What Is Known
- Vascular changes are a hallmark of inflammatory bowel diseases (IBD), but their relation to pathogenesis is unclear.
- Probe-based confocal laser endomicroscopy (pCLE) is an optical imaging technique that enables live visualization of the mucosal surface during endoscopy.
What Is New
- Probe-based confocal laser endomicroscopy (pCLE) is successfully utilized for the first time to study capillary flow rates in the duodenum of non-IBD and IBD patients.
- Capillary flow rates in the duodenum of ulcerative colitis patients are increased in the absence of inflammation as compared to non-IBD patients.
Inflammatory bowel diseases (IBD), including both Crohn disease (CD) and ulcerative colitis (UC), are major pediatric health concerns (1). IBD mainly involves the gastrointestinal tract; however, these diseases can also manifest outside the gut, indicative of potentially broad-spread systemic involvement (2,3). While disease extent is typically limited to the colon in UC (in contrast to CD, where the entire bowel can be involved), extracolonic involvement in UC, such as focal gastritis, has also been reported (4,5), suggesting that the pathogenesis may reach beyond the colon. We have recently demonstrated increased epithelial gaps (indicative of epithelial cell shedding) in the noninflamed duodenum of UC using probe-based confocal laser endomicroscopy (pCLE), an optical imaging technique that enables live visualization of the mucosal surface (6). Thus, pCLE is a technique that can assist in visualizing changes associated with IBD in vivo during endoscopy, and may help in better understanding disease pathogenesis. Mucosal barrier and microbial changes prevail in the noninflamed terminal ileum (TI) of pediatric UC patients (7). These findings indicate that pathological changes can occur outside of colon in UC, despite lack of endoscopic and microscopic inflammation.
Vascular changes have been described in IBD, although it is unclear how they relate to disease pathogenesis (8). A major challenge in studying the pathogenesis of IBD is that many features, including increased vascular flow, could be the result of present inflammation, and not the cause. Increased epithelial gaps in the noninflamed duodenum of pediatric UC indicate that a subclinical pathophysiologic process occurs in UC outside the colon (6); this could help identify factors that may potentially cause preliminary tissue damage, unrelated to or preceding inflammation. Thus, assessing circulatory changes in the duodenum in the absence of inflammation could better define the early stages of vascular pathogenesis in IBD.
We hypothesized that children with IBD would have alterations in the microvascular flow. Microvascular changes, such as tortuous vessels and fluorescein leakage into the lamina propria, have been assessed with pCLE and were found to be abnormal, even in the absence of frank lesions in UC patients (9). pCLE has been used for studying vascular changes in sepsis in porcine models and in the duodenum of patients with severe sepsis to quantify capillary length and functional capillary density (10). Vascular flow was, however, not assessed in either of these studies. We are the first group to report vascular flow changes in IBD patients, which is indeed a novel application of pCLE for understanding IBD pathogenesis. The objective of the study was to assess microvascular flow in the non-inflamed tissues of IBD patients. High resolution of mucosal circulation and single blood cell tracking can be assessed with pCLE and enable us to measure capillary flow rates during endoscopy. By measuring capillary flow rates in duodenal images obtained using pCLE during endoscopy, we show that pediatric UC patients have increased duodenal capillary flow, which was not related to disease activity, inflammatory markers, or hemoglobin levels.
IBD patients and non-IBD controls (4–17 years old) undergoing endoscopic evaluation were included. Ethics approval (University of Alberta Research Ethics Board Study-ID Pro00023820) was obtained. The non-IBD arm of the cohort included patients undergoing esophagogastroduodenoscopy and ileocolonoscopy to investigate gastrointestinal symptoms (eg, abdominal pain, diarrhea), but they were excluded if any mucosal abnormality was found. Patients with any abnormality in the upper gastrointestinal tract (endoscopic or histologic) or other mucosal lesions were excluded. A major challenge that we faced was difficulty in recruitment, as confocal imaging prolonged the endoscopy procedure and required intravenous (IV) fluorescein. Also, an ideal non-IBD cohort would have included healthy children but they do not require endoscopy. All patients received Picosalax (sodium picosulphate with magnesium citrate) for colonoscopy bowel prep. Endoscopies were performed under general anesthesia with propofol.
