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
Capsule endoscopy (SBCE) has developed a relevant role in patients with established Crohn's Disease (CD). However, evaluation of the impact in clinical management has been scarce.
Aims
To evaluate therapeutic impact of SBCE in an 11-year real-life cohort of known CD patients.
Methods
Retrospective single center study including all patients with established CD submitted to SBCE procedure from 01/01/2008 to 31/12/2019. Patency capsule was used in selected patients. Small bowel mucosal inflammation was quantified using Lewis score. Therapeutic impact was defined as a change in CD-related treatment recommended based on SBCE results. Patients were assigned to four groups regarding SBCE indication: staging, flare, post-op and remission.
Results
From the 432 SBCE performed 87.5% were conclusive. Active disease was present in 63.7 of patients; 41.6% mild inflammation and 21.9% moderate-to-severe activity. A change of management was guided by SBCE in 51.3% of procedures: 199 (46.1%) escalation and 23 (5.3%) de-escalation, with significant changes in all groups. Escalation increased with disease activity: 57.8% in mild and 89.5% in moderate-to-severe disease. De-escalation was conducted in 13.9% procedures with mucosal healing and 1.1% with mild disease.
Conclusion
SBCE is a useful tool for guiding therapeutic management in CD patients both for treatment escalation and de-escalation.
Crohn's Disease (CD) is a chronic relapsing inflammatory disease that may affect any segment of the gastrointestinal tract causing inflammatory lesions of variable depth and extension. Approximately 30% of patients have isolated small bowel (SB) involvement [
] which may hinder diagnosis by ileocolonoscopy if terminal ileum is not affected. Not only is it a diagnostic challenge but also has prognostic implications since jejunal disease has been associated with stricturing phenotype and multiple surgeries [
. Although the ileocolonoscopy remains the first step in the initial diagnostic algorithm, all newly diagnosed CD patients should undergo SB assessment [
Small Bowel Capsule endoscopy (SBCE) was introduced into clinical practice in 2001, enabling a non-invasive direct visualization of the SB mucosa and has become widespread in the evaluation of both suspected and established CD, becoming the second most frequent indication for SBCE. To date, several studies have compared the diagnostic yield of SBCE with other diagnostic procedures. While it has proven its better performance over SB follow-through and computed tomography enterography [
A meta-analysis of the yield of capsule endoscopy compared to other diagnostic modalities in patients with non-stricturing small bowel Crohn's disease.
U. Kopylov, D.E. Yung, T. Engel, et al. Diagnostic yield of capsule endoscopy versus magnetic resonance enterography and small bowel contrast ultrasound in the evaluation of small bowel Crohn's disease: systematic review and meta-analysis. Dig Liver Dis. 2017;49(8):854–63. doi:10.1016/j.dld.2017.04.013
U. Kopylov, D.E. Yung, T. Engel, et al. Diagnostic yield of capsule endoscopy versus magnetic resonance enterography and small bowel contrast ultrasound in the evaluation of small bowel Crohn's disease: systematic review and meta-analysis. Dig Liver Dis. 2017;49(8):854–63. doi:10.1016/j.dld.2017.04.013
The Impact of magnetic resonance enterography and capsule endoscopy on the re-classification of disease in patients with known Crohn's disease: a prospective Israeli IBD research nucleus (IIRN) study.
Current American and European guidelines recognize SBCE diagnostic role in patients with high clinical suspicion for CD because of its high negative predictive value [
Capsule Endoscopy, magnetic resonance enterography, and small bowel ultrasound for evaluation of postoperative recurrence in Crohn's disease: systematic review and meta-analysis.
The study aimed to analyze the impact of SBCE in the management of a large unicentric cohort of patients with established CD in clinical practice. We aimed to describe the different scenarios in which SBCE was indicated and its performance in a real-life setting. Clinical benefit measured as re-staging of the disease and therapeutic modifications that triggered SBCE findings was also a main objective. In addition, we aimed to identify predictive factors for drug therapy escalation and de-escalation. Safety of SBCE was also assessed.
