Effectiveness of Intraoperative Dexmedetomidine on Early Postoperative Cognitive Dysfunction in Elderly Patients Undergoing Video-Assisted Thoracoscopic Lobectomy: A Randomized Controlled Trial

Background: Postoperative cognitive dysfunction (POCD) is the decline in cognitive ability that occurs between one week and one year following surgery. POCD has been reported to occur in 5% to 26% of elderly surgical patients and its incidence varies with demographic, clinical, and surgical variables. Previous studies have shown that dexmedetomindine (DEX), a highly selective α 2 adrenergic receptor agonist, has many effects that include opioid-sparing properties, decreased required anesthetic medication, and neuroprotection in animal models. Ultimately, DEX may benet cognitive ability in the postoperative period. Therefore, we hypothesized that intraoperative infusion of DEX would reduce early POCD and interleukin-6 (IL-6) levels in elderly surgical patients compared to placebo. Methods: This randomized, double-blind, controlled trial enrolled 96 elderly patients that underwent video-assisted thoracoscopic lobectomy (VATL) between May 2017 and April 2018. The enrolled subjects were randomly assigned in a 1:1 ratio to receive intravenous dexmedetomidine (DEX Group) or placebo (0.9% normal saline, Saline Group) as a continuous infusion at a rate of 0.5μg/kg/hr starting at anesthesia induction until incision closure. All subjects received total intravenous anesthesia (TIVA) during surgery and patient controlled epidural analgesia (PCEA) following surgery. The primary outcomes were to determine the incidence of POCD and IL-6 expression at postoperative day (POD) 1, 2, and 7. The secondary outcomes were to evaluate the degree of pain on POD 1 and 2, as well as adverse events throughout the study. Results: 96 eligible subjects (48 per group) were randomized, dosed with study medication, and evaluated per protocol. There was a signicant difference on POCD between the DEX and saline groups on POD 1 (6.3% vs. 20.8%, p-value <0.05) and POD 2 (4.2% vs. 16.7%, p-value <0.05). The IL-6 levels were signicantly lower in the DEX group when compared with saline group during POD 1 and 2 (p-value <0.05). IL-6 returned to baseline levels on POD 7. There was no signicant difference in POCD on POD7, and overall there were no signicant differences in the incidence of adverse events between both groups. Conclusions: The administration of intraoperative DEX in elderly patients undergoing VATL decreased the incidence of early POCD. Clinical Trial Registry: This study was registered on Chinese Clinical Trial Registry with the ID ChiCTR-IPR-17010958.


Introduction
Postoperative cognitive dysfunction (POCD) is the decline in cognitive ability (especially executive functions and memory) that occurs between one week and one year after surgery [1,2]. Early POCD has been associated with higher mortality incidences, dementia and premature retirement from the workforce [2]. POCD is a main contributing factor associated with prolonged hospitalization, quality of life (QOL) deterioration, and increased morbidity/mortality [1][2][3][4]. POCD has been reported to occur in 5-26% of elderly surgical patients and its incidence varies with demographic, clinical and surgical variables [1,5].
POCD has also been demonstrated in patients undergoing video-assisted thoracoscopic lobectomy (VATL) with one lung ventilation (OLV) [6,7]. OLV increases the release of various cytokines due to perioperative organ injury [8].An in ammatory reaction, particularly in the central nervous system, is strongly associated with POCD [5,8]. A meta-analysis indicated that interleukin-6 (IL-6) could serve as a biomarker to guide the diagnosis, prevention, and treatment of POCD [9]. Dexmedetomidine (DEX), a highly selective α 2 adrenergic receptor agonist, can induce physiological sleep, inhibit in ammatory and stress response, and have brain protective effects [1,10]. Previous studies have shown that DEX has many effects that include opioid-sparing properties, decreased required anesthetic medication, and neuroprotection in animal models [1,10]. A study conducted by Monk et al. suggested that DEX prevented POCD due to its effective inhibition of the in ammatory response and neuroprotective effects both in vitro and in vivo [11]. Ultimately, DEX may bene t cognitive ability in the postoperative period [1].
We hypothesize that the intraoperative infusion of DEX will reduce POCD and IL-6 levels compared to placebo. We designed a randomized controlled trial (RCT) to assess the effect of intraoperative DEX administration on POCD and interleukin-6 (IL-6) expression in elderly patients undergoing VATL under general anesthesia and patient controlled epidural analgesia (PCEA). The primary outcomes were to assess the effect of the administration of DEX compared to placebo on POCD and expression of IL-6 in elderly patients undergoing VATL. written informed consent prior to surgery and study activities. This study has been described following the consolidated standards of reporting trials (CONSORT) guidelines for the presentation of clinical trials [13].

