Article

The RTS plus measurement of the RDW improves the prediction of 28-day mortality in trauma patients

1112 Correspondence / American Journal of Emergency Medicine 36 (2018) 10981120

William C. Dillon, MD

Start The Heart Foundation(TM), Baptist Medical Associates, 3900 Kresge

Way, Suite 60, Louisville, KY 40207, USA E-mail address: [email protected].

18 September 2017

https://doi.org/10.1016/j.ajem.2017.10.026

References

  1. Bobrow BJ, et al. Chest compression-only CPR by lay rescuers and survival from out- of-hospital cardiac arrest. JAMA 2010;304(13):1447-54.
  2. Anderson ML, et al. Rates of cardiopulmonary resuscitation training in the United

States. JAMA Intern Med 2014;174(2):194-201.

  1. American Heart Association. I. CPR in schools. [cited 2016 June 30]; available from: http://cpr.heart.org/AHAECC/CPRAndECC/Programs/CPRInSchools/UCM_475820_ CPR-in-Schools-Legislation-Map.jsp; 2016.
  2. Abella BS, et al. Quality of cardiopulmonary resuscitation during in-hospital cardiac arrest. JAMA 2005;293(3):305-10.
  3. Abella BS, et al. CPR quality improvement during in-hospital cardiac arrest using a

Real-time audiovisual feedback system. Resuscitation 2007;73(1):54-61.

  1. Swor R, et al. CPR training and CPR performance: do CPR-trained bystanders perform CPR? Acad Emerg Med 2006;13(6):596-601.

Fig. 1. The ROC curves of RDW and RTS-R.

The RTS plus measurement of the RDW improves the prediction of 28-day mortality in trauma patients

To the Editor,

The article by Taeyoung Kong et al. is indeed interesting [1]. We ap- prove the value of Red blood cell distribution width predicting the 28-day mortality in Severe trauma patients.

RDW is a measure of the range of variation of red blood cell volume [1], however, indicates nothing about the physiological findings, which is quite important for the predicting of hospital mortality in trauma pa- tients [2,3]. Normally, we use the Revised Trauma Score , consisting of three physiological indexes, in prehospital [4] and in- hospital [5,6] practice to estimate the prognosis of trauma patients. Whereas, lacking for Blood indicators is one of the disadvantages of RTS, but that is possible to be solved by integrating RDW into RTS [8]. According to the original article, we speculated that RDW could provide an significant clinical insight on the prognosis of Trauma victims, and the entire value of combining RDW with RTS could improve the Predictive power of RDW alone.

Therefore, we conducted a hospital-based pilot study, trying to cre- ate a new measuring scale, adding RDW to RTS score, named RTS-R. Firstly, we transferred RDW values into standard scores corresponding with scores of RTS. According to the cutoff value (14.4%) from the orig- inal article, we set RDW values N 14.4% were score 0, while the rest were score 1. So RTS-R is the sum of scores of RDW and RTS. After that, we col- lected retrospectively all cases in Emergency Department of the West China Hospital during 1st April 2016 to 30th September 2016. Data sources contained death certificates and medical records. A checklist was used for data gathering, which included four parts: (1) baseline characteristics such as age, gender, (2) data for calculation of RTS,

(3) RDW from the records and (4) 28-day outcome (death) of the

Table 1

The comparison of AUC of RDW and RTS-R.

AUC SEa 95% CIb Difference between areas P value RDW 0.711 0.0572 0.615 to 0.794 0.180 0.001

RTS-R 0.891 0.0457 0.815to 0.943

a Standard error.

b Confidence interval.

patients. Before data collection, the senior emergency medicine resi- dents executing the research were trained on filling the checklist and calculating RTS score. As for data analysis, a Receiver Operating Charac- teristic curve (ROC curve) was performed to compare the Predictive ability between RDW and RTS-R. Statistical analyses were performed using the MedCalc software Ver. 12.7.0.0.

Excluding the missing data (n = 6), 107 cases were enrolled in the study. Of the patients, the mean age was 45.33 +- 19.34 years with 70 male patients (65.4%), and the number of death was 17(15.9%). The RDW ranged from 12.1% to 23.5%. The RTS-R ranged from 6 to 13. Pair-wise comparison of ROC curves are showed in Fig. 1.The AUC of the RTS-R was 0.891, significantly higher than that of the RDW (AUC of RDW = 0.711; P = 0.001 b 0.05, Table 1).

