minutes (P < .001). MCCG-2 also greatly improved the image quality (P < .001) and maneuverability (P < .01). No statistical difference existed in the detection of lesions between the 2 groups, and no adverse events occurred.
Conclusions
MCCG-2 showed better performance in mucosal visualization, examination duration, and maneuverability, making better diagnosis of UGI diseases a possibility. (Clinical trial registration number: NCT 03977935.)
Magnetically controlled capsule gastroscopy (MCCG) provides good visualization of the stomach and is highly accepted because of the characteristics of painlessness, noninvasiveness, and equally favorable diagnostic accuracy as conventional endoscopy.
However, the upper GI (UGI) tract under capsule endoscopy (CE) continues to present challenges, including rapid transit through the esophagus and duodenum
compared with conventional endoscopy. Technical improvements in frame rate, field of view, dual camera, image resolution, and battery life have been developed to optimize the clinical application of CE, and some turned out to be effective.
Therefore, a second-generation MCCG (MCCG-2) highlighted with a higher and adaptive frame rate of 8 frames per second (fps), better image resolution of 720 × 720 pixels, wider field of view of 150 degrees, extended battery life of more than 12 hours, and antijamming wireless data transmission has been developed. Hence, this pilot study was conducted to determine whether MCCG-2 can further optimize the visualization of the UGI tract, thus resulting in a better diagnosis of UGI diseases.
Methods
Study design
This pilot study was a prospective, single-centered, blinded, randomized controlled study approved by the institutional review board of Shanghai Changhai Hospital and registered at clinicaltrials.gov (NCT03977935). Written informed consent was obtained from each enrolled subject.
Patients
From May 2019 to June 2019, 80 consecutive patients aged 18 to 80 years with or without abdominal complaints referring for MCCG examination were prospectively enrolled and randomly allocated into the MCCG-1 group or MCCG-2 group in a 1:1 ratio. Patients with any of the following contraindications for MCCG were excluded: pacemakers or electromedical devices implanted, which are incompatible with magnetic field; suspected or known GI stenosis, obstruction, or other known risk factors for capsule retention; scheduled magnetic resonance imaging examination before excretion of capsule; pregnancy or suspected pregnancy; and any other contraindications as determined by endoscopists.
Study intervention
The MCCG, a robotic magnetic capsule guidance system, was provided by Ankon Technologies Co Ltd (Shanghai, China). The MCCG system consisted of a guidance magnet robot, a CE, a data recorder, and a computer workstation with software for real-time view and 2 joysticks for capsule orientation control. The guidance magnet robot was of C-arm type with 5 degrees of freedom: 2 rotational (horizontal and vertical directions) and 3 translational (forward/backward, up/down, left/right).
After a standardized GI preparation regimen for MCCG, patients were placed in the left lateral decubitus position. They swallowed the MCCG-1 or MCCG-2 with a small amount of water according to a random number table with the help of nurses at the digestive endoscopic center. When the capsule reached the stomach, it was lifted away from the posterior wall, rotated, and advanced to the fundus and cardiac regions, followed by the gastric body, angulus, antrum, and the pylorus. During this procedure, position changes such as supine, prone, left, and right lateral were also helpful in achieving clear observation and smooth transition. A standardized examination procedure of MCCG is available online at videogie.org.
Patients with suspected malignancy were rechecked under conventional endoscopy.
The image capture rate in the esophagus and stomach were 2 fps or 6 fps in the MCCG-1 or MCCG-2 group, and standardized operations of MCCG were performed twice for complete gastric examination. After the capsule moved into the duodenum, patients left the hospital with the data recorder to continue with small-bowel examination. In the small bowel, MCCG-2 offered an adaptive frame rate technology that adjusted the image capture rate based on how fast the capsule was moving.
Study outcomes and definition
Basic characteristics of the enrolled patients were prospectively collected. The primary study outcome was efficacy analysis, including visualization of the esophagus and duodenum indicated by detection rate of the Z-line and duodenal papilla defined as obtaining at least 1 image of the Z-line or duodenal papilla, number of images captured for the esophagus and Z-line, circumferential visualization of the Z-line as the number of quadrants observed, and cleansing level of the Z-line as bubbles/saliva on Z-line.
