Abstract
Background and Aims:
Single balloon enteroscopy (SBE) is utilized in the evaluation of obscure gastrointestinal bleeding, but 40–50% of these patients experience continued GI blood loss, in part due to missed lesions. The utilization of a transparent cap attached to the end of the endoscope can improve mucosal visualization in other endoscopic applications, but has not yet been evaluated in SBE. The aim of this study was to evaluate the impact of a cap on the diagnostic yield of SBE.
Methods:
Consecutive adult patients scheduled for anterograde SBE for the evaluation of obscure GI bleeding were screened for inclusion from 2014 to 2017. Patients were randomized to SBE with or without a transparent cap. The primary outcome was the proportion of enteroscopies in which a P2 lesion (high potential for bleeding) was identified.
Results:
A total of 90 patients (65.7+/−12.7 years old, 47.7% female) were analyzed. There were significantly more P2 arteriovenous malformations identified in the cap group (14.8% vs. 0%, p=0.02). Additionally, the use of a cap was associated with a significantly greater depth of small bowel insertion (191.9 cm vs 156.2 cm, p=0.01). There was one perforation in the group without a cap, successfully treated with clip placement, and no adverse events in the cap group.
Conclusions:
The use of a transparent cap during SBE performed for the evaluation of obscure gastrointestinal bleeding may be an important, safe augmentation to standard SBE techniques.
Keywords: Angioectasia, Balloon Assisted Enteroscopy, Gastrointestinal Hemorrhage, Single Balloon Enteroscopy
Introduction
Obscure gastrointestinal bleeding (OGIB), defined as gastrointestinal blood loss for which no etiology can be identified on upper endoscopy, colonoscopy, and radiological evaluation, accounts for 5–10 % of gastrointestinal bleeding events, and in a majority of cases originates from the small bowel.[1, 2] The diagnosis and treatment of OGIB was revolutionized by the introduction of capsule endoscopy (CE) and balloon assisted enteroscopy (BAE) in the early 2000’s.[3–5] These modalities have allowed gastroenterologists to visualize and reach most of the small bowel. Prospective studies have demonstrated that CE and BAE have a diagnostic yield of 60–70 % in patients with OGIB.[6–8] Moreover, BAE can be used to treat small bowel lesions responsible for blood loss in 30–50% of patients with OGIB, which are typically small bowel vascular malformations.[9, 8, 6] Despite the promise of BAE, 40–50% of patients with OGIB experience continued GI blood loss following endoscopy, often despite endoscopic therapy directed at small bowel lesions.[9, 8, 6]
A potential reason for high re-bleeding rates following BAE is that pathologic lesions are missed due to incomplete visualization of small bowel mucosa. This hypothesis is supported by studies in CE literature, which indicate that significant small bowel pathology may not be visualized on initial examination in at least 20% of patients.[10] Lesions may be missed because they are located behind mucosal folds, obscured due to their position in the GI tract lumen, or obscured due to peristalsis.[11] A reason for re-bleeding beyond missed pathology is that there can be multiple culprit lesions.[12] Identifying and treating the appropriate lesion(s) in this setting can be challenging. A straightforward method of improving mucosal visualization involves attachment of a transparent cap to the end of the endoscope. Caps improve visualization by several mechanisms: they can be used to keep mucosa within range of the focal depth of the endoscope, stabilize the endoscope, push aside mucosal folds, and look over angulations.[11] During BAE, the cap may also help to anchor the scope during loop reduction, thus allowing for greater depth of insertion, allowing for identification of more pathology. Additionally, the cap can help to align a target lesion for therapy. Extensive literature exists on the use of endoscopic caps during screening colonoscopy; when a cap is used in this setting the diagnostic yield for colon polyps is increased by 5–10% when compared to conventional colonoscopy.[13, 14] Additionally, caps have been shown to be quite useful in the treatment of gastrointestinal bleeding from lesions in the stomach, as well in the resection of polyps throughout the gastrointestinal tract.[15, 16, 11] However, to date, the utility of cap assisted BAE has not been evaluated.
We hypothesized that the addition of a transparent cap to the end of the endoscope would increase the diagnostic and therapeutic yield of single balloon enteroscopy (SBE), a specific BAE system. Thus, the goal of this randomized controlled trial is to compare the diagnostic yield (DY) of cap assisted SBE (cSBE) with that of conventional SBE (sSBE) in patients with OGIB.
Methods
Study Design
We conducted a prospective, randomized controlled trial between February 2014 and May 2017 at Barnes Jewish Hospital and Barnes Jewish West County Hospital in St. Louis, Missouri. All patients were provided written informed consent before enrollment and randomization. The study was approved by the institutional review board of Washington University School of Medicine on November 11, 2013 and was registered with clinicaltrials.gov as NCT02315404.
