Evaluating video capsule endoscopy in diabetes mellitus: transit times,preparation adequacy,and the influence of insulin and GLP-1 receptor agonist use

Abstract

Background:

Delayed gastrointestinal transit time (TT), frequently observed in diabetes mellitus (DM), may hinder bowel preparation quality and visualization in video capsule endoscopy (VCE).

Objectives:

The study aim was to compare VCE preparation adequacy and completion rates between patients with and without DM.

Design:

Retrospective analysis of a prospectively maintained database, including all consecutive VCEs performed in patients aged ⩾35 years.

Methods:

Patients were divided into two groups based on a DM diagnosis reported at admission and confirmed by medication records. Clinico-demographic characteristics, bowel preparation quality, TTs, and completion rates were compared between the groups. Logistic regression analysis was performed to identify independent risk factors for inadequate bowel preparation.

Results:

Out of 672 included VCEs, 189 (28.1%) patients had DM and 483 (71.9%) were controls. Completion rates were comparable (DM-98.4% vs controls-99.0%, p = 0.553). Gastric TT was similar (DM-37.5 ± 45.5 vs controls-34.3± 48.6 min, p = 0.420), while small bowel TT was longer (DM-261.8 ± 95.6 vs controls-244.9 ± 98.4 min, p = 0.040). DM patients had higher inadequate preparation rates (31.2% vs controls-23.4%, p = 0.040). In a multivariate analysis adjusted for age and gender, DM was not independently associated with inadequate preparation. However, DM treated with insulin (26.5% of patients with DM) was significantly associated with inadequate bowel preparation (Odds ratio (OR) 2.355, p = 0.006). Glucagon-like peptide-1 receptor agonist usage (11.8% of patients with DM patients) compared to no-DM controls achieved borderline significance as a risk factor for preparation inadequacy (OR 2.148, 95% CI 0.887–5.201, p = 0.090).

Conclusion:

VCE appears to be feasible in patients with DM, with similar completion rates and gastric TTs despite slightly prolonged small bowel TTs. However, DM treated with insulin is a significant risk factor for inadequate bowel preparation, suggesting the need for a more intensive preparation protocol within this patient group.

Keywords

bowel preparation diabetes mellitus video capsule endoscopy (VCE)

Introduction

Video capsule endoscopy (VCE) is a noninvasive tool used for diagnosis and follow-up of an array of different gastrointestinal (GI) diseases.^1–3 Since the use of VCE requires only small bowel (SB) preparation, it is usually satisfactory with nutritional preparation and preemptive fasting without the need for bowel emptying solutions.⁴ Furthermore, VCE does not require anesthesia and other invasive tools used in traditional endoscopy, therefore, VCE is largely preferred by patients.^5,6

Diabetes mellitus (DM) has been on the rise in recent years, representing a significant challenge to health systems around the world.⁷ Around 75% of DM patients will present varying GI symptoms.^8–10 Gastroparesis manifests as slow emptying of the stomach and affects about 50% of patients with DM.^11–13

Reduced GI motility is not a contraindication for VCE but can affect preparation quality and test completion within battery time.^14,15 In 2007, Triantafyllou et al. observed that gastric transit times (TT) for VCE were over three times longer in DM patients, resulting in a significantly lower completion rate: less than 70% for diabetics versus 90% for controls.¹⁵ Since then, no further studies have examined VCE safety and efficacy in DM patients or the need for specific protocols. Our study aimed to evaluate VCE quality by comparing preparation adequacy, TTs, completion rates, and retention rates between DM patients and no-DM patients.

Materials and methods

Study design

A retrospective study of a prospectively maintained database of VCE examinations performed in a single tertiary center. The study protocol was approved by the institution's human research committee (Institutional Review Board number: 0394-21-TLV), data were anonymous and informed consent for participating in the study was waived.

Study population

The study population included all VCEs performed in the ambulatory gastroenterology clinic among adult patients aged ⩾35 years between January 1, 2018, and July 31, 2021. Patients with one of the following criteria were excluded: age < 35 years, inability to swallow the capsule, significant technical malfunction of the VCE recording, history of major GI surgery, and required capsule placement by esophago-gastroduodenoscopy. Repeated VCEs within the study period were also excluded (only the first study was included) unless the causes of referral were different or new pathological findings were observed.

The cohort was divided into two groups according to patients' report of DM. Patients reporting a diagnosis of DM were defined as DM group (exposed) and those ruling out the diagnosis were defined as no-DM (non-exposed). Careful documentation of anti-diabetic medication usage was performed during VCE admission, and the drugs were divided into three groups: Insulin therapy, Glucagon-like peptide-1 receptor agonist (GLP1-RA) and other anti-diabetic medications (non-GLP1-RA, non-insulin). For sub-analysis purposes, insulin usage was chosen as a possible surrogate marker for a more advanced/uncontrolled disease and GLP-1RA usage as a possible contributor for delayed TTs.

