Wysokinska, Ewa

Abstract Background Acquired Factor VIII inhibitor or autoimmune hemophilia A (AHA), has an estimated incidence of up to 1.5 cases per million/year and may result in severe hemorrhagic complications and death. Approximately 50% of cases have an underlying condition such as malignancies, autoimmune disorders and post-partum state. AHA should be suspected in any patient presenting with unexplained bleeding and an inhibited aPTT. Management consists of maintaining hemostasis and elimination of the inhibitor, however management is not standardized. We present 62 cases of AHA managed at Mayo Clinic Rochester, over the course of 36 years. We also analyzed whether aPTT at presentation correlated with the strength of inhibition measured by Bethesda Titer. Methods After IRB approval, medical records of patients with AHA were reviewed and all clinical data collected. Cumulative incidence of death was estimated by Kaplan-Meier analysis. Spearman correlation was used to calculate relation of APTT to Bethesda titer. Results Between 1976 and 2012, we identified 62 patients (male: 35), with a median age at diagnosis of AHA 69 years (mean 64, range 20-86). Clinical presentation consisted of extensive ecchymoses (n=40, 64%) in majority of cases. 29/62 (47%) patients had at least 1 identifiable predisposing condition with 12/62 (19%) patients with an underlying malignancy and 16/62 (26%) with underlying autoimmune conditions. Median Bethesda titer was 29 (range 1 to 1178). Bethesda titer was not related to the number or duration of hospitalizations. Most (69%) patients had at least one hospitalization and 12 (19%) had more than one hospitalization for bleeding complication. Inpatient therapy for bleeding consisted most commonly of FEIBA in 21 pts (34%) and rFVIIa in 6 pts (10%). Prednisone was the most common immunosuppressant used in 54 (87%) patients while Rituximab was used in 11 (18%) patients. Of 32 patients with available follow up labs most (69%) achieved remission. There was no difference in remission rates between patients treated or not treated with Rituximab (p=0.1735). Conclusion Acquired Hemophilia A is a rare condition with very heterogenous presentation. It affects mostly older male patients who present with ecchymoses and elevated APTT. The degree of APTT prolongation at the time of diagnosis does not correlate with the strength of the Bethesda titer and should not guide choice of therapy in a patient presenting with an acute bleed. Rituximab use in the 11 patients treated at Mayo did not seem to influence remission rates or survival. Disclosures: No relevant conflicts of interest to declare.

Abstract Abstract 1128 Background Dysfunctional or mismatched prosthetic heart valves can results in high shear stress related von Willebrand factor (VWF) impairment, and clinical bleeding. This mechanism is similar to that seen in aortic stenosis patients, where gastrointestinal bleeding related to VWF multimer loss is well described (Heyde syndrome). In this study, we compared the results of VWF testing of patients with normally functioning cardiac valve prostheses, dysfunctional prosthetic valves, severe native aortic stenosis, and normal controls. Methods A total of 75 patients, 31 female and 44 male with a median age of 79.5 years were recruited to this study. Thirty patients had severe aortic stenosis (mean gradient > 40 mm Hg); 39 patients had normally functioning prosthetic valves, 21 aortic (AVR) and 18 mitral valves (MVR); and 6 patients had with dysfunctional valves (4 aortic and 2 mitral). Whole blood and plasma samples were drawn from patients and tested for platelet function analyzer-100 collagen ADP closure time (PFA-CADP), VWF antigen (VWF:Ag), VWF activity by latex immunoassay (VWF:Ltx), VWF multimer analysis and VWF multimer densitometry ratio of large vs medium to low molecular weight multimers (>15 mers/2–15 mers). Results The 6 dysfunctional prosthetic valve patients included patient-prosthesis mismatch (2) and (one each), thrombosed mechanical aortic prosthesis, torn biological aortic prosthesis, severe tissue mitral prosthetic regurgitation, and post-mitral valve repair severe mitral regurgitation. Four of six patients had clinically significant bleeding. Most aortic stenosis and all 6 prosthetic dysfunction patients had abnormal VWF multimers which is rare in normally functioning prostheses and completely absent in normal donors. VWF:Ltx/Ag was lower in AS and MVR than normal. VWF:Ltx/Ag ratio, VWF:multimer ratio and PFA-CADP were all significantly abnormal in patients with dysfunctional prosthetic valve. VWF:multimer ratios in normally functioning prostheses tended to be intermediate between normals and than those of patients with severe AS or prosthetic valve dysfunction (Table). Conclusions VWF testing including quantitative von Willebrand multimer analysis and PFA-CADP reliably separate dysfunctional from normally functioning cardiac prostheses, and should be considered in the evaluation of patients with cardiac prostheses, especially those who have clinically significant bleeding. Disclosures: No relevant conflicts of interest to declare.

