Enhancing the Fever Workup Utilizing a Multi-Technique Modeling Approach to Diagnose Infections More Accurately

Abstract

Background:

Differentiation between infectious and non-infectious etiologies of the systemic inflammatory response syndrome (SIRS) in trauma patients remains elusive. We hypothesized that mathematical modeling in combination with computerized clinical decision support would assist with this differentiation. The purpose of this study was to determine the capability of various mathematical modeling techniques to predict infectious complications in critically ill trauma patients and compare the performance of these models with a standard fever workup practice (identifying infections on the basis of fever or leukocytosis).

Methods:

An 18-mo retrospective database was created using information collected daily from critically ill trauma patients admitted to an academic surgical and trauma intensive care unit. Two hundred forty-three non-infected patient-days were chosen randomly to combine with the 243 infected-days, which created a modeling sample of 486 patient-days. Utilizing ten variables known to be associated with infectious complications, decision trees, neural networks, and logistic regression analysis models were created to predict the presence of urinary tract infections (UTIs), bacteremia, and respiratory tract infections (RTIs). The data sample was split into a 70% training set and a 30% testing set. Models were compared by calculating sensitivity, specificity, positive predictive value, negative predictive value, overall accuracy, and discrimination.

Results:

Decision trees had the best modeling performance, with a sensitivity of 83%, an accuracy of 82%, and a discrimination of 0.91 for identifying infections. Both neural networks and decision trees outperformed logistic regression analysis. A second analysis was performed utilizing the same 243 infected days and only those non-infected patient-days associated with negative microbiologic cultures (n = 236). Decision trees again had the best modeling performance for infection identification, with a sensitivity of 79%, an accuracy of 83%, and a discrimination of 0.87.

Conclusion:

The use of mathematical modeling techniques beyond logistic regression can improve the robustness and accuracy of predicting infections in critically ill trauma patients. Decision tree analysis appears to have the best potential to use in assisting physicians in differentiating infectious from non-infectious SIRS.

Get full access to this article

View all access options for this article.

References

Edwards

, Peterson

, Andrus

et al. National Healthcare Safety Network (NHSN) Report: Data summary for 2006. Am J Infect Control, 2007; 35:290–301.

Hurr

, Hawley

, Czachor

et al. APACHE II and ISS scores as predictors of nosocomial infections in trauma patients. Am J Infect Control, 1999; 27:79–83.

Jarvis

. Selected aspects of the socioeconomic impact of nosocomial infections: Morbidity, mortality, cost, and prevention. Infect Control Hosp Epidemiol, 1996; 17:552–557.

Freischlag

, Busuttil

. The value of postoperative fever evaluation. Surgery, 1983; 94:358–363.

Hoover

, Bochicchio

, Napolitano

et al. Systemic inflammatory response syndrome and nosocomial infection in trauma. J Trauma Inj Infect Crit Care, 2006; 61:310–316.

Miller

, Munn

, Meredith

, Chang

. Systemic inflammatory response syndrome in the trauma intensive care unit: Who is infected? J Trauma Inj Infect Crit Care, 1999; 47:1004–1008.

Papia

, McLellan

, El-Helou

et al. Infection in hospitalized trauma patients: Incidence, risk factors, and complications. J Trauma Inj Infect Crit Care, 1999; 47:923–927.

Schey

, Salom

, Papadia

et al. Extensive fever workup produces low yield in determining infectious etiology. Am J Obstet Gynecol, 2005; 192:1729–1734.

Wallace

, Cinat

, Gornick

et al. Nosocomial infections in the surgical intensive care unit: A difference between trauma and surgical patients. Am Surg, 1999; 65:987–990.

10.

Claridge

, Sando

, Golob

et al. The “fever workup” and respiratory culture practice in critically ill trauma patients. J Crit Care, 2009.

11.

Golob

, Claridge

, Fadlalla

et al. Fever and leukocytosis in critically ill trauma patients: It's not the blood. Am Surg., 2009; 75:405–410.

12.

Golob

, Claridge

, Sando

et al. Fever and leukocytosis in critically ill trauma patients: It's not the urine. Surg Infect, 2008; 9:49–56.

13.

Wagenlehner

, Loibl

, Vogel

et al. Incidence of nosocomial urinary tract infections on a surgical intensive care unit and implications for management. Int J Antimicrob Agents, 2006; 28,Suppl 1:S86–S90.

14.

Murray

, Hoffmaster

, Schmit

et al. Evaluation of white blood cell count, neutrophil percentage, elevated temperature as predictors of bloodstream infection in burn patients. Arch Surg, 2007; 142:639–642.

15.

Theuer

, Bongard

, Klein

. Are blood cultures effective in the evaluation of fever in perioperative patients? Am J Surg, 1991; 162:615–618.

16.

Fadlalla

AMA

, Golob

, Claridge

. The Surgical Intensive Care–Infection Registry (SIC-IR): A research registry with daily clinical support capabilities. Am J Med Qual, 2009; 24:29–34.

17.

Golob

, Fadlalla

, Kan

et al. Validation of SIC-IR^©: A medical informatics system for intensive care unit research, quality of care improvement, and daily patient care. J Am Coll Surg, 2008; 207:164–173.

18.

Charalambous

, Swoboda

, Dick

et al. Risk factors and clinical impact of central line infections in the surgical intensive care unit. Arch Surg, 1998; 133:1241–1246.

19.