Data collection

All data were collected through electronic records of VCE examinations, prospectively maintained in the VCE clinic. Data retrieved included clinical and demographic data, such as age, gender, indication for VCE referral, DM diagnosis, DM treatment, and medications and disorders affecting GI motility. We also collected data regarding the VCE procedure, such as study completion, quality of bowel preparation, TTs, and relevant clinical findings. All VCEs were performed with PillCam capsule SB3 (GIVEN Imaging, Ltd., Yokneam, Israel), and the images were analyzed using RAPID 9 (PillCam™ software v9, GIVEN Imaging, Ltd., Yokneam, Israel).

Primary and secondary outcomes

Our primary outcomes were gastric and SB TTs. Gastric TT was calculated from the time of first gastric image until first duodenal image, and SB TT was calculated from the time of first duodenal image until the first cecal image. Secondary outcomes were adequate preparation rates and VCE examination completion rates. A complete VCE examination was defined by cecal visualization reported in examination report. Safety was evaluated by the rates of “true capsule retention” (>2 weeks following swallowing). Bowel preparation was prospectively graded during initial VCE reading and report according to visibility of the mucosa by using a three-step scale (good, moderate, and poor visibility), regarding each section of the SB separately (proximal, middle, and distal SB). Moderate or poor visibility was considered “inadequate preparation” and good visibility as “adequate preparation.” The pre-VCE preparation protocol was one of two according to the referring or admitting physician discretion: dietary adjustments (24 h of low fiber diet followed by 8 h of liquid diet, then 4 h of free water and finally 12 h of fasting) or the same dietary adjustments alongside 1.5 L of purgative polyethylene glycol solution. Clinically significant findings were defined as: angioectasias, signs of bleeding, aphthae/erosions/ulcerations, mucosal masses/submucosal masses/polyps.

Statistical analysis

According to the Israeli Ministry of Health, the national prevalence of type 2 DM is 2.2% among individuals aged 35–44 years, 18.5% among patients aged 55–64 years, and increases to 30.8% in those aged 66–74 years and 33.6% in those aged 75–84 years.¹⁶ Based on these figures, we estimated the prevalence of DM in our ambulatory patient population to be approximately 25%. Sample size was calculated using gastric TT data reported by Triantafyllou et al.¹⁵ Assuming a median gastric TT of 87 min in patients with DM and 24 min in no-DM controls, with estimated standard deviations of 119 and 26 min respectively, and a 1:3 group ratio, a total of approximately 160 patients (40 with DM and 120 without) would be required to detect this difference with 80% power and a two-sided α of 0.05.

All statistical analyses were performed using SPSS version 28 (SPSS Inc., Chicago, IL, USA). Continuous variables are presented as mean ± standard deviation or median (interquartile range (IQR), range) as applicable, and the t-test or Mann–Whitney test was used (respectively) for independent groups to examine the difference in continuous variables (age, gastric TT, SB TT, etc.) between patients with and without DM. Categorical variables are presented as percentages, and the difference in their prevalence was tested between the groups, using a chi-square test. Multivariate logistic regression analysis was used to identify the independent association between DM and inadequate preparation, after adjustments for potential confounders such as age and gender. Sub-analysis of primary and secondary outcomes was performed according to anti-diabetic medication usage: insulin usage (DM with insulin compared to DM without insulin and control without DM) and GLP-1RA usage (DM with GLP-1RA compared to DM without GLP-1RA and controls without DM). A p-value of < 0.05 determined statistical significance for all tests. This study adhered to the Strengthening the Reporting of Observational Studies in Epidemiology cohort reporting guidelines to ensure comprehensive and transparent reporting of our research methods and findings.¹⁷

Results

Study population

During the study period, 1184 VCEs were performed in our center. Of these examinations, 512 VCEs were excluded due to: age < 35 years (n = 339), history of major GI surgery (n = 121), esophago-gastroduodenoscopy VCE placement (n = 31), significant technical malfunction/inability to swallow the capsule (n = 13), and repeated tests (n = 8). Subsequently, a total of 672 patients were included in the study (age 62.1 ± 14.7 years, 46.6% male gender), of which 189 (28.1%) reported having a diagnosis of DM (DM group), and 483 (71.9%) did not (no-DM group) (Figure 1).

Figure 1.

Study flowchart.

The prevalence of male gender was comparable between the DM and no-DM groups (52.4% vs 44.3%, respectively, p = 0.071) while the average age was higher among the DM group compared to no-DM group (70.4 ± 9.6 years vs 58.8 ± 15.1 years, respectively, p < 0.001), both would be dealt with by adjusting in a logistic regression later in the results. Disorders affecting GI motility were generally uncommon and equally prevalent between the groups (Table 1).

