PREDICTIVE SIGNIFICANCE OF APOLIPOPROTEINS A1 AND B (apoA1 and apoB) IN DEVELOPMENT OF SEPSIS IN PATIENTS WITH POLYTRAUMA
Regional Clinical Center of Miners’ Health Protection,
Leninsk-Kuznetsky, Russia
One of the main causes of mortality in patients with polytrauma including shock and blood loss is infectious complications [1, 2]. Among the patients examined by us, only 28 % of the cases were without complications; acute respiratory distress syndrome (ARDS) was registered in 15.2 %. Traumatic disease was accompanied by various purulent and septic complications in 44.8 % with the following common complications: pneumonia (14.3 %), peritonitis (6.5 %), wound purulence (18.8 %), various types of necrosis (6.7 %), osteomyelitis (14.6 %), sepsis (3.9 %) [3].
The factors promoting the development of complications in critically ill patients are blood circulation disorders, limiting organ functions, severe metabolic changes, simultaneous appearance of big amount of own injured tissues initiating the inflammatory response; insufficient immune protection [4]. These factors are used for estimation of risk of posttraumatic mortality, but they do not clarify the mechanisms forming the basis of infection in posttraumatic period of polytrauma [5]. The indicated disorders underlie the multiple organ dysfunction syndrome (MODS) determining development of severe complications.
The consensus conference of the American College of Chest Physicians and the Society of Critical Care Medicine (ACCP/SCCM) recommended the use of the term multiple organ dysfunction syndrome – the clinical syndrome, which is characterized by development of acute potentially invertible dysfunction of organs of organ systems which are not directly involved in primary pathological process [6].
One should note that the concept of development of MODS relies on the data of the regularities of development of systemic inflammatory response syndrome (SIRS) [3, 7].
In 2012 during the revision of the terms at the conference of Surviving Sepsis Campaign (SSC) the list (Bone R.C. et al. [8]) of the signs, the symptoms and the laboratory values (C-reactive protein, procalcitonin, glucose, lactate) indicating the possibility of sepsis was supplemented [9]. The improved (the third) edition of the clinical recommendations from Surviving Sepsis Campaign (SSC12) was published in February 2013 [9].
However the offered biomarkers of sepsis are non-specific, i.e. their use is difficult for early diagnostics of sepsis in polytrauma that determines the interest to search other markers [10, 11].
Moreover, for decreasing the mortality and morbidity from posttraumatic infections and costs of treatment it is necessary to identify patients with high risk of infection and to develop the appropriate preventive measures.
At the present time no one doubts that lipoproteins and their protein components (apolipoproteins) play the own role in development of cardiovascular diseases [12, 13], promotion of immune protection from bacterial pathogens and decrease in immunological protective mechanisms of the body.
Apolipoproteins are the part of specific immunological protective mechanisms of the body against the wide range of microorganisms playing the significant role in posttraumatic infections [11]. Contois J.H., Warnick G.R., Sniderman A.D. (2012) [13] demonstrated the relationship between the decrease in circulating apolipoproteins in the blood and development of SIRS, severity of sepsis and infection. These observations showed the probable role of decreasing levels of apolipoproteins in reducing protective properties of the body after trauma [14].
However there are single studies estimating the changes in amount of circulating apolipoproteins after trauma and their role with risk of infection [11].
Thereafter, the actual issues are investigation of the pathogenetic relationship between systemic inflammatory response and estimation of clinical and predictive significance of circulating apolipoproteins in polytrauma.
The objective of the study – to estimate the clinical and predictive significance of blood levels of apolipoproteins A1 and B (ApoA1 and ApoB) in development of infection in patients with polytrauma.
MATERIALS AND METHODS
The clinical examination included 99 critically ill patients with polytrauma after road traffic accidents: 64 men and 35 women, the mean age of 25-55 (the table 1). The patients were admitted within 2 hours after injury. They had traumatic shock of degrees 2-3 (APACHE-III > 80), blood loss of 1,200-2,500 ml (20-50 % of circulating blood volume [CBV]). The individual estimation of blood loss was conducted according to the sum of external and cavitary blood loss with consideration of approximate blood loss after fractures. The inclusion criteria were the age of 16-65, presence of severe multiple or associated injuries, ISS > 30 (Injury Severity Score [15]), the volume of approximate blood loss > 20 % of CBV. The exclusion criteria were severe traumatic brain and/or abdominal injury, chronical diseases in acute phase. The control group included 15 almost healthy persons at the age of 20-25.