The RDW is an independent indicator of predicting mortality of severe trauma patients. Similar results could also be found in other studies such as Giuseppe Lippi’s [7]. Nevertheless, we consider it will be more meaningful to combine RDW with physiological parame- ters. Our results show that RTS-R is a better predictor of 28-day mortal- ity of trauma patients than RDW itself. Furthermore, RTS-R is a simple and practical tool to apply in clinical situations of emergency depart- ment. However, our pilot study was limited by its retrospective study design, small sample size and missing data. Prospective multicenter studies with adjustments for the effectiveness and application prospect of RTS-R are warranted.

Overall, RTS-R can be adopted to predict the hospital mortality of se- vere trauma patients and it is better than employing RDW solely.

Acknowledgement

This work was supported by the Research Fund of the Health and Family Planning Commission of China (201302003) and the Health and Family Planning Commission of Chengdu City (CDWSYJ-2016-01).

Zi-han Liu

West China School of Medicine, Sichuan University, Sichuan, PR China

Hai Hu

Department of Emergency Medicine, West China Hospital,

Sichuan University, Sichuan, PR China Corresponding author at: Department of Emergency Medicine, West China Hospital, Chengdu, Sichuan 610041, PR China.

E-mail address: [email protected].

Correspondence / American Journal of Emergency Medicine 36 (2018) 10981120 1113

https://doi.org/10.1016/j.ajem.2017.10.028

References

6 October 2017

the authors provide the ORs, CIs, and the area under the curve (AUC) that have been corrected for model uncertainty and optimism.

Acknowledgment

This work was not supported by any organization.

  1. Kong Taeyoung, Park Jong Eun, Park Yoo Seok, Lee Hye Sun, You Je Sung, Chung Hyun Soo, et al. Usefulness of serial measurement of the red blood cell distribution width to predict 28-day mortality in patients with trauma. Am J Emerg Med 2017 Jun 8 pii: S0735-6757(17)30445-X https://doi.org/10.1016/j.ajem.2017.06.008.
  2. Ljunggren Malin, Castren Maaret, Nordberg Martin, Kurland Lisa. The association be- tween vital signs and mortality in a retrospective cohort study of an unselected emer-

Conflict of interest

None.

Saeid Safiri

gency department population. Scand J Trauma Resusc Emerg Med 2016 Mar 3;24:21. https://doi.org/10.1186/s13049-016-0213-8.

  1. Mehmood Amber, He Siran, Zafar Waleed, Baig Noor, Sumalani Fareed Ahmed, Razzak Juanid Abdul. How vital are the vital signs? A multi-center observational study from emergency departments of Pakistan. BMC Emerg Med 2015;15(Suppl. 2):S10. https://doi.org/10.1186/1471-227X-15-S2-S10.
  2. Lichtveld RA, Spijkers ATE, Hoogendoorn JM, Panhuizen IF, van der Werken Chr. Tri- age Revised Trauma Score change between first assessment and arrival at the hospital to predict mortality. Int J Emerg Med 2008 Apr;1(1):21-6. https://doi.org/10.1007/ s12245-008-0013-7.
  3. Chawda MN, Hildebrand F, Pape HC, Giannoudis PV. Predicting outcome after multi- ple trauma: which scoring system? Injury 2004 Apr;35(4):347-58. https://doi.org/ 10.1016/S0020-1383(03)00140-2.
  4. Singh J, Gupta G, Garg R, Gupta A. Evaluation of trauma and prediction of outcome using TRISS method. J Emerg Trauma Shock 2011 Oct;4(4):446-9. https://doi.org/ 10.4103/0974-2700.86626.
  5. Lippi Giuseppe, Bovo Chiara, Buonocore Ruggero, Mitaritonno Michele, Cervellin Gianfranco. Red blood cell distribution width in patients with limb, chest and head trauma. Arch Med Sci 2017;13(3):606-11. https://doi.org/10.5114/aoms.2017.67282.
  6. Majercik Sarah, Fox Jolene, Benjamin D. red cell distribution width is predictive of mortality in trauma patients. J Trauma Acute Care Surg 2013 Aug;75(2):345-6. https://doi.org/10.1097/TA.0b013e31829957c0.