Table 1.

Clinical and demographic characteristics.

Variables	DM(n = 189)	No-DM(n = 483)	p Value
Male gender, n (%)	99 (52.4)	214 (44.3)	0.071
Age (years)	70.4 ± 9.6	58.8 ± 15.1	<0.001
Stool frequency per day	1.4 ± 1.1	1.7 ± 1.7	0.008
Indication for VCE
Anemia/bleeding	131 (69.3)	208 (43.1)	<0.001
Chronic diarrhea	48 (25.4)	254 (52.6)
Radiologic findings	10 (5.3)	21 (4.3)
Co-morbidities, n (%)
Parkinson's disease	2 (1.1)	2 (0.4)	0.292
SLE	0 (0.0)	1 (0.2)
Scleroderma	0 (0.0)	3 (0.7)
Multiple sclerosis	0 (0.0)	3 (0.7)
Mixed connective tissue disease	1 (0.5)	0 (0.0)
Medications, n (%)
Aspirin	80 (42.8)	73 (15.2)	<0.001
Narcotics	1 (0.5)	7 (1.6)	0.448
DM-medications, n (%)
Insulin	50 (26.5)	—	—
GLP-1 receptor agonist	22 (11.8)	—	—
Other anti-diabetics	159 (84.1)	—	—
Preparation protocol, n (%)
Dietary adjustments	180 (95.2)	469 (97.3)	0.226
Dietary adjustments + purgatives	9 (4.8)	13 (2.7)

DM, Diabetes mellitus; GLP-1, Glucagon-like peptide-1; SLE, Systemic lupus erythematosus.

Completion rates

A total of 8 (1.2%) VCEs were defined as incomplete (5 (1.0%) no-DM group and 3 (1.6%) in the DM group (p = 0.693)) and none of them was truly retained (seven patients completed a subsequent abdominal X-ray that did not demonstrate capsule retention, and one patient reported capsule expulsion).

Transit times

Gastric TT was comparable between the groups (37.5 ± 45.5 min in the DM group and 34.3 ± 48.6 min in the no-DM group, p = 0.424). Median values further support this similarity, with both groups showing an identical median TT of 20 min. The distributions were also comparable: IQR 12–44 min in the DM group (range 1–273) and 10–44 min in the no-DM group (range 0–751), with no significant difference (p = 0.396) (Figure 2).

Figure 2.

Distribution of gastric and small bowel transit time of VCEs between patients with and without DM.

However, in the SB, longer TT was observed among the DM group compared to the no-DM group (261.9 ± 95.6 min vs 244.9 ± 98.4 min, respectively, p = 0.044). This difference remained significant when comparing medians, with the DM group showing a median TT of 253.5 min (IQR 217–304, range 34–640) versus 236 min in the no-DM group (IQR 185–286, range 49–696; p = 0.007) (Figure 2).

Bowel preparation

The proportion of inadequate preparations was higher among the DM group compared to the no-DM group (31.2 vs 23.4%, p = 0.039). In all three SB segments, the percentage of inadequate preparations was higher in the DM group compared to the no-DM group, but statistical significance was only achieved in the middle segment (19.6% vs 10.4% p = 0.002, respectively), and borderline significance in distal segment (27.0% vs 20.7%, p = 0.082, respectively) (Figure 3). Nevertheless, in a logistic regression adjusted for age and gender, diagnosis of DM was not associated with increased rates of inadequate preparation (OR 1.373, 95% CI 0.919–2.051, p = 0.121) (Supplemental File 1).

Figure 3.

Proportion of inadequate preparation in different small bowel segments according to diagnosis of DM.

Clinically significant VCE findings

Angioectasias were significantly more prevalent in the DM group compared to no-DM controls (28.3% vs 19.7%, p = 0.021), as were findings of active bleeding or oxygenated blood (10.9% vs 6.2%, p = 0.048). In contrast, ulcers and active inflammation were less common among patients with DM than in no-DM controls (20.1% vs 34.6%, p < 0.001). The prevalence of submucosal findings was comparable between the groups (DM: 10.3% vs no-DM: 12.2%, p = 0.590). Overall, the prevalence of clinically significant findings did not differ significantly between the two groups (DM: 52.7% vs no-DM: 57.3%, p = 0.296).

Anti-diabetic medications’ use and their association with outcomes

In terms of DM treatment, 174/189 (92.1%) reported regular use of at least one anti-diabetic drug: 159 (84.1%) were treated with oral therapy, 50 (26.5%) with insulin, and 22 (11.8%) with GLP-1RA. No patient in the control group was treated with any of these drugs, including GLP-1RA for etiologies other than DM.

