Platelet-derived microparticles (PDMPs) are released from platelets along with other factors on activation. The PDMPs themselves possess procoagulant activity and are reportedly related closely with many morbid conditions, such as cardiovascular disease, in which platelets play a crucial role,1,2 although the direct toxic effect of PDMPs has not been reported to date.
Membrane plasma separators for plasmapheresis have pores of approximately 0.3 μm,3 which is slightly larger than PDMPs of 0.02–0.1 μm4 diameter. Therefore, we assume that membrane plasma separators can pass PDMPs during plasma filtration. However, the possibility of whether PDMPs can be filtrated using such a filter has not been investigated.
These cases were investigated to elucidate the PDMP dynamics during plasmapheresis.
Patients for whom we performed plasmapheresis between August 2008 and July 2009 in our department were examined. All had submitted statements of written informed consent to participate in this study. This study was approved by the Institutional Review Board of the University of Tokyo, School of Medicine (no. 2117). Table 1 presents patient characteristics. All were treated using the same membrane plasma separator (OP-08W; Asahi Kasei Kuraray Medical Co., Ltd., Tokyo, Japan). Most patients were treated for autoimmune disease (cases 1 and 3–6) to remove autoreacting antibodies. Others were treated for thrombotic microangiopathy (case 2) and for removing low-density lipoprotein (LDL) cholesterol in cases 7 and 8.
Among the eight patients, cases 1–3 underwent plasma exchange with fresh-frozen plasma (FFP) supplemented. Cases 4–6 were treated using double filtration plasmapheresis. The second filters were EC-20W for cases 4 and 5 to remove Immunoglobulin G, whereas EC-40W for case 6 to remove Immunoglobulin M: both filters are products of Asahi Kasei Kuraray Medical Co., Ltd. (Tokyo, Japan). Cases 7 and 8 underwent LDL apheresis using LA-40 (Kaneka Corp., Osaka, Japan). The dextran sulfate-conjugated column captured LDL within the plasma. The processed plasma volumes were 1.0–1.5 plasma volumes. The anticoagulant used for cases 1, 3, and 4 was nafamostat mesilate. Unfractionated heparin was used for cases 2 and 5–8.
Blood was obtained before, halfway through, and at the end of the procedure. The blood was drawn directly from the vascular access before plasmapheresis. At the remaining time points, blood was drawn from the arterial port on the circuit. Separated plasma was also obtained simultaneously when the blood sample was obtained halfway through the procedure.
The blood and plasma samples were collected to tubes containing citrate/ethylenediaminetetraacetic acid for anticoagulant (AS-0205−134; Nipro Corp., Osaka, Japan). Then the specimens were centrifuged at 9,000g for 5 minutes at room temperature. Thereafter, we transferred the supernatant to different tubes carefully to avoid contamination of platelets conveying the same antigens on their surfaces. The levels of PDMPs were measured using a commercially available enzyme-linked immunosorbent assay system (Otsuka Pharmaceutical Co. Ltd., Tokyo, Japan).5
The filtered plasma contained virtually equal levels of PDMPs to that of blood (Table 2). The mean ratio of PDMPs within the filtrated plasma was almost identical to that within blood (0.848 ± 0.141, mean ± SD).
The plasma samples for all cases except for case 1 were also centrifuged at 9,000g for 5 minutes. The levels were compared with those of the original plasma; the ratios were 0.955 ± 0.103 (mean ± SD). The fact confirmed the absence of platelets within the filtrated plasma.
However, the levels of PDMPs at the end of plasmapheresis compared with those before the therapies varied greatly (ratio: 0.290–2.315). The overall average was virtually equal to that before therapy (ratio: 1.212 ± 0.763, mean ± SD).
Results of this study indicate that the sieving coefficient for PDMPs of the filter was sufficiently high, although PDMPs at the beginning and at the end of the plasmapheresis did not differ.
Plasmapheresis has been used to remove LDL cholesterol, which is currently thought to be the largest substance. Nevertheless, PDMPs are much larger substances and have a much more complicated structure than that of LDL cholesterol. The fact that PDMPs pass through a membrane plasma separator might widen the scope of applicability of plasmapheresis.
The nature of PDMPs leads us to deduce that they are almost confined within plasma, which offers effective removal by plasmapheresis. However, the PDMP levels before and after the procedure did not mutually differ overall. We offer two explanations for this result.
We used blood products, such as FFP or albumin solution, which might contain PDMPs. The PDMP values for the healthy control population have been reported as 20.1 ± 2.9 U/ml6 or 17.2 ± 6.2 U/ml.7 Those values were almost equal to those before plasmapheresis within these cases and might also be equal to the concentration within FFP. On the other hand, in pathological conditions, the level reaches 80.4 ± 7.3 U/ml in acute coronary syndrome6 or approximately 50 U/ml in inflammatory bowel diseases.7 Future studies of whether markedly elevated levels of PDMP can be reduced effectively by plasmapheresis are necessary.
Another explanation is that the plasmapheresis itself activates platelets; thereby, the activated platelets produce PDMPs. For that reason, the PDMP concentrations might be elevated because of plasmapheresis itself. Unfortunately, no previous report presents a clinical evaluation of the production rate or half-life of PDMP. Therefore, the above speculation cannot be proved yet. Future studies exploring how rapidly PDMP is producible, especially during plasmapheresis, are also needed.
Plasmapheresis could remove PDMPs with procoagulant propensity. The fact that plasmapheresis could remove PDMPs not only suggests therapeutic approaches against thrombotic or atherosclerotic conditions but also suggests its use for research intervention against elevated PDMP concentration with minimum modulation of other factors.
Although the concentration within blood was unaffected, plasmapheresis can filter PDMPs from blood.
The authors thank Dr. Kanatani at JIMRO and Dr. Hirata at Asahi Kasei Medical for their technical and theoretical assistance.
1. Mallat Z, Benamer H, Hugel B, et al
: Elevated levels of shed membrane microparticles with procoagulant potential in the peripheral circulating blood of patients with acute coronary syndromes. Circulation
101: 841–843, 2000.
2. Nomura S, Imamura A, Okuno M, et al
: Platelet-derived microparticles in patients with arteriosclerosis obliterans: Enhancement of high shear-induced microparticle generation by cytokines. Thromb Res
98: 257–268, 2000.
3. Hirata N, Shizume Y, Shirokaze J, et al
: Plasma separator Plasmaflo OP. Ther Apher Dial
7: 64–68, 2003.
4. Nomura S, Ozaki Y, Ikeda Y: Function and role of microparticles in various clinical settings. Thromb Res
123: 8–23, 2008.
5. Osumi K, Ozeki Y, Saito S, et al
: Development and assessment of enzyme immunoassay for platelet-derived microparticles. Thromb Haemost
85: 326–330, 2001.
6. Nomura S, Uehata S, Saito S, et al
: Enzyme immunoassay detection of platelet-derived microparticles and RANTES in acute coronary syndrome. Thromb Haemost
89: 506–512, 2003.
7. Andoh A, Tsujikawa T, Hata K, et al
: Elevated circulating platelet-derived microparticles in patients with active inflammatory bowel disease. Am J Gastroenterol
100: 2042–2048, 2005.