Microbiology and Outcomes of Institutionalized Patients with Stroke-associated Pneumonia: A Prospective Observational Study

Background: The attributable mortality and microbial etiology of stroke-associated pneumonia (SAP) vary among different studies. We intednd to determine the microbiology and outcomes of SAP in lower respiratory tract (LRT) for patients with invasive mechanical ventilation (MV). Methods: In this prospective observational study, included patients were divided into SAP and non-SAP based on comprehensive analysis of symptom, imaging and laboratory results. Baseline characteristic, clinical characteristic, microbiology and outcomes were recorded and evaluated. Results: Of 185 patients, 41.6% developed SAP after onset of stroke, and they had lower proportion of non-smoker (p=0.016), lower GCS score (p<0.001), higher serum CRP score (p<0.001) at ICU admission, and higher proportion of males (p<0.001) and hypertension (p=0.018) than patients with non-SAP. Gram-negative aerobic bacilli were the predominant organisms isolated (78%), followed by Gram-positive aerobic cocci (29.9%). Out of 19 isolated pathogens, main pathogens included K. pneumoniae, S. aureus, H. inuenzae, A. baumannii, P. aeruginosa, E. aerogens, Serratia marcescens, and Burkholderia cepacia. SAP signicantly prolonged length of MV (p<0.001) and duration of ICU stays (p<0.001), shorten MV-free days by 28 (p<0.001), caused elevated vasopressor application (p=0.002) and 60-day mortality (p=0.001). Conclusion: Microbiology of SAP is similar to early-phase HAP and VAP. SAP signicantly prolongs duration of MV and lengths of ICU stays, but also markedly increase 60-day mortality.


Background
Pneumonia occurring within 7 days to onset of stroke was de ned as SAP, and it was associated with prolonged hospital stays and poor prognosis [1][2][3]. Morbidity and attributable mortality of SAP vary among different studies, and the pathophysiology of SAP are mainly explained by aspiration and strokeinduced immunodepression [1]. Risk factors included dysphagia, impaired consciousness and ineffective cough re ex, as they impaired LRT ability of eliminating microbes from oropharyngeal contents and allow pathogens enter the lung [4,5]. Diverse communities of pulmonary microbiota have been discovered, thus dysbiosis was considered related to impaired pulmonary defenses [4,6]. Besides, the alternation of oral microbial diversity was not only observed on murine pneumonia model, but also among the patients' population of post-stroke and ventilation-associated pneumonia (VAP), inspiring us new directions investigating pathogenesis [4,[7][8][9][10]. In recent studies, elevated bronchoalveolar lavage (BAL) amylase level was associated with high risk of aspiration and positive-culture, while serum procalcitonin (PCT) was associated with prognosis of aspiration pneumonia (AP) [4,5,[11][12][13]. However, diagnosis of SAP still relies on clinical features, as lack of gold standard biomarkers. According to current epidemiology of SAP, Staphylococcus aureus (S. aureus), Klebsiella pneumoniae (K. pneumoniae), Pseudomonas aeruginosa (P. aeruginosa), Escherichia coli (E. coli), Streptococcus pneumoniae (S. pneumoniae) and Haemophilus in uenzae (H. in uenzae) are considered as major pathogens, which is of high similarity with microbiology of hospital acquired pneumonia (HAP) or community acquired pneumonia (CAP) [1,2,[14][15][16]. Antibiotic prophylaxis has different effects in decreasing morbidity and mortality owing to distinctive subtypes of pneumonia as well as patient population [6,17,18]. Whether apply antibiotic prophylaxis or not and the choice of antibiotics remain controversial, and MV is also a risk factor of SAP, thus investigating microbiology of SAP of critically ill patients with MV is of high signi cance. We conduct this observative study in order to identify microbiology in the LRT and clinical features on such a strictly strati ed population, to further optimize the converge of antibiotics and nd signi cant medical indicators.

