Emergency Management of Patients with Immune Thrombocytopenia and Severe Bleeding

Background

Severe bleeding is a rare but potentially life-threatening complication of immune thrombocytopenia (ITP). In this study, we describe the management and clinical outcomes of patients with ITP presenting to the emergency room (ER) with thrombocytopenia and severe bleeding over a 9-year period from 3 Canadian tertiary care centers.

Methods

Patients with ITP who had severe thrombocytopenia (platelets <20 x10⁹/L) and severe bleeding, and who presented to any of 3 hospital emergency rooms (ER) affiliated with McMaster University in Hamilton, Canada between 2008 and 2016 were included in this retrospective study. We reviewed ER and hospital records and extracted demographic data, platelet counts, and all treatments received until discharge or day 10 of hospital admission. We described all bleeding and thrombotic events. Bleeding was graded from 0 (none) to 2 (severe) at one or more anatomical sites using the ITP Bleeding Score (Page et al, Br J Haematol 2007). Patients with skin or oral mucous membrane bleeding only were excluded. Data from a sample of 41 charts were extracted in duplicate (by SM and NS) to ensure consistency; discrepancies were resolved by the principal investigator (D.M.A.). Data were summarized descriptively as frequencies (percentage), means (standard deviation [SD]), or medians (interquartile range [IQR]). Ethics approval was obtained from the Hamilton Integrated Research Ethics Board. Funding for this study was provided by a grant from the Platelet Disorder Support Association.

Results

Thirty-one patients had platelets <20 x10⁹/L with grade 2 bleeding. Mean (SD) age was 56 years (±24) and 16 patients (52%) were female. Twenty-three patients (74%) had primary ITP; others had secondary ITP due to medications (n=2), systemic lupus erythematosus (n=2), malignancy (n=2), hepatitis (n=1), or other infection (n=1). Sixteen patients (52%) had newly-diagnosed ITP; 15 (48%) had known ITP and had previously received a median (IQR) of 3 (3-4) ITP therapies including splenectomy (n=7). Median (IQR) platelet count at presentation to the ER was 4 x10⁹/L (2-8 x10⁹/L).

There were 37 ER visits among 31 patients. Sites of grade 2 bleeding at presentation to the ER were: GI (18/37 visits, 49%), epistaxis (10/37 visits, 27%), urinary (6/37 visits, 16%), gynecological (4/37 visits, 11%) and intracranial (4/37 visits, 11%). Patients were given a median (range) of 2 (0-7) treatment modalities in hospital including corticosteroids (30/37 visits, 81%), intravenous immune globulin (25/37 visits, 68%) and platelet transfusions (17/37 visits, 46%). Other treatments were romiplostim (3/37 visits), tranexamic acid (3/37 visits); and emergency splenectomy, recombinant factor VIIa, Rh immune globulin, danazol, azathioprine and mycophenolate (1/37 visits each).

Four patients experienced new or recurrent grade 2 bleeding in-hospital, including two deaths: a 76 year-old female with a platelet count of 11 x10⁹/L on warfarin who developed ICH (received 4 treatment modalities); and a 20-year-old male with a platelet count of 8 x10⁹/L and ICH (received 7 treatment modalities). Median (IQR) length of hospital admission was 4 (3-9) days for 29 admissions.

Conclusions

Severe bleeding in patients with ITP is a high risk situation that requires prompt intervention. Treatments and outcomes were variable. A standardized approach to the treatment of bleeding emergencies in ITP is needed.