Introduction

Mechanical ventilation is the most important component of intensive care in neuro-ICU patients. The features of mechanical ventilation in patients with stroke are high percentage of invasive ventilation and its prolonged nature. Prolonged mechanical ventilation is associated with pulmonary and extrapulmonary complications including life-threating events [1-3]. Ventilator-associated pneumonia (VAP) and ventilator-associated tracheobronchitis (VAT) are the most common pulmonary infectious complications [4]. Incidence of VAP is high in patients undergoing mechanical ventilation (5-65%) [3, 4]. Risk of VAP increases by 3% daily within 5 days of ventilation, by 2% in the next 5-10 days and then by 1% daily. About 50% of pneumonias occur within 4 days of mechanical ventilation [2, 5]. Attributable mortality among patients with VAP averages about 13%. However, this value is up to 69% in some surgical patients [6]. The prerequisites for VAP include abnormal biomechanics of breathing due to pain syndrome and impaired nervous regulation, aspiration (common in bulbar disorders and impaired consciousness), intubation per se, violation of local and systemic protective mechanisms against colonization and invasion of infectious agents [1, 3]. Chronic obstructive pulmonary disease, obesity and diabetes mellitus are also risk factors of VAP [7, 8]. Pneumonia following mechanical ventilation significantly worsens treatment outcomes. Mortality rate in VAP is still high (30-70%) [5, 7, 8]. VAP is associated with prolonged mechanical ventilation, difficult weaning from ventilator and prolonged ICU-stay [7].

The main causative agents of VAP are Klebsiella pneumoniae (11.5-35.7%), Staphylococcus aureus (13-31.8%), Escherihia coli (6.5-18.2%), Staphylococcus epidermidis (9.1-15%), Pseudomonas aeroginosa (3.5-12.7%), Enterococcus faecalis (5.1-11.8%), Stenotrophomonas maltophilia (2.1-7.9%), etc. The role of Streptococcus pneumoniae in development of VAP is decreased [1-3, 5, 9]. Monocultures and gram-positive flora prevail as causative agents of early VAP. Delayed VAP is characterized by microbial associations with predominant gram-negative pathogens [2, 3, 7, 10, 11].

Importantly, modern multiple-center protocols should clarify the role of infectious complications following mechanical ventilation in patients with severe stroke.

This article is a logical continuation of the first three studies based on RETAS data.

The purpose of the study was to analyze the prevalence and impact of VAT/VAP on the course and outcomes of stroke.

Material and methods

A multiple-center observational clinical trial enrolled 14 hospitals. The study was supported by the All-Russian public organization "Federation of anesthesiologists and reanimatologists" (FAR). The FAR committee for clinical guidelines and multiple-center studies, as well as local ethics committees approved the study protocol.

Inclusion criteria: patients with verified brain stroke, age 18-90 years, need for mechanical ventilation.

Exclusion criteria: pregnancy, histologically confirmed malignancies, cardiovascular diseases (NYHA class III-IV), liver cirrhosis (terminal), chronic kidney disease (CKD) stage V (hemodialysis). Patients were included in the study between 11/01/2017 and 11/01/2019 according to inclusion criteria. We formed the register via filling out a questionnaire using a computer program (State Registration Certificate for a computer program No. 2019619217 dated May 21, 2019) [12].

We studied the prevalence of VAT, VAP, systemic inflammatory response syndrome (SIRS) and sepsis in patients with stroke undergoing mechanical ventilation.

The effect of infectious complications of stroke following mechanical ventilation on duration of ventilation, time to weaning from ventilator and ICU-stay was analyzed. We studied the impact of infectious complications of stroke following mechanical ventilation on the outcomes of disease after 28 days using the GOS scale.

The register included 1,289 patients. Of these, 1144 ones completely met the inclusion criteria. There were 609 men (53.23%).

Statistical analysis was carried out using descriptive statistics in STATISTICA 10.0 software. Categorical data are presented as absolute values and percentages. The Pearson's χ² test was used to analyze between-group differences in mortality. We also used logistic regression to determine the influence of certain factor on the risk of mortality. Results are presented as odds ratios with 95% confidence interval (OR (95% CI)). Statistical comparability of groups by age, gender, severity of stroke at admission and onset of mechanical ventilation was essential. In case of abnormal distribution, data are presented as median (Me), upper (Q1) and lower (Q3) quartiles. The Mann-Whitney test was used to confirm significant between-group differences in severity of stroke. Between-group differences were significant at p-value<0.05.

Results

Infectious complications are common in all patients with stroke. VAT was more common compared to VAP in all types of stroke. Prevalence of VAP and VAT was 13.11 and 24.91%, respectively. SIRS and sepsis were much less common (1.05%) (Table 1).

Table 1. Prevalence of infectious complications in patients with stroke undergoing respiratory support

Note. VAP — ventilator-associated pneumonia, VAT — ventilator-associated tracheobronchitis, SIRS — systemic inflammatory response syndrome.

The most common causative agents of VAP in stroke were K. pneumoniae, A. baumannii and P. aeruginosa (Table 2).

Table 2. Characteristics of causative agents of ventilator-associated pneumonia in stroke

Note. VAP — ventilator-associated pneumonia.

Mixed flora was observed in 39.29% of cases. We found 2 pathogens in 26.98% of patients with VAP, 3 microorganisms in 8.33%, 4 microorganisms in 2.38%, 5 microorganisms in 1.19% and 6 pathogens in 0.40% of patients.

