If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Delirium is common after cardiac surgery and is associated with adverse consequences, including cognitive decline. Identification of vulnerable older adults might allow for early implementation of delirium-prevention strategies. Brain MRI findings provide insight into structural brain changes that may identify vulnerable patients. The purpose of this study was to examine the association between brain MRI characteristics potentially associated with delirium vulnerability and the development of postoperative delirium in a nested cohort of patients undergoing cardiac surgery.
Methods
We identified 79 cardiac surgery patients who had brain MRI imaging after cardiac surgery, as part of an ongoing randomized trial evaluating the efficacy of blood pressure management based on cerebral autoregulation monitoring versus standard management for improving neurological outcomes. Cerebral lateral ventricular size, cortical sulcal width, and white matter hyperintensities (WMH) on brain MRI scans were graded on a validated 0 to 9 scale, and categorized into tertiles. New ischemic lesions were characterized as present or absent. Delirium was assessed using a validated chart-review. Neuropsychological testing performed before surgery was used to establish preoperative cognitive baseline. Multivariable logistic regression was used to assess the independent association between MRI characteristics and postoperative delirium.
Findings
The average age of patients was 70.1 ± 7.8 years old, and 72% were male. Twenty-eight of 79 (35.4%) patients developed postoperative delirium. Patients with delirium had higher unadjusted ventricular size (median 4 vs. 3, P = 0.003), and there was a trend towards higher sulcal sizes and WMH grades. Increasing tertiles of ventricular size (Odds Ratio [OR] 3.59; 95% Confidence Interval [CI] 1.59-8.12; P = 0.002) and sulcal size (OR 2.15; 95%CI 1.13-4.12; P = 0.02) were associated with postoperative delirium, with a trend for tertiles of WMH grade (OR 1.91; 95%CI 0.99-3.68; P = 0.05). In multivariable models adjusted for logistic EuroSCORE, baseline cognitive status, bypass time, and any postoperative complication, each tertile of ventricular size was associated with increased odds of postoperative delirium (OR 3.23 per tertile increase in ventricular size; 95%CI 1.21-8.60; P = 0.02). There were no differences in odds of delirium by tertiles of sulcal grade, tertiles of white matter grade, or presence of new ischemic lesions, in adjusted models.
Implications
Increased brain ventricular size was independently associated with delirium after cardiac surgery. These results suggest that cerebral atrophy may contribute to increased vulnerability for postoperative delirium. Baseline brain MRIs may be useful in identifying cardiac surgery patients at high risk for postoperative delirium, who might benefit from targeted perioperative approaches to prevent delirium. ClinicalTrials.gov identifier: NCT00981474.
Thus, increasing efforts have focused on characterizing the pathophysiology of postoperative delirium and, in particular, identifying vulnerable older adults who might benefit from targeted delirium prevention strategies.
In patients undergoing cardiac surgery, the characteristics most strongly associated with postoperative delirium reflect cerebral pathology, including prior stroke,
Because these brain measures are somewhat nonspecific, better characterization of brain structure and function, through modalities such as brain magnetic resonance imaging (MRI), may yield insights into the pathophysiology of delirium and also identify vulnerable patients at highest risk for postoperative delirium so that anesthetic approaches could be modified.
Despite the potential of MRI findings as biomarkers, the reports of an association between brain MRI findings and risk for postoperative delirium have not been consistent.
Association of pre-operative brain pathology with post-operative delirium in a cohort of non-small cell lung cancer patients undergoing surgical resection.
have been associated with postoperative delirium in some studies. Similarly, cerebral atrophy has been associated with postoperative delirium in heterogeneous patient populations.
However, weaknesses of previous studies include the use of poorly validated delirium assessments, imprecise characterization of MRI findings, lack of accounting for potentially confounding variables, and heterogeneous patient populations. Recently, a prospective, well-conducted study in noncardiac surgery patients addressed many limitations of the earlier studies, and it found no difference in cerebral atrophy or WMH according to delirium status after surgery.
