Cytokine Levels in Critically Ill Children With Severe Sepsis and Their Relation With the Severity of Illness and Mortality

Abstract

Objective:

To study the baseline cytokine levels and their relation with the severity of illness and mortality in critically ill children with severe sepsis.

Design:

Subgroup analysis of a randomized, double-blind, placebo-controlled trial.

Setting:

Pediatric intensive care unit of a tertiary level teaching hospital in India.

Patients:

Fifty children with severe sepsis aged 3 months to 12 years.

Material and Methods:

Blood was collected at admission for estimation of pro-inflammatory (interleukin 6 [IL-6], IL-12p70, IL-17, and tumor necrotic factor α [TNF-α]) and anti-inflammatory (IL-10 and transforming growth factor β1 [TGF-β1]) cytokines.

Primary Outcome:

To find out correlation between cytokine levels and severity of illness scores (Pediatric Risk of Mortality [PRISM] III score, Sequential Organ Failure Assessment [SOFA], and Vasoactive-Inotropic Score [VIS]).

Secondary Outcomes:

To compare cytokine levels among survivors and nonsurvivors.

Results:

Baseline pro-inflammatory cytokine levels (median [interquartile range]) were IL-6: 189 (35-285) pg/mL, IL-12p: 48 (28-98) pg/mL, IL-17: 240 (133-345) pg/mL, and TNF-α: 296 (198-430) pg/mL; anti-inflammatory cytokine levels were IL-10: 185 (62-395) pg/mL and TGF-β1: 204 (92-290) ng/mL. Pro-inflammatory cytokines showed positive correlation with PRISM III score: IL-6 (Spearman correlation coefficient, ρ = 0.273, P = .06), IL-12 (ρ = 0.367, P = .01), IL-17 (ρ = 0.197, P = .17), and TNF-α (ρ = 0.284, P = .05), and anti-inflammatory cytokines showed negative correlation: IL-10 (ρ = −0.257, P = .09) and TGF-β (ρ = −0.238, P = .11). Both SOFA and VIS also showed weak positive correlation with IL-12 (ρ = 0.32, P = .03 and ρ = 0.31, P = .03, respectively). Among nonsurvivors (n = 5), the levels of all the measured pro-inflammatory cytokines were significantly higher as compared to survivors, IL-6: 359 (251-499) pg/mL versus 157 (97-223) pg/mL, P < .0001, IL-12p70: 167 (133-196) pg/mL versus 66 (30-100) pg/mL, P < .0001, IL-17: 400 (333-563) pg/mL versus 237 (122-318) pg/mL, P = .009, and TNF-α: 409 (355-503) pg/mL versus 330 (198-415) pg/mL, P = .002, respectively.

Conclusion:

In critically ill children with severe sepsis, pro-inflammatory cytokines (especially IL-12p70) showed a weak positive correlation with severity of illness and were significantly higher among nonsurvivors.

Keywords

cytokines severe sepsis critical illness children PRISM III

Introduction

Pediatric severe sepsis is characterized by high morbidity and mortality.^1

–4 With recent advances in treatment, the trends of case fatality rate (CFR) of pediatric severe sepsis and septic shock had declined over time, but the difference persists between developing and developed nations with pooled CFRs of 31.7% and 19.3%, respectively, according to a recent meta-analysis.⁵ In India, the mortality due to severe sepsis is around 60%.⁶

The severe sepsis is characterized by sequential cascade of hyperactive and dysregulated inflammatory cytokines, and outcome is linked to the extent of inflammatory response.^7

–12 Effective inflammatory response by pro-inflammatory cytokines (tumor necrotic factor α [TNF-α], interferon γ [IFN-γ], interleukin 1α [IL-1α], IL-1β, IL-6, IL-8, IL-12, and IL-17) is important for effective elimination of pathogens.^8,13,14 The anti-inflammatory cytokines (IL-4, IL-10, IL-13, and transforming growth factor β [TGF-β]) and cytokine inhibitors (soluble TNF-RI and II, soluble IL-1ra) are needed to counterbalance the inflammatory response.^8,13,14 On the other hand, the excess of inflammation leads to multiple organ dysfunction syndrome (MODS) and death, and excessive anti-inflammatory response can lead to immune paralysis and risk of infections.^8,13

–16 These cytokines start increasing within few hours of onset of sepsis, and the levels remain elevated for many days depending on baseline host factors, duration and severity of sepsis, treatment modalities used, timing of sampling in relation to the onset of sepsis, and interindividual variation.^14
–16

