Abstract
Introduction: Aortic stenosis (AS) is sometimes associated with gastrointestinal bleeding, and this phenomenon is known as Heyde’s syndrome. Such bleeding is most often considered to originate from gastrointestinal angiodysplasias, but the frequency and endoscopic features of such bleeding remain unclear. This study aimed to determine the frequency and endoscopic features of gastrointestinal angiodysplasia in patients with severe AS. Patients and Methods: In this multicenter, retrospective study, we evaluated consecutive patients who underwent transcatheter aortic valve implantation (TAVI) with severe AS from May 2016 to December 2019. We extracted the data on the clinicopathological features according to the status of anemia, the proportion of patients who underwent gastrointestinal endoscopic examinations and demonstrated gastrointestinal angiodysplasia, and identified the endoscopic features associated with such patients. Results: In 325 patients, the rates of moderate/severe anemia (hemoglobin < 11 g/dL) were 52%. Regarding medicine, there were no significant differences between the patients with and without moderate/severe anemia. Patients were examined by esophagogastroduodenoscopy (21%), colonoscopy (12%), and balloon-assisted enteroscopy or small bowel capsule endoscopy (1.5%). Patients with moderate/severe anemia had significantly more angiodysplasia (38.3% vs. 7.7%; p < 0.0001) and active bleeding (23.4% vs. 0%; p < 0.01). Angiodysplasia was detected in 21 patients (stomach, n = 9; small intestine, n = 5, and colon, n = 10). Conclusions: The results suggest, for the first time, that patients with severe AS who underwent TAVI and moderate/severe anemia frequently had gastrointestinal angiodysplasia and active bleeding throughout the entire gastrointestinal tract.
Introduction
Aortic stenosis (AS) is the most common valvular heart disease [1]. A previous report showed that the prevalence of severe AS among elderly individuals was 3.4% [2]. Transcatheter aortic valve implantation (TAVI) is used to treat patients with severe, symptomatic AS and an inoperable status or a high-risk but operable status [3, 4]. TAVI has been associated with reduced mortality and improved overall functional status and quality of life [3].
Anemia is one of the most common comorbidities in patients with severe AS [5]. Heyde’s syndrome is the association of gastrointestinal angiodysplasia bleeding with AS [6‒8]. Angiodysplasia, namely, fragile abnormal small vessels that form just underneath the mucosa, develops due to certain hemostatic disorders [9, 10]. There are several biological mechanisms underlying Heyde’s syndrome. First, a narrow valve causes abnormal shear stress in the blood flow, and high-molecular-weight multimers (von Willebrand factor [VWF]) are reduced in size and become less hemostatically competent than their parent polymers [6]. In 67% of patients with severe AS, a loss of high-molecular-weight multimers is noted [7]. Second, the loss of VWF causes angiogenesis via an Ang-2/Tie-2/VEGFR-2 pathway [11]. Previous studies have shown that approximately 20% of patients with critical AS who underwent aortic valve replacement had a history of dermal or mucosal bleeding [12].
However, in Heyde’s syndrome, the frequency of gastrointestinal angiodysplasia and bleeding and their endoscopic features are unclear. Therefore, in the present study, we conducted a retrospective survey evaluating the characteristics of patients with severe AS who required TAVI, as well as the corresponding endoscopic features.
Patients and Methods
A multicenter retrospective cohort study was conducted at 10 institutes. The 10 institutes were Kyoto Prefectural University of Medicine, Kyoto Prefectural University of Medicine North Medical Center, Kyoto Okamoto Memorial Hospital, Fukuchiyama City Hospital, Kyoto City Hospital, Japanese Red Cross Kyoto Daini Hospital, Nishijin Hospital, Matsushita Memorial Hospital, Rakusai Shimizu Hospital, and Rakuwakai Otowa Hospital. We reviewed the preprocedural records from our database for all adults who were diagnosed with severe AS who required TAVI and who underwent TAVI at 10 institutes between May 2016 and March 2019. All patients provided their written informed consent to undergo TAVI. This study was approved by the Ethical Review Committee of Kyoto Prefectural University of Medicine (ERB-C-1446-2) and carried out in accordance with the Declaration of Helsinki of the World Medical Association.
Procedures
Consecutive patients who were diagnosed with severe AS and required TAVI at 10 institutes between May 2016 and March 2019 were eligible for inclusion in this study. Severe AS was defined as an aortic valve area ≦ 1.0 cm2, peak aortic velocity >4.0 m/s, or a mean pressure gradient >40 mm Hg as determined by Doppler echocardiography. The peak aortic pressure gradient was calculated with the modified Bernoulli equation. In this study, the patients were categorized into 2 groups based on the lowest hemoglobin (Hb) value up to 1 year prior to treatment with TAVI according to anemia, which was defined as moderate/severe anemia or no moderate/severe anemia according to the definition of the World Health Organization of anemia: moderate/severe anemia (Hb < 10.9 g/dL) and no moderate/severe anemia (Hb ≥ 11.0 g/dL) [13]. In this study, angiodysplasia was determined according to the definition of vascular superficial lesions and typically presented as a bright red, irregular, round lesion [9, 10].
Outcome Parameters
The main outcomes of this study were the proportion of patients who underwent gastrointestinal endoscopic examinations (esophagogastroduodenoscopy [EGD], colonoscopy [CS], balloon-assisted enteroscopy, and small bowel capsule endoscopy), the proportion of patients who had gastrointestinal angiodysplasia and active bleeding, and the corresponding endoscopic features.
Statistical Analyses
The χ2 test, paired t test, and Mann-Whitney U test were used to compare categorical variables. p values <0.05 were considered statistically significant. All statistical analyses were performed using the SPSS software program (version 22.0 for Windows; IBM Japan, Ltd., Tokyo, Japan).