| Table 1. Characteristics of patients | |
| 42.2 ± 2.23 | |
| Mean age, years1 | |
| 64 / 35 | |
| Gender: male/female, abs. | |
| Types of injuries, abs. (%): | |
| Head | 2 (2.02 %) |
| Face | 5 (5.05 %) |
| Chest | 26 (26.3 %) |
| Abdomen | 9 (9.1 %) |
| Extremity | 46 (46.5 %) |
| External | 12 (11.03 %) |
| Injury severity: | |
| ISS, points1 | 46.3 ± 2.21 |
| Severity of state on admission: | |
| APACHE III, points1 | 80.9 ± 12.10 |
| SAPS II, points1 | 36.5 ± 17.10 |
| SOFA, points1 | 6.6 ± 0.44 |
| 14 (6.5-21.6) | |
| Length of stay in ICU, days2 | |
| Duration of artificial lung ventilation, days1 | 8.4 ± 0.91 |
| Notes: 1 – arithmetic mean ± error of the mean; | |
| 2 – median (interquartile range: 25%-75%) | |
| APACHE III – Acute Physiology and Chronic Health (Knaus W., 1985) | |
| SAPS II – New Simplified Acute Physiology Score (Le Gall J. R. et al., 1993; Lemeshow S., Saulnier F., 1994) | |
| SOFA – Sequential Organ Failure Assessment (Vincent JL et al., 1996) | |
| ISS – Injury Severity (Baker S.P., O’Neill B., Haddon W., Long W.B., 1974) | |
The study was conducted in compliance with the World Medical Association Declaration of Helsinki – Ethical Principles for Medical Research Involving Human Subjects 2013. The written consent from the patients and the approval from the local ethical committee were received.
The study group was realized with use of the microbiological and laboratory techniques on the days 1-3, 5-7, 8-10, 11-14, 17-21 after admission to the intensive care unit. The findings of microbiological and clinical infections and use of antibiotics were recorded each day.
Bacterial contamination was estimated with inoculation of various biomaterials (blood, urine, sputum etc.) according to the actual order No.535, Health Ministry of USSR, April 22, 1985. Microorganisms were identified with the bacteriological analyzer Vitek 2 (bioMerieux, France). Apolipoprotein A1 (ApoA1) and apolipoprotein B (ApoB) in the blood serum were measured with the analytic module platform Cobas 6000 SWA (Roche Diagnostics, Germany) with the immunoturbidimetric technique and the reagents Roche Diagnostics (Germany).
By the end of observation (21st day) all patients were retrospectively distributed into the groups according to maximal manifestations of SIRS, with use of the criteria from the consensus conference of ACCP/SCCM [9]: SIRS (n = 180), local infection (n = 36), sepsis (n = 27), severe sepsis (n = 12), septic shock (n = 6). The classification was retrospectively conducted by two physicians who did not participate in treatment. The diagnosis SIRS was concluded if more than one symptoms were identified: body temperature < 36° or > 38°, respiratory rate > 20 per minute or PaCO2 < 32 mm Hg, heart rate > 90 per minute, leukocyte count < 4,000/ml or > 12,000/ml or more than 10 % of neocytes. A case was considered as infection after confirmation of infection source and microbiological confirmation, as well as after identification of microorganisms in sterile tissues. Sepsis was diagnosed on the basis of a suspected or confirmed infection accompanied by at least two SIRS criteria; severe sepsis – after identification of the signs of acute dysfunction of one or multiple organ systems. Severe sepsis with signs of tissue or organ hypoperfusion and arterial hypotony (non-efficient correction with infusion therapy) was classified as septic shock.
The statistical analysis of the data was conducted with IBM SPSS Statistics 20. The quantitative variables were presented as Ìå (LQ-UQ), with Ìå – median, (LQ-UQ) – interquartile range (LQ – 25 %, UQ – 75 % quartiles). The qualitative signs were presented as absolute and relative (%) values. Fisher’s exact test and χ2-test were used for comparison of the qualitative values. Identification of relationships was realized with multiple logistic regression. Dunnett’s procedure of multiple comparisons was conducted after identification of differences (the comparison group was SIRS (the first case) or the control findings (the second case)). ROC-curve was analyzed for estimating the diagnostic information capacity of the tests (ApoA1 and ApoB). P < 0.05 was statistically significant.