Probe-based Confocal Laser Endomicroscopy and Capillary Flow Rate Analysis
Duodenal vascular imaging of patients was conducted using pCLE during endoscopy. Mucosal imaging with pCLE was performed before extracting biopsies in all cases and shortly after initiating anesthesia (upper endoscopy always preceded colonoscopy) to reduce variation in anesthesia and procedure time that could affect outcomes. After injection of IV fluorescein (10%, 5 mg/kg, maximum of 250 mg; Alcon, Mississauga, ON, Canada) as contrast media, the confocal probe was inserted through the endoscopy working channel. Mucosal images were collected by positioning the probe directly against the mucosa (Fig. 1A). Images were obtained, anonymously stored, and analyzed, using the Cellvizio viewer software (Muana Kea Technologies Inc, Suwanee, GA), separately by 2 reviewers who were blinded to the study groups.
Images were selected where at least one-fourth of the villus was visible. Video sequences in which the blood cells were present in a single file were selected for capillary flow rate analysis since more than one parallel file would likely represent larger blood vessels with higher flow rates. Blood vessels with a diameter greater than 20 μm were excluded for the same reason. Blood cells needed to be visible over at least 3 consecutive frames in the video sequence to be included in the analysis. A minimum of 3 individual blood cell samples were tracked per patient (more when possible). The averages of the distances travelled by blood cells were used to determine flow rates for each patient. Distance travelled by blood cells was measured using an epithelial cell as an anatomical reference point with the Cellvizio software diameter tool. The distance travelled was divided by sequence duration to calculate the capillary flow rate (μm/ms or mm/s; Fig. 1B).
Confocal endomicroscopy prolonged the endoscopy procedure by 5 to 7 minutes. Fluorescein is commonly used in ophthalmology practice and it has been safely administered during confocal endomicroscopy procedures in pediatric patients (11). Some reported adverse reactions to intravenous injection of fluorescein include rash, hypotension, nausea, and vomiting, but occur rarely (12). The risk of severe side effects, such as bronchospasm and cardiac arrest, is extremely low (0.05%) (13).
Correlation of Capillary Flow With Disease Activity
The correlation between capillary flow and disease features was conducted by plotting capillary flow rates against serum inflammatory markers (erythrocyte sedimentation rate [ESR], C-reactive protein) (14) and against the standardized IBD scoring index, Pediatric Ulcerative Colitis Activity Index (PUCAI for UC) and Pediatric Crohn's Disease Activity Index (PCDAI for CD) (15,16). Serum hemoglobin and albumin levels were also compared with flow rates.
As described in our recently published study (6), cytokine profiles in the duodenal biopsies of all non-IBD and IBD patients were analyzed using the V-Plex pro-inflammatory panel 1 kit (Mesoscale Diagnostics, Rockville, MD) and included the following cytokines: interferon (IFN)-γ, interleukin (IL)-1β, IL-2, IL-4, IL-6, IL-8, IL-10, IL-12p70, IL-13, and tumor necrosis factor (TNF).
Data were analyzed using Graph Pad Prism (San Diego, CA). Comparison between study groups was done using analysis of variance (ANOVA) and correlations were conducted using Pearson's correlation coefficient. Interobserver and intraobserver variability were determined by Cohen kappa (κ) coefficient. Statistical significance was determined as P < 0.05.
Confocal images were obtained from 57 patients, of which 12 were excluded as they did not meet the inclusion criterion (low image resolution; tissue inflammation). Of the 12 patients who were excluded, 3 IBD and 5 non-IBD patients had poor resolution of images. All 4 patients who were excluded for evidence of inflammation in the duodenum were non-IBD patients. Complete data were available for 45 patients, including 22 non-IBD controls, 14 CD, and 9 UC patients. Of the CD cohort, 8 were newly diagnosed and 6 were previously diagnosed. All UC patients were established cases. Patient characteristics are described in Supplemental Digital Content, Table 1, http://links.lww.com/MPG/A879.
