The effect of GM-CSF and predictors of treatment outcome in pediatric septic shock patients

doi:10.21203/rs.3.rs-4790762/v1

Background

Septic shock in pediatric patients frequently leads to high mortality rates, primarily attribute to sepsis-induced immunosuppression. Granulocyte-macrophage colony-stimulating factor (GM-CSF) has emerged as a potential therapy to mitigate this immunosuppression. However, the effect of GM-CSF in pediatric septic shock remains uncertain. This study aimed to investigate the effect of GM-CSF administration and identify the potential treatment outcome predictors in pediatric patients with septic shock.

Methods

The retrospective cohort study was structured to include pediatric patients admitted to the Pediatric Intensive Care Unit (PICU) of Children’s Hospital of Fudan University in Shanghai, between January 1^st, 2019 and January 1^st, 2024, who were diagnosed with septic shock during hospitalization. Data were collected on the demographic characteristics of patients, treatments of GM-CSF and other combined therapies, laboratory findings, and clinical outcomes. Univariate and multivariate logistic regression were utilized to evaluate the effect of GM-CSF on 28-day mortality and identify its potential treatment outcome predictors.

Results

A total of 200 patients were included in the study: 66 in GM-CSF treatment group and 134 in no GM-CSF treatment group. The 28-day mortality of GM-CSF group showed higher risk compared with no GM-CSF group (59.1% vs. 35.1%, P=0.001). In the multivariate logistic regression analysis, the effect of GM-CSF showed decreased the risk of 28-day mortality (OR=0.472, 95% CI 0.153-1.457, P=0.192) after adjusting the confounding factors. Potential outcome predictors of GM-CSF treatment included hematopoietic stem cell transplantation (HSCT), lactic acid (LAC) levels, hospital-acquired septic shock (HASS), red blood cell (RBC) count, and platelet (PLT) count.

Conclusions

GM-CSF treatment showed clinical benefit in pediatric septic shock patients, especially with higher LAC, lower RBC, lower PLT. The potential treatment outcome predictors can be monitored during the treatment of pediatric septic shock.

Granulocyte-macrophage colony-stimulating factor

Pediatric

Septic shock

Mortality

Sepsis, defined as a life-threatening organ dysfunction resulting from an impaired host response to infection(1), represents a serious global health problem and ranks as the leading cause of death in hospitals(2). In 2017, an estimated 25 million children suffered from sepsis worldwide, leading to more than 3 million fatalities(3). Global epidemiological research demonstrates that the incidence of pediatric sepsis in the pediatric intensive care unit (PICU) reached 8.2%, with a high mortality of 25%(4). Septic shock is a specific subtype of sepsis characterized by profound alterations in circulation, cellular function, and metabolic processes(1). These significant changes contribute to the severity and complexity of the condition, necessitating prompt and comprehensive medical intervention to improve outcomes.

Immunologically, the interplay between pathogen-induced pro-inflammatory responses and associated anti-inflammatory reactions often results in severe immunosuppression among a substantial subset of patients(5). This sepsis-induced immunosuppression sets the stage for secondary infections and multiple organ dysfunction syndrome (MODS), thereby exacerbating the risk of adverse outcomes(6).

The potential of granulocyte-macrophage colony-stimulating factor (GM-CSF) in mitigating sepsis-induced immunosuppression has garnered interest due to its ability to revitalize immune responses and enhance immune function(7, 8). In a study conducted by Orozco et al.(9), GM-CSF demonstrated a positive effect in reducing the duration of antibacterial treatment and minimizing associated complications in patients with abdominal sepsis, although it did not lead to a reduction in in-hospital mortality. Nevertheless, its therapeutic effect in the context of sepsis remains uncertain in pediatric patients. A randomized controlled trial by Bilgin et al.(10) in infants with neutropenia and sepsis showed GM-CSF was associated with a reduction in mortality. However, a retrospective study on pediatric sepsis have indicated a higher frequency of GM-CSF use among non-survivors without further analyzing the therapeutic effect of GM-CSF(11). Given the existing uncertainty surrounding the therapeutic effect of GM-CSF in pediatric sepsis, this single-center retrospective study aims to investigate the effect of GM-CSF and identify the potential treatment outcome predictors in pediatric septic shock patients.