Brain atrophy and white-matter hyperintensities are not significantly associated with incidence and severity of postoperative delirium in older persons without dementia.
However, it is unclear if these findings would be generalizable to patients undergoing cardiac surgery, because the pathophysiology of delirium in cardiac surgery patients may be different because of underlying patient characteristics (eg, high prevalence of cerebrovascular disease) and different surgical insults (eg, cardiopulmonary bypass and associated inflammatory burden).
In the present study, we hypothesized that specific brain MRI characteristics (ventricular and sulcal size, WMH, and new ischemic lesions) would be associated with postoperative delirium in patients undergoing cardiac surgery.
Patients and Methods
The study procedures met with the approval of the institutional review board and were performed after receiving individual written informed consent.
Patients
This study was a prospective observational trial, nested in an ongoing multiyear randomized controlled study evaluating the association between cerebral blood flow autoregulation
Duration and magnitude of blood pressure below cerebral autoregulation threshold during cardiopulmonary bypass is associated with major morbidity and operative mortality.
and brain injury after cardiac surgery. The objectives of this study were not the primary outcome of the original trial, and thus the present study is a secondary data analysis.
Patients underwent surgery between October 2009 and August 2012. The primary inclusion criterion was primary or re-operative coronary artery bypass graft (CABG) and/or valve or aortic surgery that required cardiopulmonary bypass bypass, as well as having an elevated risk of stroke or encephalopathy, based on a Johns Hopkins risk score model composed of history of stroke, presence of carotid bruit, hypertension, diabetes, and age, which generally excluded patients in the lowest quartile of risk.
Exclusion criteria were renal failure requiring dialysis, non–English speaking, contraindications to MRI (eg, pacemaker), and emergency surgery. As part of the main randomized controlled trial, patients were randomized 1:1 to blood pressure targets during cardiopulmonary bypass based on measures of cerebral autoregulation versus standard-of-care targets. Data on postoperative outcomes (major morbidity/mortality, acute kidney injury, and delirium) in a subset of these patients have been reported separately.
During the time period of the nested cohort study, 1337 patients were screened, of whom 777 (58.1%) did not meet enrollment criteria, 246 (18.4%) were not approached for logistical reasons (eg, staff availability on weekends or emergent procedure), 134 (10.0%) declined participation, and 180 (13.5%) were enrolled. Of the enrolled patients, 101 did not complete the MRI (25 refused, 20 were unable to schedule, 17 had retained epicardial pacemaker leads or other metal identified with screening head radiographs, 10 were unable to tolerate the MRI procedure, 10 had scans that were unavailable or uninterpretable for analysis, 9 were discharged early, 7 died, and 3 withdrew). A total of 79 patients remained in the analysis. A CONSORT diagram is included as Appendix I.
Perioperative Care
Patients received standard institutional monitoring, including radial arterial blood pressure. General anesthesia was induced and maintained with midazolam (0.15 mg/kg), fentanyl (5–20 μg/kg), and isoflurane, with pancuronium or vecuronium for muscle relaxation. Cardiopulmonary bypass was conducted by using a nonocclusive roller pump and a membrane oxygenator; the circuit included a ≤40-μm arterial line filter. Nonpulsatile flow was maintained between 2.0 and 2.4 L/min/m. Patients were managed by using α-stat pH management, and rewarming was based on institutional standards with a goal of maintaining a nasopharyngeal temperature <37oC. Postoperative sedation was performed with propofol 20 to 75 μg/kg/min until appropriate for tracheal extubation or for 24 hours after surgery. Patients requiring >24 hours of mechanical ventilation received fentanyl and midazolam.
