Postoperative Outcomes in Elderly Patients Undergoing Cardiac Surgery With Preoperative Cognitive Impairment: A Systematic Review and Meta-Analysis : Anesthesia & Analgesia

Secondary Logo

Journal Logo

Advanced Search
Featured Articles: Meta-Analysis

Postoperative Outcomes in Elderly Patients Undergoing Cardiac Surgery With Preoperative Cognitive Impairment: A Systematic Review and Meta-Analysis

Au, Emily BSc*; Thangathurai, Gowtham BHSc†; Saripella, Aparna MSc*; Yan, Ellene HBSc*,‡; Englesakis, Marina MLIS§; Nagappa, Mahesh MBBS∥; Chung, Frances MBBS, MD, FRCPC*

Author Information

From the *Department of Anesthesia and Pain Management, Toronto Western Hospital, University Health Network, Toronto, Ontario, Canada

†Department of Medicine, McGill University, Montreal, Quebec, Canada

‡Department of Medical Science, University of Toronto, Toronto, Ontario, Canada

§Department of Library & Information Services, University Health Network, Toronto, Ontario, Canada

∥Department of Anesthesia and Perioperative Medicine, London Health Sciences Centre and St Joseph Health Care, Schulich School of Medicine and Dentistry, Western University, London, Ontario, Canada.

Accepted for publication November 15, 2022.

Funding: None.

Conflicts of Interest: See Disclosures at the end of the article.

Listen to this Article of the Month podcast and more from OpenAnesthesia.org® by visiting http://journals.lww.com/anesthesia-analgesia/pages/default.aspx.

Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s website.

Address correspondence to Frances Chung, MBBS, MD, FRCPC, Department of Anesthesia and Pain Medicine, Toronto Western Hospital, University Health Network, University of Toronto, 399 Bathurst St, Toronto, ON M5T 2S8, Canada. Address e-mail to [email protected].

Anesthesia & Analgesia 136(6):p 1016-1028, June 2023. | DOI: 10.1213/ANE.0000000000006346

KEY POINTS

  • Question: Do older cardiac surgical patients with preoperative cognitive impairment have an increased risk of adverse postoperative outcomes?
  • Findings: After cardiac surgery, preoperative cognitive impairment is associated with an increased risk for postoperative delirium, major bleeding, and a longer length of hospital/intensive care unit (ICU) stay in the older population.
  • Meaning: Routine neurocognitive screening in the preoperative setting may help identify at-risk individuals among older cardiac surgical patients, and its feasibility should be assessed in future studies.

The proportion of older patients (≥65 years of age) undergoing surgery in industrialized nations has been rising in recent decades.1,2 It is estimated that approximately 30% to 50% of older surgical patients have preoperative cognitive impairment.3 Despite the high prevalence of cognitive impairment in the aging population, baseline cognition is not routinely assessed in the preoperative setting. Cognitive impairment encompasses a spectrum of neurocognitive disorders (NCDs), ranging from mild cognitive impairment or mild NCD, in which independence with activities of daily living (ADLs) is preserved, to dementia or major NCD, in which major ADLs are significantly compromised.4 A recent systematic review and meta-analysis found that preoperative cognitive impairment in older noncardiac surgical patients was associated with increased risk of postoperative delirium (POD), 1-year mortality, 30-day readmissions, and postoperative complications.5

Cardiovascular disease has been identified as the leading cause of morbidity and mortality in the older population, with an increasing number of patients requiring cardiac surgery.6,7 Cardiac surgery varies in invasiveness and complexity, ranging from minimally invasive surgery (ie, transcatheter valve replacement [TAVR]), to open heart surgery involving coronary artery bypass graft (CABG). More invasive cardiac surgeries are associated with systemic inflammation and a high prevalence of postoperative complications.8

The overall impact of preoperative cognitive impairment on postoperative outcomes after various cardiac surgeries is not fully known. This systematic review and meta-analysis aims to evaluate the association between cognitive impairment and postoperative outcomes in older patients undergoing cardiac surgery. To minimize adverse outcomes, preoperative neurocognitive screening may help identify at-risk individuals, who require closer monitoring, additional interventions, and more excellent follow-up care.

METHODS

Study Design and Registration

The protocol for this systematic review and meta-analysis was registered through the International Prospective Register of Systematic Reviews (PROSPERO; CRD42022312594). The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guideline was used for reporting this study.9

Search Strategy

Development and implementation of the search strategy were conducted by an experienced information specialist (M.E.). MEDLINE, MEDLINE ePubs and In-Process Citations (daily), Embase & Embase Classic, Cochrane Central Register of Controlled Trials (CCTR), Cochrane Database of Systematic Reviews (CDSR), APA PsycINFO, WHO ICTRP, and ClinicalTrials.gov were searched from their inception dates to March 9, 2021, and updated January 4, 2022. The search followed the Cochrane Handbook10 and the Cochrane Methodological Expectations of Cochrane Intervention Reviews (MECIR)11 for searching, the PRISMA 2020 guidelines9 for reporting, the PRISMA-S12 extension for searches, and the PRESS guideline for peer-reviewing the search strategies.13

Preliminary searches were conducted, and full-text literature was mined for keywords and appropriate controlled vocabulary terms. The search strategy concept blocks were built using the following topics: (Preoperative) AND (Cognitive assessment) AND (Elderly) AND (Postoperative or Surgery) AND (Functional Outcomes), using both free-text and controlled vocabulary terms. Database search results were limited to the English language, humans, and adults. All conference materials were excluded. Furthermore, forward and backward citation searching was completed using Google Scholar. Duplicates were removed using EndNote X8. The detailed search strategy is provided in Supplemental Digital Content 1, Appendix 1, https://links.lww.com/AA/E172.

Included studies were divided into 2 groups, preoperative cognitive impairment or dementia, and data were analyzed separately. Preoperative cognitive impairment was identified using a validated neurocognitive screening tool. In contrast, dementia was recognized by International Classification of Diseases (ICD) coding or practitioner assessment with the Diagnostic and Statistical Manual of Mental Disorders, 4th ed text revision (DSM-IV-TR).

Study Selection and Data Extraction

Title and abstract screening were independently completed on Rayyan by 2 reviewers (E.A. and G.T.).14 Full-text screening and data extraction were independently conducted by 2 reviewers (E.A. and G.T.). Any disagreements were resolved by another reviewer (A.S.). The inclusion criteria were: (1) older patients (≥60 years of age); (2) presence of preoperative cognitive impairment (as detected using a validated screening tool, eg, Mini-Mental State Examination [MMSE] and Montreal Cognitive Assessment [MoCA]); (3) patients undergoing cardiac surgery; (4) at least one reported postoperative complication; (5) comparator group with no cognitive impairment was present; and (6) English language. Conference abstracts, case series, and case reports were excluded.

