Platelet function analyzer (PFA)‐100® closure time in the evaluation of platelet disorders and platelet function – a rebuttal

Abstract

The Platelet function analyzer (PFA)-100® (Dade-Behring, Marburg, Germany) device has now been used for several years in hospitals to access bleeding tendency of patients. Recently, the Platelet Physiology Subcommittee of the SSC, International Society on Thrombosis and Haemostasis (ISTH), published the first official recommendations [1] on the PFA-100® system. The article comprehensively reviews the published data about the application of the PFA-100® for the evaluation of several congenital and acquired platelet function disorders. The authors conclude: (1) ‘although the PFA-100® CT is abnormal in some forms of platelet disorders, the test does not have sufficient sensitivity or specificity to be used as a screening tool for platelet disorders’. However, we are concerned that the Subcommittee that published this report might have been misguided in their conclusion because its members did not realize that this test had not been developed to detect platelet disorders. This may be due to the fact that several seminal papers that reflect the development of this test over the past 25 years have not been adequately appraised in this review. According to Popper [2] the Main Conclusion 1 of the ISTH publication is not a scientific statement, because it can never be contradicted. But we could transform it into the following scientific statement (2): ‘The PFA-100® test has only a sensitivity of 25 % for platelet disorders. This is not sufficient for using the test as a screening tool for platelet disorders’. Statement 2 can easily be falsified using material in the ISTH publication (Table 1) [1]. If we assume a patient collective containing patients with ‘Glanzmann thrombasthenia, Bernard Soulier syndrome, Platelet-type von Willebrand disease, Grey platelet syndrome’, then according to Table 1 the sensitivity of the PFA-100® increases to 100 % in contrast to another collective with patients with ‘Platelet procoagulant defect’. Here the sensitivity decreases to 0%. For that reason Statement 2 must be wrong. In addition, Popper would certainly agree that Table 1 [1] is not a scientific statement, because it contains nine ambiguous elements such as: ‘PFA-100® CT can be pathological or normal in Wiskott Aldrich syndrome’. To make this table scientifically valid, percentages of sensitivity and prevalence would have to be inserted for every platelet disorder to support the heavy weight conclusion of the ISTH. The question arises: ‘What would have been a fair judgement of the ISTH?’ Obviously the most important thing to do first is to look up the original papers of those scientists who developed the method, who are surprisingly not cited in the ISTH paper. Usually ISTH members recognize the seminal contribution of a former ISTH president. The principle of the PFA-100® system was developed in the laboratory of G. V. R. Born at King's College London around 1980. Using the fire fly technique, it was discovered that, besides collagen, adenosine 5′-diphosphate (ADP) originating from injured vessel wall cells is a possible candidate for initiating primary hemostasis [3]. To test this hypothesis, the complex in vivo situation was reduced to an ex vivo set-up by perfusing an artificial capillary (200 μm) and an aperture (150 μm) with anticoagulated blood. The porous aperture was covered with type I collagen and soaked with ADP, in order to mimic the injured part of a vessel [4]. This reduction was at that time quite radical, because everybody thought that primary hemostasis could only take place in vivo. However, investigations in J. Caen's laboratory in Paris showed that the ex vivo model correlated surprisingly well with deficiencies of primary hemostasis [5]. The laboratory system was refined to yield the first commercial instrument ‘Thrombostat 4000®’ (V. von der Goltz, Seeon, Germany), which was used predominantly in hospitals. In particular, in blood donation facilities the need emerged for a technique to detect donors who had ingested aspirin. To address this question, apertures covered only with collagen were applied. This set-up is highly aspirin-sensitive and yields results similar to the collagen-induced platelet aggregation [6]. To improve reproducibility, but maintain the aspirin sensitivity, epinephrine was added to the collagen aperture [7]. The Thrombostat instrument was replaced by the PFA-100® [8]. However, it is important to note that the core (or measurement set-up) of the PFA-100® is very similar to that of the Thrombostat 4000® system. Only the capillary has a greater radius, which causes an increased initial flow. In studies comparing both devices very similar results were obtained [9, 10]. Thus, the literature on the Thrombostat technique [11] could have been employed for the evaluation of the