Stroke-associated pneumonia (SAP) is one of the most common complications among patients with acute ischemic stroke (AIS), taking place most frequently within the first week of stroke onset (1–3). The incidence of SAP ranges from 6.7 to 37.98% (4–6). Patients with SAP are more likely to have a worse outcome than patients with non-SAP, namely, poor functional prognosis, excessive time in hospital, and an increased risk of disability, 30-day and 1-year mortality (7–9). In addition, a retrospective cohort study including 14,702 patients with AIS found that pneumonia was significantly associated with the development of non-pneumonia medical complications, such as gastrointestinal bleeding, urinary tract infection, and recurrent stroke (10). Several clinical trials have shown that prophylactic use of antibiotics was ineffective to prevent SAP (11, 12).

Numerous studies have discovered various risk factors for SAP such as old age, being male, stroke severity, dysphagia, and diabetes (5, 9, 13). Moreover, researchers created several predictive models with these risk factors (4, 14–17), among which the A2DS2 score [age, atrial fibrillation (AF), dysphagia, sex, and severity] was considered to have a good predictive capacity (18). In a consecutive cohort of 1,569 patients with AIS, Gong et al. proved that the A2DS2 score could effectively predict the development of SAP in the Chinese population (19). However, SAP prediction is still a challenge due to its atypical clinical symptoms and the low accuracy of X-ray images (20). Therefore, it is necessary for us to look for valid predictors to diagnose SAP as early as possible.

It has long been known that there exists a wide range of interplay between inflammation and coagulation. For instance, inflammation spikes immediately after injury. Aggregated platelets and neutrophils release factors that stimulate the coagulation cascade in acute lung disease (21).

The fibrinogen-to-albumin ratio (FAR) which combines coagulation with nutritional status, is a new vital inflammatory biomarker for a variety of diseases, such as cervical cancer, oligodendroglial gliomas, acute coronary syndrome, and stroke (22–25). Zheng et al. found that a high FAR level on admission was highly associated with 3-month mortality and disability in patients with acute lacunar stroke (22). Yet, the link between FAR and the risk of developing SAP in patients after AIS remains unclear. The pneumonia severity index (PSI) is one of the best-known predictors for prognosis, containing 20 variables covering demography, clinical features, physical examination, laboratory examination, and chest radiography (26). A decade after PSI was founded, Aujesky et al. proved that PSI could effectively predict mortality and other adverse outcomes of low-risk patients and provided useful feedback to guide initial treatment (27).

Here, we aimed to explore the relationship between SAP and FAR in a retrospective cohort, which might provide an easily available and economical predictor for SAP.

To the best of our knowledge, this is the first study to explore the role of FAR in the development of SAP. In this retrospective study, we found that coagulation functions differ significantly between patients with SAP and non-SAP. Patients with SAP had a higher level of fibrinogen, PT, aPTT, D-dimer, and lower albumin levels at admission. We also found an independent relationship between higher FAR in the first hours after hospitalization and the development of SAP after adjusting for potential contributory factors. This suggested the possible utility of FAR as a marker for the risk of developing SAP. In addition, high FAR levels correlated with pneumonia severity. Furthermore, our data demonstrated that FAR could effectively predict the incidence of SAP with an optimal cutoff point of 0.0977 that could identify the risk of SAP, with remarkable sensitivity and specificity. Therefore, FAR could be a useful and valid biomarker to identify potential patients with SAP at admission.

Compared with other literature, our study first described the coagulation parameters between SAP and non-SAP and found high FAR was an independent risk for the prevalence of SAP. In our study, 100 (10.7%) patients were diagnosed with SAP, which was consistent with previous reports (6). The study showed age, baseline NIHSS score, AF, dysphagia, and length of hospital stay were contributory factors toward SAP, which was also consistent with previous reports (9). In addition, our study showed that antiplatelet therapy was related to a low risk of SAP. The link has not been studied yet. Studies regarding coagulation functions and SAP are rare. In an animal experiment carried out on models of acute lung injury, researchers found nebulized heparin could reduce pulmonary coagulopathy and inflammation effectively by modulating alveolar macrophages (31). During the COVID-19 pandemic, Bi et al. found that fibrinogen, FAR, and D-dimer were significantly elevated in severe patients with COVID-19 with inferior prognosis and accelerated clinical progression (32). A recent study found high levels of D-dimer could increase the risk of 3-month mortality and death/severe disability in AF-related patients with AIS (33). More evidence indicates that an extensive link exists between coagulation and inflammation. The acute phase of inflammation response leads to excessive activation of the coagulation system, and coagulation also amplifies the inflammation process considerably (34, 35). Despite this, the effect of FAR on acute patients with IS, of whom a high proportion suffer from coagulation dysfunction, had not been studied until now.