Results
Baseline Characteristics
A total of 325 patients with severe AS who underwent TAVI were enrolled in this study. There were 151 patients (46%) with moderate/severe anemia and 174 (54%) without moderate/severe anemia. The median hemoglobin levels were 9.9 g/dL in the moderate/severe anemia group and 12.2 g/dL in the no moderate/severe anemia group. Baseline characteristics are summarized in Table 1. Overall, patients with moderate/severe anemia had significantly lower body mass index values (p < 0.01), higher Society of Thoracic Surgeons scores (STS) (p < 0.001), higher frailty scores (≥5, evaluated by Clinical Frailty Score) (p < 0.05), and a higher rate of NYHA III/IV (p < 0.01). Regarding medication, there were no significant differences between patients with and without moderate/severe anemia. Patients with moderate/severe anemia had higher serum creatinine values (p < 0.0001), lower estimated glomerular filtration rates (p < 0.0001), higher BNP values (p < 0.0001), lower albumin values (p < 0.0001), and lower Fe values (p < 0.0001) than those without moderate/severe anemia. Regarding the echocardiographic parameters, Vmax, AVA, left ventricular ejection fraction, wall thickness, and mean PG in patients with moderate/severe anemia and without moderate/severe anemia did not differ to a statistically significant extent.
Patient characteristics
| Variables | Hb ≥11.0 g/dL (N = 151) | Hb <11.0 g/dL (N = 174) | p value |
|---|---|---|---|
| Age, years | 86 (60–96) | 87 (69–95) | 0.0618 |
| Male, n (%) | 45 (29.2) | 41 (24.0) | 0.285 |
| BMI, kg/m2 | 21.7 (14.0–35.6) | 20.6 (14.1–32.7) | 0.0012 |
| STS score, % | 5.6 (0.79–21.1) | 6.5 (1.8–30.9) | 0.0003 |
| Frailty scale ≥ 5, n (%) | 59 (38.3) | 85 (49.7) | 0.0389 |
| NYHA III/IV, n (%) | 44 (28.6) | 73 (42.7) | 0.0081 |
| Hypertension, n (%) | 109 (70.8) | 114 (66.7) | 0.425 |
| Diabetes, n (%) | 35 (22.7) | 38 (22.2) | 0.913 |
| PAD, n (%) | 25 (16.2) | 47 (27.5) | 0.0147 |
| CAD, n (%) | 49 (31.8) | 72 (42.1) | 0.0073 |
| CVA, n (%) | 24 (13.3) | 26 (15.2) | 0.925 |
| AF, n (%) | 18 (11.7) | 30 (17.5) | 0.137 |
| Medication data | |||
| Antiplatelet drug, n (%) | |||
| Aspirin | 51 (33.1) | 73 (42.7) | 0.0761 |
| Thienopyridine | 24 (15.6) | 31 (18.1) | 0.541 |
| Anticoagulant drug, n (%) | |||
| DOAC | 28 (18.2) | 27 (15.8) | 0.566 |
| Warfarin | 4 (2.6) | 8 (4.7) | 0.321 |
| PPI, n (%) | 78 (50.6) | 96 (56.1) | 0.322 |
| Oral iron agent, n (%) | 15 (9.7) | 22 (12.9) | 0.376 |
| NSAIDS, n (%) | 7 (4.5) | 4 (2.4) | 0.272 |
| Laboratory data | |||
| Hb, g/dL | 12.2 (11.0–15.7) | 9.9 (3.7–10.9) | <0.0001 |
| Cr, mg/dL | 0.83 (0.37–2.02) | 0.99 (0.33–8.06) | <0.0001 |
| eGFR, mg/dL | 53.7 (20.0–120.1) | 43.7 (5.7–134.1) | <0.0001 |
| BNP, pg/mL | 178.9 (8.4–2,202) | 327.4 (23.6–6,049) | <0.0001 |
| Alb, g/dL | 3.85 (2.6–5.1) | 3.6 (2.2–4.7) | <0.0001 |
| Fe, μg/dL | 70 (9.0–146) | 54 (8.8–167) | <0.0001 |
| Ferritin, ng/mL | 64 (25–717) | 58 (20–984) | 0.215 |
| Total cholesterol, mg/dL | 189 (104–134) | 175 (61–277) | 0.0011 |
| TTE data | |||
| Peak velocity, m/s | 4.5 (1.9–6.4) | 4.4 (2.5–6.4) | 0.945 |
| Mean pressure gradient, mm Hg | 49.0 (8–100) | 48.0 (15–102) | 0.805 |
| AVA, cm2 | 0.60 (0.20–1.20) | 0.59 (0.20–2.10) | 0.704 |
| LVEF, % | 62 (22–85) | 62 (18–83) | 0.212 |
| Variables | Hb ≥11.0 g/dL (N = 151) | Hb <11.0 g/dL (N = 174) | p value |
|---|---|---|---|
| Age, years | 86 (60–96) | 87 (69–95) | 0.0618 |
| Male, n (%) | 45 (29.2) | 41 (24.0) | 0.285 |
| BMI, kg/m2 | 21.7 (14.0–35.6) | 20.6 (14.1–32.7) | 0.0012 |
| STS score, % | 5.6 (0.79–21.1) | 6.5 (1.8–30.9) | 0.0003 |
| Frailty scale ≥ 5, n (%) | 59 (38.3) | 85 (49.7) | 0.0389 |
| NYHA III/IV, n (%) | 44 (28.6) | 73 (42.7) | 0.0081 |
| Hypertension, n (%) | 109 (70.8) | 114 (66.7) | 0.425 |
| Diabetes, n (%) | 35 (22.7) | 38 (22.2) | 0.913 |
| PAD, n (%) | 25 (16.2) | 47 (27.5) | 0.0147 |
| CAD, n (%) | 49 (31.8) | 72 (42.1) | 0.0073 |
| CVA, n (%) | 24 (13.3) | 26 (15.2) | 0.925 |
| AF, n (%) | 18 (11.7) | 30 (17.5) | 0.137 |
| Medication data | |||
| Antiplatelet drug, n (%) | |||
| Aspirin | 51 (33.1) | 73 (42.7) | 0.0761 |
| Thienopyridine | 24 (15.6) | 31 (18.1) | 0.541 |
| Anticoagulant drug, n (%) | |||
| DOAC | 28 (18.2) | 27 (15.8) | 0.566 |
| Warfarin | 4 (2.6) | 8 (4.7) | 0.321 |
| PPI, n (%) | 78 (50.6) | 96 (56.1) | 0.322 |
| Oral iron agent, n (%) | 15 (9.7) | 22 (12.9) | 0.376 |
| NSAIDS, n (%) | 7 (4.5) | 4 (2.4) | 0.272 |
| Laboratory data | |||
| Hb, g/dL | 12.2 (11.0–15.7) | 9.9 (3.7–10.9) | <0.0001 |
| Cr, mg/dL | 0.83 (0.37–2.02) | 0.99 (0.33–8.06) | <0.0001 |
| eGFR, mg/dL | 53.7 (20.0–120.1) | 43.7 (5.7–134.1) | <0.0001 |
| BNP, pg/mL | 178.9 (8.4–2,202) | 327.4 (23.6–6,049) | <0.0001 |
| Alb, g/dL | 3.85 (2.6–5.1) | 3.6 (2.2–4.7) | <0.0001 |
| Fe, μg/dL | 70 (9.0–146) | 54 (8.8–167) | <0.0001 |
| Ferritin, ng/mL | 64 (25–717) | 58 (20–984) | 0.215 |
| Total cholesterol, mg/dL | 189 (104–134) | 175 (61–277) | 0.0011 |
| TTE data | |||
| Peak velocity, m/s | 4.5 (1.9–6.4) | 4.4 (2.5–6.4) | 0.945 |
| Mean pressure gradient, mm Hg | 49.0 (8–100) | 48.0 (15–102) | 0.805 |
| AVA, cm2 | 0.60 (0.20–1.20) | 0.59 (0.20–2.10) | 0.704 |
| LVEF, % | 62 (22–85) | 62 (18–83) | 0.212 |
Values are the mean ± SD, median (interquartile range), or number (%).