Patients with hemorrhagic and ischemic stroke developed VAT after 1-2 [0; 6] and 4 [0; 5] days, respectively (p=0.0022). VAT significantly increased duration of mechanical ventilation in patients with stroke (4 [2; 7] vs. 10 [5; 17] days, p<0.001), time to weaning from ventilator (0 [0; 2] vs. 0 [0; 2], p=0.003) and ICU-stay (7 [3; 12] vs. 13 [7; 20] days, p<0.001).

VAT influenced the outcomes of stroke in patients with similar severity of this complication. Indeed, VAT reduced the probability of favorable outcomes (GOS grade 4/5) (OR 0.73, 95% CI 0.49; 1.08, p = 0.048). Favorable outcomes in patients with VAT were 1.4 times less probable. Accordingly, VAT increased the probability of unfavorable outcomes (GOS grade 1-3) (OR 1.05, 95% CI 0.99; 1.10, p = 0.048). The risk of adverse outcomes in patients with VAT was 1.05 times higher. Moreover, the risk of non-fatal adverse outcomes was higher in patients with VAT (GOS grade 2/3) (OR 1.46, 95% CI 1.24; 1.72, p<0.001). The probability of favorable outcomes in patients with VAT was 2.6 times lower, and risk of adverse non-fatal outcome was 1.46 times lower.

Most patients with hemorrhagic and ischemic strokes developed VAP after 5 [3; 7] and 4 [3; 5] days, respectively (p=0.02). At the same time, lethal outcomes in patients with VAP developed later (10 [5; 13] vs. 6 [3; 9] days, p<0.001). VAP significant increased duration of mechanical ventilation and ICU-stay in comparable groups (6 [4; 12] vs. 18 [12; 28] days and 12 [10; 20] vs. 20 [14; 32] days, respectively, p<0.001).

VAP influenced the outcomes of stroke in patients with similar severity of this complication. Thus, VAP was associated with higher mortality for the entire sample of patients with stroke (p<0.001) (Figure) and patients with hemorrhagic stroke (p=0.026). VAP was associated with higher risk of non-fatal adverse outcomes (GOS grade 2/3) (OR 1.56, 95% CI 1.32; 1.83, p<0.001). Moreover, the risk of adverse outcomes in patients with VAP was 1.56 times higher.

Effect of VAP on mortality in patients with stroke.

Discussion

The modern paradigm of cerebral stroke is such that it is advisable to consider this event as a disease of the whole organism. Along with obvious primacy of damage to the central nervous system followed by acute cerebral insufficiency in patients with severe stroke, extracerebral syndromes and complications of stroke are also of particular importance [13]. Cerebrovisceral syndromes include cerebrocardial, cerebroabdominal and cerebropulmonary syndromes, as well as neurogenic bladder and stress hyperglycemia [13–15]. Cerebropulmonary syndrome accompanied by severe acute respiratory distress syndrome occurs, in particular, in severe stroke. This complication is associated with poor prognosis.

We should also consider infectious extracerebral complications. Indeed, extracerebral complications of stroke make up 30% of all unfavorable outcomes [1, 4]. Acute cerebral insufficiency absolutely dominates in acute period of stroke, while extracerebral causes of mortality predominate after 2 weeks [16]. Our data confirm that ventilator-associated infectious complications dominate after the acute period of stroke. According to some studies, mean incidence of VAP is 27% among all patients undergoing prolonged mechanical ventilation. This value is almost the same over the past 20 years [17, 18].

The main causative agents of VAP are K. pneumoniae, A. baumanni and P. aeruginosa. S. aureus and P. mirabilis are also common. These findings confirm no specific bacterial landscape in these patients. Causative agents of nosocomial infections widespread among various groups of ICU patients are typical for VAT and VAP in severe stroke. Mixed strains and high resistance to various antibiotics are common. Importantly, gram-negative flora prevails throughout the recent years (Table 3). The earlier studies emphasized that gram-positive pathogens were more common [10]. Microbial associations make up a significant proportion too.

Table 3. Pathogens of VAP according to various studies

Note. VAP — ventilator-associated pneumonia.

Ventilator-associated infections have a tremendous impact on all aspects of the course and outcomes of disease. Patients with VAP and VAT require prolonged mechanical ventilation. Moreover, they are poorly weaned from ventilator. These features increase ICU-stay in patients with stroke. These data are fully consistent with international and national studies [5, 7, 9, 10, 17–19]. Importantly, we confirmed the role of ventilator-associated infections in stroke outcomes. Moreover, if VAT influences the likelihood of positive outcomes of disease, then VAP is associated with high risk of mortality. Thus, study results confirm the literature data in these aspects too [1].

Thus, we can conclude that ventilator-associated respiratory infections as a part of nosocomial infections influence mortality after acute period of stroke. Along with acute cerebral insufficiency, this problem is extremely important for patients with brain damage and requires intensive measures against infectious complications of stroke.

Conclusion

1. The prevalence of ventilator-associated pneumonia and ventilator-associated tracheobronchitis was 13.11 and 24.91%, respectively. The most common causative agents of ventilator-associated pneumonia in patients with stroke were Klebsiella pneumonia (53.12%), Acinetobacter baumannii (38.89%) and Pseudomonas aeruginosa (15.62%). These findings correspond to general trend of dominant gram-negative flora in nosocomial pneumonia.

2. Ventilator-associated tracheobronchitis increased duration of respiratory support (p<0.001), weaning from ventilator (p=0.003) and ICU-stay in patients with stroke (p<0.001). VAT also reduced the probability of favorable outcomes (p = 0.048).

3. Ventilator-associated pneumonia increased duration of mechanical ventilation (p<0.001), ICU-stay (p<0.001) and risk of mortality (p<0.001) in patients with stroke.

Author contribution

All authors equally participated in development of the concept of the study, collection and analysis of data, writing and editing the manuscript.

The authors declare no conflicts of interest.