In this study, we included 32 abdominal infection patients with sepsis that were recruited from the Beijing Ditan Hospital ICU between February 2021 and September 2021. These patients were divided into two groups, the elderly group (≥65 years, 14 patients) and the non-elderly group (<65 years, 18 patients), based on the reported age threshold (Narci et al., 2009) for epidemiological analysis. We compared the neutrophil and monocyte phagocytic function, neutrophil and monocyte respiratory burst function, neutrophil and monocyte tumor necrosis factor- alpha (TNF-α) release function, and neutrophil and monocyte phenotypes between patients and healthy controls (n = 18). We also compared the clinical features, neutrophil and monocyte functions, phenotype, and outcome measures of patients on the first day of diagnosis with abdominal sepsis between the elderly group and the non-elderly group. Patients were tracked until December 2021 to record their infectious events, the development of organ failure, 90-day survival, hospitalization, and ICU stay.

In this study we included patients between the ages of 18 and 80 who were diagnosed with abdominal sepsis during hospitalization. Diagnostic criteria for abdominal sepsis were: 1) deemed septic by the Sequential Organ Failure Assessment (SOFA) score where sepsis is defined by a score increase of 2 or equal to 2. For screening purposes, rapid SOFA scores based on three criteria for neurologic (Glasgow Coma Scale <15), cardiovascular (systolic blood pressure <100 mmHg), and respiratory (respiratory rate >22/min) dysfunction are available clinically identified patients at high risk of sepsis. The diagnostic criteria for septic shock were that vasopressor drug therapy is initiationly and continuously needed to maintain a mean arterial pressure >65 mmHg or a serum lactate level >2 mmol/L, and that the infection site is in the abdominal cavity. 2) There are at least two ways to diagnose abdominal infection: such as history taking, physical examination, laboratory examination, imaging examination and abdominal paracentesis. History taking and physical examination can help diagnose patients with suspected IAI, and laboratory tests can further identify the degree of disease progression. Among them, symptoms and signs such as abdominal pain, rebound pain, cessation of flatus and defecation by anus, and fever all have a strong suggestive effect on abdominal infection, while white blood cell (WBC) count, C-reactive protein (CRP), and procalcitonin (PCT) can predict the degree of infection progression. Imaging examination and paracentesis is important for the localization of the source of infection, the severity of the lesion, and the selection of subsequent treatment measures. 3) Healthy, age- and sex-matched, non-smoking volunteers with no history of liver disease served as healthy controls (HC). HC alcohol intake was <20 g/day and volunteers did not drink alcohol or exercise excessively within 24 h before blood sampling. 4) The above tests were carried out with the informed consent of the patients and their families and were approved by the Medical Ethics Committee.

The following patients were excluded from this study: 1) patients with incomplete clinical data; 2) patients who took antibiotics for at least 5 days due to suspected infection in the past 30 days; 3) patients with a history of any immunodeficiency including malignancy or human immunodeficiency virus (HIV); 4) patients treated with immunosuppressive agents (prednisolone, azathioprine); 5) patients with successful resuscitation after sudden cardiac arrest; and 6) pregnant women.

This study was approved by Committee of Ethics at Beijing Ditan Hospital, Capital Medical University [Department of medical ethics; Beijing Ditan Hospital (2021)-005]. Blood samples, clinical, biochemical data, and physiological data of patients were collected after obtaining fully informed consent. The blood samples of the HC group were from the Health Examination Center of Beijing Ditan Hospital.

We collected 4 ml of peripheral blood from sepsis patients and HC, added it into Ethylene Diamine Tetraacetic Acid (EDTA) anticoagulant tubes and immediately pre-cooled samples to 0°C–4°C. We performed the neutrophil and monocyte function analysis within 2 h of blood being drawn. The solution was aspirated into a centrifuge tube and the erythrocytes were lysed with ammonium chloride lysis buffer (BioLegend, United States) and washed with phosphate buffered saline (PBS) (Corning, United States). Single cells were obtained by adding 10% fetal bovine serum (Gibco, United States) and Roswell Park Memorial Institute (RPMI) 1640 medium containing 1% penicillin and streptomycin (Corning, United States) for in vitro culture and incubated in a carbon dioxide incubator for 30 min. Flow cytometry was performed on cells with the following antibodies: PE anti-CD16 (BioLegend), APC anti-CD14 (BioLegend, United States), PerCP anti-CD45 (BioLegend, United States), FITC anti-CD16 (BioLegend, United States), BV421 anti-TNF-α (BioLegend, United States). Latex Beads (Carboxylate-modified, Yellow-green, Sigma) were used for cellular phagocytic function analysis. The respiratory burst of cells was detected with 2,7-dichlorofluorescein diacetate (DCFH-DA. Sigma). Samples were processed through a BD FACSCanto flow cytometer and analyzed. After obtaining 50,000–100,000 cells per sample, cells were selectively gated by forward scattering versus side scatter dot plots. Flowjo V10 software was used to analyze the flow cytometry data, and GraphPad Prism7 was used to perform the graphing and statistical analysis.