Hb, hemoglobin; BMI, body mass index; STS, Society of Thoracic Surgeons Score; NYHA, New York Heart Association; PAD, peripheral artery disease; CAD, coronary artery disease; CVA, cerebrovascular attack; AF, atrial fibrillation; DOAC, direct oral anticoagulant; PPI, proton pump inhibitor; NSAID, nonsteroidal anti-inflammatory drugs; eGFR, estimated glomerular filtration rate; BNP, brain-derived natriuretic peptide; TTE, transthoracic echocardiography; AVA, aortic valve area; LVEF, left ventricular ejection fraction; LVEDV, left ventricular end-diastolic volume; LVESV, left ventricular end-systolic volume.
Endpoints
Regarding the endoscopic examinations, among all patients, 22.5% (73/325) underwent endoscopic examinations. Regarding the ratio of the patients who underwent endoscopic examinations according to the status of anemia, 27.0% (47/174) of patients in the moderate/severe anemia group and 17.2% (26/151) in the no moderate/severe anemia group had endoscopic examinations (p = 0.035). The results are summarized in Table 2. Regarding the stomach and duodenum, 16.3% (7/43) of patients in the moderate/severe anemia group and 8.3% (2/24) in the no moderate/severe anemia group had angiodysplasia, and 4.7% (2/43) of patients in the moderate/severe anemia group and 0% (0/24) in the no moderate/severe anemia group had active bleeding. Regarding the jejunum and ileum, 71.4% (5/7) of patients in the moderate/severe anemia group and 0% in the no moderate/severe anemia group had angiodysplasia, and 57.1% (4/7) of patients in the moderate/severe anemia group and 0% in the no moderate/severe anemia group had active bleeding. Regarding the colon, 37.0% (10/27) of patients in the moderate/severe anemia group and 0% (0/11) in the no moderate/severe anemia group had angiodysplasia, and 22.2% (6/27) of patients in the moderate/severe anemia group and 0% (0/11) in the no moderate/severe anemia group had active bleeding. The results are summarized in Table 3.
The ratio of the patients who underwent endoscopic examinations according to the status of anemia
| All 325 cases | Hb >11 g/dL (151 cases) | Hb <11 g/dL (174 cases) | p value | |
|---|---|---|---|---|
| Total | 22.5 (73/325) | 17.2 (26/151) | 27.0 (47/174) | 0.035 |
| Esophagogastroduodenoscopy | 20.6 (67/325) | 15.9 (24/151) | 24.7 (43/174) | 0.050 |
| Balloon-assisted enteroscopy | 1.5 (5/325) | 0 (0/151) | 2.9 (5/174) | 0.036 |
| Small bowel capsule endoscopy | 1.2 (4/325) | 0 (0/151) | 2.3 (4/174) | 0.061 |
| Colonoscopy | 11.7 (38/325) | 7.3 (11/151) | 15.5 (27/174) | 0.021 |
| All 325 cases | Hb >11 g/dL (151 cases) | Hb <11 g/dL (174 cases) | p value | |
|---|---|---|---|---|
| Total | 22.5 (73/325) | 17.2 (26/151) | 27.0 (47/174) | 0.035 |
| Esophagogastroduodenoscopy | 20.6 (67/325) | 15.9 (24/151) | 24.7 (43/174) | 0.050 |
| Balloon-assisted enteroscopy | 1.5 (5/325) | 0 (0/151) | 2.9 (5/174) | 0.036 |
| Small bowel capsule endoscopy | 1.2 (4/325) | 0 (0/151) | 2.3 (4/174) | 0.061 |
| Colonoscopy | 11.7 (38/325) | 7.3 (11/151) | 15.5 (27/174) | 0.021 |
Hb, hemoglobin. Values are percentages with numbers in parentheses.
