Author + information
- Received November 2, 2016
- Revision received April 5, 2017
- Accepted April 11, 2017
- Published online December 4, 2017.
- Venkat Vuddanda, MDa,
- Mohammad-Ali Jazayeri, MDa,
- Mohit K. Turagam, MDb,
- Madhav Lavu, MDa,
- Valay Parikh, MDa,
- Donita Atkins, BSa,
- Sudharani Bommana, MPhila,
- Madhu Reddy Yeruva, MDa,
- Luigi Di Biase, MD, PhDc,
- Jie Cheng, MDd,
- Vijay Swarup, MDe,
- Rakesh Gopinathannair, MDf,
- Mojtaba Olyaee, MDa,
- Vijay Ivaturi, PhDg,
- Andrea Natale, MDc and
- Dhanunjaya Lakkireddy, MDa,∗ ()
- aCardiovascular Research Institute, University of Kansas Hospital, Kansas City, Kansas
- bDivision of Cardiology, University of Missouri, Columbia, Missouri
- cTexas Cardiac Arrhythmia Institute, St. David’s Medical Center, Austin, Texas
- dTexas Heart Institute, St. Luke’s Hospital, Houston, Texas
- eArizona Heart Rhythm Center, Phoenix, Arizona
- fDivision of Cardiology, University of Louisville, Louisville, Kentucky
- gDepartment of Pharmacy Practice and Science University of Maryland, Baltimore
- ↵∗Address for correspondence:
Dr. Dhanunjaya Lakkireddy, Division of Cardiovascular Diseases, Cardiovascular Research Institute, The University of Kansas Hospital, 3901 Rainbow Boulevard, Kansas City, Kansas 66160.
Objectives The present study describes the use of octreotide (OCT) in patients with atrial fibrillation (AF) receiving oral anticoagulation (OAC) who have gastrointestinal (GI) bleeding related to arteriovenous malformations (AVMs), as well as its effect on OAC tolerance and subsequent rebleeding.
Background AVMs cause significant GI bleeding, especially in patients with AF who are receiving OAC for stroke prevention. OCT has been shown to minimize recurrent GI bleeds related to AVMs.
Methods In a multicenter, observational study, 38 AF patients with contraindications to OAC because of AVM-related GI bleeding were started on 100 μg of subcutaneous OCT twice daily. OAC was resumed in all patients within 48 h. Incidence of recurrent GI bleeds was calculated, and hemoglobin levels were recorded at enrollment and at 3 and 6 months’ follow-up.
Results After a median follow-up of 8 months, 36 patients (mean age 69 ± 8.0 years; mean CHA2DS2-VASc score 3 ± 1 and mean HAS-BLED score 3 ± 1) were available for analysis. All were able to successfully resume OAC, and 28 of 36 (78%) remained on OAC at the conclusion of the study, whereas 8 underwent left atrial appendage closure with subsequent OAC discontinuation. No systemic thromboembolic events occurred in follow-up. Of the 28 patients who continued receiving OAC, 19 (68%) were free of recurrent GI bleed, 4 had minor GI bleeds, 4 required transfusion, and 1 required colectomy for GI bleeding. Mean hemoglobin levels in all patients receiving OAC were significantly higher at 3- and 6-month follow-up than at baseline (p < 0.001).
Conclusions Subcutaneous OCT therapy is an attractive option in AF patients receiving OAC who have AVM-related GI bleeds. It allows successful reinitiation of OAC as a bridge to left atrial appendage exclusion or short-term relief from bleeding.