Study design and subjects

This 5-year, single-center, retrospective cohort study was designed to conduct from January 1, 2019, to December 31, 2023, in the PICU of Children’s Hospital of Fudan University. The inclusion criteria were as follows: (1) pediatric patients were 29 days to 18 years old; (2) all patients were diagnosed with pediatric septic shock according to the 2015 Chinese Pediatric Expert Consensus: 1) hypotension (blood pressure was lower than the 5th percentile for age, or systolic blood pressure was more than 2 standard deviations below the normal value for age); 2) blood pressure needed to be maintained with vasoactive medications (dopamine > 5 μg/kg/min) or any dose of dobutamine, norepinephrine, or epinephrine; 3) at least three manifestations of hypoperfusion among (i) peripheral arterial pulsations weakened, increased heart rate and pulse rate, (ii) pale or ashen skin, wet and cold, marbled pattern, (iii) the capillary refilling time (CRT) was prolonged for more than 3 seconds (except for the influence of environmental temperature), (iv) early irritability or malaise, apathy, late stage of confusion, even coma, convulsion, (v) after fluid resuscitation, the urine volume remains less than 0.5 mL/kg/h for at least 2 hours, (vi) lactic acidosis, arterial blood lactic acid more than 2 mmol/L(12). Patients with incomplete medical records were excluded. In the PICU ward, GM-CSF treatment is administered when a patient's absolute neutrophil count is less than 0.5×10^9/L. The retrospective study was approved by the Ethics Committee of the Children’s Hospital of Fudan University [No.431 (2021)]. Written informed consent of all participants was obtained.

Groups

In this study, patients were mainly categorized into the GM-CSF group and the No GM-CSF group based on whether they received GM-CSF treatment.

Data collection

The following clinical data were collected from the electronic medical record system: clinical and demographic information such as age, sex, diagnosis, and laboratory test results; administered treatments (antibiotic therapy, blood transfusion, vasopressors or steroids, mechanical ventilation, and renal replacement therapy); the source (emergency department or hospital ward) and the reason of admission to the PICU, and the Pediatric Critical Illness Score (PCIS)(13) when diagnosed with septic shock. The study outcomes included 28-day mortality, in-hospital mortality, and PICU hospitalization time.

Definitions

28-day mortality was referred to the mortality within 28 days after being diagnosed with septic shock(12). The definitions of community-acquired septic shock (CASS) and hospital-acquired septic shock (HASS) were determined based on the time of infection that developed septic shock upon admission. CASS was within 48 hours while HASS was after 48 hours of admission, including 48 hours(12). Underlying diseases were defined as diseases that existed before the onset of septic shock and are not directly related to septic shock.

Statistical analysis

All statistical analyses were performed using IBM SPSS Statistic 26. The normality of the continuous data was verified by the Kolmogorov–Smirnov test. The differences in quantitative data with a normal distribution of the two groups were compared using the Student’s t-test. While in non-normal distribution, the Mann-Whitney U test was used. Categorical variables were analyzed by Pearson’s chi-square test. The quantitative data with normal distribution were expressed as mean ± standard deviation, whereas those with a non-normal distribution were represented as median and quartiles. The categorical variables were reported as count and percentage. Univariate and multivariable logistic regression was performed to identify the potential predictive factors of GM-CSF in the treatment outcomes of septic shock. The Receiver Operating Characteristic (ROC) Curve was used to determine the cutoff value, and the two-classification were performed. Due to some variables cannot be classified using ROC curve, we classify these data such as WBC into quartiles. Results were presented as adjusted odds ratio (OR) and 95% confidence interval (CI). P<0.05 was considered as statistical significance.

Demographic characteristics of patients

The study involved 213 pediatric patients with septic shock admitted over a 5-year period, with 13 patients excluded due to incomplete data, resulting in a final analysis of 200 patients (Fig. 1). Among these, 66 (33%) received GM-CSF treatment, while 134 (67%) did not. In the GM-CSF group, a significant proportion of patients (98.5%) had underlying diseases, including hematologic or oncologic diseases (66.7%), HSCT (27.3%), and so on. The median age of the GM-CSF group was 6.8 years (2.4, 11.8) years, and 66.7% (44/66) were male. Demographic characteristics and laboratory tests are detailed in Table 1. Baseline levels differ significantly between the two groups.

Supportive therapies in pediatric patients with septic shock

In terms of supportive therapies, the analysis revealed no statistically significant variances between the two groups concerning the utilization of diverse interventions. These interventions encompassed empirical antibacterial treatment administered on the initial day, respiratory support, renal replacement therapy, and the administration of steroids. The utilization rate of norepinephrine in the GM-CSF group (98.5%) was significantly more significant compared to the group without GM-CSF (85.8%) in the context of vasoactive drugs (P=0.005), with no discernible disparity observed in the usage of other vasoactive agents (Table 2).