Results of endoscopic examinations according to the status of anemia
| All 325 cases | Hb >11 g/dL (151 cases) | Hb <11 g/dL (174 cases) | p value | |
|---|---|---|---|---|
| Total | 73 cases | 26 cases | 47 cases | |
| Angiodysplasia | 28.8 (21/73) | 7.7 (2/26) | 38.3 (19/47) | <0.0001 |
| Active bleeding | 15.1 (11/73) | 0 (0/26) | 23.4 (11/47) | 0.0074 |
| Stomach – duodenum | 67 cases | 24 cases | 43 cases | |
| Angiodysplasia | 13.4 (9/67) | 8.3 (2/24) | 16.3 (7/43) | 0.3605 |
| Active bleeding | 3.0 (2/67) | 0 (0/24) | 4.7 (2/43) | 0.2834 |
| Jejunum – ileum | 7 cases | 0 cases | 7 cases | |
| Angiodysplasia | 71.4 (5/7) | 0 (0/0) | 71.4 (5/7) | - |
| Active bleeding | 57.1 (4/7) | 0 (0/0) | 57.1 (4/7) | - |
| Colon | 38 cases | 11 cases | 27 cases | |
| Angiodysplasia | 26.3 (10/38) | 0 (0/11) | 37.0 (10/27) | 0.0187 |
| Active bleeding | 15.8 (6/38) | 0 (0/11) | 22.2 (6/27) | 0.0884 |
| All 325 cases | Hb >11 g/dL (151 cases) | Hb <11 g/dL (174 cases) | p value | |
|---|---|---|---|---|
| Total | 73 cases | 26 cases | 47 cases | |
| Angiodysplasia | 28.8 (21/73) | 7.7 (2/26) | 38.3 (19/47) | <0.0001 |
| Active bleeding | 15.1 (11/73) | 0 (0/26) | 23.4 (11/47) | 0.0074 |
| Stomach – duodenum | 67 cases | 24 cases | 43 cases | |
| Angiodysplasia | 13.4 (9/67) | 8.3 (2/24) | 16.3 (7/43) | 0.3605 |
| Active bleeding | 3.0 (2/67) | 0 (0/24) | 4.7 (2/43) | 0.2834 |
| Jejunum – ileum | 7 cases | 0 cases | 7 cases | |
| Angiodysplasia | 71.4 (5/7) | 0 (0/0) | 71.4 (5/7) | - |
| Active bleeding | 57.1 (4/7) | 0 (0/0) | 57.1 (4/7) | - |
| Colon | 38 cases | 11 cases | 27 cases | |
| Angiodysplasia | 26.3 (10/38) | 0 (0/11) | 37.0 (10/27) | 0.0187 |
| Active bleeding | 15.8 (6/38) | 0 (0/11) | 22.2 (6/27) | 0.0884 |
Hb, hemoglobin. Values are percentages with numbers in parentheses.
Outcome Assessment
Patients with moderate/severe anemia had significantly more angiodysplasia and active bleeding. The reasons for undergoing endoscopy can be categorized as suspected gastrointestinal bleeding (anemia, melena, hematochezia, and a positive fecal occult blood test) or unsuspected gastrointestinal bleeding (screening, surveillance, body weight loss, abdominal discomfort, lesion detected by computed tomography, constipation, and others). In the moderate/severe anemia group, regarding the reasons for undergoing EGD and CS, the ratio of reasons categorized as suspected gastrointestinal bleeding was significantly higher than the ratio of reasons categorized as unsuspected gastrointestinal bleeding. The results are summarized in Table 4.
The reasons for undergoing endoscopic examinations
| All | Hb > 11 g/dL | Hb < 11 g/dL | p value | |
|---|---|---|---|---|
| Esophagogastroduodenoscopy | 67 cases | 24 cases | 43 cases | <0.0001 |
| Suspected GI bleeding | 58.2 (39/67) | 20.8 (5/24) | 79.1 (34/43) | |
| Anemia | 28.4 (19/67) | 0 (0/24) | 44.2 (19/43) | |
| Melena | 23.9 (16/67) | 16.7 (4/24) | 27.9 (12/43) | |
| Hematochezia | 6.0 (4/67) | 4.2 (1/24) | 7.0 (3/43) | |
| Unsuspected GI bleeding | 41.8 (28/67) | 79.2 (19/24) | 20.9 (9/43) | |
| Screening | 10.4 (7/67) | 12.5 (3/24) | 9.3 (4/43) | |
| Surveillance | 17.9 (12/67) | 37.5 (9/24) | 7.0 (3/43) | |
| Body weight loss | 3.0 (2/67) | 8.3 (2/24) | 0 (0/43) | |
| Abdominal pain | 1.49 (1/67) | 41.7 (1/24) | 0 (0/43) | |
| Lesion detected by CT | 1.49 (1/67) | 4.2 (1/24) | 0 (0/43) | |
| Others | 4.5 (3/67) | 12.5 (3/24) | 4.7 (2/43) | |
| Balloon-assisted enteroscopy | 5 cases | 0 | 5 cases | - |
| Suspected GI bleeding | ||||
| Anemia | 40 (2/5) | 0 | 40 (2/5) | |
| Melena | 60 (3/5) | 0 | 60 (3/5) | |
| Small bowel capsule endoscopy | 4 cases | 0 | 4 cases | - |
| Suspected GI bleeding | ||||
| Anemia | 100 (4/4) | 0 | 100 (4/4) | |
| Colonoscopy | 38 cases | 11 cases | 27 cases | <0.0001 |
| Suspected GI bleeding | 39.5 (25/38) | 27.3 (3/11) | 81.5 (22/27) | |
| Anemia | 31.6 (12/38) | 0 (0/11) | 44.4 (12/27) | |
| Melena | 13.2 (5/38) | 9.1 (1/11) | 14.8 (4/27) | |
| Hematochezia | 15.8 (6/38) | 9.1 (1/11) | 18.5 (5/27) | |
| FOBT positive results | 5.3 (2/38) | 9.1 (1/11) | 40.7 (11/27) | |
| Unsuspected GI bleeding | 34.2 (13/38) | 72.7 (8/11) | 18.5 (5/27) | |
| Screening | 15.8 (6/38) | 36.4 (4/11) | 7.4 (2/27) | |
| Surveillance | 7.9 (3/38) | 18.2 (2/11) | 3.7 (1/27) | |
| Constipation | 2.6 (1/38) | 9.1 (1/11) | 0 (0/27) | |
| Lesion detected by CT | 7.9 (3/38) | 9.1 (1/11) | 7.4 (2/27) | |
| All | Hb > 11 g/dL | Hb < 11 g/dL | p value | |