Yellow-green-labeled Latex beads were added to the peripheral blood single cell suspension, incubated in a carbon dioxide incubator at 37°C for 3 h, placed on ice for 10 min to stop phagocytosis, and then washed twice with PBS buffer. Next, the corresponding cell surface fluorescent antibody combination was added (CD16-PE/CD14-APC/CD45-PerCP staining) and allowed to incubate at room temperature for 15 min in the dark. PBS buffer was added after the incubation period for washing. The ratio of Beads+ cells in neutrophils and monocytes was detected and analyzed by flow cytometer to determine the phagocytic function.

Peripheral blood single-cell suspensions from patients and age-matched healthy controls were simultaneously assayed with 0.5 mg/ml DCFH-DA. The peripheral blood single cell suspension was incubated in a carbon dioxide incubator at 37°C for 15 min, fixed with 4% paraformaldehyde for 10 min, washed twice with PBS buffer, and then received a combination of neutrophil fluorescent antibodies (CD16-PE/CD14-APC/CD45-PerCP staining), Samples were incubated at room temperature for 15 min in the dark, washed with PBS buffer, and then run through the flow cytometer to analyze the proportion of DCFH+ cells in neutrophils and monocytes, and measure the generation of reactive oxygen species (ROS) in that percentage of cells.

Golgi plug (Thermal Fisher, United States) was added to the peripheral blood single-cell suspension to block the secretion of cytokines. Golgi plug was incubated with cells at 37°C in a carbon dioxide incubator for 3 h, and the neutrophil fluorescent antibody combination (CD16-FITC/CD14-APC/CD45-PerCP staining) was added, incubated at room temperature for 15 min in the dark, and then washed with PBS buffer. Next, we added the intracellular fluorescent staining kit Fix-Perm (BD bioscience, United States) membrane breaking agent, incubated at 4°C for 20 min in the dark, used Perm Wash (Becton Dickinson, United States) to rinse twice, added intracellular protein fluorescent antibody TNF-α-BV421, incubated at 4°C in the dark for 30 min, and, lastly, washed twice with Perm Wash. We used flow cytometry to analyze the ratio of TNF-α+ in neutrophils and monocytes, and measure the function of producing TNF-α.

The data of subjects with normal distribution were expressed in the form of mean ± standard deviation. The difference in unpaired data between two groups was analyzed by independent sample t-test, and the paired data was analyzed by paired sample t-test, Data for more than two groups were analyzed by one-way ANOVA. Data that was not normally distributed was expressed by median and interquartile range. The unpaired data between two groups was analyzed by Mann-Whitney test, and the paired data was analyzed by Wilcoxon matched-pairs signed rank test. For all data, p < 0.05 was considered statistically significant.

Among the 32 recruited patients, 14 (43.8%) were elderly and 18 (56.2%) were non-elderly, with no significant difference in gender between these two groups (p = 0.458). As shown in Table1, 64.3% (9/14) of the elderly patients but only 44.4% (8/18) of the non-elderly patients had underlying diseases (p = 0.308). Septic shock was more common in the elderly group (78.6%) than in the non-elderly group (50%), but no significant difference was detected for this trend (p = 0.098). By comparing the basic vital signs of patients on the first day of diagnosis, it was found that the mean arterial pressure distribution of the elderly group and the non-elderly group was 60 (56–70) mmHg and 73 (64–81) mmHg, respectively, which was significantly different (p = 0.020). In addition, the comparison of routine blood parameters between the two groups demonstrated that the percentage of neutrophils in the elderly group on the first day [88.0% (83.1%–91.7%)] were significantly higher than those in the non-elderly group [83.7% (73.3%)]-87.1%)] (p = 0.017). The monocytes percentage [5.3% (3.7%–7.1%)], however, was significantly lower in the elderly group when compared to the non-elderly group [8.3% (5.0%–13.2%)] (p = 0.040). The neutrophil to lymphocyte ratio (NLR), derived from complete blood cell count, has been shown to be a widely used non-specific inflammatory marker, and recent studies have found that the addition of platelet count to this ratio, (the neutrophil to lymphocyte and platelet ratio, NLPR), is expected to further improve its clinical value. In this study, our results showed that the NLRs of elderly patients [14.07 (9.48–27.40)] were significantly higher than those of non-elderly patients [11.38 (5.08–15.18)] (p = 0.003). Meanwhile, the NLPRs of patients in the elderly group [14.59 (8.70–24.95)] were also higher than those of patients in the non-elderly group [10.59 (7.16–16.66)], but the difference was not statistically significant (p = 0.067). In addition, the red blood cell volume distribution widths in the elderly group [15.4% (13.1%–17.6%)] were significantly lower than those in the non-elderly group [18.2% (14.3%–22.3%) (p = 0.044)].