It has been demonstrated that in adults and children with severe sepsis, the heat shock protein 72 (HSP-72), IL-6, IL-8, IL-10, and TNF-α are elevated.^12,17,18 In adults with sepsis, it has been demonstrated that nonsurvivors had persistently high or increasing levels of cytokines (both pro- and anti-inflammatory), and survivors had decreasing levels.^{8,14,19

–25} Similar results have been noted in children with septic shock where low cytokine levels within 24 hours of admission were associated with high likelihood of survival.^12,26
–28

The ability of physicians to predict severity of illness and mortality at admission based on clinical parameters is sometimes limited. Moreover, there is heterogeneity in pediatric patient population in terms of different presentations and severity of the same disease, different response to same treatment, and different outcome.¹¹ The identification of pediatric patients with severe sepsis who are at higher risk of severe disease and mortality may be beneficial as far as specific treatment, anticipated outcome, and prognosis are concerned. The prediction of severity of illness and mortality by inflammatory parameters may allow for early anticipation about the prognosis and more aggressive personalized clinical care to optimize the organ functions and potential for better survival. Among critically ill children with sepsis, only a few studies from developed world assessed the role of cytokines in assessing the illness severity and prognostication.^12,26
–28 We therefore planned this study to demonstrate if pro-inflammatory and anti-inflammatory cytokine levels at admission to PICU were able to predict the severity of illness and mortality.

The rationale for using pro-inflammatory cytokines (IL-6, IL-12p70, IL-17, and TNF-α) and anti-inflammatory cytokines (IL-10 and TGF-β) in the original RCT was to study the in vivo effect of probiotics on both pro-inflammatory and anti-inflammatory cytokines (modulation of inflammation) in critically ill children with severe sepsis.

Materials and Methods

This study is a secondary analysis of data from a randomized, double-blind, placebo-controlled trial conducted in a 15-bedded pediatric intensive care unit (PICU) of a tertiary care teaching hospital in North India from November 2014 to October 2015.²⁹ The 100 children in age-group of 3 months to 12 years with severe sepsis were enrolled and randomized by computer-generated randomization table to probiotic (n = 50) and placebo (n = 50) groups. The probiotic group received probiotic mix (Lactobacillus paracasei, L plantarum, L acidophilus, L delbrueckii, Bifidobacterium longum, B breve, B infantis, and Streptococcus salivarius), maltose, and silicon dioxide, and the placebo group received only maltose and silicon dioxide. Probiotics or placebo were in powdered form with identical appearance, taste, and smell and packed in identical looking sachet. The 14 sachets per enrolled patient of either probiotic or placebo were packed in opaque packets with random allocation number written on it. The dose was 1 sachet twice daily for 7 days. Random number allocation and preparation of the packets containing probiotics or placebo were done by a person not involved in patient enrollment, data collection, data entry, and analysis. Randomization codes were broken only after analysis. Two microliters of blood was obtained from peripheral vein using aseptic precautions in a plain vial from each patient within 12 hours of admission to PICU for estimation of serum cytokine levels. The levels of pro-inflammatory cytokines (IL-6, IL-12p70, IL-17, TNF-α) and anti-inflammatory cytokines (IL-10, TGF-β) were determined by enzyme-linked immunosorbent assay (ELISA), as per the recommendations of the kit manufacturer (RayBio Human Cytokine ELISA kits; Ray Biotech Inc, Norcross, GA). The minimum detectable limits of IL-6, IL-12p70, IL-17, TNF-α, IL-10, and TGF-β were 3, 10, 80, 30, 1, and 80 ng/mL, respectively.²⁹ For this secondary analysis, we enrolled 50 children from the placebo group. Sepsis and severe sepsis were defined as per standard definitions.³⁰ Readers can refer to original RCT for more details on methodology.²⁹

Baseline demographic data including age, sex, diagnosis, indication of ICU admission, complications, interventions (mechanical ventilation, vasoactive support, dialysis, and antibiotics), duration of stay in PICU, and final outcome (survival/death) were noted. Severity of illness was measured by Pediatric Risk of Mortality III (PRISM-III) score³¹ within 24 hours of admission, Sequential Organ Failure Assessment (SOFA) score³² on day 1, and maximum Vasoactive-Inotropic Score (VIS).^33,34 Children were followed till PICU discharge or death. We followed uniform protocols for managing children with severe sepsis and septic shock,³⁰ acute lung injury and acute respiratory distress syndrome (ARDS),^30,35 and central nervous system (CNS) infections.³⁶

Outcomes

Primary outcome was to find out correlation between cytokine levels and illness severity (PRISM III, SOFA, and VIS scores), and secondary outcome was to compare cytokine levels among survivors and nonsurvivors.