PFA-100®. It has to be stressed that the PFA-100® principle was introduced 20 years ago [4] as a method for simulating primary hemostasis in vitro. Similar to the in vivo bleeding time technique, it is dependent on the typical platelet functions of (1) platelet adhesion (dependent on von Willebrand factor [VWF] concentration and quality [5, 12]), and (2) platelet aggregation (dependent on GP IIb/IIIa receptor). The platelet release reaction (function 3) becomes important especially in the EPI-disposable. In addition, the global PFA-100® test is dependent on blood hematocrit, platelet count and blood viscosity, similar to its in vivo counterpart. The blood flow causes a relatively high shear rate between 20 000 and 50 000 s−1 at the aperture [13] (not 5000 as stated by the ISTH). The approach of the inventors of the method was to provide a fast global technique, which correlates well to defective primary hemostasis, helping the doctor in a typical clinical situation to quickly decide whether a bleeding risk perisurgically exists that is due to disturbances of the primary hemostasis. The instrument has been called ‘Platelet Function Analyser’ by Dade-Behring mainly for Food and Drug Administration reasons: it was thereby possible to present correlations of the PFA-100® to the ‘Born Platelet Aggregometer’ [14], the gold standard of platelet function techniques. However, according to its principle PFA-100® is only able to detect platelet functions 1–3. It will never detect deficiencies of secondary hemostasis such as ‘Platelet procoagulant defect’, as is expected by the ISTH. Therefore, if the PFA-100® system is judged according to its claims it must be judged for detecting platelet functions 1–3. Not all platelet disorders display defects in functions 1–3. The new instrument has not been named ‘Platelet Disorder Detector’, an ability focused on by the Subcommittee. In accordance to Table 1 of the ISTH report [1], the sensitivity for abnormalities of platelet adhesion (including von Willebrand disease) and aggregation is excellent, more than 90%. Clinical experiences support the view that the PFA-100® is even too sensitive, as it detects a single dose of 100 mg aspirin in about 50% of controls. This might be due to the relatively high shear rates. This characteristic usually leads to overtreatment under practical circumstances. The sensitivity of the EPI-disposable to detect abnormalities of the platelet release reaction seems to be somewhat lower. It has to be stressed that the global system will measure the product of functions: platelet adhesion and aggregation. Increased VWF concentration can thereby compensate for deficiencies in platelet aggregation; certainly a helpful property for predicting bleeding. In a typical clinical setting of a big hospital with patients with positive bleeding history, the sensitivity of the PFA-100® reaches more than 90% in detecting patients with disturbances of the global platelet function [15]. At the university hospital of Marburg, the PFA-100®, and its predecessor, the Thrombostat 4000®, have been used in the hemostasis laboratory for more than 15 years now (> 100 000 determinations). The test is routinely applied in patients with a history of bleeding symptoms or complications, or when recent aspirin ingestion prior to invasive procedures is reported. In addition, PFA-100® is performed in patients with possible disturbed platelet function as liver or renal failure, and before complicated or major surgery. Only in case of a clearly suspicious history, negative PFA tests are followed by a platelet aggregation test. Abnormal PFA results lead to further testing and to the recommendation of hemostatic treatment. This includes desmopressin, antifibrinolytics and/or platelet concentrates, dependent on the laboratory results and the clinical situation. We are not aware of any life-threatening bleeding complications in patients having been managed this way. We admit, though, that this approach may have led to overtreatment in several cases. All considerations strongly support the view that the first official ISTH statement on the PFA-100 technology is very much pointing in the wrong direction. The article has caused much confusion to doctors, because they interpret the global statement of the ISTH in a way that gives the PFA-100® test no value in screening procedures at the hospital. On the other hand, we also support the conclusion of the SSC report that the PFA-100® should be further evaluated in prospective clinical trials in order to substantiate the current knowledge and experiences particularly with respect to monitoring of therapeutic measures. Professor M. Kratzer invented the PFA-100® principle. He sold all rights to the technique to Baxter Diagnostic Inc. in 1992. Professor V. Kretschmer has performed several evaluation studies for Dade-Behring, but has never been a consultant of this company.