Patients were identified from the endoscopy schedule on the day of SBE. All adult (≥18 years old) patients scheduled to undergo anterograde SBE for the evaluation of OGIB were screened for inclusion in the study. Screened patients included inpatients and outpatients. The exclusion criteria were as follows: (1) unable to provide written informed consent; (2) pregnancy or lactation; (3) suspected bowel obstruction or GI perforation; (4) unable to tolerate sedation or general anesthesia due to medical co-morbidities; (5) uncorrected coagulopathy (platelet count <50,000, international normalized ratio (INR)> 2, partial thromboplastin time (PTT)> 2x upper limit of normal); (6) patient undergoing retrograde SBE.
One of the investigators discussed the study with potential participants in the pre-endoscopy area. A screening questionnaire was administered by a member of the research team to identity eligible patients. Those subjects who qualified for the study based on inclusion/ exclusion criteria were invited to participate. Those patients who wished to participate were consented in the pre-endoscopy area. Subsequently, patients were randomized to a group and proceeded to SBE with or without cap.
We collected categorical and quantitative demographic variables and endoscopy procedural details including procedure duration, diagnostic findings, and procedure complications. Pathology results were recorded and the electronic medical record was reviewed to extract any relevant clinical information.
Randomization
The study research assistant with no clinical involvement in the study performed the randomization assignment. Randomization was conducted after enrollment. Patients were randomized using Research Randomizer, a web based random number generator.[17]
Pre-procedure evaluation
There was no standardized preprocedure workup. All patients had previous standard endoscopic evaluation. The majority, but not all patients, had a CE study prior to SBE, which idenfitified suspicous pathology that could likely be reached with SBE. In the minority of patients that did not have a CE performed prior to SBE, pathology was idenfitied on CT or MRI, or SBE was performed with a high clinical suspicion for a lesion in the small bowel distal to the area examined with standard antegrade endoscopy. Only patients undergoing antegrade SBE were included in this study. Patients were scheduled for antegrade SBE rather than retrograde SBE due to findings on previous CE, imaging, or due clinical suspicion for a lesion in the proximal GI tract.
Intervention
Eligible patients were randomized to undergo balloon assisted enteroscopy using the single balloon enteroscopy system with the Olympus SIF-180 enteroscope/ overtube with or without a transparent cap. For the cSBE group, a transparent cap with a diameter of 11.35 mm (D-201–10704; Olympus Medical Systems) was attached to the distal end of the enteroscope to protrude 4.0 mm ahead of the tip of the enteroscope (Supplemental Figure 1). All procedures were performed by one of the eight faculty experienced in SBE or by an advanced endoscopy fellow under direct supervision. Given the nature of the study, blinding was not possible.
Procedures were performed at the endoscopy laboratories at Barnes Jewish Hospital or at the endoscopy center at Barnes Jewish West County Hospital. Procedures were performed under deep sedation or general anesthesia, administered at the discretion of the attending anesthesiologist. Carbon dioxide was used for insufflation. Glucagon, 0.5 mg was administered at time of anesthesia induction to paralyze the small bowel and improve visualization, with additional doses of glucagon given at the discretion of the attending gastroenterologist in 0.25 mg aliquots. Changes in patient position and use of abdominal pressure were permitted at the discretion of the attending gastroenterologist. The point of maximum insertion into the small bowel was defined by failure to advance the enteroscope after at least 3 consecutive attempts.
Outcomes
The primary outcome was the DY of the cSBE group compared with the sSBE group, with DY defined as proportion of enteroscopies in which clinically significant findings were identified. Clinically significant findings were defined as P2 lesion (lesion considered to have high potential for bleeding, such as typical angiomata, large ulceration, tumor or varices) (Supplemental Figures 4–5).[18] A secondary outcome was the proportion of enteroscopies in which any abnormality was detected.[18]
Abnormalities were classified as the following:
P0: lesion with no bleeding potential, including visible submucosal veins, diverticula without the presence of blood or nodules without mucosal break (Supplemental Figure 2)
P1: lesion considered to have uncertain bleeding potential, such as a red spot on the intestinal mucosa or small erosions (Supplemental Figure 3).
P2: lesion considered to have high potential for bleeding; such as typical angiomata, large ulceration, tumor or varices (Supplemental Figures 4–5).
Additional secondary outcomes included the proportion of enteroscopies in which any clinically significant (P2) arteriovenous malformation (AVM) was found and the proportion of enteroscopy in which any AVM was found. We also analyzed the therapeutic yield (TY) of cSBE compared to sSBE. TY was defined as proportion of enteroscopies in which a therapeutic intervention was undertaken. Interventions included in this calculation were polypectomy, argon plasma coagulation (APC), bipolar coagulation, dilation of strictures, and endoscopic clipping. Biopsy was considered to be a therapeutic intervention if histopathology results lead to the initiation of a medical or surgical therapy (i.e: resection of mass, medical therapy for Crohn’s disease). Depth of small bowel insertion was compared between groups and calculated using the method of Efthymiou.[19] Here the number of complete folds (valvulae conniventes) was counted during withdrawal of the endoscope; with the assumption that the distance between folds is approximately 0.9 centimeters (cm). Additional secondary outcomes included total procedure duration in minutes (min), withdrawal time, defined as time from beginning of enteroscope withdrawal from the maximum insertion point to removal of the enteroscope from the patient (in minutes), number of pathologic lesions identified per patient, and procedure related adverse events.
