Early Difference in Apoptosis of Intestinal Mucosa of Rats with Severe Uncontrolled Hemorrhagic Shock after Three Fluid Resuscitation Methods
Lu Yuan-qiang,Cai Xiu-jun,Gu Lin-hui,Wang Qi,Huang Wei-dong,Bao De-guo
DOI: https://doi.org/10.1097/00029330-200605020-00010
2006-01-01
Abstract:Hemorrhagic shock is a common clinical emergency case. Successful treatment is usually accomplished by surgical control of hemorrhage and restoration of tissue perfusion. The fluid resuscitation method in the presurgical care of the hypotensive trauma patients is controversial. Current guidelines for presurgical treatment of patients with hemorrhagic shock recommend rapid volume resuscitation to normal blood pressure as quickly as possible. The practice is controversial because aggressive restoration of intravascular volume and rapid increase of blood pressure before securing hemostasis may exacerbate hemorrhage and worsen outcome. Controlled fluid resuscitation allows pre-hospital treatment to work with compensatory mechanisms. The concept is to restore some intravascular fluid while respecting hemostatic mechanisms. Such a scheme would balance the mutually exclusive processes of tissue perfusion and hemostasis. In recent years, some studies have indicated that intestinal mucosa is the first-affected and rapidly-changed site when shock or other low-flow conditions occur. Consequently, intestinal tract was defined as “the canary of the body” by Dantzker.1 In this study we used a model of rat with severe hemorrhagic shock and active bleeding to observe the early effects of three resuscitation methods on apoptosis of intestinal mucosa. METHODS Animals Forty-two male Sprague-Dawley (SD) rats, weighing 270 g to 460 g, were obtained from the Medical Institute of Zhejiang Province, China. Animal model This study was approved by the Ethics Committee of Sir Run Run Shaw Hospital Affiliated to Medical College, Zhejiang University. The rats had unlimited access to food and water before the experiments. After being weighed, the rats were anesthetized with pentobarbital (40 mg/kg intraperitoneally) and were placed in a supine position on a warming pad (25°C). After povidone-iodine solution was used, the right carotid artery was isolated and cannulated with polyethylene catheter through a neck incision. The arterial catheter was used for blood withdrawal and was connected to a pressure transducer and computerized physiograph system for continuous hemodynamic monitoring. In the same way, the left femoral vein was cannulated for fluid infusion. The animals were heparinized (500 U/kg). Blood losses of the procedure were measured by mopping all blood from the incision with preweighed gauze sponges, which were then reweighed. A transformation formula of 1 g=0.9 ml of blood was used. Only animals whose blood losses were lower than 0.2 ml during the above-mentioned procedure and that were spontaneously breathing 10 minutes after the procedure were included in the study. Among the test rats, thirty-two fulfilled the inclusion criteria. According to the method created by Capone,2 the model of rat with severe hemorrhagic shock and active bleeding was established. Under light anesthesia, the injury began (time = zero) with blood withdrawal through the carotid arterial cannula for four times (at a rate of 1 ml per 100 g per 5 minutes in the first two times, 0.5 ml per 100 g per 5 minutes in the late two times). The shed blood was collected in glass syringes with heparin and reinfused during resuscitation. At 30 minnutes, uncontrolled hemorrhagic shock was added to the volume-controlled shock by amputation of the tail at 75 percent of its length, measured from the tip. The bleeding tail was immediately directed into a container (with heparin) and the amount of shed blood was measured. This phase was called “pre-hospital phase” and continued for over 60 minutes. During this period, the rats were early resuscitated by infusing normal saline (NS). Fluids were administered via the femoral vein with an infusion pump at the rate of 2 ml·kg-1·min-1. The pump was turned on and off to maintain the mean arterial pressure (MAP) goal. At 90 minutes, a phrase simulating hospital treatment (hospital phrase) began. Hemostasis was achieved by tail wound closure. Simultaneously, resuscitation began with infusion of blood and normal saline solution. The “hospital phrase” lasted 60 minutes and the end points were hematocrit (Hct) of 30 percent and MAP of 80 mmHg. Animals grouping Thirty-two rats were randomly divided into four groups of eight rats each with the sequence of the experiments randomized in blocks of four (one from each group): group 1 (control group), neither fluid resuscitation nor hemostasis in the “pre-hospital” or “hospital phases” (no treatment); group 2 (no fluid resuscitation group, NF group), no fluid resuscitation in the “pre-hospital phase” (no field fluid resuscitation); group 3 (controlled fluid resuscitation group, NS40 group), NS infusion during the “pre-hospital phase” to reach and maintain MAP at 40 mmHg, beginning immediately after the tail cut (field fluid resuscitation to MAP 40 mmHg); and group 4 (aggressive fluid resuscitation group, NS80 group), NS infusion during the “pre-hospital phase” beginning immediately after the tail cut, to reach and sustain MAP of 80 mmHg (field fluid resuscitation to MAP 80 mmHg). Groups 2, 3 and 4 had a “hospital phase” with the same end points: control of bleeding, fluid resuscitation with blood and NS to Hct of 30 percent and MAP of 80 mmHg. Collection and detection of samples Blood samples (0.3 ml/sample) were taken separately from rats for complete blood count at 0, 120 and 150 minutes, and blood samples (0.2 ml/sample) were collected separately for determining serum lactate level at 0, 30, 60 and 90 minutes. Rats living for 150 minutes were regarded as survivors. The surviving rats were immediately sacrificed after resuscitation and hemostasis. The small intestine was taken out quickly and flushed with 0.01 mol/L cold phosphate buffer solution (pH 7.4). A part of small intestine was fixed with 10% buffered formaldehyde for terminal deoxynucleotidyl transferase dUTP nick end labelling (TUNEL). All histological specimens were examined by a pathologist who was blinded to the animals' resuscitation protocol. The TUNEL detection kit was purchased from Promega Corp., USA. The detection procedure was mainly done according to the instructions provided by the corporation. Under high power field (400 × magnification) of fluorescent microscope, the number of apoptotic cells characterized by the positive staining nucleus in TUNEL staining sections was also counted in 10 randomly selected fields per section. The mean number of apoptotic cells per one field for each rat was calculated for further statistical analysis. The mucosa was scraped from the other part of small intestine, and was immediately sent to the central laboratory of Sir Run Run Shaw Hospital for apoptosis assaying by flow cytometer (FCM, Beckman Coulter EPICS XL, USA), and the method of Annexin-V/PI staining was used. The Annexin-V/PI staining kit was purchased from Bender MedSystems Corp., Austrian. A total of 5000 cells were measured in each sample. Statistical analysis Data were expressed as mean±standard deviation (SD). With SPSS 10.0 software package, statistical analysis for comparing mean values from the four groups was made by one way analysis of variance (ANOVA) and least significant difference t test (LSD-t). Survival rate was compared using Fisher's exact test. A P value less than 0.05 was considered statistically significant. RESULTS Characteristics of the animal model The rats of the severe uncontrolled hemorrhagic shock model suffered approximately 50% blood volume loss based on animal body weight. Eight rats without treatment in control group all survived at 30 minutes, four rats survived at 90 minutes, no one survived at 150 minutes. ANOVA showed that there were no statistically significant differences in weight, Hct, (PLT) platelet and hemoglobin among the rats of the four groups. Variation of MAP Pre-hemorrhage MAP was (114 ± 12) mmHg for control group, (124±9) mmHg for NF group, (128± 10) mmHg for NS40 group, and (129±7)mmHg for NS80 group. At this point, MAP in NS40 or NS80 group was higher than that in control group (P<0.05 ), but there was no significant difference among NF, NS40, and NS80 groups. At the conclusion of blood withdrawal (at 20 minutes), MAP decreased to (24±3) mmHg in control group, (21±3) mmHg in NF group, (22±2) mmHg in NS40 group, and (21±2) mmHg in NS80 group. Thirty minutes before the tail cut, MAP slightly increased to (33±6) mmHg in control group, (30±4) mmHg in NF group, (31±4) mmHg in