Study characteristics and outcomes of interest were independently collected using a standardized Excel sheet. Patient and study characteristics collected during the data extraction phase included the year of publication, study design, type of surgery, sample size, age, sex, the prevalence of preoperative cognitive impairment, and type of diagnostic tool. The outcomes included any postoperative complications (ie, cardiac complications, pulmonary complications, renal impairment, infection, bleeding, hospital length of stay [LOS], postoperative cognitive dysfunction [POCD], POD, mortality, hospital readmissions, etc) in older patients with preoperative cognitive impairment and undergoing cardiac surgery (Supplemental Digital Content 1, Table 1, https://links.lww.com/AA/E172).

Quality Assessment of Studies

The quality of each included study was critically evaluated using the Newcastle-Ottawa Scale (NOS) and the Meta-Analysis of Observational Studies in Epidemiology (MOOSE) checklist.15,16 The NOS assessed the risk of bias in nonrandomized studies across 3 broad domains: selection of cohorts, comparability of cohorts, and outcome assessment. A total of 8 question items were evaluated using a star system, rendering a maximum score of 9. The study was considered good quality when the assigned score was ≥8. The study was regarded as moderate quality when the assigned score was equal to 5, 6, or 7. The study was considered poor quality when the assigned score was ≤4. The MOOSE checklist assessed study design, study population identification, outcome and outcome assessment definition, selective loss during follow-up, and identification of important confounding and prognostic variables. Two reviewers independently appraised the assessment of study quality (G.T. and E.Y.). A third reviewer (A.S.) resolved conflicts between the 2 reviewers.

GRADE: Quality of Evidence

The Grading of Recommendations Assessment, Development, and Evaluation (GRADE)17 method was used to assess the quality of evidence.17 The GRADE system includes the risk of bias, inconsistency, indirectness, inaccuracy, and publication bias. The baseline rating of GRADE for randomized controlled trials (RCTs) is high (4 points) and low (2 points) for observational studies. The outcome rating is adjusted or downgraded after considering the 5 assessment criteria.

Statistical Analysis

Qualitative analysis was performed for the included studies. We created tables for demographics, study characteristics, and postoperative complications. For quantitative analysis, we determined odds ratios (ORs) with a 95% confidence interval (CI) for dichotomous outcomes (POD, postoperative complications, and mortality) and the standardized mean difference (SMD) with 95% CI for the continuous outcome: length of hospital stay (LOS) and intensive care unit length of stay (ICU LOS). Meta-analyses were performed when we had a minimum of 3 or more appropriate studies and were illustrated in the form of forest plots. We included studies with different types of cardiac surgeries (eg, CABG, transcatheter aortic valve implantation or replacement [TAVI or TAVR], and surgical aortic valve replacement [SAVR]). Trial sequential analysis (TSA) was performed to assess the risk of type 1 errors.18

F1
Figure 1.:
PRISMA flow diagram. PRISMA indicates Preferred Reporting Items for Systematic Reviews and Meta-Analyses
Table 1. - Demographics and Study Characteristics of Cognitively Impaired and Dementia Patients Undergoing Elective Cardiac Surgeries
First author, year, country Type of surgery Study design Total patients (N) Age (y) Sex (female)‚ n (%) Preoperative cognitive impairment prevalence‚ n (%) Cognitive assessment test and cutoff
Cognitive impairment
Mixed cardiac surgery
 Itagaki et al, 22 2020, Japan Mixed a RC 89 74.9 ± 5.5 32 (35.9) 55 (61.8) MoCA <26
 Boureau et al, 23 2018, France b Mixed RC 197 81.35 ± 3.5 102 (51.8) 29 (14.7) MMSE <24
 Terazawa et al, 24 2018, Japan Mixed PC 490 71.9 ± 7.2 160 (32.7) 51 (10.4) MMSE <24
 Harrington et al, 25 2011, USA Mixed RC 181 68.1 ± 10.7 2 (1.0) 52 (29.0) CIB ≤5
 Maekawa et al, 26 2011, Japan Mixed a PC 362 69.5 ± 9.7 141 (39.0) 40 (11.0) HDS <24
 Veliz-Reissmüller et al, 27 2007, Sweden Mixed PC 107 71.6 ± 6.0 41 (38.3) 4 (3.7) MMSE <25
CABG surgery
 Kazmierski et al, 28 2014, UK c CABG PC 113 64 ± 8.9 23 (20.4) CI: 28 (24.8); dementia: 6 (5.3) CI: MoCA <26; dementia: DSM-IV-TR criteria
 Aykut et al, 29 2013, Turkey CABG PC 48 71.4 ± 3.7 26 (54.0) 25 (52.0) MoCA <26
 Baba et al, 30 2007, Japan CABG d PC 218 71.2 ± 5.5 66 (30.3) 14 (6.4) HDS <24
 Millar et al, 31 2001, UK CABG PC 81 76.4 ± 24.4 17 (21.0) 13 (16.0) Stroop neurological screening test
TAVI/TAVR
 Luque et al, 32 2021, Spain TAVR PC 501 83.5 ± 5.0 289 (57.7) 18 (3.6) MMSE <24
 Kapadia et al, 33 2020, USA TAVR PC 142 84.2 ± 6.0 69 (48.6) 46 (32.4) MMSE <24
 Saji et al, 34 2019, Japan TAVR PC 455 84.5 ± 5.0 309 (67.9) 177 (38.9) HDS-R <24
 Yanagisawa et al, 35 2018, Japan TAVI PC 1111 85.0 ± 5.2 791 (71.2) 420 (38.0 MMSE <25
 Schoenenberger et al, 36 2016, Switzerland TAVI PC 279 (cognitive impairment data available for 229) 83.5 ± 4.3 161 (57.7) 63 (22.6) MMSE <26
Dementia
TAVR
 Jain et al, 23 2021, USA TAVR RC 57,805 82.3 ± 6.4 27,685 (47.9) 2910 (5.0) Previous dementia diagnosis (ICD codes)
Abbreviations: CABG, coronary artery bypass graft surgery; CIB, Clock-in-the-Box; CPB, cardiopulmonary bypass; DSM-IV-TR, Diagnostic and Statistical Manual of Mental Disorders, 4th ed text revision; HDS, Hasegawa Dementia Score; HDS-R, Hasegawa Dementia Score revised; ICD, International Classification of Diseases; MMSE, Mini-Mental Status Examination; MoCA, Montreal Cognitive Assessment; PC, prospective cohort; RC, retrospective cohort; TAVI, transcatheter aortic valve implantation; TAVR, transcatheter aortic valve replacement.
aMixed surgery: CABG, valve repair/replacement, and thoracic aneurysm repair.
bIncludes some percentage of emergency surgeries.
cIncludes some percentage of dementia patients.
dIncludes some percentage of off-pump CABG surgery.

