Systemic immune inflammatory index (SII) and systemic inflammatory response index (SIRI) as predictors of postoperative delirium in patients undergoing off-pump coronary artery bypass grafting (OPCABG) with cerebral infarction

Abstract

Objective

Postoperative delirium (POD) is a common complication following off-pump coronary artery bypass grafting (OPCABG) and is associated with significant morbidity. This study aims to evaluate the correlation of systemic immune inflammatory index (SII) and systemic inflammatory response index (SIRI) with postoperative delirium (POD) in patients with cerebral infarction undergoing OPCABG.

Methods

The perioperative cohort study included 321 patients who underwent OPCABG. Patients were divided into two groups based on the occurrence of POD: the delirium group (n = 113) and the non-delirium group (n = 208). Baseline characteristics, including gender, left ventricular ejection fraction (LVEF), surgery duration, hypertension, age, and smoking history were analyzed. SII and SIRI values were calculated preoperatively, and their association with POD was assessed using univariate and multivariate logistic regression analyses. Receiver operating characteristic (ROC) curves were used to evaluate the predictive accuracy of SII and SIRI.

Results

Statistical differences between SII and SIRI in the two groups (P < 0.05) were observed. Multivariate analysis confirmed that SII and SIRI, age and preoperative smoking history were predictors of POD. ROC curve analysis demonstrated that SII and SIRI had considerable predictive power, with AUC values of 0.73 (0.67–0.79) for SII and 0.75 (0.69–0.81) for SIRI.

Conclusion

SII and SIRI were found to be associated with an increased risk of POD in patients undergoing OPCABG, but further research is needed to confirm these findings and determine their independence as risk factors.

Background

Postoperative delirium (POD) is a common yet serious complication observed in patients following surgical procedures. It is characterized by inattention, acute onset and fluctuating course, altered level of consciousness, and disorganized thinking [1]. While the general incidence of POD in surgical patients ranges from 2 to 3%, the rate escalates dramatically between 26 and 52% following cardiac and major vascular surgery, with significant correlations to increased mortality rates [2]. Studies indicate that in patients undergoing off-pump coronary artery bypass surgery (OPCABG), the emergency of POD is often linked to pre-existing conditions such as multiple cerebral infarctions and a decline in overall cognitive function prior to surgery [3]. The pathogenesis of POD is predominantly associated with systemic inflammation [4, 5], triggered by surgical trauma, which catalyzes the release of cytokines such as histone, IL-6 and C-reactive protein (CRP). This cascade leads to both systemic and neurological inflammatory response. Surgical stress facilitates the transportation of inflammatory mediators to the brain through various pathways, intensifying the risk of developing POD [6,7,8]. Particulary in patients with cerebral infarction, the compromised integrity of the blood-brain barrier allows inflammatory factors to infiltrate the central nervous system, further promoting inflammatory reaction within the pain [9,10,11]. Despite the prevalence and the severe impact of POD, it remains a largely preventable condition [12]. Accurate preoperative assessment and the implementation of targeted preventative measures are crucial for mitigating the risks associated with this syndrome. However, clinical practice still largely lacks effective preoperative prognostic indices for POD.

Recent research has highlighted the potential of systemic immune inflammation index (SII) and systemic inflammatory response index (SIRI) as promising prognostic markers. These indices, calculated by platelet, neutrophil, lymphocyte, and monocyte counts, in peripheral blood have shown predictive value for POD across various surgical fields, including abdominal and orthopedic surgeries [13, 14]. Nonetheless, there is a notable gap in the literature regarding their effectiveness predicting POD in patients undergoing OPCABG with pre-existing cerebral infarction [15]. The aim of this study is to fill the gap by investigating the correlation between SII and SIRI and the occurrence of POD in this specific patient cohort through a comprehensive clinical retrospective analysis. The findings are expected to enhance our understanding of predictive value of these indices in managing and preventing POD in patients subjected to OPCABG.