Brain MRI Scans
Patients underwent brain MRI scans generally within 5 to 8 days after surgery (range, 2–23 days). All MRI imaging was performed on a 3-T Siemens MRI research scanner (Munich, Germany), and the following MRI sequences were obtained and used in the analysis: T1-weighted, T2/fluid attenuation inversion recovery, and diffusion-weighted imaging (DWI). Cerebral ventricular size was assessed on a 0 to 9 scale by comparison to 8 reference images that had successively increasing lateral ventricular size, ranging from normal (grade 1) to severe (grade 8) enlargement. These methods were previously developed and validated for the Cardiovascular Health Study.
Patients with ventricles smaller than grade 1 received grade 0, whereas patients with ventricles larger than grade 8 received grade 9. Similarly, cerebral sulcal widening and WMH were assessed on a scale of 0 to 9 of increasing severity compared with reference images as previously described.
Acute ischemic lesions on MRI were determined by the presence of any increased signal on the DWI sequence with corresponding low signal on the apparent diffusion coefficient maps. All images were assessed by a board-certified neurologist (R.F.), who was blinded to the patients’ delirium assessment. A second investigator (M.B.), also blinded to the patients’ delirium assessment, reviewed randomly selected images from 20% of the sample. Both reviewers were masked to all patient characteristics, including delirium status, neuropsychological test results, and randomization status. A weighted κ statistic was calculated to assess inter-rater agreement (κ = 0.69 for WMHs; κ = 0.62 for sulcal widening; and κ = 0.63 for ventricular size).
Delirium and Neuropsychological Assessment
Delirium was assessed by using a validated chart review method.
Because clinical judgment was required, a research assistant trained in formal delirium assessment by 1 of the authors (K.J.N., a psychiatrist) performed all the chart abstractions by using a chart-based instrument. The abstractor was masked to MRI findings and postoperative neuropsychological test results. According to the methods of Inouye et al
(as specifically outlined in Appendix I of the article), the abstractor searched the medical record for any mention of key terms, with evidence of acute onset or change, to support a diagnosis of delirium. Key terms and pertinent evidence were based on the following question: “Is there any evidence from the chart of acute confusional state (eg, delirium, mental status change, inattention, disorientation, hallucinations, agitation, inappropriate behavior, etc.)?” All sections of the medical record were searched, including but not limited to progress notes, nursing notes, social work notes, physical/occupational therapy notes, and consultant notes. For each possible episode of delirium, the abstractor recorded the source of information, time of onset, and a verbatim description of the episode. A 3-person panel (K.J.N., C.B., and L.M.) with training in formal delirium assessment and active involvement in delirium research reviewed the chart abstractions and determined the final diagnosis of delirium. Both the research assistant who abstracted the data and the participants in the delirium consensus panel were masked to randomization status, MRI results, and neurocognitive test results.
Neuropsychological testing was performed 1 to 3 days before surgery. The tests assessed a number of cognitive domains known to be associated with vascular disease and thus potentially affected in patients undergoing cardiac surgery.
Covariates and other variables were recorded from the anesthesia preoperative or intraoperative record and from other clinical documentation by research staff.
Data Analysis
Statistical analyses were conducted by using Stata Statistical Software: Release 12 (StataCorp LP, College Station, Texas). Brain MRI ventricular size grade, cerebral sulcal grade, and WMH grade were nonnormally distributed and were thus categorized into tertiles. New ischemic lesion(s) were considered as a binary (yes/no) variable. The primary outcome was any episode of delirium, considered as a binary variable.
Several important covariates were examined for inclusion in the final regression models. To estimate baseline cognition, a composite cognitive z score was calculated by using prior methods from the individual cognitive test scores.
Cognitive outcomes three years after coronary artery bypass surgery: a comparison of on-pump coronary artery bypass graft surgery and nonsurgical controls.