Cochrane Review Manager version 5.4 was used with a random-effects model to conduct the meta-analysis.19 Mantel-Haenszel (M-H) method combined dichotomous events, and the inverse variance method was used for continuous events. To measure whether stability in solutions is achieved on the significant outcome, a Bonferroni test and false discovery rate were calculated. Leave-one-out meta-analysis was performed to evaluate the effect of each included study on the combined estimates. The I2 statistic was calculated for each outcome to quantify the heterogeneity across the included studies. Publication bias was assessed by the rank correlation test by Begg and Mazumdar and Egger’s regression asymmetry test.20,21

RESULTS

Search Strategy

Our search strategy retrieved 15,058 articles for review (Figure 1). After removing 4429 duplicate studies, abstract and title screening was performed, yielding 73 full-text articles. Of these, 57 articles were excluded for various reasons, such as a lack of preoperative cognitive impairment and postoperative complications, noncardiac surgery, and wrong population. Sixteen articles met our inclusion criteria and were included in the qualitative synthesis.22–37 Eight articles were included in the quantitative synthesis.22,24,25,27,28,32,34,35

Study Characteristics

Study characteristics and demographic data of the included studies are summarized in Table 1. Studies were conducted in Japan (n = 6),22,24,26,30,34,35 the United States (n = 3),25,33,37 the United Kingdom (n = 2),28,31 France (n = 1),23 Sweden (n = 1),27 Turkey (n = 1),29 Spain (n = 1),32 and Switzerland (n = 1).36 Study design consisted of prospective cohort studies (n = 12)24,26–36 and retrospective cohorts (n = 4).22,23,25,37 All studies were elective cardiac surgery, with one including a small number of emergency cardiac surgeries.23 In total, there are 16 studies with 62,179 patients. The mean age (SD) was 82.1 ± 6.7 years, and 51.8% were men. Cognitive impairment and dementia were identified in 25.1% (1028 of 4089) and 5.0% (2916 of 57‚918) of patients, respectively. Studies were classified as mixed surgery (n = 6),22–27 CABG (n = 4),28–31 and TAVI/TAVR (n = 6).32–37

Measures of Cognitive Impairment and Dementia

Cognitive impairment was investigated in 15 studies, and dementia was investigated in 2. Cognitive impairment was evaluated using several validated cognitive screening instruments, including the MMSE (n = 7),23,24,27,32,33,35,36 the Hasegawa Dementia Score (HDS; n = 3),26,30,34 the MoCA (n = 3),22,28,29 the Stroop neurological screening test (n = 1),31 and the Clock-in-the-Box (CIB; n = 1).25 Dementia was identified using the DSM-IV-TR criteria (n = 1)28 and from a previous formal diagnosis using ICD coding (n = 1).37 The prevalence of cognitive impairment ranged from 3.7% to 61.8%, whereas the prevalence of dementia ranged from 5.0% to 5.3%.

Quality Assessment

Quality assessment of each included study is shown in Supplemental Digital Content 1, Table 2, https://links.lww.com/AA/E172. The NOS scores ranged from 6 to 9 out of a maximum score of 9 across all observational cohort studies. The included studies scored well in the following domains: representativeness of exposed cohort, selection of nonexposed cohort, ascertainment of exposure, comparability of cohorts based on design or analysis, follow-up for outcomes, and adequacy of follow-up of cohorts. While most studies scored 2 stars for the comparability domain, 1 received 0 stars.29 Only 5 studies demonstrated that the outcome of interest was present before study onset,22,23,25,34,37 and 7 adopted adequate outcome assessment.22,23,28,31–33,36 Following the MOOSE checklist, all studies identified their study population and defined their outcome of interest and outcome assessment. Most studies identified and adjusted for confounding and pertinent prognostic variables and did not demonstrate a selective loss of participants during the follow-up periods.

GRADE Evaluation

GRADE evaluation of the quality of evidence was performed for the following outcomes: POD, major postoperative bleeding, ICU LOS, hospital LOS, and 30-day mortality (Supplemental Digital Content 1, Table 3, https://links.lww.com/AA/E172). All the included studies were observational cohorts, and the quality of evidence was rated low for hospital LOS, ICU LOS, and bleeding outcomes. POD and 30-day mortality were rated very low due to wide CIs.

Postoperative Outcomes in Cognitive Impairment and Dementia Studies

Postoperative Delirium

Five studies (1300 patients) reported on the association between cognitive impairment and POD (Supplemental Digital Content 1, Table

4a, https://links.lww.com/AA/E172).22,24,27,28,32 POD was evaluated using a variety of measures, such as the confusion assessment method (CAM),27,32 CAM-ICU,28 and the Intensive Care Delirium Screening Checklist (ICDSC) (Supplemental Digital Content 1, Table 4a, https://links.lww.com/AA/E172).22 A meta-analysis of 3 cognitive impairment studies (715 patients) using CAM or CAM-ICU showed that preoperative cognitive impairment (50 patients) was associated with higher rates of POD after cardiac surgery (70.0% vs 20.5%; OR, 8.35; 95% CI, 4.25–16.38; I2, 0%; P < .00001) (Figure 2A).27,28,32

F2
Figure 2.:
Association between cognitive impairment and postoperative outcomes after cardiac surgery. A, Postoperative delirium. B, Postoperative bleeding. C, Intensive care unit—length of stay. D, Hospital length of stay. E, 30-d mortality. CI indicates confidence interval.

Two studies (57,918 patients) assessed the impact of dementia on POD with CAM and ICD codes (Supplemental Digital Content 1, Table 4a, https://links.lww.com/AA/E172).28,37 Dementia was associated with a higher prevalence of POD compared to patients with normal cognition in one study (7.4% vs 3.6%; P < .01).37

We confirmed our results by Bonferroni correction of alpha (α = 0.01), and the corrected probability related to the false discovery rate was 4.9%. We recalculated the CI by increasing it to 99%, which widened a little (3.44 to 20.25) but did not impact the final inference of our results. After Bonferroni corrections, our results remained significant, with a P value of <.00001 for both CIs. There was no evidence for publication bias, as confirmed by Begg’s test (P = .296) or Egger’s test (P = .380). Finally, we did a TSA to calculate the required information size (RIS). TSA with a type 1 error of 5% and type 2 error of 20% demonstrated that the RIS is 589 participants (data not shown in the figure).