Method

Population

The retrospective study was approved by the ethic committee approval of Tianjin Chest Hospital before the conduction. The signed informed consent forms were collected from all the subjects. From June 2021 to March 2023, perioperative medical records of 1389 patients who had preoperative combined cerebral infarction and underwent OPCABG at Tianjin Chest Hospital were analyzed. The patients were confirmed by neurologists through findings from CT or MRI scans in conjunction with our research team. In the OPCABG procedures, several types of grafts were used, primarily arterial grafts such as the internal mammary artery and, where necessary, venous grafts like the saphenous vein. Each graft choice was tailored based on the patient’s vascular condition and specific coronary anatomy. Special care was taken to minimize or avoid aortic manipulation, particularly in patients classified under CABG 1 and 2, to potentially reduce the incidence of postoperative neurological complications associated with cerebral embolism from aortic handling. However, patients were excluded considering the factors of preoperative creatinine value higher than 133 umol/L, haemoglobin less than 100 g/L in the laboratory tests, left ventricular ejection fraction (LVEF) < 40%, preoperative need for intra-aortic balloon counterpulsation (IABP), comorbid chronic obstructive pulmonary disease (COPD), preoperative presence of hyperthyroidism or hypothyroidism, need for emergency surgery, recent interventional therapy for acute myocardial infarction and a history of previous cardiac surgery. 321 patients in total were finally enrolled.

Data collection and analysis

Data on patient demographics and clinical characteristics were comprehensively collected. This included gender, age, body mass index (BMI), smoking history, diabetes mellitus, hypertension, hyperlipidaemia, LVEF, cardiac output (CO), metabolic equivalent (MET), European System for Cardiac Operative Risk scoring system (EuroSCORE). Additionally we documented each patient’s history such as arrhythmia, as well as surgery. Laboratory test results both pre- and post-operatively were gathered, encompassing haemoglobin levels and serum creatinine. We also recorded the duration of surgery, intraoperative blood loss, and urine output. Crucially, the initial counts of neutrophil, lymphocyte, and platelet post-admission were noted for each patient were recorded. The SII was calculated using the formula: (platelet count × neutrophil count)/ lymphocyte count. Similarly, the SIRI was determined by the formula: (neutrophil count × monocyte count)/lymphocyte count.

Patient grouping and delirium assessment

Patients were divided into delirium group (POD group) and non-delirium group (non-POD group) based on the occurrence of POD. POD was assessed within the five days post-surgery. Delirium detection was performed using the Consciousness Ambiguity Assessment Method for ICU (CAM-ICU). Assessment were conducted twice daily, in the morning and afternoon, from the time of emergence from anesthesia up to five days postoperatively. A positive delirium dilagnosis was confirmed if any sings of delirium were observed during this period. In the CAM-ICU protocol, the Richmond Agitation and Sedation Scale (RASS) was conducted primarily to ensure patients were responsive enough for evaluation, specifically maintaining a RASS score of ≥-3. The results of CAM-ICU were evaluated based on four criteria: (1) acute changes in mental status or a fluctuating course, (2) inattention, (3) disorganized thinking, and (4) altered level of consciousness. A positive CAM-ICU assessment was for delirium equired the presence of both the first two features and either of the latter two.

Statistic analysis

Statistical analyses were performed using R software, version 4.3.0. Continuous variables conforming to a normal distribution were presented as mean ± standard deviation (SD) and comparisons between groups were performed using the independent samples t-test. Non-normally distributed measures were presented as median (interquartile range: P25, P75) and analyzed using the Mann-Whitney test. Categorical variables were expressed as percentages, and analyzed using Pearson chi-square test. Variables with p < 0.05 in the results in univariate analysis were subsequently included in the unconditional logistic regression models to identify the risk factors associated with POD. Further sensitivity analyses utilized propensity score matching to adjust for potential confounders between the two groups. P < 0.05 indicated that the difference was statistically significant.

Results

Characteristics

Significant differences between the POD and non-POD groups were observed in age and smoking history (p < 0.05), indicating older age and a higher incidence of smoking among those who developed POD. No significant differences were found in other parameters such as body mass index (BMI), left ventricular ejection fraction (LVEF), cardiac output (CO), metabolic equivalent (MET), and EuroSCORE. Surgical time and gender distribution approached but did not reach statistical significance (p = 0.056 for gender) (Table 1). The comparison highlights the complexity of risk factors contributing to the development of POD in this patient population.