Outcomes and complications of patients

The GM-CSF group had more disseminated intravascular coagulation (DIC) patients than the No GM-CSF group (45.5% vs. 24.6%, P=0.003). However, no significant disparities were noted between the two groups in terms of the proportion of patients experiencing respiratory failure or other complications. In addition, the analysis indicated a significantly higher 28-day mortality rate (59.1% vs. 35.1%, P=0.001) and a higher in-hospital mortality rate (43.9% vs. 26.1%, P=0.011) among patients who received GM-CSF treatment, contrasting with those who did not. Nevertheless, no statistically significant difference was found in the duration of ICU stay (P=0.351) (Table 3).

Potential predictive factors of GM-CSF in the treatment outcomes of septic shock

In the univariate logistic regression analysis, GM-CSF emerged as a risk factor for 28-day mortality (OR 2.674; CI 0.459-4.899, P=0.001). In contrast, multivariable analysis showed when demographic characteristics, the severity of disease, and related laboratory indicators were adjusted, the use of GM-CSF had a protective effect on 28-day mortality although the p-value is not statistically significant (OR 0.472; CI 0.153-1.457, P=0.192). Univariate regression model and Model 1-3 suggested that HSCT (P=0.024), LAC (P<0.001), HASS (P=0.018), RBC (P=0.022), and PLT (P<0.001) were important predictors associating to the effect of GM-CSF on 28-day mortality (Table 4).

In our retrospective single-center cohort study involving 200 septic shock patients, we identified GM-CSF treatment as a protective factor associated with 28-day mortality in our multivariable regression model. In addition, we also identified several predictors of the treatment outcomes, including higher LAC, lower RBC and lower PLT.

As a pro-inflammatory cytokine, GM-CSF plays a crucial role in stimulating the bone marrow to produce granulocytes and macrophages, thereby counteracting immune suppression in septic shock patients(14). Various studies have explored the impact of GM-CSF on sepsis-induced immunosuppression(15, 16), yet their findings are inconclusive due to methodological differences. For instance, in a randomized controlled trial, C.-H. Vacheron found that using GM-CSF could not decrease the 28-day mortality and ICU-acquired infection (15). Contrary to the research, some retrospective analyses of G-CSF or GM-CSF demonstrated a positive effect on sepsis(17). However, these studies were conducted on adults, leaving the effect of GM-CSF on 28-day mortality in pediatric septic shock patients unknown.

In the beginning, we observed a higher 28-day mortality rate, in-hospital mortality, and a greater proportion of DIC complications among individuals treated with GM-CSF therapy. However, we observed differences in the baseline characteristics of patients between the two groups, suggesting the above results may have been influenced by confounding factors. Therefore, we employed multivariable logistic regression analysis in an attempt to mitigate the impact of these confounding factors on the results. Due to the characteristics of septic shock progression, we incorporate patients' baseline conditions, infection status, and hematologic indicators into our regression model. Finally, the OR for GM-CSF is 0.472 (95% CI: 0.153-1.457), indicating that it is a protective factor although not statistically significant. This finding contrasts with our initial observation. To explore this distinction, we then focused on significant predictors associated with outcomes in our model, including HSCT, high LAC, HASS, low RBC, and low PLT, which may influence the association between GM-CSF and 28-day mortality.

The first set of predictors associated to the outcome of GM-CSF treatment includes patients' baseline conditions, chiefly involving HSCT and elevated lactate levels. In our study, patients with a history of tumor chemotherapy or HSCT constituted the two main groups significantly associated with GM-CSF usage in septic shock cases. Although the overall objective of using GM-CSF as adjunctive therapy in these two patient groups is consistent – namely, to enhance the immune function in neutropenic patients(18-20) – only HSCT emerged as a significant predictor associating with the effect of GM-CSF treatment. Patients with HSCT generally have poorer baseline conditions and are susceptible to various risk factors for mortality because of graft-versus-host disease and infections(21). Elevated lactate levels is another crucial predictor, indicating tissue hypoxia and exacerbating the suppressive effects on immune cells, ultimately compromising the body's immune response capabilities(22). A retrospective cohort study of 1,060 patients diagnosed with septic shock revealed a significant correlation between lactate levels at 6 hours and mortality rates (OR with 95% CI, 1.27 [1.21-1.34]). Therefore, it makes sense that high lactate level is associated with adverse outcomes. Next, we focus on the infection-related factors, with Hospital-acquired septic shock (HASS) emerging as the most significant. Previous research has consistently highlighted a higher 28-day mortality among HASS patients compared to those with community-acquired septic shock (CASS)(12, 23), underscoring the importance of considering HASS as a predictor of outcomes.