|---|---|---|---|---|
| Esophagogastroduodenoscopy | 67 cases | 24 cases | 43 cases | <0.0001 |
| Suspected GI bleeding | 58.2 (39/67) | 20.8 (5/24) | 79.1 (34/43) | |
| Anemia | 28.4 (19/67) | 0 (0/24) | 44.2 (19/43) | |
| Melena | 23.9 (16/67) | 16.7 (4/24) | 27.9 (12/43) | |
| Hematochezia | 6.0 (4/67) | 4.2 (1/24) | 7.0 (3/43) | |
| Unsuspected GI bleeding | 41.8 (28/67) | 79.2 (19/24) | 20.9 (9/43) | |
| Screening | 10.4 (7/67) | 12.5 (3/24) | 9.3 (4/43) | |
| Surveillance | 17.9 (12/67) | 37.5 (9/24) | 7.0 (3/43) | |
| Body weight loss | 3.0 (2/67) | 8.3 (2/24) | 0 (0/43) | |
| Abdominal pain | 1.49 (1/67) | 41.7 (1/24) | 0 (0/43) | |
| Lesion detected by CT | 1.49 (1/67) | 4.2 (1/24) | 0 (0/43) | |
| Others | 4.5 (3/67) | 12.5 (3/24) | 4.7 (2/43) | |
| Balloon-assisted enteroscopy | 5 cases | 0 | 5 cases | - |
| Suspected GI bleeding | ||||
| Anemia | 40 (2/5) | 0 | 40 (2/5) | |
| Melena | 60 (3/5) | 0 | 60 (3/5) | |
| Small bowel capsule endoscopy | 4 cases | 0 | 4 cases | - |
| Suspected GI bleeding | ||||
| Anemia | 100 (4/4) | 0 | 100 (4/4) | |
| Colonoscopy | 38 cases | 11 cases | 27 cases | <0.0001 |
| Suspected GI bleeding | 39.5 (25/38) | 27.3 (3/11) | 81.5 (22/27) | |
| Anemia | 31.6 (12/38) | 0 (0/11) | 44.4 (12/27) | |
| Melena | 13.2 (5/38) | 9.1 (1/11) | 14.8 (4/27) | |
| Hematochezia | 15.8 (6/38) | 9.1 (1/11) | 18.5 (5/27) | |
| FOBT positive results | 5.3 (2/38) | 9.1 (1/11) | 40.7 (11/27) | |
| Unsuspected GI bleeding | 34.2 (13/38) | 72.7 (8/11) | 18.5 (5/27) | |
| Screening | 15.8 (6/38) | 36.4 (4/11) | 7.4 (2/27) | |
| Surveillance | 7.9 (3/38) | 18.2 (2/11) | 3.7 (1/27) | |
| Constipation | 2.6 (1/38) | 9.1 (1/11) | 0 (0/27) | |
| Lesion detected by CT | 7.9 (3/38) | 9.1 (1/11) | 7.4 (2/27) | |
Values are percentages with numbers in parentheses. Hb, hemoglobin; GI bleeding, gastrointestinal bleeding; CT, computed tomography; FOBT, fecal occult blood test.
All 21 patients with angiodysplasia or active bleeding are summarized in Table 5. Figure 1 (a–o) shows gastric and duodenal angiodysplasias. Figure 2 (a–m) shows jejunal and ileal angiodysplasias. Figure 3 (a–v) shows colonic angiodysplasias. Nine patients (case Nos. 1–9) had angiodysplasia of the stomach or duodenum, with 44.4% (4/9) of the patients having angiodysplasia of the lower stomach. Five patients (case Nos. 10–14) had angiodysplasia of the small intestine. These cases involved multiple sites, with angiodysplasia located at the jejunum in 80% of the cases (4/5). In all 10 cases (case Nos. 3, 5, 14–21) of angiodysplasia of the colon, the location was the right colon. Six patients had hemorrhages after the treatment of severe AS (case Nos. 3, 5, 10, 12, 15, and 17).
Twenty-one cases of severe AS patients with angiodysplasia
| Case No. | Age/sex | Hb, g/dL | Indication for endoscopy | Location of angiodysplasia | Solitary or multiple | Active bleeding | Hemostasis | Active bleeding date from TAVI | Blood transfusion | Figure |
|---|---|---|---|---|---|---|---|---|---|---|
| 1 | 89/F | 9.4 | Melena | Stomach (lower) | Multiple | + | APC | Before 69, 59 days | − | 1a, 1b, 1c |
| 2 | 89/F | 7.8 | Melena | Stomach (lower) | Multiple | − | - | + | 1d | |
| 3 | 90/F | 10.3 | Hematochezia | Stomach (lower) | Multiple | − | - | + | 1e | |
| Colon (right) | Multiple | + | APC, clip | Before 120 days | 3j, 3k | |||||
| After 64 days | ||||||||||
| 4 | 84/M | 10.0 | Melena | Stomach (lower) | Multiple | + | APC | Before 770, 570 days | + | 1f, 1g |
| Duodenum | Solitary | − | - | 1o | ||||||
| 5 | 94/F | 10.0 | Hematochezia | Stomach (upper, middle) | Multiple | − | - | 1h, 1i | ||
| Colon (right) | Multiple | + | APC | After 45 days | + | 3i | ||||
| 6 | 91/F | 10.2 | Melena | Stomach (middle) | Multiple | − | - | + | 1j, 1k | |
| 7 | 92/F | 9.9 | Screening | Stomach (middle) | Solitary | − | - | + | 1l | |
| 8 | 84/F | 12.6 | Surveillance | Stomach (middle) | Solitary | − | - | − | 1m | |
| 9 | 83/F | 12.1 | Melena | Stomach (upper) | Solitary | − | - | − | 1n | |
| 10 | 87/F | 3.7 | Hematochezia, melena | Jejunum | Multiple | + | APC, clip | Before 40, 160, 270 days | + | 2a, 2b, 2c, 2d, 2e, 2f |
| After 30 days | ||||||||||
| 11 | 87/F | 10.4 | Melena | Jejunum, Ileum | Multiple | + | APC | Before 120, 200 days | + | 2g, 2h, 2i, 2j |
| 12 | 87/F | 6.4 | Hematochezia | Jejunum | Multiple | + | APC | After 30 days | + | 2k |
| 13 | 86/F | 7.2 | Melena | Jejunum | Multiple | + | APC, clip | Before 200 days | + | 2l |