Sample Size Estimation
Based on prior studies of second look small bowel endoscopy and studies of cap assisted colonoscopy we assumed that using a cap during SBE will increase the DY by 20%.[10, 11] With this assumption in mind, and requiring a power of 80% and alpha of 0.05 (two tailed) we calculated that we would need a total of 212 patients (106 per group). We planned to enroll this sample size over three years given that our endoscopy unit performs >90 SBE procedures per year.
Statistical Analysis
The categorical variables were summarized using frequencies and percents, while the quantitative variables were summarized using means and standard deviations. The chi square test was used for categorical variables while student’s t test was used for quantitative variables to test the difference between the study groups. Additionally, multivariable logistic regression analysis was used in order to identify factors associated with success in detecting small bowel pathology. Continuous variables with a non-normal distribution were compared using the Wilcoxon rank-sum test. A two sided p value of < 0.05 was considered significant. Statistical analyses were performed using IBM SPSS v 25.0 (Armonk, NY).
RESULTS
Participants and baseline characteristics
93 patients presented to our endoscopy centers and were assessed for eligibility. One patient was excluded due to uncorrected coagulopathy; this patient did not undergo SBE. One patient declined to participate. 91 patients were randomized; 55 to cSBE and 36 to sSBE. One patient was excluded from the study after randomization due to failure to place the cap after randomization. A total of 54 patients were analyzed in the cSBE arm and 36 in the sSBE arm (Figure 1).
Figure 1.
Study Flow Diagram
The estimated sample size to detect a significant difference in our primary outcome was 212. However, we were unable to enroll this sample size in the proposed three year enrollment period. Based on our average enrollment rate, we would need to continue the study for 7 more years. The last patient was enrolled in May, 2017 and we stopped enrollment due to lower than expected procedure volume.
Baseline demographics and characteristics are summarized in Table 1. Overall, the patients’ average age was 65.7+/−12.7 years old. 47.7% of the patients were female. More patients underwent procedures for overt bleeding in the sSBE group (22.2%) than in the cSBE group (3.7%) (p=0.01). Additionally, more patients were on antiplatelet agents (> than aspirin 81 mg) or anticoagulants in the cSBE group (33.3%) than in the sSBE group (13.9 %) (p=0.049). Other demographics, comorbid medical conditions, and pre-procedure laboratory values were similar between the groups. Preprocedure diagnostic workup was similar between groups. There was no difference in the rate of fellow participation, blood transfusions required prior to procedures, or iron supplementation between groups.
Table 1.
Baseline Characteristics
| SBE with cap (n=54) | SBE without Cap (n=36) | P value | |
|---|---|---|---|
| Age years ± SD | 66.7±12.1 | 64.3±13.2 | 0.40 |
| Gender n female (%) | 22(40.7) | 21(58.3) | 0.13 |
| Body Mass Index ±SD | 31.5±7.1 | 31.1±6.3 | 0.80 |
| Outpatient n (%) | 47(87.0) | 28(77.8) | 0.26 |
| Occult Bleeding n (%) | 52(96.3) | 28(77.8) | 0.01* |
| Fellow n (%) | 12(22.2) | 8(23.5) | 1.00 |
| Co-morbidities n (%) | |||
| Coronary artery disease n (%) | 25(46.3) | 14(38.9) | 0.52 |
| Cerebrovascular accident n (%) | 5(9.3) | 3(8.3) | 1.00 |
| Congestive heart failure n (%) | 5(9.3) | 8(22.2) | 0.13 |
| Valvular disease n (%) | 5(9.3) | 3(8.3) | 1.00 |
| Diabetes mellitus n (%) | 26(48.1) | 16(44.4) | 0.83 |
| Chronic kidney disease n (%) | 8(14.8) | 6(16.7) | 1.00 |
| Cirrhosis n (%) | 2(3.7) | 3(8.3) | 0.39 |
| Anti-platelet/Anticoagulation n (%) | 18(33.3) | 5(13.9) | 0.049* |
| Blood Transfusion n (%) | 38(70.4) | 23(63.9) | 0.65 |
| Prior Capsule n (%) | 46(85.2) | 34(94.4) | 0.31 |
| Prior CT or MRI n (%) | 12(22.2) | 6(17.6) | 0.60 |
| Pre-Procedure Labs | |||
| Creatinine mg/dl mean ± SD | 1.1± 0.4 | 1.1± 0.7 | 0.59 |
| INR mean ± SD | 1.2± 0.2 | 1.1± 0.1 | 0.15 |
| Hgb g/dl mean ± SD | 9.6± 1.8 | 9.5± 1.5 | 0.77 |
| Platelet k/cumm mean ± SD | 258.6± 96.5 | 236.5± 128.4 | 0.46 |
p<0.05
Abbreviations: CT, computed tomography; MRI, magnetic resonance imaging; SBE, single balloon enteroscopy; SD, standard deviation.