RESULTS AND DISCUSSION
Two or three SIRS criteria (increasing respiratory and cardiac rate, leukocytosis) were identified at the moment of admission or during 3 days according to the analysis of intensity of clinical manifestations of systemic inflammatory response syndrome in 83.8 % of the patients with polytrauma. 15.2 % of the patients demonstrated the signs of acute respiratory distress syndrome, but any clinical signs of infection were absent. The bacteriological examination of various biomaterials did not identify the increase in microflora. By the moment of the days 5-7 81 patients demonstrated the signs of infection overlay in view of inflammation of the urinary tract (n = 22), respiratory organs (bronchitis, pneumonia, n = 50), wound purulence (n = 9), with identification of microflora in the diagnostically significant titer. Microorganisms were identified as monocultures (34 %) and as a part of associations (66 %). The patients with local infection and generalized infection process demonstrated gram-positive and gram-negative microflora with the equal rate of isolation (the table 2). On the days 8-10 the inoculation of various biomaterials did not identify the increase in microflora in the patients with local infection, whereas the patients with sepsis demonstrated 100 % rate of positive results with prevailing gram-negative bacteria. Severe course of sepsis in the ICU was characterized by gradual replacement of antibiotic-sensitive microflora to hospital multiple resistant microflora: gram-negative (Pseudomonas aeruginosa, Acinetobacter spp, Klebsiella ðneumoniae), with extended spectrum beta lactamases and gram-positive strains, in association with methicillin-resistant Staphylococcus aureus (MRSA). It corresponds to the data from other researchers [1, 11, 16].
| Table 2. The number of positive inoculations (abs.) in patients with polytrauma in different terms of observation and in dependance on severity of sepsis | ||||||||||||
| Separated microorganisms | ÑÑÂÎ SIRS | Local infection (n = 36) | Sepsis | Severe sepsis | Septic shock | |||||||
| (n = 18) | (n = 27) | (n = 12) | (n = 6) | |||||||||
| Days of follow-up | ||||||||||||
| 5-7 | 5-7 | 5-7 | 8-10 | 5-7 | 8-10 | 5-7 | 8-10 | |||||
| Enterococcus faecalis | - | 12 | - | - | - | - | ||||||
| Escherichia coli | - | 11 | 6 | 6 | 2 | 1 | - | - | ||||
| Streptococcus pyogenes | - | 8 | - | - | - | - | ||||||
| Streptococcus millerii | - | 2 | - | - | - | - | ||||||
| Staphylococcus åpidermidis | - | 15 | 6 | 6 | 2 | 2 | - | - | ||||
| Staphylococcus aureus | - | - | 6 | 6 | 3 | 3 | 2 | 3 | ||||
| Acinetobacter species | - | 6 | 6 | 12 | 2 | 5 | 1 | 2 | ||||
| Pseudomonas aeruginosa | - | - | - | 6 | 1 | 2 | 2 | 3 | ||||
| Klebsiella ðneumoniae | - | 9 | 3 | 3 | 2 | 2 | 1 | 2 | ||||
| Candida albicans | - | - | - | - | - | 1 | - | - | ||||
| Notes: Here and in Table 3: SIRS - systemic inflammatory response syndrome | ||||||||||||
On the days 17-21 the growth of microflora was noted only in the group of the patients with septic shock, in 100 % of the cases – microbial associations (P. aeruginosa and Acinetobacter spp, in 66.6 % of the cases in association with K. pneumoniae, in 33.4 % – with MRSA).
Moreover, the posttraumatic period was accompanied by infectious complications in 81 % of the critically ill patients with polytrauma; sepsis was diagnosed in 45 % on the days 8-10, with severe course because of gradual overlay of multiple resistant opportunistic gram-negative microflora (P. Aeruginosa and Acinetobacter spp., in 66.6 % of the cases in association with Kl. Pneumoniae, in 33.4 % – with MRSA).
The results of the examination of ApoA1 showed its decrease within 3 days of observation as compared to the values in the healthy persons: 116 (105-127 µg/ml). It supposes translocation of endogenous microflora after severe injury (the table 3). The level of ApoA1 only slightly changed during the following follow-up. The lower levels of ApoA1 were found in the patients of the septic groups on the days 5-7 of observation, but without the statistically significant differences as compared to the SIRS group (p = 0.053). It could be determined by massive translocation of endogenous microflora and overlay of secondary infection [3] and it probably reflects the excessive lipopolysaccharide-induced stimulation of bacteria of the cells of monocytic-macrophage link. Thereafter it could lead to disruption of the autoregulatory mechanisms and promotion of generalization of the infectious process [19].
