Centrifugation is a crucial step impacting microparticle measurement

To the Editor

In the original article entitled “Flow cytometric measurement of microparticles: Pitfalls and protocol modification” Shah et al. [1] address an interesting and crucial issue on how variations in sample preparation affect microparticle measurements. Indeed, several pre-analytical variables such as blood collection, sample processing, transportation and centrifugation conditions may have a major impact on microparticle measurement, and have not been adequately addressed in the literature [2]. This specific question is of high importance for proper microparticle enumeration in clinical studies since the current lack of standardization hinders data interpretation [3].

While we agree that standardization of microparticle detection and quantitation is an important priority, in the Shah et al. article, the authors have employed a method for microparticle isolation that does not appear to conform with current published techniques. Specifically, the ‘Methods’ section describes the isolation of plasma microparticles using three successive centrifugation steps (160 g × 20 min, 1500 g × 20 min and 1500 g × 20 min) in order to obtain cell free plasma, and a final centrifugation step to spin down microparticles (13,000 g × 2 min). The pellet is then re-suspended and analysed by flow cytometry. Consequently, it may be possible that the majority of microparticles remain in the final discarded supernatant. Methods to pellet plasmatic microparticles before analysis, while still controversial, have been applied by several teams [4]. However, we are not aware that any centrifuge for only 2 minutes to pellet microparticles; more typically, centrifugation conditions have ranged from about 18,000 g × 30 min [5] to 100,000 g × 60 min [6]. We disagree with the authors when they claim that “a high-speed centrifugation of 13,000 g for 2 minutes is widely reported as an appropriate means of isolating microparticles”. In order to support this statement, Shah et al. improperly quote the Forum article published in Journal of Thrombosis and Haemostasis in 2004 [4], in which the 13,000 g × 2 min centrifugation step used by several groups was intended to fully deplete plasma of platelets, not to isolate microparticles. In this case, the supernatant is directly used to analyse microparticles by flow cytometry or by solid phase capture, whereas the pellet (composed of remaining platelets and centrifugation-induced platelet microparticles) is discarded. Moreover, Shah et al. state that their protocol was compared to one “adapted from B. Hugel et al. protocol” (protocol P2). However, Shah et al. measured microparticles from the second centrifugation (13,000 g × 2 min) pellet which is at odds with the paper by Hugel et al. [7], where microparticles were measured in the supernatant fraction.

In summary, in the absence of a systematic study to define the optimal centrifugation conditions, we believe that the pre-analytical steps used by Shah at al. may have resulted in the loss of most of their microparticles possibly accounting for the very low microparticle counts presented.