| 14 | 91/M | 6.8 | Hematochezia | Ileum | Multiple | − | - | + | 2m | |
| Colon (right) | Multiple | − | - | 3v | ||||||
| 15 | 87/F | 9.5 | Hematochezia | Colon (right) | Multiple | + | Coagulation | Before 10, 25 days | + | 3a, 3b, 3c, 3d, 3e |
| After 15 days | ||||||||||
| 16 | 87/F | 6.9 | Hematochezia | Colon (right) | Multiple | + | APC, clip | Before 93 days | − | 3f, 3g, 3h |
| After 377 days | ||||||||||
| 17 | 89/F | 9.7 | Hematochezia | Colon (right) | Multiple | + | Clip | Before 40 days | + | 3l |
| 18 | 85/M | 8.3 | Hematochezia | Colon (right) | Multiple | − | - | + | 3m, 3n | |
| 19 | 83/F | 8.7 | Anemia | Colon (right) | Multiple | − | - | − | 3o, 3p | |
| 20 | 80/M | 8.3 | Screening | Colon (right) | Multiple | − | - | − | 3q, 3r, 3s | |
| 21 | 69/F | 8.9 | Hematochezia | Colon (right) | Multiple | − | - | − | 3t, 3u |
| Case No. | Age/sex | Hb, g/dL | Indication for endoscopy | Location of angiodysplasia | Solitary or multiple | Active bleeding | Hemostasis | Active bleeding date from TAVI | Blood transfusion | Figure |
|---|---|---|---|---|---|---|---|---|---|---|
| 1 | 89/F | 9.4 | Melena | Stomach (lower) | Multiple | + | APC | Before 69, 59 days | − | 1a, 1b, 1c |
| 2 | 89/F | 7.8 | Melena | Stomach (lower) | Multiple | − | - | + | 1d | |
| 3 | 90/F | 10.3 | Hematochezia | Stomach (lower) | Multiple | − | - | + | 1e | |
| Colon (right) | Multiple | + | APC, clip | Before 120 days | 3j, 3k | |||||
| After 64 days | ||||||||||
| 4 | 84/M | 10.0 | Melena | Stomach (lower) | Multiple | + | APC | Before 770, 570 days | + | 1f, 1g |
| Duodenum | Solitary | − | - | 1o | ||||||
| 5 | 94/F | 10.0 | Hematochezia | Stomach (upper, middle) | Multiple | − | - | 1h, 1i | ||
| Colon (right) | Multiple | + | APC | After 45 days | + | 3i | ||||
| 6 | 91/F | 10.2 | Melena | Stomach (middle) | Multiple | − | - | + | 1j, 1k | |
| 7 | 92/F | 9.9 | Screening | Stomach (middle) | Solitary | − | - | + | 1l | |
| 8 | 84/F | 12.6 | Surveillance | Stomach (middle) | Solitary | − | - | − | 1m | |
| 9 | 83/F | 12.1 | Melena | Stomach (upper) | Solitary | − | - | − | 1n | |
| 10 | 87/F | 3.7 | Hematochezia, melena | Jejunum | Multiple | + | APC, clip | Before 40, 160, 270 days | + | 2a, 2b, 2c, 2d, 2e, 2f |
| After 30 days | ||||||||||
| 11 | 87/F | 10.4 | Melena | Jejunum, Ileum | Multiple | + | APC | Before 120, 200 days | + | 2g, 2h, 2i, 2j |
| 12 | 87/F | 6.4 | Hematochezia | Jejunum | Multiple | + | APC | After 30 days | + | 2k |
| 13 | 86/F | 7.2 | Melena | Jejunum | Multiple | + | APC, clip | Before 200 days | + | 2l |
| 14 | 91/M | 6.8 | Hematochezia | Ileum | Multiple | − | - | + | 2m | |
| Colon (right) | Multiple | − | - | 3v | ||||||
| 15 | 87/F | 9.5 | Hematochezia | Colon (right) | Multiple | + | Coagulation | Before 10, 25 days | + | 3a, 3b, 3c, 3d, 3e |
| After 15 days | ||||||||||
| 16 | 87/F | 6.9 | Hematochezia | Colon (right) | Multiple | + | APC, clip | Before 93 days | − | 3f, 3g, 3h |
| After 377 days | ||||||||||
| 17 | 89/F | 9.7 | Hematochezia | Colon (right) | Multiple | + | Clip | Before 40 days | + | 3l |
| 18 | 85/M | 8.3 | Hematochezia | Colon (right) | Multiple | − | - | + | 3m, 3n | |
| 19 | 83/F | 8.7 | Anemia | Colon (right) | Multiple | − | - | − | 3o, 3p | |
| 20 | 80/M | 8.3 | Screening | Colon (right) | Multiple | − | - | − | 3q, 3r, 3s | |
| 21 | 69/F | 8.9 | Hematochezia | Colon (right) | Multiple | − | - | − | 3t, 3u |
Hb, hemoglobin; TAVI, transcatheter aortic valve implantation; APC, argon plasma coagulation.
Typical endoscopic findings of angiodysplasia of the stomach. a Case No. 1. EGD showed multiple angiodysplasia with active bleeding in the lower stomach. b, c Case No. 1. EGD showed multiple angiodysplasia in the lower stomach. d Case No. 2. EGD showed multiple angiodysplasia in the lower stomach. e Case No. 3. EGD showed multiple angiodysplasia in the lower stomach. f, g Case No. 4. EGD showed multiple angiodysplasia with active bleeding in the lower stomach. h, i Case No. 5. EGD showed multiple angiodysplasia in the upper and middle stomach. j, k Case No. 6. EGD showed multiple angiodysplasia in the middle stomach. l Case No. 7. EGD showed solitary angiodysplasia in the middle stomach. m Case No. 8. EGD showed solitary angiodysplasia in the upper stomach. n Case No. 9. EGD showed solitary angiodysplasia in the upper stomach. o Case No. 4. EGD showed solitary angiodysplasia in the duodenum. EGD, esophagogastroduodenoscopy; APC, argon plasma coagulation.