The examination of ApoB identified the significant decrease in all groups in the patients with polytrauma as compared to the control values (p = 0.02; the table 3). The mean values of ApoB in the patients of the septic groups were lower than the lowest reference. The maximal decrease in ApoB was found in the blood of the patients of the septic groups on the days 1-3 and 5-7 (the table 3).
| Table 3. Dynamics of levels of A1/ (ApoA1) and B (ApoB) (mg/dl) in blood serum of patients with polytrauma | |||||||
| Days | Indices | ÑÑÂÎ SIRS | Local infection | Sepsis | Severe sepsis (n = 12) | Septic shock | Control group |
| (n = 18) | (n = 36) | (n = 27) | (n = 6) | (n = 15) | |||
| 1-3 | ApoA1 | 105 (92-127) | 107 (119-78) | 101 (96-148) | 103 (85-174) | 89 (88-99) *+ | 116 (105-127) |
| ApoB | 78 (73-82) | 67 (65-89) + * | 52 (57-75) +* | 48 (47-58) + * | 46 (39-54) +* | 82 (78-92) | |
| 5-7 | ApoA1 | 112 (90-120) | 96 (92-123)+ | 84 (57-75) *+ | 83 (81-108) *+ | 85 (87-121) *+ | |
| ApoB | 71 (73-89) | 50 (48-53)* + | 45 (47-61)*+ | 31 (29-65)*+ | 35 (33-50)*+ | ||
| 11-14 | ApoA1 | 95 (110-120) * | 105 (110-135) | 93 (91-125) * | 95 (87-110) * | 96 (83-20) * | |
| 17-21 | Apo | 81 (73-98) | 98 (87-120) | 38 (45-55) + * | 35 (43-58) + * | 27 (40-47) + * | |
| 17-21 | ApoA1 | 95 (33-40) * | 107 (33-39) | 103 (90-110) | 103 (103-107) | 108 (145-160)+ | |
| ApoB | 80 (75-85) | 87 (78-89) | 37 (45-60) + * | 30 (43-59) + * | 31 (45-62) + * | ||
| Note: Data are presented as median (25-75 percentiles). ApoA1 and ApoB - apolipoprotein A1 and B. | |||||||
| * - statistically significant differences compared to the control group; | |||||||
| + - statistically significant differences compared to SIRS group at p<0.001 | |||||||
The strong direct correlation relationship was found between the serum level of ApoB and severity of systemic inflammation syndrome (Spearman’s rank correlation coefficient ρ = 0.76, ð < 0.001). It shows a possibility for the use of ApoB for estimating the severity of SIRS in the patients with polytrauma. The relationship between apolipoproteins AI and the infection was not statistically significant with α = 0.05 (p = 0.051).
Because of the decrease in the serum levels of ApoB in the critically ill patients with polytrauma on the days 1-3 and the relationship between such decrease and severity of developing sepsis, the possibility for using this value as an early biomarker of sepsis was estimated. The ROC-curve was constructed for demonstrating the likelihood ratio for positive results of the test. The ROC-curve analysis showed the cutoff value of ApoB (the days 1-3) for early prediction of sepsis (43.7 mg/dl). The sufficiently high diagnostic efficiency of the test in relation to sepsis was found: the area under the ROC-curve was 0.72 (95 % CI: 1.1-3.6). The diagnostic sensitivity of the cutoff value of ApoB was 98 %, diagnostic specificity – 88 % (the table 4). The incidence of the diagnostic levels of ApoB was 100 % in the septic groups on the days 1-3 and 5-7, whereas the fact of contamination with gram-negative microflora was confirmed for 58 % of the patients on the days 5-7.
