Impact of Early Postoperative Delirium on Long-Term Outcome in Patients after Coronary Artery Bypass Graft Surgery: A 2-Year Follow-Up Cohort Study

Background: The negative influence of postoperative delirium on long-term outcomes is still underestimated. The purpose of this study was to investigate the impact of early post-operative delirium on long-term outcome of patients after coronary artery bypass graft (CABG) surgery. Methods: This was a 2-year follow-up cohort study. Two hundred and forty-three patients who underwent CABG were enrolled. Delirium was assessed using the Confusion Assessment Method for the Intensive Care Unit (CAM-ICU) in the first 5 days after surgery. Patients were followed-up for 2 years after the surgery. Long-term survival rate was recorded. Cognitive function was assessed with Telephone Interview for Cognitive Status-modified (TICS-m). Quality of life was assessed with 36-Item Short Form Health Survey (SF-36). Results: Early postoperative delirium developed in 50.6% of patients. Multivariable regression analysis showed that prolonged duration of delirium (relative risk [RR] 1.322, 95% confidence interval [CI] 1.074-1.628; P=0.009) and prolonged length of stay in ICU (RR 1.071, 95% CI 1.008-1.139; P=0.027) were the independent predictors of 2-year mortality. Patients who developed delirium had a lower score on TICS-m. Furthermore, they had lower scores on physical functioning, physical role, general health and social role of SF-36 health survey. Conclusions: Early postoperative delirium was common in patients after CABG surgery. Prolonged duration of delirium was associated with increased risk of 2-year mortality. In survival patients at 2-year follow-up, those who developed early postoperative delirium had worse cognitive function and quality of life than those who did not. ABSTRACT

Delirium is a common neurological complication in patients after cardiac surgery.The reported incidence of early postoperative delirium ranged from 3% to 72% (2)(3)(4)(5).In our prospective cohort study, the incidence of early postoperative delirium after coronary artery bypass graft (CABG) surgery was 50.6% (6).The occurrence of postoperative delirium is associated with worse shortterm outcomes, such as increased postoperative complications, prolonged in-Original Article Volume 3 March, 2016 Number 2 tensive care unit (ICU) or hospital stay, increased health care costs, high mortality rate and declined cognitive function (7)(8)(9)(10).Delirium is also associated with worse longterm outcomes in patients after cardiac surgery.Koster et al. (11) reported that early postoperative delirium was correlated with long-term dysfunction of memory, attention and sleep.In a 36month follow-up study of 302 patients after CABG surgery, Loponen et al. (12) found that the quality of life was much worse in delirium patients than in non-delirium ones at 0.5, 1.5 and 3 years after surgery, respectively.After ten years' follow-up of patients receiving CABG surgery, the study of Gottesman et al. (13) showed that early postoperative delirium was an independent predictor of long-term mortality.
So far there have been few studies of Chinese patients in this field.The purpose of our research was to observe the impact of early postoperative delirium on long-term outcome including survival rate, cognitive function and quality of life in Chinese patients after CABG surgery.

MATERIALS AND METHODS
This was a long-term follow-up study of patients recruited in a prospective cohort study (6).The study protocol was approved by the local Clinical Research Ethics Committees (No. [2007] 077).All patients gave written informed consent.

Patients
The inclusion criteria were patients who were referred for elective CABG surgery at Peking University First Hospital and Beijing Fuwai Hospital from March 2008 to July 2008.Patients who met any of the following criteria were excluded, i.e., 1) previous cardiothoracic surgery; 2) history of psychiatric disease; 3) history of adrenal gland disease; 4) history of glucocorticoid therapy for more than 7 consecutive days within one year; 5) preoperative left ventricular ejection fraction (LVEF) < 25% (echocardiography, Simpson's method); or 6) concomitant surgery other than CABG, such as valvular replacement.

Perioperative Data Collection
Preoperative data including demographic characteristics, previous medical history, chronic smok-ing, alcoholism, habitual benzodiazepines use, previous surgical anesthesia, preoperative LVEF, Canadian Cardiovascular Society (CCS) class, New York Heart Association (NYHA) functional class, and European System for Cardiac Operative Risk Evaluation (EuroSCORE) were collected.
Intraoperative data including duration of anesthesia, duration of surgery, dosage of anesthetics, use of anticholinergics, surgery with cardiopulmonary bypass, and blood transfusion were recorded.
Postoperative data including Acute Physiology and Chronic Health Evaluation (APACHE) II score, use of benzodiazepines in the ICU, duration of sedation in the ICU, duration of mechanical ventilation, length of stay (LOS) in ICU and hospital, occurrence of delirium during the first 5 days after surgery, occurrence of non-delirium complications during the first 28 days after surgery (Table 1), LVEF before discharge, and 2year mortality were documented.