Atrial fibrillation (AF) is the most common cardiac arrhythmia worldwide (1). It imparts significant stroke risk and is associated with a >2-fold increase in the odds of developing silent cerebral infarctions (2). Systemic oral anticoagulation (OAC) is currently the mainstay of therapy to reduce the thromboembolic complications associated with AF (3); however, its use may be limited in patients at high risk of bleeding, particularly gastrointestinal (GI) bleeding (4–6). Arteriovenous malformations (AVMs) such as angiodysplasia and hemorrhagic telangiectasias of the GI tract account for 40% to 60% of lower GI bleeds (7,8). GI bleeding often forces withdrawal of OAC and places patients at high risk of systemic thromboembolism. First-line management options for AVMs include endoscopic therapy with argon plasma coagulation, fluoroscopy-guided vascular embolization, and surgery. Most of the aforementioned options can be limited by patient factors, procedural risks, and anatomic factors such as the multifocal nature of the AVMs or poor accessibility of the culprit lesion, resulting in recurrent GI bleeds in 30% to 40% of cases despite the above interventions (9).
There is some evidence that pharmacotherapy with somatostatin analogues such as octreotide (OCT) is an effective and well-tolerated option to prevent recurrent GI bleeds in cases where endoscopic or surgical therapy is not feasible or is unsuccessful. The benefit is much more profound in patients with coagulopathies or obligate need for OAC use (10). Current literature on this topic is limited to anecdotal case reports. We sought to investigate whether OCT therapy can facilitate safe reinitiation of OAC in AF patients with a high risk of stroke and GI bleeding due to AVMs.
In this multicenter, observational study, 150 AF patients with contraindication to OAC because of GI bleeding (defined as the appearance of melena, hematochezia, or guaiac-positive stool and a new drop in hemoglobin) were identified at the cardiac electrophysiology clinic. Sixty patients with GI bleeding related to AVMs (Figure 1) or of obscure etiology (no pathology identified on endoscopy) were screened. Small-bowel AVMs are hard to identify and have been shown to account for a large percentage of obscure GI bleeds (11). Patients with GI bleeding secondary to other causes were excluded. Patients with a ventricular assist device, malignancy, thrombocytopenia (platelet count <150,000/μl), chronic liver disease, chronic kidney disease, and active infection were excluded. Of 60 patients screened, 55 met inclusion and exclusion criteria. Seventeen patients could not obtain OCT because of insurance coverage–related issues. Thirty-eight patients were ultimately enrolled in the study (Figure 2). Approval was obtained from the local institutional review boards.
Patients were started on OCT therapy (100 μg twice daily given subcutaneously). Although monthly intramuscular injection of a long-acting OCT formulation is available, daily injections were chosen because of cost implications. There were no significant differences between the 2 formulations in terms of efficacy or safety. OAC was resumed 48 h after initiation of OCT therapy. Choice of OAC was based on physician preference.
Baseline demographic information, patient characteristics, medical history, medication details, data on recurrent GI bleed and systemic thromboembolism (stroke, transient ischemic attack, and splenic infarct), and hemoglobin levels at baseline and at 3 and 6 months’ follow-up were collected.
Recurrent GI bleed was defined as any clinically suspected or documented bleeding from the GI tract as indicated by a new drop in hemoglobin and the appearance of melena, hematochezia, or guaiac-positive stools. All GI bleeding events were further characterized based on the need for blood transfusion or invasive intervention. GI bleeds that required blood transfusion or colectomy were considered for left atrial appendage closure (LAAC) using the LARIAT Suture delivery system (SentreHEART Inc., Palo Alto, California) or WATCHMAN endocardial occlusion device (Boston Scientific, Marlborough, Massachusetts). If patients were not candidates for LAAC, their OCT dose was up-titrated to either 200 or 300 μg at the physician’s discretion, and OAC therapy was continued. If patients experienced recurrent GI bleeding despite the above interventions, their OAC therapy was discontinued.
Categorical variables are presented as a frequency (n) or percentage, and continuous variables are expressed as mean ± SD or median (interquartile range). Categorical variables were compared with chi-square test or Fisher exact test. Repeated-measures data analysis was performed with a mixed-effects regression model in the entire cohort who continued receiving OAC, and cohorts were stratified by presence or absence of GI bleed if assumptions for normality were met. If assumptions of normality were not met, continuous variables were compared with the Friedman test. A value of p < 0.05 was considered statistically significant. Statistical analysis was performed with IBM SPSS Statistics version 23.0 (IBM, Armonk, New York).