| Table 4. Estimation of predictive significance of apolipoproteins A1 (ApoA1) and B (ApoB) (mg/dl) in blood serum and relationships between inflammation values in patients with polytrauma | ||||||||
| ROC-curve | OR | CI 0.95 | P | Sensitivity | Specificity | Probability ratio + | Probability ratio - | |
| Àðî Â < 50 ìã/äë / mg/dl | 0.72 | 1.9 | 1.1-3.6 | 0.03 | 0.98 | 88 % | 2.9 | 0.41 |
| Àðî À < 100 ìã/äë / mg/dl | 0.69 | 3.65 | 1.2-9.82 | 0.0051 | 80 % | 47 % | 1.42 | 0.40 |
| SIRS criteria (3 and more) according to ACCP/SCCM -1992 | 0.816 | 12.03 | 5.13-28.20 | < 0.001 | 58 % | 90 % | 5.61 | 0.47 |
| Note: OR - odds ratio; CI - confidence interval. | ||||||||
DISCUSSION
Our study showed the significant decrease in the serum levels of ApoB in the patients of the septic groups (sepsis, severe sepsis, septic shock) as compared to the patients in the groups with SIRS and local infection. These findings showed the relationship between the low level of apolipoprotein B and the risk of infection after polytrauma.
There is an actual issue about the decrease in the blood level of apolipoproteins and its association with the reduction in the immunological protective mechanisms of the body or significant metabolic changes resulting in attenuation of the mechanisms of immunological responsiveness of the body.
Apolipoproteins play the significant role in the inborn immune response to bacterial infection, especially against Staphylococcus aureus and Escherichia coli. In our study Escherichia coli was a predominant gram-negative microorganism leading to an infection according to the causes of most nosocomial infections. The role of apolipoproteins AI in inborn immune protection against Escherichia coli includes the suppression of bacterial growth and clearance of endotoxin. Apolipoproteins AI are involved in complement activation and elimination of gram-negative pathogen [19]. Apolipoproteins AI and AII support the activity of paraoxonase, which play the role in destruction of regulation of agressin by means of Pseudomonas aeruginosa [20]. Also apolipoproteins B destructs the regulation of virulence of Staphylococcus aureus and suppress an invasive infection [21]. High density lipoproteins containing apolipoproteins AI destruct the toxic function of Staphylococcus aureus. Our observations show that the decrease in control of bacterial virulence in combination with defects of phagocytosis worsens the coagulation, and deactivation of monocytes creates the medium for disseminated infection.
Our study showed the maximal risk of development of posttraumatic hospital infections in the patients with polytrauma with the blood level of apolipoproteins B at least 50 % of the lowest reference.
Currently, time of estimation of the blood levels of apolipoproteins is more available, taking into account the modern laboratory techniques, it does not depend on calculation of complex multifactorial clinical estimation (such as APACHE II) and presents the value with clear normal range of values.
Moreover, it is possible to use the calculated value – ApoA1/ApoB index, which we previously offered for estimation of severity of patients’ condition in acute period of polytrauma and efficiency of medical procedures [17]. The combined estimation of apolipoproteins A1 and B in the blood serum will increase their predictive significance in estimation of condition severity and risk of infection in patients with polytrauma.
In our study the patients were admitted to the ICU, where despite of the preventive measures including a separate room for each patient, strict adherence to hand hygiene and preventive administration of antibiotics, we could observe the significant increase in the rate of infections with prolonged hospital stay and, as result, increasing costs of treatment. The high probability of infection, despite of optimal infection control, indicates that patients could be infected with bacteria, which are the part of normal flora. It is in agreement with the previous data, where Staphylococcus aureus and Escherichia coli are the primary nosocomial pathogens. If patients have some specific immune defects, which are the cause of increased rate of infections, it is necessary to use the additional measures for prevention of infections. Our findings offer an additional potential factor of posttraumatic immune suppression that should be considered during development and estimation of a protocol for decreasing the rate of posttraumatic nosocomial infections.
Therefore, ApoA1 and ApoB decrease in critically ill patients within 24 hours after polytrauma and the intensity of polytrauma is associated with severity of developing sepsis. Patients with polytrauma demonstrate the maximal decrease in the serum levels of ApoB on the days 1-3 and 5-7 after injury, whereas microbiological examinations of various biomaterials give the first positive results only on the days 5-7 in 58 % of patients with evident infections. The received results show the diagnostic and predictive significance of ApoB and the possible use as an early marker of infectious complications. The monitoring of ApoB allows estimating the severity of SIRS: the decrease in its level < 50 mg/dl show the generalization of an infectious process.
CONCLUSION
1. The decrease in the serum levels of apolipoprotein B (ApoB) in patients with polytrauma indicates the generalization of the inflammatory process and can be used as an additional diagnostic criterion of sepsis.
2. The high diagnostic sensitivity (100 %) of the posttraumatic (the days 1-3 and 5-7) cutoff level of ApoB (<50 mg/dl) allows its use as an early marker of septic complications determined by gram-negative microflora.