Diastolic but Not Systolic Heart Failure Is Associated with Multiple Abnormalities on Platelet Aggregation Testing

Introduction: The effect(s) of co-morbid medical conditions on platelet function is poorly understood. In this retrospective EMR-based study, we analyzed the influence of various diseases on in vitro measures of platelet function - platelet function analyzer-100 (PFA-100) closure times, platelet aggregation (using light transmission aggregometry (LTA)), platelet dense granule release (using lumi-aggregometry), and platelet flow-cytometry for surface glycoproteins. We also examined their influence on VWF testing.

Methods: Four hundred ninety seven patients who had platelet aggregation testing performed using LTA between August 2008 and August 2013 were included in our study. Co-morbidities at the time of testing were recorded. Propensity score matching for each individual disease was used to adjust for relevant covariates. We used a 1:1 nearest neighbor match without replacement, with caliper width set to 0.2 times the standard deviation of the logit of the propensity score. Following matching, Fisher's exact test or Chi square test was used as appropriate to assess the association between categorical variables, while the Mann-Whitney test was used to test the association between categorical and continuous measures. Pearson co-efficient was used to assess the correlation between continuous variables. P < 0.05 was considered significant.

Results: 1) Congestive heart failure (n = 44) was associated with impaired platelet aggregation in the presence of arachidonic acid (p = 0.001) and collagen (p = 0.009), as well as impaired dense granule release in the presence of collagen (p = 0.002) and epinephrine (0.012). It was also associated with abnormal aggregation (p = 0.024) and release (p = 0.028) in the presence of ≥ 2 agonists in the respective panels. Diastolic heart failure (n = 25) was found to be associated with impaired aggregation in the presence of ADP (p = 0.007), collagen (p = 0.001), or arachidonic acid (p = 0.007), and to ≥ 2 agonists in the aggregation panel (p = 0.008). Systolic heart failure (n = 26) was not associated with abnormalities in aggregation or release. 2) Severe aortic stenosis (n = 17) was associated with prolonged collagen/ADP (p = 0.003) and collagen/epinephrine (p <0.001) closure times with PFA-100, but not with any abnormalities in the platelet aggregation/release panels. Severe aortic stenosis was associated with a decreased ristocetin cofactor/VWF antigen ratio (0.66±0.17 vs. 0.90±0.37; p = 0.030), but not with any other abnormalities in VWF testing. 3) Diabetes mellitus (n = 65) was associated with impaired platelet aggregation in the presence of collagen (p = 0.034) and impaired platelet release in the presence of epinephrine (p = 0.027). However, glycated hemoglobin level (HbA1C) was not found to correlate with impairments in either aggregation or release in the presence of any agonist. Hypothyroidism (n = 71) or vitamin D deficiency (n = 39) were not found to be associated with abnormalities in any of the platelet function assays. Finally, biochemical parameters reflecting hepatic or renal function did not correlate with any abnormalities in platelet function assays. However, the total number of co-morbidities in any patient correlated with the number of abnormalities in the platelet aggregation as well as release panels.

Conclusion: Diastolic heart failure was associated with impaired platelet aggregation in the presence of multiple agonists. Though the mechanism remains unclear, we postulate that this could be related to shear stress to which the platelets are subjected in the non-compliant ventricles. Severe aortic stenosis was associated with prolonged collagen/ADP as well as collagen/epinephrine PFA-100 closure times and with lower ristocetin co-factor/VW antigen ratio suggesting functional impairment of VWF. Though diabetes mellitus was associated with impaired platelet aggregation in the presence of collagen and impaired dense granule release in the presence of epinephrine, no correlation was found between these abnormalities and HbA1C levels, making the significance of the association unclear.