Impact of a Pharmacist-Driven Prothrombin Complex Concentrate Protocol on Time to Administration in Patients With Warfarin-Associated Intracranial Hemorrhage

Jessica L. Corio, PharmD; Jonathan H. Sin, PharmD; Bryan D. Hayes, PharmD; Joshua N. Goldstein, MD, PhD; Lanting Fuh, PharmD


Western J Emerg Med. 2018;19(5):849-854. 

In This Article


We conducted a single-center, retrospective review of consecutive patients issued 4F-PCC for warfarin-associated ICH. Patient characteristics collected included patient age, sex, type of ICH, and dose of 4F-PCC administered. Data points collected included ED registration time, initial INR, time INR was drawn and resulted, time ICH was confirmed on imaging, and documented time of 4F-PCC administration. We obtained institutional review board approval, and the need for informed consent was waived.

Our institution is a 1000-bed, Level I trauma and major regional referral center with more than 110,000 annual ED visits. During the study period, clinical pharmacists were physically present in the ED daily from 07:30 a.m. to midnight. During the overnight period, pharmacy services were provided from a separate, centralized location. The blood bank was staffed 24/7. Emergency medicine clinical pharmacy services were established in our ED prior to development of this protocol; therefore, no changes in the pharmacy-staffing model were required to support implementation of the pharmacist-driven 4F-PCC protocol.

We identified patients issued 4F-PCC between September 2015 and February 2017 from the hospital's EMR and the blood bank data system. We chose parallel six-month pre- and post-implementation time periods for investigation. Patients treated between September 2015 and February 2016 were considered part of the pre-protocol group, and those treated between September 2016 and February 2017 were considered part of the pharmacist-driven protocol group. The new protocol was implemented in April 2016, but full education of clinicians was not yet complete and operational components were still being optimized. For this reason, we excluded this transitional period and included patients starting from September 2016 for the purposes of this analysis.

Patients were included in the analysis if they met the following criteria: ≥18 years of age; ICH confirmed on imaging; documented warfarin use; and initial INR ≥2 (Figure). Patients were excluded if they received more than one dose of 4F-PCC during the same hospitalization or if they received 4F-PCC under the pharmacist-driven protocol outside of ED clinical pharmacist coverage hours. To account for potential changes in staffing and blood bank workflow during the overnight time period, patients in the pre-protocol group were also excluded if they presented between midnight and 7:30 a.m. While hospital policy permitted 4F-PCC use for indications other than warfarin-associated ICH during both study periods, such use was rare and required extra levels of approval, introducing excess variability. As a result, we only analyzed those with warfarin-associated ICH for ease of analysis.


Inclusion and exclusion criteria.
4F-PCC, four-factor prothrombin complex concentrate; ICH, intracranial hemorrhage INR, international normalized ratio.

The primary outcome of the study was the amount of time from when the patient met criteria for 4F-PCC to the time of administration. The criteria required both an initial INR ≥2 and confirmation of ICH on imaging; the latter of the two recorded times was designated as the earliest time 4F-PCC was indicated for use. All time stamps were determined from the EMR, including the resulting time of the INR documented by the laboratory, and the final read time on neuroimaging results. For patients who were transferred to our facility, if both the patient's INR and neuroimaging were already available from outside hospital records and used for 4F-PCC criteria, then the arrival time to our ED was designated as the starting time point. This starting time point was chosen since our ED clinicians receive the INR and neuroimaging information telephonically prior to the patient's arrival. Once the patient arrived to the ED, they were already considered a 4F-PCC candidate. Our protocol's intention was for the 4F-PCC procurement process to begin immediately upon arrival. However, if the transferred patient did not have both INR and neuroimaging results readily available and communicated to the clinicians, then the starting time point was pushed back until the patient was officially deemed a 4F-PCC candidate, after the missing information was resulted in our ED.

Secondary outcomes included dose of 4F-PCC in concordance with INR and weight-based FDA-label dosing recommendations and hospital protocol, as well as concomitant IV vitamin K administration. We also evaluated in-hospital mortality between the two protocol groups.

For our study purposes, appropriate dosing of 4F-PCC was determined based on the patient's pre-treatment INR and recorded weight at the time of administration. Administered 4F-PCC doses exceeding five units per kilogram above or below the recorded weight were operationally defined as "inappropriate." This was to account for potential differences in estimation of patient weight, in cases when an accurate weight was not easily attainable.

We considered that if our protocol shortened time by 20 minutes it would be deemed clinically relevant. A power calculation showed that we would need 32 patients, 16 in each arm, to detect this difference at the p <0.05 level. We analyzed the primary outcome using the Mann-Whitney U test. Baseline characteristics, secondary outcomes, and clinical outcomes were assessed using Student's t-test, chi-square test, or Fisher's exact test where appropriate. A p-value <0.05 was noted to be statistically significant.