Thirty-eight patients who met the inclusion criteria were enrolled in the study. Two patients were lost to follow-up, whereas 8 underwent LAAC and had OAC and OCT discontinued before the 3-month follow-up visit. Baseline characteristics of the study population are shown in Table 1.
The mean age was 69 ± 8.0 years, and patients were predominantly white women (69.4%) with nonparoxysmal AF (52.8%). The mean CHA2DS2VASc score (a clinical stroke risk prediction model) was 3 ± 1, and the HAS-BLED score (which estimates risk of major bleeding for patients receiving anticoagulation) was 3 ± 1. Before discontinuation of OAC, 44% of patients were taking warfarin, 22.2% were taking apixaban, 16.7% were taking rivaroxaban, and 16.7% were taking dabigatran. A left atrial appendage clot was identified in 5 of 36 patients (13.8%), and systemic thromboembolic events were reported in 8 of 36 patients (22%) before enrollment. Median left atrial size was 4.25 cm (interquartile range: 3.70 to 4.75 cm), and median left ventricle ejection fraction was 60% (interquartile range: 55% to 65%). The most common source of GI bleed before enrollment was small intestine (42%), followed by colon (25%). Angiodysplasia was the most commonly identified pathology (76%). One third of the patients (33.3%) underwent prior endoscopic intervention with argon plasma coagulation of identifiable lesions before enrollment.
Median follow-up duration was 8 months (range 6 to 13 months). Among the study population, 8 of 36 patients (22%) underwent LAAC for stroke prevention (LARIAT in 4 patients, WATCHMAN in 3 patients, and AtriClip [AtriCure, West Chester, Ohio] in 1 patient) with subsequent discontinuation of OAC and OCT therapy. Before LAAC, while still receiving OAC/OCT therapy, 4 of 8 (50%) had recurrent GI bleeds, and of these, 3 (37.5%) were major bleeds that required blood transfusion, whereas 1 was a minor bleed that required no intervention (Figure 3).
The remaining 28 of 36 patients (77.8%) continued OAC with apixaban in 15 cases (53.5%), rivaroxaban in 11 (39.3%), and warfarin in 2 (7.1%). Among the 28 patients who continued to receive OAC, 19 (67.8%) had no recurrent GI bleeds, whereas 4 (14.2%) had minor GI bleeds and 4 (14.2%) had major GI bleeds that required blood transfusion. In 1 other case, a patient had a major GI bleed that ultimately required colectomy (Figure 4, Table 2).
In those patients who remained on OAC throughout the study period (n = 28), the mean hemoglobin levels were significantly higher at 3 months (9.33 g/dl vs. 7.49 g/dl; p < 0.001) and 6 months (11.10 g/dl vs. 7.49 g/dl; p < 0.001) than at baseline (Figure 5). In 19 of the 28 patients free of recurrent GI bleeding while receiving OAC/OCT therapy (68%), mean hemoglobin levels were significantly higher at 3 months (9.62 ± 0.87 g/dl) and 6 months (11.69 ± 0.91 g/dl) than at baseline (7.55 ± 1.08 g/dl). A nonparametric Friedman test of differences among mean hemoglobin levels at baseline and 3- and 6-month follow-up was conducted and rendered a chi-square value of 36, which was significant (p < 0.01) (Figure 6). In 9 of the 28 patients with recurrent GI bleeding while undergoing OAC/OCT therapy (32%), mean hemoglobin levels were not significantly higher at 3-month follow-up than at baseline (8.72 g/dl vs. 7.55 g/dl; p = 0.56), but there was a statistically significant difference at 6-month follow-up (9.84 g/dl vs. 7.55 g/dl; p = 0.02). Of 28 patients who remained on OAC, 8 had no identifiable cause on endoscopy and were labeled as having “obscure etiology.” Because the majority of these patients were clinically similar to patients with confirmed AVMs, we included them as a single cohort. The results discussed herein retained statistical significance even in the cohorts stratified on the basis of the cause of GI bleed (AVMs vs. obscure etiology). Please see the Online Appendix for further details and comparisons.