Comments on Predictive performance of serum S100B for neuronal damage and poor clinical outcomes

Dear Editor,

We read with great interest the study by Yilmaz J and colleagues [1]. After reading this article carefully and critically, we have some concerns to point out as follows:

The second end points in the study were composite clinical out- comes in SC users, which combine the need for endotracheal intubation, incidence of seizures, the need for intensive care unit admission, and in- hospital mortality [1]. The results have demonstrated that odds ratio (OR) (95% confidence interval [CI]) for effect of S100B on composite clinical outcomes was 13.2 (2.1 to 28.1) [1]. Although the results were very interesting, however, for application in clinical practice it is neces- sary to estimate association between S100B and each specific outcome such as in-hospital mortality separately.

Commonly used strategy to have effective sample size and gain in precision is combining specific outcomes into a broader category such as composite clinical outcomes. However, nowadays, advanced simple statistical methods such as hierarchical regression approach are avail- able for estimating the association between an exposure and a number of different outcomes separately [2,3]. Here, the authors could reanalyze their data with aforementioned method and provide association between S100B and the need for endotracheal intubation, incidence of seizures, the need for intensive care unit admission, and in-hospital mortality separately.

Another shortage of the study is that the prediction model in the study fully specified the based on observed data, and the model uncer- tainty following statistical modelling have been ignored. Such model uncertainty can be limited with bootstrap validation [4,5]. We suggest

Managerial Epidemiology Research Center, Department of Public Health,

School of Nursing and Midwifery, Maragheh University of Medical Sciences,

Maragheh, Iran

Erfan Ayubi Department of Epidemiology, School of Public Health, Shahid Beheshti University of Medical Sciences, Tehran, IranDepartment of Epidemiology & Biostatistics, School of Public Health, Tehran University of Medical Sciences,

Tehran, Iran Corresponding author at: Department of Epidemiology, School of Public Health, Shahid Beheshti University of Medical Sciences, Tehran, Iran.

E-mail address: [email protected]

7 September 2017

https://doi.org/10.1016/j.ajem.2017.10.031

References

  1. Yilmaz S, Karakayali O, Kale E, Akdogan A. Could serum S100B be a predictor of neu- ronal damage and clinical poor outcomes associated with the use of synthetic canna- binoids? S100B to predict neuronal damage of SC in the ED. Am J Emerg Med 2018; 182:435-41.
  2. Richardson DB, Hamra GB, MacLehose RF, Cole SR, Chu H. Hierarchical regression for

analyses of multiple outcomes. Am J Epidemiol 2015;182:459-67.

  1. Ashrafi-Asgarabad A, Behroozi A, Safiri S. The predictive value of plasma catestatin for all-cause and cardiac deaths in chronic heart failure patients: methodological issues. Peptides 2017 (in press).
  2. Steyerberg E. Clinical prediction models: A Practical Approach to Development, Vali- dation, and Updating. Springer Science & Business Media; 2008.
  3. Ashrafi-Asgarabad A, Ayubi E, Safiri S. Predictors of health-related quality of life in people with amyotrophic lateral sclerosis: methodological issues. J Neurol Sci 2017; 372:228.

Ability of emergency medicine residents in the diagnosis of CHF with a preserved ejection fraction by echocardiogram

To the Editor,

Heart failure is a syndrome with significant mortality and mor- bidity including frequent hospitalizations and poor quality of life. Many patients with heart failure develop symptoms that prompt them to visit the emergency department for evaluation. Previous re- search has shown that emergency medicine physicians are able to accurately diagnosis heart failure with reduced left ventricular ejec- tion fraction [1-3]. However, half of all patients with heart failure have a preserved left ventricular ejection fraction (HF/pEF) [4]. Addi- tional articles have demonstrated emergency medicine physicians are able to diagnose HF/pEF [5-6], but no studies thus far have looked at emergency medicine residents’ ability to accurately diagnose HF/ pEF.

We set out to assess whether emergency medicine residents could be taught to diagnose HF/pEF based on echocardiographic im- ages as well as to see how well they retained this knowledge. This study was a prospective longitudinal cohort study following

Leave a Reply

Your email address will not be published. Required fields are marked *