Aneurysmal subarachnoid hemorrhage (aSAH) represents a devastating and life-threatening type of hemorrhagic stroke (1–3). Despite the advancements in the current treatment of intracranial aneurysms, including microsurgery and endovascular coil embolization, both disability and mortality rates of aSAH have remained high (4–6).

The aSAH in patients was also found to be accompanied by vasospasm and other non-neurologic complications, including postoperative pneumonia (POP), a central venous catheter (CVC)-associated infections, and urinary tract infection, among others (7), whereas studies have shown that from 13 to 37% of patients with aSAH may have POP impacted after surgical treatment (7–9).

The current research is thus aimed at finding early risk factors for POP, which might help identify high-risk patients for aggressive monitoring and therapeutic interventions.

Numerous risk factors of POP have been identified, most are based on clinical features, including age, aSAH severity, and ventilator use (7, 10, 11). Biomarkers of inflammation, such as the neutrophil-to-lymphocyte ratio (NLR), have recently gained particular interest during the last years (12) with plenty of evidence indicating that the neutrophil-to-albumin ratio (NAR) is an independent risk factor for cancer, sepsis, and in-stent restenosis undergoing both carotid angioplasty and stenting and ST-segment elevation myocardial infarction (STEMI) (13–16).

However, the mechanistic link between NAR and aSAH remains unstudied, which is why we sought to investigate in the present study the relationship between NAR and POP after aSAH, on the one hand, and whether NAR at an early stage may help to identify patients at high risk of POP, on the other (17).

Several patients with aSAH (n = 395) were considered for this retrospective analysis, out of which 78 (19.7%) patients had POP after operative treatment (Table 1). The group with POP was characterized by older age and severe clinical conditions upon admission (higher WFNS grade, mFisher grade, and intraventricular hemorrhage; Table 1). They also proved to have a higher rate of ventilator use and higher neutrophil, lymphocyte, D-dimer, and NLR on admission.

The POP group showed a significantly higher NAR on admission than that of the non-POP group (0.31 [0.25–0.39] vs. 0.23 [0.18–0.28]; p < 0.001; Table 1; Figure 1A). In addition, patients with severe aSAH (WFNS ≥ 3) exhibited significantly higher NAR levels (0.23 [0.22–0.24] vs. 0.34 [0.31–0.37]; p < 0.001) and higher rates of POP than the rest (12.2 vs. 52.0%; p < 0.001) (Figures 1B,C).

A multivariate logistic regression analysis revealed that after adjusting for confounders NAR upon admission was associated with POP [odds ratio (OR) = 1.907; 95% confidence interval (CI), 1.232–2.953; p = 0.004; Table 2]. Age (OR = 1.068; 95%CI, 1.029–1.108; p < 0.001), initial mFisher grade ≥ 3 (OR = 2.512; 95%CI, 1.199–5.264; p = 0.015), initial NLR (OR = 1.078; 95%CI, 1.034–1.123; p < 0.001), and the rate of mechanical ventilator use (OR = 7.680; 95%CI, 2.518–23.430; p < 0.001) remained as independent factors of POP. In the ROC curve analysis used for predicting POP after aSAH, NAR value of 0.27 was determined to be the cut-off value (AUC [95% CI] 0.775 [0.717–0.832]; p < 0.001; Youden's index = 0.41; Figure 2), we also computed the AUC of WFNS grade (95% CI, 0.725 [0.654–0.796]; p < 0.001; Figure 2) and NLR (95% CI, 0.722 [0.663–0.780]; p < 0.001; Figure 2) in order to compare the predictive power between NAR and conventional markers. The results were able to show that the predictive performances of NAR were comparable to WFNS grade (NAR vs. WFNS grade: Z = 1.587, p = 0.112; Figure 2) and NLR (NAR vs. NLR: Z = 1.335, p = 0.182; Figure 2).

