Predicting risk of death in ED resuscitation room among trauma patients is the focus of clinical practice. A prediction tool that is accurate and easy-to-use is expected by clinicians.
qSOFA, as a simple novel prediction score, is used to predict in-hospital mortality in non-ICU patients with suspected infection [11]. Studies have also verified the validity of qSOFA in predicting outcome of ED patients with and without suspected infections [18], as well as patients with burn, cancer, pesticide poisoning, and blunt trauma [14-17].
We performed a single center, retrospective study of adult trauma patients triaged to ED resuscitation room. In the current study, our purpose was to explore the predictive validity of qSOFA for death in ED resuscitation room among trauma patients. This study demonstrates that qSOFA score has good correlation with death, higher qSOFA score means higher risk of death.
RTS has been developed for use in triage and outcome prediction among injury patients [4]. High score of RTS indicates high survival probability [4]; but high score of qSOFA means low survival probability. qSOFA score was negatively correlated with RTS (r=-0.38, p<0.001). This result was also confirmed by R.S. Jawa et al. [17], supporting our result. These data further suggest that high qSOFA score is indicative of a risk of an unfavorable prognosis following trauma.
In the present study, patients with a qSOFA score of 0 had a 0.6% incidence rate of death. As the qSOFA score increased from 1 to 3, the rate of death significantly increased from 3.28% to 15.38%. The death proportions were significantly higher in patients with qSOFA of 2 or more. We found that elevated qSOFA scores were directly associated with increased death proportion in Likelihood-ratio Chi squared test. This tendency was in concordance with those reported by Singer AJ et al. [18] and R.S. Jawa et al. [17]. In the research by Singer AJ et al., qSOFA score were associated with mortality (0 [0.6%], 1 [2.8%], 2 [12.8%], and 3[25.0%]). While in the research by R.S. Jawa et al., qSOFA scores were associated with in-hospital mortality (1.7% with qSOFA equals to 0; 8.7% with qSOFA equals to 1; 22.4% with qSOFA equals to 2; 23.1% with qSOFA equals to 3; p < 0.001). SilvioA et al. [15] revealed that for patients with qSOFA score<2 vs qSOFA score ≥ 2, the hospital mortality rate was 7.36% vs 35.7% (28.3%; 95% CI, 13%-47.7%, p<0.001). But an interesting result was reported by A. Prasad, et al. [14], the highest mortality rate was in the group of qSOFA score = 2 (12.2%), with none in the group of score = 3, which is inconsistent with our finding, but this result was not discussed by authors.
As higher qSOFA scores were more correlated with death on univariate analysis, we examined its performance by multivariate analysis. Using the multivariate logistic regression (after adjustment for sex, age, RTS), we found qSOFA scores of 2 and 3 were independently associated with death. We cautiously suggest that clinicians should pay more attention and give more frequent monitoring to trauma patients with qSOFA scores of 2 or more at presentation.
Among patients with suspected infection in non-ICU, the AUC of qSOFA for predicting in-hospital mortality was 0.81 (95% CI, 0.80-0.82) [13]. While in our study, the AUC of qSOFA for predicting death was 0.78 [95% CI, 0.72-0.85]. The two predictive values are close. The performance of the qSOFA in the current study was also similar to that reported by Singer AJ et al. [18]. In which, the AUC for predicting mortality among patients with and without suspected infection were 0.75 (95% CI 0.71 to 0.78) and 0.70 (95% CI 0.65 to 0.74), respectively.
In addition, our objects were confined to trauma patients triaged to ED resuscitation room. Trauma victims triaged to consulting room were excluded, they were usually less severely ill and generally at lower risk of death. And our study endpoint was death occurred in ED resuscitation room. Death after leaving the ED resuscitation room was not covered by this study. These might lead to an underestimation of total death toll, and an over- or underestimation of death proportion, even an over- or underestimation of predictive capacity of qSOFA for death.
Several limitations exist in our study. This was a retrospective study from a single center, which is subject to selection bias. The results may not be representative. In the future, large, multi-center retrospective reviews as well as prospective researches may be required to determine whether the qSOFA scores can accurately predict death in ED resuscitation room among trauma patients.