Procalcitonin levels in children with bloodstream infections caused by different species: a cohort study

Background: Timely diagnosis and accurate identification of the causative microorganism in sepsis is crucial in order to offer targeted treatment and increase survival rates. Previous studies have aimed to identify biomarkers that could potentially predict blood culture positivity in patients with bacteremia; however, most of the research has been performed in adult populations. The aims of this study were to analyze procalcitonin (PCT) levels in confirmed bloodstream infections by species in children and assess their utility in immunocompromised patients. Methods: Medical records of children younger than 18 years admitted from 2011 to 2018 were reviewed. Subjects who met the diagnostic criteria for sepsis, with PCT levels collected within a 72-hour period prior to obtaining a blood culture were included. Kruskal-Wallis test was used to compare differences among groups. Receiver-operating characteristic curves were used to evaluate PCT cut-offs. Results: A total of 120 patients were included. Mean age was 55 months. Mean PCT levels in immunosuppressed patients was 26.68 mcg/L, compared to 8.78 in the immunocompetent group. Subjects with bacteremia by Gram-negative bacilli (GNB) had the highest mean PCT levels (18.2 ± 34.2) (p < 0.001). Sensitivity and specificity were 78% and 53% for Gram-positive cocci (GPC), 60.9% and 33.3% for GNB, and 75% and 25% for yeasts, respectively. Subgroup analysis showed 87.5% sensitivity and 16.7% specificity of PCT for predicting documented GNB bacteremia in immunodeficient children. Conclusions: PCT may be considered as a surrogate biomarker in immunocompromised children, and a viable tool to differentiate etiology by species.


Background
Sepsis is a leading cause of morbidity and mortality in the pediatric population (1). Timely diagnosis and accurate identification of the causative microorganism is crucial in order to offer targeted treatment and increase survival rates. Previous studies have aimed to identify biomarkers that could potentially predict blood culture positivity in patients with bacteremia; however, most of the research has been performed in adult populations (1,2). Procalcitonin (PCT) is a 116-aminoacid prohormone synthetized and secreted by the thyroid C-cells (3)(4)(5). High levels of PCT are observed in critically ill infected patients (7)(8)(9)(10). It has been reported that PCT levels higher than 3.61 ng/mL predict blood culture-confirmed bacteremia in adults, with a 75% sensitivity and 72% specificity (9). Likewise, previous studies have described a high negative predictive value (95.4%) of normal levels of PCT for predicting bacteremia (10). Furthermore, it has been suggested that PCT levels could be helpful in differentiating among bacterial species in adults, with Gram-negative bacilli (GNB) causing a greater increase of PCT levels compared to Gram-positive cocci (GPC) (11). The objectives of this study were to analyze PCT levels in confirmed bloodstream infections by species in children and assess their utility in immunocompromised patients.

Methods
Medical records of all subjects admitted to the National Institute of Pediatrics (INP) in Mexico City from 2011 to 2018 were reviewed. Subjects younger than 18 years of age who met the diagnostic criteria for sepsis, with PCT levels collected within a 72-hour period prior to obtaining a blood culture were included in the analysis. Subjects with polymicrobial blood cultures and/or with isolation of commensal bacteria (coagulase-negative Staphylococci, Gram-positive bacilli, and Micrococcus spp) in a single peripheral blood culture were excluded. Blood cultures with commensal bacteria were only included in central line-associated bloodstream infections (CLABSI), and if the same microorganism was documented in two or more peripheral blood cultures. Samples were processed by the automated blood culture system BD BACTEC™. Bacterial identification and susceptibility testing were performed by BD Phoenix™ 100. PCT levels were obtained using the Thermo Scientific Fisher™ system, with a cut-off value of 0.5 mcg/L.