In addition, 13 of 36 patients (36%) underwent successful cardioversion and were given antiarrhythmic drugs for maintenance of sinus rhythm and continued receiving OAC/OCT. Throughout the follow-up period, there were no reported events of systemic thromboembolism or intracranial hemorrhage. OCT therapy enabled systemic thromboembolism prevention by either continued OAC use (78%) or LAAC (22%). Among the patients in whom OAC was continued, 68% remained free of recurrent GI bleeding during the study period. Patients reported no side effects from OCT (hypothyroidism, bradycardia, or gallbladder dysfunction) during the study period.
GI bleeding is a common side effect of OAC (odds ratio: 1.45; 95% confidence interval: 1.07 to 1.97) (4–6). Often, diagnostic endoscopy reveals either no identifiable pathology or intestinal pathology such as AVMs. In such situations, continued OAC use can result in higher rebleeding rates. No clinical trials currently address the safety of reinitiating OAC after major GI bleeding (12–16). Our study is the first multicenter, observational study evaluating the safety of reinitiation of OAC in AF patients with AVM-related GI bleeds using concomitant OCT therapy. We showed that mean hemoglobin levels were significantly higher at 3 and 6 months compared with baseline.
AVM-related GI bleeds
AVMs account for ∼5% of upper GI bleeds and 40% to 60% of lower GI bleeds (17). They are pathologically dilated communications between thin-walled veins, venules, and capillaries located in the mucosa and submucosa of the GI system (18). Their pathogenesis remains unclear. Increased expression of angiogenic factors (basic fibroblast growth factor and vascular endothelial growth factor [VEGF]) is likely to play a major role (19). Acquired von Willebrand disease due to valvular heart disease (e.g., aortic stenosis) in elderly patients is considered to be another possible mechanism and is mediated by negative modulation of the VEGF receptor (20,21). Increased expression of VEGF exerts direct action on endothelial cells, resulting in increased production of tissue factor or plasminogen activators. This increase in local fibrinolytic activity could contribute to the increased bleeding tendency and high rebleeding rates despite interventions (22). Rebleeding might be higher in patients who require OAC therapy.
Pharmacotherapy with somatostatin analogues (OCT), hormone therapy (estrogen analogues), and thalidomide has emerged as a treatment option for refractory AVM-related GI bleeds. Data from hormone therapy trials showed no significant differences compared with placebo in rates of recurrent GI bleeds and transfusion requirements (8). Data from OCT and thalidomide studies showed promising results, but thalidomide use was limited by its high incidence of central nervous system side effects (71%) (23).
OCT, a synthetic analogue of somatostatin, has been studied in AVM-related GI bleeding. It acts by inhibition of angiogenesis (24). Indeed, disappearance or decrease in the size of AVMs has been reported with OCT therapy (25). Some studies demonstrated decreased incidence of recurrent GI bleeding and an increase in hemoglobin levels in a sustained fashion (26–29). However, in most of these studies, only a small number of patients (≤10) were receiving concomitant antithrombotic therapy.
Dosage and formulations
OCT is available in 2 injectable formulations for long-term use (i.e., twice-daily subcutaneous injection and monthly intramuscular injection). Oral formulation of the drug is not yet approved but could be available soon (30). The therapeutic dose ranges from 100 to 500 μg twice daily for the subcutaneous formulation and 10 to 30 mg monthly for the intramuscular formulation. In the present study, the subcutaneous formulation was preferred because of cost. Although the maximum daily dose was 500 μg thrice daily, this dose was associated with higher adverse events in prior studies (31). In light of this, we decided to limit patients to a maximum daily dose of 300 μg twice daily.