Notably, the hematologic status including RBC, and PLT was also an important predictor of patients’ outcomes in our study. RBCs not only have the function of transporting and exchanging oxygen and carbon dioxide but also play an essential role in cellular blood immunity such as chemokine regulation, complement binding, and pathogen immobilization(24, 25). In the context of sepsis, decreased RBC counts can stem from various mechanisms, including functional iron deficiency, infection, inflammation, and the onset of multiple organ dysfunction(26). Although studies have shown that RBC count alone may not serve as a diagnostic or prognostic indicator in sepsis, it can serve as a valuable marker reflecting the disease's severity(27). In our retrospective study, the majority of patients had underlying diseases, and those who experienced septic shock with low red RBC levels tended to have worse clinical outcomes. Platelets, on the other hand, are pivotal in hemostasis and exert significant modulatory effects on inflammatory responses(28). Thrombocytopenia, commonly observed in sepsis and septic shock, often results from decreased production and increased consumption(29). Notably, lower platelet counts have been consistently linked to heightened mortality among patients admitted to the ICU(30, 31). For instance, a prospective study involving 931 septic shock patients revealed that individuals with low and intermediate-low platelet counts exhibited a heightened risk of 30-day mortality compared to those with normal platelet levels (hazard ratios with 95% CI, 2.00 [1.32-3.05] and 1.72 [1.22-2.44], respectively(32).

Despite these insights, our study faced several limitations. Firstly, this study is an observational study. Although it is a retrospective cohort study, factors of the efficacy of GM-CSF treatment in pediatric septic shock need to be tested by RCTs. Secondly, our study is a single-center study, which may limit the generalizability of our findings. Finally, we still need to increase the days of follow-up to examine the long term clinical outcome.

In conclusion, GM-CSF treatment showed clinical benefit in pediatric septic shock patients. Additionally, this study reveals several predictors of the outcomes of pediatric septic shock including higher LAC, lower RBC and lower PLT, which indicates the importance of monitoring these indicators during the treatment of pediatric septic shock.

Ethics approval and consent to participate

The study protocol was approved by the Medical Ethics Committee of the Children's Hospital of Fudan University [No.431 (2021)] and was conducted in accordance with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards. The patient’s informed consent has been obtained from the patient’s legal guardian.

Consent for publication

Informed consent was obtained from legal guardians of all patients who participated in the study for publication of anonymous patients’ data.

Availability of data and materials

The datasets generated and analyzed during the current study are available from the corresponding author on reasonable requests.

Competing interests

The authors declare no competing interests.

Funding

This work was supported by grants from the National Key R&D Program of China (2021YFC2701800, 2021YFC2701805), Shanghai Municipal Science and Technology Major Project (ZD2021CY001), National Natural Science Foundation of China (82202374), Shanghai Committee of Science and Technology (22ZR1408500), Shanghai Municipal Health System major supports discipline projects (2023ZDFC0103),Clinical Research Plan of Shanghai Hospital Development Center (SHDC2022CRS027).

Authors' contributions

ZCY and LGP participated in the design and concept of the study, supervised data collection, and revised the manuscript for important intellectual content. YZH, TGX and WYD were involved in the study’s conceptualization and design, data collection and analysis, and drafted the manuscript. LTY and CWM made a huge contribution to the acquisition of data. SP participated in the study design and reviewed the manuscript. YJY and ZYF revised the manuscript for important intellectual content. All authors read and approved the final manuscript.

Acknowledgments

We are grateful to all the nurses, respiratory therapists, and doctors in the pediatric intensive care unit of Children’s Hospital of Fudan University. In particular, we thank all the patient family members for trusting us over the years.