Statistical Analysis

Data entry and statistical analysis were done by using Microsoft Excel 2013 (Microsoft, Redmond, Washington) and SPSS software version 21 (IBM Corp 2012; IBM SPSS Statistics for Windows, Version 21, Armonk, New York). Baseline variables were expressed by using descriptive statistics, median (interquartile range, IQR), range, and percentages. The correlation between cytokine levels and PRISM III, SOFA, and VIS score was demonstrated by Spearman correlation coefficient. The cytokine levels at admission among survivors and nonsurvivors were compared by Mann-Whitney U test. All tests were 2 tailed and P value <.05 was taken as significant. The formal sample size calculation was not carried out as this study is a secondary analysis of previously published trial.

Results

The median age at admission was 3.4 (IQR: 0.95-7) years and 60% were males. The most common diagnoses were CNS infection (n = 13), pneumonia (n = 10), gastrointestinal sepsis (n = 6), and scrub typhus (n = 6). The median PRISM III, SOFA, and VIS scores were 15 (12-20), 6 (3-8), and 30 (10-80), respectively. Various complications noted were septic shock (60%), acute kidney injury (32%), MODS (30%), health-care-associated infections (20%), and ARDS (18%; Table 1). Thirty children (60%) underwent mechanical ventilation for a median duration of 7 (5-12) days, 31 (62%) required vasoactive drugs for a median duration of 76 (48-96) hours, and 7 (14%) needed dialysis. The most commonly used antibiotics were ceftriaxone, aminoglycosides, and cloxacillin for a median duration of 8 (7-10) days. Five (10%) cases died (Table 1).

Table 1.

Demographic and Clinical Characteristics, Interventions, and Complications in Children With Severe Sepsis.

Characteristics	Children With Severe Sepsis (n = 50)	Survivors (n = 45)	Nonsurvivors (n = 5)	P
Age in years, median (IQR)	3.4 (0.95-7)	4 (1-7.5)	1.75 (0.5-3.25)	.03
Boys:girls	30:20	27:18	3:2	.69
Primary diagnosis				.51
Central nervous system infection, n (%)	13 (26)	11 (24.4)	2 (40)
Pneumonia, n (%)	10 (20)	9 (20)	1 (20)
Gastrointestinal sepsis, n (%)	6 (12)	6 (13.3)	0
Scrub typhus, n (%)	6 (12)	5 (11.1)	1 (20)
Disseminated staphylococcal sepsis, n (%)	4 (8)	3 (6.7)	1 (20)
Others, n (%)	11 (22)	11 (22)	0
Duration of illness (days), median (IQR)	4 (2.5-7)	3.9 (3-8)	4 (2-5)	.92
PRISM III, median (IQR)	15 (12-20)	15 (12.5-19.5)	29 (24-35.5)	.00
SOFA score at admission, median (IQR)	6 (3-8)	6 (3-7)	16 (9-17.5)	.00
VIS, median (IQR)	30 (10-80)	10 (0-30)	100 (28-135)	.02
GCS, median (IQR)	12 (7-15)	14 (7-15)	5 (5-11)	.03
Acute respiratory distress syndrome, n (%)	9 (18)	6 (13.3)	3 (60)	.03
Septic shock, n (%)	30 (60)	25 (55.6)	5 (100)	.07
Acute kidney injury, n (%)^a	16 (32)	11 (24.4)	5 (100)	.02
MODS, n (%)	15 (30)	11 (24.4)	4 (80)	.02
CRP at admission (mg%), median (IQR)	81 (35-140)	80 (35-130)	88 (41-232)	.57
Health-care-associated infection, n (%)	10 (20)	8 (17.8)	2 (40)	.25
Bloodstream infection, n (%)	5 (10)	4 (8.9)	1 (20)
Ventilator-associated pneumonia, n (%)	2 (4)	1 (2.2)	1 (20)
Urinary tract infection, n (%)	3 (6)	3 (6.7)
Required mechanical ventilation, n (%)	30 (60)	26 (57.8)	5 (100)	.08
Duration of ventilation in days, median (IQR)	7 (5-12)	7 (5-12)	6 (3-9)	.31
Required vasoactive drugs, n (%)	31 (62)	26 (57.8)	5 (100)	.08
Duration of vasoactive drugs in hours, median (IQR)	76 (48-96)	54 (34-86)	110 (77-144)	.04
Central line, n (%)	20 (40)	15 (33.3)	5 (100)	.007
Required dialysis, n (%)	7 (14)	3 (6.7)	4 (80)	.001
Received steroids	13 (26)	9 (20)	4 (80)	.01
Antibiotics used
Ceftriaxone, n	38	33	5
Gentamicin/amikacin, n	16	11	5
Cloxacillin, n	13	10	3
Doxycycline/azithromycin, n	8	7	1
Imipenem, n	7	5	2
Clindamycin, n	5	4	1
Duration of antibiotics in days, median (IQR)	8 (7-10)	9 (6-12)	6 (3-9)	.90
Duration of PICU stay in days, median (IQR)	9 (6-11)	9 (6-12)	6 (3-9)	.17

