035 UPPER GASTROINTESTINAL BLEEDING IN THE POST PROCEDURAL PERIOD IN PATIENTS UNDERGOING TRANSCATHETER AORTIC VALVE IMPLANTATION: A RETROSPECTIVE ANALYSIS D Stanger, F Wong, S Alipour, B Bressler, D Wood, O Takach, J Webb Vancouver, British Columbia BACKGROUND: Symptomatic aortic stenosis (AS) is an indi- cation for aortic valve replacement. Transcatheter aortic valve implantation (TAVI) is an increasingly common technique used to replace the valve for patients deemed high risk for conventional open sternotomy approach. Given the comorbidities and demographics of the patient population undergoing TAVI, they are at significant risk of upper gastrointestinal bleeding (UGIB). Furthermore, con- current use of antithrombotic medication may further the risk of GI bleeding. In addition, TAVI typically utilizes intraoperative trans- esophageal echocardiography (TEE) to obtain visualization during valve placement which may add to further risks of UGIB. As TAVI is a relatively new technique, its extra-cardiac complications are being elucidated. The objective of the current study was to determine the incidence of significant UGIB requiring endoscopic evaluation in patients undergoing TAVI. METHODS: A retrospective analysis of UGIB incidence in admitted patients to St. Paul’s Hospital (SPH) from the specified timeline (Jan 2005-Aug 2014) was conducted. Subjects were patients who were referred to the Gastroenter- ology service at SPH with clinical evidence of UGIB and/or required an esophagogastroduodenoscopy (EGD) during their admission for TAVI. Patients who underwent a TAVI procedure were identified from a database maintained by the Division of Cardiology at SPH. RESULTS: A total of 845 TAVI procedures were included in the analysis. Of these, 703 procedures were via transfemoral (TF-TAVI) approach and 142 were via transapical (TA- TAVI) approach. 2.01% (n¼17) of all TAVI procedures had UGIB. For TF- TAVI and TA-TAVI, 2.13% (15/703) and 1.41% (2/142) had UGIB, respectively. The average hemoglobin drop in those with UGIB was 27.7g/L. Of the patients who had significant UGIB, 58.8% (10/17) were on triple antith- rombotic therapy. 12 EGDs were performed for further evaluation of UGIB. Of the EGDs performed, 2 were normal, 2 had clean-based duodenal ulcers, 3 had clean-based gastric ulcers, and 5 had high-risk lesions in the esophagus and all of these required endoscopic therapy. CONCLUSION: TAVI is a new minimally invasive for aortic valve replacement and carries a small risk of upper gastroin- testinal bleed. When UGIB does occur in the setting of TAVI, it is usually significant and requires pRBC transfusion. The most common lesion found is a distal esophageal or gastro- esophageal junction ulceration with active bleeding. Postu- lated reasons for bleeding may stem from the use of TEE intraoperatively in which there may be local trauma. Furthermore, bleeding may be more severe in some patients due to antiplatelet and/or anticoagulation use perioperatively. 036 EXAMINING THE EFFECT OF PATIENT-CENTERED CARE ON PATIENT OUTCOMES: A SYSTEMATIC REVIEW SM Fredericks, J Lapum, G Hui Toronto, Ontario BACKGROUND: Patient centered care (PCC) is a dynamic and individualized approach to care delivery. It encompasses the integration of an individual’s values, beliefs, and preferences regarding their overall health and well-being in the design and implementation of health care related activities. While there has been recent interest in conducting systematic reviews to examine the effectiveness of PCC interventions, various studies fall short in explaining the type of intervention that is most effective in producing significant changes to desired outcomes. Specifically, the characteristics of these in- terventions that include the mode and dose is rarely presented. This information is vital to facilitate the design of future PCC interventions that will result in increased quality of care during the hospitalization period, as well as increased perfor- mance of self-care behaviours post-hospital discharge. METHODS: The purpose of this systematic review was to determine the characteristics of PCC interventions that have demonstrated effectiveness in enhancing the quality of care during in-patient hospitalization, and the performance of self- care behaviors during the post-hospital discharge period. The PRISMA framework was used to ensure the accurate and complete conduct and reporting throughout this review. A total of 40 studies met the selection criteria of involving samples over the age of 18 years of age, and being published between 1995 and 2014 and were included in this systematic review. Descriptive statistics were used to delineate study, participant, and intervention characteristics. RESULTS: The effectiveness of PCC interventions were noted in a quarter of these trials (approximately 5 studies), even though these studies were designed to reflect a rigorous pro- cess, (i.e. use of randomized controlled trial design) and had a low attrition rate. Of particular interest, the intervention studies that reported non-statistically significant findings in relation to the effectiveness of PCC interventions, contained populations that were diagnosed and living with chronic conditions. These studies reported small effect sizes suggesting PCC interventions may not be worthwhile to implement to individuals diagnosed and living with chronic illnesses, as these individuals, over time, may have developed their own self-care behaviours and may have formed a unique, long-term relationship with their immediate health care provider. S18 Canadian Journal of Cardiology Volume 31 2015