Singer M, Deutschman CS, Seymour CW, Shankar-Hari M, Annane D, Bauer M, et al. The Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3). Jama. 2016;315(8):801-10.
Rudd KE, Johnson SC, Agesa KM, Shackelford KA, Tsoi D, Kievlan DR, et al. Global, regional, and national sepsis incidence and mortality, 1990-2017: analysis for the Global Burden of Disease Study. Lancet. 2020;395(10219):200-11.
Weiss SL, Peters MJ, Alhazzani W, Agus MSD, Flori HR, Inwald DP, et al. Surviving sepsis campaign international guidelines for the management of septic shock and sepsis-associated organ dysfunction in children. Intensive Care Med. 2020;46(Suppl 1):10-67.
Wang S, Yin F, Zhang Y, An K, Xi Y, Lu X, et al. Epidemiology and clinical characteristics of pediatric sepsis in PICUs of China: A national cross-sectional study. MedComm (2020). 2023;4(1):e211.
Venet F, Rimmelé T, Monneret G. Management of Sepsis-Induced Immunosuppression. Crit Care Clin. 2018;34(1):97-106.
Liu D, Huang SY, Sun JH, Zhang HC, Cai QL, Gao C, et al. Sepsis-induced immunosuppression: mechanisms, diagnosis and current treatment options. Mil Med Res. 2022;9(1):56.
Hotchkiss RS, Monneret G, Payen D. Sepsis-induced immunosuppression: from cellular dysfunctions to immunotherapy. Nat Rev Immunol. 2013;13(12):862-74.
Bo L, Wang F, Zhu J, Li J, Deng X. Granulocyte-colony stimulating factor (G-CSF) and granulocyte-macrophage colony stimulating factor (GM-CSF) for sepsis: a meta-analysis. Crit Care. 2011;15(1):R58.
Orozco H, Arch J, Medina-Franco H, Pantoja JP, González QH, Vilatoba M, et al. Molgramostim (GM-CSF) associated with antibiotic treatment in nontraumatic abdominal sepsis: a randomized, double-blind, placebo-controlled clinical trial. Arch Surg. 2006;141(2):150-3; discussion 4.
Bilgin K, Yaramiş A, Haspolat K, Taş MA, Günbey S, Derman O. A randomized trial of granulocyte-macrophage colony-stimulating factor in neonates with sepsis and neutropenia. Pediatrics. 2001;107(1):36-41.
Weiss SL, Fitzgerald JC, Pappachan J, Wheeler D, Jaramillo-Bustamante JC, Salloo A, et al. Global epidemiology of pediatric severe sepsis: the sepsis prevalence, outcomes, and therapies study. Am J Respir Crit Care Med. 2015;191(10):1147-57.
Su GY, Fan CN, Fang BL, Xie ZD, Qian SY. Comparison between hospital- and community-acquired septic shock in children: a single-center retrospective cohort study. World J Pediatr. 2022;18(11):734-45.
Zhang L, Huang H, Cheng Y, Xu L, Huang X, Pei Y, et al. [Predictive value of four pediatric scores of critical illness and mortality on evaluating mortality risk in pediatric critical patients]. Zhonghua Wei Zhong Bing Ji Jiu Yi Xue. 2018;30(1):51-6.
Becher B, Tugues S, Greter M. GM-CSF: From Growth Factor to Central Mediator of Tissue Inflammation. (1097-4180 (Electronic)).
Vacheron CH, Lepape A, Venet F, Monneret G, Gueyffier F, Boutitie F, et al. Granulocyte-macrophage colony-stimulating factor (GM-CSF) in patients presenting sepsis-induced immunosuppression: The GRID randomized controlled trial. (1557-8615 (Electronic)).
Pinder EM, Rostron AJ, Hellyer TA-O, Ruchaud-Sparagano MH, Scott J, Macfarlane JG, et al. Randomised controlled trial of GM-CSF in critically ill patients with impaired neutrophil phagocytosis. (1468-3296 (Electronic)).
Manjappachar NK, Cuenca JA, Ramírez CM, Hernandez M, Martin P, Reyes MP, et al. Outcomes and Predictors of 28-Day Mortality in Patients With Hematologic Malignancies and Septic Shock Defined by Sepsis-3 Criteria. J Natl Compr Canc Netw. 2022;20(1):45-53.
Mathias B, Szpila BE, Moore FA, Efron PA, Moldawer LL. A Review of GM-CSF Therapy in Sepsis. Medicine (Baltimore). 2015;94(50):e2044.
Mehta HM, Malandra M, Corey SJ. G-CSF and GM-CSF in Neutropenia. J Immunol. 2015;195(4):1341-9.
Cetean S, Căinap C, Constantin AM, Căinap S, Gherman A, Oprean L, et al. The importance of the granulocyte-colony stimulating factor in oncology. Clujul Med. 2015;88(4):468-72.
Hill GR, Koyama M. Cytokines and costimulation in acute graft-versus-host disease. Blood. 2020;136(4):418-28.
Nolt B, Tu F, Wang X, Ha T, Winter R, Williams DL, et al. Lactate and Immunosuppression in Sepsis. Shock. 2018;49(2):120-5.
Dabar G, Harmouche C, Salameh P, Jaber BL, Jamaleddine G, Waked M, et al. Community- and healthcare-associated infections in critically ill patients: a multicenter cohort study. Int J Infect Dis. 2015;37:80-5.
Minasyan H. Sepsis: mechanisms of bacterial injury to the patient. Scand J Trauma Resusc Emerg Med. 2019;27(1):19.
Lam LKM, Murphy S, Kokkinaki D, Venosa A, Sherrill-Mix S, Casu C, et al. DNA binding to TLR9 expressed by red blood cells promotes innate immune activation and anemia. Sci Transl Med. 2021;13(616):eabj1008.
Effenberger-Neidnicht K, Hartmann M. Mechanisms of Hemolysis During Sepsis. Inflammation. 2018;41(5):1569-81.
Agnello L, Giglio RV, Bivona G, Scazzone C, Gambino CM, Iacona A, et al. The Value of a Complete Blood Count (CBC) for Sepsis Diagnosis and Prognosis. Diagnostics (Basel). 2021;11(10).
Assinger A, Schrottmaier WC, Salzmann M, Rayes J. Platelets in Sepsis: An Update on Experimental Models and Clinical Data. Front Immunol. 2019;10:1687.
Harbi GA, Chaari A. Platelets parameters in septic shock: clinical usefulness and prognostic value. Blood Coagul Fibrinolysis. 2020;31(7):421-5.
Cox D. Sepsis - it is all about the platelets. Front Immunol. 2023;14:1210219.
Xie Y, Tian R, Xie H, Jin W, Du J, Huang P, et al. The clinical significance of thrombocytopenia complicating sepsis: A meta-analysis. J Infect. 2019;78(4):323-37.
Claushuis TA, van Vught LA, Scicluna BP, Wiewel MA, Klein Klouwenberg PM, Hoogendijk AJ, et al. Thrombocytopenia is associated with a dysregulated host response in critically ill sepsis patients. Blood. 2016;127(24):3062-72.