Diagnostic yield
In the cSBE group, there were more procedures in which any type of lesion was found compared to the sSBE group (75.9% vs. 55.6%, p=0.04). There were significantly more P2 AVMs identified in the cSBE group than the sSBE group (14.8% vs. 0%, p=0.02) (Table 2). For our primary outcome, there was a nonsignificant trend towards greater DY of identifying clinically significant (P2) lesions in the cSBE group (20.4%) compared with the sSBE group (8.3%) (p=0.12) (Table 2). There was no difference in the proportion of studies in which any AVM was found (38.9% vs. 25.0%, p=0.23) (Table 2).
Table 2.
Diagnostic Yield
| SBE with cap (n=54) | SBE without Cap (n=36) | P Value | |
|---|---|---|---|
| P2 lesion n (%) | 11(20.4) | 3(8.3) | 0.12 |
| Any lesion n (%) | 41(75.9) | 20(55.6) | 0.04* |
| P2 AVM n (%) | 8(14.8) | 0(0.0) | 0.02* |
| Any AVM n (%) | 21(38.9) | 9(25.0) | 0.23 |
| Number of lesions per procedure mean ± SD | 2.7 ± 5.9 | 1.3 ± 3.0 | 0.15 |
| Number of AVMS per procedure mean ± SD | 2.6 ± 6.0 | 1.1 ± 3.0 | 0.12 |
p<0.05
Abbreviations: AVM, arteriovenous malformation; SBE, single balloon enteroscopy; SD, standard deviation.
Secondary outcomes
There was no difference in TY in the small bowel between the cSBE or sSBE groups (38.9% vs 32.4%, p=0.55) or in any part of the GI tract (51.9% vs. 50%, p=0.87). APC was used in 100% of the cases where treatment was performed in the small bowel. Additional treatment with a clip was performed in 3 (5.6%) cases in the cSBE group and in 2 (5.6%) cases in the sSBE group. There was no difference in treatment success in the small bowel (38.9% vs 32.4%, p=0.55) (Table 3).
Table 3.
Therapeutic Yield
| SBE with cap (n=54) | SBE without Cap (n=36) | P Value | |
|---|---|---|---|
| SB treatment n (%) | 21(38.9) | 11(32.4) | 0.55 |
| Any treatment n (%) | 28(51.9) | 17(47.2) | 0.83 |
| SB treatment success n (%) | 21(38.9) | 11(32.4) | 0.55 |
Abbreviations: SBE, single balloon enteroscopy; SB, small bowel
There was a non-significant increase in mean number of lesions per procedure identified in the cSBE group compared with sSBE (2.7± 5.9 vs. 1.3± 3.0, p=0.15) (Table 2). Additionally, there was a similar trend towards an increase in mean number of AVMS identified per procedure in the cSBE group compared with sSBE (2.6± 6.0 vs. 1.1± 3.0, p=0.12) (Table 2).
The use of a cap was associated with a significantly greater depth of small bowel insertion (191.9± 50.2 cm vs. 156.2± 78.2 cm, p=0.01) (Table 4). There was a trend towards shorter overall procedure time in the cSBE group (34.6± 12.7 min vs. 38.2± 11.4 min, p=0.18) (Table 4). There was no difference in withdrawal time between the groups (11.2± 7.9 min vs. 12.6± 6.9 min, p=0.41) (Table 4).
Table 4.
Secondary Outcomes
| SBE with cap (n=54) | SBE without Cap (n=36) | P Value | |
|---|---|---|---|
| Insertion depth in cm mean ±SD | 191.9± 50.2 | 156.2± 78.2 | 0.01* |
| Total procedure time in min mean ±SD | 34.6± 12.7 | 38.2± 11.4 | 0.18 |
| Withdrawal time in min mean ±SD | 11.2± 7.9 | 12.6± 6.9 | 0.41 |
p<0.05
Abbreviations: SBE, single balloon enteroscopy; SD, standard deviation.
There was one perforation in the proximal jejunum in the sSBE group, treated successfully with clip placement. In this case, active bleeding from an AVM was successfully treated with APC prior to recognition of the perforation. A fellow was involved in this case. Two procedures were aborted for looping in the stomach and small bowel in the sSBE group; the proximal small bowel was reached in each. A fellow was involved in one of the aborted procedures and no active bleeding was seen in either of these cases. There were no procedural complications in the cSBE group (p= 0.03 for rate of adverse events for cSBE vs. sSBE) (Table 5).
Table 5.
Adverse Events
| SBE with cap, n=54 | SBE without Cap, n=36 | P Value | |
|---|---|---|---|
| Any Adverse Event n (%) | 0(0.0) | 3(8.3) | 0.03* |
| Perforation n (%) | 0(0.0) | 1(2.8) | 0.4 |
| Aborted Procedure n (%) | 0(0.0) | 2(5.5) | 0.16 |
p<0.05
Abbreviations: SBE, single balloon enteroscopy; SD, standard deviation.
