Intensive plasma exchange increases a disintegrin and metalloprotease with thrombospondin motifs-13 activity and reverses organ dysfunction in children with thrombocytopenia-associated multiple organ failure

Abstract

Background: Thrombocytopenia-associated multiple organ failure (TAMOF) is a poorly understood syndrome in critically ill children. A disintegrin and metalloprotease with thrombospondin motifs (ADAMTS-13), formerly known as von Willebrand factor (VWF) cleaving protease, is decreased in adults with VWF-mediated thrombotic microangiopathy, and intensive plasma exchange (PEx) both replenishes ADAMTS-13 and improves outcome in these patients.

Objectives: To determine whether: 1) critically ill children with TAMOF syndrome have decreased ADAMTS-13 activity, 2) ADAMTS-13 activity correlates with platelet counts and VWF antigen, 3) the autopsies from patients who died with reduced ADAMTS-13 activity have VWF-rich microthrombi, and 4) intensive PEx will restore ADAMTS-13 activity and facilitate organ failure resolution.

Design: First study: observational. Second study: randomized control trial.

Setting: Single center university pediatric intensive care unit.

Patients: First study: thirty-seven consecutive children (17 males and 20 females; ages ranging from 9 days to 23 years) identified with > or = 2 organs dysfunction were enrolled. Seventy-six percent of these children had thrombocytopenia (platelet counts < 100,000/mm3). Five additional critically ill children without MOF were also enrolled. In the second study, children with severe TAMOF (platelet counts < 100,000/mm3 and > 3 organ failure) were randomized to PEx or standard therapy. Primary physicians and parents agreed to enrollment in 10 of the 20 eligible patients with ages ranging from 1 year to 18 years. Five patients received PEx and 5 patients received standard therapy.

Results: First study: children with TAMOF (n = 28) had decreased ADAMTS-13 activity, but similar plasminogen activator inhibitor-1 activity and prothrombin time compared to children with MOF without thrombocytopenia (n = 9, p < 0.05). All non-survivors (n = 7) had TAMOF, reduced ADAMTS-13 activity, and VWF-rich microvascular thromboses at autopsy. In the second study, PEx (n = 5, median 12 days, 4-28 days) restored ADAMTS-13 activity and organ function, compared to standard therapy (n = 5, p < 0.05).

Conclusions: Children with TAMOF syndrome can have VWF-mediated thrombotic microangiopathy. Similar to adult experience, PEx can replenish ADAMTS-13 activity and reverse organ failure.

Figures

Figure 1. Platelet counts recovered in children with multiple organ failure who survived compared to those who did not survive.[1–8] Von Willebrand Factor (VWF)-rich thrombi are seen in the small blood vessels of the brains, lungs, and kidneys of non-survivors with thrombocytopenia-associated multiple organ failure

(A) Black circles represent platelet counts (mean +/− S.D.) in children with multiple organ failure who survived (n = 30). White circles represent platelet counts in children who did not survive (n = 7). Patients who survived recovered their platelet counts compared to children who did not survive (2 Factor ANOVA p < 0.05). (B). Brain (I, IV), kidney (II, V), and lung (III, VI) tissue from a representative patient stained for VWF (brown color) (polyclonal rabbit anti-human VWF antibody). Panels I, II, and III serve as negative controls from the same patients as in Panels IV, V, and VI respectively, labeled with isotype antibody (non-specific rabbit anti-human IgG). The secondary antibody was conjugated with horse radish peroxidase (brown color).

Figure 2. ADAMTS-13 activity, presence of ULVWF multimers and prothrombin time (PT) in critically ill patients with thrombocytopenia associated MOF, MOF without thrombocytopenia, and no MOF

There were a total of 37 MOF and 5 non-MOF critically ill patients enrolled in the first study period. (A) There were 28 MOF patients with thrombocytopenia. 89% (25/28), 53% (15/28) and 46% (13/28) of these patients had low ADAMTS-13 activity, the presence of ULVWF multimers, and elevated PT respectively. All thrombocytopenic MOF patients with ULVWF multimers had low ADAMTS-13 activity. Only 4% (1/28) of thrombocytopenic MOF had normal ADAMTS-13 activity, normal PT, and no ULVWF multimers. (B) There were 9 MOF patients without thrombocytopenia. All 9 of these patients had normal PT. 67% (6/9) of patients had normal ADAMTS-13 of whom 33% (2/6) had the presence of ULVWF multimers. 33% (3/9) patients had decreased ADAMTS-13 of whom 11% (1/9) patients had the presence of ULVWF multimers. (C) There were 5 non-MOF critically ill controls. These patients were not thrombocytopenic and had normal ADAMTS-13 activity, normal PT, and no ULVWF multimers.