Typical endoscopic findings of angiodysplasia of the stomach. a Case No. 1. EGD showed multiple angiodysplasia with active bleeding in the lower stomach. b, c Case No. 1. EGD showed multiple angiodysplasia in the lower stomach. d Case No. 2. EGD showed multiple angiodysplasia in the lower stomach. e Case No. 3. EGD showed multiple angiodysplasia in the lower stomach. f, g Case No. 4. EGD showed multiple angiodysplasia with active bleeding in the lower stomach. h, i Case No. 5. EGD showed multiple angiodysplasia in the upper and middle stomach. j, k Case No. 6. EGD showed multiple angiodysplasia in the middle stomach. l Case No. 7. EGD showed solitary angiodysplasia in the middle stomach. m Case No. 8. EGD showed solitary angiodysplasia in the upper stomach. n Case No. 9. EGD showed solitary angiodysplasia in the upper stomach. o Case No. 4. EGD showed solitary angiodysplasia in the duodenum. EGD, esophagogastroduodenoscopy; APC, argon plasma coagulation.
Typical endoscopic findings of angiodysplasia of the small intestine. a Case No. 7. Capsule endoscopy showed multiple angiodysplasia in the jejunum. b–f Case No. 10. Balloon-assisted enteroscopy showed multiple angiodysplasia with active bleeding in the jejunum. g–j Case No. 11. Balloon-assisted enteroscopy showed multiple angiodysplasia with active bleeding in the jejunum and ileum. k Case No. 12. Balloon-assisted enteroscopy showed multiple angiodysplasia with active bleeding in the jejunum. l Case No. 13. Balloon-assisted enteroscopy showed multiple angiodysplasia with active bleeding in the jejunum. m Case No. 14. Capsule endoscopy showed angiodysplasia in the ileum. APC, argon plasma coagulation.
Typical endoscopic findings of angiodysplasia of the small intestine. a Case No. 7. Capsule endoscopy showed multiple angiodysplasia in the jejunum. b–f Case No. 10. Balloon-assisted enteroscopy showed multiple angiodysplasia with active bleeding in the jejunum. g–j Case No. 11. Balloon-assisted enteroscopy showed multiple angiodysplasia with active bleeding in the jejunum and ileum. k Case No. 12. Balloon-assisted enteroscopy showed multiple angiodysplasia with active bleeding in the jejunum. l Case No. 13. Balloon-assisted enteroscopy showed multiple angiodysplasia with active bleeding in the jejunum. m Case No. 14. Capsule endoscopy showed angiodysplasia in the ileum. APC, argon plasma coagulation.
Typical endoscopic findings of angiodysplasia of the colon. a–e Case No. 15. Colonoscopy showed multiple angiodysplasia with active bleeding in the right colon. f–h Case No. 16. Colonoscopy showed multiple angiodysplasia with active bleeding in the right colon. i Case No. 5. Colonoscopy showed multiple angiodysplasia with active bleeding in the right colon. j, k Case No. 3. Colonoscopy showed multiple angiodysplasia with active bleeding in the right colon. l Case No. 17. Colonoscopy showed multiple angiodysplasia with active bleeding in the right colon. m, n Case No.18. Colonoscopy showed multiple angiodysplasia in the right colon. o, p Case No.19. Colonoscopy showed multiple angiodysplasia in the right colon. q–s Case No. 20. Colonoscopy showed multiple angiodysplasia in the right colon. t, u Case No. 21. Colonoscopy showed multiple angiodysplasia in the right colon. v Case No. 11. Capsule endoscopy showed solitary angiodysplasia in the right colon. APC, argon plasma coagulation.
Typical endoscopic findings of angiodysplasia of the colon. a–e Case No. 15. Colonoscopy showed multiple angiodysplasia with active bleeding in the right colon. f–h Case No. 16. Colonoscopy showed multiple angiodysplasia with active bleeding in the right colon. i Case No. 5. Colonoscopy showed multiple angiodysplasia with active bleeding in the right colon. j, k Case No. 3. Colonoscopy showed multiple angiodysplasia with active bleeding in the right colon. l Case No. 17. Colonoscopy showed multiple angiodysplasia with active bleeding in the right colon. m, n Case No.18. Colonoscopy showed multiple angiodysplasia in the right colon. o, p Case No.19. Colonoscopy showed multiple angiodysplasia in the right colon. q–s Case No. 20. Colonoscopy showed multiple angiodysplasia in the right colon. t, u Case No. 21. Colonoscopy showed multiple angiodysplasia in the right colon. v Case No. 11. Capsule endoscopy showed solitary angiodysplasia in the right colon. APC, argon plasma coagulation.
Discussion
This study investigated the frequency of and endoscopic features of Heyde’s syndrome among patients with severe AS. The results showed that although only 21% of patients who underwent TAVI underwent gastrointestinal endoscopic examinations, the patients with moderate/severe anemia more often had gastrointestinal angiodysplasia than those without moderate/severe anemia. We also showed the endoscopic features of a large series of 21 patients with Heyde’s syndrome.
A previous study showed that in 3,403 patients with severe AS, 38% (1,282 patients) had moderate/severe anemia (Hb ≤ 10.9 g/dL) [5]. Heart disease could be the cause of gastrointestinal angiodysplasia bleeding and anemia [14]. Previous studies also showed that 21% of patients with critical AS who underwent aortic valve replacement had a history of dermal or mucosal bleeding [12]. As expected, our results showed that 52% of patients who underwent TAVI had moderate/severe anemia (Hb ≤ 10.9 g/dL). However, our study showed that only approximately 21% of patients who underwent TAVI and had moderate/severe anemia underwent gastrointestinal endoscopic examinations. Thirty-eight percent of those with moderate/severe anemia who underwent gastrointestinal endoscopic examinations had gastrointestinal angiodysplasia. Fifty-eight percent of the patients with moderate/severe anemia who had angiodysplasia had active bleeding. The patients with moderate/severe anemia had significantly more angiodysplasia (38.3% vs. 7.7%; p < 0.0001) and more active bleeding (23.4% vs. 0%; p < 0.01) than patients with no moderate/severe anemia. Angiodysplasia and active bleeding were more often found in the jejunum, ileum, and colon than in the stomach. However, almost half of the patients underwent EGD alone and did not undergo CS and enteroscopy. The positive finding rate may have increased if more patients underwent CS or enteroscopy.