Definitions
Sepsis was defined as the presence of systemic inflammatory response syndrome (SIRS) in addition to a documented infection via blood culture. SIRS was defined according to the Society of Critical Care Medicine (4). Subjects who met two or more of the following criteria were diagnosed with SIRS: fever ≥ 38ºC, hypothermia < 36ºC, tachycardia > 90 beats per minute, tachypnea > 20 breaths per minute, hypocapnia PaCO 2 < 32 mmHg, and leukocytosis/leukopenia adjusted by age. In order to avoid overestimation of infectious episodes, a second infectious event in the same subject was defined as sepsis after a minimum of six days of hemodynamic stability without antibiotics. Patients with immunodeficiencies, including solid organ transplantation, hematopoietic stem cell transplantation (HSCT), primary immunodeficiency, solid tumors, nephrotic syndrome, Down's syndrome, severe malnutrition, hematologic immunodeficiency (i.e. leukemia, hemophagocytic lymphohistiocytosis), and drug-induced immunodeficiency were eligible for inclusion in the study.

Statistical analysis
For the statistical analysis, chi-squared test was used to analyze categorical variables. Kruskal-Wallis test was used to compare differences among groups. Receiver-operating characteristic (ROC) curves were used to evaluate PCT cut-offs. Statistical significance was assumed if the null hypothesis could be rejected at p < 0.05. Statistical analysis was performed using SPSS v.21 (IBM Corp., USA).

Demographic characteristics
A total of 41,836 blood cultures were obtained from January 2011 to April 2018, of which 5,059 cultures (12.09%) were positive. PCT levels were available for 311 subjects; 191 of these were excluded due to polymicrobial (n = 2), contaminated (n = 68), and duplicated (n = 121) blood cultures.
The final sample included 120 subjects with sepsis and documented PCT levels within a 72-hour period before blood culture collection (Fig. 1). Mean age was 55 months, 54% were male, and 44.2% had an immunodeficiency. Mean PCT level was 15.3 mcg/L.

PCT levels in immunocompromised children
Mean PCT level in immunodeficient children was 26.68 mcg/L, compared to 8.78 mcg/L in immunocompetent hosts (p < 0.05). PCT level distribution is shown in Table 1. The most frequent immunodeficiencies were hematologic (53.8%), primary immunodeficiency (9.6%), and solid tumors (9.6%). Subjects with hematologic immunodeficiency had the highest PCT mean level (31.4 mcg/L), followed by HSCT recipients (16.9 mcg/L); however, no statistical significance was observed.

Discussion
The current knowledge about PCT as a biomarker for sepsis in children has been described in previous studies (12)(13). Elevation of PCT levels usually occurs earlier during the course of infection, even before the elevation of other biomarkers, peaking at 24-36 hours (12). Pontrelli et al (13) showed a moderate accuracy for the diagnosis of sepsis in neonates with a PCT cut-off of 2.0-2.5 ng/mL (14). A 2015 meta-analysis showed that PCT is highly accurate in differentiating bacterial and viral meningitis in children with 96% sensitivity (12). PCT leves in blood culture positivity by different microorganism groups in children is scarce, especially in immunocompromised hosts.
Our results show a statistically significant difference between mean PCT values for each microorganism group. Mean PCT levels in children with GNB infections were significantly higher than those with GPC and fungal infections. These findings are consistent with previous studies performed in adults (14)(15)(16) Consistent with previous studies, we report lower PCT levels in fungal infections compared to bacterial events (26). It has been suggested that fungal infections could trigger an alternate inflammatory response route that does not involve PCT, explaining its modest rise.
Studies on PCT in immunocompromised patients are scarce (30)(31)(32)(33)(34). A recent systematic review and meta-analysis in children with chemotherapy-induced neutropenic fever showed that PCT levels > microbiologically defined infections (34). In our study according to mean PCT levels, we found a statistically significant difference between immunocompromised (26.68 mcg/L) and immunocompetent (8.78 mcg/L) children with sepsis (p < 0.05). We also report an 87.5% sensitivity of PCT for predicting blood culture-proven GNB infection, making PCT a useful resource in clinical practice. PCT levels were also increased in different types of immunosuppression.
Our study has several limitations. A prospective design would aid in having better control of the variables and include a larger sample. Likewise, PCT measurements were not serial, which would have allowed us to analyze PCT behavior in relation to variables such as time, isolated microorganism, treatment, and outcome.

Conclusions
Our study found that PCT could be a viable tool to predict blood-culture proven sepsis, particularly in