Early changes in white blood cell, C-reactive protein and procalcitonin levels in children with severe multiple trauma

Abstract

BACKGROUND:

To analyze early changes in white blood cells (WBCs), C-reactive protein (CRP) and procalcitonin (PCT) in children with multiple trauma, before secondary inflammation develops.

METHODS:

This single-center retrospective study collected data from patients with blunt traumatic injury admitted to the pediatric intensive care unit (PICU). According to the prognostic outcome of 28 d after admission to the PICU, patients were divided into survival group (n=141) and non-survival group (n=36). Characteristics between the two groups were compared. Receiver operation characteristic (ROC) curve analysis was conducted to evaluate the capacity of different biomarkers as predictors of mortality.

RESULTS:

The percentages of children with elevated WBC, CRP, and PCT levels were 81.36%, 31.07%, and 95.48%, respectively. Patients in the non-survival group presented a statistically significantly higher injury severity score (ISS) than those in the survival group: 37.17±16.11 vs. 22.23±11.24 (t=6.47, P<0.01). WBCs were also higher in non-survival group than in the survival group ([18.70±8.42]×10⁹/L vs. [15.89±6.98] ×10⁹/L, t=2.065, P=0.040). There was no significant difference between the survival and non-survival groups in PCT or CRP. The areas under the ROC curves of PCT, WBC and ISS for predicting 28-day mortality were 0.548 (P=0.376), 0.607 (P=0.047) and 0.799 (P<0.01), respectively.

CONCLUSIONS:

Secondary to multiple trauma, PCT levels increased in more patients, even if their WBC and CRP levels remained unchanged. However, early rising WBC and ISS were superior to PCT at predicting the mortality of multiple trauma patients in the PICU.

INTRODUCTION

Injury is one of the leading causes of childhood mortality, according to data published by the World Health Organization (WHO).[1] Motor vehicle crashes are the most common cause of injury-related deaths, followed by falls, both in developed countries and developing countries.[2-5] After injury, injured cells/tissues release an alarmin, damage-associated molecular patterns (DAMPs), causing a local pro-inflammatory reaction. The local production of alarmins could lead to systemic inflammation,[6,7] resulting in many bimolecular changes after injury, including white blood cell (WBC), interleukin 6 (IL-6), tumor necrosis factor α (TNF-α), fibrinogen (FIB), C-reactive protein (CRP) and procalcitonin (PCT).[8-12] Several previous studies have revealed that PCT rises faster than CRP and has a longer half-life (24 h) than IL-6 (14 h) and TNF-α (9 h).[10,13]

PCT is produced by thyroid C-cells, lung and intestinal neuroendocrine cells, with a very low level in the healthy individuals (<0.05 ng/mL).[14] The synthesis of PCT can be stimulated after the pathogen-associated molecular patterns (PAMPs) of lipopolysaccharide (LPS) from gram-negative bacteria identified by pattern recognition receptors (PRRs) on lymphocytes and antigen-presenting cells.[15] DAMPs, produced by damaged tissues, are similar to PAMPs. As a result, when trauma occurs, DAMPs from injured tissues can work as PAMPs to activate innate immune and inflammatory processes leading to elevated PCT levels.[6] There have been numerous studies on PCT in various types of traumatic injuries. According to some studies in adults, admission or baseline PCT was not meaningful to prognosis after trauma, and postponing PCT detection was recommended.[16] Others affirmed the early value of PCT in traumatic patients.[17,18] In addition, limited evidence is available in pediatric populations. We designed this study to describe the changes of PCT in pediatric patients with multiple trauma within 24 h of onset of trauma, analyze the association among PCT, CRP, ISS and GCS, and estimate the predictive ability of PCT. We assumed that post-traumatic changes in PCT could predict adverse outcomes, yet secondary inflammation was induced.

METHODS

Study design

This retrospective study analyzed the data from patients with injuries primarily due to motor vehicle accidents and falls who were admitted to the Pediatric Intensive Care Unit (PICU) of Shengjing Hospital of China Medical University, between January 2014 and December 2018. Ethical board approval and informed consent were obtained. All data were anonymised.

Inclusion criteria

Patients with the following were included: age between 1 month and 14 years old; injury mechanisms with traffic accident, falling, or crush without open injury; admission time within 24 h from initial injury.

Exclusion criteria

Patients who were post-surgery; patients died within 6 h after admission; patients with underlying diseases, including genetic metabolic diseases, hepatic disease, other chronic organ dysfunction or detected infectious disease.

Data collection

Age, gender, onset time, involved site(s), systemic manifestations, blood test, imaging examination, mechanical ventilation (MV), diagnostics and outcome were collected.

The main biomarkers including PCT, CRP, WBC, Neut%, platelets (PLT), and hemoglobin (HB) were tested simultaneously within 30 min after the patients were admitted to the PICU. The severity of injury was evaluated by the injury severity score (ISS). The injuries were classified as brain, face, chest, abdominal, limbs, and skin, according to the ISS. The Glasgow Coma Score (GCS) was used to evaluate the neurological status of children. ISS and GCS scores were assessed within 2 h after admission. WBC>10.0×10⁹/L, CRP>8.0 mg/L, and PCT>0.1 ng/mL were regarded as positive results. The diagnosis criteria of trauma-induced coagulopathy (TIC) were defined as activated partial thromboplasting time (APTT)>60 s, prothrombin time (PT)>18 s, or international normalized ratio (INR)>1.6.[19] According to the prognosis outcome on the 28^th day after admission to the PICU, patients were divided into the survival and non-survival groups. The primary endpoint of this research was the 28-day mortality of the children after trauma, and the secondary endpoints were other severe complications, including respiratory failure, shock, TIC and multiple organ dysfunction (MODS).

