The association between Monocyte‐to‐Lymphocyte ratio and postoperative delirium in ICU patients in cardiac surgery

Abstract Objective To analyze the relationship between monocyte‐to‐lymphocyte ratio (MLR) and postoperative delirium (POD). Methods This cohort study was conducted in the Medical Information Mart for Intensive Care‐III (MIMIC‐III) version 1.4 database. MLR was measured according to the complete blood count. ICD‐9 was used to measure postoperative delirium. Multivariable logistic regression was utilized to examine the relationship between MLR and POD. Results Three thousand eight hundred sixty‐eight patients who had received cardiac surgery were retrospectively enrolled, including 2171 males and 1697 females, with a mean age of 63.9 ± 16.2 years. The univariate analysis suggested that high MLR (as a continuous variable) as associated with a 21% higher risk of POD (O R: 1.12, 95% CI, 1.02, 1.43, p = 0.0259), After adjustments for other confounding factors, gender, age, race, temperature, SBP, DBP, MAP, respiratory rate, SOFA, peripheral vascular disease, AG, psychoses, drug, and alcohol addiction, the results showed that high MLR (as a continuous variable) independently served as a risk factor for POD (OR: 1.21; 95% CI: 1.01–1.44; p = 0.0378). MLR was assessed as quintile and tertiles, high MLR was an independent risk factor for POD. In the subgroup analysis, there were no differences in MLR for patients with POD in pre‐specified subgroups. Conclusions Monocyte‐to‐lymphocyte ratio was a risk factor for POD. More research is necessary to thoroughly examine the function of MLR in POD.

stays in the intensive care unit (ICU) and the hospital, as well as higher post-surgical morbidity and death. 6 Furthermore, POD has an influence on long-term cognitive and functional deterioration, resulting in an increased demand for health care resources and expenses. 7 Even though the exact mechanism behind the development of POD is not known at this time, neuroinflammation generated by the surgery-induced systemic inflammatory process has been suggested to participate in this condition. [8][9][10] While an elevated level of c-reactive protein and interleukin-6 (IL-6) has also been shown to be correlated with POD, 11 whereas one cohort research found no correlation between the level of plasma IL-6 and delirium in elderly hospitalized patients. Consequently, the relationship between plasma markers of inflammation and POD is still unclear. 12 POD might possibly be predicted using biomarkers associated with inflammation. 13 It has been shown that the aberrant elevation in inflammation blood cell variables, including the neutrophil count, neutrophil-to-lymphocyte ratio, and platelet-tolymphocyte ratio, functions as a basic indicator of inflammatory response; all of these parameters have been evaluated for their potential in POD. [14][15][16] However, the predictive value of inflammatory biomarkers in POD has not been investigated independently and remains unknown.
Hemogram derived inflammatory markers have been reported to be associated with outcome in ICU population. These include platelet-to-lymphocyte count ratio 17 , neutrophil-to-lymphocyte count ratio, 18 and mean platelet volume. 19 Therefore, another hemogram derived marker, monocyte-to-lymphocyte ratio (MLR), could be associated with worse outcome in ICU patients after cardiac surgery. The MLR is a newly developed and integrative inflammatory biomarker that is based on monocyte and lymphocyte counts 20,21 .
The increased MLR was initially used to assess the diabetic kidney injury 22 ,liver steatosis, 23 irritable bowel syndrome, 24 cancer, 25 and

| Source of data and sample
We conducted a single-center retrospective cohort study; we collected all relevant data from Medical Information Mart for Intensive

| Selection criteria
Among the more than 50,000 different patients in the database, the subjects included in this study had to meet the following criteria: (1) The current procedural terminology (CPT) was utilized to identify patients undergoing cardiac surgery: Ideally, the CPT number ought to fall between 33,010 and 37,799, and (2) age ≥16 years. Exclusion criteria of this study were as follows: (1) younger than 16 years of age; (2) monocytes and lymphocyte count data were lost on the first day of admission to ICU; (3) data were missing by more than 20%; (4) intubated patients; and (5) patients had a hematologic neoplasm diagnosis.

| Evaluation of MLR
The blood count was recorded, which included the absolute numbers of lymphocytes and monocytes. In addition, MLR was defined as MLR = M/L.

| Baseline variables
Data were extracted by structured query language PostgreSQL 9.6.
Demographic variables such as age, gender, race, and complications included hypertension, arrhythmia, heart valve disease, congestive heart failure, pulmonary circulation disorders, peripheral vascular disease, diabetes were acquired. All vital signs, results of blood gas (anions gap, lactate, and bicarbonate concentrations), lab findings, mechanical breathing time, and perioperative transfusion data were obtained. The sequential organ failure assessment (SOFA) score and s simplified acute physiology score II (SAPS II) were all computed while being admitted to the ICU. to represent the effect. We also made adjustments for the factors that were correlated with the dominant and secondary outcomes.