Postoperative Complications

Five studies (2466 patients) evaluated cognitive impairment and postoperative complications after cardiac surgery (Table 2).24,26,29,34,35 The main postoperative complications reported were major bleeding (n = 3),24,34,35 stroke (n = 2),26,34 acute kidney injury (n = 2),34,35 vascular complications (n = 1),34,35 and pneumonia (n = 2)24,29 (Supplemental Digital Content 1, Table 4b, https://links.lww.com/AA/E172). A meta-analysis of 3 studies (2056 patients) indicated an association between preoperative cognitive impairment (648 patients) and increased postoperative major bleeding (12.0% vs 7.2%; OR, 1.46; 95% CI, 1.06–2.02; I2, 0%; P = .02) (Figure 2B).24,34,35

Table 2. - Other Postoperative Outcomes.
First author, year, country Type of surgery Cognitive impairment‚ % No cognitive impairment‚ % Postoperative major complications Postoperative delirium LOS Nonhome discharge Cognitive deterioration
Itagaki et al, 22 2020, Japan Mixed cardiac 61.8 31.9 NR OR, 3.11 (1.04–9.32) NR NR NR
Terazawa et al, 24 2018, Japan Mixed cardiac 10.4 89.6 OR, 3.02 (1.30–7.03) a OR, 2.26 (1.04–1.13) a NR OR, 2.39 (1.07–5.35) a NR
Harrington et al, 25 2011, USA Mixed cardiac 29.0 71.0 NR NR NR HR, 0.93 (0.89–0.98) a NR
Maekawa et al, 26 2011, Japan Mixed cardiac 11.0 89.0 NR NR NR NR NR
Veliz-Reissmüller et al, 27 2007, Sweden Mixed cardiac 3.7 96.3 NR NR NR NR NR
Kazmierski et al, 28 2014, UK CABG 24.8 75.2 NR OR, 6.33 (31.93–20.76) a NR NR NR
Aykut et al, 29 2013, Turkey CABG 52.0 48.0 NR NR NR NR NR
Baba et al, 30 2007, Japan CABG 6.4 93.6 NR NR NR NR NR
Millar et al, 31 2001, UK CABG 16.0 84.0 NR NR NR NR NR
Luque et al, 32 2021, Spain TAVR 3.6 96.4 NR OR, 4.17 (1.11–15.71) a NR NR NR
Kapadia et al, 33 2020, USA TAVR 32.4 67.6 NR NR NR NR NR
Schoenenberger et al, 36 2016, Switzerland TAVI 22.6 77.4 NR NR NR NR OR, 0.87 (0.34–2.26) a
Jain et al, 37 2021, United States TAVR 5.0 95.0 NR OR, 2.13 (1.26–3.61) OR, 1.11 (1.03–1.19) OR, 2.27 (1.67–3.08) NR
Mixed cardiac surgery: CABG, valve repair/replacement, and thoracic aneurysm repair.
Abbreviations: CABG; coronary artery bypass graft surgery; LOS, length of stay; NR, not reported; OR, odds ratio (95% confidence interval); HR, hazard ratio (95% confidence interval); TAVI, transcatheter aortic valve implantation; TAVR, transcatheter aortic valve replacement.
aMultivariate analysis.

The association between dementia and postoperative complications was reported in one study (57,805 patients) (Supplemental Digital Content 1, Table 4b, https://links.lww.com/AA/E172).37 Of the postoperative complications (eg, stroke, acute kidney injury, and vascular complications), dementia was only associated with increased postoperative major bleeding (14.7% vs 8.6%; P < .01).

Compared to the 95% CI (1.06–2.02), the 99% CI for postoperative bleeding slightly widened (0.96–2.23), impacting the final inference of our results. The lower CI slightly crossed the null value of 1, suggesting the effect was insignificant. Similarly, after the Bonferroni correction of alpha (α = 0.01), the postoperative bleeding was not statistically significant (P = .02). There was no evidence for publication bias, as confirmed by Begg’s test (P = .601) or Egger’s test (P = .666). In TSA, keeping the risk of type 1 and type 2 errors at 5% and 20%, respectively, and the current incidence of events at 12% and 7%, respectively, in the cognitive impairment and normal cognition groups, the current sample size (2056) almost reached the required information sample size of 2356 (not shown in the figure).

Length of Stay

Five studies (1328 patients) reported on the association between cognitive impairment and LOS (Table 2).22,24,25,28,34 LOS outcomes included both ICU LOS and hospital LOS (Supplemental Digital Content 1, Table 4c, https://links.lww.com/AA/E172). The pooled analysis of 4 studies (1120 patients) demonstrated that cognitive impairment was associated with a significant increase in ICU LOS (days; SMD, 0.39; 95% CI, 0.09–0.68; I2, 70%; P = .01) (Figure 2C).22,24,28,34 Leave-one-out meta-analysis showed that Kazmierski et al28 contributed to the maximum heterogeneity, which decreased to 0% after it was removed. For hospital LOS, a meta-analysis of 5 studies (1301 patients) indicated an association between cognitive impairment (363 patients) and longer hospital LOS (SMD, 0.36; 95% CI, 0.20–0.51; I2, 22%; P < .00001) (Figure 2D).22,24,25,28,34 Leave-one-out analysis showed that the removal of Kazmierksi et al or Saji et al decreased heterogeneity to 0%.28,34 One study (57,805 patients) evaluated dementia and hospital LOS (Table 2).37 Dementia was associated with longer hospital LOS compared to cognitively normal patients (6.75 ± 0.07 days vs 6.11 ± 0.06; P < .01) (Supplemental Digital Content 1, Table 4c, https://links.lww.com/AA/E172).

For hospital LOS, we confirmed our results by Bonferroni correction of alpha (α = 0.01), and the corrected probability related to the false discovery rate was 4.9%. No evidence for publication bias was found by Begg’s test (P = .220) or Egger’s test (P = .172) for length of stay. We found the RIS that estimates the sample size needed in case the effect curve does not cross any bounds was 575. We set the risk of type 1 and type 2 errors to 5% and 20% respectively. We set the mean difference as 2.69 in TSA and used the TSA viewer software version 0.9.5.5 beta for analysis.

Similarly for ICU LOS, we confirmed our results by Bonferroni correction of alpha (α= 0.01), and the corrected probability related to the false discovery rate was 4.9%. The 99% CI slightly widened (0.00–0.77) without impacting the final inference of our results. After Bonferroni corrections, our results remained significant (P = .01) for both CIs. No evidence for publication bias was found by Begg’s test (P = .174) or Egger’s test (P = .296) for ICU LOS. We found the RIS that estimates the sample size needed in case the effect curve does not cross any bounds was 1683. We set the mean difference as 0.83 in TSA, while the risk of type 1 and type 2 errors to 5% and 20%, respectively (not shown in the figure).

