Comparison of Multiple Electrode Impedance-Based Platelet Aggregometry to Light Transmission Aggregometry for Diagnosis of Qualitative Platelet Defects

Background: Platelet function testing remains a challenge in evaluation of bleeding disorders. The literature suggests a highly variable rate of qualitative platelet defects (QPDs) among individuals with bleeding symptoms. However, diagnosis of platelet disorders is complicated by the need for testing fresh platelets and the requirement of specialized expertise and equipment for performing the assays. Thus, typically only tertiary care centers have the ability to perform comprehensive platelet testing. Platelet testing may be especially difficult in pediatrics, when required blood volume and challenging phlebotomy may limit the age at which patients can be tested. Thus, a test that would allow effective screening of patients without the need for specialized expertise would be of benefit, especially if testing could be accomplished on a smaller quantity of blood. Multiple electrode aggregometry (MEA) is a whole blood, impedance based platelet function assay using standardized reagents and automated pipetting. Indeed, MEA is currently utilized as point-of-care testing for both ASA and P2Y12 inhibitors. MEA has the further advantage of requiring significantly less blood volume for aggregation testing.

Objective: We sought to determine the sensitivity and specificity of the MEA for the diagnosis of QPDs.

Methods: To determine if MEA is an adequate screen for QPDs, we analyzed patients with abnormal bleeding symptoms that underwent comprehensive platelet function testing (CPFT) at our institution from 2011-2013. During this time frame, MEA was performed as part of the CPFT. CPFT consisted of LTA, ATP secretion, glycoprotein expression, total ATP/ADP content, and quinacrine uptake and release. We compared the direct relationship of the MEA area under the curve to the maximal aggregation on LTA. Additionally, the proportion of agreement in abnormalities between MEA and LTA was assessed. This study was approved by the Cincinnati Children's Hospital Medical Center IRB.

Results: 120 patients met inclusion criteria and had a total of 128 evaluable platelet function assays during this time frame. Of these assays, 30 demonstrated abnormalities that were of likely clinical significance. MEA area under the curve (AUC) generally demonstrated poor linear correlation to maximal aggregation (MA) on LTA for each agonist. Linear correlation of MEA AUC with LTA MA was the highest with collagen, and lowest with arachidonic acid. We next assessed the finding of an abnormality by LTA versus MEA. MEA sensitivity for abnormalities on LTA was highest for ADP, at 73.7%. Sensitivity did not exceed 60% for any other agonist. Specificity for MEA was highest for detecting abnormalities with collagen, ristocetin, and arachidonic acid (97.9%, 100%, and 76% respectively). We next examined the ability of MEA to detect abnormalities of likely clinical significance. Abnormalities of clinical significance were determined by review of CPFT results by two hematologists. Sensitivity and specificity of MEA for likely clinically relevant platelet abnormalities was 80% and 54.0%, respectively. Negative predictive value was 89.8% and positive predictive value was 26.7%. MEA performed best in identifying patients on anti-platelet agents.

Conclusions: Based on our data, MEA did not appear to have sufficient sensitivity or specificity to utilize as a primary screening for QPDs.