Briefly, individual neuoropsychological test scores were converted to z scores, using the mean and SD of the baseline scores from the control group from the randomized parent study as the reference. After each individual cognitive test score was converted to a z score, the scores were combined into an average z score, calculated from the average of the nonmissing individual test z scores. Finally, the average z scores were renormalized by using the control group’s mean and SD to develop a composite cognitive z score outcome. Postoperative complications were prospectively assessed, and a composite comorbidity outcome was defined as new clinically diagnosed stroke, new dialysis requirement, sepsis, mechanical ventilation >48 hours, inotropic medications >24 hours, new intra-aortic balloon pump insertion, cardiac arrest, or reoperation because of bleeding.
Analysis
We estimated our sample would have 86% power with an α level of 0.05 to detect an odds ratio of 2.3, based on increasing tertile of ventricular size and adjusted for other variables in the model. Baseline patient and surgical characteristics were compared by using t tests for normally distributed variables, rank sum tests for nonnormally distributed variables, and χ2 tests and Fisher’s exact tests for categorical variables. Logistic regression models were used to assess the association between brain MRI characteristics and postoperative delirium. For all analyses, P values <0.05 were considered significant. Covariates to include in the final adjusted regression model were chosen a priori, as well as based on subsequent comments by reviewers, and included a logistic European System for Cardiac Operative Risk Evaluation score (which includes age, type of surgery, and history of prior cardiac surgery), composite cognitive z score, total cardiopulmonary bypass time, and a composite of any complication. Several sensitivity analyses were also conducted, in which the following variables were added to the final model: (1) age and previous cardiac surgery; (2) surgical procedure; and (3) individual postoperative complications (mechanical ventilation >48 hours, inotrope >24 hours, postoperative stroke, new intra-aortic balloon pump insertion, cardiac arrest, and reoperation for bleeding). In all models, we also assessed whether randomization affected the outcomes, and we found no differences in inferences; this factor was therefore not included in the final models presented.
Results
There were no significant differences in patient or surgical characteristics between the 79 patients included in this study (who underwent MRI scanning) and patients in the parent study who did not undergo MRI scanning (Appendix II). In particular, the mean (SD) age of participants without MRI scans (71.4 [8.6] years) was similar to the age of participants with MRI scans (70.1 [7.8] years; P = 0.29), as was the incidence of postoperative delirium (26.7% vs 35.4% [P = 0.23], respectively).
The characteristics of patients in this cohort are shown in Table I. The mean age of patients was 70.1 (7.8) years, and there was a high prevalence of hypertension, coronary artery disease, history of smoking, and diabetes. When stratified according to median ventricular size on brain MRI, patients with ventricular grade >3 (median value) were older, more likely to have undergone prior cardiac surgery, had longer cardiopulmonary bypass times, more mechanical ventilation >48 hours, and more intra-aortic balloon pumps inserted. Differences in patients according to sulcal size, white matter volume, and the presence or absence of new ischemic lesions are shown in Appendices III,IV, and V, respectively. Significantly, patients with sulcal size and white matter grade greater than median values were older than patients with corresponding MRI characteristics less than median values.
Table IPatient and surgical characteristics according to median ventricular grade (larger grade indicating larger ventricles). Unless otherwise indicated, values are no. (%) or median (interquartile range).
Composite complications include stroke, inotropic drug >24 hours, new intra-aortic balloon pump, mechanical ventilation >48 hours, sepsis, new dialysis requirement, cardiac arrest, or reoperation due to bleeding.
COPD = chronic obstructive pulmonary disease; EuroSCORE = European System for Cardiac Operative Risk Evaluation; CABG = coronary artery bypass graft surgery; NA = not applicable.
The χ2 test was used for significance.
† Fisher’s exact test was used for significance.
‡ Composite complications include stroke, inotropic drug >24 hours, new intra-aortic balloon pump, mechanical ventilation >48 hours, sepsis, new dialysis requirement, cardiac arrest, or reoperation due to bleeding.