Postoperative Mortality

Five studies (2532 patients) assessed postoperative mortality in patients with cognitive impairment after cardiac surgery (Table 3).23,24,34–36 The mortality outcomes were measured at various time intervals: in-hospital (n = 2),24,35 30 days (n=3),24,34,35 3 months (n = 1),35 6 months (n = 2),35,36 9 months (n = 1),35 1 year (n = 2),23,35 and 3 years (n = 1) (Supplemental Digital Content 1, Table 4d, https://links.lww.com/AA/E172).34 Meta-analysis of 3 studies (2056 patients) demonstrated no significant association between cognitive impairment and 30-day mortality (1.69% vs 1.06%; OR, 2.58; 95% CI, 0.64–10.44; I2, 55%; P = .18) (Figure 2E).24,34,35

Table 3. - Mortality Outcomes of Included Studies.
First author, year, country Cognitive impairment‚ % No cognitive impairment‚ % In-hospital morality 30-day mortality 6-month mortality 1-year mortality
Terazawa et al, 24 2018, Japan 10.4 89.6 OR, 5.72 (1.80–18.17) NR NR NR
Boureau et al, 23 2018, France 14.7 85.3 NR NR NR HR, 2.47 (0.87–7.01)
Saji et al, 34 2019, Japan 38.9 61.1 NR HR, 2.11 (1.21–3.69) a NR NR
Yanagisawa et al, 35 2018, Japan 38.0 62.0 NR NR NR HR, 2.1 (1.10–4.00) a
Schoenenberger et al, 36 2016, Switzerland 22.6 77.4 NR NR NR NR
Jain et al, 37 2021, United States 5.0 95.0 OR, 1.26 (0.57–2.79) NR NR NR
Abbreviations: CI‚ confidence interval; HR, hazard ratio (95% CI); NR, not reported; OR, odds ratio (95% CI).
aMultivariate analysis.

Compared to the 95% CI (0.64–10.44), the 99% CI for 30-day mortality slightly widened (0.41–16.20), and the final inference of our results remained nonsignificant. After the Bonferroni correction of alpha (α = 0.01), the 30-day mortality remained statistically nonsignificant (P = .18). There was no evidence for publication bias, as confirmed by Begg’s test (P = .601) or Egger’s test (P = .450). In the TSA analysis, the risk of type 1 and type 2 errors was kept at 5% and 20%, respectively. The incidence of 30-day mortality was 1.69% and 1.06% in the cognitive impairment and normal cognition groups, respectively. The current sample size (n = 2056) remained far behind the RIS size of 4608 (not shown in the figure).

Leave-one-out meta-analysis showed that Yanagisawa et al35 contributed to the maximum heterogeneity, its removal decreasing heterogeneity to 0%. One study (57,805) examined the association between dementia and in-hospital mortality (Table 2).37 No significant associations were found between dementia and in-hospital mortality (Supplemental Digital Content 1, Table 4d, https://links.lww.com/AA/E172).

Other Postoperative Outcomes

Secondary outcomes are outlined in Table 2 and Supplemental Digital Content 1, Table 4a–f, https://links.lww.com/AA/E172. Postoperative outcomes included postoperative cognitive dysfunction (POCD; n = 4),26,30,31,36 functional status (n = 1),33 and discharge outcomes (n = 2).24,25 One study found an association between cognitive impairment and increased incidence of POCD at 1 week (38% vs 19%, P = .006),26 whereas another study found that cognitive impairment was not significantly associated with POCD at 6 months.36 For discharge outcomes, cognitive impairment was associated with an increased prevalence of discharge to rehabilitation facilities than cognitively normal individuals (31.4% vs 14.6%, P = .005).24 Dementia was associated with significantly higher discharge rates to rehabilitation facilities compared to normal cognition (Table 2).37

DISCUSSION

Our systematic review and meta-analysis evaluated the evidence on adverse postoperative outcomes in older cardiac surgical patients with preoperative cognitive impairment. We pooled data from 16 prospective and retrospective studies, which included 62,179 patients. The prevalence of preoperative cognitive impairment ranged from 3.7% to 61.8%. Preoperative cognitive impairment was associated with an 8-fold increased risk of POD in older patients undergoing cardiac surgery. In contrast, preoperative cognitive impairment was associated with a 5% increase in absolute risk of major postoperative bleeding and an increase in hospital and ICU LOS by around 0.4 days.

We found that preoperative cognitive impairment is associated with an 8-fold increased risk of POD after cardiac surgery, much higher than 4-fold higher odds of POD in noncardiac surgery.5 The prevalence of delirium ranges from 26% to 52% after cardiac surgery.38 Delirium is theorized to result from a breakdown of functional neural networks.39–41 Patients with impaired baseline network connectivity that are exposed to events that further disrupt network connectivity can precipitate delirium. Previous research has identified an association between cognitive impairment, reduced functional connectivity, and reduced white matter integrity.42,43 Inflammation, metabolic abnormalities, and certain medications also cause an acute breakdown in network connectivity.39 Cardiac surgery causes substantial systemic inflammation that can precipitate the development of delirium in patients with preoperative cognitive impairment.8 Additionally, nutritional deficiencies and higher cortisol concentrations in cognitively impaired individuals have been associated with the development of delirium.28,32

The importance of recognizing risk factors for POD and optimizing them perioperatively cannot be understated as POD has been associated with higher morbidity and mortality, cognitive dysfunction, decreased quality of life, increased hospital LOS, and increased costs for the health care system.44–47

Major bleeding has been identified as a significant complication in both open and closed cardiac surgery,48,49 estimated at 2% to 10% incidence and associated with considerable mortality.50,51 We found that preoperative cognitive impairment is associated with a 5% absolute increase in the risk of major bleeding after cardiac surgery. Although no clear mechanistic link exists, it has been hypothesized that medication noncompliance, interference in anticoagulation levels due to polypharmacy-related drug interactions, or elevated risk of falls in cognitively impaired patients may lead to substantial bleeding.52

In our review, preoperative cognitive impairment was associated with increased hospital and ICU LOS after cardiac surgery. Previous research has identified hemodynamic instability and organ dysfunction as major causes of prolonged LOS after cardiac surgery, with approximately 10% of cardiac surgical patients requiring lengthy postoperative care.53 Prolonged hospital LOS has been associated with higher rates of hospital-acquired infections, decreased patient satisfaction, and increased hospital costs.54 Similarly, prolonged ICU LOS has been associated with lower long-term survival rates, reduced quality of life, and higher resource consumption.55,56

We did not find an association between preoperative cognitive impairment and increased 30-day mortality after cardiac surgery. This is consistent with a previous meta-analysis on preoperative cognitive impairment and noncardiac surgery.5 However, the sample size of 2056 patients may be too small for this outcome. It has been proposed that patients may be more closely followed in the early postoperative period, which may account for the lack of difference in 30-day mortality.5 While 30-day mortality has been validated as a hospital performance metric, long-term postoperative mortality following cardiac surgery may be underestimated.57,58 Because of a small number of studies reporting longer mortality outcomes, we could not perform a meta-analysis. Nevertheless, 2 studies found an association between cognitive impairment and increased 1-year and 3-year mortality.34,35

Implications for Perioperative Care

A large proportion of cardiac surgical patients consist of older individuals. The prevalence of preoperative cognitive impairment was approximately 30% to 50% in older surgical patients.3 To date, preoperative screening is not a part of routine practice. A recent survey of American Society of Anesthesiologists members suggested that the respondents offer preoperative cognitive impairment screening to <10% of surgical patients.59 In cardiac surgery, the European System of Cardiac Operative Risk Evaluation (EuroSCORE) and the Society of Thoracic Surgeons (STS) score are widely used to assess the risk of perioperative mortality.60,61 Although these models include important predictor variables (eg, comorbidities, patient-related, cardiac-related, and operation-related variables), they fail to incorporate risk factors associated with adverse postoperative outcomes, such as cognitive impairment or frailty.59–60

Potential benefits can be derived from establishing routine preoperative neurocognitive testing. Screening presents an opportunity to stratify patients based on the risk of adverse outcomes and utilize targeted therapies perioperatively to potentially mitigate these negative effects.62 Previous research has identified successful perioperative interventions that reduce the occurrence of adverse outcomes, such as early geriatric consultation, avoidance of polypharmacy, orientation protocols, perioperative sleep enhancement, adequate nutrition, and visual or hearing aids.63,64 Moreover, identifying patients at risk for poor postoperative outcomes with cognitive testing may assist physicians with clinical decision-making and better inform patients and caregivers about the risks associated with surgery and recovery.