| Statistical Analysis
In addition, multivariate analysis was used to control for the corresponding confounding factors; in model I, the confounding factors, including age, gender, and race were adjusted, while in model II, confounding factors, including age, sex, race, diastolic blood pressure, heart rate, respiratory rate, temperature, SpO2, heart failure anion gap, platelet, serum chloride. Were adjusted. Furthermore, subgroup analyses were conducted in order to corroborate the validity of our results. STATA (version: 15.0) (STATA Corp LLC) was employed to conduct the statistical analyses. Statistical significance was considered to have been attained when p < 0.05.

| Patient characteristics
A total of 3868 patients with cardiac surgery were retrospectively enrolled, including 2171 males and 1697 females, with a mean age of 63.9 ± 16.2 years. According to the absence or presence of POD, the patients were classified into POD and non-POD groups; data for a sum of 562 patients in the POD cohort and 3306 in the non-POD cohort were analyzed. The patient baseline information is shown in Table 1. The patients in the POD group had significantly higher SOFA scores, SAPS II scores, Elixhauser comorbidity index scores, and MLR as compared to those in the POD group. Vital signs, including mean heart rate, respiration rate, temperature, SBP, DBP, and MAP were higher in the POD group, Laboratory indicators were used for the assessment of organ functions for both groups of patients. The results demonstrated that there was no significant difference between the two groups. Patients in the Pod group had higher rates of drug and alcohol addiction. We further compared three groups based on MLR, and the patient baseline information shown in Table 2. As opposed to the patients in the low MLR group, those in the high MLR group are dominantly female, the white race, elevated CHF rate, arrhythmia, delirium, hypertension, kidney failure disease, greater platelet, and WBC counts, elevated heart rate, PT, APTT, anion gap, BUN, creatinine, lactate, and respiratory rate, reduced levels of SPO2, chloride, bicarbonate, MAP, hemoglobin, and elevated SOFA scores, SAPS II scores, Elixhauser comorbidity index scores.

| Relationship between MLR and POD
Following adjustments for the potential confounding variables, we developed distinct models for the purpose of evaluating the independent impacts of MLR on the POD. As shown in Table 3

| subgroup analysis
The results of subgroup analysis are shown in Table 4. There were no differences in MLR for patients with POD in pre-specified subgroups.

| DISCUSS ION
As far as we know, this study is the first to demonstrate the strong correlation between MLR and POD. We observed that patients shown that when the central nervous system's homeostasis is disrupted, a large number of inflammatory mediators produced by activated microglia cause neuroinflammation to occur. 32,33 Despite the fact that elevated levels of inflammatory cytokine have been correlated with the occurrence and progression of POD, measuring inflammatory markers is a costly procedure that cannot be performed in many institutions. 33 However, since the inflammatory mediators used in the present research could be derived from the findings of a full blood count examination, they are simple to use and affordable to obtain. It has been shown that the MLR may function as a biomarker for both systematic inflammatory responses and neuroinflammation in individuals suffering from depression. 34 It is common practice to utilize inflammatory markers found in regular blood testing to predict the prognosis of POD, and these indicators represent the extent of systemically low-intensity inflammatory response. As a result, they are used to treat a broad range of clinical disorders in different settings.
Nevertheless, the link between MLR and POD remains unclear.
Local inflammation in injured brain areas contributes to secondary brain injury by enhancing the disintegration of the blood-brain barrier, neural death, oxidative stress, cerebral edema, and microvascular failure that occurs due to the damage. 35,36 The inflammation could manifest itself throughout the brain, with long-term consequences for the patient's cognitive functioning. 37 The primary damage caused by POD delivers a chain of occurrences, such as the secretion of excitotoxic compounds, oxidative stress, and mitochondrial illnesses, all of which contribute to secondary brain injury, manifesting as compaction of brain tissue, impaired blood clotting, an intracellular biochemistry sequence reaction, inflammation, and other symptoms.

TA B L E 4 (Continued)
simple; nonetheless, the absence of M and L in the database remains prevalent, resulting in selection bias.

| CON CLUS ION
In conclusion, we offered the first proof that MLR is correlated with an elevated chance of developing POD. MLR might be an accessible and reliable marker that can be used to predict POD in ICU patients in cardiac surgery. This finding should be confirmed in prospective studies.

ACK N OWLED G EM ENT
None.

CO N FLI C T O F I NTE R E S T
None.

DATA AVA I L A B I L I T Y S TAT E M E N T
All the data used to support this study are available from the corresponding author (E-mail: kapalu1979@sina.com) upon request. Note: ORs (95% CIs) were derived from logistic regression models. Covariates were adjusted as in model 1 ( Table 3).