Operation-related parameters included esophageal transit time, GET, gastric transit time, small-bowel transit time, and total running time. Image quality, maneuverability, and detection of lesions were also prospectively documented. All procedure-related adverse events were closely recorded.
GET was defined as the time for examination of gastric primary anatomic landmarks twice. Image quality ranged from 1 to 10 denoting the worst to the best.
Maneuverability was classified as fluency (the response to operation and video effect), stability (the ability of holding the capsule at 1 position for at least 1 minute), and comfortableness (the operator’s fatigue degree during the examination), and each index was graded from 1 to 5, with 1 as the worst and 5 as the best.
All examinations and maneuverability evaluation were performed by an endoscopist (W.Z.) with experience of more than 1000 cases of MCCG operation. Two other endoscopists blinded to the grouping independently evaluated the image quality with a mean value as the final score, and a third endoscopist made the final judgement when a discrepancy of more than 3 occurred. The randomization schedule was generated by the investigator using a random number table, so that patients and endoscopists were all blinded to the treatment protocol assigned.
Statistical analysis
As a pilot study to evaluate the clinical utility of MCCG-2, sample size was not calculated for this study. Quantitative data were summarized with parametric statistics, mean and standard deviation, or nonparametric statistics, median, and interquartile range (IQR), whereas categorical data were presented as frequency (percentage). Data with a normal distribution were compared using parametric analysis, and non-normally distributed data were compared using nonparametric statistical analyses. Categorical variables were analyzed with the χ2 exact test, and quantitative data were analyzed using the Mann-Whitney Wilcoxon test with a final 2-sided P < .05 indicating statistical difference. SPSS 13.0 software (IBM Corp, Armonk, NY, USA) was used.
Results
Patient characteristics and safety analysis
Eighty subjects (40 in each group; mean age, 46 years [range, 24-78]) were enrolled. Characteristics of the study population are shown in Table 1. No statistical difference of baseline characteristics or indications for MCCG existed between the 2 groups. No capsule retention or adverse events were observed in all patients during the 2-week follow-up period.
Table 1Demographics and indications for MCCG examination of enrolled patients in 2 groups
As illustrated in Table 2 and Figure 1, MCCG-2 greatly improved the visualization of the esophagus as the frames captured for esophageal mucosa and Z-line were 171.00 (IQR, 81.25-409.25) and 2.00 (IQR, 0-13.00), a significant increase from those in the MCCG-1 group with 97.00 (IQR, 42.00-160.25) and .00 (IQR, 0-2.75; P = .002 and P = .028, respectively). Circumferential viewing of the Z-line was also greatly improved with complete observation of the Z-line in 5 patients (12.5%) under MCCG-2 compared with none under MCCG-1, and at least 2 quadrants of the Z-line were successfully viewed in 12 patients (30.0%) in the MCCG-2 group, greatly improved from only 5 patients (12.5%) in the MCCG-1 group (P = .043). Although there was no significant difference in the detection of the Z-line and duodenal papilla, an obvious upward trend still existed in MCCG-2 group.