Recent studies have begun to evaluate the use of multicomponent preoperative assessments that incorporate neurocognitive testing. One example is the comprehensive geriatric assessment (CGA), which assesses the physiological, social, psychological, and functional status of older patients.65 Although lower CGA assessment scores have been associated with increased postoperative complications, current CGA models that combine CGA assessment and optimization have not been associated with improved postoperative outcomes.66,67 In 2017, the European Society of Anesthesiology published evidence-based and consensus-based guidelines on POD, calling for clinicians to implement preoperative POD risk assessments, with cognitive impairment being the most important risk factor.68

Considering the growing body of evidence demonstrating associations between cognitive impairment and adverse postoperative outcomes in older surgical patients, future research should focus on evaluating the feasibility of implementing neurocognitive screening into routine preoperative care. A greater number of studies are needed to assess various screening tools based on metrics such as ease of use, clinical utility, and whether their establishment may favorably affect postoperative outcomes.

Limitations

There are several limitations in this systematic review and meta-analysis. First, there was moderate heterogeneity among the included studies due to variability in study design (eg, different neurocognitive screening tools with varying cutoff scores) and types of cardiac surgical procedures. Second, we could not perform meta-analyses for certain postoperative outcomes (eg, mortality and POCD) due to a limited number of studies. Results were often reported with varying time frames making pooling impossible. Similarly, we could not perform subgroup analyses based on the type of surgery (eg, TAVR and cardiopulmonary bypass), which may have yielded important conclusions due to significant differences in invasiveness and perioperative complications associated with such procedures. Third, GRADE evaluation of the postoperative outcomes consisted of very low-to-low-quality of evidence. Fourth, it is possible that some patients would have stayed for a longer time in the hospital than others. This may have caused the skewed distribution in the LOS outcome. Finally, our search strategy included only English-language publications, which may have introduced bias.

CONCLUSIONS

In older patients undergoing cardiac surgery, cognitive impairment was associated with 8-fold increased risk of delirium, a 5% increase in absolute risk of major postoperative bleeding, and an increase in hospital and ICU LOS by approximately 0.4 days. Given the higher incidence of cardiovascular disease and cardiac surgery in older patients, the feasibility of implementing routine neurocognitive screening should be evaluated.

DISCLOSURES

Name: Emily Au, BSc.

Contribution: This author helped with study conception and design, acquiring and interpreting study data, and writing the manuscript.

Conflicts of Interest: None.

Name: Gowtham Thangathurai, BHSc.

Contribution: This author helped acquire and interpret study data and write the manuscript.

Conflicts of Interest: None.

Name: Aparna Saripella, MSc.

Contribution: This author helped acquire study data and with statistical analysis.

Conflicts of Interest: None.

Name: Ellene Yan, HBSc.

Contribution: This author helped acquire study data.

Conflicts of Interest: None.

Name: Marina Englesakis, MLIS.

Contribution: This author helped develop the search strategy.

Conflicts of Interest: None.

Name: Mahesh Nagappa, MBBS.

Contribution: This author helped with statistical analysis and writing the manuscript.

Conflicts of Interest: M. Nagappa reports grants from the Academic Medical Organization of Southwestern Ontario Opportunity Fund, Ontario Ministry of Health and Long-Term Care Innovation Fund, and Lawson Research Fund.

Name: Frances Chung, MBBS, MD, FRCPC.

Contribution: This author helped with study conception and design, interpreting study data, and writing the manuscript.

Conflicts of Interest: F. Chung reports research support from the University Health Network Foundation, ResMed Foundation, consultant to Takeda Pharma, Up-to-Date royalties, and STOP-Bang proprietary to University Health Network.

This manuscript was handled by: Stefan G. De Hert, MD.

GLOSSARY

ADL
activity of daily living
CABG
coronary artery bypass graft
CAM
confusion assessment method
CCTR
Cochrane Central Register of Controlled Trials
CDSR
Cochrane Database of Systematic Reviews
CGA
comprehensive geriatric assessment
CI
confidence interval
CIB
Clock-in-the-Box
CPB
cardiopulmonary bypass
DSM-IV-TR
Diagnostic and Statistical Manual of Mental Disorders 4th ed text revision
EuroSCORE
European System of Cardiac Operative Risk Evaluation
GRADE
Grading of Recommendations Assessment, Development, and Evaluation
HDS
Hasegawa Dementia Score
HDS-R
Hasegawa Dementia Score revised
HR
hazard ratio
ICD
International Classification of Diseases
ICDSC
Intensive Care Delirium Screening Checklist
ICU
intensive care unit
LOS
length of stay
MECIR
Methodological Expectations of Cochrane Intervention Reviews
M-H
Mantel-Haenszel method
MMSE
mini-mental state examination
MoCA
Montreal Cognitive Assessment
MOOSE
Meta-Analysis of Observational Studies in Epidemiology
NCD
neurocognitive disorder
NOS
Newcastle-Ottawa Scale
NR
not reported
OR
odds ratio
PC
prospective cohort
POCD
postoperative cognitive dysfunction
POD
postoperative delirium
PRISMA
Preferred Reporting Items for Systematic Reviews and Meta-Analyses
PROSPERO
International Prospective Register of Systematic Reviews
RC
retrospective cohort
RCT
randomized controlled trial
RIS
required information size
SAVR
surgical aortic valve replacement
SD
standard deviation
SMD
standardized mean difference;
STS
Society of Thoracic Surgeons
TAVI
transcatheter aortic valve implantation
TAVR
transcatheter valve replacement; TSA = trial sequential analysis