2. Patients and methods
2.1 Patients and study design
We included all SBCE performed from 01/01/2008 to 31/12/2019 in patients with known CD followed in the Inflammatory Bowel Diseases Unit of Complejo Hospitalario de Navarra (Pamplona, Spain). Patients with suspected CD but without a previous diagnosis based on ECCO (European Crohn's and Colitis Organization) criteria [
] were excluded. In addition, patients with suspected CD in which SBCE confirmed the diagnosis were also excluded. Demographic, clinical, laboratory and SBCE results were collected by case record review of electronic notes.
2.2 Group classification
Patients were assigned to four groups. The first one (Staging group) included patients whose indication was to assess disease extension in the small bowel (SB). The second group (Flare group) included patients with clinical symptoms, raised inflammatory markers or anemia in which a flare was suspected. The third group (Post-op) included patients with previous ileocecal resection referred to SBCE for detection of endoscopic post-operative recurrence. The fourth group (Remission group) included asymptomatic patients with no altered inflammatory markers to assess mucosal healing for evaluation of therapeutic strategy.
The first and second groups were not exclusive, and patients that would be flaring at the time that SB was assessed were included in both groups. This situation was not uncommon, especially after the initial diagnosis with ileocolonoscopy.
2.3 Definitions
A complete SBCE was defined as a capsule that reached the colon or stoma bag. We defined a conclusive SBCE as a procedure that enabled an adequate assessment for decision-making and was not followed by other diagnostic techniques. A CE was considered as retained when it persisted in the bowel after 2 weeks from ingestion or medical, endoscopic or surgical treatment was needed for extraction.
Mucosal inflammation was graded using the Lewis Score (LS) [
] using the calculator included in the RAPID® software. Normal SBCE or clinically insignificant lesions were defined as LS < 135, mild-to-moderate inflammation as LS between 135 and 790, and moderate-to-severe inflammatory activity as LS ≥ 790. Small bowel lesions were considered to have proximal location if they were located in the upper two thirds of the small bowel (first two tertiles of the SBCE) and mucosal inflammation graded using the LS.
Clinical impact was defined as a change in CD related treatment. This includes 5-ASA, steroids (first and second generation) inmunomodulators (thiopurines and methotrexate) and biologics (infliximab, adalimumab, ustekinumab, vedolizumab and certoliumab). Escalation, de-escalation, dose adjustment or referral to surgery recommended based on CE results, at the next clinic visit or within 3 months after the CE, were considered therapeutic changes.
2.4 SBCE procedure
SBCE was performed using either PillCam® SB, SB2 (Given Imaging, Covidien Ltd., Yoqneam, Israel) or PillCam SB3® (Medtronic, Minneapolis, USA). Prior to SBCE procedure, small bowel patency was assessed with a PillCam® Patency Capsule (PC) (Medtronic, Minneapolis, USA) only in those patients in which the referring physician deemed necessary based on retention risk assessment (history of SB obstruction, abdominal surgery or significant elevated inflammatory markers) [
]. An intact PC excreted in its original shape within 30 h after swallowing was considered a positive patency test and underwent SBCE. All procedures in the present study were performed as an outpatient basis.
Regarding preparation all patients were informed to follow dietary modifications with a low-fiber diet on the day before the procedure with clear liquids only in the evening and a 12 h fast. No purgatives, prokinetics or antifoaming agents were given per protocol. Patients were allowed to drink clear liquids 2 h after capsule ingestion and to eat solid food 4 h after capsule ingestion.
