978 Correspondence

Nwawka OK, Kabutey NK, Locke CM, Castro-Aragon I, Kim D. Ultrasound-guided nee- dle localization to aid foreign body removal in pediatric patients. J Foot Ankle Surg 2014;53(1):67-70. http://dx.doi.org/10.1053/j.jfas.2013.09.006 [Epub 2013 Nov 13].

Borgohain B, Borgohain N, Handique A, Gogoi PJ. Case report and brief review of literature on Sonographic detection of accidentally implanted wooden foreign body causing persistent sinus. Crit Ultrasound J 2012;4(1):10. http://dx.doi.org/10. 1186/2036-7902-4-10.

Jeckovic M, Anupindi SA, Barbir SB, Lovrenski J. Is ultrasound useful in detection and follow-up of gastric foreign bodies in children? Clin Imaging 2013;37(6):1043-7.

Tahmasebi M, Zareizadeh H, Motamedfar A. Accuracy of ultrasonography in detec- ting radiolucent soft-tissue foreign bodies. Indian J Radiol Imaging 2014;24(2): 196-200.

Utility of shock index calculation in hemorrhagic trauma

We read with great interest the article by Edla et al [1] comparing Heart rate variability metrics vs routine vital signs as diagnostic tests to improve trauma patient management focusing on the identifica- tion of trauma patients with major hemorrhage. They conducted a mul- tivariate analysis using routine vital signs (heart rate, respiratory rate, systolic blood pressure, and Pulse pressure) as the comparator to test the hypothesis that HRV metrics can improve the identification of pa- tients with major hemorrhage. However, when combined with routine vital signs, HRV added negligible additional discriminatory value. The authors addressed a very important question as far as the most substantial clinical problem facing physicians being the identifica- tion of hemorrhagic trauma. In prehospital setting, current trauma triage relies on abnormal physiological criteria to determine the patient’s Mode of transport, priority of treatment, destination for treat- ment, and need for possible Life-saving interventions.

We would like to go further into the debate and speculate that calculation of the Shock Index may be more useful for caregivers than isolated measurements of systolic blood pressure (SBP) and heart rate (HR) in the compensatory phase of shock. The SI is defined as the ratio of HR to SBP. This easily calculable score in the field has been demonstrated to be a pragmatic and useful guide for diagnosing acute hypovolemia in the presence of normal HR and blood pressure. Shock index has been shown to correlate with other indices of end- organ perfusion such as Central venous oxygen saturation and arteri- al lactic acid concentration [2]. Compared with HR or SBP alone, SI has been suggested to be a better measure of hemodynamic stability [3]. Rady et al [4] evaluated a SI cutoff point of 0.9 in a cohort of 275 adult patients presenting to an emergency department with stable vital signs. The authors found that a SI greater than 0.9 was associat- ed with an illness that was treated immediately, admission to the hospital, and intensive therapy on admission. A given set of vital signs may on initial interpretation appear unalarming, but calcula- tion of SI added additional perspective that could influence Clinical decisions [5].

To conclude, we would like to know if the authors, maybe based on a retrospective analysis of the data set of 402 subjects, could test the use- fulness of SI (with a cutoff value of 0.9) in initial assessment of patients with ongoing exsanguinations?

Marilyn Franchin, MD^?

Daniel Jost, MD Hugues Lefort, MD Stephane Travers, MD Jean-Pierre Tourtier, MD

Emergency Medical Department, Fire Brigade of Paris, Paris, France

References

1. Edla S, Reisner AT, Liu J, et al. Is heart rate variability better than routine vital signs for pre-hospital identification of major hemorrhage? Am J Emerg Med 2015;33(2): 254-61.
2. Rady MY. The role of central venous oximetry, lactic acid concentration and shock index in the evaluation of clinical shock: a review. Resuscitation 1992;24:55-60.
3. Birkhahn RH, Gaeta TJ, Terry D, Bove JJ, Tloczkowski J. Shock index in diagnosing early acute hypovolemia. Am J Emerg Med 1994;23:323-6.
4. Rady MY, Smithline HA, Blake H, Nowak R, Rivers E. A comparison of the shock index and conventional vital signs to identify acute, critical illness in the emergency depart- ment. Ann Emerg Med 1994;24:685-90.
5. Vandromme MJ, Griffin RL, Kerby JD, et al. Identifying risk for massive transfusion in the relatively normotensive patient: utility of the prehospital shock index. J Trauma 2011;70:384-8.

   In reply to “Utility of shock index calculation in hemorrhagic trauma”?

   

   To the Editor,

   We wish to thank the correspondents for their interest and com- ments regarding our report [1]. We agree that multivariate vital-sign analysis is a powerful tool. The Shock Index , which scales the heart rate (HR) to the systolic blood pressure (SBP), is attractive because it can be computed mentally at the bedside. At least in theory, by exam- ining multiple vital signs, one may better distinguish Abnormal vital signs due to psychological distress (typically tachycardia with hyper- tension) vs blood loss and shock (relative tachycardia with normal or reduced blood pressure). In addition to the reports cited by the corre- spondents, SI has been studied in trauma registries of more than 16000 [2] and 21000 [3] patients, demonstrating that blood transfusion requirement and mortality are associated with increasing SI.

   To address the question posed by the correspondents, we computed the areas under receiver operating characteristic curves (ROC AUCs) for SI using the same data set of 402 subjects from Edla et al [1]. We used that report’s methodology for excluding unreliable vital signs and ana- lyzed the average vital-sign values from each subject’s initial 15 minutes of physiological data. The ROC AUCs for SI were 0.76, 0.80, and 0.81 for predicting 24-hour red blood cell transfusion greater than or equal to 1, 5, and 9 units, respectively. These ROC AUCs for SI trend higher than the ROC AUCs for HR and SBP (available in Table 2 from Edla et al [1]), although the differences were not statistically significant. The sensitivity and specificity of SI greater than 0.9 as a predictor of massive transfusion (defined as 24-hour red blood cell transfusion >= 9 units) were 63% and 83%, respectively, using the 15-minute average of SBP and HR.

   One challenge of SI is that its value changed minute by minute be- cause the patient’s HR fluctuated. Many patients developed SI greater than 0.9 at least at some time point during early trauma care. In a sepa- rate analysis of 855¹ subjects during prehospital transport [4], we found that 57% of the patients with no significant bleeding nonetheless dem- onstrated SI greater than 0.9, at least transiently. We found that SI great- er than 1.4 was a more practical cutoff, with a false-positive rate of only 12% in patients without bleeding; and it was sensitive to 59% of massive transfusion patients. (For comparison, note that SBP b 90 mmHg had a false-positive rate of 10% in patients without bleeding; and it was sensi- tive to 50% of massive transfusion patients.)

   At the bedside, clinicians should consider computing SI using a time- averaged value of HR and SBP from a multiminute observation interval to reduce False alarms [5]. There are also statistical techniques that can

   ?Corresponding author at: Emergency Medical Department

   Fire Brigade of Paris. 1, place Jules Renard F-75017 Paris, France

   E-mail address [email protected] http://dx.doi.org/10.1016/j.ajem.2015.04.001

   ? Conflicts of interest: None of the authors have any conflicts of interest to disclose.

   ¹ The 402 subjects from Edla et al [1] comprise the subset of this larger data set of 855 subjects [4] with a full set of reliable vital signs and at least 5 minutes of reliable electrocar- diogram waveform data for heart rate variability analysis.