Capillary Flow Rates
Confocal images were analyzed to quantify capillary vascular flow (example shown in Fig. 1B). A minimum of 3 distance points travelled by a blood cell were recorded per patient (more where possible). Datasets from both observers were analyzed and the average measurements reported. The number of capillary flow measurements recorded per patient were non-IBD: 4.4 ± 0.25, CD: 3.90 ± 0.3, UC: 4.2 ± 0.4. Duodenal capillary flow rates were significantly higher in UC patients (Fig. 1C; non-IBD controls: 0.57 ± 0.03; UC: 0.75 ± 0.02; CD: 0.65 ± 0.02 [mean ± SEM; capillary flow rate mm/s]; ANOVA, P < 0.05, N = 22 non-IBD, 14 CD, and 9 UC patients). Cohen κ coefficient interobserver variability was 0.82, suggesting strong agreement between the blinded observers. Mean values of capillary flow rate measurements by reviewer 1 were non-IBD: 0.55 ± 0.04, UC: 0.74 ± 0.06, CD: 0.63 ± 0.04, and those of the second reviewer were non-IBD: 0.59 ± 0.02, UC: 0.8 ± 0.05, CD: 0.67 ± 0.04 mm/s. Intraobserver reliability of the measurements was determined by a second blinded analyses by reviewer 1, and showed a high degree of correlation between the first and second sets of readings (Cohen κ coefficient was 0.98). None of the included patients had duodenal inflammation on endoscopy or histology; this suggests that increased vascular flow occurs in these patients in the absence of frank lesions.
Correlation of Capillary Flow Rates With Inflammation
There was no correlation between ESR, C-reactive protein, PUCAI, and capillary flow rates in UC patients. Rather, ESR and PUCAI showed a trend for negative correlation with flow, as did PCDAI. Although our study may not have been large enough to link changes in vascular flow with inflammation, there was no correlation between capillary flow rates and inflammatory markers in non-IBD and UC patients (P > 0.05, Fig. 2A–C). This further supports that increased blood flow in UC is present even in the absence of inflammation. While duodenal IL-2 and IL-8 were high in UC patients, there was no correlation between duodenal IL-2 and IL-8 levels and capillary flow rates (data not shown). None of the patients had active inflammation in the duodenum; however, 10 of 14 CD and 6 of 9 UC patients were found to have active tissue inflammation on mucosal biopsies taken during ileocolonoscopy. In the CD group, 4 patients had inflammation in the TI, 2 in the colon, and 4 in both TI and colon. Three UC and 4 CD patients of the entire cohort were in remission at the time of endoscopy. In a subgroup analysis, we found no difference in the vascular flow rates of patients with active or inactive disease (data not shown).
Serum Hemoglobin and Albumin Levels
Increased capillary flow rates could be secondary to anemia and increased cardiac output. Thus, we compared serum hemoglobin levels between groups. There was a trend for lower hemoglobin in IBD patients but this was not statistically significant. Hypoprotenemia also affects cardiac output, but there was no difference in albumin levels between groups (Supplemental Digital Content, Table 1, http://links.lww.com/MPG/A879). None of our patients had hemoglobinopathies, Heinz bodies, red cell defects, or other vascular disorders.
The increased capillary flow was not explained by changes in hemoglobin or albumin as there was no correlation between serum hemoglobin and albumin of UC patients and capillary flow rates (Fig. 3; hemoglobin: Pearson r coefficient = 0.32, P > 0.05; albumin: Pearson r coefficient = 0.04, P > 0.05).
Vascular changes are a hallmark of IBD and have been associated with inflammation (8); the question of whether these changes contribute to, or occur as a consequence of, inflammation remains.
The goal of our study was to identify alterations in the microvascular circulation in unaffected areas in IBD patients by quantifying capillary flow rates using pCLE. Our study is the first to measure capillary blood flow rates in the IBD patients overall, and specifically in the duodenum, and to report an increase in UC patients, compared with non-IBD and CD patients. This novel finding in our pilot study is demonstrated in the absence of local inflammation, suggesting that it is not secondary to inflammation. Increased vascular flow was unrelated to inflammatory markers or disease activity (although this was a secondary outcome that our study was likely not powered enough to determine). Thus, our data suggest that vascular changes can occur independent of inflammatory parameters and may represent a fundamental intestinal change in UC patients outside the colon. Capillary flow rates were unchanged in CD patients, which suggest that different pathogenic pathways could be present in UC and CD. Hyperemia in the gut occurs as a result of many stimuli, such as increased tissue oxygen consumption and serotonin, possibly mediated by nitric oxide (17).