Side effects profile
The most frequently reported side effects included pain at the injection site (10% to 20%) and mild to moderate GI disturbances (5% to 15%) such as nausea, flatulence, loose stools, and abdominal cramping that were transient and self-limiting (32–38).
Long-term therapy (>12 months) had been reported to result in hepatobiliary dysfunction, most commonly through asymptomatic gallstone formation (20% to 40%) that usually requires no therapeutic intervention. Etiopathogenesis of these gallstones is unclear but might involve change in bile composition, inhibition of gallbladder emptying, hepatic bile secretion, and sphincter of Oddi motility. Timing of OCT injection in relation to meals might mitigate this risk to a certain extent (39,40). Anecdotal case reports of OCT-induced biliary hepatitis and pancreatitis have also been published (41,42).
Cardiovascular side effects include conduction disturbances (∼2%) ranging from sinus bradycardia to complete heart block, more prominently with intravenous rather than subcutaneous or intramuscular injection (43–45). Intravenous infusions were reported to have systemic and pulmonary vasopressor effects (46). Long-term subcutaneous OCT therapy has been reported to cause injection-site lipoatrophy (47). Other rare laboratory abnormalities (<1%) include reversible thrombocytopenia and hyperkalemia, especially in patients undergoing hemodialysis (48,49).
Most of the safety data associated with long-term use of OCT come from the acromegaly population. Going forward, more prospective studies are needed to pursue exploratory analysis to examine the dose-time-response relationship of its safety and efficacy profiles in the AF population. In addition, with the current choice of monthly OCT injections available, differences in the drug’s safety and efficacy profiles based on route of delivery need to be further investigated in a prospective fashion.
Despite the above-mentioned side effects, OCT is often well tolerated and is considered a valuable therapeutic option in many hypersecretory states (acromegaly, carcinoid syndrome, VIPoma). It is also used as an antidote for sulfonylurea-induced hypoglycemia, especially in patients with heart failure who cannot tolerate intravenous dextrose infusions (50). Recent reports suggested that OCT therapy showed a favorable response in decreasing blood transfusions, number of endoscopic procedures, and readmissions because of GI bleeding in patients with continuous-flow left ventricular assist devices (CF-LVADs) (51,52). GI bleed is the most common cause of readmissions in patients with end-stage heart disease treated with CF-LVADs (53). It is hypothesized that the lack of pulse pressure in CF-LVADs might lead to the development of AVMs, and the exact etiopathogenesis is multifactorial (54).
A recent phase I study by Malhotra et al. (55) evaluating the safety and tolerability of octreotide acetate long-acting release 20 mg depot injection every 4 weeks until week 16 after CF-LVAD placement had 8 patients in the study. None of the patients experienced side effects or safety concerns related to OCT, nor did GI bleeding occur in the study population (55).
In our study, we used OCT, a known therapy for AVM-related GI bleeding, in the novel setting of AF patients with high stroke risk warranting OAC use. By mitigating the risk of rebleeding, OCT enabled the safe continuation of OAC in more than one-half of the patients (52.8%), as evidenced by the steady improvement of their hemoglobin levels while receiving OCT plus OAC therapy. Even in patients with GI bleeding while undergoing therapy, OCT helped enable the continuation of short-term periprocedural OAC for LAAC. Nevertheless, before this therapy is considered as a long-term measure, the benefits of the treatment must be weighed against the risk of adverse effects such as hepatobiliary dysfunction associated with long-term therapy.