Twenty-eight (35.4%) patients developed postoperative delirium. Characteristics of patients with and without delirium are shown in Table II. Patients with delirium were older and had more baseline asthma or chronic obstructive pulmonary disease and more prior cardiac surgery compared with patients without delirium. In the postoperative period, patients with delirium had a higher incidence of composite postoperative complications compared with patients without delirium. Specifically, patients with delirium had a higher incidence of mechanical ventilation >48 hours, inotropic drug >24 hours, and new intra-aortic balloon pump insertion. There was no difference in incidence of stroke in patients with delirium versus without delirium, and no patients developed sepsis or a new need for dialysis.
Table IIPatient and surgical characteristics according to patients’ delirium status. Unless otherwise indicated, values are no. (%) or median (interquartile range).
Composite complications include stroke, inotropic drug >24 hours, new intra-aortic balloon pump, mechanical ventilation >48 hours, sepsis, new dialysis requirement, cardiac arrest, or reoperation due to bleeding.
4 (7.8)
13 (46.4)
<0.001
COPD = chronic obstructive pulmonary disease; EuroSCORE = European System for Cardiac Operative Risk Evaluation; CABG = coronary artery bypass graft surgery; NA = not applicable.
The χ2 test was used for significance.
† Fisher’s exact test was used for significance.
‡ Composite complications include stroke, inotropic drug >24 hours, new intra-aortic balloon pump, mechanical ventilation >48 hours, sepsis, new dialysis requirement, cardiac arrest, or reoperation due to bleeding.
Brain MRI scans were obtained at a median of 6 days (interquartile range [IQR], 5–8) after surgery. Median ventricular grade was 3 (IQR, 3–4), median sulcal grade was 3 (IQR, 2–5), and median white matter grade was 2 (IQR, 1–3). New ischemic lesions were present in 44 (55.7%) patients. As shown in Table III, patients with delirium had significantly higher (unadjusted) ventricular size, and a trend toward higher sulcal grade and WMH grade (both P = 0.05), compared with patients without delirium.
Table IIICharacteristics of brain magnetic resonance imaging according to patients’ delirium status. Unless otherwise indicated, values are given as median (interquartile range).
As shown in Table IV, Table V, in unadjusted models, each increasing tertile of ventricular size, sulcal size, and WMH grade (trend only), but not the presence of new ischemic lesions, was associated with an increased risk of postoperative delirium. The dose–response relationships between postoperative delirium and brain MRI characteristics are shown in the Figure. In models adjusted for logistic European System for Cardiac Operative Risk Evaluation score, baseline cognitive status, cardiopulmonary bypass time, and presence of any postoperative complication, each increasing tertile of ventricular size was associated with increased odds of postoperative delirium (odds ratio, 3.23 per tertile increase in ventricular size [95% CI, 1.21–8.60]; P = 0.02). There were no differences in the odds of postoperative delirium according to tertiles of sulcal size, tertiles of WMH grade, or presence of new ischemic lesions in the adjusted models. Sensitivity analyses were also conducted in which we included: (1) age and history of previous cardiac surgery; (2) the type of surgical procedure; (3) individual complications (in lieu of a composite variable); and (4) number of units of blood products transfused. We found that the inferences for all adjusted models were unchanged.
Table IVUnadjusted and adjusted odds of delirium according to tertile of ventricular size on brain magnetic resonance imaging.
Composite complications include stroke, inotropic drug >24 hours, new intra-aortic balloon pump, mechanical ventilation >48 hours, sepsis, new dialysis requirement, cardiac arrest, or reoperation because of bleeding.
10.2
2.88–35.99
<0.001
12.2
2.56–58.4
0.002
EuroSCORE = European System for Cardiac Operative Risk Evaluation.
Adjusted for all characteristics presented in the table.
† Composite complications include stroke, inotropic drug >24 hours, new intra-aortic balloon pump, mechanical ventilation >48 hours, sepsis, new dialysis requirement, cardiac arrest, or reoperation because of bleeding.
Adjusted for logistic European System for Cardiac Operative Risk Evaluation score, cognitive z score, cardiopulmonary bypass time, and any complication.