Table 1 Study cohort characteristics and laboratory indicators at septic shock diagnosis

Characteristics	GM-CSF (n=66)	No GM-CSF (n=134)	P
Age, year	6.8 (2.4, 11.8)	3.5 (0.8, 8.9)	0.001
Male, n (%)	44 (66.7)	85 (63.4)	0.653
Underlying diseases, n (%)
≥1 obvious underlying disease	65 (98.5)	95 (70.9)	<0.001
Hematologic/oncologic diseases	44 (66.7)	20 (14.9)	<0.001
Gene deficiency	17 (25.8)	34 (25.4)	0.953
Autoimmune disease	4 (6.1)	3 (2.2)	0.330
HSCT	18 (27.3)	12 (9.0)	0.001
Nervous system disease	0 (0.0)	15 (11.2)	0.011
Digestive tract disease	0 (0.0)	9 (6.7)	0.073
Congenital heart disease	0 (0.0)	3 (2.2)	0.544
ICU admission source, n (%)			<0.001
Emergency	18 (27.3)	79 (59.0)
Hospital Ward	48 (72.7)	55 (41.0)
Reason for ICU admission, n (%)			0.311
Septic shock	27 (40.9)	65 (48.5)
Other	39 (59.1)	69 (51.5)
PCIS at SS diagnosis, median (IQR)	72.0 (68.0, 80.0)	75.0 (60.0, 80.0)	0.208
Vital signs at SS diagnosis
Temperature (℃), median (IQR)	37.2 (36.5, 38.3)	37.0 (36.5, 37.8)	0.375
Abnormal RR, n (%)	28 (42.4)	57 (42.5)	0.988
Abnormal HR, n (%)	38 (57.6)	76 (56.7)	0.908
Source of infection, n (%)			<0.001
HASS	45 (68.2)	35 (26.1)
CASS	21 (31.8)	99 (73.9)
Positive pathogen detection, n (%)	37 (56.1)	77 (57.5)	0.851
G+	7 (10.6)	23 (17.2)	0.222
G-	24 (36.4)	48 (35.8)	0.940
Virus	5 (7.6)	17 (12.7)	0.398
Fungus	12 (18.2)	11 (8.2)	0.038
Indicators, Median (IQR)
RBC (×10⁹/L)	2.34 (2.05, 2.97)	3.45 (2.69, 4.05)	<0.001
Hb (g/L), mean (SD)	72.7 (19.8)	93.8 (27.8)	<0.001
WBC (×10⁹/L)	0.59 (0.19, 1.41)	9.70 (5.64, 15.00)	<0.001
PLT (×10⁹/L)	17 (5, 32)	135 (75, 213)	<0.001
N (×10⁹/L)	0.00 (0.00, 0.57)	7.13 (3.83, 13.05)	<0.001
M (×10⁹/L)	0.00 (0.00, 0.10)	0.56 (0.22, 1.29)	<0.001
L (×10⁹/L)	0.00 (0.00, 0.47)	1.40 (0.60, 3.00)	<0.001
CRP (mg/L)	119.0 (23.0, 141.8)	47.9 (12.0, 125.0)	0.012
LAC (mmol/L)	4.7 (2.1, 9.2)	2.9 (1.5, 5.3)	0.503
PH	7.33 (7.06, 7.46)	7.39 (7.26, 7.49)	0.471
GLU (mmol/L)	7.7 (6.4, 13.9)	8.9 (6.8, 12.3)	0.848
Na (mmol/L)	136 (132, 140)	135 (131, 140)	0.253
K (mmol/L)	3.4 (2.9, 4.2)	3.1 (2.6, 5.1)	0.445
INR	1.45 (1.24, 1.78)	1.50 (1.29, 1.94)	0.164
PT (s)	17.8 (15.9, 20.2)	18.3 (16.1, 22.5)	0.180
APTT (s)	41.9 (38.2, 49.8)	40.4 (34.2, 55.6)	0.746
d-Dimer (mg/L)	1.84 (1.14, 7.43)	3.40 (1.62, 7.19)	0.047
ALB (g/L)	28.2 (24.5, 31.9)	29.8 (26.9, 33.1)	0.037
TBIL (μmol/L)	28.5 (12.5, 100.1)	12.5 (6.7, 44.7)	0.014
ALT (U/L)	38.4 (14.6, 114.2)	37.6 (14.5, 129.2)	0.242
AST (U/L)	108.5 (36.2, 280.6)	77.8 (37.4, 348.6)	0.031
CR (μmol/L)	30.8 (24.0, 56.0)	43.8 (24.0, 84.0)	0.045
Urea (mmol/L)	6.08 (4.75, 8.51)	7.00 (3.81, 10.55)	0.742

