Journal of Emergencies, Trauma, and Shock
Home About us Editors Ahead of Print Current Issue Archives Search Instructions Subscribe Advertise Login 
Users online:1105   Print this pageEmail this pageSmall font sizeDefault font sizeIncrease font size   

EDITORIAL Table of Contents   
Year : 2010  |  Volume : 3  |  Issue : 4  |  Page : 309-310
What's new in Emergencies Trauma and Shock? Still searching for a scoring system for sepsis!

Trauma, Surgical Critical Care and Emergency Surgery, St Lukes Health Network, Bethlehem, PA

Click here for correspondence address and email

Date of Web Publication28-Sep-2010

How to cite this article:
Grossman MD. What's new in Emergencies Trauma and Shock? Still searching for a scoring system for sepsis!. J Emerg Trauma Shock 2010;3:309-10

How to cite this URL:
Grossman MD. What's new in Emergencies Trauma and Shock? Still searching for a scoring system for sepsis!. J Emerg Trauma Shock [serial online] 2010 [cited 2022 Aug 16];3:309-10. Available from:

Severity of illness scoring systems for critically ill and injured patients have been in existence for more than 20 years. [1] These tools were envisioned as a means of stratifying severity of illness among ICU patients in order to compare performance and outcomes. They use mortality as the discriminate end-point and have been utilized for ICU performance assessment, resource allocation, and clinical research. Most of those in common use including APACHE II and III, SAPS II, and MPM I and II, are based upon a series of physiologic variables coupled with chronic disease components and patient age. The ability of any such scoring system to predict mortality (or any other outcome) requires statistical analysis of an initial clinical database followed by validation of the model in the same data set or another population of patients. The most common method of assessing the validity of predictive models plots the relationship between true and false positive predictions for all predicted probabilities of the outcome in question and calculates the area under the curve (receiver-operator curve) as the C-statistic. A C-statistic in excess of 0.90 indicates a high proportion of true-positive predictions for any given calculated risk of the outcome. The ICU mortality prediction models noted previously have C-statistics between 0.83 and 0.90. [2]

In this issue of JETs Crowe and co-workers attempt to evaluate the effectiveness of three scoring systems developed for predicting hospital mortality based upon data available in the Emergency Department (ED). [3] These scores were evaluated for a series ED patients with severe sepsis and septic shock who received early goal-directed therapy (EGDT) and for whom observed in-hospital mortality was 32.9%. They suggest that these tools might be useful to stratify patients in order to define subsets that would benefit from treatment and that the specific scoring systems tested have the advantage of simplicity compared with more cumbersome and complex ICU scoring systems. In this study the authors found C-statistics between 0.59 and 0.74 with the MEDS score performing best. The authors conclude that these scores had relatively low validity in the current data set and should not be used to make decisions about patient care.

While these conclusions are supported by the authors' data there are other important considerations implied by this work. Age, malignancy, and certain other variables common to all scoring systems might profoundly impact decisions to continue aggressive management following the ED phase of care. The frequency with which care is withdrawn as a byproduct of patient predetermination and/or failure to respond to resuscitative measures following the ED phase of care is often an unknown, as it was in the current study. Next, the idea that hospital mortality is equated with therapy provided in the ED is an important concept that has received only indirect analysis. Nonetheless, despite the myriad factors that potentially impact on outcome from severe sepsis following the ED phase of care, there appears to be a direct correlation between the presence of early and appropriate care and hospital mortality. [4] These observations also seem to hold true for the management of ST segment elevation myocardial infarction and thrombo-embolic stroke. [5],[6] Perhaps lessons learned 40 years ago in the earliest trauma outcome studies describing a "golden hour" for trauma patients might be generalized to patients with critical illness. [7] Indeed commonly used scoring systems for predicting mortality including those used in the present study generally lack a "time since onset" variable and efforts to improve validity of ICU scoring systems have focused upon elements of care prior to arrival in the ICU (i.e. direct admission vs. transfer from a general care ward). Finally, given these assumptions regarding timely and aggressive care in the absence of patient pre-determination, the concept of scoring patients during the emergency phase of care might be somewhat counter intuitive. The sickest patients, most at risk of dying, would seem to be those most in need of early aggressive care.

In general, scoring systems in both critical care and emergency medicine may be more useful for research and quality improvement than clinical care. They do afford the opportunity to stratify and compare interventions, outcomes. and cost-effectiveness and in this sense are very useful in the administration of programs caring for acutely injured and ill patients. It seems however that repeated efforts to use these scores to determine which patients should be treated or when they should be treated are unlikely to be fruitful given the uncertainties of the scoring systems when applied to individual patients.

It has now been well established that the timely care of acutely ill and injured patients using basic principles of resuscitation offers the best opportunity for a favorable outcome. Current iterations of ICU scoring systems include measures of the response to these interventions. Since the capacity to reverse physiologic abnormalities and the time course over which cellular homeostasis is restored seems to be the best determinant of observed outcome, it seems unlikely that any of the currently available scoring systems provide adequate sensitivity to make decisions about whether or not to initiate aggressive measures during the early resuscitative phase of care in the ED. In short, these systems provide too little discrimination too soon in the continuum of care.

   References Top

1.Knaus WA, Draper EA, Wagner DP, Zimmerman JE. APACHE II: A Severity of Disease Classification System. Crit Care Med 1985;13:818-29.  Back to cited text no. 1  [PUBMED]    
2.Castella X, Artigas A, Bion J, Kari A. A comparison of severity of illness scoring systems for ICU patients: Results of a multicenter, multinational study. The European / North American Severity Study Group. Crit Care Med 1995;23:1327-35.  Back to cited text no. 2  [PUBMED]  [FULLTEXT]  
3.Colleen A. Crowe, Erik B. Kulstad, Chintan D. Mistry, Christine E. Kulstad. Evaluation of severity of illness scoring systems in the prediction of hospital mortality in severe sepsis and septic shock. J Emerg Trauma Shock 2010 in press.  Back to cited text no. 3      
4.Rivers E, Nguyen B, Havstad S, Ressler J, Muzzin A, Knoblich B, et al. Early Goal-Directed Therapy in the Treatment of Severe Sepsis and Septic Shock. N Engl J Med 2001;345:1368-77.  Back to cited text no. 4  [PUBMED]  [FULLTEXT]  
5.Cannon CP, Gibson CM, Lambrew CT, Shoultz DA, Levy D, French WJ, et al. Relationship of symptom onset to balloon time and door-to-balloon time with mortality in patients undergoing angioplasty for acute myocardial infarction. JAMA 2000;238:2941-7.  Back to cited text no. 5      
6.Sauer JL. Time is Brain-Quantified. Stroke 2006;37:263-6.  Back to cited text no. 6      
7.West JG, Trunkey DD, Lim RC. Systems of Trauma Care. A Study of Two Counties. Arch Surg 1979;114:455-60.  Back to cited text no. 7  [PUBMED]  [FULLTEXT]  

Correspondence Address:
Michael D Grossman
Trauma, Surgical Critical Care and Emergency Surgery, St Lukes Health Network, Bethlehem, PA

Login to access the Email id

Source of Support: None, Conflict of Interest: None

DOI: 10.4103/0974-2700.70740

Rights and Permissions


    Similar in PUBMED
   Search Pubmed for
   Search in Google Scholar for
    Email Alert *
    Add to My List *
* Registration required (free)  


 Article Access Statistics
    PDF Downloaded88    
    Comments [Add]    

Recommend this journal