2.5 Statistical analysis
Continuous variables were described as the mean value and standard deviation or as median and interquartile range if data did not follow normal distribution. For comparisons between groups, t-test or Mann-Whitney U test was performed as appropriate. Categorical variables were described as proportion or percentage and compared using the Pearson χ² test or Fisher exact test as appropriate. Logistic regression was used to determine associations between clinical, biochemical and endoscopic measures of disease activity, with escalation and de-escalation of therapy. All analyses were calculated using SPSS statistical software (version 20, IBM corporation, Chicago, USA). P values less than 0.05 were considered to be statistically significant.
2.6 Ethical considerations
The present study was approved by the Clinical Research Ethics Committee of Navarra on the 26th October 2018 (Registry number 61/2018). Written informed consent was obtained in all cases. All identifiable medical information was removed and all analyses performed using anonymized data.
3. Results
3.1 Patients included and basal characteristics
A patient flow diagram is illustrated in Fig. 1. A total of 465 SBCE were requested in 333 CD patients followed in the IBD unit during the study period. The median SBCE per patient was 1.89 (IQR 1–9). In most patients (61.9%) small bowel patency was tested with a previous PC and in 33 (6.5%) SBCE was contraindicated to a deformed or delayed PC excretion. A total of 432 SBCE procedures in 305 patients were included. Most patients were integrated in the flare group (n = 226), followed by the staging group (n = 161), the remission group (n = 92) and the post-op (n = 36). Of note, 83 patients were included in both flare and staging group. Baseline characteristics and treatment at the time of SBCE are shown in Table 1.
The SBCE reached caecum and was considered complete in 384 procedures (88.9%). A total of 378 SBCE (87,5%) were considered as conclusive and allowed clinical decision-making, while the rest required additional diagnostic workup (35.1% magnetic resonance, 24.6% ileocolonoscopy, 19.3% repeated SBCE, 10.5% intestinal ultrasound and 10.5% double-balloon enteroscopy). Bowel cleanliness according to a 4-point scale was 59.7% excellent, 7.6% good, 25.9% fair and 5.1% poor. All patients receive dietary recommendations per protocol as previously described and only 2 patients (0.5%) received prokinetic agents. Small gaps were present in 26 (6%) of the studies but did not compromise the diagnostic yield as all of them were considered conclusive. No statistically significant differences between groups were found regarding conclusiveness (p = 0.53). Globally no differences related mucosal visualization quality were found, except for poor visualization category that was more frequently reported in the post-op group compared to the rest of the groups (13.9% vs 4.29, p = 0.028).
3.3 Disease activity and extent
Active disease was present in 275 (63.7%) of procedures; 180 (41.6%) presented mild inflammation and 95 (21.9%) moderate-to-severe inflammatory activity.
Disease activity (LS > 135) was found in 74.3% of SBCE in the post-op group, 72.3% of the staging group, 70.4% of the flare group and 48.4% of the remission group. Inflammatory activity based on median LS was significantly lower in the remission group (median LS= 112) compared to the other groups; flare (median LS= 229, p = 0.001), post-op (median LS= 237, p = 0.42) and staging (median LS= 331, p < 0.001). No differences were found between flare and post-op (p = 0.82), flare and staging (p = 0.52) and post-op and staging (p = 0.19)
Proximal activity was found in 131 (30.3%) patients (16.4% mild, 13.9% moderate-to-severe), which led to an upstage in the disease extent (Montreal L4 notification) in 92 patients (21.3%) since 39 patients had previously been diagnosed with proximal lesions. There was more frequent proximal disease in flare (36.3%, p = 0.005) and staging group (38.5%, p = 0.004) and significantly less frequent in the remission group (19.6%, p = 0.01).
Proximal and terminal ileum stenosis were present in 18 (4.2%) and 53 (12.9%) respectively. There was a change in 22 (5.1%) patients in the Montreal Classification Behavior (B), with no differences between groups.