Abbreviations: CRP, C-reactive protein; GCS, Glasgow coma score; IQR, interquartile range; MODS, multiple organ dysfunction syndrome; PICU, pediatric intensive care unit; PRISM III, Pediatric Risk of Mortality III score; SOFA, Sequential Organ Failure Assessment score; VIS, Vasoactive-Inotropic Score.

^a pRIFLE criteria.

When compared to survivors, nonsurvivors were significantly younger (P = .03); had lower GCS at admission (P = .03); had higher PRISM III score (P < .001), SOFA score at admission (P < .001), VIS score (P = .02); higher proportion had ARDS (P = .03), AKI (P = .02), and MODS (P = .02); and require need of central venous catheter (P = .007), dialysis (P = .001), and steroids (P = .01). The proportion of cases having shock (P = .07) and requiring vasoactive drugs (0.08) and mechanical ventilation (P = .08) were higher among nonsurvivors, although the difference was not statistically significant (Table 1).

Primary Outcome

At admission, the levels of pro-inflammatory cytokine, median (IQR), were IL-6: 189 (35-285) pg/mL, IL-12p: 48 (28-98) pg/mL, IL-17: 240 (133-345) pg/mL, and TNF-α: 296 (198-430) pg/mL. The levels of anti-inflammatory cytokine were IL-10: 185 (62-395) pg/mL and TGF-β1: 204 (92-290) ng/mL (Table 2).

Table 2.

Secondary Outcome: Cytokine Levels at Admission in All Cases of Severe Sepsis and Among Survivors and Nonsurvivors.

Cytokine Levels at Admission	Total Cases (N = 50)	Survivors (n = 45)	Nonsurvivors (n = 5)	P ^a
IL-6 (pg/mL), median (IQR)	189 (35-285)	157 (97-223)	359 (251-499)	.000
IL-12p70 (pg/mL), median (IQR)	48 (28-98)	66 (30-100)	167 (133-196)	.000
IL-17 (pg/mL), median (IQR)	240 (133-345)	237 (122-318)	400 (333-563)	.009
TNF-α (pg/mL), median (IQR)	296 (198-430)	330 (198-415)	409 (355-503)	.002
IL-10 (pg/mL), median (IQR)	185 (62-395)	200 (62-408)	163 (50-240)	.31
TGF-β1 (ng/mL), median (IQR)	204 (92-290)	206 (98-292)	187 (52-309)	.62

Abbreviations: IQR, interquartile range; IL, interleukin; TGF, transforming growth factor; TNF, tumor necrosis factor.

^a P value between cytokine levels among survived and died (Mann-Whitney U test).

P value <0.05 is significant.

Pro-inflammatory cytokines showed weak positive correlation with PRISM III score: IL-6 (Spearman’s correlation coefficient, ρ = 0.273, P = .06), IL-12 (ρ = 0.367, P = .01), IL-17 (ρ = 0.197, P = .17), and TNF-α (ρ = 0.284, P = .05; Figure 1), and anti-inflammatory cytokines showed negative correlation: IL-10 (ρ = −0.257, P = .09) and TGF-β (ρ = −0.238, P = .11; Figure 2). Both SOFA and VIS also showed weak positive correlation with IL-12 (ρ = 0.32, P = .03 and ρ = 0.31, P = .03, respectively; Table 3).

Table 3.

Correlation Between Cytokines and Severity of Illness Parameters.^a

	PRISM III		SOFA		VIS
	ρ	P Value	ρ	P Value	ρ	P Value
IL-6	0.273	.06	0.17	.25	0.06	.71
IL-12p70	0.367	.01	0.31	.03	0.32	.03
IL-17	0.197	.17	0.09	.53	0.02	.90
TNF-α	0.284	.05	0.16	.29	0.16	.28
IL-10	−0.257	.09	−0.21	.17	0.01	.93
TGF-β1	−0.238	.11	−0.09	.56	0.12	.44

Abbreviations: IL, interleukin; TGF, transforming growth factor; TNF, tumor necrosis factor; PRISM III, Pediatric Risk of Mortality III score; SOFA, Sequential Organ Failure Assessment score; VIS, Vasoactive-Inotropic Score.