Design and Population
We conducted a prospective single-center observational study in an intensive care unit (ICU) of a Zhejiang University a liated hospital between January 2020 and December 2020. The study complied with the current version of the Helsinki Declaration and appropriate clinical practice guidelines. The trial was approved by ethics committee of our hospital (2019 Ethical Review No.343), and registered in Chinese Clinical Trial Rsgistry (ChiCTR2000028849). Written informed consent was obtained from all of the included patients or the next of kin. Criteria for enrollment included ICU inpatient ≥ 18, stroke onset within 72 hours, invasive MV at least for 24 hours. Acute ischemic stroke was de ned as acute onset, focal neurological de cits or panfacial nerve dysfunction, presence of responsible lesion on imaging or duration of symptoms or signs for at least 24 hours, exclusion of non-vascular causes and cerebral hemorrhage. Acute hemorrhagic stroke was de ned as acute onset, focal neurological de cit symptoms often accompanied by headache, vomiting, elevated blood pressure and varying degrees of disturbance of consciousness, head imaging showed bleeding lesions, exclusion of non-vascular cerebral etiology.
SAP was de ned as new or progressive pulmonary in ltrative lesions after stroke with at least two clinical symptoms, body temperature ≥ 38°C, new cough, expectoration, purulent discharge, with or without chest pain, signs of lung consolidation, and/ or wet rales, leukocytosis (> 10,000/mm 3 ) or leukopenia (< 4,000/mm 3 ) with or without nuclear shift to the left. Exclusion criteria were infectious diseases (including pneumonia) within 3 months before the onset of stroke, antibiotics use within 3 months, comorbidities such as chronic obstructive pulmonary disease (COPD), interstitial lung disease, lung tumor, atelectasis, pulmonary edema, pulmonary embolism and autoimmunity diseases, medical history of dysphagia or gastroesophageal re ux, pregnant, pulmonary imaging did not accord with the manifestations of pneumonia after discussion by the research team, unclear medical history. Participants can withdraw from the study when they or reject to continue to take part in the study without any speci c reason.

Procedures and Statistical Analysis
All enrolled patients were followed prospectively and managed according to the following protocol.
Characteristic such as age, sex, type of stroke, previous medical history, time from onset of stroke to intubation as well as smoking status, and baseline data at ICU admission including Glasgow Coma Scale (GCS) score, white cell count (WBC), PCT, CRP were collected and recorded. Besides, outcomes such as duration of MV, duration of MV-free days by 28, duration of ICU stay, duration of hospital stay, vasopressor during ICU stay 60-day mortality were also observed and recorded. Presence of chest imaging examination indicted pulmonary in ammation when ICU admission, specimen should be obtained within 2 hours to ICU admission. Otherwise, continuous surveillance based on clinical symptoms were need, and chest imaging examination would be applied to these patients if symptoms appeared within 48 hours. Once positive imaging evidence supported in ammation, sample should be obtained as soon as possible. Quantitative analysis of berscope or semi-quantitative analysis of suction in endotracheal tube are used to collect specimens. The process of obtaining specimen, storage in sterile container and sending to laboratory should be nished within 2 hours (at room temperature). All specimens should be processed as soon as possible to ensure the activity of pathogenic bacteria, and strictly follow standard inspection procedures such as inoculation, culture, smear, staining and observation. The sputum samples were also inoculated in fungal culture medium. The automatic bacterial identi cation system (Merieux VITEK 2 Compact) was used for strain identi cation. After sample collection, empirical antibiotic therapy was started according to local epidemiology and antibiotic regimen would be modi ed if needed based on the bacterial culture results. Patients were monitored daily until discharge from the hospital or death after the diagnosis of pneumonia. Included patients were divided into SAP and non-SAP based on comprehensive analysis of symptom, imaging and laboratory results. Baseline charateritics and outcomes were analyzed and described as mean ± SD, median (Interquartile range, IQR), percentage as appropriate. Contiunous variables were compared by Student's t test if they were normally distributed, otherwise, were compared by Wilcoxon Mann-Whitney U test. Pearson c2 test or Fisher exact test were applied to categorical variables to compare differences between SAP and non-SAP. All tests were 2-tailed and statistical signi cance was determined at an α a level of 0.05. Statistical analyses were performed with the SPSS 20.0.

Comparison of Patients With SAP and With Non-SAP
A total 392 patients with acute stroke were admitted at ICU during the study, 207 were excluded according to inclusion and exclusion criteria. Finally, 185 patients were eligible for further analysis and those were divided into two groups as SAP (n = 77) and non-SAP (n = 108) according to the microbiological results of sputum samples (Fig. 1). No participants were withdrawn from the trial, and all included patients completed the study protocol and assessment of the main outcomes. Comparing to patients with non-SAP, patients with SAP had lower GCS score (median: 8 vs 13, IQR: 4.25-10 vs 8-15, p < 0.001), higher serum CRP score (median:10.6 vs 5.2, IQR:4.6-30.5 vs 1.8-13.4, p < 0.001) at ICU admission, higher proportion of males (62% vs 36.1%, p < 0.001) and higher proportion of hypertension (58.4% vs 40.7%, p = 0.018). Besides, comparing to non-SAP group, more patients in SAP group had smoking history (45.5% vs 25.9%, P = 0.016) ( Table 1). No other baseline characteristics differed signi cantly between two groups.   Antimicrobial Therapy of Post-stroke Patients With SAP or Non-SAP In terms of antibiotic use in patients with SAP, post-operative application predominated, followed by preoperative and post-operative antibiotic coverage. As for antimicrobial therapy in patients with non-SAP, post-operative application predominated, followed by without antibiotic coverage, then pre-operative and post-operative antibiotic coverage (Table S1). The difference of the medication regimen of antibiotics between the two groups was statistically signi cant (p < 0.001). Antibiotics use included cefuroxime, piperacillin-tazobactam, clindamycin, cefotaxime, cefoperazone-sulbactam, latamoxef and ceftriaxone. Besides, microbiology of patients with SAP and the corresponding antibiotics regimen were illustrated in the In addition, signi cant difference in 60-day mortality was also found between patients with SAP (n = 17, 22%) and patients with non-SAP (n = 6, 5.6%, p = 0.001) ( Table 4).

