Profile of Thrombosis Distribution Following Haploidentical and Identical-Sibling Transplantation: A 15-Year Experience at a Single Center in China

Abstract

Introduction:

Thrombotic episodes are a serious cause of morbidity and mortality in hematopoietic stem cell transplantation (HSCT) patients. There have been many studies on the incidence and risk factors of thrombosis after HSCT, but no systematic and comprehensive summary of the distribution, frequency and occurrence time of thrombosis in allogeneic HSCT recipients is available. Given that the complications after HSCT correlate with the transplant type, the incidence of thrombosis may also differ in different transplant protocals. Our center has explored and carried out haploidentical HSCT for a long time, and this study aimed to compare the thrombosis characteristics between haploidentical HSCT and identical-sibling HSCT.

Methods:

A retrospective study of allogeneic HSCT recipients at the Peking University Institute of Hematology was conducted from 2003, the initiation of haploidentical HSCT, to 2017. All haploidentical HSCT donors were treated with G-CSF, and the HSCT recipients received a modified busulfan (BU) / cyclophosphamide (CY) + antihuman thymocyte immunoglobulin (ATG) regimen for myeloablative conditioning and a cyclosporine A (CsA) + mycophenolate mofetil (MMF) + short-term methotrexate (MTX) regimen for GVHD prophylaxis. Thrombotic episodes in our study were classified into groups including deep-vein thrombosis (DVT), pulmonary embolism (PE), catheter-induced thrombosis (CIT), stroke, portal vein thrombosis (PVT) and others. DVT and PE belonged to venous thromboembolic events (VTE). Univariate and multivariable Cox models were used to explore the risk factors, including hematopoietic cell transplant comorbidity index (HCT-CI), for the development of thrombotic events.

Results:

We identified 2392 haploidentical HSCT recipients and 727 identical-sibling HSCT recipients, and a total of 219 (7.02%) thromboembolic events were registered. Among them, 166 thromboembolic events developed after haploidentical HSCT, including 51 cases (2.13%) of DVT, 3 cases (0.13%) of PE, 63 cases (2.63%) of CIT, 40 cases (1.67%) of stroke, 4 cases (0.17%) of PVT and 5 cases (0.21%) of other thrombotic episodes. Additionally, 42 thromboembolic events developed after identical-sibling HSCT, including 14 cases (1.93%) of DVT, 0 cases of PE, 17 cases (2.34%) of CIT, 10 cases (1.38%) of stroke, and 1 case (0.14%) of PVT. The incidence of thrombosis in haploidentical and identical-sibling HSCT recipients was not significantly different (P>0.05 ).

CIT appeared on day 26 of BMT (median value; range, 15 to 48), VTE (PE, DVT) on day 215 (range, 20 to 732), stroke on day 47 (range, 22 to 268), and PVT on day 55 (range, 40 to 274). The overall survivals (OSs) of these thrombosis groups were 92.1%, 90.7% , 37.5% and 25%. Patients with VTE and CIT have a significantly longer OS than those with PVT and stroke (P<0.05).

Based on our nested-control study, HCT-CI ≥ 3 (P<0.05) and GVHD (P<0.05)were significant risk factors for thrombosis occurrence in HSCT recipients in multivariable Cox models.

Conclusion:

The incidence of thrombosis in haploidentical and identical-sibling HSCT recipients was not significantly different. CIT was the most common thromboembolic event in HSCT recipients at our center, followed by VTE, stroke, and PVT. Among them, patients with VTE and CIT have a significantly longer OS than those with PVT and stroke. HCT-CI ≥ 3 and GVHD were significant risk factors for thrombosis occurrence in HSCT recipients.