The comparison of organ function indicators on the first day of diagnosis of sepsis in the two groups showed that the levels of total bilirubin (liver function indicator) in the non-elderly group were higher than those in the elderly group, which were 60.7 (21.0–117.2) mmol/L and 15.5 (9.0–32.3) mmol/L, respectively (p = 0.015). However, the differences in renal function, coagulation function, cardiac function, acid-base electrolyte, and some infection indexes were not found to be statistically significant (p > 0.05) (Table 2).

Neutrophils and monocytes are important components of the innate immune response in sepsis, both of which have phagocytic function. Once phagocytes find and recognize pathogens, phagocytosis will occur, and then phagocytic enzyme activity increases. In order to understand the phagocytic function of patients with abdominal sepsis in different age groups, the phagocytic function of phagocytes collected on the first day of diagnosis of abdominal sepsis was detected by flow cytometry (Figures 1A,B), and compared with that of healthy volunteers (Figure 2). We found that the phagocytic function of both neutrophils and monocytes was elevated in patients with peritoneal sepsis compared to healthy volunteers (p < 0.001). In terms of neutrophils (Figure 2A), the phagocytic function of elderly abdominal sepsis patients was lower than that of non-elderly patients (p > 0.05). Similarly, the phagocytic function of monocytes in the elderly group was lower than that of the non-elderly group (p > 0.05) (Figure 2B).

In order to clarify the changes of respiratory burst function of phagocytes and further explore their actual ability to eliminate pathogenic microorganisms, we detected the production of cellular ROS (Figures 1C,D). The production of ROS by neutrophils was significantly reduced in both elderly and non-elderly patients with abdominal sepsis compared with HC group (Figure 3A) (p < 0.001), but the decrease degree between the elderly group and non-elderly group was not statistically significant (p > 0.05). As seen from Figure 3B, the monocyte ROS production in elderly patients was significantly higher than that of the non-elderly group (p < 0.05), and the production of ROS by monocytes in the non-elderly group was higher than healthy volunteers, however the difference was not statistically significant (p > 0.05). In addition, we employed the mean fluorescence intensity to profile the results of cellular ROS production (Figure 4).

Neutrophils and monocytes are the main effector cells of the innate immune system. During infection, neutrophils eliminate pathogens mainly through phagocytosis and the release of ROS, but their ability to release inflammatory factors is weaker than monocytes. Here we further explored the effect of age on the intracellular TNF-α secretion capacity of neutrophils and monocytes (Figures 1E,F). The positive proportion of monocytes and neutrophils capable of secreting TNF-α was significantly lower in healthy volunteers compared with patients with abdominal sepsis (p < 0.05) (Figure 5). Concerning age, the proportion of neutrophils able to secrete TNF was significantly higher in elderly patients with abdominal sepsis than in nonelderly patients (Figure 5A). However, previous clinical routine blood data showed that the percentage of neutrophils in peripheral blood was significantly higher than that of monocytes, and the percentage of neutrophils in elderly patients was higher, (p < 0.05) (Table 1). The percentage of monocytes in the non-elderly group was significantly higher than that in the elderly group.

As shown in Table 3, the proportion of patients needing mechanical ventilation was higher in the elderly group (71.4%) than that in the non-elderly group (50.0%). The incidence of acute kidney injury (AKI) was higher in the elderly group (50.0%) than that in the non-elderly group (33.3%) and the length of the hospital stay was 40 (19–44) days in the elderly group and 25 (Esper et al., 2006; Mees et al., 2008; Turnbull et al., 2009; Alves-Filho et al., 2010; Gentile et al., 2012; Prakash et al., 2012; Sadaka et al., 2013; Shaw et al., 2013; Chen et al., 2014; Nacionales et al., 2014; Vanzant et al., 2014; Nacionales et al., 2015; Shalova et al., 2015; Singer et al., 2016; Mira et al., 2017; Kumar, 2018; Raymond et al., 2018; Wang et al., 2018; Coakley et al., 2019; Guo et al., 2019; Peerapornratana et al., 2019; Bartsch et al., 2020; Gameiro et al., 2020; Levey et al., 2020; Mae et al., 2020; Shao et al., 2020; Chen et al., 2021; Yu et al., 2021) days in the mostly non-elderly group. Additionally, the 90-day mortality rate was higher in the elderly group (28.6%) than that in the non-elderly group (16.7%), but no significant differences in these outcomes were observed (p > 0.05). We also found that patients in the elderly group spent more time in the ICU [30 (11–41) days] than those in the non-elderly group [21 (12–42) days] (p < 0.05).