Table 2Efficacy analysis between first-generation and second-generation MCCG
Characteristics
MCCG-1 group (n = 40)
MCCG-2 group (n = 40)
P value
Visualization of the esophagus
Frames of esophagus
97.00 (42.00-160.25)
171.00 (81.25-409.25)
.002
Frames of Z-line
0 (0-2.75)
2.00 (0-13.00)
.028
Detection of Z-line
17 (42.5)
24 (60)
.117
Circumferential viewing of the Z-line
.043
100%
0 (.0)
5 (12.5)
≥50%
5 (12.5)
7 (17.5)
<50%
12 (30.0)
12 (30.0)
0%
23 (57.5)
16 (40)
Cleansing level of the Z-line area
.950
No bubbles
13 (32.5)
11 (27.5)
A few bubbles
18 (45.0)
22 (55.0)
Increased amount of bubbles
7 (17.5)
5 (12.5)
Severe bubbles
2 (5.0)
2 (5.0)
Detection of duodenal papilla
3 (7.5)
6 (15)
.479
Examination-related parameters
Esophageal transit time, s
59.50 (24.75-82.75)
44.50 (23.0-107.5)
.904
Mean GET, min
7.78 ± .97
5.27 ± .74
<.001
Gastric transit time, min
53.09 (22.88-90.84)
48.35 (19.09-78.83)
.364
Small-bowel transit time, min
327.60 (260.73-394.55)
297.22 (239.68-369.27)
.349
Total running time, min
702.83 (648.47-856.10)
1001.99 (774.42-1207.38)
<.001
Image quality score
7.90 ± .61
8.63 ± .57
<.001
Maneuverability score
Fluency
3.04 ± .51
4.59 ± .42
<.001
Comfortableness
3.16 ± .45
4.51 ± .43
<.001
Stability
3.84 ± .55
4.17 ± .50
.009
Total score
10.05 ± 1.04
13.28 ± .90
<.001
Values are median (range), n (%), or mean ± standard deviation.
Figure 1Comparison of representative images under magnetically controlled capsule gastroscopy (MCCG) and conventional endoscopy. Complete viewing of the Z-line under MCCG-1 (A, B) and MCCG-2 (C, D). Close-up image of midesophageal cancer and gastric greater curvature cancer under conventional endoscopy (E, F) and MCCG-2 (G, H).
Figure 1Comparison of representative images under magnetically controlled capsule gastroscopy (MCCG) and conventional endoscopy. Complete viewing of the Z-line under MCCG-1 (A, B) and MCCG-2 (C, D). Close-up image of midesophageal cancer and gastric greater curvature cancer under conventional endoscopy (E, F) and MCCG-2 (G, H).
For operation-related parameters, MCCG-2 dramatically shortened the GET from 7.78 ± .97 to 5.27 ± .74 minutes and extended the total running time from 702.83 (IQR, 648.47-856.10) to 1001.99 minutes (IQR, 774.42-1207.38; P < .001). Esophageal transit time, gastric transit time, and small-bowel transit time were similar between the 2 groups.
Image quality and maneuverability analysis
The mean score of the image quality was 8.63 ± .57 in the MCCG-2 group and 7.90 ± .61 in the MCCG-1 group; thus, the image quality was enhanced greatly using MCCG-2 (P < .001). As displayed in Video 1 (available online at www.giejournal.org) and Figure 2, anatomic landmarks of the stomach can be clearly observed using MCCG-2. Operators in MCCG-2 group definitely had a better experience elucidated by the improved maneuverability-related parameters: 4.59 ± .42 versus 3.04 ± .51 for fluency, 4.51 ± .43 versus 3.16 ± .45 for comfortableness, and 4.17 ± .50 versus 3.84 ± .55 for stability, respectively (all P < .01).
Figure 2Representative images of anatomic landmarks under second-generation magnetically controlled capsule gastroscopy. A, Middle esophagus. B, Cardiac and fundus. C and D, Gastric body. E, Angulus. F, Antrum. G, Pylorus. H, Duodenal papilla.
Figure 2Representative images of anatomic landmarks under second-generation magnetically controlled capsule gastroscopy. A, Middle esophagus. B, Cardiac and fundus. C and D, Gastric body. E, Angulus. F, Antrum. G, Pylorus. H, Duodenal papilla.
Regarding positive findings, 25 (62.5%) and 24 (60.0%) patients were diagnosed with inflammation and 12 (30.0%) and 13 (32.5%) focal lesions were detected in the MCCG-1 and MCCG-2 groups, respectively. There was no significant difference in detection of inflammation and focal lesions (Table 3). Of note, 1 esophageal cancer and 1 gastric cancer were detected in the MCCG-2 group, and both were confirmed by subsequent conventional endoscopy within 24 hours (Fig. 1).
MCCG is now attracting more attention and especially has a potential role in the GI disease screening modality because of its noninvasiveness and satisfactory diagnostic accuracy. However, there is still a long way to go before replacing conventional gastroscopy as a diagnostic modality, which remains the criterion standard diagnostic tool in the foregut.
