Sublingual microcirculation and internal environment changes as early indications of sepsis: A prospective observational study

This study aims to investigate the changes in microcirculation and internal environment before sepsis in patients with infectious diseases.


| BACKG ROU N D
Sepsis is currently described as a life-threatening organ dysfunction caused by an imbalance in the host response due to infection. The Sequential Organ Failure Assessment (SOFA) increased by ≥2 points relative to the baseline, as the basis for the diagnosis of sepsis 1,2 This definition is currently standardized and widely accepted. Several studies claim that the SOFA score is more complete, but too extensive, relies more on laboratory tests, and typically difficult to apply outside the ICU. 3,4 Using this method to diagnose sepsis may reduce diagnostic sensitivity in the very early stages of the disease, which is not conducive to the early diagnosis of sepsis, and may delay the initiation of treatment [3][4][5][6] This is because, while the sepsis 3.0 diagnostic criteria can better explain the pathophysiological results following this disease, they are still unable to effectively identify persons at high risk of developing sepsis, primarily due to acute infection. Delay in diagnosis and treatment has been associated with a poor prognosis for patients with sepsis, and early treatment significantly improves the survival rate of patients. [7][8][9] The pathophysiology of sepsis is thought to be closely related to the uncontrolled inflammatory response of the host induced by pathogenic microorganisms 10,11 Microcirculation disorder is crucial during this process. Pathogenic microorganisms can cause the host to produce a variety of inflammatory mediators. These inflammatory mediators, along with microorganisms and their components, activate leukocytes, endothelial cells, and PLT, causing the excessive activation of coagulation and damage to the physiological anticoagulant pathway, thus promoting microvascular thrombosis. [12][13][14][15] Microcirculation thrombosis is associated with hyperfibrinolysis, [15][16][17] which further aggravates coagulation function abnormalities and eventually leads to microcirculation disorder. The persistence of microcirculation disorder causes tissue hypoperfusion, which increases cell hypoxia and anaerobic fermentation, as well as the associated internal environment disorder reflected by changes in base excess (BE) and lactate (Lac) and tissue and organ injury [18][19][20][21] This microcirculation disorder particularly occurs earlier than the increase in the SOFA score observed in patients with organ dysfunction following trauma and heart surgery 22,23 We hypothesized that using microcirculation disorder and changes in internal environmental parameters as a diagnostic index is superior to the current sepsis diagnosis method of organ injury selection. This can also help to identify and diagnose sepsis earlier, hence, the motivation to conduct this clinical experiment.

| Trial design and setting
This single-center prospective observational study took place from January 1, 2021, to September 30, 2021. Patients with infectious diseases were enrolled in the study at the Emergency Department(ED), Beijing Chao-Yang Hospital, affiliated with Capital Medical University. Patients were included in the cohort based on the inclusion and exclusion criteria, and were assessed daily for changes in SOFA scores. According to Sepsis 3.0, patients were classified into two groups based on their condition 5 days after admission: the sepsis and the non-sepsis groups. This study aims to determine the changes in sublingual microcirculation and the internal environment before sepsis.
The treatments received by the patients were followed the requirements for clinical routine. Once the patient is diagnosed with sepsis, liquid therapy, antibiotic therapy, and vasopressors were given according to Surviving Sepsis Campaign.

| Exclusion criteria
1. Patients with a primary disease that required surgical intervention.
2. Patients who were suffering from diseases with an abnormal anticoagulant mechanism.
3. Patients with long-term anticoagulants use.

| Observation items
The observation items involved were Gender, age, basic disease his-

| Specimen collection
The patients signed an informed consent and completed the relevant laboratory examinations, including blood routine (leukocyte, neutrophil, hemoglobin, and PLT count, etc.), PCT, coagulation and Ddimer, blood gas analysis, and monitored sublingual microcirculation

| Image acquisition and analysis of sublingual microcirculation
The MicroSee T200 ( SPSS 23.0 software was used to analyze the data, and the normal distribution measurement data were expressed as Mean ± SD. The chi-square test was used for counting data. A two-way repeatedmeasures ANOVA was used to compare the before and after data between the two groups. Independent sample t-test was used for inter-group comparison. Paired sample t-test was used to compare the data at different time points within the same group. The variables with a statistical difference in univariate analysis were further analyzed by logistic regression. Receiver operating characteristics (ROC) curves were used to test diagnostic sensitivity and specificity. p < .05 was considered to be statistically significant. Graphpad Prism 9.4 software was employed for mapping.

| Patient characteristics
A total of 122 patients participated in this study, of which 21 patients withdrew from the study due to failure to complete 48 h related monitoring or failure to follow-up on whether sepsis occurred. In total, 101 patients completed the study, of which 46 patients had an increase in SOFA score ≥2 from the baseline, and were diagnosed as sepsis within 5 days after admission, according to the sepsis 3.0 diagnostic criteria ( Figure 1). Sepsis occurred on 2.20 ± 0.54 days after admission. No significant difference was reported for age, gender, basic disease history, primary infection site, PCT, leukocyte count, hemoglobin, creatinine, blood urea nitrogen, or SOFA score between the two groups at admission. As COVID-19 has been well-controlled in China, especially in general hospitals, no one positive for COVID-19 participated in this study.
The difference in 28-day mortality between the sepsis and nonsepsis groups was statistically significant (p < .05). See Table 1 for details.

