Intra-operative hyperoxia and the risk of delirium in elderly patients after cardiac surgery

Introduction

Delirium is one of the most common complications after cardiac surgery. The incidence rates range from 11% up to 52% [1]. The presence of delirium after cardiac surgery is associated with a higher mortality rate, increased length of Intensive Unit (ICU) stay [2], higher hospitalization costs [3] and a risk of readmission [4]. Furthermore, the occurrence of delirium during the postoperative period correlates with the development of long-term consequences after surgery such as cognitive decline and functional impairment, which in turn results in the deterioration of the patient’s quality of life [5].

The etiology of delirium is multifactorial and it is always a combination of predisposing and precipitating risk factors [6]. Several pathophysiological mechanisms that contribute to the development of postoperative delirium have been presented, such as inflammation, disturbances in cerebral autoregulation, aortic plaque disruption, microemboli, Alzheimer-like cerebral pathology, and the neurotoxicity of anesthetics [7].

The role of low oxygen concentration in the development of delirium has been previously described. Severe hypoxia during CPB is one of the risk factors for postoperative delirium [8] and intraoperative cerebral oxygen desaturation and disturbances in the oxygen supply-demand balance may contribute to acute brain failure [7, 9]. One of the main tasks for the anesthesiologist during cardiac surgery is to prevent cellular hypoxia and high concentrations of oxygen are often administered. As a consequence, hyperoxia episodes may occur that are as dangerous to the patient as the occurrence of hypoxia. Increasing amounts of data are demonstrating worse clinical outcomes related to hyperoxia. The presence of hyperoxia episodes were reported in over 43% of emergency department patients [10], 36% of patients after cardiac arrest [11] and up to 70% of patients hospitalized in the ICU [12].

The role of hyperoxia as a risk factor for the development of delirium in patients undergoing cardiac surgery with cardiopulmonary bypass (CPB) has not yet been determined. The aim of this study was to determine the impact of hyperoxia episodes occurring during CBP on the incidence of delirium in the immediate postoperative period.

Results

The characteristics of the study population are shown in Table 1. Out of 93 patients included in the study, 11 (11.8%) developed postoperative delirium (group POD (+)) and 82 did not (group POD (-)). There were no significant differences between groups in age, gender, or in factors related to the patient’s clinical condition assessed with the Euroscore II and Charlson Comorbidity Index. There was no statistical difference between the study groups in the incidence of comorbidities such as recent myocardial infarction, lung disease, peripheral vascular disease, history of atrial fibrillation, diabetes mellitus or cerebral vascular disease (Table 1). There were no significant differences in the incidence of postoperative complications such as bleeding, need for re-thoracotomy, need for red blood cell transfusions, or acute kidney injury. 2 patients (18%) in the POD (+) group and 5 patients (6%) in the POD (-) group received milrinone infusion (P=0.15). No one needed adrenaline infusion and noradrenaline infusion was given to 6 patients (54.5%) in the POD (+) group and 40 patients (48.7%) in the POD (–) group (P=0.71). The highest dose of noradrenaline was similar in both groups (P=0.65) (Table 2).

The mean ICU length of stay was 3 days in both groups. There were no significant differences between groups in factors related to the operation such as CBP time, aortic cross clamping time, and mean hemoglobin level and cardiac index during CBP (Table 2). There was no significant difference between groups in blood loss or red blood cell transfusions and detailed information is provided in Table 2. No patient was given steroids perioperatively. After terminating sedation all patients were extubated on the day of surgery and the mean mechanical ventilation time was 292 minutes in group POD (+) and 231 minutes in the POD (-) group (n.s.). The hospital survival rate was 100% in both groups.

Discussion

The study showed that despite the short time of CPB and without complications during surgery and during the ICU stay, up to 11.8% of patients experienced delirium in the period immediately after the surgery.

In the group of patients with delirium, the incidence rate of severe hyperoxia during CPB was significantly higher than in patients with an uneventful postoperative course. Thus, CBP hyperoxia episodes may be a risk factor associated with the occurrence of postoperative delirium.

Data on the negative consequences of hyperoxia are becoming more widely known. The results of previously published studies indicate that in a wide range of medical conditions such as myocardial injury, cardiac arrest, stroke, and respiratory failure requiring mechanical ventilation, hyperoxia is associated with increased morbidity and hospital mortality [11, 13, 14]. In our study, the occurrence of hyperoxia during surgery did not change mortality, and the survival rate in the hospital was 100%. However, this study group was a low risk population; all patients could be extubated on the day of surgery, without the need for a prolonged ICU stay.