The DM group was divided according to insulin usage: DM with insulin usage n = 50, DM without insulin n = 139, no-DM n = 483. SB TT was significantly longer among insulin-treated patients compared to non-insulin-treated DM patients, both in terms of mean values (285.0 ± 104.7 vs 253.4 ± 91.0 min, p = 0.046) and medians (274.5 min (IQR 235.8–318.0, range 56–640) vs 247.5 min (IQR 212.8–291.8, range 34–568), p = 0.027). In contrast, gastric TTs were comparable between the groups. The mean GTT was 32.5 ± 37.8 min in insulin users and 39.4 ± 47.9 min in non-insulin users with DM (p = 0.363), with corresponding medians of 19.5 min (IQR 10–39.8, range 2–167) and 21.5 min (IQR 12–50.3, range 0–273) (p = 0.361) (Figure 4).

Figure 4.

Transit times according to DM diagnosis and insulin usage.

The rates of inadequate preparation were similar between no-DM (23.4%) and DM without insulin (26.6%) and twofold higher among those with DM using insulin (44.0%) (p = 0.006).

In a multivariate logistic regression, adjusted for age and gender, DM treated with insulin was associated a two-fold increased risk for inadequate preparation compared to patients with DM not treated with insulin (OR 2.156, 95% CI 1.096–4.239, p = 0.026) and almost 2.5-fold increased risk compared to controls without DM (OR 2.355, 95% CI 1.277–4.344, p = 0.006) (Table 2).

Table 2.

Multivariate logistic regression (dependent variable—inadequate preparation).

Variable	OR	95% CI	p Value
Gender (male vs female)	1.187	0.835–1.687	0.339
Age (years)	0.994	0.981–1.007	0.338
Insulin usage
DM (insulin vs no insulin)	2.156	1.096–4.239	4.239
DM with insulin vs no-DM	2.355	1.277–4.344	0.006

DM, diabetes mellitus.

Another sub-analysis was performed by dividing the cohort according to DM diagnosis and GLP-1RA usage (DM with GLP-1RA n = 22, DM without GLP-1RA n = 139, no-DM n = 483, no-DM with GLP-1RA n = 0). Among the 22 patients receiving GLP-1RA therapy, 14 were treated with dulaglutide and 8 with liraglutide. No patients in the cohort were treated with semaglutide or tirzepatide.

There was a non-significant trend toward longer gastric and SB TTs among patients treated with GLP-1RAs, compared to DM patients not receiving GLP-1RA therapy and no-DM controls. Gastric TT was 36.2 ± 45.2 min in non-GLP-1RA users and 47.5 ± 47.1 min in GLP-1RA users (p = 0.273), with median values of 20 min (IQR 11–42, range 0–273) and 30.5 min (IQR 12–74, range 0–181), respectively (p = 0.206). Similarly, SB TT was 260.4 ± 99.4 min in non-users and 273.4 ± 57.9 min in GLP-1RA users (p = 0.560), with corresponding medians of 248 min (IQR 215–302.5, range 34–640) and 270 min (IQR 231–321, range 155–374), respectively (p = 0.244).

Although the rates of inadequate preparation were almost double among DM using GLP-1RA (40.9%) compared to DM without GLP-1RA (29.9%) and No-DM (23.4%), only borderline significance was achieved (p = 0.061) (Supplemental File 1). According to a logistic regression, adjusted for age and gender, GLP-1RA usage among patients with DM was not associated with increased proportions of inadequate preparation compared to patients with DM not using GLP-1RA (OR 1.674, 95% CI 0.670–4.178, p = 0.270). However, a trend was observed compared to no-DM controls (OR 2.148, 95% CI 0.887–5.201, p = 0.090) (Supplemental File 1).

Discussion

In this study, we aimed to evaluate whether VCE performance could be affected by DM diagnosis. DM has been on the rise globally in recent years, with the eastern Mediterranean region demonstrating one of the most drastic climbs in prevalence.¹⁸ The national DM prevalence according to the Ministry of Health is 3%–8% among ages 35–55 years and 30%–35% among ages 65 years and above.¹⁶ In our large cohort of 672 patients aged 35 years or more, 28.1% had a diagnosis of DM and were treated with various anti-diabetic medications.

The most expected pitfall regarding VCE performance among patients with DM is delayed gastric TT due to diabetic related dysmotility.^11–13 To the best of our knowledge, until recently, only a single study investigated VCE studies among patients with DM compared to controls. This was a small sample size study including 87 patients performing VCE studies (with only 29/87 patients with DM) and demonstrating a threefold longer gastric TTs and shorter SB TTs among patients with DM compared to matched controls.¹⁵ In our study, gastric TT did not differ between patients with or without DM. Gastric TTs were similar even when controls were selectively compared to insulin users as a surrogate marker of a more advanced or uncontrolled disease. The discrepancies between our findings (study period 2018–2021) and those of Triantafyllou et al.¹⁵ (study period 2002–2006) may be attributed to differences in VCE technology, study methodology, and patient populations, between study periods. First, advances in VCE technology, such as extended battery life and enhanced imaging, likely contributed to the markedly higher completion rates observed in our cohort (>98%), compared to the 69%–89% reported in the earlier study cohort. Second, most patients in Triantafyllou et al. study underwent purgative bowel preparation, which may prolong gastric TT, and only a small proportion of them were insulin-treated patients. On the other hand, most patients in our study used standard dietary preparation, and the cohort included a more heterogeneous diabetic population, with 25% receiving insulin therapy—reflecting a broader spectrum of disease severity. These methodological and clinical differences may account for the shorter and more consistent gastric TTs in our cohort and support the broader generalizability of our findings.