Previous case reports showed that patients with severe AS had angiodysplasia of the stomach, small intestine, and colon [15‒19]. As expected, our study showed that 21 of 73 patients with severe AS who underwent endoscopic examinations and TAVI had angiodysplasia located in the stomach, small intestine, or colon. Among cases involving the stomach, 44% (4/9) had angiodysplasia of the lower stomach. Among cases involving the small intestine, 80% (4/5) had angiodysplasia of the jejunum. Among cases involving the colon, all patients (10/10) had angiodysplasia of the right colon. The differences in the frequency of gastrointestinal angiodysplasia depending on the part of the gastrointestinal tract could be associated with the differences in blood flow depending on the part of the gastrointestinal tract.
The present study showed that among the patients who underwent TAVI, approximately 21% underwent EGD, approximately 12% underwent CS, and only 1.5% underwent balloon-assisted enteroscopy or 1.2% capsule endoscopy. This may be plausible, as the median age of patients who underwent TAVI was 86 years old, and heart function is generally in decline in such elderly individuals.
The present study also showed that 29% of patients who underwent endoscopic examinations had angiodysplasia, and 15% of patients had active bleeding. This may be plausible by several mechanisms. First, mucosal and submucosal hypoxia due to reduced cardiac output causes enhanced angiogenesis through the increased proliferation of vascular endothelial growth factor [9, 20]. Second, the loss of VWF causes angiogenesis via an Ang-2/Tie-2/VEGFR-2 pathway [11]. Third, the loss of VWF causes hemorrhaging [12]. Angiodysplasia is a basal disease, most likely congenital, while bleeding can be triggered by severe AS. Although we did not evaluate VWF multimers and changes in gastrointestinal angiodysplasia after TAVI, we are preparing a prospective study to examine these subjects. Although previous studies showed that gastrointestinal angiodysplasia in patients with severe AS disappeared after aortic valve replacement and TAVI [15‒19, 21], 6 patients had hemorrhages after TAVI, from 15 days to 377 days after treatment of severe AS in our study. The resolution of high shear stress at the site of the stenotic aortic valve could improve anemia and lead to the reduction or disappearance of angiodysplasia through these mechanisms. Further studies will be needed to evaluate the effects of the resolution of high shear stress on angiodysplasia.
Several limitations associated with the present study warrant mention. First, our study was a retrospective study. Second, not all patients underwent gastrointestinal endoscopic examinations. Third, we did not evaluate VWF multimers. We were confident in the diagnosis of Heyde’s syndrome in our 21 cases because they all showed severe AS with gastrointestinal angiodysplasia. Fourth, we did not evaluate the changes in gastrointestinal angiodysplasia after TAVI. Fifth, we did not evaluate patients who underwent surgical aortic valve replacement.
In summary, we performed a retrospective study to examine the frequency of Heyde’s syndrome and its endoscopic features. We showed that severe AS patients with moderate/severe anemia frequently had gastrointestinal angiodysplasia and active bleeding throughout the entire gastrointestinal tract. Almost all patients had multiple lesions. Therefore, we should carefully monitor gastrointestinal angiodysplasia in severe AS patients with moderate/severe anemia. Based on the results, CS and enteroscopy should be considered in patients with negative EGD results. Further studies will be needed to evaluate the precise frequency of Heyde’s syndrome, the correlation between VWF large multimers and the number of angiodysplasia lesions, and the efficacy of resolving high shear stress for gastrointestinal angiodysplasia.
Acknowledgments
We thank all members of the Department of Molecular Gastroenterology and Hepatology and the Department of Cardiovascular Medicine, Kyoto Prefectural University of Medicine, for their help with this study. We thank Dr. Hiroshi Fujita and Dr. Kazuhiro Katada of Kyoto Prefectural University of Medicine North Medical Center, Dr. Daisuke Kanbayashi and Dr. Toshifumi Tsuji of Fukuchiyama City Hospital, Dr. Satoshi Akabame and Dr. Masatoshi Miyata of Kyoto Okamoto Memorial Hospital, Dr. Takashi Okada and Dr. Takayuki Motoyoshi of Kyoto City Hospital, Dr. Keiji Inoue and Dr. Takuji Kawamura of Japanese Red Cross Kyoto Daini Hospital, Dr. Shin Nakamori and Dr. Yoshikazu Inagaki of Nishijin Hospital, Dr. Tatsuya Kawasaki and Dr. Kazuhiro Kamada of Matsushita Memorial Hospital, Dr. Yasuhiro Yamahara of Rakusai Shimizu Hospital, and Dr. Hirokazu Yokoi of Rakuwakai Otowa Hospital. We would like to thank Brian Quinn, Japan Medical Communication for English language editing.
Statement of Ethics
All patients provided their written informed consent to undergo TAVI. The study was approved by the Clinical Ethics Committee on Human Experiments of Kyoto Prefectural University of Medicine (IRB registration number, ERB-C-1446-2). All procedures followed were in accordance with the Ethical Standards of the Responsible Committee on Human Experimentation (institutional and national) and with the 1964 Declaration of Helsinki and later versions.
Conflicts of Interest Statement
Kan Zen received a Scholarship donation from Abbott Japan and was a paid speaker and teacher for Edwards Lifescience and Medtronic. Naohisa Yoshida and Osamu Dohi received a research grant from Fujifilm Co. (J162001222). The other authors declare no conflicts of interest.
Funding Sources
This work was supported by the Japanese Foundation for Research and Promotion of Endoscopy (JFE) Grant, a multicenter clinical research Grant of the Japanese Gastrointestinal Association, and JSPS KAKENHI Grant No. JP21K07967.
Author Contributions
Conception and design: Satoshi Sugino, Ken Inoue, Kan Zen, and Masaki Yashige. Analysis and interpretation of the data: Reo Kobayashi, Kazuaki Takamatsu, Nobuyasu Ito, Naoto Iwai, Ryohei Hirose, Toshifumi Doi, Osamu Dohi, Naohisa Yoshida, Kazuhiko Uchiyama, Tomohisa Takagi, Takeshi Ishikawa, and Hideyuki Konishi. Drafting of the article: Satoshi Sugino and Ken Inoue. Critical revision of the article for important intellectual content: Ken Inoue and Kan Zen. Final approval of the article: Satoaki Matoba and Yoshito Itoh.
Data Availability Statement
All data generated or analyzed during this study are included in this article. Further inquiries can be directed to the corresponding author.
References
