출혈성 쇼크를 일으킨 흰쥐에서 선형회귀 분석모델을 이용한 출혈량 추정

Blood Loss Prediction of Rats in Hemorrhagic Shock Using a Linear Regression Model

  • 이탁형 (연세대학교 생체공학협동과정) ;
  • 이주형 (연세대학교 생체공학협동과정) ;
  • 최재림 (연세대학교 생체공학협동과정) ;
  • 양동인 (연세대학교 생체공학협동과정) ;
  • 김덕원 (연세대학교 생체공학협동과정)
  • Lee, Tak-Hyung (Graduate Program in Biomedical Engineering, Yonsei University) ;
  • Lee, Ju-Hyung (Graduate Program in Biomedical Engineering, Yonsei University) ;
  • Choi, Jae-Rim (Graduate Program in Biomedical Engineering, Yonsei University) ;
  • Yang, Dong-In (Graduate Program in Biomedical Engineering, Yonsei University) ;
  • Kim, Deok-Won (Graduate Program in Biomedical Engineering, Yonsei University)
  • 발행 : 2010.01.25

초록

출혈성 쇼크는 응급실에서 일어나는 사망 원인의 많은 부분을 차지하고 있다. 본 연구의 목적은 출혈량에 따라 변화하는 생리적인 변수들의 특징을 알아보는 것이다. 또한 이를 이용하여 전체 혈액량 대비 손실된 혈액의 비율을 산출하는 선형회귀분석 모델을 만드는 것이다. 총 60마리의 흰쥐를 출혈량에 따라 체중 100g 당 15분 동안 0ml, 2ml, 2.5ml 3ml로 정하여 총 4그룹으로 나누었다. 출혈 중에 변화하는 심박수, 수축기혈압, 이완기혈압, 호흡수, 체온 등을 분석하였다. 분석한 데이터를 무작위로 나누어 360개의 데이터 세트를 선형회귀 분석모델을 만드는데 사용했고 이 모델의 R (결정계수) 제곱 값은 0.80이었다. 나머지 360개의 데이터를 이용하여 만든 모델을 시험한 결과, 추정된 손실 혈액의 비율의 RMS (root mean square) 오차 값은 5.7%가 나왔다. 비록 선형회귀분석모델이 직접적으로 실제 임상에서 사용될 수 없지만 추가적인 연구를 통해 이 방법이 출혈성 쇼크의 소생술을 시행하는데 필요한 용액의 양을 결정하는데 도움을 줄 수 있을 것으로 생각된다.

Hemorrhagic shock is a common cause of death in the emergency department. The purpose of this study was to investigate the relationship between blood loss as a percent of the total estimated blood volume (% blood loss) and changes in several physiological parameters. The other goal was to achieve an accurate prediction of percent blood loss for hemorrhagic shock in rats using a linear regression model. We allocated 60 Sprague-Dawley rats into four groups: 0ml, 2ml, 2.5ml, 3 mL/100 g during 15 min. We analyzed the heart rate, systolic and diastolic blood pressure, respiration rate, and body temperature in relation to the percent blood loss. We generated a linear regression model predicting the percent blood loss using a randomly chosen 360 data set and the R-square value of the model was 0.80. Root mean square error of the tested 360 data set using the linear regression was 5.7%. Even though the linear regression model is not directly applicable to clinical situation, our method of predicting % blood loss could be helpful in determining the necessary fluid volume for resuscitation in the future.

키워드

참고문헌

  1. World Health Organization, "World Health Statistics 2008," World Health Organization Press, pp. 46-47, 2008.
  2. Deakin CD, Hicks IR. "AB or ABC: pre-hospital fluid management in major trauma," J Accid Emerg Med, Vol. 11, no. 3, pp. 154–157,September 1994. https://doi.org/10.1136/emj.11.3.154
  3. Collins. T, "Understanding Shock". Nurs Stand, Vol. 14, no. 49, pp. 35-41, August 2000.
  4. Wilson M, Davis DP, Coimbra R, "Diagnosis and monitoring of hemorrhagic shock during the initial resuscitation of multiple trauma patients: a review," J Emerg Med, Vol 24, no. 4, pp. 413-422, May 2003. https://doi.org/10.1016/S0736-4679(03)00042-8
  5. Gutierrez G, Reines HD, Wulg-Gutierrez ME, "Clinical review: hemorrhagic shock," Crit Care, Vol 8, no. 5, pp. 373-381, October 2004. https://doi.org/10.1186/cc2851
  6. Yang S, Zou LY, Bounelis P, Chaudry I, ChathamJC, MarchaseRB, "Glucosamine administration during resuscitation improves organ function after trauma hemorrhage," Shock, Vol. 25, no. 6, pp. 600-607, June 2006. https://doi.org/10.1097/01.shk.0000209563.07693.db
  7. Akira Takasu, Toshihisa Sakamoto, Yoshiaki Okada, "Effect of induction rate for mild hypothermia on survival time during uncontrolled hemorrhagic shock in rats." J Trauma, Vol. 61 no. 6, pp. 1330-1335, December 2006. https://doi.org/10.1097/01.ta.0000245980.12711.6a
  8. Chen L, McKenna TM, Reisner AT, Gribok A, Reifman J, "Decision tool for the early diagnosis of trauma patient hypovolemia." J Biomed Inform, Vol 41, no. 3, pp. 469-478, January 2008. https://doi.org/10.1016/j.jbi.2007.12.002
  9. Jefferson MF, Pendleton N, Mohamed S, Kirkman E, Little RA, Lucas SB, Horan MA, "Prediction of hemorrhagic blood loss with a genetic algorithm neural network." J Appl Physiol, Vol. 84, no. 1, pp. 357-361, January 1998. https://doi.org/10.1152/jappl.1998.84.1.357