Moreover, a recent study on delayed gastric TT in VCE found a higher prevalence of DM among patients with delayed emptying, but multivariate analysis showed no significant association.¹⁹ This may be explained by the fact that VCE is not designed to measure and evaluate delayed gastric emptying or gastroparesis. Studies comparing gastric TT measured by Wireless Motility Capsule (WMC), breath testing or VCE have found that VCE TT is shorter compared to the other two tests, perhaps due to the fact that the VCE is performed by drinking a small amount of water after proper fasting and the other tests require eating a meal which further delays gastric emptying.^20–22

As for SB TT, our results showed longer SB TT in the DM group, contradictory to the results described by Triantafyllou et al.,¹⁵ where the DM group had shorter SB TT. Nevertheless, median SB TT for patients with DM are similar in both studies (approximately 260 min), and the major difference is practically between SB TT of the control groups of both studies (236 vs 300 min in Triantafyllou et al.). There is no normal upper limit for SB TT, however, in most studies including a previously published study conducted in our center regarding older adult patients, the mean/median SB TTs are usually between 200 and 250 min, much less than Triantafyllou et al.’s cohort.^23,24 Longer TTs may eventually result in incomplete studies. Triantafyllou et al. reported lower VCE completion rates among the DM group compared to controls (69.0% vs 89.4%, respectively, p = 0.016). Not only that, in our study, the completion rates were comparable between the DM and no-DM groups, but they were also much higher (98.4% in DM group and 99.0% in the control group, p = 0.693). In fact, in our entire cohort, only 8/672 VCEs (1.2%) were defined as incomplete examinations with no true documented retention. It is important to mention that the study by Triantafyllou et al. was published in 2007. In that time, the first generation of VCE was utilized, allowing a video of no more than 8 h, while current generation VCE provides videos of up to 12 h. This may explain the difference in the total completion rate, but not the similarity between the groups with or without DM.

The quality of the VCE study is not established just by its completion. Another implication of longer TTs can be increased rates of inadequate preparation and impaired diagnostic yield. As expected, DM patients in our cohort demonstrated a higher prevalence of angioectasias and active bleeding, and a lower prevalence of ulcers and active inflammation, which is consistent with known pathophysiological patterns in this population. While SBTT was modestly prolonged in the DM group, this was accompanied by a higher rate of inadequate bowel preparation, which may have compromised mucosal visibility and limited the detection of additional lesions. Therefore, although angioectasias and bleeding were more common among patients with DM, these findings may in fact be underdiagnosed due to suboptimal visualization.

Inadequate preparation was more prevalent among patients with DM compared to controls (31.2% vs 23.4%, p = 0.039), while insulin usage was found to be an independent risk factor for inadequate preparation (OR 2.355, p = 0.006) adjusted for age and gender. Sub-analysis according to GLP-1RA usage showed the same trend but was probably underpowered due to the small sample size of only 22 patients (OR 2.148, p = 0.090). In the work by Triantafyllou et al., inadequate preparation rates were comparable but were much higher and approximated 50%. This was despite purgative usage by 82% of patients while in our study, more than 95% of patients used dietary adjustments alone. The ideal protocol of bowel preparation prior to VCE study is still debatable. Although various guidelines tend to recommend purgatives’ inclusion in the protocol,^1,25–27 the evidence is weak, and multiple studies demonstrate no effect on diagnostic yield in spite of a more intense preparation.^4,28,29 Furthermore, there is no common practice or specific guidelines regarding patients with DM. In a meta-analysis from 2018, DM was shown to be a significant risk factor for inadequate bowel preparation during colonoscopy in Western countries.³⁰ This supports the notion of a need for a tailored preparation protocol for patients with DM. In our study, we have used a low-residue diet followed by a liquid diet and a 12 h fast, in most patients, but perhaps a more intense protocol should be used for patients with a diagnosis of DM, especially with a more advanced disease.

Exogenic insulin is a known DM therapy, today usually used in the more advanced patients who lack glycemic control under other anti-diabetic drugs.³¹ Insulin therapy itself has no specific effect on GI motility, but unstable glycemic control does. Uncontrolled DM has a well-established effect, with hyperglycemia slowing gastric emptying.^32,33 However, our results showed the insulin using group had both longer SB TT and lower adequate preparation rates, and not necessarily delayed gastric emptying as mentioned above.