^a ρ: Spearmen correlation coefficient.

Secondary Outcomes

Nonsurvivors had significantly higher levels of pro-inflammatory cytokines, IL-6: 359 (151-499) versus 157 (97-223), P < .0001, IL-12p70: 167 (133-196) versus 66 (30-100), P < .0001, IL-17: 400 (333-503) versus 237 (122-318), P = .009, and TNF-α: 409 (355-503) versus 330 (198-415), P = .002, and statistically nonsignificant trend toward lower levels of anti-inflammatory cytokines, IL-10: 163 (500-240) versus 200 (62-408), P = .31, and TGF-β1: 187 (52-309) versus 206 (98-292), P = .62 (Table 3).

The median PRISM III, SOFA, and VIS scores were significantly higher among nonsurvivors than survivors, 29 (22-37) versus 15 (13-18), P < .0001, 15 (9-17) versus 6 (3-7), P < .0001, and 100 (28-135) versus 10 (0-30), P = .03, respectively.

Also, in total sample of 100 patients from original trial, nonsurvivors had significantly higher PRISM III, 22 (16-29) versus 15 (13-19.5), P = .009, SOFA, 9 (6-12) versus 5 (3-8), P = .004, and VIS score, 115 (45-125) versus 10 (0-45), P = .04, than survivors.

Discussion

Advances in diagnostic modalities and therapeutic strategies (improved resuscitation, appropriate infection control, and better supportive critical care therapies) have led to improved outcome in children with sepsis. Even then, the morbidity and mortality due to sepsis remains high, especially in resource-limited setting.^7,37,38 Early identification of patients with severe sepsis with higher severity of illness and increased risk of death is crucial. This may help in providing aggressive individualized care to optimize the organ functions among high-risk cases and potential for better survival. In this study, we tried to address this issue and demonstrated a weak positive correlation between baseline pro-inflammatory cytokines (especially IL-12p70) and severity of illness scores (PRISM III, SOFA, and VIS scores) and evidence of profound inflammation at admission to PICU among the nonsurvivors of severe sepsis. The patient population included in this study is critically ill children with severe sepsis due to infectious diseases unique to Indian subcontinent and which are associated with high mortality when compared to Western world.^5,6

Few studies involving critically ill adults and children have used various cytokines to predict the mortality^{8,14,19

–28} and disease severity.^8,28 Gogos et al¹⁴ noted that serum levels of IL-10 and TNF-α were significantly higher in nonsurvivors with severe sepsis (n = 65). Bozza et al⁸ demonstrated that in 60 adults with severe sepsis, the levels of IL-1β, IL-6, IL-7, IL-8, IL-10, IL-13, IFN-γ, monocyte chemoattractant protein 1 (MCP-1), and TNF-α were significantly higher in patients with shock than in those without shock. They also observed that IL-8 and MCP-1 had positive correlation with SOFA score, and high IL-1β, IL-4, IL-6, IL-8, MCP-1, and Granulocyte-Colony Stimulating Factor (G-CSF) significantly predicted mortality. In patients with postoperative (cardiovascular) multi-organ failure (n = 14), Hamano et al¹⁹ demonstrated that the serum level of IL-8 was significantly higher in nonsurvivors when compared to survivors. Heper et al²¹ compared C-reactive protein, procalcitonin, IL-10, and TNF-α at admission, 24th, 48th, and 72nd hours in cases of sepsis (n = 21) and severe sepsis (n = 18 cases). Procalcitonin levels were significantly higher in severe sepsis group at all times points and were significantly higher in nonsurvivors at the 72nd hour. The TNF-α levels were significantly higher among nonsurvivors at admission as well as at the 24th hour. Levels of IL-10 were higher in severe sepsis group and among nonsurvivors at all time points. Andaluz-Ojeda and Bobillo²³ demonstrated that in adults with severe sepsis or septic shock (n = 29), the levels of IL-6, IL-8, MCP-1, and IL-10 (within first 24 hours of ICU admission) were higher in nonsurvivors (n = 12). They devised a combined cytokine score which showed strongest association with mortality (much stronger than individual cytokines).