| Sublingual microcirculation
A two-way repeated-measures ANOVA test indicated a significant difference in MFI (F = 36.34, p = .00), FHI (F = 50.55, p = .00), and PPV (F = 64.13, p = .00) between the two groups. A significant difference was also reported in MFI, FHI, and PPV between the sepsis and non-sepsis groups at admission, 24 and 48 h (p < .05). MFI and PPV were significantly lower, while FHI was significantly higher in the sepsis patients. However, no significant difference was reported in terms of the change value and change rate of MFI, PPV, and FHI between the two groups. The parameters of sublingual microcirculation at 24 and 48 h in the two groups did not significantly improve or deteriorate compared with those at admission. See Table 2 and

| PLT, Lac, D-dimer, BE, SOFA
A two-way repeated-measures ANOVA test showed that the PLT of the two groups was significantly different (F = 27.45, p = .00). The PLT in the sepsis group was significantly lower than that in the nonsepsis group at admission, 24 and 48 h (p < .05). Compared with the time of admission, the PLT of the two groups decreased dynamically, especially in the sepsis group (p < .05). The 24 h PLT change value, 24 h PLT change rate, 48 h PLT change value, and 48 h PLT change rate in sepsis group were significantly higher than that in the nonsepsis group (p < .05). See Table 3 and Figure 2 for more details.
The Lac of the two groups was significantly different (F = 37.19, p = .00). The Lac in the sepsis group was significantly higher at admission, 24 and 48 h (p < .05). No significant difference was reported for the Lac change value and change rate between the two groups.
See Table 3 and Figure 2 for more details.
A significant difference was reported in the D-dimer between the two groups (F = 5.59, p = .02). The D-dimer in the sepsis group was significantly higher at admission, 48 h after admission (p < .05), but not 24 h after admission (p = .05). No significant difference was reported for the D-dimer change value and change rate between the two groups. See Table 3 and Figure 2 for more details.
The BE of the two groups was significantly different (F = 4.78, p = .03). There was no significant difference in BE between the two groups at admission, but the BE in the sepsis group was significantly lower than that in non-sepsis group 24 and 48 h after admission. The BE change value and change rate between the two groups reported no significant difference. See Table 3 and Figure 2 for more details.
There was no significant difference in the SOFA score between the two groups at admission. The SOFA score of the sepsis group increased significantly after admission (p < .05). See Table 3 and Figure 2 for more details.

| Independent risk factors for sepsis
The PLT, Lac, D-dimer, MFI, FHI, and PPV were statistically different at admission; this was further analyzed by logistic regression.
Decreased levels of MFI and PLT were independent risk factors for sepsis (p < .05). See Table 4 for more details.

| Early warning indicators of sepsis
The ROC curve was used to determine whether sepsis occurred within 5 days after admission. The 24 h PLT change rate, 24 h PLT change value, PLT, MFI, FHI, and PPV at admission were found to be useful as early warning indicators of sepsis. See Table 5 and Figure 3 for more details.  In this study, when compared to admission, the PLT of the two groups significantly decreased. In comparison with non-sepsis patients, sepsis patients had a more severe microcirculation disorder and dynamic decrease in PLT, and their SOFA score significantly increased after admission. This finding implies that when acute infection develops, a sustained inflammatory reaction can result in an increase PLT consumption and a decrease in PLT count. [31][32][33] Regardless of whether sepsis is ultimately diagnosed, a decrease in PLT count can be detected, but the decrease in PLT is more obvious in the sepsis group. This is because sepsis has more serious vascular endothelial injury caused by excessive inflammatory reaction, especially the abnormal state of coagulation function characterized by microcirculation disturbance by microthrombosis, which leads to more obvious platelet consumption.

| DISCUSS ION
The ROC curve was used to determine whether sepsis occurred within 5 days after admission. MFI, FHI, and PLT at admission were found to be useful as early warning indicators of sepsis, but their Firstly, regardless of whether sepsis is diagnosed, patients with a severe infection are considered to be in the same group based on the mechanism of microcirculation lesions caused by infection.
Secondly, some non-sepsis patients may have serious microcirculation disorders at admission, but their SOFA score did not gradually This is an important cause of organ dysfunction, but not a determining factor, since the severity of organ dysfunction is determined by several factors, including injury, repair, and compensation. 34 Fourthly, the relevant laboratory indicators of an organ injury are continuous variables, which are artificially scored and summarized into SOFA scores. This leads to insensitivity to minor organ damage that has not yet reached the level that causes changes in the SOFA score. The SOFA score cannot reflect the dynamic relationship between pathogenic factors and organ dysfunction, 3-6 resulting in the limitations in its use.

| CON CLUS I ON AND PROS PEC T
Before sepsis, patients with severe infection had microcirculation disorders and deterioration of internal environmental parameters.
The deterioration of MFI and PLT are independent risk factor for this disease. The early changes in microcirculation and PLT are included in the diagnostic criteria for sepsis, which aids in its early detection and diagnosis. This single-center study requires a larger sample size and a larger population structure to further confirm the above conclusions.

FU N D I N G I N FO R M ATI O N
There is no fund support for this study.

CO N FLI C T O F I NTER E S T S TATEM ENT
There are no conflicts of interest reported by any of the authors.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.