Another important anti-diabetic drug with a steep increase in its prevalent use is GLP-1RA. The recent FDA approval of GLP-1RA for weight loss caused a sharp rise in its use, become more prevalent, not only in DM patients but also in overweight and obese patients as well.³⁴ One of the known effects of GLP-1RA is delaying gastric emptying time, which begs the question whether the VCE test could be affected.

A recent study by Odah et al. included 68 diabetic patients undergoing VCE while on GLP-1RA, matched 1:1 with diabetic controls not receiving GLP-1RAs.³⁵ They reported significantly delayed gastric TTs and reduced completion rates in the GLP-1RA group. Likewise, our study found a trend toward longer gastric and small bowel transit times among GLP-1RA-treated diabetics, but differences between DM groups were non-significant. Several factors may account for this discrepancy. First, the GLP-1RA subgroup in our study was relatively small (n = 22), limiting statistical power. Second, our cohort included only users of dulaglutide and liraglutide, both of which have comparatively modest effects on GI motility. In contrast, the study by Odah et al. included newer agents such as semaglutide (33%) and tirzepatide (1.5%), which are associated with a more pronounced delay in gastric emptying.³⁵ Additionally, our study population comprised ambulatory, real-world patients who may have less controlled diabetes and more variable metabolic profile.

Interestingly, the median gastric TT among GLP-1RA users in the Odah study (37.5 min) was slightly longer than in our cohort (30.5 min).³⁵ However, their non-GLP-1RA control group had a notably shorter median gastric TT (16 min), compared to both our non-GLP-1RA group (20 min) and no-DM controls (also 20 min), potentially amplifying the contrast. Nonetheless, all these differences are modest and likely not clinically significant. Surely, further studies on the subject are needed before making the GLP-1RA a non-issue regarding the yield of the VCE.

Of note, a recent study has found no association between GLP1 agonist usage and delayed gastric emptying among patients with DM including sub-analyses according to specific populations such as age, gender, smoking status, etc. Moreover, the strongest predictor for delayed gastric emptying was the duration of DM (more or less than 10 years) and not the level of hemoglobin A1C.³⁶ It should be noted though that there may be a selection bias whereas diabetics with gastroparesis tend to stop those medications, as adherence rates are low with long-term antidiabetic drugs generally and specifically with GLP1 agonists.^37,38

Our study has several limitations, mostly resulting from the nature of retrospective studies. Although most of the data were prospectively documented, important objective measures for glycemic control such as hemoglobin A1C and fasting blood glucose levels could not be found. Nevertheless, the data is not readily available through referral letters, thus our results reflect real-world decision-making, by using the patient's medication list as a surrogate marker. Our results might be impaired by classification bias, since the classification of patients was according to their own report regarding the diagnosis of diabetes. There may be patients in the control group with undiagnosed DM, but in that case, it is probably a relatively “new” diabetes, and therefore the chance of damage to target organs is low. Furthermore, our study did not differentiate between type 1 and type 2 diabetes, and data regarding glycemic control and maintenance were not available. At the time of cohort assembly, data on diabetes types were not systematically collected. Nevertheless, given the low prevalence of type 1 diabetes in the general adult population and the age threshold of ⩾35 years in our cohort, we believe that the findings predominantly reflect individuals with type 2 diabetes. As such, the conclusions drawn remain relevant and applicable to this population. Another limitation is the small sample size of patients treated with GLP-1RA (n = 22), which limits the statistical power to detect significant differences in this subgroup. Although these patients showed numerically higher rates of inadequate bowel preparation compared to both no-DM and DM patients not receiving GLP-1RA therapy, the difference did not reach statistical significance. Larger, adequately powered studies are needed to clarify the potential impact of GLP-1RA treatment on VCE performance. Finally, although this was a single-center study, it represents the largest VCE cohort in patients with DM to date. The relatively large and diverse population may help mitigate this limitation and enhance the generalizability of our findings.

Conclusion

In conclusion, the prevalence of DM is rising globally, and attention should be given to special considerations regarding this patient population. We have found the VCE procedure to be safe for patients with DM, regardless of the duration of the disease or anti-diabetic treatment. As for the efficacy of the test, it seems that gastric TTs do not differ from controls without DM, and the SB TTs are slightly longer but not in a way that significantly impairs the completion of the exam, allowing detection of typical significant findings. However, considering higher percentages of inadequate preparation, especially in patients regularly using insulin, we suggest consideration of a more intense preparation protocol and naturally testing its effectiveness in future prospective studies.

