Risk Factors for Bleeding in Coronavirus Disease 2019 Patients on Extracorporeal Membrane Oxygenation and Effects of Transcatheter Arterial Embolization for Hemostasis

Abstract

Purpose: To evaluate risk factors for bleeding events in coronavirus disease 2019 (COVID-19) patients on extracorporeal membrane oxygenation (ECMO) and to share the initial results of transcatheter arterial embolization (TAE) for hemostasis.

Material and methods: Forty-three COVID-19 patients who received ECMO from May 2020 to September 2021 were enrolled in this study. Patients with sudden onset anemia immediately underwent computed tomography to assess bleeding. We compared laboratory data, duration of ECMO, hospitalization period, and fatality of patients' groups with and without significant hemorrhagic events using the chi-square test and Mann-Whitney U test. We also assessed the results of TAE in patients who received hemostasis.

Results: A total of 25 bleeding events occurred in 24 of the 43 patients. Age was a risk factor for bleeding events and fatality. The average duration of ECMO and hospitalization period were significantly longer in those with bleeding events (42.9 and 54.3 days) than in those without bleeding events (16.2 and 25.0 days) (p < 0.05). In addition, those with bleeding had higher fatality (45.8%) than those without (15.8%) (p < 0.05). Active extravasation was confirmed for 5 events in 4 of 24 patients. TAE was attempted and performed successfully in all but one of these four cases, in whom bleeding ceased spontaneously.

Conclusions: Elderly COVID-19 patients on ECMO had a greater risk of bleeding complications and fatal outcomes. TAE was effective in providing prompt hemostasis for patients who have the treatment indication.

Keywords: bleeding; coronavirus disease 2019 (COVID-19); extracorporeal membrane oxygenation (ECMO); extravasation; transcatheter arterial embolization (TAE).

Figure 1.

A summary flowchart of patient selection.

Figure 2.

CECT of Case 1 at the first bleeding event. (a) The arterial-phase CECT image shows a jet of contrast medium within the hematoma of the left psoas muscle (arrow). (b) The jet of contrast medium is enlarged on the delayed-phase image (arrow). This is the classic pattern of active extravasation.

Figure 3.

Angiography of Case 1 at the first bleeding event. (a) 2D-DSA of the left internal iliac artery (IIA) was obtained after a 5.0-French catheter (Twist B, Medikit Co., Ltd., Tokyo, Japan) was advanced into the left IIA. 2D-DSA shows active extravasations from the branch of the left iliolumbar artery (arrows). (b) The 2D-DSA image of the branch of the left iliolumbar artery was obtained after a 1.9-French microcatheter (Nadeshiko Akane, JMS Co., Ltd., Hiroshima, Japan) navigated by a 0.014-inch guidewire (Transend, Boston Scientific Japan, Tokyo, Japan) was coaxially advanced. 2D-DSA shows active extravasations from the branch of the left iliolumbar artery (arrows). (c) The 2D-DSA image of the left IIA after TAE shows the disappearance of active extravasation. TAE was performed using a mixture of N-butyl-2-cyanoacrylate and lipiodol at a volume ratio of 1:3 after the microcatheter was advanced as close as possible to the bleeding points.

Figure 4.

CT and angiography of Case 1 at the second bleeding event. (a) The delayed-phase image of CECT shows the active extravasation of contrast medium in the ileal lumen (arrow). (b) 2D-DSA of the superior mesenteric artery (SMA) was obtained after a 5.0-French catheter (Twist B) was advanced into the SMA. 2D-DSA shows the active extravasation of contrast medium from the limbs of the vasa recta of the ileal artery (arrow). (c) The 2D-DSA image of SMA after TAE shows the disappearance of active extravasation (arrow). TAE was performed using two pushable metallic microcoils after a 1.9-French microcatheter (Nadeshiko Akane) navigated by a 0.014-inch guidewire (Transend) was coaxially advanced as close as possible to the bleeding points.

Figure 5.

CECT of Case 2. (a) The arterial-phase CECT image shows a jet of contrast medium within the hematoma of the left rectus abdominis muscle (arrow). (b) The jet of contrast medium is enlarged on the delayed-phase image (arrows). This is the classic pattern of active extravasation.

Figure 6.

Angiography of Case 2. (a) 2D-DSA of the left external iliac artery (EIA) was obtained after a 5.0-French catheter (Twist B) was advanced into the left EIA. 2D-DSA shows active extravasations from the branch of the inferior epigastric artery (arrow). (b) 2D-DSA of the branch of the inferior epigastric artery was obtained after a 1.9-French microcatheter (Nadeshiko Akane) navigated by a 0.014-inch guidewire (Transend) was coaxially advanced. (c) 2D-DSA of the left EIA after TAE shows the disappearance of active extravasation (arrows). TAE was performed using gelatin sponge particles after the microcatheter was advanced as close as possible to the bleeding point.