Participants
We screened patients older than 65 years old undergoing elective VATL for lung cancer between May 2017 and April 2018. All patients received total intravenous anesthesia (TIVA) and PCEA. Exclusion criteria included history of mental disorders, mental impairment (all patients were consulted by a psychiatrist one day before surgery), substance or alcohol addiction, smoking (14 days prior to surgery), sleep disorders, and severe chronic obstructive pulmonary disease that required oxygen therapy. Subjects with partial pressure of oxygen (PaO2) < 60 mmHg in arterial blood, preoperative bradycardia (heart rate (HR) < 45 bpm), severe visual or hearing impairment, protocol violation, baseline Mini-Mental State Examination (MMSE) score < 24, prolonged extubation during anesthesia emergence (> 30 minutes), and failure of PCEA (failure of epidural puncture or pain score > 5) were also excluded.

Randomization and blinding
Subjects were randomized in a 1:1 ratio to receive dexmedetomidine 0.5 µg/kg/hr arm or saline placebo using a computer-generated randomization schedule. The pharmacist that prepared the study drug was the only un-blinded personnel. The respective continuous infusion of DEX or placebo were administered during surgery.

Parameters for general anesthesia
All patients enrolled in the study were scheduled to receive TIVA. Patients fasted overnight and without premedication. Midazolam or other benzodiazepines were avoided. After entering the operating room, electrocardiograph and pulse oxygen were monitored. Vital signs including MAP, HR, SpO2, and temperature were monitored in the operating room and recorded. During the Intra-operative Period, esophageal or other core temperature was monitored and recorded. Invasive blood pressure was monitored by a left radial artery catheter under local anesthesia. Thoracic epidural catheters were placed at T6-7 interspace and advanced 3-4 cm cephalically for PCEA prior to anesthesia induction.
Fentanyl, 0.2-0.4 µg/kg, etomidate, 0.3 mg/kg, and rocuronium 0.6 mg/kg were used to induce anesthesia. Neuromuscular blocking agents were used for intubation and anesthesia maintenance at the anesthesiologist's discretion. After double lumen intubation, mandatory volume ventilation was applied throughout the procedure with expiratory tide carbon dioxide (ETCO 2 ) 35-45 mmHg. Anesthesia maintenance consisted of propofol and remifentanil; and Bispectral index (BIS) was maintained between 40 and 60 during the surgery. Cis-atracurium titrated according to the need of surgery process. Propofol and remifentanil were discontinued at the start of skin closure, and the analgesia pump was connected. Patients were extubated after muscle relaxant reversed, conscious level achieved, and adequate tide volume gained. PCEA was started after wound closure and a loading dose of 5 ml 0.15% ropivacaine was injected. Then, the PCEA with 0.15% ropivacaine and 0.12 µg/ml sufentanil was started. Analgesia pump parameters were set to a background ow of 5 ml/hr, PCA of 3 ml and a lockout time of 15 min. If the patient's rest visual analogue scale (VAS) was above 3, 3 ml of analgesia pump drug was given by the investigator. PCEA was removed on the morning of POD 3.
Intervention Subjects were randomly assigned to DEX group or Saline group (placebo). The study drug or placebo was infused using a controlled infusion device (standard intravenous pump system) that precisely infused study drug. Subjects assigned to the DEX group received a continuous infusion of DEX that was pumped through a peripheral vein assess at a rate of 0.5 µg/kg/hr starting at anesthesia induction without bolus or loading dose and stopping when the chest closure started. Subjects in the saline group received the same dose rate of 0.9% normal saline.

