Search for a command to run...
Introduction: Experimental data indicate angiotensin II induces plasminogen activator-inhibitor 1 and may promote platelet activation, however the clinical relevance of this in patients receiving synthetic angiotensin II for shock remains unclear. The objective of this study was to assess the timing and incidence of thromboembolic events in the angiotensin II for the treatment of high-output shock (ATHOS-3) clinical trial. Methods: This was a post-hoc analysis of ATHOS-3, a phase 3 registration trial that randomized patients with vasopressor-refractory vasodilatory shock to angiotensin II or placebo. Thromboembolic events were captured through day 28. Cox proportional hazards regression was used to compare the probability of surviving without a thromboembolic event within 28-days between the groups. Results: Of the 321 patients randomized, 21 (12.9%) of those receiving angiotensin II and 8 (5.1%) of those receiving placebo had a thromboembolic event documented during follow-up. The median duration of study drug exposure was 48 hours in both groups. At 28-days, 88 (54%) of angiotensin II and 73 (46%) of placebo patients were alive (p = 0.12). Patients in the angiotensin II group (28 days, IQR 6-28) had non-significantly more median observable days alive during the study period than the placebo group (17 days, IQR 5-28), p = 0.08. Numerically, less thromboembolic events occurred during study drug infusion or within 3 days of study drug discontinuation in the angiotensin II (n = 12; 57%) compared to placebo (n = 7; 88%) group (p = 0.12). The distribution of thromboembolic event severity graded as 1-2 or 3-4 was similar between the groups. The probability of surviving without a thromboembolic event at 28-days was similar between the angiotensin II and placebo groups (HR 1.12, 95% CI 0.83-1.52). Conclusions: Thromboembolic events in angiotensin II recipients tended to occur later, with nearly half occurring more than 3 days after study drug discontinuation. The probability of surviving without a thromboembolic event at 28-days was similar between those randomized to angiotensin II and placebo. Epidemiological studies evaluating thromboembolic events in angiotensin II recipients are warranted and should consider competing risks and confounders.