Endothelial cells prevent platelet aggregation and excessive leukocyte infiltration in the healthy state, and also maintain low vascular permeability. Changes in blood flow can occur due to sheer stress that can cause vasodilation of arterioles. Vascular flow is affected by factors secreted by the endothelium, such as prostacyclin, nitric oxide, and endothelium-derived hyperpolarizing factor, which act as vasodilators, or by vasoconstrictors, such as endothelin-1 and thromboxane (18). Vascular endothelial growth factor-A and B induce angiogenesis in IBD (19). In addition, changes in endothelial cells occur in response to inflammation as various cytokines, such as TNF-α, induce VCAM-1 and MAdCAM-1 and promote platelets adherence in IBD. Cytokines involved in IBD also affect gut vasculature, as TNF-α and IL-8 both possess pro-angiogenic properties (8). We did find IL-8 to be elevated in the duodenal biopsies of our UC patients in a previous study (6), but what triggered its increase is yet unknown. Our recent study did find increased epithelial gaps in the same patients (6); however, there was no correlation between the increased blood flow and increased gaps (data not shown).
It is tempting to speculate that changes in these (and other, yet unidentified) parameters could potentially be a reason for increased blood flow in the absence of inflammation and could ultimately trigger inflammation. This was, however, not directly addressed in this study and such conclusions cannot be made without additional studies with larger cohorts and molecular analysis.
Anemia occurs commonly in IBD due to blood or iron loss, and/or malabsorption (20) and can affect cardiac output and blood flow. Tissue hypoxia can result in increased cardiac output, decreased blood viscosity, and peripheral vascular resistance (21). To determine whether increased blood flow in our UC patients could be explained by anemia, we compared hemoglobin levels of all groups and did not find any difference between groups. We correlated the flow rates with hemoglobin levels and found that variation in flow could not be explained by hemoglobin; in fact, hemoglobin showed a trend for a positive correlation with flow. Having a direct measure of cardiac output would have been a more reliable control, but this was not clinically indicated during the procedures and, therefore, not available. As protein-losing enteropathy is a rare complication of UC (22), and hypoalbuminemia affects cardiac output, we compared the pre-procedure albumin levels but found no differences between groups.
Our study has several limitations. While patients in the CD group included both newly diagnosed and chronic patients, the UC cohort only had follow-up cases; follow-up cases were on therapy, which could affect our results. The effects of 5-ASA on vascular dynamics is well established, including prevention of platelet aggregation and reducing factor VIII activation, thus preventing thrombin formation at microvascular lesion sites (23,24). Oral administration of 5-ASA attenuates activation of platelets in IBD patients (25).
Recruitment of treatment-naïve patients, especially with UC, in future studies are important to address the effect of treatment on our findings. Prolongation of the endoscopy procedure with confocal imaging and the requirement of IV injection of fluorescein did, unfortunately, negatively impact recruitment, thus our sample size is small. We would have liked to measure flow rates in inflamed areas as well but had to limit the study to one site to reduce endoscopy time and the need for a second injection of fluorescein. In addition, given that this method has not been described to date in humans, we did not have a reference to compare to. Remarkably, a mouse study has shown a mean intestinal capillary flow rate of 2.6 nL/min (26), which is almost completely identical to the flow rate we found (after unit conversion and assuming a capillary diameter of 10 μm, 0.57 mm/s = 2.686 nL/min). While capillary flow rates can be successfully quantified using pCLE, we were not successful in measuring capillary density to assess angiogenesis due to technical challenges. It is possible that the flow changes observed are secondary to changes in the vascular bed, but one would expect increased capillary branching to slow flow down, not enhance it.
In conclusion, our study is the first to demonstrate quantification of microvascular blood flow in the duodenum of IBD patients, and showed increased capillary flow rates in the unaffected duodenum in UC patients using pCLE. Assessing duodenal blood flow in IBD patients over time in subsequent studies may help determine if disease course or responses to therapy are predicted by blood flow rates. Using this novel application of pCLE technology, future studies need to focus on associating changes in capillary flow rates with underlying vascular pathology in UC to better define disease pathogenesis and possibly identify new treatment targets.
The authors are grateful for the patients and their families for participating in the study and the endoscopy nurses, anesthetists, and staff for their assistance with carrying out the pCLE studies. Special thanks to Cheryl Kluthe, RN, and Pavel Medvedev, EPIC research coordinator, funded through a WCHRI capacity building award, for their assistance in recruiting patients.
