Kharkov National Medical University,

Kharkov, Ukraine

INTRODUCTION

In the beginning of the XXI century trauma takes the fourth place as mortality cause among all age categories, but for individuals at the age before 50 this cause takes the first place [3]. About 30 % of patients with concomitant injury have severe chest injuries [10]. After traumatic brain injury thoracic injury is the most frequent cause of lethal outcomes after accidents [11].

Thoracic injury pathogenesis includes all possible initiating factors for development of oxidative stress: hypoxia after ventilation disorders and direct injury to the lung parenchyma [7]; ischemia/reperfusion phenomenon associated with decreased cardiac output after intrathoracic pressure increase; blood loss, which can reach 1-1.5 L, even in case of isolated chest injury [8]. The most dramatic oxidative stresses take a place during the first 1-2 days after trauma, in so called shock period of traumatic disease [4, 12], because at this time infusion-transfusion therapy for circulating blood volume, decompression of pleural cavities, oxygen therapy and hardware support for respiratory function are performed. Often such patients require adrenoceptive support for sufficient level of arterial pressure. Early postsurgical period makes its own contribution to general volume of traumatic injuries (conception of two hits) [9]. All the above suggests that oxidative stress markers should show severity of homeostasis disorder in patients with severe thoracic injury in shock period of traumatic disease. However, there are no clear criteria among the indices characterizing oxidative stress and assisting evaluation of traumatic shock severity and functional state in patients with severe concomitant thoracic injury.

 

Objective – to search the evaluation criteria for severity of state of patients in shock period of severe concomitant thoracic injury among the markers of oxidative injury to lipids and proteins, which could predict outcome.

 

MATERIALS AND METHODS

There was an examination of 73 male patients at the age of 20-68, who had severe concomitant chest injury and received treatment in the department of anesthesiology and resuscitation for patients with concomitant injury in Kharkov City Clinical Hospital of Emergency Medical Aid by the name of O.I. Meshchaninov. The study included patients with severe closed concomitant chest injury with intrathoracic afflux, lung and heart contusions, skeletal injuries. All patients received intensive infusion-transfusion analgesic, anti-inflammatory, antibacterial and metabolic therapy, prevention of gastrointestinal stress ulcers according to severity of injuries and state of the patients. The study was performed at days 1-2 (3-33.5 hours) after trauma. The controls included 15 healthy male volunteers of similar age.

The levels of malondialdehyde in the blood plasma were evaluated with TBA-activity of deproteinized plasma [6], and the levels of protein carbonyl groups were evaluated using reaction with dinitrophenylhydrazine extracted from protein plasma [2]. For correction for hemodilution at the background of significant volume of infusion-transfusion therapy the levels of markers of oxidative injury to biomolecules were divided by total protein concentration in blood plasma, which was defined with biuret method [5]. Severity of injuries of individual anatomic locations of the body was evaluated with Abbreviated Injury Score (AIS), concomitant injury – with Injury Severity Score (ISS). Revised Trauma Score (RTS) [1] was used for calculation of severity of patients’ state on admission.

For stratification of the patients the cluster analysis was used with STATGRAPHICS Plus 5.0. Square of Euclidean distance was used as distance measure between the objects. Cluster division was performed with unweighed centroid method. With a dendrogram the amount of clusters for population distribution was defined. The statistical analysis was performed with GraphPad Prism 5.03. Kruskal-Wallis test was used for assessment of statistical significance between the groups.

The analysis of statistical confidence of the contingency tables was performed with χ² test. For cut-off values between the groups ROC-analysis was performed. P < 0.005 was statistically significant. The patients and the volunteers were included into the study after their consent according to the principles of Helsinki declaration of World Association of Physicians “Ethic principles of conduction of scientific and medical studies with human participation”, with amendments from 2000 and “The order of conduction of clinical trials of medical drugs and expertise of materials of clinical studies and Ordinary position of ethics commission”, confirmed by the Order of the Ministry of Health of the Ukraine from 23.09.2009, #690. The study protocol was approved by the ethics and bioethics committee, Kharkov National Medical University, #5, 17.05.2011.

 

RESULTS AND DISCUSSION  

Considering the fact that oxidative processes of lipids and proteins do not go separately, but are the parts of one process – oxidative stress, we decided to use cluster analysis for distribution of the patients behind both indices simultaneously.

The figure 1 shows the results of cluster analysis with survival rate.  