The median LS was significantly higher in incomplete procedures than complete SBCE (3361 vs 225, p < 0.001). More strictures (19, 35.8%) were found in the incomplete group compared to the completed SBCE (11, 3.1%) (p < 0.001). Because LS is very dependent of strictures we compared LS excluding procedures with strictures and we also found significant differences between incomplete and complete procedures (900 vs 168, p < 0.10). Incomplete procedures rates varied according to the group: 15.5% in flare, 10.6% in staging, 5.6% in the post-op and 5.4% in the remission (p = 0.032).
3.4 Change in disease management
Overall, SBCE results guided changes in the treatment in 51.3% of procedures, 199 (46.1% of total) with an escalation of CD treatment and 23 (5.3% of total) with de-escalation. Disease activity demonstrated by SBCE was correlated with therapeutic changes. We found that 6.3% increased therapy with no significant inflammatory activity, whereas 57.8% and 89.5% increased therapy with mild or moderate-to-severe disease, respectively (p < 0.001). On the contrary, we found that de-escalation was conducted only in 13.9% of patients with no inflammation in SBCE and 1.1% of procedures with mild disease, but not if moderate-to-severe lesions were found (p < 0.001) (Fig. 2).
Fig. 2Changes in drug therapy by disease activity in SBCE.
The proportion of procedures followed by drug escalation and therapeutic de-escalation was significantly different between groups. The staging and flare group were the ones with more escalation of therapy after the SBCE, with 63.4% and 54.9% respectively. In the post-op group a 41.7% of SBCE increased medications whereas 58.3% remained unchanged. On the contrary, in the remission group, only 15.4% escalated and a significant 22% of procedures de-escalated treatment (Fig. 3). From the 199 SBCE followed by escalation of therapy, 37.2% initiated immunosuppressants, 28.1% initiated steroids (26.6% of second generation), 16.1% initiated biologics, 5% intensified biologic, 4% changed biologic agent (3.5% out-of-class and 0.5% within class) and 1 patient (0.5%) was referred to surgery. Regarding de-escalation most cases (69.5%) stopped combo therapy, 39.1% resulting in immunosuppressant monotherapy and 30.4% maintaining biologics as the only treatment. The remaining changes were withdrawing biologics (8.7%), immunosuppressants (8.7%) and 5-ASA (4.3%). The percentage of patients in each therapeutic group before and after SBCE is shown in Table 2.
The presence of proximal small bowel involvement and active disease, remarkably moderate-to-severe lesions, were independent predictors for therapy escalation (Table 3). Mucosal healing assessed by SBCE (LS < 135) was the only independent factor that predicted de-escalation of therapy (OR 6.86 [1.42–33]) unlike clinical remission by Harvey-Bradshaw index < 4 (p < 0.4), CRP < 5 mg/l (p = 0.7) and FC < 250 microg/g (p = 0.7).
Table 3Predictors of escalation of drug therapy in all patients undergoing SBCE.
We had a total of 11 SBCE retentions (2.5%). All were managed non-surgically but in one case despite an escalation of medical therapy with a course of corticosteroids, endoscopic retrieval was necessary. There was a statistically significant difference between the rate of retention in the group of SBCE with and without previous PC test (0.39% vs 5.6%, p < 0.001). The retention rate per procedure was 2,54% (1.06–4.03) and per patient was 3.6% (1.51–5.69). No other adverse events were reported.
4. Discussion
Our research has confirmed that SBCE results impact clinical decision-making in a large cohort of patients with established CD. Previous studies have focused the definition of clinical impact in prognosis [
Assessment of small bowel mucosal healing by video capsule endoscopy for the prediction of short-term and long-term risk of Crohn's disease flare: a prospective cohort study.
. However, we considered a broader definition of its influence by defining a conclusive procedure as one that provides enough information for decision-making. In a real-life setting, not only the demonstration of inflammatory lesions is valid for the approach of CD patients, but the absence of them is also equally relevant. In our cohort, only 12.5% of procedures did not provide enough information for the managing of the disease and required further diagnostic workup.
