Of the 140 patients with sepsis, 18 (12.9%) had uncomplicated sepsis, 22 (15.7%) had severe sepsis, and 100 (71.4%, p < 0.05) developed septic shock. As for the main risk factors, 79 patients with sepsis were older than 65 years (56.4%) and all progressed to death, showing that a higher age favored the susceptibility of these patients.

In total, 110 (78.6%) died regardless of the worsening of the disease. As expected, the greater the aggravation of this pathology, the longer the hospital stay. In addition, the greater the severity of sepsis, the greater the exposure to invasive procedures, thus all patients with septic shock underwent some invasive procedure during their stay in the ICU, mainly through a central vascular catheter, urinary catheter, mechanical ventilation or tracheostomy, which may explain the worsening of sepsis (Table 1).

Table 1. Clinical features of patients admitted to the ICU.

Variables	n
Patients admitted to the ICU	200
Patients admitted to the ICU with sepsis	140 (70%)
Male gender	122 (61%)
Average age, years (range)	60.75 (28–89)
Average length of stay in the ICU for patients without sepsis (days)	5
Average length of stay in the ICU for patients with sepsis (days)	15.7
APACHE II score average	35
Type of ICU admission—n (%) *
Medical admission	132 (66%)
Surgical admission	68 (34%)
Infection Site—n (%) *
Pulmonary	91 (65%)
Abdominal	42 (30%)
Urinary bacteremia	7 (5%)
Main Comorbidities—n (%) *
Neoplasms	112 (80%)
Diabetes mellitus	15 (10.7%)
Systemic arterial hypertension (PAH)	13 (9.3%)
Invasive Procedures—n (%) *,#
Urinary catheterization	77 (55%)
Central vascular catheterization	116 (82.9%)
Bladder probe	70 (50%)
Tracheostomy	105 (75%)
Mechanical ventilation	111 (79.3%)

According to the worsening of the disease, bacteria were the main etiological agents (75%), of which Gram-negative bacilli were the most frequent (26%), especially in patients with greater disease severity. From this group, the non-fermenting bacilli (58%) and the Enterobacteriaceae family (52%) stand out. Gram-positive cocci represented 7% of the total of isolated microorganisms, of which 60% were Staphylococcus aureus, 20% Enterococcus sp., and 20% coagulase-negative Staphylococcus. Furthermore, in 67% of cases, other bacteria were also found, such as mycobacteria (69%), Clostridium tetani (19%), and Neisseria meningitidis (12%). Viruses and fungi accounted for 90% of the samples, with COVID-19 being the most frequent (92%), while HIV, H1N1 virus, and Candida albicans yeast were found in 8% of the samples.

Among the microorganisms resistant to the main antimicrobials used in the clinic, Gram-positive bacteria were found in 20% of clinical isolates, of which 50% were Staphylococcus aureus (S. aureus), mainly methicillin-resistant S. aureus (MRSA, 70%). Furthermore, among the coagulase-negative Enterococcus and Staphylococcus (coagulase-negative S.), two clinical isolates of vancomycin-resistant Enterococcus (VRE) and two methicillin-resistant S. coagulase-negative (MRSCON) isolates were identified. Regarding Gram-negative bacteria, 25 clinical isolates were identified, highlighting enterobacteria resistant to third- and fourth-generation cephalosporins (48% of enterobacteria), followed by Pseudomonas aeruginosa (P. aeruginosa) resistant to carbapenems (35% of P. aeruginosa isolates). Acinetobacter baumannii (A. baumannii) resistant to third-generation cephalosporins and carbapenems and Stenotrophomonas maltophilia (S. maltophilia) resistant to sulfamethoxazole and trimethoprim corresponded to 50% and 100% of these species, respectively.

It is noteworthy that 40% (56) of the 140 patients who developed sepsis had the first negative culture results. In these patients, the presence of microorganisms was detected and clinically characterized when sepsis was in full development. Empirical antibiotic treatment was initiated mostly with broad-spectrum drugs such as carbapenems (imipenem, meropenem), third- and fourth-generation cephalosporins, and vancomycin.

The basal level of expression of P2Y12, CD62P, CD41, and CD61 antigens on the surface of platelets showed a significant increase (p < 0.01) between platelets of patients before admission to the ICU and after the development of sepsis (Table 2). This difference was also found in relation to platelets from healthy volunteers (negative control).

In all 100 volunteers who donated healthy platelets, as expected, the highest quantity was that of miR-127 in relation to miR-320a. Only the 140 patients who developed sepsis had the relative quantity of these miRNAs inverted—that is, the quantity of miR-320a was greater than that of miR-127 (Table 2, Figure 1). The relationship between the expression levels of these two miRNAs is shown in the heatmap in Figure 2, made for 20 patients who developed sepsis.

Importantly, the quantitative change in miRNAs was significant (p < 0.01) when comparing the values of patients who developed the three different types of sepsis versus the values of the same patients before sepsis and the values of healthy volunteers. The change in miR-127 and miR-320a also coincided with the development of thrombocytopenia, as the 140 patients who developed sepsis had platelets below 50,000/µL and increased procalcitonin (above 2.0 ng/mL).