Odds Ratio
95% CI
P
Odds Ratio
95% CI
P
Sulcal size (tertile)
2.15
1.13–4.12
0.02
1.62
0.74–3.53
0.23
White matter hyperintensity grade (tertile)
1.91
0.99–3.68
0.05
1.28
0.54–3.06
0.58
New ischemic lesions
1.73
0.67–4.67
0.26
2.25
0.69–7.29
0.18
Adjusted for logistic European System for Cardiac Operative Risk Evaluation score, cognitive z score, cardiopulmonary bypass time, and any complication.
In the present study, tertile of ventricular size was independently associated with postoperative delirium. Although tertiles of both sulcal size and severity of WMH were associated (or trended to association) with postoperative delirium in unadjusted models, neither of these parameters, nor the presence of new ischemic lesions, was independently associated with postoperative delirium.
Because brain MRIs are objective and widely available, they have emerged as a powerful tool to investigate structural brain abnormalities with high spatial resolution and may be particularly useful in prediction of surgical outcomes (including delirium).
Our findings of a relationship between brain ventricular volume size and delirium is consistent with previous reports from studies of patients admitted to medical and noncardiac surgical intensive care units.
Enlarged brain ventricular size reflects cerebral atrophy, and thus our data support the hypothesis that neuronal loss may predispose patients to delirium after a precipitating insult such as cardiac surgery.
Association of pre-operative brain pathology with post-operative delirium in a cohort of non-small cell lung cancer patients undergoing surgical resection.
WMH volumes in the present study were not independently associated with postoperative delirium. There may be several reasons for these differing results. First, previous studies have generally not accounted for postoperative complications and baseline cognitive status. Postoperative complications may be an important confounding variable because they are associated both with baseline patient disease severity and with postoperative delirium (P < 0.001 in adjusted models in our data). Similarly, few studies have adjusted for baseline cognitive status, although baseline cognitive status is highly associated with postoperative delirium and also with cerebral atrophy and WMH. In this study, because we were particularly interested if brain MRI characteristics would be associated with postoperative delirium even after accounting for cognitive status, we did account for cognition in our models. In our study, it is noteworthy that severity of WMH trended toward association with postoperative delirium in univariate analysis but was not independently associated with postoperative delirium, suggesting either the importance of confounding variables or that our study may have been underpowered to detect this particular adjusted association. In addition, measurement of WMH was not identical to prior studies, and misclassification error may therefore have contributed to the discrepancy in results.
Recently, a well-conducted study in noncardiac surgery patients found no association between WMH or global brain volumes and postoperative delirium.
Brain atrophy and white-matter hyperintensities are not significantly associated with incidence and severity of postoperative delirium in older persons without dementia.
However, the mechanism of delirium may be different for patients undergoing cardiac surgery for several reasons. First, patients undergoing cardiac surgery have high baseline morbidity, including a high prevalence of cerebrovascular and cardiovascular disease and high illness severity scores. In addition, cardiac surgery involves exposure to cardiopulmonary bypass, which results in heightened inflammation, changes in cerebral blood flow, and cerebral emboli. Thus, our results show that decreases in ventricular size may be most important in patients with cardiac and cerebrovascular risk factors undergoing cardiac surgery.
Our findings highlight that traditional assessments of older adults in the preoperative period can miss important pathology that would inform postoperative management. Indeed, the American Geriatrics Society and the American College of Surgeons recently partnered to develop guidelines for the preoperative assessment of older adults,
and the recommendations move beyond traditional organ-based approaches toward evaluation of cognitive, functional, nutritional, and frailty status. For patients at high risk for delirium based on preoperative assessment, tailored perioperative strategies are crucial, and general principles include avoidance of precipitating agents (eg, benzodiazepines), emphasis on mobilization, screening for the development of delirium, and consideration of avoiding excessive depth of anesthesia.