Table 1 Baseline characteristics and clinical parameters of patients undergoing OPCABG

Laboratory test

The analysis revealed no significant differences between the two groups in terms of neutrophil and platelet counts. However, statistically significant differences were noted in lymphocyte counts and monocyte counts, with the POD group showing lower lymphocyte counts (p = 0.003) and slightly higher SIRI values (p = 0.034). Both SII and SIRI, which were calculated from these cellular components, differed significantly between the groups (p = 0.003 for SII), indicating a stronger systemic inflammatory response in patients who developed POD (Table 2). These findings suggested that heightened preoperative inflammation, as quantified by SII and SIRI, may be associated with an increased risk of delirium post-surgery.

Table 2 Preoperative hematological profiles and inflammatory indices of patients undergoing OPCABG

POD multi factor logistic regression analysis

The analysis identified significant associations between several variables and the risk of POD. Smoking was found to significantly increase the risk of developing POD, with an OR of 2.32 (95% CI: 1.24–4.35; p = 0.009). Age was another significant predictor, with each additional year increasing the risk by 10% (OR: 1.10; 95% CI: 1.05–1.16; p < 0.001). The SII was positively associated with an increased risk of POD, indicating that higher systemic inflammation levels correlate with higher POD risk (OR: 1.01 per unit increase; p < 0.001). In contrast, higher SIRI values were associated with a lower risk of POD (OR: 0.53; 95% CI: 0.35–0.79; p = 0.002), suggesting a complex interplay between different components of the systemic inflammatory response in the development of delirium (Table 3). The ROC curve demonstrates that the SII has a moderate predictive value for POD, with an AUC of 0.73 (95% CI: 0.67–0.79) (Fig. 1). This indicates a fair level of discrimination in identifying patients at risk of developing delirium post-surgery. The curve stays well above the line of no discrimination, suggesting that SII could be a useful clinical tool to assess POD risk preoperatively. The ROC curve in Fig. 2 demonstrates that SIRI has a moderate predictive value for identifying the risk of POD, with an AUC of 0.75 (95% CI: 0.69–0.81). This suggests a reasonable capability of SIRI to distinguish between patients at high and low risk of developing delirium post-surgery. The accuracy indicated by the AUC supports the use of SIRI as a potential tool in the preoperative risk assessment for POD.

Table 3 Postoperative renal injury multivariate analysis

Fig. 1

Receiver operating characteristic (ROC) curve for systemic immune inflammatory index (SII) predicting postoperative delirium

Fig. 2

ROC curve for SIRI predicting postoperative delirium

Secondary analysis

The analysis revealed no significant differences in most baseline characteristics between the POD and non-POD groups, including age, BMI, cardiac output (CO), left ventricular ejection fraction (LVEF), metabolic equivalent (MET), EuroScore, surgery time, bleeding volume, and urine volume. Smoking status showed no statistically significant differences, nor did the presence of arrhythmia, diabetes, high blood pressure, pulmonary disease, kidney disease, and the number of bypass vessels. The lack of significant differences across these variables suggests a homogeneous distribution between the two groups, providing a balanced comparison for assessing the effects of POD. Statistically significant differences were not observed in major clinical factors, indicating that other unmeasured or inherent patient characteristics might influence the development of POD (See Table 4).

Table 4 Patients’ propensity score in two groups

In Fig. 3 before adjustment, significant imbalances were evident in several key variables, which could potentially confound the outcomes of the study. Post-adjustment, the propensity score method effectively reduced the imbalances across all measured covariates, bringing them below the negligible threshold (AMD < 0.1). This adjustment ensures a more reliable and unbiased comparison between the POD and non-POD groups, particularly highlighting the method’s effectiveness in balancing the variables such as age, smoking history, and LVEF, which are critical in studies involving cardiovascular surgery outcomes. The graph illustrates a marked improvement in covariate balance, affirming the robustness of the analytic adjustments made to account for potential confounders in the study.