Results of above studies involving critically ill adults are somewhat similar to the results shown in index study. Very few pediatric studies have demonstrated the role of cytokines levels in predicting the severity of illness and mortality in children with severe sepsis. Wong et al²⁶ demonstrated that in children with septic shock, a serum IL-8 level of ≤220 pg/mL (measured within 24 hours of admission) predicted a high likelihood of survival (95% negative predictive value [NPV] for 28-day mortality]. Nowak et al²⁷ noted that in 156 children with septic shock, the admission serum levels of chemokine (C-C motif) ligand 4 (CCL4) of >140 pg/mL had 92% sensitivity and 40% specificity for predicting mortality. A level of ≤140 pg/mL had NPV of 98% for mortality. Vermont et al²⁸ enrolled 58 children with meningococcal sepsis and noted that levels of MCP-1, macrophage inflammatory protein 1α, growth-related gene product α, and IL-8 (at admission and after 24 hours) were extremely high in all children, correlated strongly with disease severity (higher in septic shock group), and significantly higher in nonsurvivors (n = 9) than in survivors. Wong et al¹² noted that in children with septic shock, the pediatric sepsis biomarker risk model (CCL3, HSP-A1B, IL8, neutrophil elastase 2, and lipocalin 2 measured within 24 hours of PICU admission) reliably identifies children at risk of death and greater illness severity. Similar to results of the index study, authors from developed countries also noted that inflammatory cytokines and biomarkers were extremely high in pediatric sepsis and septic shock, correlated strongly with disease severity, and significantly higher among nonsurvivors.^12,26
–28 Therefore, the markers of inflammation could be a potential tool to predict severity of disease and risk of death among children with severe sepsis at admission.

We did not notice a good correlation of anti-inflammatory cytokines with illness severity scores, and the levels of anti-inflammatory cytokines showed a statistically nonsignificant trend toward lower values among nonsurvivors. In contrast, few studies involving adults demonstrated that anti-inflammatory cytokines (IL-10) were higher in nonsurvivors of severe sepsis.^{14,21,23,39,40} It has been hypothesized that the continuous dominant immunosuppressive effect of anti-inflammatory cytokines triggered by pro-inflammatory cytokines at beginning of illness lead to immune suppression, enhanced susceptibility to bacterial infections, and increased mortality.^21,41,42 The higher levels of pro-inflammatory and low levels of anti-inflammatory cytokines in index study could be due to unbalanced and unchecked excessive inflammatory response in children with severe sepsis.

The major strength of this study is that this is the first study of its kind where multiple (both pro- and anti-inflammatory) cytokines were studied in children with severe sepsis to correlate with the severity of illness and mortality. Multiple cytokines are studied in only a few studies including critically ill adults.^8,9,23 It has been suggested that simultaneous evaluation of both pro- and anti-inflammatory cytokines to predict mortality is justified based on the physiological behavior and effect of cytokines on the sepsis outcome.²³ We enrolled a homogenous cohort of children with severe sepsis due to infections commonly seen in Indian subcontinent, and results may be extrapolated to this group of critically ill children.

The limitations of this study are: We studied cytokine profile at single time point (ie, at admission). Measuring cytokines at different time points could have given the information about trajectories and trends of these cytokines in relation to disease progression and better differentiation between survivors and nonsurvivors. Though in the original trial, cytokine levels were also measured on day 7, but there were only 2 deaths after 7 days, thus making comparison between survivors and nonsurvivors irrelevant. There are few pediatric studies involving critically ill children with septic shock where measurement of cytokines levels only once (within 24 hours of admission to PICU) had reliably identified the children at greater risk of death and illness severity.^12,26
–28 So, we assumed that single value at admission in children with severe sepsis might also be useful. Moreover, this will also be useful in planning the resources at admission in resource-limited settings or in busy PICUs. More research is needed to be conducted on this subject, and serial measurement of cytokines may give a better picture. Another limitation is inability to perform stratification of cytokine levels to predict outcome due to small sample size as the process of stratification requires generation of cutoff values at first place in a group of patients and then validation of cutoff values generated on another prospective cohort. Although we performed multiple comparisons in the study, the adjustment was not technically possible because of small size of this single-center study. Also, ELISA has not been done in duplicates.

Conclusion

We have demonstrated a weak positive correlation between pro-inflammatory (especially IL-12p70) cytokines and severity of illness scores (PRISM III, SOFA, and VIS scores) in critically ill children with severe sepsis. The nonsurvivors had significant higher levels of pro-inflammatory cytokines at admission. This information may be valuable in devising scoring system for assessing severity of disease and predicting mortality.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Financial assistance from Department of Pediatrics for buying ELISA kits of cytokines.

