SummaryThe clinical characteristics and long-term outcomes of patients presenting with acute pulmonary embolism (PE) during treatment with warfarin have not been described. Clinical details of all patients admitted to a tertiary institution from 2000-2007 with acute PE were retrieved retrospectively, baseline warfarin status and the international normalised ratio (INR) were recorded, and their outcomes tracked using a statewide death registry. Of 923 patients with clearly documented warfarin status included in this study, 83 (9%) were taking warfarin. Mean (± standard deviation) day-1 INR of those taking warfarin was 2.3 ± 0.9, with 67% of patients therapeutically anti-coagulated (INR ≥2.0) at presentation (49 patients with INR <2.5 and 34 with INR ≥2.5). Patients taking warfarin on admission were more likely to have heart failure, atrial fibrillation and valvular heart disease, with similar prevalence of malignancy and ischaemic heart disease, compared to patients not on warfarin. Total mortality of the cohort (mean follow-up 4.0 ± 2.5 years) was 31.6% (in-hospital mortality 1.5%), and was similar between warfarin and no warfarin groups. There was however a greater than four-fold increased risk of post-discharge death due to recurrent PE for the patients taking warfarin on admission (hazard ratio [HR] 4.43, 95% confidence interval [CI] 1.36-14.42, p=0.01). Among patients taking warfarin on admission, day-1 INR <2.5 significantly increased long-term all-cause mortality compared to INR ≥2.5 (adjusted HR 2.51, 95% CI 1.08-5.86, p=0.03). In conclusion, patients presenting with PE during treatment with warfarin have an increased risk of death from recurrent PE. Admission INR appears to have independent long-term prognostic importance in these patients.