Multivariate Analysis
We used multivariate logistic regression to identify predictors of any lesion found on SBE. Input variables were cap use, indication (overt or occult), anticoagulation use, and insertion depth in cm. The use of a cap was an independent predictor of identifying any lesion during SBE with an odds ratio (OR) of 2.91 (1.21, 8.58, p=0.045). Greater insertion depth also predicted detecting any lesion with an OR of 1.008 per cm (1.001, 1.016, p=0.035). Anticoagulation use and overt bleeding did not predict increased DY (Table 6). There were no significant independent predictors of identifying any P2 lesion, however the OR for cap use in predicting P2 lesions was 3.39 (0.71, 16.17, p=0.13). Multivariate analysis could not be performed with P2 AVMs as the outcome, as there were no events in the sSBE group.
Table 6.
Multivariate logistic regression to identify predictors of finding any lesion
| Variable | OR | 95% CI | P Value |
|---|---|---|---|
| Cap Use | 2.91 | 1.21, 8.58 | 0.045* |
| Overt Bleeding | 3.52 | 0.58, 21.26 | 0.17 |
| Anticoagulation use | 0.43 | 0.14, 1.33 | 0.14 |
| Insertion Depth (CM) | 1.008 | 1.001, 1.016 | 0.035* |
Abbreviations: CI, confidence interval; CM, centimeter; OR, odds ratio
p<0.05
DISCUSSION
The results of our study show that a transparent cap placed on the distal tip of small bowel enteroscope is a useful augmentation to standard equipment, as it led to increased identification of high risk AVMs, identification of any lesion, and led to further insertion depth. On multivariate analysis, the use of a cap and greater insertion depth predicted the identification of any lesion on SBE with ORs of 2.91 (1.21, 8.58) and 1.008 (1.001, 1.016), respectively. However, there was no significant difference in our primary DY of all P2 lesions with the use of a cap. Though this difference was not statistically significant, there was a trend towards greater DY of P2 lesions in the cSBE group and it is possible that this difference would have reached statistical significance if we reached our predetermined sample size. The use of cap also led to increased yield of finding any AVM, total number of AVMs, and shorter procedure time. Although these findings did not reach statistical significance, the trend was towards improved outcomes with the use of a cap for each measure. There were no adverse events in the cSBE, compared to 3 in the sSBE group. Overall, our findings would support a benefit of using a transparent cap in SBE.
Complete inspection of the mucosa of the gastrointestinal tract is challenging in the colon and small bowel as even with meticulous technique, lesions can be located behind mucosal folds or flexures and a significant proportion of the mucosal surface can remain unexamined.[19] As such, various methods have been used to improve mucosal visualization. Transparent caps improve polyp detection, other procedural performance metrics, and patient tolerance/acceptance.[19–21] The use of transparent cap has been shown to reduce cecal intubation time, improve polyp detection and removal, and reduce pain.[19–24] Our study attempted to evaluate whether a transparent cap improved the performance of SBE and we showed improved DY and amount of small bowel examined with the use of a cap. In other modalities, the use of a cap is well tolerated by patients.[25] Transparent caps are commonly available and inexpensive as it fits standard endoscopes.[26] As such, there is little downside in the use of a cap. We did not assess for improved patient comfort or satisfaction. However, we did show that using a cap led to shorter procedure time, which is advantageous, particularly in the patients typically requiring SBE, who are typically older and have multiple comorbidities. Describing these results would likely lead to increased satisfaction among patients, although this has yet to be studied.
The mechanism by which a transparent cap improved DY yield in other endoscopic modalities has been explored. Transparent caps depress folds, allowing visualization of blind areas behind folds, which assists in finding lesions.[19, 27] Transparent caps also allow for continuously good visual fields.[26, 28] In comparison to colonoscopy, single balloon enteroscopy is technically more challenging and visualization is more difficult due to number of folds and looping. Therefore, applying the same theoretically benefits of a cap in colonoscopy makes intuitive sense in the small bowel. Our results supported this hypothesis.
We found that the insertion depth was greater in the cSBE group compared to the sSBE group. We hypothesize that use of a distal attachment facilitates greater depth of small bowel insertion by anchoring helping to anchor the scope during loop reduction, as well as allowing the scope to traverse acute turns in the small bowel lumen without looping. On multivariate analysis, greater insertion depth was associated with finding any lesion on SBE, but was not associated with finding high risk, P2 lesions. We have used one method of estimation of insertion depth using the fold counting method for its relative ease of use and validation in prior studies. [19] There other methods to estimate insertion depth, which have their own limitations and strengths, and our results may have been changed if another method were used. The method we used correlates well with at least one other method of calculating insertion depth, which is performed by estimating the number of 5 cm increments of bowel during withdrawal.[19] Insertion depth is not a direct, strict clinical outcome, but it is assumed that greater depth of insertion will increase DY and TY. [19] Some studies to date have found no increase in TY and DY, despite improvement in procedural performance measures, such as insertion depth.[29, 30] These studies were all relatively small and designed to compare procedural performance metrics such as insertion depth and examination time, not DY or TY. Therefore, the ability to make conclusions regarding the correlation of insertion depth and DY or TY is limited. It may be that greater insertion depth does lead to the identification of more pathology, as we have shown, but the increase may be in clinically insignificant pathology.