Supplemental Material

sj-doc-1-cmg-10.1177_26317745251359459 – Supplemental material for Evaluating video capsule endoscopy in diabetes mellitus: transit times, preparation adequacy, and the influence of insulin and GLP-1 receptor agonist use

Supplemental material, sj-doc-1-cmg-10.1177_26317745251359459 for Evaluating video capsule endoscopy in diabetes mellitus: transit times, preparation adequacy, and the influence of insulin and GLP-1 receptor agonist use by Nimrod Akerman, Naomi Fliss-Isakov, Tamar Thurm, Mati Shnell, Yael Sofer, Oren Shibolet and Liat Deutsch in Therapeutic Advances in Gastrointestinal Endoscopy

Supplemental Material

sj-docx-2-cmg-10.1177_26317745251359459 – Supplemental material for Evaluating video capsule endoscopy in diabetes mellitus: transit times, preparation adequacy, and the influence of insulin and GLP-1 receptor agonist use

Supplemental material, sj-docx-2-cmg-10.1177_26317745251359459 for Evaluating video capsule endoscopy in diabetes mellitus: transit times, preparation adequacy, and the influence of insulin and GLP-1 receptor agonist use by Nimrod Akerman, Naomi Fliss-Isakov, Tamar Thurm, Mati Shnell, Yael Sofer, Oren Shibolet and Liat Deutsch in Therapeutic Advances in Gastrointestinal Endoscopy

Footnotes

Acknowledgements

None.

Declarations

ORCID iD

Nimrod Akerman

Supplemental material

Supplemental material for this article is available online.

References

Enns

Hookey

Armstrong

, et al. Clinical practice guidelines for the use of video capsule endoscopy. Gastroenterology 2017; 152: 497–514.

Blanco-Velasco

Zamarripa-Mottú

Solórzano-Pineda

, et al. Comparison in the diagnostic yield between ‘Pillcam SB3’ capsule endoscopy and ‘OMOM Smart Capsule 2’ in small bowel bleeding. A randomized head-to-head study. Digestive Dis 2020; 39: 211–216.

Koffas

Laskaratos

F-M

Epstein

Training in video capsule endoscopy: current status and unmet needs. World J Gastrointest Endosc 2019; 11: 395–402.

Hookey

Louw

Wiepjes

, et al. Lack of benefit of active preparation compared with a clear fluid–only diet in small-bowel visualization for video capsule endoscopy: results of a randomized, blinded, controlled trial. Gastrointest Endosc 2017; 85: 187–193.

Collins

PD.

Video capsule endoscopy in inflammatory bowel disease. World J Gastrointest Endosc 2016; 8: 477–488.

Van Gossum

Munoz-Navas

Fernandez-Urien

, et al. Capsule endoscopy versus colonoscopy for the detection of polyps and cancer. New Engl J Med 2009; 361: 264–270.

Zimmet

Alberti

Magliano

, et al. Diabetes mellitus statistics on prevalence and mortality: facts and fallacies. Nat Rev Endocrinol 2016; 12: 616–622.

Yarandi

Srinivasan

Diabetic gastrointestinal motility disorders and the role of enteric nervous system: current status and future directions. Neurogastroenterol Motility 2014; 26: 611–624.

Chiu

Y-C

Kuo

M-C

Rayner

, et al. Decreased gastric motility in type II diabetic patients. Biomed Res Int 2014; 2014: 894087.

10.

Asgharnezhad

Joukar

Fathalipour

, et al. Gastrointestinal symptoms in patients with diabetes mellitus and non-diabetic: a cross-sectional study in North of Iran. Diabetes Metab Syndr 2019; 13: 2236–2240.

11.

Chakraborty

Halland

Burton

, et al. GI dysfunctions in diabetic gastroenteropathy, their relationships with symptoms, and effects of a GLP-1 antagonist. J Clin Endocrinol Metab 2019; 104: 1967–1977.

12.

Kong

Horowitz

Jones

, et al. Natural history of diabetic gastroparesis. Diabetes Care 1999; 22: 503–507.

13.

Horowitz

Fraser

Disordered gastric motor function in diabetes mellitus. Diabetologia 1994; 37: 543–551.

14.

Bandorski

Kurniawan

Baltes

, et al. Contraindications for video capsule endoscopy. World J Gastroenterol 2016; 22: 9898–9908.

15.

Triantafyllou

Kalantzis

Papadopoulos

, et al. Video-capsule endoscopy gastric and small bowel transit time and completeness of the examination in patients with diabetes mellitus. Digest Liver Dis 2007; 39: 575–580.

16.

ICDC-Israel Center for Disease Control. Diabetes registry report 2016. Ministry of Health Israel, 2016. https://www.health.gov.il/UnitsOffice/ICDC/disease_Registries/Pages/diabetes_reg.aspx

17.

von Elm

Altman

Egger

, et al. The strengthening the reporting of observational studies in epidemiology (STROBE) statement: guidelines for reporting observational studies. J Clin Epidemiol 2008; 61: 344–349.