NS40 group, and (30± 2) mmHg in NS80 group. There was no statistically significant difference in MAP among the four groups at 20 and 30 minutes. Fluid volumes administered The volume of normal saline given during the “pre-hospital phase” was (3.86±1.12) ml per 100 g in NS40 group and (13.95±3.46) ml per 100 g in NS80 group (P<0.01 for NS80 group compared with NS40 group). Control and NF groups did not receive fluid in the “pre-hospital phase”. Survival rate All rats survived in the first 30 minutes of the experiment. When the “hospital phase” began (at 90 minutes), the number of survivors was four in control group, three in NF group, eight in NS40 group, and six in NS80 group. At the conclusion of the “hospital phase” (at 150 minutes), the number of survivors was zero in control group, three in NF group, eight in NS40 group, and six in NS80 group. Compared with NF group (3/8), the higher survival rate of NS40 and NS80 groups (14/16) showed significant difference (P<0.05). The survival rate of NS40 group (8/8) was significantly higher than that of NF group (3/8), with the comparison showing significant difference (P<0.05). The survival rate of NS40 group (8/8) was slightly higher than that of NS80 group (6/8), the survival rate of NS80 group (6/8) was slightly higher than NF group (3/8), with the comparison showing no statistically significant difference. Serum lactate level There was no statistically significant difference in serum lactate level at 0 and 30 minutes of the “pre-hospital phase”. After fluid resuscitation, serum lactate levels in the NS40 and NS80 groups obviously decreased. Compared with control and NF groups, there was statistically significant difference in serum lactate level at 60 and 90 minutes of the “pre-hospital phase”. The dynamic changes of serum lactate level are listed in Table 1.Table 1: Serum lactate level changes of four groups rats in the “pre-hospital phase” (mmol/L)TUNEL staining In TUNEL stained sections of the intestinal mucosae of NF, NS40 and NS80 groups, some positive staining cells (apoptotic cells) were observed. Fluorescent microscopy revealed that TUNEL positive cells presented green fluorescence and contracted or fragmented nuclei. Occasionally, apoptotic bodies could be found (Fig.).Fig.: Small intestinal mucosa tissue, apoptosis detection (TUNEL, original magnification×400)The TUNEL staining results in the intestinal mucosae of NF, NS40 and NS80 groups are shown in Table 2. The number of TUNEL positive cells was obviously more in NS80 group than that in NF or NS40 group, with the differences showing statistical significance (P<0.01). However, between NS40 group and NF group, there was no significant difference in the number of TUNEL positive cells.Table 2: Results of apoptosis in the intestinal mucosae of rats in 3 groupsFlow cytometric analysis Table 2 particularly shows the FCM results of apoptosis in the intestinal mucosae of NF, NS40 and NS80 groups. The apoptotic rate was obviously higher in NS80 group than that in NF or NS40 group, and the comparison showed statistically significant difference (P<0.01). In addition, the apoptotic rate in NS40 group was higher than that of NF group (P<0.05). DISCUSSION Conventional guidelines for pre-hospital treatment of hypotension secondary to hemorrhage after trauma recommend rapid infusion of crystalloid solution to restore normal blood pressure as quickly as possible. This premise is based, in part, on clinical studies and on substantial laboratory data that showed hemorrhagic shock in animals of a controlled hemorrhage model was reversible when shed blood was replaced for three to four times that volume of crystalloid solution. Although controlled hemorrhage is a well-defined laboratory model, resuscitation of the patients with multiple injuries and active bleeding may present very different pathophysiology. Sometimes, increased blood pressure from rapid fluid resuscitation can lead to a worsening outcome by disruption of early soft thrombus, coagulopathy, hemodilution and rebleeding. The body is good at compensating for hemorrhagic shock. Left undisturbed, it will decrease blood loss and maintain organ perfusion to some extent through a series of stress reactions. However, compensatory mechanisms are limited in magnitude and duration. Profound initial blood losses or conditions of prolonged transport to the