There are several limitation of this study. We did not reach of predetermined sample size due to slow enrollment. This was primarily due to lower than predicted procedure volume. Our center has many endoscopists capable of performing SBE, but the procedure is often deferred until absolutely indicated due to various reasons. The procedure is time consuming and the TY can be low relative to time requirements. In our study, the use of a cap did not increase the TY or success of therapy. If the sample size were increased it is possible that there would have been a higher rate of therapeutic success with cap. However further studies are needed.
The study could not be blinded by design. It is possible that this led to a bias on the part of the endoscopists. Specifically, with a cap present, it may be that an endoscopist would unconsciously work harder to insert to a greater depth or examine the small intestine more carefully. There are short nosed caps made out of similar material as the cap used in this study that could be used as a comparator instead of no cap. This would have reduced some of the unconscious bias mentioned above, but could have provided useful information about the utility of different cap types SBE. This type of cap was not available at our institution at the time of the study. Additionally, if another cap improved diagnostic yield compared to no cap, the sample size needed to differentiate the diagnostic yields of the two cap types would have increased significantly.
At baseline, the two groups were similar in all demographics, comorbid medical conditions, initial labs, and transfusion requirements. However, more patients in the group without a cap underwent more procedures for overt bleeding. On multivariate analysis, the odds ratio for overt bleeding in predicting the detection of any lesion was 3.52, but the confidence interval crossed 1. We would expect the DY to increase in the setting of overt bleeding, so the consequence of more cases of overt bleeding in the sSBE group would favor the null hypothesis and reduce the difference in yield between the groups. On the other hand, overt bleeding could reduce visibility leading to missed lesions, increasing the difference in DY between the cSBE and sSBE groups. None of the patients in the study had massive, active, hemodynamically significant bleeding at the time of SBE, so with careful inspection and washing, the overall effect of more cases of overt bleeding in the sSBE group would bias the results towards the null hypothesis, if anything.
The cap group also had more patients on anticoagulant or antiplatelet agents. It is unclear why this difference occurred. It is possible that greater use of these medications would lead to more lesions such as ulcers or increase the risks of bleeding from lesions, and would upstage the severity of the lesions. However, the main significant finding of greater DY of P2 AVMs, should not increase with these medications. Additionally, on multivariate analysis, anticoagulation use did not predict identification of any lesion or any P2 lesion.
The one perforation in the sSBE group occurred in the setting of overt bleeding, which may have obscured visibility, but this is the exact mechanism by which cap use should improve SBE performance. A fellow was involved in this case and in one of the two cases aborted due to looping. Fellow involvement may increase the chance of an adverse event, but the proportion of cases involving fellows was no different between the two groups.
In the end, the results of this study show that the addition of a transparent cap to the distal tip of a small bowel enteroscope improves the performance and DY of small bowel examinations for OGIB. We were able to insert the SBE to a greater depth and identify more P2 AVMs. We were unable to find a significant difference in the primary outcome between our two groups, but this may be due to our sample size. Despite this result, our data support the augmentation of standard equipment with a transparent cap because of our other findings and because the addition of a transparent cap to standard SBE equipment is a simple and cheap measure, which has been proven safe in this and in prior studies.
Supplementary Material
Supplemental Figure 1. Single balloon enteroscopy with transparent cap fitted to distal tip
Supplemental Figure 2. P0 lesion. Small bowel lymphangiectasia
Supplemental Figure 3. P1 lesion. Red spots in small bowel
Supplemental Figure 4. P2 lesion. Small bowel ulcers
Supplemental Figure 5. P2 lesion. Small bowel AVMS
Acknowledgements
We thank Osama Altayar, MD for providing assistance in data analysis
This study was partially funded through NIH/NIDDK (T32 DK007130 and P30DK052574) SH
Abbreviations:
- SBE
single balloon enteroscopy
- cSBE
single balloon enteroscopy with cap
- sSBE
single balloon enteroscopy without cap
- BMI
body mass index
- CE
capsule endoscopy
- OGIB
obscure gastrointestinal bleed
- BAE
balloon assisted enteroscopy
- DY
Diagnostic yield
- INR
international normalized ratio
- PTT
partial thromboplastin time
- AVM
arteriovenous malformation
- CM
centimeters
- Min
minutes
- APC
argon plasma coagulation
- OR
odds ratio
Footnotes
Publisher's Disclaimer: This Author Accepted Manuscript is a PDF file of an unedited peer-reviewed manuscript that has been accepted for publication but has not been copyedited or corrected. The official version of record that is published in the journal is kept up to date and so may therefore differ from this version.