Discussion
This was the rst survey investigating microbiology of SAP in LRT for post-stroke patients treated with invasive MV. In our study, the main attributed pathogens were K. pneumoniae (39%), S. aureus (26%), H. in uenzae (10.4%), A. baumannii (7.8%), P. aeruginosa (6.5%), E. aerogens (5.2%), Serratia marcescens (5.2%), Burkholderia cepacia (5.2%), similar to the microbiology of culture-positive pneumonia for hospitalized neurologic patients and early-onset VAP [19][20][21]. Besides, a systemic review involving 7968 patients with acute stroke indicted that the commonly isolated organisms of SAP included K. pneumoniae (12.8%), E. coli (9%), S. aureus (10.1%), P. aeruginosa (6%), A. baumanii (4.6%) and S. pneumoniae (3.5%) [22]. As for the type of pathogens, our conclusion was of essential agreement with the systemic review, however, the speci c frequency for each pathogen calculated by different studies varied. This difference might be explained by many divergences among studies including heterogeneity of patients' population, heterogeneity of epidemics in different region, heterogeneity of clinical environment, whether treated with invasive treatment or prophylactic antibiotic therapy (PAT) and so on. For instance, this systemic analysis pointed out that studies whose patients were at relatively higher-risk such as dysphagic or ICU admission had a high proportion of Gram-negative aerobic bacilli and S. aureus, which is consistent with ndings in our study [22].
Critically ill patients deserve more attention, as they have many risk factors of SAP, including hypertension, dysphagia, MV, male, a lower consciousness level and bedridden [23]. In our study, hemorrhagic stroke was the main type of stroke. Actually, ischemic stroke is the more common type of stroke, and most researchers always recruited patients from stroke units (ischemic stroke predominated) or directly selected patients with ischemic stroke [22,24,25]. The difference between ischemic stroke and hemorrhagic stroke was lying on many aspects including mechanism, symptom as well as prognosis, whereas whether these divergences would further impact SAP are still unclear. For example, strokeinduced immunodepression was found correlated with mechanism and morbidity risk of SAP for patients with ischemic stroke [26,27]. However, whether immunodepression induced by hemorrhagic stroke is identical to ischemic stroke still uncertain. In our prospective cohort study, serum CRP and PCT, increased vigorously comparing with the baseline level at ICU admission. Serum CRP and PCT were considered to be independently associated with SAP in some studies, but not others. The diagnostic value of serum CRP and PCT in discrimination of SAP was not de nite yet [28]. Improvement of diagnostic sensitivity and speci city need comprehensive consideration of symptoms and biomarkers. For afebrile patients with chest ndings and leucocytosis or new radiological chest in ltrates, Kalra et al highly recommended CRP ≥ 30 mg/L as a supplement to diagnosed SAP [29]. This strategy remedied missed diagnosis owing to afebrile to some extent. In our study, CRP at ICU admission elevated mildly in SAP group (10.6 vs 5.2 mg/L, p < 0.001) than non-SAP, which might be a somatic response to acute stroke or correlate with stroke-induced immunodepression. However, whether such CRP change in early phase was a valuable sign helping recognize potential patient with SAP, are still unknown. In addition, dynamic change process of biomarkers (eg. CRP and PCT) re ect the progression and prognosis of disease to some extent. PCT were considered to have higher diagnostic and prognostic accuracy of SAP than CRP in recent studies [30,31]. Interestingly, PCT-guided antimicrobial strategies in ICU attracted a lot attention, and PCT-guided cessation of antibiotics were found to reduce both antibiotic exposure and mortality [32,33]. PCT might be a good indicator re ecting progression of SAP and guiding clinical medication. More studies should be designed to clarify the details about role of the serum CRP and PCT in SAP. We found patients with SAP spent more time on MV and staying in ICU, were applied more vasopressors, and had higher 60-day mortality than patients with non-SAP. This conclusion is consistent with the ndings in previous studies, in which SAP was associated with poor functional outcome and mortality [34,35]. It is really important to gure out how to recognize potential patients as early as possible and how to take active strategies to avoid the occurrence of SAP. Some patients were applied antibiotics before operation because they indeed faced high-risk of infection and had a poor physical status in our study. Although antibiotics use was discouraged to prevent SAP  [18,37]. The former found PAT could reduce incidence of infection, while the latter illustrated that PAT did not reduce the frequency of pneumonia. The consistent conclusion in these two high-quality studies was that PAT did not improve functional outcome at 3 months. The above ndings suggested PAT was only capable of preventing urinary tract infections, not SAP [39]. In summary, in terms of bactericidal effect of antibiotics, no antibiotics have been found to have bene cial effects in preventing SAP so far. However, speci c antibiotic protocols in each study succumb to local hygiene policies based on local epidemiology, and in uenced by local empiric administration habits. Several factors such as type of antibiotics, time point of administration, dose, mode of administration, type of population will impact the nal effects of antibiotics. We haven't found the positive effects of PAT in SAP so far, probably because we didn't choose the optimal antibiotics strategies in previous studies. Another point attracted our attention is the neuroprotective effects of some antibiotics. Previously, ceftriaxone was found to signi cantly reduce acute stroke mortality in animal models, and these positive results might be due to its neuroprotective effects including increasing glutamate uptake and inducing neurotrophins [40]. Recently, Dai et al found that minocycline could attenuate brain injury and iron overload after intracerebral hemorrhage in aged female rats [41]. Besides, Yang et al also demonstrated that minocycline could attenuate acute stroke injury in rat brain by promoting blood-brain barrier remodeling and facilitating neuroprotective phenotype alternative activation of microglia / macrophages. Treatment with minocycline signi cantly reduced levels of proin ammatory factors tumor necrosis factors-α (TNF-α) and interleukin-1β (IL-1β) and increased levels of anti-in ammatory factors transforming growth factor-β (TGF-β) and interleukin-10 (IL-10)[42]. As we all known, neurovascular injury is the primary etiology of SAP, antibiotics would have application value in early phase of stoke if they have neuroprotective effects. Thus, this direction needs more attention and more investigations should be conducted to clarify more details. In summary, it is insu cient to totally negate the effects of PAT in SAP just depending on evidence from current studies.
There are some limitations in our study. Firstly, our study is single-center observational cohort study and the sample size was relatively small. Since microbiology of SAP associate with local epidemiology, our ndings need to be further validated in a larger cohort based on multi-center cooperation. Secondly, our study only focused on patients' short-term outcomes such as duration of ICU stays and 60-day mortality.
Long-term follow-up should continue to gure out the effects of SAP in long-term outcome. Thirdly, as microbiologically con rmed bacterial pneumonia was the focus in our study, determination of infection strictly followed results of sputum bacterial culture, whereas virus detection was not performed in our study. It would be better when broader pathogen (eg. bacterium and viruses) distribution and frequency were gured out. Finally, some patients were applied pre-operative prophylactic antibiotics since they had high-risk of infection and serious status, which might impact the results of bacterial culture. The main antibiotics in our study was cefuroxime, which is a broad-spectrum antibiotic. Thus, the actual morbidity of SAP might be higher than results obtained in our study.

