Earlier time to hemostasis is associated with decreased mortality and rate of complications: Results from the Pragmatic Randomized Optimal Platelet and Plasma Ratio trial

Abstract

Backdrop: Clinicians intuitively recognize that faster time to hemostasis is important in bleeding trauma patients, but these times are rarely reported.

Methods: Prospectively collected data from the Pragmatic Randomized Optimal Platelet and Plasma Ratios trial were analyzed. Hemostasis was predefined as no intraoperative bleeding requiring intervention in the surgical field or resolution of contrast blush on interventional radiology (IR). Patients who underwent an emergent (within 90 minutes) operating room (OR) or IR procedure were included. Mixed-effects Poisson regression with robust error variance (controlling for age, Injury Severity Score, treatment arm, injury mechanism, base excess on admission [missing values estimated by multiple imputation], and time to OR/IR as fixed effects and study site as a random effect) with modified Bonferroni corrections tested the hypothesis that decreased time to hemostasis was associated with decreased mortality and decreased incidence of acute kidney injury (AKI), acute respiratory distress syndrome (ARDS), multiple-organ failure (MOF), sepsis, and venous thromboembolism.

Results: Of 680 enrolled patients, 468 (69%) underwent an emergent procedure. Patients with decreased time to hemostasis were less severely injured, had less deranged base excess on admission, and lower incidence of blunt trauma (all p < 0.05). In 408 (87%) patients in whom hemostasis was achieved, every 15-minute decrease in time to hemostasis was associated with decreased 30-day mortality (RR, 0.97; 95% confidence interval [CI], 0.94-0.99), AKI (RR, 0.97; 95% CI, 0.96-0.98), ARDS (RR, 0.98; 95% CI, 0.97-0.99), MOF (RR, 0.94; 95% CI, 0.91-0.97), and sepsis (RR, 0.98; 95% CI, 0.96-0.99), but not venous thromboembolism (RR, 0.99; 95% CI, 0.96-1.03).

Conclusion: Earlier time to hemostasis was independently associated with decreased incidence of 30-day mortality, AKI, ARDS, MOF, and sepsis in bleeding trauma patients. Time to hemostasis should be considered as an endpoint in trauma studies and as a potential quality indicator.

Level of evidence: Therapeutic/care management, level III.

Figures

Figure 1.

Schematic representation of overall time to hemostasis, which consists of time from hospital arrival to OR/IR and time from OR/IR to hemostasis.

Figure 2A–D.

Between-site comparisons for patients requiring emergent OR/IR (n=468). Incidence of blunt (versus penetrating) trauma varied significantly between sites (A). Time from hospital arrival to emergent OR/IR (B), time from emergent OR/IR to hemostasis (C), and overall time to hemostasis (D) also varied significantly between sites. Time from emergent OR/IR to hemostasis accounted for median 73% (IQR 63%–87%) of overall time to hemostasis. *, p

Figure 3.

Locally weighted scatterplot smoothing (LOWESS) regression of incidence of any complication (mortality, AKI, ARDS, MOF, sepsis, or VTE) versus time to hemostasis.

Figure 4A–B.

Mixed effects Poisson regression with robust error variance for patients requiring emergent OR/IR and who achieved hemostasis (n=408). Unadjusted models (A) control for study site as a random effect. Adjusted models (B) control for age, ISS, injury mechanism, time to OR/IR, treatment group, and admission base excess as fixed effects and study site as a random effect. *, statistically significant after modified Bonferroni correction.