REFERENCES

1. Etzioni DA, Liu JH, Maggard MA, Ko CY. The aging population and its impact on the surgery workforce. Ann Surg. 2003;238:170–7.
2. Fowler AJ, Abbott TEF, Prowle J, Pearse RM. Age of patients undergoing surgery. Br J Surg. 2019;106:1012–1018.
3. Kapoor P, Chen L, Saripella A, et al. Prevalence of preoperative cognitive impairment in older surgical patients: a systematic review and meta-analysis. J Clin Anesth. 2022;76:110574.
4. Sachs-Ericsson N, Blazer DG. The new DSM-5 diagnosis of mild neurocognitive disorder and its relation to research in mild cognitive impairment. Aging Ment Health. 2015;19:2–12.
5. Chen L, Au E, Saripella A, et al. Postoperative outcomes in older surgical patients with preoperative cognitive impairment: a systematic review and meta-analysis. J Clin Anesth. 2022;80:110883.
6. Torbiak L, Stammers A, van Ineveld C, Arora RC. Cardiac surgery in the older adult. Rosenthal RA, Zenilman ME, Katlic MR, eds. In: Principles and Practice of Geriatric Surgery. Springer International Publishing; 2019:1–32.
7. Roger VL, Go AS, Lloyd-Jones DM, et al.; American Heart Association Statistics Committee and Stroke Statistics Subcommittee. Heart disease and stroke statistics—2012 update. Circulation. 2012;125:e2–e220.
8. van Ton AMP, Duindam HB, van Tuijl J, et al. Neuroinflammation in cognitive decline post-cardiac surgery (the FOCUS study): an observational study protocol. BMJ Open. 2021;11:e044062.
9. Moher D, Shamseer L, Clarke M, et al.; PRISMA-P Group. Preferred reporting items for systematic review and meta-analysis protocols (PRISMA-P) 2015 statement. Syst Rev. 2015;4:1.
10. Higgins JP, Thomas J, Chandler J, et al. Cochrane handbook for systematic reviews of interventions. John Wiley & Sons; 2019.
11. Higgins JPT, Lasserson T, Chandler J, Tovey D, Churchill R. Methodological Expectations of Cochrane Intervention Reviews (MECIR). Accessed October 1, 2022. https://community.cochrane.org/sites/default/files/uploads/MECIR%20PRINTED%20BOOKLET%20FINAL%20v1.01.pdf.
12. Rethlefsen ML, Kirtley S, Waffenschmidt S, et al.; PRISMA-S Group. PRISMA-S: an extension to the PRISMA Statement for Reporting Literature Searches in Systematic Reviews. Syst Rev. 2021;10:39.
13. McGowan J, Sampson M, Salzwedel DM, Cogo E, Foerster V, Lefebvre C. PRESS peer review of electronic search strategies: 2015 guideline statement. J Clin Epidemiol. 2016;75:40–46.
14. Ouzzani M, Hammady H, Fedorowicz Z, Elmagarmid A. Rayyan—a web and mobile app for systematic reviews. Syst Rev. 2016;5:210.
15. Wells GA, Shea B, O’Connell D, et al. The Newcastle-Ottawa Scale (NOS) for Assessing the Quality of Nonrandomised Studies in Meta-Analyses. Oxford; 2000.
16. Stroup DF, Berlin JA, Morton SC, et al. Meta-analysis of observational studies in epidemiology: a proposal for reporting. JAMA. 2000;283:2008–2012.
17. GRADE handbook. Accessed May 30, 2022. https://gdt.gradepro.org/app/handbook/handbook.html
18. Wetterslev J, Thorlund K, Brok J, Gluud C. Trial sequential analysis may establish when firm evidence is reached in cumulative meta-analysis. J Clin Epidemiol. 2008;61:64–75.
19. Review Manager (RevMan) [Computer program]. Version 5.4, The Cochrane Collaboration, 2020.
20. Begg CB, Mazumdar M. Operating characteristics of a rank correlation test for publication bias. Biometrics. 1994;50:1088–101.
21. Egger M, Smith GD, Schneider M, Minder C. Bias in meta-analysis detected by a simple, graphical test. BMJ. 1997;315:629–634.
22. Itagaki A, Sakurada K, Matsuhama M, Yajima J, Yamashita T, Kohzuki M. Impact of frailty and mild cognitive impairment on delirium after cardiac surgery in older patients. J Cardiol. 2020;76:147–153.
23. Boureau AS, Chapelet G, Paille M, et al. Geriatric factors associated with 1-year mortality after aortic valve replacement. Eur Geriatr Med. 2018;9:845–851.
24. Terazawa S, Oshima H, Narita Y, et al. Strategy of cardiovascular surgery for patients with dementia as evaluated by mini-mental state examination. Circ J. 2018;82:2998–3004.
25. Harrington MB, Kraft M, Grande LJ, Rudolph JL. Preoperative cognitive status is independently associated with discharge location after cardiac surgery. Am J Crit Care. 2011;20:129–137.
26. Maekawa K, Goto T, Baba T, Yoshitake A, Katahira K, Yamamoto T. Impaired cognition preceding cardiac surgery is related to cerebral ischemic lesions. J Anesth. 2011;25:330–336.
27. Veliz-Reissmüller G, Agüero Torres H, van der Linden J, Lindblom D, Eriksdotter Jönhagen M. Pre-operative mild cognitive dysfunction predicts risk for post-operative delirium after elective cardiac surgery. Aging Clin Exp Res. 2007;19:172–177.
28. Kazmierski J, Banys A, Latek J, et al. Mild cognitive impairment with associated inflammatory and cortisol alterations as independent risk factor for postoperative delirium. Dement Geriatr Cogn Disord. 2014;38:65–78.
29. Aykut K, Albayrak G, Guzeloglu M, Baysak A, Hazan E. Preoperative mild cognitive dysfunction predicts pulmonary complications after coronary artery bypass graft surgery. J Cardiothorac Vasc Anesth. 2013;27:1267–1270.
30. Baba T, Goto T, Maekawa K, Ito A, Yoshitake A, Koshiji T. Early neuropsychological dysfunction in elderly high-risk patients after on-pump and off-pump coronary bypass surgery. J Anesth. 2007;21:452–458.
31. Millar K, Asbury AJ, Murray GD. Pre-existing cognitive impairment as a factor influencing outcome after cardiac surgery. Br J Anaesth. 2001;86:63–67.
32. Luque T, Noriega FJ, McInerney A, et al. Impact of delirium in acute cardiac care unit after transcatheter aortic valve replacement. Int J Cardiol. 2021;330:164–170.
33. Kapadia M, Shi SM, Afilalo J, et al. Cognition, frailty, and functional outcomes of transcatheter aortic valve replacement. Am J Med. 2020;133:1219–1222.
34. Saji M, Tobaru T, Higuchi R, et al. Cognitive assessment using the revised Hasegawa’s dementia scale to determine the mid-term outcomes following transcatheter aortic valve replacement. J Cardiol. 2019;74:206–211.
35. Yanagisawa R, Tanaka M, Yashima F, et al. Frequency and consequences of cognitive impairment in patients underwent transcatheter aortic valve implantation. Am J Cardiol. 2018;122:844–850.
36. Schoenenberger AW, Zuber C, Moser A, et al. Evolution of cognitive function after transcatheter aortic valve implantation. Circ Cardiovasc Interv. 2016;9:e003590.