Conflicts of Interest: The authors have no conflicts of interest to declare
References
- 1.Zuckerman GR, Prakash C, Askin MP, Lewis BS. AGA technical review on the evaluation and management of occult and obscure gastrointestinal bleeding. Gastroenterology 2000;118(1):201–21. [DOI] [PubMed] [Google Scholar]
- 2.Fujita M, Manabe N, Honda K, Tarumi K, Murao T, Katada S et al. Long-term outcome after double-balloon endoscopy in patients with obscure gastrointestinal bleeding. Digestion 2010;82(3):173–8. doi:000313360 [pii] 10.1159/000313360. [DOI] [PubMed] [Google Scholar]
- 3.Liu K, Kaffes AJ. Review article: the diagnosis and investigation of obscure gastrointestinal bleeding. Aliment Pharmacol Ther 2011;34(4):416–23. doi: 10.1111/j.1365-2036.2011.04744.x. [DOI] [PubMed] [Google Scholar]
- 4.Yamamoto H, Sekine Y, Sato Y, Higashizawa T, Miyata T, Iino S et al. Total enteroscopy with a nonsurgical steerable double-balloon method. Gastrointest Endosc 2001;53(2):216–20. doi:S0016510701525390 [pii]. [DOI] [PubMed] [Google Scholar]
- 5.Iddan G, Meron G, Glukhovsky A, Swain P. Wireless capsule endoscopy. Nature 2000;405(6785):417. doi: 10.1038/35013140. [DOI] [PubMed] [Google Scholar]
- 6.Kushnir VM, Tang M, Goodwin J, Hollander TG, Hovis CE, Murad FM et al. Long-term outcomes after single-balloon enteroscopy in patients with obscure gastrointestinal bleeding. Dig Dis Sci 2013;58(9):2572–9. doi: 10.1007/s10620-013-2588-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Domagk D, Mensink P, Aktas H, Lenz P, Meister T, Luegering A et al. Single-vs. double-balloon enteroscopy in small-bowel diagnostics: a randomized multicenter trial. Endoscopy 2011;43(6):472–6. doi: 10.1055/s-0030-1256247. [DOI] [PubMed] [Google Scholar]
- 8.Gerson LB. Long-term outcomes after single-balloon enteroscopy: are they any different from double-balloon enteroscopy for vascular lesions? Dig Dis Sci 2013;58(9):2441–3. doi: 10.1007/s10620-013-2759-x. [DOI] [PubMed] [Google Scholar]
- 9.Samaha E, Rahmi G, Landi B, Lorenceau-Savale C, Malamut G, Canard JM et al. Long-Term Outcome of Patients Treated With Double Balloon Enteroscopy for Small Bowel Vascular Lesions. Am J Gastroenterol 2011. doi: 10.1038/ajg.2011.325 ajg2011325 [pii]. [DOI] [PubMed] [Google Scholar]
- 10.Viazis N, Papaxoinis K, Vlachogiannakos J, Efthymiou A, Theodoropoulos I, Karamanolis DG. Is there a role for second-look capsule endoscopy in patients with obscure GI bleeding after a nondiagnostic first test? Gastrointestinal endoscopy 2009;69(4):850–6. doi: 10.1016/j.gie.2008.05.053. [DOI] [PubMed] [Google Scholar]
- 11.Sanchez-Yague A, Kaltenbach T, Yamamoto H, Anglemyer A, Inoue H, Soetikno R. The endoscopic cap that can (with videos). Gastrointestinal endoscopy 2012;76(1):169–78 e1–2. doi: 10.1016/j.gie.2012.04.447. [DOI] [PubMed] [Google Scholar]
- 12.Jeon SR, Byeon JS, Jang HJ, Park SJ, Im JP, Kim ER et al. Clinical outcome after enteroscopy for small bowel angioectasia bleeding: A Korean Associateion for the Study of Intestinal Disease (KASID) multiceter study. J Gastroenterol Hepatol 2017;32(2):388–94. doi: 10.1111/jgh.13479. [DOI] [PubMed] [Google Scholar]
- 13.Rastogi A Cap-assisted colonoscopy. Gastroenterol Clin North Am 2013;42(3):479–89. doi: 10.1016/j.gtc.2013.05.008. [DOI] [PubMed] [Google Scholar]
- 14.Westwood DA, Alexakis N, Connor SJ. Transparent cap-assisted colonoscopy versus standard adult colonoscopy: a systematic review and meta-analysis. Dis Colon Rectum 2012;55(2):218–25. doi: 10.1097/DCR.0b013e31823461ef. [DOI] [PubMed] [Google Scholar]
- 15.Yap CK, Ng HS. Cap-fitted gastroscopy improves visualization and targeting of lesions. Gastrointest Endosc 2001;53(1):93–5. doi: 10.1067/mge.2001.110453. [DOI] [PubMed] [Google Scholar]
- 16.Kim JI, Kim SS, Park S, Han J, Kim JK, Han SW et al. Endoscopic hemoclipping using a transparent cap in technically difficult cases. Endoscopy 2003;35(8):659–62. doi: 10.1055/s-2003-41512. [DOI] [PubMed] [Google Scholar]
- 17.Urbaniak GC, & Plous S. Research Randomizer (Version 4.0) http://www.randomizer.org/. 2013. http://www.randomizer.org/. Accessed June 2013.