Outcomes
The primary outcomes were to assess the effect of the administration of DEX compared with saline on POCD and expression of IL-6 in elderly patients undergoing VATL. Both neurocognitive scales and serum level of IL-6 for each patient were assessed at baseline (prior to surgery) and consequently on the postoperative day (POD) 1, 2, 7. POCD was diagnosed by Z-score in a comprehensive test scale recommended by the International Study of POCD (ISPOCD) [14]. Patients were de ned as having POCD when Z-score were above 1.96 using cross-reference [15].
The secondary outcomes were postoperative analgesia effect (scored using a VAS, where 0 represented no pain and 10 represented the strongest pain) and adverse events (including bradycardia, nausea or vomiting, skin itching, respiratory depression, and shiver) within 2 days after surgery.

Statistical analysis
Based on preliminary results not published, we calculated that 17% of difference with POCD between groups, 86 patients (43 each group) will provide 80% power and a type I error of 0.05 (two-sided test). In this trial, 48 patients in each group were recruited to allow for drop out. All data was processed by the statistical package, SPSS v. 22.0, for Windows (SPSS, Inc., Chicago, IL, USA). Quantitative variables were presented as mean ± standard deviation. Incidences were analyzed using χ2 or Fisher's exact tests. Continuous variables were tested with Student's t-tests for normal distribution data; otherwise, Mann-Whitney U tests were used. Repeat analysis of variance (ANOVA) was applied to compare the difference among the different times in the two groups. The signi cance level was set at 5%.

Results
A total of 130 subjects were enrolled and assessed for potential participation in the study and underwent screening procedures. After eligibility assessments, 34 subjects were considered ineligible due to not meeting inclusion criteria (20), declined their participation (8) and canceled surgery (6). The remaining 96 subjects ful lled the screening procedures and were randomized to be allocated in either group. After completing all enrollment activities and blinding removal, 48 subjects were allocated to the DEX group and 48 subjects were allocated to the saline group. The ow diagram of this RCT following the CONSORT 2010 guidelines is shown in Figure 1 [13]. Therefore, the data analysis of 96 subjects was conducted.
There were no signi cant differences in American Society of Anesthesiology (ASA) Physical Status Classi cation, age, sex, body mass index (BMI), education years, anesthesia length (from induction to anesthesia stop), OLV length and length of extubation after anesthesia stop found between the groups (P>0.05) ( Table 1). Note: Data are shown as mean ± standard deviation (SD) or n.
There was a signi cant difference on POCD between the DEX group and saline group on POD 1 (6.3% vs. 20.8%, p-value 0.037) and POD 2 (4.2% vs. 16.7%, p-value 0.045). (Table 2).  There were no signi cant differences on baseline IL-6 levels between both groups. The IL-6 levels were signi cantly lower on the DEX group when compared with saline group during POD 1 and 2 (p-value <0.05). IL-6 returned to baseline levels on POD 7. (Figure. 2) There were no signi cant differences on pain VAS scores at rest between the groups during POD 1 and 2, while the pain VAS scores at movement were signi cantly lower in the DEX group (p-value <0.05, Table 3). Adverse events collected for the study (bradycardia, nausea or vomiting, skin itching, respiratory depression and shiver) were rare and there were no signi cant differences between the groups (p-value > 0.05, Table 4). Note. Data are expressed as n (%).