Delirium Assessment
Prior to the study, the physician performing the assessment of delirium (D -LM) was trained by a psychiatrist to use the Confusion Assessment Method for the Intensive Care Unit (CAM-ICU) (14,15).During the study phase, patients were assessed for delirium twice daily (from 6 to 8 a. m. and from 6 to 8 p.m.).For each patient, delirium assessment was performed until the fifth postoperative day or the disappearance of delirious symptoms for two consecutive days.Study personnel who assessed delirium were blinded to the treatment details.
Delirium assessment was performed in two steps.Firstly, level of sedation (level of arousal) was assessed by means of the Richmond Agitation Sedation Scale (RASS) (16,17).Secondly, delirium was diagnosed by means of the CAM-ICU, which detected four features of delirium: (a) acute onset of mental status changes or a fluctuating course, (b) inattention, (c) disorganized thinking, and (d) altered level of consciousness.To be diagnosed as delirium, a patient must display both (a) and (b) and either (c) or (d).

Long-Term Follow-Up
The primary outcome was the survival state at 2 years after surgery.Long-term follow-up was performed via telephone interview.For those who died at the time of follow-up, the cause and exact date of death were recorded.
For patients who survived at 2-year followup, the following items were evaluated.Cognitive function was assessed with Chinese version of Telephone Interview for Cognitive Status-Modified (TICS-M).It evaluates 4 domains of cognitive function, i.e., 1) orientation; 2) recent memory and delayed recall (memory); 3) attention/calculation; 4) semantic memory, comprehension and repetition (language).The total score of TICS-M is 50, with high score indicating better function (18).Quality of life was assessed with the Chinese version of 36-Item Short Form Health Survey (SF-36).It evaluates 8 different domains of quality of life, i.e., physical functioning (PF), role physical (RP), bodily pain (BP), general health (GH), vitality (VT), social functioning (SF), role emotional (RE), and mental health (MH).The scores of 8 domains are calculated separately, with high score indicating better quality of life (19,20).The degrees of anxiety and depression were assessed with Self-Rating Anxiety/Depression Scales (SAS/SDS) (21).The occurrence of long-term postoperative complications was also recorded.

Statistical Analysis
Continuous variables were presented as mean ± standard deviation (SD) or median (inter-quartile range).Data were compared with independent samples t rest or Mann-Whitney U test.Categorical variables were presented as number of patients (percentage).Data were compared with chi-square test or Fisher's exact test.
To screen predictors of long-term mortality, variables that differed in univariate analyses (P< 0.05) were included in Kaplan-Meier analysis and compared with log-rank test.Cox proportional hazard model was then used to identify independent predictors, and calculated the relative risk (RR) value and 95% confidence interval (CI).
Two-sided P values of less than 0.05 were regarded as statistically significant.All statistical analyses were performed with the SPSS statistical package version 17.0 (SPSS Inc, Chicago, IL, USA).

RESULTS
Two hundred forty-three patients were enrolled in this study.Among them, two did not complete delirium assessment because of deep sedation/coma and early death after surgery.A total of 123 patients (50.6% ) developed delirium during the early postoperative period.Postoperative follow-up was performed from January 2010 to July 2010.The mean (± SD) duration from surgery to follow-up was 24.3 (± 1.0) months.Thirty patients were lost at 2-year follow-up.At last, 211 patients completed both early delirium assessment and 2-year follow-up after surgery, 190 patients completed both early delirium and 2-year quality of life assessment (Figure).
At baseline, patients who developed delirium were older, suffered more comorbid diabetes, and had lower preoperative LVEF and higher EuroSCORE.During the perioperative period, patients who developed delirium had longer durations of anesthesia/surgery, higher APACHE II score at ICU arrival, higher postoperative serum cortisol level, longer durations of sedation/mechanical ventilation, longer LOS in ICU and hospital after surgery, lower postoperative LVEF, and higher incidence of 28-day complications.At 2-year follow-up after surgery, patients who developed delirium had higher mortality rate (Tables 2 and 3).
Variables that were statistically significant in Kaplan-Meier survival analysis (P<0.05) were listed in Table 4. Univariate Cox proportional hazard model analysis identified eight factors that were possibly associated with long-term mortality.After testing for multicollinearity, four factors were excluded from multivariate Cox proportional hazard model analysis.As a result, two factors were identified as independent predictors of long-term mortality.Duration of delirium was associated with increased risk of long-term mortality in this risk-adjusted analysis (RR 1.322, 95% CI 1.074-1.628;P=0.009) (Table 4).
At 2-year follow-up after surgery, patients who developed delirium had a higher score on Self-Rating Depression Scale and a lower score on TICS-m.Furthermore, they had lower scores on physical functioning, physical role, general health and social role on SF-36 health survey (Table 5).