MCCG-2 also shows superiority especially in control method, image resolution, frame rate, battery life, and stable and powerful magnetic force. In addition, antijamming technology promises effective, steady, and smooth wireless data transmission, given the capability of spatial signal processing and error correction. The EndoCapsule system (Olympus Medical Systems Corporation, Tokyo, Japan, and Siemens Healthcare AG, Erlangen, Germany) under magnetic resonance imaging guidance has a dual-camera capsule to ensure complete visualization of gastric mucosa.
MCCG provides complete visualization of the stomach, and the dual camera may require extra consumption of battery. In addition, a novel magnetic-assisted CE system consisting of a cable CE and hand-held magnetic field navigator was reported. Despite satisfactory maneuverability and visualization completeness in the esophagus, stomach, and duodenum, it may increase patient discomfort during the procedure compared with wireless CE.
In our study, the new upgraded MCCG-2 greatly improved the visualization of the esophagus and promoted the visualization of the duodenal papilla. Previous studies have shown the Pillcam ESO achieved satisfactory visualization in the esophagus, elucidated by a whole circumference of the Z-line of 92.5% for Pillcam UGI (Medtronic Ltd, Dublin, Ireland) with 35 fps,
However, Pillcam ESO 1 and 2 were specialized for esophageal examination, and the Pillcam UGI with a higher frame rate and dual camera can be power-consuming. The detachable string MCCG has been used and proved to be feasible, safe, and well tolerated for viewing the esophagus.
Thus, the detachable string MCCG-2 can undoubtedly balance excellent visualization of the UGI tract with complete examination of the small bowel, which is our next investigation target.
Another breakthrough was the dramatically shortened mean GET from 7.78 to 5.27 minutes using the MCCG-2, which was comparable with conventional endoscopy with an examination of about 5.00 minutes.
and this study further proposed MCCG-2 as a more-suitable modality for gastric examination with favorable diagnostic accuracy and reduced examination time. This could potentially benefit countries with gastric cancer screening programs the most by reducing the burden of the gastroscopy requirement.
The diagnostic yield of CE in the small bowel may be hampered by incomplete small-bowel examination. Studies revealed a pooled completion rate of 83.5%, with the possibility of missed diagnoses in the terminal ileum.
thus significantly promoting gastric transit time. It was suggested that the adaptive frame rate technology of the Pillcam SB3 and Pillcam Colon 2 may enhance diagnostic yield.
MCCG-1 can provide a traditional image capture rate of 2 fps, whereas MCCG-2 also introduced an adaptive technology to combat CE limitations during periods of rapid transit in the small bowel. Thus, an extended battery life and adaptive frame rate in the small bowel further guarantee complete examination and diagnostic accuracy of the small bowel and even more inspection of the colon.
This trial has limitations. First, as a pilot study to access the efficacy and safety of the updated MCCG, lesion detection rate was not significantly different between the 2 groups mainly because of the small sample size, warranting further large-scale studies to test the diagnostic ability compared with conventional endoscopy. Second, the evaluation of maneuverability and image quality was somehow subjective, which may cause over-interpretation.
In conclusion, this pilot study was the first trial for efficacy and safety evaluation of the updated MCCG, proposing an optimized visualization and diagnosis of UGI tract. Also, it established the foundation for more large-scale investigations validating MCCG as a promising examination modality for the entire GI tract, especially under the introduction of deep learning–based artificial intelligence models, a hotspot introduced both in conventional endoscopy
DISCLOSURE: The following authors received research support for this study from the Chang Jiang Scholars Program by the Chinese Ministry of Education (grant Q2015190) and the “Ten Thousand Plan” National High Level Talents Special Support Plan: Z. Liao; and Shanghai Sailing Program, China: Y.-Y. Qian (no. 19YF1446700), J. Pan (no. 18YF1422800). All other authors disclosed no financial relationships.
If you would like to chat with an author of this article, you may contact Dr Liao at [email protected]
30053-5&id=gr1ad.jpg){kind=link}
30053-5&id=gr1eh.jpg){kind=link}
30053-5&id=gr2ad.jpg){kind=link}
30053-5&id=gr2eh.jpg){kind=link}