- 18.Saurin JC, Delvaux M, Gaudin JL, Fassler I, Villarejo J, Vahedi K et al. Diagnostic value of endoscopic capsule in patients with obscure digestive bleeding: blinded comparison with video push-enteroscopy. Endoscopy 2003;35(7):576–84. doi: 10.1055/s-2003-40244. [DOI] [PubMed] [Google Scholar]
- 19.Efthymiou M, Desmond PV, Brown G, La Nauze R, Kaffes A, Chua TJ et al. SINGLE-01: a randomized, controlled trial comparing the efficacy and depth of insertion of single- and double-balloon enteroscopy by using a novel method to determine insertion depth. Gastrointest Endosc 2012;76(5):972–80. doi: 10.1016/j.gie.2012.06.033. [DOI] [PubMed] [Google Scholar]
- 20.Matsushita M, Hajiro K, Okazaki K, Takakuwa H, Tominaga M. Efficacy of total colonoscopy with a transparent cap in comparison with colonoscopy without the cap. Endoscopy 1998;30(5):444–7. doi: 10.1055/s-2007-1001305. [DOI] [PubMed] [Google Scholar]
- 21.Morgan J, Thomas K, Lee-Robichaud H, Nelson RL, Braungart S. Transparent cap colonoscopy versus standard colonoscopy to improve caecal intubation. Cochrane Database Syst Rev 2012;12:CD008211. doi: 10.1002/14651858.CD008211.pub3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Kondo S, Yamaji Y, Watabe H, Yamada A, Sugimoto T, Ohta M et al. A randomized controlled trial evaluating the usefulness of a transparent hood attached to the tip of the colonoscope. Am J Gastroenterol 2007;102(1):75–81. doi: 10.1111/j.1572-0241.2006.00897.x. [DOI] [PubMed] [Google Scholar]
- 23.Shida T, Katsuura Y, Teramoto O, Kaiho M, Takano S, Yoshidome H et al. Transparent hood attached to the colonoscope: does it really work for all types of colonoscopes? Surg Endosc 2008;22(12):2654–8. doi: 10.1007/s00464-008-9790-6. [DOI] [PubMed] [Google Scholar]
- 24.Rastogi A, Bansal A, Rao DS, Gupta N, Wani SB, Shipe T et al. Higher adenoma detection rates with cap-assisted colonoscopy: a randomised controlled trial. Gut 2012;61(3):402–8. doi: 10.1136/gutjnl-2011-300187. [DOI] [PubMed] [Google Scholar]
- 25.Lee YT, Lai LH, Hui AJ, Wong VW, Ching JY, Wong GL et al. Efficacy of cap-assisted colonoscopy in comparison with regular colonoscopy: a randomized controlled trial. Am J Gastroenterol 2009;104(1):41–6. doi: 10.1038/ajg.2008.56. [DOI] [PubMed] [Google Scholar]
- 26.Dafnis GM. Technical considerations and patient comfort in total colonoscopy with and without a transparent cap: initial experiences from a pilot study. Endoscopy 2000;32(5):381–4. doi: 10.1055/s-2000-637. [DOI] [PubMed] [Google Scholar]
- 27.Lee YT, Hui AJ, Wong VW, Hung LC, Sung JJ. Improved colonoscopy success rate with a distally attached mucosectomy cap. Endoscopy 2006;38(7):739–42. doi: 10.1055/s-2006-925238. [DOI] [PubMed] [Google Scholar]
- 28.Tada M, Inoue H, Yabata E, Okabe S, Endo M. Feasibility of the transparent cap-fitted colonoscope for screening and mucosal resection. Dis Colon Rectum 1997;40(5):618–21. [DOI] [PubMed] [Google Scholar]
- 29.Messer I, May A, Manner H, Ell C. Prospective, randomized, single-center trial comparing double-balloon enteroscopy and spiral enteroscopy in patients with suspected small-bowel disorders. Gastrointest Endosc 2013;77(2):241–9. doi: 10.1016/j.gie.2012.08.020. [DOI] [PubMed] [Google Scholar]
- 30.Moran RA, Barola S, Law JK, Amateau SK, Rolshud D, Corless E et al. A Randomized Controlled Trial Comparing the Depth of Maximal Insertion Between Anterograde Single-Balloon Versus Spiral Enteroscopy. Clin Med Insights Gastroenterol 2018;11:1179552218754881. doi: 10.1177/1179552218754881. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supplemental Figure 1. Single balloon enteroscopy with transparent cap fitted to distal tip
Supplemental Figure 2. P0 lesion. Small bowel lymphangiectasia
Supplemental Figure 3. P1 lesion. Red spots in small bowel
Supplemental Figure 4. P2 lesion. Small bowel ulcers
Supplemental Figure 5. P2 lesion. Small bowel AVMS