SBCE is a safe procedure for established CD patients when retention risk is properly evaluated. Initial data suggested retention rates of 3.6 and 8.2% for patients with suspected and established CD [
]. A recent update of the literature found significantly lower retention rates of 2.3 and 4.6% for suspected and established CD, which may drop to 2.9% when patency capsule was used [
Small-bowel capsule endoscopy and device-assisted enteroscopy for diagnosis and treatment of small-bowel disorders: European society of gastrointestinal endoscopy (ESGE) technical review.
] underscore the importance of carefully assessing the patient's past medical history to identify risk factors and recommend performing a patency test (either PC or small bowel enterography) in those with higher risk. Routine use of patency capsule has not been shown to reduce the risk of retention compared to selective use in patients with risk factors such as stricturing or penetrating disease phenotypes [
]. Our result of a 2.5% retention rate is consistent with the latest data, although our cohort was initiated in 2008 when higher retention rates were described. Most patients in our series had SB patency screened following a selective strategy and in only 0.39% of them the SBCE was retained, a lower rate than expected. As a result, we have confirmed that SBCE is safe in patients with established CD and that PC lowers retention risk when accurately used.
Regarding completeness of examinations, Niv et al. [
] showed that CD patients have significantly higher incomplete study rates in comparison with patients with other diagnoses. Focused on known CD patients and based on our results we can conclude that both strictures and higher inflammatory activity were found more in incomplete procedures. Furthermore, patients with more inflammatory disease irrespective of having strictures are more likely to have incomplete SBCE.
Existing literature has already assessed the therapeutic impact of CD patients undergoing SBCE. The largest retrospective cohort to date from Kopylov et al. [
] with 187 patients in a very similar setting of a tertiary center, reported a recommended therapeutic change following SBCE of 52.3% which is consistent with our results of 51.3% therapeutic modification. A similar study conducted by Santos-Antunes et al. [
. As reasonably expected, the proportion of patients in which therapeutic changes were proposed increased with disease activity, which is found throughout all previous studies.
In all considered groups the proportion of patients under the different therapeutic agents changed before and after SBCE. While Santos-Antunes failed to achieve a significant change in the post-op group maybe due to a small number of patients (n = 10), we found significant changes in that group, mainly the decrease of patients without treatment. In the staging group, the most relevant changes were the increase in immunosuppressants and combo therapy. Similarly, in the flare group, the main variation was the rise in combo therapy while also increasing both immunosuppressants and biologics. These changes are consistent with their results in which there was an increase in patients under azathioprine and anti-TNF both in staging and flare group. Furthermore, we included the remission group which was not considered in their study and where we found a significant modification after SBCE decreasing the proportion of patients under combo therapy. Therefore, in all clinical situations considered a SBCE exam impacts therapeutic management.
Effective therapies for the maintenance of remission in CD are available at the moment but involve a cumulative risk of adverse events. The so-called “exit strategies” in which a therapeutic agent is reduced or withdrawn should be approached with caution and there remains uncertainty about the risks, benefits and timing of stopping treatment when patients are in remission [
]. In SBCE studies that evaluate therapeutic impact there is paucity of data regarding this downstage. For example in the aforementioned cohorts, only 1.6% [
] (3 patients in each series) medication was withdrawn. In our data we found that 23 patients de-escalated treatment, 5.3% of the total cohort but 21.7% of the remission group. This high proportion of de-escalation can be explained because we have a higher number of asymptomatic patients in which the referring physician indicated the SBCE for finding more information to approach the therapeutic strategy (the remission group).