Conclusion
Overall, this is the rst prospective cohort study to investigate the microbiology of SAP in LRT for critically post-stroke patients treated with invasive MV. Those patients are at high risk of developing SAP, especially when patients have the following characteristic: male, have smoking history, hypertension, unconsciousness, elevated CRP at ICU admission etc. We found the main pathogens responsible for SAP in LRT were K. pneumoniae, S. aureus, H. in uenzae, A. baumanni, P. aeruginosa, E. aerogens, Serratia marcescens, Burkholderia cepacia and etc, which is similar to microbiology of early-phase HAP and VAP. Besides, SAP would signi cantly not only prolong duration of MV and duration of ICU stay, but also increase vasopressor during ICU stays as well as mortality.

Declarations
Ethics approval and consent to participate The trial was approved by ethics committee of our hospital (2019 Ethical Review No.343), and registered in Chinese Clinical Trial Rsgistry (ChiCTR2000028849). Written informed consent was obtained from all of the included patients.

Consent for publication
Not applicable Availability of data and materials The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Competing interests
The authors declare that they have no competing interests 42. Yang Y, Salayandia VM, Thompson JF, Yang LY, Estrada EY, Yang Y. Attenuation of acute stroke injury in rat brain by minocycline promotes blood-brain barrier remodeling and alternative microglia/macrophage activation during recovery. J Neuroin ammation. 2015;12:26-6. Figure 1