Pawar

, Kapoor

, Sharma

et al. Central venous catheter-related blood stream infections: Incidence, risk factors, outcome, and associated pathogens. J Cardiothorac Vasc Anesth, 2004; 18:304–308.

20.

Angstwurm

, Gaertner

, Schopohl

. Outcome in elderly patients with severe infection is influenced by sex hormones but not gender. Crit Care Med, 2005; 33:2786–2793.

21.

Bochicchio

, Joshi

, Bochicchio

et al. Incidence and impact of risk factors in critically ill trauma patients. World J Surg, 2006; 30:114–118.

22.

Bochicchio

, Joshi

, Knorr

et al.

Impact of nosocomial infections in trauma: Does age make a difference?

J Trauma Inj Infect Crit Care, 2001; 50:612–617.

23.

Crabtree

, Pelletier

, Gleason

et al. Gender-dependent differences in outcome after the treatment of infection in hospitalized patients. JAMA, 1999; 282:2143–2148.

24.

Jamulitrat

, Narong

, Thongpiyapoom

. Trauma severity scoring systems as predictors of nosocomial infection. Infect Control Hosp Epidemiol, 2002; 23:268–273.

25.

Coopersmith

, Kollef

. ACS surgery online. Dale

, Federman

. Critical Care, Postoperative and Ventilator-Associated Pneumonia. WebMD Inc., 2000.

26.

Bleeker

, Moll

, Steyerberg

et al. External validation is necessary in prediction research: A clinical example. J Clin Epidemiol, 2003; 56:826–832.

27.

Terrin

, Schmid

, Griffith

et al. External validity of predictive models: A comparison of logistic regression, classification trees, and neural networks. J Clin Epidemiol, 2003; 56:721–729.

28.

Vergouwe

, Steyerberg

, Eijkemans

et al. Substantial effective sample sizes were required for external validation studies of predictive logistic regression models. J Clin Epidemiol, 2005; 58:475–483.

29.

Becalick

, Coats

. Comparison of artificial intelligence techniques with UKTRISS for estimating probability of survival after trauma. UK Trauma and Injury Severity Score. J Trauma Inj Infect Crit Care, 2001; 51:123–133.

30.

Clermont

, Angus

, DiRusso

et al. Predicting hospital mortality for patients in the intensive care unit: A comparison of artificial neural networks with logistic regression models. Crit Care Med, 2001; 29:291–296.

31.

DiRusso

, Chahine

, Sullivan

et al. Development of a model for prediction of survival in pediatric trauma patients: Comparison of artificial neural networks and logistic regression. J Pediatr Surg, 2002; 37:1098–1104.

32.

DiRusso

, Sullivan

, Holly

et al. An artificial neural network as a model for prediction of survival in trauma patients: Validation for a regional trauma area. J Trauma Inj Infect Crit Care, 2000; 49:212–220.

33.

Gabbe

, Cameron

, Wolfe

et al. Prehospital prediction of intensive care unit stay and mortality in blunt trauma patients. J Trauma Inj Infect Crit Care, 2005; 59:458–465.

34.

Wolfe

, McKenzie

, Black

et al. Models developed by three techniques did not achieve acceptable prediction of binary trauma outcomes. J Clin Epidemiol, 2006; 59:26–35.

35.

Boyd

, Tolson

, Copes

. Evaluating trauma care: The TRISS method. J Trauma Inj Infect Crit Care, 1988; 27:370–378.

36.

Baxt

. Use of an artificial neural network for the diagnosis of myocardial infarction. Ann Intern Med, 1991; 115:843–848.

37.

Doyle

, Dvorchik

, Mitchell

et al. Predicting outcomes after liver transplantation: A connectionist approach. Ann Surg, 1994; 219:408–415.

38.

Tourassi

, Floyd

, Sostman

et al. Artificial neural network for diagnosis of acute pulmonary embolism: Effect of case and observer selection. Radiology, 1995; 194:889–893.

39.

Vyborny

, Giger

. Computer vision and artificial intelligence in mammography. AJR Am J Roentgenol, 1994; 162:699–708.

40.

Leibovici

, Gitelman

, Yehezkelli

et al. Improving empirical antibiotic treatment: Prospective, nonintervention testing of a decision support system. J Intern Med, 1997; 242:395–400.

41.

Leibovici

, Paul

, Nielsen

et al. The TREAT project: Decision support and prediction using causal probabilistic networks. J Antimicrob Chemother, 2007; 30,Suppl 1:S93–S102.

42.

Paul

, Andreassen

, Tacconelli

et al. Improving empirical antibiotic treatment using TREAT, a computerized decision support system: Cluster randomized trial. J Antimicrob Chemother, 2006; 58:1238–1245.

43.

Paul

, Nielsen

, Goldberg

et al. Prediction of specific pathogens in patients with sepsis: Evaluation of TREAT, a computerized decision support system. J Antimicrob Chemother, 2007; 59:1204–1207.

44.

Kass

. An exploratory technique for investigating large quantities of categorical data. Appl Stat, 1980; 29:119–127.

45.

Breiman

, Friedman

, Olshen

, Stone

. Classification and Regression Trees. New York: Chapman and Hall, 1984.

46.

Loh

, Shih

. Split selection methods for classification trees. Stat Sin, 1997; 7:815–840.

47.

Quinlan

. C4.5: Programs for Machine Learning. San Francisco: Morgan Kaufmann Publishers, 1993.