Discussion
Surgical interventions can activate a cascade of reactions which could then result in an in ammatory response and pain [16]. The increase of proin ammatory cytokines may be an essential risk factor in cognitive decline after surgery [17]. Over-expression of in ammatory cytokines may lead to POCD [18].
POCD is a risk factor that results in a longer hospital stay, a lower likelihood of a return to independent living, and increased mortality [19]. An international trial of elderly patients (median age 68, range 60-81 years) demonstrated the incidence of POCD could be up to 26% within 7 days after non-cardiac surgery, and advanced age and length of anesthesia time were highly related with POCD [20]. Patients with OLV during thoracic surgery were at higher risk of developing POCD [6]. Our study selected subjects who presented with high risks, including elderly patients undergoing VATL with OLV.
DEX, an alpha 2 adrenal receptor agonist, has a range of effects on in ammation, anesthesia, analgesia, and cerebral protection. DEX used in monitored anesthesia can signi cantly reduce requirements for both midazolam and fentanyl, and provide better patient satisfaction [21]. DEX exhibited signi cant effects against oxygen and glucose deprivation-induced injury [22]. The increase of pro-in ammatory cytokines in the hippocampus may be an important risk factor for cognitive decline after surgery [23]. The aim of this study was to investigate the effectiveness of DEX on POCD and the levels of main in ammatory marker (IL-6) in elderly patients undergoing VATL for lung cancer.
This study demonstrated that the incidence of POCD on POD1 was higher than that at the other time points, with the incidence of POCD decreasing over time. This trial suggests that patients with POCD at POD1 are prone to develop POCD. Moreover, the patients were administered DEX from incision to chest closure, with no infusion occurring during the postoperative period. It is possible that the dose of intraoperative DEX was associated with the setting of the neurochemical milieu of a general anesthetic. The expression of IL-6 reached the climax on POD 1, which is consistent with the occurrence of POCD. This result might be due to the relation between the infusion time and the acting nature of the drug, or the in ammation activated by trauma.
We excluded patients with perioperative severe cardiovascular and pulmonary comorbidities, such as coronary artery disease, arrhythmia, heart dysfunction, chronic obstructive pulmonary disease, asthma, respiratory failure, which provided a homogeneous group for which we could make conclusive statements. Conversely, Deiner et al. conducted a RCT infusing dexmedetomidine (0.5 μg/kg/hr) vs. placebo during surgery and 2 hours after surgery in patient undergoing major elective cardiac surgery; then the study assessed postoperative delirium for up to 5 days and POCD at 3 and 6 months [1]. The study concluded that the infusion of dexmedetomidine at a rate of 0.5 mcg/kg/hr did not prevent POCD or postoperative delirium after major cardiac surgery [1]. The dose of dexemedetomidine used for this study was similar with the dose from our study; however, our study assessed early POCD and controlled postoperative pain with PCEA.
Thoracic epidural analgesia is a gold standard for pain control after thoracic surgery. PCEA improves outcomes, including: the in uence on POCD, the opioid dose and its related side effects, such as nausea and vomiting, skin itching, and respiratory depression [24]. Although PCEA controlled analgesia effect for both groups in our study, DEX could signi cantly improve pain at movement.
This result might be implicated by another trial [21]. DEX was a safe and effective primary sedative alternative [21]. Furthermore, we combined neurocognitive score and IL-6 simultaneously. We used the recommended tests by ISPOCD in the present study and calculated Z-score above 1.96 to diagnose POCD [15]. This method avoided oor or ceiling effects. The evaluators were well trained and blinded to the grouping. There were no signi cances between the groups in adverse events. We did not use a loading dose of DEX as the other study, but bradycardia rarely occurred.
Our study had a few limitations to be considered. First, we followed up the patients within 7 days after surgery when they were at a high risk for developing POCD. This might overlook the patients that could develop POCD after 7 days postoperatively. Second, although there were many biochemical indicators to predict POCD, we chose a popular clinical biomarker IL-6 because of limited funding. Third, we excluded the patients with severe pulmonary disease, which may decrease the incidence of POCD. Fourth, the values of oxygen pressure in the artery blood were normal, therefore we did not collect and analyze the values in artery blood gas. Lastly, this study did not assess for postoperative delirium and that might interfere with POCD assessments.

Conclusion
The administration of intraoperative DEX at 0.5 μg/kg/hr infusion in elderly patients undergoing VATL decreased the incidence of early POCD. In addition, PCEA provided satis ed analgesic effect. The use of DEX and PCEA might reduce the IL-6 expression and pain VAS at movement within the rst two days after surgery.

Declarations
Ethics approval and consent to participate: The ethics committee of the A liated Hospital of Inner Mongolia Medical University approved the study protocol (no: 2017-003). All patients signed written informed consent prior to surgery and study activities.
Consent for publication: The authors consent for publication.
Availability of data and materials: The datasets used and analysed during the current study are available from the corresponding author on reasonable request.

Figure 2
Comparison of IL-6 levels. Note: Data are expressed as mean ± standard deviation (SD) and frequency.