Physiologic Response to Hemorrhagic Shock Depends on Rate and Means of Hemorrhage David A. Z. Frankel, M.D., José A. Acosta, M.D., Devashish J. Anjaria, M.D., Rafael Dibs Porcides, M.D., Paul L. Wolf, M.D., Raul Coimbra, M.D., Ph.D., and David B. Hoyt, M.D. 1 University of California, San Diego and Irvine, California Submitted for publication August 15, 2006 Background. Traditional models of shock classify se- verity based on the volume of hemorrhage. Clinically, hemorrhage occurs at a variable rate, usually slowing as blood pressure drops; however most animal exper- imental models use a constant rate of hemorrhage. Our hypothesis was that rapid bleeding followed by slower bleeding using a fixed total volume would re- sult in a greater physiologic insult. Materials and Methods. Yorkshire pigs (S and S Farms, Ranchita, CA) underwent placement of jugular and femoral catheters after anesthesia. All animals were hemorrhaged a total of 30 mL/kg. The animals were di- vided into constant rate hemorrhage over 10 min (Constant-10) (3 mL/kg/min), constant rate hemorrhage over 20 min (Constant-20) (1.5 mL/kg/min), or a varying rate of hemorrhage of 2.15 mL/kg/min over 7 min, and then 1.15 mL/kg/min over the remaining 13 min (Physiologic-20). Shock, mean arterial pressure (MAP) < 20mmHg, was maintained for 60 min. Resuscitation was performed with Ringer’s lactate (RL) and shed blood (2:1 ratio), until shed blood was exhausted and then only RL to maintain a MAP >60 mmHg for 3 h. Results. Physiologic-20 shock resulted in signifi- cantly increased maximal heart rate, peak serum lac- tate, and volume of required RL resuscitation. Ade- quacy of resuscitation was ensured by MAP, urine output, and clearance of serum lactate. Conclusions. A more physiologic method of fixed vol- ume hemorrhagic shock results in a significantly in- creased physiologic response as demonstrated by in- creased volume of fluid resuscitation. This differential physiologic response may represent an improved hem- orrhagic shock model, and could have implications for future hemorrhagic shock studies. © 2007 Elsevier Inc. All rights reserved. Key Words: shock; hemorrhage. INTRODUCTION Traditional models of hemorrhagic shock create hy- potension by withdrawing a fixed volume of blood or by targeting a specific blood pressure and serially with- drawing blood to maintain this target pressure. The problem is that neither model duplicates the physio- logic profile of true clinical hemorrhage. Clinically, we use the Advanced Trauma Life Sup- port (ATLS) classification to describe the severity of shock based on percentage of blood volume lost. Shock is graded class I to class IV based on a hemorrhage of 15%, 15 to 30%, 30 to 40%, and 40% of total blood volume respectively [1]. Classes III and IV are associ- ated with an approximately 30% mortality in the clin- ical scenario and it is this level of shock that most animal models attempt to recreate. These attempts to translate the clinical scenario into a laboratory model, however, do not take into account the varying rates of hemorrhage. Important to remember in creating a more clinically relevant model is that physiologic hemorrhage is not linear. Initial rapid bleeding eventually slows due to normal physiologic mechanisms including vasocon- striction, local tamponade factors, primary hemostasis, and a progressively decreasing blood pressure. Taking these factors into account, a recent study has created a computer generated model of clinical hemorrhage and response to resuscitation [2]. Using a carefully con- structed computer algorithm, a more physiologically accurate blood pressure curve was created, and proved to be different from that of traditional models. The objective of this study was to determine if there are physiologic differences between a more clinically rele- vant model of hemorrhagic shock compared with tra- ditional fixed volume models. Our hypothesis was that rapid bleeding followed by slower bleeding using a fixed total volume would result in a greater physiologic insult in our model. 1 To whom correspondence and reprint requests should be addressed at Department of Surgery, University of California, Irvine, 333 City Blvd. West, Suite 700, Orange, CA 92868. E-mail: dhoyt@uci.edu. Journal of Surgical Research 143, 276 –280 (2007) doi:10.1016/j.jss.2007.01.031 276 0022-4804/07 $32.00 © 2007 Elsevier Inc. All rights reserved.