Figure 3. Relationships between quartiles of ADAMTS-13 activity and VWF antigen levels, and platelet counts are similar to those reported in adults with secondary thrombotic microangiopathic syndromes.[16;18]

White circles represent critically ill children without thrombocytopenia (platelet counts ≥ 100K) and without multiple organ failure (n = 5). Black circles represent children with multiple organ failure (OFI ≥ 2) (n = 37). ADAMTS-13 activity in quartiles 1, 2, 3, and 4 are defined respectively as 0–24%, 25–49%, 50–74%, 75–100% relative to healthy adult donor plasma activity. Similar to adult reports, ADAMTS-13 activity correlated inversely with VWF antigen levels (A), and positively with platelet counts (B) (Linear Regression, p<0.05).

Figure 4. Plasma exchange replenishes ADAMTS13, and reduces organ failure

Plasma exchange patients are represented by black circles (n = 5), and Standard therapy patients are represented by white circles (n = 5). (A) There was a robust reduction (p<0.001 2 Factor Repeated Measures ANOVA group × time) over the 28 day study period in PELOD scores in patients randomized to the plasma exchange arm compared to those given standard therapies. (B) There was a significant increase in ADAMTS-13 activity with plasma exchange (44 ± 10% on day 1 increased to 56 ± 11.5% on day 7; mean ± SEM) compared to standard therapy (25 ± 11% on day 1 decreased to 17 ± 11% on day 7; mean ± SEM) (p<0.05, 2 Factor ANOVA group × time).

Figure 5. A 3 y.o. male with gram negative sepsis and thrombocytopenia associated multiple organ failure who had decreased ADAMTS-13 activity with ultra-large VWF multimers and increased PT and PAI-1 activity which did not resolve with randomization to standard therapy

(A) Ultra-large VWF multimers were present on day 1 and increased over a 5-day period. C represented healthy control sample. Autopsy showed VWF-rich microvascular thrombi. The brain (B), and kidney (C) are labeled with VWF (brown color) (polyclonal rabbit anti-human VWF antibody). ADAMTS-13 activity was persistently low. VWF antigen levels were more than 200% of normal control. Platelet count decreased and ADAMTS-13 inhibitors and PT (seconds) increased over time.

Figure 6. A 16 y.o. female with unrecognized chronic granulomatous disease, aspergillus infection, P. aeruginosa pneumonia, and thrombocytopenia associated multiple organ failure who had reduced ADAMTS 13 activity with circulating ultra-large VWF multimers and ADAMTS 13 inhibitors was randomized to the plasma exchange arm

Ultra-large VWF multimers, VWF antigen level, and ADAMTS-13 inhibitor levels decreased and ADAMTS-13 activity, and platelet counts increased with three days of plasma exchange. When three organ failure had resolved for 48 hours, plasma exchange was stopped on day four. Thereafter, an increase in VWF antigen and ADAMTS-13 inhibitor levels and a reduction in ADAMTS-13 activity were observed. Plasma exchange therapy was subsequently restarted on day 8 for a total of 14 days of therapy with eventual resolution of thrombocytopenia associated multiple organ failure. PT (seconds) was normal throughout. This child survived.

Figure 7. A 7 y.o. male wilt ALL s/p bone marrow transplant with E. faecalis sepsis. The patient developed thrombocytopenia associated multiple organ failure with decreased ADAMTS-13 activity and increased VWF antigen. Clinical and laboratory abnormalities resolved with randomization to plasma exchange therapy but recrudesced after cessation of the protocol

(A) Plasma exchange was associated with restoration of ADAMTS-13 activity, reduced VWF antigen and ADAMTS-13 inhibitor levels, reduced LDH, and increasing platelet count. Cessation of plasma exchange at 7 days when the OFI was < 3 for 48 hours was associated with a recrudescence of coagulation abnormalities. ADAMTS-13 activity decreased with a persistence of ADAMTS-13 inhibitor levels. Thrombocytopenia worsened and VWF antigen increased. At the request of the primary service, plasma exchange therapy was never reinstituted after the 14 day study period. The child recrudesced and died after 45 days with unremitting thrombocytopenia associated multiple organ failure and an ADAMTS-13 level that was 27% of normal. (B) Autopsy showed VWF-rich (brown color) (polyclonal rabbit anti-human VWF antibody) microvascular thrombi in the brain.