operating room can overwhelm compensation and result in death. On the basis of the above-mentioned theories, some scholars advocate the concept of “controlled fluid resuscitation”, which only administers moderate fluid infusion to prolong the compensatory time of patients with traumatic hemorrhagic shock before surgical hemostasis.2-4 Controlled fluid resuscitation means restoring some intravascular volume while hemostatic mechanisms were taken into consideration, which allows pre-hospital treatment to work with compensatory mechanisms, balances the seemingly mutually exclusive processes of tissue perfusion and hemostasis. The model of blood withdrawal combined with active bleeding by tail cut produced severe hemorrhagic shock, in which initial blood loss was about 50 percent of the circulation blood volume. Early fluid infusion could obviously improve short-term survival rate of rats. In addition to increasing survival rate, fluid resuscitation partly improved tissue perfusion and metabolic parameter as indicated by serum lactate level in our study. The main causes of death in the late stage of patients with severe trauma or hemorrhagic shock include infection, multiple organ dysfunction syndrome (MODS) or multiple organ failure (MOF). Recently, several researches have found that apoptosis is obviously induced in visceral organs in the early stage of polytrauma combined with shock, which may play a role in early organ injury and later multiple organ failure.5-7 The apoptosis in pathological conditions can also reflect the severity of disordered internal environment of the body. In particular, significant apoptosis of intestinal mucosa can result in impairment of mucosa barrier and immune function, which is bound up with endogenous infection and MODS. Our study showed clearly that a great deal apoptosis of intestinal mucosa of the rat survivors happened in the early period of hemorrhagic shock and resuscitation. Hotchkiss and associates also demonstrated that the apoptosis of intestinal epithelial and lymphoid tissues occured extremely rapidly after injury and shock.8 Apoptotic loss of intestinal epithelial cells may compromise bowel wall integrity and be a mechanism for bacterial or endotoxin translocation into the systemic circulation. Apoptosis of lymphocytes may impair immunologic defenses and predispose to infection. More recently, Diebel and associates revealed the pivotal role of tumor necrosis factor-alpha in signaling apoptosis in intestinal epithelial cells under shock conditions by Caco2 intestinal cell monolayers.9 Furthermore, the apoptosis of intestinal mucosa was statistically higher in NS80 group than that in NF or NS40 group. This implies that, in pre-hospital treatment of severe and uncontrolled hemorrhagic shock, brief and moderate hypotension may be helpful for maintaining organ function and decreasing the later complications of traumatic patients. In recent years, a few clinical observations also supported the same opinions.10-12 The inherent mechanism is not completely clear, but may be related to the following factors: ① aggressive fluid infusion seriously disturbs the internal environment and exacerbates cellular metabolism, which can induce abnormal apoptosis of visceral organs; ② high perfusion pressure of tissue and organ followed with aggressive fluid resuscitation can cause severe ischemia-reperfusion injury by inducing a large amount of oxygen-derived free radicals. Several laboratory studies have shown that ischemia-reperfusion injury can induce apoptosis.13-15 In addition, the apoptosis of small intestinal mucosa in NS40 group was higher than that of NF group, probably because controlled fluid resuscitation can still induce ischemia-reperfusion injury to some extent. In conclusion, our results showed that in severe uncontrolled hemorrhagic shock, some fluid must be given at proper time for improving tissue perfusion and avoid early death. The data also confirmed that aggressive fluid resuscitation to restore MAP of 80 mmHg during uncontrolled hemorrhage obviously increased the apoptosis of intestinal mucosa in rats. Controlled fluid resuscitation to maintain MAP of 40 mmHg in the treatment of severe uncontrolled hemorrhagic shock can obviously improve the early survival rate and reduce the apoptosis of intestinal mucosa, which may benefit improvement of prognosis.