18.

Global report on diabetes, 2016. https://www.who.int/publications-detail-redirect/9789241565257

19.

Freitas

Macedo Silva

Xavier

, et al. Prolonged gastric transit time in small-bowel capsule endoscopy: which patients are at risk and what are the implications? Turk J Gastroenterol. Epub ahead of print 9 January 2023. DOI: 10.5152/tjg.2023.22191.

20.

Diaz Tartera

Webb

D-L

Al-Saffar

, et al. Validation of SmartPill® wireless motility capsule for gastrointestinal transit time: intra-subject variability, software accuracy and comparison with video capsule endoscopy. Neurogastroenterol Motil 2017; 29: 1–9.

21.

Hejazi

Bashashati

Saadi

, et al. Video capsule endoscopy: a tool for the assessment of small bowel transit time. Front Med (Lausanne) 2016; 3: 6–10.

22.

Nonaka

Inamori

Endo

, et al. Correlation between gastric transit time measured by video capsule endoscopy and gastric emptying determined by the continuous real-time 13C breath test (BreathID system). Hepatogastroenterology 2014; 61: 2159–2162.

23.

Thurm

Gluck

Barak

, et al. Octa-nonagenarians can perform video capsule endoscopy safely and with a higher diagnostic yield than 65-79-year-old patients. J Am Geriatr Soc 2022; 70: 2958–2966.

24.

O’Grady

Murphy

Barry

, et al. Defining gastrointestinal transit time using video capsule endoscopy: a study of healthy subjects. Endosc Int Open 2020; 8: 396–400.

25.

Pennazio

Rondonotti

Despott

, et al. Small-bowel capsule endoscopy and device-assisted enteroscopy for diagnosis and treatment of small-bowel disorders: European Society of Gastrointestinal Endoscopy (ESGE) Guideline – Update 2022. Endoscopy 2023; 55: 58–95.

26.

Leighton

Brock

Semrad

, et al. Quality indicators for capsule endoscopy and deep enteroscopy. Am J Gastroenterol 2022; 117: 1780.

27.

Squirell

Ricci

Hookey

Preparation, timing, prokinetics, and surface agents in video capsule endoscopy. Gastrointest Endosc Clin N Am 2021; 31: 251–265.

28.

Rahmi

Cholet

Gaudric

, et al. Effect of different modalities of purgative preparation on the diagnostic yield of small bowel capsule for the exploration of suspected small bowel bleeding: a multicenter randomized controlled trial. Am J Gastroenterol 2022; 117: 327–335.

29.

Lamba

Ryan

Hwang

, et al. Clinical utility of purgative bowel preparation before capsule endoscopy: a multicenter, blinded, randomized controlled trial. Gastrointest Endosc 2022; 96: 822–828.

30.

Mahmood

Farooqui

Madhoun

MF.

Predictors of inadequate bowel preparation for colonoscopy: a systematic review and meta-analysis. Eur J Gastroenterol Hepatol 2018; 30: 819–826.

31.

Hamaty

Insulin treatment for type 2 diabetes: when to start, which to use. Cleve Clin J Med 2011; 78: 332–342.

32.

Russo

Stevens

Chen

, et al. Insulin-induced hypoglycemia accelerates gastric emptying of solids and liquids in long-standing type 1 diabetes. J Clin Endocrinol Metab 2005; 90: 4489–4495.

33.

Riddle

MC.

Relief of gastrointestinal symptoms by correcting insulin excess. Diabetes Care 1981; 4: 296–298.

34.

Adhikari

Jha

Dardari

, et al. National trends in use of sodium-glucose cotransporter-2 inhibitors and glucagon-like peptide-1 receptor agonists by cardiologists and other specialties, 2015 to 2020. J Am Heart Assoc 2022; 11: e023811.

35.

Odah

Vattikonda

Stark

, et al. Glucagon-like peptide-1 receptor agonists and capsule endoscopy in patients with diabetes: a matched cohort study. Gastrointest Endosc 2025; 101: 393–401.

36.

Kalas

Dang

Galura

, et al. Frequency of GLP-1 receptor agonists use in diabetic patients diagnosed with delayed gastric emptying and their demographic profile. J Investig Med 2023; 71: 11–16.

37.

Weiss

Yang

Carr

, et al. Real-world weight change, adherence, and discontinuation among patients with type 2 diabetes initiating glucagon-like peptide-1 receptor agonists in the UK. BMJ Open Diabetes Res Care 2022; 10: e002517.

38.

Durden

Liang

Fowler

, et al. The effect of early response to GLP-1 RA therapy on long-term adherence and persistence among type 2 diabetes patients in the United States. J Manag Care Spec Pharm 2019; 25: 669–680.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.09 MB

0.04 MB