HSCT, Hematopoietic stem cell transplant; GM-CSF, Granulocyte-macrophage colony-stimulating factor; PCIS, Pediatric critical illness score; SS, Septic shock; RR, Respiratory rate; HR, Heart rate; HASS, Hospital-acquired septic shock; CASS, Community-acquired septic shock; RBC, Red blood cell; HB, Hemoglobin; WBC, White blood cell; PLT, Platelet; N, Neutrophil; M, Monocyte; L, Lymphocyte; CRP, C-reactive protein; LAC, Lactic acid concentration; Na, Sodium; K, Potassium; INR, International normalized ratio; PT, Prothrombin time; APTT, Activated partial thromboplastin time; ALB, Albumin; TBIL, Total bilirubin; ALT, Alanine aminotransferase; AST, Aspartate aminotransferase; CR, Creatinine.

Table 2 Antimicrobial and supportive therapies in pediatric patients with septic shock

Characteristics	GM-CSF (n=66)	No GM-CSF (n=134)	P
Types of empirical antimicrobial therapy on day 1, n (%)
Carbapenems	48 (72.7)	95 (70.9)	0.787
Glycopeptides	30 (45.5)	64 (47.8)	0.759
Oxazolidinone	12 (18.2)	21 (15.7)	0.653
Beta-lactamase inhibitors	1 (1.5)	4 (3.0)	0.885
Cephalosporin	13 (19.7)	17 (12.7)	0.192
Aminoglycosides	3 (4.5)	4 (3.0)	0.876
Quinolones	9 (13.6)	13 (9.7)	0.403
Nitroimidazoles	0 (0.0)	7 (5.2)	0.139
Sulfanilamide	13 (19.7)	17 (12.7)	0.192
Glycyl tetracyclines	6 (9.1)	6 (4.5)	0.196
Macrolide antibiotics	1 (1.5)	4 (3.0)	0.885
Polypeptide antibiotic	8 (12.1)	5 (3.7)	0.050
Antiviral drugs, n (%)	18 (27.3)	20 (14.9)	0.036
Antifungal drugs, n (%)	32 (48.5)	36 (26.7)	0.002
Respiratory support	65 (98.5)	133 (93.3)	0.607
Ventilator days, median (IQR)	13.0 (7.0, 22.0)	13.0 (6.0, 36.0)	0.543
Renal replacement therapy, n (%)	23 (34.8)	59 (44.0)	0.214
Transfusions, n (%)	22 (33.3)	57 (42.5)	0.211
Vasoactive agents, n (%)
Norepinephrine	65 (98.5)	115 (85.8)	0.005
Epinephrine	42 (63.6)	66 (49.3)	0.055
Dobutamine	60 (90.9)	108 (80.6)	0.061
Terlipressin	9 (13.6)	17 (12.7)	0.851
Milrinone	12 (18.2)	13 (9.7)	0.088
Vasopressor days, median (IQR)	5.0 (4.0, 8.0)	5.0 (3.0, 8.0)	0.636
Steroids, n (%)	39 (59.1)	62 (46.3)	0.088

GM-CSF, Granulocyte-macrophage colony-stimulating factor.

