Abstract 1162

Introduction:

Since the advent of plasma exchange, mortality from thrombotic thrombocytopenic purpura (TTP) has gone from approximately 90% to less than 20%. However, due to thrombocytopenia, placement of a central venous catheter for plasma exchange can cause concern on the part of the proceduralist. Traditional teaching has been that platelet transfusion is contraindicated in TTP unless there is severe bleeding. However, a more recent review shows that the evidence for harm from platelet transfusion is uncertain. Due to the risk of bleeding, proceduralists often ask for platelet transfusion before catheter insertion. We conducted a review in our institution of bleeding episodes after cathether insertion in patients with suspected TTP.

Methods:

A single institution retrospective review was performed at SUNY Upstate Medical University on patients with presumed TTP from January 1999 until July of 2011. Patients were identified using both pheresis records and diagnosis codes of admitted patients in order to ensure that patients with catastrophic bleeding episodes prior to plasma exchange were not omitted. Each event was defined as placement of a pheresis catheter in a patient who presented with presumed TTP. Data was collected on patient age, platelet count prior to line insertion, the service performing the insertion, use of image guidance, bleeding complications, and survival.

Results:

There were 55 patients with a total of 57 catheter insertion attempts in thrombocytopenic patients with presumed TTP. One patient had an unsuccessful insertion attempt and another had catheter malfunction; both required a second insertion attempt while thrombocytopenic. Overall there were no major bleeding complications and no bleeding that required invasive intervention or removal of the catheter. There were 3 episodes of bleeding that resolved with pressure. Otherwise, there was minor bleeding that did not require any intervention documented in 14 cases. These included oozing, saturation of dressing, and 6 hematomas which did not require treatment. There was a single unsuccessful attempt at blind catheter placement which resulted in a hematoma.

The median platelet count on catheter insertion was 26K with a range from 3K to 128K. Platelet transfusion was given prior to catheter placement in 14 of the episodes. The median platelet count in this population was 12K before transfusion. 5 of these 14 patients had minor bleeding complications (35%). In the 43 attempts in patients who did not receive platelet transfusion prior to line insertion, the median platelet count was 25K. 12 of these patients had minor bleeding (28%). All 3 patients with minor bleeding which required noninvasive intervention were in this group. Of the 57 attempts at line insertion, 32 were in the femoral vein, 23 in the internal jugular vein and 2 in the subclavian vein. Image guidance was used in 25 of the 57 attempts. Of the bleeding episodes that required noninvasive intervention, 2 occurred after blind placement of the catheter and 1 was after image guidance was used.

Of the 55 patients treated for presumed TTP, 8 died during admission (15%). Mortality in the group of patients receiving platelet transfusion prior to catheter placement was 43% versus 5% in the patients without platelet transfusion beforehand. When adjusted for 4 patients with other causes for their thrombotic microangiopathy, the mortality in the transfused group was 40%. In general, patients receiving platelet transfusion prior to catheter insertion seemed more acutely ill, including 1 patient with HIV and pancreatitis, 1 patient with HIV, and 1 patient with pancreatitis alone.

Conclusion:

In this single institution retrospective review there were no major bleeding complications associated with plasma exchange catheter insertion in thrombocytopenic patients with presumed TTP. In light of the uncertain risk of platelet transfusion in patients with TTP, it is reasonable to forgo prophylactic platelet transfusion prior to catheter placement.

Disclosures:

No relevant conflicts of interest to declare.

This icon denotes a clinically relevant abstract

Sign in via your Institution