Statistical analysis

Discrete variables and continuous variables were expressed as counts (percentages) and means ± standard deviations (or medians and quartiles for abnormally distributed variables), respectively. To analyze the difference between two groups, the Student’s t-test or Mann-Whitney U-test was used for continuous variables, and the Chi-square was used for categorical variables. Receiver operation characteristic (ROC) curve analysis was conducted to evaluate the capacity of different biomarkers as predictors of mortality. All tests were two-tailed, and a P-value<0.05 was considered statistically significant. All data were statistically analyzed using SPSS Version 21.0 (IBM Corp., USA).

RESULTS

Demographics and clinical manifestations of all patients

During the four-year study period, a total of 284 pediatric patients with blunt trauma were admitted to our PICU, of whom 216 were admitted within 24 h of onset, meeting the inclusion criteria. There were 39 patients excluded from this study: 30 patients were post operation; 6 patients died within 6 h; and 3 patients had underlying diseases. Ultimately, 177 patients were enrolled in this study. A total of 118 children were injured in traffic accidents, 56 children were injured in falls, and 3 children sustained crush injuries. After 28 d of trauma, 36 patients died, with a mortality rate of 20.34%, among whom 21 patients died due to severe central nervous system injury. Eight patients died due to hemorrhoea, while 3 patients died of MODS. In addition, secondary infection caused 3 deaths and one patient died of a suspected pulmonary embolism. A total of 71 (40.11%) patients had severe brain injury with a low GCS score (≤8). Forty-four (24.86%) patients underwent respiratory failure, while 18 (10.17%) patients experienced shock. Besides, 22 (12.43%) patients were diagnosed with TIC (Table 1).

Table 1.

General characteristics of the 177 included pediatric patients

In the 177 patients, the percentages of children with increased WBC, CRP, and PCT were 81.36%, 31.07%, and 95.48%, respectively. The most common primary injury was head injury (146 cases, 82.49%), followed by chest (106 cases, 59.89%) and extremities (59 cases, 33.33%). The levels of PCT were determined in patients with different injured regions, significant differences were found between the groups with and without injured region in abdomen, chest, extremities and face regions. But the PCT levels were not significant difference between the head injury and non-head injury patients, as well as in the surface injury and non-surface injury patients (Table 2).

Table 2.

PCT in different injured regions

Characteristics of patients in the survival and non-survival groups

According to the prognostic outcome on the 28^th day after admission to the PICU, patients were divided into survival group (n=141) and non-survival group (n=36). As shown in Table 3, patients in the non-survival group presented a significantly higher ISS than those in the survival group: 37.17±16.11 vs. 22.23±11.24 (t=6.47, P<0.01). WBCs were also higher in the non-survival group than those in the survival group ([18.70±8.42]×10⁹/L vs. [15.89±6.98]×10⁹/L, t=2.065, P=0.040). The differences in PCT and CRP were not statistically significant between the survival and non-survival groups (P>0.05).

Table 3.

Comparison of general characteristics in survival group and non-survival group

The prediction of prognosis

ROC curves were used to predict mortality with ISS, WBC, CRP, and PCT. The areas under the ROC curves for CRP and PCT were 0.439 (P=0.262) and 0.548 (P=0.376), respectively. The areas under the ROC curves for WBC and ISS were 0.607 (P=0.047) and 0.799 (P<0.01), with statistically significant (Figure 1).

Figure 1.

Receiver operating characteristic (ROC) curves of ISS, WBC, CRP, and PCT for predicting 28-day mortality. WBC: white blood cell; CRP: C-reactive protein; PCT: procalcitonin; ISS: Injury Severity Score.

DISCUSSION

In our study, we described the early changes in PCT compared with other biomarkers, to predict major complications and outcomes in children with multiple traumas. The effects of PCT on predicting the complications of sepsis and MODS in adults following traumatic injury have been investigated in many studies,[10,20-25] but in multi-trauma pediatric patients, its role has not yet been established. Considering the post-traumatic infection, the onset time of all of the patients enrolled in this research was within the initial 24 h of injury, and open trauma patients were excluded, in accordance with a previous research.[23] After trauma occurred, physiological responses were triggered, increased PCT levels were observed in 95.48% of the patients, while only 31.07% of the children were found to have elevated CRP level. PCT was more sensitive than CRP on trauma patients, which was in agreement with other studies.[18,23] In this study, we analyzed the correlation between ISS and PCT. It showed a good relationship between ISS and PCT (r=0.423, P<0.01), which is consistent with previous studies, indicating that the more serious the trauma is, the higher the PCT level is.[18,26] However, PCT showed little difference in survival group and non-survival group. Furthermore, based on the ROC curve analysis, the predictive ability of PCT was not statistically significant.

There are several limitations in this study. First, the serum markers were measured at a single time point after trauma, so a prospective study with dynamic determination of PCT in pediatric traumatic patients should be carried out in the future. A previous research suggested that surgery, massive transfusion, and intensive care might influence the PCT concentration,[27] but in our study, we only excluded the post-surgery patients, and all the patients were under intensive care. A secondary peak of PCT was found in the infected group, but not the non-infected group,[28] which could be a good biomarker to predict outcome. However, the significance of the early PCT changes cannot be overlooked.

CONCLUSION

Secondary to multiple trauma, PCT levels increased in more patients, even if their WBC and CRP levels remained unchanged. However, early rising WBC and ISS were better than PCT at predicting the mortality of multiple trauma patients in the PICU.