Fig. 3

Covariate balance before and after adjustment

The analysis confirmed that higher SII values were significantly associated with an increased risk of postoperative delirium, with an OR of 1.01 per unit increase (95% CI: 1.01–1.01, p < 0.001). In contrast, higher SIRI values were associated with a reduced risk, as indicated by an OR of 0.45 (95% CI: 0.27–0.76, p = 0.003). Smoking status and age did not significantly affect the risk of developing postoperative delirium (OR for smokers: 1.36, 95% CI: 0.58–3.18, p = 0.478; OR for age per year increase: 1.04, 95% CI: 0.96–1.13, p = 0.360). These results highlight the importance of systemic inflammation as measured by SII and SIRI in predicting postoperative delirium in this patient population (See Table 5).

Table 5 Multivariate logistic regression analysis of risk factors for postoperative delirium

Discussion

Delirium is an acute, fluctuating psychiatric dysfunction accompanied by a variety of adverse consequences, including prolonged hospitalization, increased healthcare costs, and increased complications [16]. The activation of the brain’s immune system by inflammatory factors is recognized as a key mechanism in the onset of POD. Particularly following open coronary artery bypass grafting (CABG) surgery, the accompanying inflammation and oxidative stress may induce endothelial dysfunction and disrupt the blood-brain barrier, leading to neuronal damage and microglial cell activation, which will lead to POD [16,17,18].

The SII and SIRI, which incorporate neutrophils, lymphocytes and platelets, have emerged as significant predictors of POD. Neutrophils, short-lived phagocytes, signal an acute inflammatory response when elevated, often exacerbated by a concurrent decrease in lymphocytes [19, 20]. Platelets have been studied for their role in their role in bioenergetic functions and disease states like Parkinson’s disease and Alzheimer’s disease. Their manipulation during surgery, particularly through transfusions, has been linked to a reduced incidence of POD [21, 22], suggesting their role in maintaining the integrity of the blood-brain barrier and moderating inflammatory transmigration into the brain.

Elevated SII levels, reflecting heightened inflammation, have correlated with increased disease severity in patients undergoing cardiac surgery and non-cardiac surgeries [23, 24]. Initially defined by Hu et al., SII’s relevance to endothelial dysfunction and its predictive value in cardiovascular outcomes have been well-documented and its prognosis [25]. Another study also suggested that SII was closely associated with short-term adverse outcome in stroke patients after surgery, and that SII could be used as a predictor of adverse outcome in patients with acute ischemic stroke [26]. Our findings affirm that a high SII serves as an independent risk factor for POD in patients with preoperative cerebral infarctions. Similarly, SIRI has shown strong association with POD outcomes in our study. Discovered by Qi et al. in 2016 SIRI was initially intended to help identify candidates for aggressive chemotherapy in patients with pancreatic cancer and has since been identified as a prognostic marker in various malignancies and cardiovascular disease [24, 27]. Its relevance to cognitive impairments post-stroke has also been established, reinforcing its utility in predicting adverse outcomes post-surgery [28].

However, some limitations exist in this study. Details on the mortality and distribution among patients were not included. In fact, mortality rates could be related to severity of the inflammatory status and the distribution of mortality rates across different patient groups provide further insights into the complex interplay between systemic inflammation and patients outcomes following OPCABG. Hence, in future study, mortality rates and its distribution should be included. Additionally, this study focused on preoperative variable, other risk factors including clinical condition and neurological status, degree of atherosclerotic damage, medication, postoperative pain were not mentioned, which may interfere with the result. Besides, the number of cases included in this study was relatively insufficient, and the number of cases was further compressed after the secondary analyses. Moreover, this study was designed to be a single-center retrospective study, where some bias may exist. A more comprehensive outcome may be remained to conclude in a multi centre research that cover wider region.

Conclusion

This study confirms the Systemic Immune-Inflammation Index (SII) and Systemic Inflammatory Response Index (SIRI) as effective predictors of postoperative delirium (POD) in patients with cerebral infarction undergoing off-pump coronary artery bypass grafting (OPCABG). The association of higher SII and SIRI scores with increased POD risk highlights the importance of these indices in preoperative risk assessment. Future research should focus on validating these findings in larger, multicentric studies to further integrate these indices into clinical practice, aiming to reduce POD incidence and improve surgical outcomes.