37. Jain V, Kalra A, Panhwar MS, et al. Outcomes of transcatheter aortic valve replacement in patients with cognitive dysfunction. J Am Geriatr Soc. 2021;69:1363–1369.
38. Brown CH. Delirium in the cardiac surgical intensive care unit. Curr Opin Anaesthesiol. 2014;27:117–122.
39. Sanders RD. Hypothesis for the pathophysiology of delirium: role of baseline brain network connectivity and changes in inhibitory tone. Medical Hypotheses. 2011;77:140–143.
40. Maldonado JR. Delirium pathophysiology: an updated hypothesis of the etiology of acute brain failure. Int J Geriatric Psychiatry. 2018;33:1428–1457.
41. Shafi MM, Santarnecchi E, Fong TG, et al. Advancing the neurophysiological understanding of delirium. J Am Geriatr Soc. 2017;65:1114–1118.
42. Root JC, Pryor KO, Downey R, et al. Association of pre-operative brain pathology with post-operative delirium in a cohort of non-small cell lung cancer patients undergoing surgical resection. Psycho-Oncology. 2013;22:2087–2094.
43. Gomez C, Stam CJ, Hornero R, Fernandez A, Maestu F. Disturbed beta band functional connectivity in patients with mild cognitive impairment: an MEG study. IEEE Trans Biomed Eng. 2009;56:1683–1690.
44. Liang CK, Chu CL, Chou MY, et al. Interrelationship of postoperative delirium and cognitive impairment and their impact on the functional status in older patients undergoing orthopaedic surgery: a prospective cohort study. PLoS One. 2014;9:e110339.
45. Rudolph JL, Marcantonio ER, Culley DJ, et al. Delirium is associated with early postoperative cognitive dysfunction. Anaesthesia. 2008;63:941–947.
46. Sugimura Y, Sipahi NF, Mehdiani A, et al. Risk and consequences of postoperative delirium in cardiac surgery. Thorac Cardiovasc Surg. 2020;68:417–424.
47. Abelha FJ, Luís C, Veiga D, et al. Outcome and quality of life in patients with postoperative delirium during an ICU stay following major surgery. Crit Care. 2013;17:R257.
48. Généreux P, Head SJ, Van Mieghem NM, et al. Clinical outcomes after transcatheter aortic valve replacement using valve academic research consortium definitions: a weighted meta-analysis of 3519 patients from 16 studies. J Am Coll Cardiol. 2012;59:2317–2326.
49. Mack MJ, Leon MB, Smith CR, et al. 5-year outcomes of transcatheter aortic valve replacement or surgical aortic valve replacement for high surgical risk patients with aortic stenosis (PARTNER 1): a randomised controlled trial. Lancet. 2015;385:2477–2484.
50. Vuylsteke A, Pagel C, Gerrard C, et al. The Papworth bleeding risk score: a stratification scheme for identifying cardiac surgery patients at risk of excessive early postoperative bleeding. Eur J Cardiothorac Surg. 2011;39:924–930.
51. Dyke C, Aronson S, Dietrich W, et al. Universal definition of perioperative bleeding in adult cardiac surgery. J Thorac Cardiovasc Surg. 2014;147:1458–1463.e1.
52. Bendayan M, Messas N, Perrault LP, et al. Frailty and bleeding in older adults undergoing TAVR or SAVR: insights from the FRAILTY-AVR study. JACC Cardiovasc Interv. 2020;13:1058–1068.
53. Tribuddharat S, Sathitkarnmanee T, Ngamsangsirisup K, Nongnuang K. Efficacy of intraoperative hemodynamic optimization using FloTrac/EV1000 platform for early goal-directed therapy to improve postoperative outcomes in patients undergoing coronary artery bypass graft with cardiopulmonary bypass: a randomized controlled trial. MDER. 2021;14:201–209.
54. Rojas-García A, Turner S, Pizzo E, Hudson E, Thomas J, Raine R. Impact and experiences of delayed discharge: a mixed-studies systematic review. Health Expect. 2018;21:41–56.
55. Lagercrantz E, Lindblom D, Sartipy U. Survival and quality of life in cardiac surgery patients with prolonged intensive care. Ann Thorac Surg. 2010;89:490–495.
56. Mahesh B, Choong CK, Goldsmith K, Gerrard C, Nashef SAM, Vuylsteke A. Prolonged stay in intensive care unit is a powerful predictor of adverse outcomes after cardiac operations. Ann Thorac Surg. 2012;94:109–116.
57. Edwards MB, Taylor KM. Is 30-day mortality an adequate outcome statistic for patients considering heart valve replacement? Ann Thorac Surg. 2003;76:482–485.
58. Siregar S, Groenwold RHH, de Mol BAJM, et al. Evaluation of cardiac surgery mortality rates: 30-day mortality or longer follow-up?. Eur J Cardiothorac Surg. 2013;44:875–883.
59. Deiner S, Fleisher LA, Leung JM, et al.; ASA Committee on Geriatric Anesthesia and the ASA Perioperative Brain Health Initiative. Adherence to recommended practices for perioperative anesthesia care for older adults among US anesthesiologists: results from the ASA Committee on Geriatric Anesthesia-Perioperative Brain Health Initiative ASA member survey. Perioper Med (Lond). 2020;9:6.
60. Nashef SAM, Roques F, Sharples LD, et al. EuroSCORE II. Eur J Cardiothorac Surg. 2012;41:734–745.
61. Shahian DM, Jacobs JP, Edwards FH, et al. The Society of Thoracic Surgeons National Database. Heart. 2013;99:1494–1501.
62. Vacas S, Canales C, Deiner SG, Cole DJ. Perioperative brain health in the older adult: a patient safety imperative. Anesth Analg. 2022;135:316–328.
63. Janssen T, Alberts A, Hooft L, et al. Prevention of postoperative delirium in elderly patients planned for elective surgery: systematic review and meta-analysis. Clin Interv Aging. 2019;14:1095–1117.
64. Brown C IV, Deiner S. Perioperative cognitive protection. Br J Anaesth. 2016;117(suppl_3):iii52–iii61.
65. Partridge JSL, Harari D, Martin FC, et al. Randomized clinical trial of comprehensive geriatric assessment and optimization in vascular surgery. Br J Surg. 2017;104:679–687.
66. Saripella A, Wasef S, Nagappa M, et al. Effects of comprehensive geriatric care models on postoperative outcomes in geriatric surgical patients: a systematic review and meta-analysis. BMC Anesthesiol. 2021;21:127.
67. Lee YH, Oh HK, Kim DW, et al. Use of a comprehensive geriatric assessment to predict short-term postoperative outcome in elderly patients with colorectal cancer. Ann Coloproctol. 2016;32:161–169.
68. Aldecoa C, Bettelli G, Bilotta F, et al. European Society of Anaesthesiology evidence-based and consensus-based guideline on postoperative delirium. Eur J Anaesthesiol. 2017;34:192–214.

Supplemental Digital Content

Copyright © 2023 International Anesthesia Research Society