Abstract Background: Aortic stenosis-associated acquired von Willebrand syndrome (AS-AVWS) is caused by an accelerated degradation of the highest molecular weight von Willebrand factor (VWF) multimers (HMWM). We recently reported that the severity of the HMWM loss correlates with the severity of AS and patients’ bleeding tendency (Am J Cardiol, 2013;111:374-81). As for the missing HMWM, it is still uncertain if they are cleared from circulation or redistributed with medium-low MWMs, which can be assessed by VWF propeptide (Pro)/antigen (Ag) ratio. Besides the tedious VWF multimer analysis, it is uncertain if VWF activity by immunoturbidic method (Lx)/ Ag ratio or VWF collagen binding activity (CBA)/Ag ratio can be used as a screening test to predict AS-AVWS. It has been speculated in the literature that VWF:CBA is more sensitive than VWF:Lx for detecting a HMWM loss. Our goal was to exam the laboratory characteristics of various VWF tests in evaluating AS-AVWS and the potential mechanism of HMWM loss. Method: Sixty-six patients (between years 2010-2012, 43% male) with varying degrees of AS (16 mild, 20 moderate, and 30 severe) and separate 21 patients who had aortic valve replacement were assessed with VWF:Ag, VWF:Lx, VWF:CBA, platelet function analyzer collagen plus adenosine diphosphate (PFA-CADP), VWF multimer analysis and multimer ratio by densitometry (Am J Cardiol, 2013;111:374-81) after their echocardiography. The clinical endpoints of this study are AS severity measured by mean gradient (MG) by echocardiography and clinically significant bleeding. Statistical analyses including Spearman rank correlation test, estimation of the area under the receiver operating characteristics (ROC) curve, and Wilcoxon rank sum tests were performed. Results: VWF:Lx and VWF:CBA had strong correlation (Spearman r=0.94, P<0.001), and the correlation of VWF:CBA/Ag and VWF:Ltx/Ag was 0.49, p<0.001). As previously published, in patients with AS, MG correlated with VWF:Lx/Ag (Spearman r = -0.41, p <0.001), PFA-CADP (r = 0.49, p <0.001), and VWF multimer ratio (r = -0.76, p <0.001). MG also correlated with VWF:CBA/Ag (r= -0.38, p=0.002), but not with VWF: Pro/Ag (r=-0.20, p=0.12). Among AS and AVR patients, the area under the ROC curve for detection of the loss of HMWM was 0.88 (95% CI: 0.80-0.95) by VWF multimer ratio, 0.76 (95% CI: 0.65-0.86) by PFA-CADP, 0.70 (95% CI: 0.58-0.82) by VWF:CBA/Ag ratio, 0.67 (95% CI: 0.56-0.70) by VWF:Lx/Ag ratio, and 0.53 (95% CI: 0.41-0.65) by VWF:Pro/Ag . In addition to previously published results, clinically significant bleeding, in 14% of AS patients, was not associated with VWF:CBA/Ag (p=0.25) or VWF:Pro/Ag ratio (p=0.83). VWF: CBA/Ag ratio was significantly lower in patients with severe AS (MG > 40 mmHg) compared to AVR patients (median: 0.84 vs. 1.06, p<0.001), while there was no evidence of a difference in VWF:Pro/Ag (p=0.45). Conclusion: HMWM losses are highly prevalent in patients with moderate to severe AS. Of the VWF tests, PFA-CADP and VWF multimer analysis and ratio had the highest predictive ability for AS-AVWS followed by VWF:CBA/Ag and VWF:Lx/Ag. The VWF:CBA is not overly superior than VWF:Lx for predicting a HMWM loss. The normal VWF:Pro/Ag suggests that AS-AVWS is different from other types of AVWS due to accelerated clearance of HMWM. Disclosures No relevant conflicts of interest to declare.

Abstract Abstract 2230 Background: Heyde syndrome describes patients with severe aortic stenosis (AS) who have bleeding diathesis, especially gastrointestinal bleeding. Although acquired won Willebrand syndrome (AVWS) due to loss of the highest molecular weight multimers (HMWM) presents in most patients with severe AS, not all such patients develop Heyde syndrome. The goal of this study was to explore whether the severity of HMWM loss was significantly associated with Heyde syndrome. Methods: We first performed chart reviews on patients referred for double balloon enteroscopy (DBE) for GI bleeding or iron-deficiency anemia (GIB/FeDefic) in two time periods, from 2005–2008, and from 2009–2011. Next, patients who could have AVWS due to AS or other cardiovascular or hematologic disorders were identified. Laboratory tests including VWF antigen (VWF:Ag), VWF ristocetin cofactor activity or VWF latex immunoturgidic activity (VWF:Ltx) and/or VWF multimer analysis were performed as requested by physicians. From this population, patients with Heyde syndrome were identified. Their clinical and VWF laboratory testing results were compared to a separated group of patients with severe AS patients who had no clinically significant bleeding, but had matched gender and hemodynamic severity by echocardiogram (Matched AS Controls). These patients all had VWF:Ag, VWF:Ltx, and VWF multimers performed. In both Heyde syndrome patients and Matched AS controls, VWF multimer densitometry ratio of the HMWM (>15 mers) to low to intermediate multimers (2–15 mers) was assessed (VWF:multi ratio). Results: Overall, 437 patients underwent DBE for GIB/FeDef. Between 2005 and 2008, 1% of 318 patients had VWF testing. From 2009 to 2011, 23% of 119 patients were tested. Heyde syndrome was noted in 15 patients. Of the 437 patients, 134 patients (31%) likely had AVWS based on diagnoses of moderate-severe AS (30%), moderate to severe mitral regurgitation (25%), moderate to severe pulmonary hypertension (16%), prosthetic valve regurgitation or stenosis (13%), hypertrophic cardiomyopathy (10%), and hematologic disorders (4%). Among these 134 patients, 29 patients had VWF tests performed, and 22 had VWF multimer analysis. VWF antigen levels were similar in Heyde syndrome and Matched AS Controls (176±89% versus 155±77% respectively). Despite similar hemodynamic severity, more severe HMWM loss as indicated by lower VWF:Ltx/Ag and VWF:multi ratios was present in Heyde syndrome patients than matched AS controls (table, P<0.0007). Conclusion: If AVWS is suspected in patients with GIB/FeDef, laboratory testing for VWF indices is important to confirm the diagnosis of AVWS. Despite similar hemodynamic severity, Heyde syndrome patients have more severe VWF HMWM loss indicated by significantly decreased VWF:Ltx/VWF:Ag and VWF:multi ratios. Conversely, though highly sensitive for AVWS, PFA-CADP does not predict bleeding tendency in patients with severe AS. Disclosures: No relevant conflicts of interest to declare.