Table 3 Complications and outcomes in pediatric patients with septic shock

Characteristics	GM-CSF (n=66)	No GM-CSF (n=134)	P
Prognosis
Length of ICU stay, median (IQR)	13.5 (8.0, 26.0)	15.0 (8.0, 36.0)	0.351
28-day mortality, n (%)	39 (59.1)	47 (35.1)	0.001
In-hospital mortality, n (%)	29 (43.9)	35 (26.1)	0.011
Complications, n (%)
Respiratory failure	49 (74.2)	106 (79.1)	0.439
Acute renal failure	16 (24.2）	26 (19.4)	0.429
Liver dysfunction	10 (15.2)	26 (19.4)	0.462
DIC	30 (45.5)	33 (24.6)	0.003
Cerebral dysfunction	9 (13.6)	30 (22.4)	0.142
Digestive failure	11 (16.7)	14 (10.4)	0.211
MODS	53 (80.3)	117 (87.3)	0.192

GM-CSF, Granulocyte-macrophage colony-stimulating factor; DIC, Disseminated intravascular coagulation; MODS, Multiple organ dysfunction syndrome.

Table 4 Univariate and multivariate logistic regression analysis of 28-day mortality in children with septic shock

Variable	Univariate logistic regression		Multivariate logistic regression
	OR (95% CI)	P	Model1		Model2		Model3
	OR (95% CI)	P	OR (95% CI)	P	OR (95% CI)	P	OR (95% CI)	P
Age, year	1.003 (0.947-1.064)	0.907	1.028 (0.953-1.109)	0.471	1.061 (0.975-1.155)	0.168	1.059 (0.971-1.155)	0.197
Sex	1.048 (0.584-1.883)	0.874	0.965 (0.484-1.925)	0.919	1.043 (0.494-2.204)	0.912	1.022 (0.474-2.205)	0.956
GM-CSF	2.674 (0.459-4.899)	0.001	1.278 (0.501-3.259)	0.608	0.515 (0.196-1.358)	0.180	0.472 (0.153-1.457)	0.192
HSCT	4.690 (2.055-10.704)	<0.001	3.851 (1.458-10.170)	0.006	3.799 (1.342-10.751)	0.012	3.569 (1.183-10.767)	0.024
PCIS
>80	Ref.		Ref.		Ref.		Ref.
71-80	2.277 (1.039-4.989)	0.040	1.947 (0.737-5.140)	0.179	2.259 (0.792-6.444)	0.127	2.018 (0.694-5.874)	0.198
≤70	2.110 (0.964-4.618)	0.062	1.325 (0.491-3.580)	0.578	1.399 (0.487-4.015)	0.533	1.208 (0.395-3.693)	0.740
LAC, mmol/L
≤2.3	Ref.		Ref.		Ref.		Ref.
>2.3	4.849 (2.581-9.110)	<0.001	4.594 (2.256-9.357)	<0.001	4.360 (2.049-9.277)	<0.001	4.457 (2.046-9.708)	<0.001
WBC, 10⁹/L
P₂₅-P₇₅(0.81-13.42)	Ref.		Ref.		-		Ref.
<P₂₅ (0.81)	2.160 (1.083-4.039)	0.029	1.214 (0.465-3.171)	0.692	-		0.671 (0.229-1.968)	0.468
>P₇₅ (13.42)	0.880 (0.435-1.778)	0.721	1.234 (0.546-2.790)	0.614	-		1.663 (0.667-4.145)	0.275
HASS v.s. CASS	3.546 (1.958-6.420)	<0.001	2.909 (1.382-6.125)	0.005	-	-	2.728 (1.189-6.259)	0.018
CRP, mg/L
≤8	Ref.		Ref.		-	-	Ref.
>8	0.940 (0.245-3.612)	0.929	1.385 (0.510-3.762)	0.523	-	-	1.046 (0.339-3.233)	0.937
RBC, 10⁹/L
≤2.80	3.942 (2.177-7.139)	<0.001	-	-	2.535 (1.158-5.549)	0.020	2.571 (1.147-5.761)	0.022
>2.80	Ref.		-	-	Ref.		Ref.
PLT, 10⁹/L
≤74	7.765 (4.089-14.745)	<0.001	-	-	6.725 (2.844-15.902)	<0.001	7.369 (2.947-18.523)	<0.001
>74	Ref.		-	-	Ref.		Ref.

GM-CSF, Granulocyte-macrophage colony stimulating factor; HSCT, Hematopoietic stem cell transplant; PCIS, Pediatric critical illness score; LAC, Lactic acid concentration; WBC, White blood cell; HASS, Hospital-acquired septic shock; CASS, Community-acquired septic shock; CRP, C-reactive protein; RBC, Red blood cell; PLT, Platelet.

The effect of GM-CSF and predictors of treatment outcome in pediatric septic shock patients

Status:

Version 1

Abstract

Figures

Introduction

Methods

Results

Discussion

Conclusion

Declarations

References

Tables

Status:

Version 1