ORCID iD

Arun Bansal, MD, FIAP, FICCM

Jayashree Muralidharan, MD, DNB, FIAP, FICCM

References

Weiss

Fitzgerald

Pappachan

, et al. Global epidemiology of pediatric severe sepsis: the sepsis prevalence, outcomes, and therapies study. Am J Respir Crit Care Med. 2015;191(10):1147–1157.

Kaukonen

Bailey

Suzuki

Pilcher

Bellomo

. Mortality related to severe sepsis and septic shock among critically ill patients in Australia and New Zealand, 2000-2012. JAMA. 2014;311(13):1308–1316.

Marshall

. Understanding the global burden of pediatric sepsis. Am J Respir Crit Care Med. 2015;191(10):1096–1098.

Giuliano

Jr Markovitz

Brierley

, et al. Comparison of pediatric severe sepsis managed in U.S. and European ICUs. Pediatr Crit Care Med. 2016;17(6):522–530.

Tan

Wong

Sultana

Koh

Jit

Mok

, et al. Global case-fatality rates in pediatric severe sepsis and septic shock: a systematic review and meta-analysis. JAMA Pediatr. 2019;173(4):352–362.

Kaur

Vinayak

Mittal

Kaushik

Aamir

. Clinical outcome and predictors of mortality in children with sepsis, severe sepsis, and septic shock from Rohtak, Haryana: a prospective observational study. Indian J Crit Care Med. 2014;18(7):437–441.

Bateman

Seed

. Procession to pediatric bacteremia and sepsis: covert operations and failures in diplomacy. Pediatrics. 2010;126(1):137–150.

Bozza

Salluh

Japiassu

, et al. Cytokine profiles as markers of disease severity in sepsis: a multiplex analysis. Crit Care. 2007;11(2):R49.

Mera

Tatulescu

Cismaru

, et al. Multiplex cytokine profiling in patients with sepsis. APMIS. 2011;119(2):155–163.

10.

Wong

Weiss

Giuliano

Jr Wainwright

Cvijanovich

Thomas

, et al. The temporal version of the pediatric sepsis biomarker risk model. PLoS One. 2014;9(3):e92121.

11.

Carlton

McHugh

McDonough

Sturza

Desch

Cornell

. Markers of endothelial dysfunction and cytokines in high-risk pediatric patients with severe sepsis. Am J Respir Crit Care Med. 2020;201(3):380–384.

12.

Wong

Salisbury

Xiao

, et al. The pediatric sepsis biomarker risk model. Crit Care. 2012;16(5):R174.

13.

Tamayo

Fernandez

Almansa

, et al. Pro- and anti-inflammatory responses are regulated simultaneously from the first moments of septic shock. Eur Cytokine Netw. 2011;22(2):82–87.

14.

Gogos

Drosou

Bassaris

Skoutelis

. Pro- versus anti-inflammatory cytokine profile in patients with severe sepsis: a marker for prognosis and future therapeutic options. J Infect Dis. 2000;181(1):176–180.

15.

Hackett

Thomson

Hart

. Cytokines, chemokines and other effector molecules involved in meningococcal disease. J Med Microbiol. 2001;50(10):847–859.

16.

Dinarello

. Proinflammatory and anti-inflammatory cytokines as mediators in the pathogenesis of septic shock. Chest. 1997;112(6 suppl):321S–329S.

17.

Vardas

Ilia

Sertedaki

, et al. Increased glucocorticoid receptor expression in sepsis is related to heat shock proteins, cytokines, and cortisol and is associated with increased mortality. Intensive Care Med Exp. 2017;5(1):10.

18.

Fitrolaki

Dimitriou

Venihaki

Katrinaki

Ilia

Briassoulis

. Increased extracellular heat shock protein 90alpha in severe sepsis and SIRS associated with multiple organ failure and related to acute inflammatory-metabolic stress response in children. Medicine (Baltimore). 2016;95(35):e4651.

19.

Hamano

Gohra

Noda

, et al. Increased serum interleukin-8: correlation with poor prognosis in patients with postoperative multiple organ failure. World J Surg. 1998;22(10):1077–1081.

20.

Panacek

Marshall

Albertson

, et al. Efficacy and safety of the monoclonal anti-tumor necrosis factor antibody F(ab’)2 fragment afelimomab in patients with severe sepsis and elevated interleukin-6 levels. Crit Care Med. 2004;32(11):2173–2182.

21.