Finally, the patients were split into two groups, the high and the low NAR groups, according to the best cut-off value of NAR. Patients in the former NAR group had higher hypertension, higher initial WFNS, higher rates of POP, and a higher rate of mechanical ventilator use than those within the low NAR group (Table 3).

To our knowledge, no other studies have explored the association between NAR and POP after aSAH, in this study, we have found that a higher NAR value on admission was related to POP in patients with aSAH. Interestingly, NAR remained a significant factor of POP after adjusting for traditional risk factors such as age, clinical grade, and a higher rate of mechanical ventilator use.

Postoperative pneumonia is a common condition with potentially severe complications within 30 days after surgical treatment of patients with aSAH, directly affecting the prognosis (8, 21, 22). Although there is a variety of classic predictors of POP, most are still based on clinical parameters such as age and WFNS grade (7), whereas the relevant biomarkers that have been used in recent years to predict the occurrence of POP after aSAH include cerebral lactate and NLR (12, 23). We suppose that hematology inspection data might be more productive in showing objective subtle changes in patients, especially when both hematology indicators were administered in combination.

Previous studies have shown that NAR was an inflammatory marker for cancers, sepsis, in-stent restenosis undergoing carotid angioplasty, and STEMI (13–16), but the underlying mechanisms have remained unclear. Other studies have also shown that neutrophil counts increase as part of the post-stroke immunodepression phenomenon (24–26), which is activated by the sympathetic nervous system and the hypothalamic–pituitary–adrenal axis (27, 28). Not only is serum albumin, the most abundant antioxidant in the living body, but it is also an inflammatory biomarker that is involved in systemic inflammation through high levels of pro-inflammatory cytokines and growth factors (29–31).

Moreover, other studies have shown that albumin therapy can enhance organ function and minimize complications in patients with aSAH, which may lead to a better outcome (32–34). We argue that pathological mechanisms of elevated NAR related both to an excessive neuroinflammation response, indicated by increased neutrophil, and albumin decrease that might result in susceptibility to infection as POP. Some of these studies also demonstrated a close relationship between POP and severe aSAH (7, 35).

The results presented in this study showed that NAR levels in patients with aSAH with higher WFNS and mFisher grades were higher than those with lower WFNS and mFisher grades, which conclude that the severity of aSAH could be associated with NAR and POP. Additionally, the incidence of pneumonia in patients with aSAH with higher WFNS grade was higher than those with lower WFNS grade.

The ROC curves were further applied to discriminate between patients with aSAH at risk of POP, having observed that NAR was significantly capable of predicting the occurrence of POP. Although inflammatory conditions may be dynamically changing during the early acute period, these predictive abilities were comparable to the NLR and WFNS grades. Our findings, in turn, suggest that the NAR may be potent enough to predict POP.

At the same time, it is interesting to find a higher level of NAR in the POP group than that found in the non-POP group, regardless of using the ventilator or not. This is probably due to the internal inflammation states remaining an imperative cause of pneumonia in patients, despite the risk of lung infection due to mechanical ventilation, and NAR might ultimately be a promising predictor of POP.

The present study has several limitations that ought to be addressed. It consists of a retrospective single-center analysis in which the possibility of a selection bias may exist, making the generalization of these results and their relation to clinical conditions a cautionary matter. Moreover, whether NAR changes over time are correlated with the onset of POP needs to be further examined.

The current study suggests that NAR might be considered a robust and readily available biomarker whose usefulness aims to predict POP occurrence after aSAH. This study can provide an insight into preventive antibiotic therapy and albumin therapy, which ultimately suggests the usefulness of further clinical trials on this matter.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

XZ and CW helped with data collection. XZ and AL contributed significantly to the analysis and manuscript preparation. XZ, SZ, RL, and CW performed the data analysis and wrote the manuscript. XZ, SZ, AL, and RL helped perform the analysis with constructive discussions. All authors contributed to the article and approved the submitted version.

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.