As fibrinogen and albumin, two critical factors in the coagulation system that care for biomarkers for nutrition state and inflammation, are delivered by the liver, the combined application of two factors (FAR) could be a more powerful predictor than single biomarker fibrinogen or albumin. An elevated FAR level may be a result of increased fibrinogen levels or/and decreased albumin levels. In this study, the FAR levels of patients with SAP were higher than patients with non-SAP (0.113 vs. 0.087, P < 0.001; Table 1). The FAR level of patients with hemorrhagic transformation (HT) was higher than the patients with non-HT after stroke in a previous study (8.60 vs. 10.29, respectively; Table 1) (36). This may be due to the inflammatory response caused by the occurrence of stroke. When the immune system participates in the ischemic brain processes, stroke-induced immunodepression can protect our body from excessive inflammatory reactions. However, it promotes the risk of concurrent infections and affects the prognosis of stroke patients (37). Nonetheless, the association between FAR and AIS requires further study.

Fibrinogen, known as a 340 kDa glycoprotein, can be deposited within and around damaged tissue zones. In research conducted in Spain, Miranda Acuña et al. found elevated fibrinogen deposition in the brain in a wide range of neurological diseases such as multiple sclerosis (38). However, the underlying mechanisms between fibrinogen and SAP remain unclear. There could be a potential mechanism to explain this association. First, fibrinogen is identified as a ligand for various cell surface receptors. It could serve as an intermediate molecule to boost cell-to-cell adhesion between leukocytes and the endothelium. In this way, fibrinogen can facilitate the migration of leukocytes outside the blood vessel to specific tissues (39, 40), and induce changes in leukocytes function. This can take place through cell movement, phagocytosis, release of chemokines and cytokines, degranulation, and nuclear factor-Kappa B-mediated transcription (41–45). Second, plasma fibrinogen levels increase by 2–3 times in response to inflammation, causing cell aggregation, higher blood viscosity, and an increase in endothelin-1 (46). Moreover, during the fibrinolysis process, plasmin, the major fibrinolytic protease, can degrade extracellular matrix proteins and activates matrix metalloproteinases resulting in tissue damage (47, 48).

Serum albumin levels represent the state of nutrition status of an individual and indicate the presence of inflammation. Inflammation increases capillary permeability and flee of serum albumin, causing the distribution volume of albumin to increase while total albumin mass decreases. Albumin accumulates at inflammation sites so it has been widely used as an ideal drug delivery platform in anti-inflammation therapy (49). Several studies revealed that low serum albumin was considered a high-risk factor for deterioration and poor prognosis in patients with nosocomial infection (50, 51). Growing research evidence showed that albumin was a useful, independent prognostic factor in patients after stroke (52, 53). In a cohort study of 759 patients with AIS followed for 3 months, a poor outcome was independently related to low serum albumin level, ischemic heart disease, and infarction size (53). Based on physiological characteristics of albumin, namely, antioxidant, anti-inflammatory, anticoagulant, and antiplatelet aggregation activity and regulation of microvascular permeability, albumin could be used to predict the pulmonary infection after stroke.

Notably, FAR baseline levels correlated with poststroke pneumonia during hospitalization. It also seems plausible that there may be an association between FAR and pneumonia severity, although inflammation response changes dynamically overtime during AIS. It is interesting that this relationship continued despite the effects of the inflammation process.

There remained some limitations in our study. First, our study was based on a single-center retrospective database and some inevitable subjective selections bias might have existed, reducing the reliability of the investigation. Second, FAR was only recorded once at admission and could not be recorded at other times (beyond 24 h), which limited the analysis of the dynamic association between FAR and SAP. It is necessary to conduct a further longitudinal study to verify the predictive value of FAR measured at multiple times for SAP. Third, our study lacked data on the use of the nasogastric tubes and mechanical ventilation, which were related to a high risk of respiratory infections. Finally, other important blood parameters such as antithrombin and leukocyte were not included in this study.

Despite the limitations, our findings are very meaningful. Our study revealed that high FAR was an independent predictive factor related to a higher risk of SAP in patients with AIS. It was also related to the severity of pneumonia. The study provides new references for further clinical trials to determine whether the levels of FAR may serve as a new marker to select patients with AIS at a higher risk of SAP and thus, to guide targeted treatment at an early stage.

The original contributions presented in the study are included in the article/supplementary material, further inquiries can be directed to the corresponding author/s.

The studies involving human participants were reviewed and approved by the Ethics Committee of First Affiliated Hospital of Wenzhou Medical University. Written informed consent for participation was not required for this study in accordance with the national legislation and the institutional requirements.

GL, MH, JS, ZW, and JH conceived and designed the study. GL and MH interpreted data. GL, MH, and JS wrote the manuscript. GL, JS, and XX prepared figures. GL and JS did the statistical analyses. GL, MH, JS, XX, HL, LQ, HZ, MX, DG, and LY screened and extracted data. ZW, JH, and GH supervised the study. All authors have made an intellectual contribution to the manuscript and approved the submission.

This work was supported by the Projects of Provincial Natural Science Foundation of Zhejiang (no. LY19H090013), the Jiaxing Science and Technology Plan Project (no. 2019AD32181) and Institute of Aging, Key Laboratory of Alzheimer's Disease of Zhejiang Province, Wenzhou Medical University, Wenzhou, Zhejiang, China.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.