INTRODUCTION: Therapeutic plasma exchange (TPE) is a widely accepted therapy for TTP. A leading hypothesis for its benefit is that TPE replenishes ADAMTS13 with fresh frozen plasma (FFP) and removes inhibitory antibodies. We have reported that endothelial microparticles (EMP) positive for ULvWF are elevated in acute TTP, decline with repeated TPE, and can inhibit ADAMTS13. Presence of platelet microparticles (PMP) in FFP has been described. This led us to the hypothesis that the benefit of TPE is partly due to its effect on microparticles (MP). We investigated the effect of TPE on MP and ADAMTS13 in TTP.

METHODS: Nine patients were studied longitudinally during an episode of TTP. Blood samples were obtained upon admission and daily pre- and post-TPE. Additionally, the infused and removed plasma were studied. EMP defined by CD31+/CD42b− (EMP31), CD62E+ (EMP62) and CD62E+/vWF+ (EMPVWF); PMP were defined by CD31+/CD42b+ (PMP42) and CD41+ (PMP41), all measured in platelet-poor plasma by flow cytometry; concentrations are in millions / mL. ADAMTS13 activity was assayed by a FRETS-vWF73 method of

Kokame et al [
Br J Haematol
. vWF multimers were analyzed following
Raines et al [
Thromb Res

RESULTS: At admission, TTP patients exhibited very low ADAMTS13 levels, 0–10% of controls (ctl), and elevated EMP31 (2.5 ±0.7 vs. 0.35 ±0.09 ctl) and EMP62 (6.3 ±1.9 vs. 0.25 ±0.07 ctl), both p<0.001. After the first TPE, EMP fell significantly, to 1.76 ±0.85 for EMP31 and 3.4 ±1.3 for EMP62, both p<0.02. Similarly, EMPVWF dropped from 3.7 ±1.8 at admission (ctl = 0.28 ±0.20) to 2.6 ±1.0 after the first TPE, p=0.02. In parallel, ADAMTS13 activity slowly increased with each TPE, finally to near normal, 85 ±22% of ctl. These findings suggest a relationship between ADAMTS13 activity and MP counts. The MP fraction isolated from TTP patients had no detectable ADAMTS13 activity. When the MP fraction from TTP patients was incubated with normal particle free plasma (PFP) for 6hr, then centrifuged to remove MP, a significant fraction of activity was lost (25 ±8%, p=0.02), and no activity could be detected in the resuspended MP. Examination of the vWF multimer composition of the MP fraction of TTP patients exhibited enrichment of large vWF. When we assayed FFP for EMP, PMP and ADAMTS13 activity, we found EMP and LMP comparable to controls (EMP31 = 0.4 ±0.25; EMP62 = 0.31 ±0.2). Only PMP were significantly elevated in FFP (PMP42 = 4.4 ±1.8; PMP41 = 9.0 ±3.2). When we compared ADAMTS13 activity in the MP fraction and the PFP of FFP, the MP fraction accounted for 35 ±12% of the total activity. MP from FFP did not exhibit ULvWF. We consistently observed that immediately following each TPE, both EMP31 and EMP62 were reduced by 15–30%, with a parallel increase in both species of PMP of 10–15%. Concomitantly, EMP and PMP profiles in plasma removed through TPE from TTP patients remained comparable to samples obtained prior to the TPE.

CONCLUSION: These results suggest that one action of TPE in TTP is to remove EMP which tend to bind and inactivate ADAMTS13. Thus, the basis of TPE is two-fold: infusion of ADAMTS13 by FFP infusion, and reduction of high levels of inhibitors including inhibitory antibodies as well as EMP.

Disclosure: No relevant conflicts of interest to declare.

Author notes


Corresponding author