Association between the plasma-to-red blood cell ratio and survival in geriatric and non-geriatric trauma patients undergoing massive transfusion: a retrospective cohort study

Abstract

Background: The benefits of a high plasma-to-red blood cell (RBC) ratio on the survival of injured patients who receive massive transfusions remain unclear, especially in older patients. We aimed to investigate the interaction of age with the plasma-to-RBC ratio and clinical outcomes of trauma patients.

Methods: In this retrospective study conducted from 2013 to 2016, trauma patients who received massive transfusions were included. Using a generalized additive model (GAM),we assessed how the plasma-to-RBC ratio and age affected the in-hospital mortality rates. The association of the plasma-to-RBC ratio [low (< 0.5), medium (0.5-1.0), and high (≥ 1.0)] with in-hospital mortality and the incidence of adverse events were assessed for the overall cohort and for patients stratified into non-geriatric (16-64 years) and geriatric (≥ 65 years) groups using logistic regression analyses.

Results: In total, 13,894 patients were included. The GAM plot of the plasma-to-RBC ratio for in-hospital mortality demonstrated a downward convex unimodal curve for the entire cohort. The low-transfusion ratio group was associated with increased odds of in-hospital mortality in the non-geriatric cohort [odds ratio 1.38, 95% confidence interval (CI) 1.22-1.56]; no association was observed in the geriatric group (odds ratio 0.84, 95% CI 0.62-1.12). An increase in the transfusion ratio was associated with a higher incidence of adverse events in the non-geriatric and geriatric groups.

Conclusion: The association of the non-geriatric age category and plasma-to-RBC ratio for in-hospital mortality was clearly demonstrated. However, the relationship between the plasma-to-RBC ratio with mortality among geriatric patients remains inconclusive.

Keywords: Damage control resuscitation; Geriatric trauma; Massive transfusion; Trauma registry.

Fig. 1

Diagramatic representation of patients who received massive transfusion in the Trauma Quality Improvement Program database

Fig. 2

Generalized additive model evaluating the association between plasma-to-red blood cell ratios and outcomes. Association of plasma-to-red blood cell ratios and study outcomes in the entire cohort (A–C) and in the non-geriatric group (D–F) and geriatric group (G–I). Mortality and adverse events were analyzed using the non-linear logistic generalized additive model adjusted for sex, injury mechanisms, Revised Trauma Score, Injury Severity Score, and Abbreviated Injury Scale Score in each body region (head, neck, thorax, abdomen, upper extremities, and pelvis/lower extremities), total prehospital time, trauma center level, and hospital type (university, community, and non-teaching). Adverse events included cardiac, respiratory, and renal failure, as well as thrombotic events and sepsis. The shaded region represents the 95% confidence intervals for the estimated points

Fig. 3

Cumulative survival curves using Cox proportional hazard regression model with a time-dependent covariate. The cumulative survival curves for the low (< 0.5), medium (0.5–1.0), and high (≥ 1.0) plasma-to-red blood cell transfusion ratio groups, which were estimated using the Cox proportional hazard regression model for non-geriatric group (A) and geriatric group (B) patients. The hazard ratio for 28-day mortality was controlled for age, sex, injury severity score, Revised Trauma Score, injured region, and hospital type. The plasma-to-RBC ratios at 4 h and 24 h were treated as time-dependent covariates