Our study presents the obvious limitations of an observational study with median follow-up of only 8 months (range 6 to 13 months). All patients received OCT therapy, and there was no control arm. It is hard to justify continued OAC without an intervention in such a high-risk group of patients with GI bleeding while receiving anticoagulation therapy. We were therefore unable to find a control group, because most patients with recurrent GI bleeding were not receiving OAC. Effects of long-term OCT therapy in this population are unknown. Further studies with larger sample sizes and appropriate comparator groups, despite the stated challenges with identifying the latter, are needed to confirm our findings. Despite these limitations, our study points toward a potential novel therapy for patients with AF and a history of GI bleeding who require continued OAC.
Subcutaneous OCT therapy is a potential therapeutic option in patients with AVM-related GI bleeding who require OAC therapy for stroke prevention. Treatment with OCT offers a safer way to reinitiate OAC by mitigating the risk of recurrent GI bleeds in the majority of patients. OCT therapy could serve as a bridge to performing LAAC procedures and enable continuation of OAC after cardioversion and antiarrhythmic drug therapy.
COMPETENCY IN PATIENT CARE: Pharmacotherapy with somatostatin analogues such as octreotide is an attractive option in people with lower GI bleeds related to vascular malformations (AVMs) and an obligate need for OAC use. This enables successful reinitiation of OAC with a decreased risk of recurrent GI bleeds and provides a window to explore options such as rhythm control strategy and LAAC in people with atrial fibrillation.
TRANSLATIONAL OUTLOOK: Large-scale, multicenter prospective studies and controlled trials comparing GI bleed event rates, dose-response effect, and other clinical parameters that influence the risk of recurrent GI bleeding are needed to validate the above-mentioned conclusions.
Dr. Di Biase is a consultant to Stereotaxis, Biosense Webster, and St. Jude Medical; and has received speaker honoraria/travel reimbursement from Biotronik, Medtronic, Boston Scientific, Janssen, Pfizer, and Epi EP. Dr. Swarup has served as a consultant to Abbot Vascular and Biosense Webster; and is on the speakers bureaus for St. Jude Medical, Boston Scientific, Janssen, and Pfizer. Dr. Gopinathannair has served as a consultant to St. Jude Medical and Boston Scientific; has served on the speakers bureau for the American Heart Association, Pfizer, Bristol-Myers Squibb, Zoll Medical, and AltaThera Pharmaceuticals; and has served on an advisory board for HealthTrust PG. Dr. Natale is a consultant for Stereotaxis, Biosense Webster, and St. Jude Medical; and has received speaker honoraria/travel reimbursements from Biotronik, Medtronic, Boston Scientific, Janssen, Pfizer, and Epi EP. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
All authors attest they are in compliance with human studies committees and animal welfare regulations of the authors’ institutions and Food and Drug Administration guidelines, including patient consent where appropriate. For more information, visit the JACC: Clinical Electrophysiology author instructions page.
- Abbreviations and Acronyms
- atrial fibrillation
- arteriovenous malformation
- continuous-flow left ventricular assist device
- left atrial appendage closure
- oral anticoagulation
- Received November 2, 2016.
- Revision received April 5, 2017.
- Accepted April 11, 2017.
- Chugh S.S.,
- Havmoeller R.,
- Narayanan K.,
- et al.
- Chang H.-Y.,
- Zhou M.,
- Tang W.,
- Alexander G.C.,
- Singh S.
- Kovacs R.J.,
- Flaker G.C.,
- Saxonhouse S.J.,
- et al.
- Danesi R.,
- Del Tacca M.
- von Essen R.,
- Ostermaier R.,
- Grube E.,
- et al.
- Cozzi R.,
- Attanasio R.
- Welin S.V.,
- Janson E.T.,
- Sundin A.,
- et al.
- Catnach S.M.,
- Anderson J.V.,
- Fairclough P.D.,
- et al.
- Sakamoto H.
- Glatstein M.,
- Scolnik D.,
- Bentur Y.
- Dang G.,
- Grayburn R.,
- Lamb G.,
- Umpierrez De Reguero A.,
- Gaglianello N.
- Malhotra R.,
- Shah K.B.,
- Chawla R.,
- et al.