1. Burisch J, Munkholm P. Inflammatory bowel disease epidemiology. Curr Opin Gastroenterol
2. Patil SA, Cross RK. Update in the management of extraintestinal manifestations of inflammatory bowel disease. Curr Gastroenterol Rep
3. Dotson JL, Hyams JS, Markowitz J, et al. Extraintestinal manifestations of pediatric inflammatory bowel disease and their relation to disease type and severity. J Pediatr Gastroenterol Nutr
4. Lin J, McKenna BJ, Appelman HD. Morphologic findings in upper gastrointestinal biopsies of patients with ulcerative colitis
: a controlled study. Am J Surg Pathol
5. Tobin JM, Sinha B, Ramani P, et al. Upper gastrointestinal mucosal disease in pediatric Crohn disease and ulcerative colitis
: a blinded, controlled study. J Pediatr Gastroenterol Nutr
6. Zaidi D, Jorgensen M, Huynh HQ, et al. Increased epithelial gap density in the non-inflamed duodenum of children with inflammatory bowel diseases. J Pediatr Gastroenterol Nutr
7. Alipour M, Zaidi D, Valcheva R, et al. Mucosal barrier depletion and loss of bacterial diversity are primary abnormalities in paediatric ulcerative colitis
. J Crohns Colitis
8. Cromer WE, Mathis JM, Granger DN, et al. Role of the endothelium in inflammatory bowel diseases. World J Gastroenterol
9. Buda A, Hatem G, Neumann H, et al. Confocal laser endomicroscopy for prediction of disease relapse in ulcerative colitis
: a pilot study. J Crohn's Colitis
10. Schmidt C, Lautenschlger C, Petzold B, et al. Confocal laser endomicroscopy reliably detects sepsis-related and treatment-associated changes in intestinal mucosal microcirculation. Br J Anaesth
11. Venkatesh K, Cohen M, Evans C, et al. Feasibility of confocal endomicroscopy in the diagnosis of pediatric gastrointestinal disorders. World J Gastroenterol
12. Kwiterovich KA, Maguire MG, Murphy RP, et al. Frequency of adverse systemic reactions after fluorescein angiography. Results of a prospective study. Ophthalmology
13. Yannuzzi LA, Rohrer KT, Tindel LJ, et al. Fluorescein angiography complication survey. Ophthalmology
14. Solem CA, Loftus EV, Tremaine WJ, et al. Correlation of C-reactive protein with clinical, endoscopic, histologic, and radiographic activity in inflammatory bowel disease. Inflamm Bowel Dis
15. Turner D, Otley AR, Mack D, et al. Development, validation, and evaluation of a pediatric ulcerative colitis
activity index: a prospective multicenter study. Gastroenterology
16. Turner D, Griffiths AM, Walters TD, et al. Mathematical weighting of the pediatric Crohn's disease activity index (PCDAI) and comparison with its other short versions. Inflamm Bowel Dis
17. Kolios G, Valatas V, Ward SG. Nitric oxide in inflammatory bowel disease: a universal messenger in an unsolved puzzle. Immunology
18. Sandoo A, van Zanten JJCSV, Metsios GS, et al. The endothelium and its role in regulating vascular tone. Open Cardiovasc Med J
19. Scaldaferri F, Vetrano S, Sans M, et al. VEGF-A links angiogenesis and inflammation in inflammatory bowel disease pathogenesis. Gastroenterology
20. Weiss G, Gasche C. Pathogenesis and treatment of anemia in inflammatory bowel disease. Haematologica
21. Hebert PC, Hu LQ, Biro GP. Review of physiologic mechanisms in response to anemia. CMAJ
22. Ungaro R, Babyatsky MW, Zhu H, et al. Protein-losing enteropathy in ulcerative colitis
. Case Rep Gastroenterol
23. Undas A. Blood coagulation at the site of microvascular injury: effects of low-dose aspirin. Blood
24. Undas A, Brummel-Ziedins KE, Mann KG. Antithrombotic properties of aspirin and resistance to aspirin: beyond strictly antiplatelet actions. Blood
25. Carty E, Macey M, Rampton DS. Inhibition of platelet activation by 5-aminosalicylic acid in inflammatory bowel disease. Aliment Pharmacol Ther
26. Foitzik T, Kruschewski M, Kroesen A, et al. Does microcirculation play a role in the pathogenesis of inflammatory bowel diseases? Answers from intravital microscopic studies in animal models. Int J Colorectal Dis