] recommends an appropriate re-evaluation of disease activity considering clinical, biochemical, endoscopic and/or radiological techniques before withdrawing treatment. As we can infer from our results SBCE may play a key role in this regard. Mucosal healing in SBCE was the only independent factor that predicted de-escalation of therapy in the logistic regression. A remission measured by the Harvey-Bradshaw index or inflammatory markers such as FCP or CRP within range were not independently associated with stopping therapy. In other words, the physician needs an endoscopic assessment for a proper risk evaluation and cannot rely on indirect parameters. There is growing evidence of the discordance between inflammatory activity assessed by endoscopy and biochemical markers. In a study by Mitselos et al. [
] with 50 patients with isolated small bowel disease, CD activity found an AUC for the performance of CRP of 0.69 and in other study the correlation and performance of CRP and FCP was still poor even when both parameters were combined [
]. Our findings do not diverge from existing literature and underscores the importance of assessing mucosal healing with SBCE.
To the best of our knowledge, this is the largest study assessing therapeutic impact of SBCE on the management of established CD in a routine clinical setting. This is an 11-year experience of a single tertiary hospital. Unlike multicentric studies, in our research only a limited group of gastroenterologists were involved in the management of the CD patients and followed similar criteria for the indication of the procedure thus limiting heterogenicity. The same SBCE protocol was followed in all patients, which did not vary during the study period. Our study provides consistent results with the previous data but with a larger and real-life cohort.
The most relevant limitation of our study is its retrospective design. For example, we could not evaluate the degree of confidence which the referring physician had pre- and post-SBCE. This has been explored in a small cohort of 50 patients by Hansel et al. in which a pre- and post-CE questionnaire with the managing plan with a 5-point level of confidence (LOC) was completed and a change of ≥ 2 points considered relevant. LOC changes were related to the CE in 82% (56% clinically meaningful) [
]. However, missing information was prevented by accessing a digital platform available in our institution since 2001 where all clinical data is stored including clinical notes, analysis, endoscopy and imaging studies results. Despite the abovementioned limitations, our research provides relevant insight and useful data on the clinical benefit of SBCE for the management of patients with established CD.
An emerging clinical and scientific attention has been devoted to the use of more panoramic capsule devices. While the lateral panoramic view capsule has been reported to increase small-bowel accuracy [
Panenteric capsule endoscopy identifies proximal small bowel disease guiding upstaging and treatment intensification in Crohn's disease: a European multicentre observational cohort study.
. Future studies will determine the role of the different devices in clinical practice
In conclusion, SBCE is a safe and useful tool when approaching established CD patients due to its guidance in therapeutic management in a real-life setting in all of the clinical scenarios considered. Its positive impact does not limit to treatment escalation but also helps to rationalize medications in those who can benefit from it.
A meta-analysis of the yield of capsule endoscopy compared to other diagnostic modalities in patients with non-stricturing small bowel Crohn's disease.
U. Kopylov, D.E. Yung, T. Engel, et al. Diagnostic yield of capsule endoscopy versus magnetic resonance enterography and small bowel contrast ultrasound in the evaluation of small bowel Crohn's disease: systematic review and meta-analysis. Dig Liver Dis. 2017;49(8):854–63. doi:10.1016/j.dld.2017.04.013
The Impact of magnetic resonance enterography and capsule endoscopy on the re-classification of disease in patients with known Crohn's disease: a prospective Israeli IBD research nucleus (IIRN) study.
Capsule Endoscopy, magnetic resonance enterography, and small bowel ultrasound for evaluation of postoperative recurrence in Crohn's disease: systematic review and meta-analysis.
Assessment of small bowel mucosal healing by video capsule endoscopy for the prediction of short-term and long-term risk of Crohn's disease flare: a prospective cohort study.
Small-bowel capsule endoscopy and device-assisted enteroscopy for diagnosis and treatment of small-bowel disorders: European society of gastrointestinal endoscopy (ESGE) technical review.
Panenteric capsule endoscopy identifies proximal small bowel disease guiding upstaging and treatment intensification in Crohn's disease: a European multicentre observational cohort study.
00459-X&id=gr1.jpg){kind=link}
00459-X&id=gr2.jpg){kind=link}
00459-X&id=gr3.jpg){kind=link}