DISCUSSION
The results of our study showed that, in patients after CABG surgery, early postoperative delirium was a common complication.Prolonged duration of delirium was associated with increased risk of 2-year mortality.In survival patients, those who developed early postoperative delirium had worse cognitive function and quality of life than those who did not at 2 years after surgery.
In a 10-year follow-up cohort study of patients after CABG surgery, Gottesman et al. (13) found that, even after adjustment for age, duration of cardiopulmonary bypass and several vascular risk factors, the occurrence of delirium was still associated with increased risk of longterm mortality.In another study of medical (post-acute care facility) patients, Kiely et al. (22) reported that persistent delirium was an independent predictor of 1-year mortality.Our results further confirmed that, even after adjustment with confounding factors, prolonged duration of early postoperative delirium was a predictor of 2-year mortality.
The exact mechanism underlining the association between the duration of delirium and longterm mortality remains unclear.It is possible that the causes of delirium, rather than the delirium itself, lead to the increased long-term mortality.In other words, the pathophysiological conditions leading to delirium are presumably responsible for causing higher mortality rate.Another possible explanation is that persistent delirium indicates a maladaptation of patients to the environment, which then lead to a severe pathophysiologic injury and a worse long-term outcome (22).In clinical practice, it is possible to improve patients' long-term survival by decreasing the occurrence and duration of delirium.But this deserves study confirmation.
As in medical ICU patients (23), the occurrence of postoperative delirium was associated with long-term cognitive dysfunction.In elderly patients after hip surgery, Kat et al. (24) found that delirious patients had a twofold risk to develop mild dementia and cognitive dysfunction at 30-month follow-up.Koster et al. (11) reported that occurrence of delirium was associated with the development of dysfunction of memory, attention and sleep in patients after cardiac surgery.The results of our study confirmed again that long-term cognitive function was      worse in delirious patients than in non-delirious ones.Studies showed that persistent delirium was correlated with decreased brain volume and lowered white matter integrity (25,26).These might be the pathological mechanisms of the correlation between delirium duration and long-term cognitive decline.
In the present study, early postoperative delirium was associated with poor long-term quality of life.This is also confirmed in other patient populations.For patients who underwent hip fracture surgery, those who developed postoperative delirium had significantly declined daily life ability at the end of oneyear follow-up (27).For patients who underwent CABG surgery, development of delirium was associated with worse quality of life at 3 years after surgery (12).When explaining the present results, we must note that in patients who developed delirium, more were dead or could not accomplish quality of life assessment at the time of 2-year followup.And the reasons that led to withdraw from quality of life assessment might indicate poor

D
elirium is a state of transient mental disorder.According to the Diagnostic and Statistical Manual of Mental Disorders (4th edition, DSM-IV), delirium was characterized by global impairment of cognitive functions, reduced level of consciousness, abnormalities of attention, increased or decreased psychomotor activity, and disordered sleep-wake cycle.It develops over a short period of time and tends to fluctuate during the course of the day (1).

Lost at 2 -
Figure.Flow Diagram of the Study.aThese two patients died early after surgery, no delirium assessment was performed.QoL, quality of life.
as mean ± standard deviation, number of patients (percentage), or median (interquartile range).a Complications that occurred after the 28th postoperative day.The definitions of long-term complications were listed in Table1.QoL, quality of life; TICS-m, Telephone Interview for Cognitive Status-modified; SDS, Self-Rating Depression Scale; SAS, Self-Rating Anxiety Scale; SF-36, 36-Item Short-Form Health Survey.

Table 2 . Preoperative Variables. Variable
Quality of life; MI, myocardial infarction; CCS, Canadian Cardiovascular Association; COPD, chronic obstructive pulmonary disease; EuroSCORE, European System for Cardiac Operative Risk Evaluation; LVEF, left ventricular ejection fraction; NYHA, New York Heart Association.

Table 4 . Risk Factors of Postoperative Long-Term Mortality Variable
Meier analysis to estimate cumulative survival rates between groups, then log-rank test was done to calculate P value; b Univariable and multivariable, which were screened by the log-rank test (P<0.05),were analyzed by using Cox proportional hazard model (backward LR) respectively; c The definition of postoperative complications were listed in table 1; d Variable that were not included in multivariate Cox proportional hazard model.RR, relative risk; CI, confidence interval; LVEF, left ventricular ejection fraction; LOS, length of stay; ICU, intensive care unit.