The influence of hyperoxia on brain activity has been the subject of preclinical and clinical studies; however, the results of these studies are often contradictory. The results of preclinical studies indicate that prolonged hyperoxia induces a disturbance in the balance between the production of reactive oxygen species (ROS) and antioxidant defense [15, 16]. As a consequence, neuron apoptosis and cerebral tissue ROS increase, leading to persistent cerebral damage. In a clinical study investigating the effect of hyperoxia in patients with severe traumatic brain injury, a PaO₂ above 19.9 kPa was shown to contribute to an increase in the glutamate level in brain tissue, suggesting secondary brain damage [17]. These results were not confirmed in a recently published retrospective analysis of over 24,000 patients after traumatic brain injury; hyperoxia > 39.8 kPa during the first 24 hours in the ICU, although present in 13% of patients, was not associated with a worse outcome [18]. In another study, Rincon et al. demonstrated the detrimental effect of hyperoxia in ventilated ICU patients after stroke; a PaO₂ above 39.9 kPa was associated with higher in-hospital mortality and hyperoxia was even more harmful than hypoxia below 7.9 kPa [19]. These data underline the need for studies of controlled oxygenation in ventilated ICU patients.

To date, the impact of hyperoxia on the brain activity of patients undergoing cardiac surgery with CPB has been the subject of very few studies and this problem requires thorough analysis. The results of our study indicate a direct relationship between the high intraoperative PaO₂ level and the occurrence of delirium in the immediate postoperative period in elderly patients. During CBP, severe arterial hyperoxia (PaO₂ ≥ 26.6 kPa) was recorded in all patients who postoperatively developed delirium, and the incidence rate of severe hyperoxia recorded during CPB was significantly higher in patients with delirium than in those with an uneventful postoperative course.

In a study by Lopez et al., cerebral oxygenation was measured with oximetry monitors in cardiac surgery patients. The results indicated that prolonged cerebral hyperoxia during CBP was responsible for the occurrence of delirium following surgery in 30% of patients. What is more, the risk of developing delirium significantly increased when an episode of hypoxia was followed by prolonged hyperoxia [20]. In a recently published study by Thudium et al. a relationship was found between cerebral hyperperfusion and post cardiac surgery delirium. In the study the authors examined right middle cerebral artery blood flow during CBP using the transcranial Doppler [21]. In our study we did not plan to measure cerebral blood flow, but the cardiac index during CBP was similar in both groups, with and without postoperative delirium (mean CI=2,5 l/min/m², P=0.69).

Although data on the negative effects of hyperoxia are increasing, toxic oxygen levels can still be found in the operating room. The safety of maintaining near-physiological levels of oxygenation during CPB was confirmed in a randomized clinical trial. It was demonstrated that the procedure of using an almost physiological level of oxygenation during CPB did not increase the level of lactate and troponin after surgery and did not increase the frequency of hypoxic episodes [22].

Delirium is always a result of many factors the majority of which are unpreventable. One important risk factor is prolonged CBP time. A detailed analysis of 1863 cardiac surgery procedures indicated that CPB time ≥90 minutes was considered to be prolonged and was associated with a longer hospital stay and a higher mortality rate [26]. In another study CPB ≥90 was associated with an increased release of the biomarkers of ischemia/reperfusion intestinal damage and endotoxemia [27]. Since we have no influence on CBP duration, careful attention to even the smallest details during surgery that may potentially have an impact on the postoperative course, are therefore, very important. Control and prevention of perioperative hyperoxia can be one such task. In our study, samples of blood-gases were controlled every 20-30 minutes and gas flow settings were corrected accordingly. The implementation of real-time PaO₂ measuring devices can help reduce both hypoxia and hyperoxia episodes.

We are aware of our study limitations. The study was deliberately limited to elderly patients, a population at higher risk of postoperative delirium. We did not plan to analyze the long-term consequences of hyperoxia during CPB on brain injury. However, in a recently published retrospective analysis of 1018 patients undergoing cardiac surgery with CPB no association was found between hyperoxia >26.6 kPa or >39.9 kPa during surgery and neurocognitive dysfunction 6 weeks after surgery [23].

Additional research conducted on a larger population, with the assessment of cerebral oximetry and cerebral blood flow, and an examination of the immediate and long-term consequences of hyperoxia, will shed light on the question of the effects of hyperoxia on brain function after cardiac surgery. Future research in this area is needed and should also include the impact of hyperoxia on brain damage and the release of brain-specific biomarkers. Changes in biomarkers in patients experiencing hyperoxia episodes could be associated with cognitive decline and delirium after surgery.

Considering the lack of statistically significant differences between groups in the values of the other parameters studied in the perioperative period, it can be assumed with high probability that intraoperative hyperoxia may be the cause of postoperative delirium.

Delirium is always a result of many factors. As we have no influence on many of them, preventing delirium is always a difficult task. Since hyperoxia during CBP is an important risk factor for developing postoperative delirium, avoiding hyperoxia above 26.6 kPa during CBP, along with avoiding large variations in oxygen tension, seems an easy and safe measure to lower the risk of this complication.

Materials and Methods

This retrospective study was conducted at the 6-bed cardiosurgical ICU at the Department of Anesthesiology and Intensive Therapy, Wroclaw Medical University in Poland between December 2016 and March 2018. The Bioethics Committee of Wroclaw Medical University approved the study (permission no. 219/2016). Inclusion criteria were: age ≥ 65 years and elective on-pump cardiac surgery. Exclusion criteria were: CBP time ≥ 90 minutes, emergency procedure, a different type of anesthesia (TIVA, total intravenous anesthesia), or refusal to provide written informed consent.