Heper

Akalin

Mistik

, et al. Evaluation of serum C-reactive protein, procalcitonin, tumor necrosis factor alpha, and interleukin-10 levels as diagnostic and prognostic parameters in patients with community-acquired sepsis, severe sepsis, and septic shock. Eur J Clin Microbiol Infect Dis. 2006;25(8):481–491.

22.

Rivers

Kruse

Jacobsen

, et al. The influence of early hemodynamic optimization on biomarker patterns of severe sepsis and septic shock. Crit Care Med. 2007;35(9):2016–2024.

23.

Andaluz-Ojeda

Bobillo

, et al. A combined score of pro- and anti-inflammatory interleukins improves mortality prediction in severe sepsis. Cytokine. 2012;57(3):332–336.

24.

Tschaikowsky

Hedwig-Geissing

Braun

Radespiel-Troeger

. Predictive value of procalcitonin, interleukin-6, and C-reactive protein for survival in postoperative patients with severe sepsis. J Crit Care. 2011;26(1):54–64.

25.

Hotchkiss

Moldawer

Opal

Reinhart

Turnbull

Vincent

. Sepsis and septic shock. Nat Rev Dis Primers. 2016;2:16045.

26.

Wong

Cvijanovich

Wheeler

, et al. Interleukin-8 as a stratification tool for interventional trials involving pediatric septic shock. Am J Respir Crit Care Med. 2008;178(3):276–282.

27.

Nowak

Wheeler

Harmon

Wong

. Admission chemokine (C-C motif) ligand 4 levels predict survival in pediatric septic shock. Pediatr Crit Care Med. 2010;11(2):213–216.

28.

Vermont

Hazelzet

de Kleijn

van den Dobbelsteen

de Groot

. CC and CXC chemokine levels in children with meningococcal sepsis accurately predict mortality and disease severity. Crit Care. 2006;10(1):R33.

29.

Angurana

Bansal

Singhi

, et al. Evaluation of effect of probiotics on cytokine levels in critically ill children with severe sepsis: a double-blind, placebo-controlled trial. Crit Care Med. 2018;46(10):1656–1664.

30.

Dellinger

Levy

Rhodes

, et al. Surviving sepsis campaign: international guidelines for management of severe sepsis and septic shock: 2012. Crit Care Med. 2013;41(2):580–637.

31.

Pollack

Patel

Ruttimann

. PRISM III: an updated pediatric risk of mortality score. Crit Care Med. 1996;24(5):743–752.

32.

Vincent

de Mendonca

Cantraine

, et al. Use of the SOFA score to assess the incidence of organ dysfunction/failure in intensive care units: results of a multicenter, prospective study. Working group on “sepsis-related problems” of the European Society of Intensive Care Medicine. Crit Care Med. 1998;26(11):1793–1800.

33.

McIntosh

Tong

Deakyne

Davidson

Scott

. Validation of the Vasoactive-Inotropic Score in pediatric sepsis. Pediatr Crit Care Med. 2017;18(8):750–757.

34.

Wernovsky

Wypij

Jonas

, et al.

Postoperative course and hemodynamic profile after the arterial switch operation in neonates and infants. A comparison of low-flow cardiopulmonary bypass and circulatory arrest

. Circulation. 1995;92(8):2226–2235.

35.

Saharan

Lodha

Kabra

. Management of acute lung injury/ARDS. Indian J Pediatr. 2010;77(11):1296–1302.

36.

Jindal

Singhi

. Non-traumatic coma and altered mental status. Indian J Pediatr. 2012;79(3):367–375.

37.

Watson

Carcillo

. Scope and epidemiology of pediatric sepsis. Pediatr Crit Care Med. 2005;6(3 suppl):S3–S5.

38.

Hua

Lin

Lai

Kong

Chang

Chen

. Probiotic bio-three induces Th1 and anti-inflammatory effects in PBMC and dendritic cells. WJG. 2010;16(28):3529–3540.

39.

Wunder

Eichelbronner

Roewer

. Are IL-6, IL-10 and PCT plasma concentrations reliable for outcome prediction in severe sepsis? A comparison with APACHE III and SAPS II. Inflamm Res. 2004;53(4):158–163.

40.

Monneret

Finck

Venet

, et al. The anti-inflammatory response dominates after septic shock: association of low monocyte HLA-DR expression and high interleukin-10 concentration. Immunol Lett. 2004;95(2):193–198.

41.

Adib-Conquy

Cavaillon

. Compensatory anti-inflammatory response syndrome. Thromb Haemost. 2009;101(1):36–47.

42.

Chiche

Forel

Thomas

, et al. The role of natural killer cells in sepsis. J Biomed Biotechnol. 2011;2011:986491.