The population was distributed into 5 clusters in concordance with distribution. The clusters were combined into 3 groups in dependence on survival level. The first group included the patients of A cluster, the second one – the patients of B and C clusters. The third group included the patients of D and E clusters. It is interesting that the patients of D cluster were characterized with the highest levels of indices of oxidative injury to both proteins and lipids, but the patients of E cluster had the lowest indices in the population; however, lethality was 100 % in both clusters (Table 1). These data can be explained with the fact that decrease in oxidative processes below the normal values is unfavorable in predictive relation, because it testifies severe disorders in oxygen consumption processes [8].

 The statistically significant difference in mortality between the groups was observed (Table 1) that testifies a possibility of predicting outcomes of traumatic diseases at 1-2 days after trauma. These criteria are able to reflect severity of blood loss (traumatic shock), as one can see with the presence of statistically significant difference between the groups of the patients who needed hemotransfusion therapy. The level of malondialdehyde is less specific value for differentiation of patients between the groups I and II. Possibly, it is related to the fact that malondialdehyde is able to overcome renal barrier and come out of the blood, especially during intensive infusion therapy. Also it is possible that lipid peroxidation processes do not occur proportionally to the processes of oxidative modification of proteins.

Table 1

Characteristics of patient groups

Note: * χ2 test, ** Kruskal-Wallis test, R – reliable, NR – not reliable, ALV – artificial lung ventilation, M – mean, SE – standard  error.

error.

For estimation of cut-off values between the groups ROC-analysis was performed. Considering the absence of statistically significant difference in malondialdehyde levels between the groups I and II (Table 1), we combined two groups (A. B, C clusters) into one. The results are given in the table 2.

Table 2
Separation points between groups of patients

As result, the cut-off values became statistically significant, with high sensitivity and specificity, and it confirms their diagnostic importance. Therefore, according to the statistical data it is possible to suppose the following criteria for evaluation of oxidative processes of proteins and lipids, based on mortality level (Table 3).

Table 3
Criteria for severity of oxidative stress during shock period of severe concomitant thoracic injury

The patients with possible level of mortality of 20 % and 50 % are characterized with the same relative concentration of malondialdehyde – 0.102-0.1448 mcM/g of protein. Differentiation between the groups is possible by the means of evaluation of relative levels of protein carbonyl groups – the group with 20 % mortality prediction is characterized with the level of carbonyl groups 14.14-17.46 mcM/g of protein, the group of 50 % - 10.77-14.14 mcM/g of protein. In case of cluster analysis with using relative levels of malondialdehyde or carbonyl groups of proteins alone, we received neither equivalent nor similar dendrograms compared to that one which was created using both markers of oxidative stress simultaneously. It testifies the impossibility of evaluation of oxidative processes with one index, because free radicals, as one knows, affect all classes of biomolecules by means of some intermediate stages and chemical reactions, which are exclusive for each of them [5].

The normal values of oxidative injury to biomolecules, which were obtained with the examination of the volunteers of the control groups, are 0.1005 ± 0.003 mcM/g of protein for malondialdehyde and 11.998 ± 0.5793 mcM/g of protein for protein carbonyl groups. In case of traumatic disease with severe concomitant chest injury the similar values are common for the patients’ group with probable lethal outcomes of 50 % and cannot described as satisfactory. The good predictive criteria are slightly increased (compared to the normal values) levels of malondialdehyde and protein carbonyl groups at days 1-2 after trauma.

Considering the obtained statistical data from the table 1, these criteria could be reviewed as additional diagnostic criteria for severity of state of patients with severe thoracic injury in shock period of traumatic disease.

 

CONCLUSION:

1. Prediction of lethal outcome of traumatic disease in severe concomitant chest injury is possible at days 1-2 after trauma based on evaluation of intensity of oxidative processes using relative levels of malondialdehyde and protein carbonyl groups.

2. A favorable outcome of traumatic disease (expected mortality 20 %) in patients with malondialdehyde level from 0.1023 to 0.1448 mcM/g of protein is expected in case of combination with the levels of protein carbonyl groups from 14.14 to 17.46 mcM/g of protein, and uncertain outcome (expected mortality 50 %) – from 10.77 to 14.14 mcM/g of protein.

3. Very high and very low levels of oxidative injury to proteins (< 10.77 or > 17.46 mcM/g of protein) and lipids (< 0.1023 or > 0.1448 mcM/g of protein) are statistically significant predictive signs of unfavorable outcome of traumatic disease.

4. These criteria are able to reliably predict probability of lethal outcome and do not depend on patients’ age, severity of injuries to individual anatomic regions and duration of hospital period.