American Geriatric Society Expert Panel on Postoperative Delirium in Older Adults Postoperative delirium in older adults: best practice statement from the American Geriatrics Society.
Although brain imaging is not practical in all surgical patients, it is noteworthy that in the present study, ventricular size was linearly associated with baseline cognitive performance. Thus, although brain MRI scans may not be available on all patients, formal cognitive assessment may shed insight into important brain pathology and inform postoperative management. However, formal cognitive assessment in the preoperative period is not widely practiced.
Nevertheless, with a growing appreciation of the importance of preventing delirium and cognitive decline, formal neuroimaging may play a future role in risk stratification of patients or targeting vulnerable patients for perioperative optimization.
Our findings also highlight that postoperative complications are strongly associated with the development of postoperative delirium, with the implication that postoperative events may play an important role in the pathogenesis of delirium. As an example, transition state models examining corticosteroid administration in the intensive care unit have reported an increased risk of delirium in the 24 hours after steroid administration.
and do not incorporate important postoperative information. Our results raise the possibility that prevention of postoperative complications may also be an important delirium prevention strategy, although further research needs to be done.
Strengths of this study include the use of validated delirium assessments and objective scales to characterize brain MRI findings, rigorous assessment of baseline cognitive status and covariates, and a focus on cardiac surgery patients. There are several limitations to this study. Very importantly, brain MRI scans were obtained in the postoperative period and thus do not formally precede the incidence of delirium. It is therefore possible that delirium contributed to postoperative MRI changes. However, ventricular size and sulcal width are considered relatively static and robust features that change slowly over relatively long periods of time.
Similarly, our WMH grading accounted for and excluded new ischemic lesions seen on DWI, ensuring that graded WMH reflected presurgical status. Second, we used a rigorous validated method of delirium assessment; however, we did not conduct in-person assessments and, compared with the well-known Confusion Assessment Method,
Thus, misclassification error may be present. Our delirium assessment was likely more sensitive for hyperactive delirium (which is clinically recognizable) compared with hypoactive delirium (which may not be as clinically apparent).
; thus, identifying patients with hypoactive delirium is important for future studies. Because our study size was small, we were limited in the number of covariates that could be considered; thus, our results could be explained by residual confounding, although we used multiple models as a sensitivity analysis. Finally, only 44% of enrolled patients underwent an MRI. Although patient characteristics were generally similar between these patient groups, this limits the generalizability of the results and may introduce selection bias.
Conclusions
Results from this study show that increased ventricular size may be independently associated with delirium after cardiac surgery. These results imply that patients with increased ventricular size on preoperative imaging may benefit from targeted delirium prevention strategies. Further study is needed to confirm these findings in a cohort with in-person delirium assessments that are sensitive to hypoactive delirium.
Conflicts of Interest
The authors have indicated that they have no conflicts of interest regarding the content of this article. For unrelated studies, Dr. Neufeld has received research support from Ornim Medical, and Dr. Hogue has received research support from Covidien, Inc, and served on the advisory board for Ornim Medical.
Author Contributions
Charles Brown contributed to the study design, data acquisition, data interpretation and analysis, and manuscript preparation. Roland Faigle contributed to the study design, data acquisition, data interpretation and analysis, and manuscript preparation. Lauren Klinker contributed to manuscript preparation and revision. Mona Bahouth contributed to data acquisition and manuscript revision. Laura Max contributed to data acquisition and manuscript revision. Andrew LaFlam contributed to data acquisition and manuscript revision. Karin Neufeld contributed to data acquisition, data interpretation, and manuscript revision. Kaushik Mandal contributed to manuscript preparation, data interpretation, and manuscript revision. Rebecca Gottesman contributed to study design, data interpretation, and manuscript revision. Charles Hogue contributed to study design, data collection, data interpretation, and manuscript revision.
Acknowledgments
This work was supported by a National Institutes of Health (NIH) grant (RO3 AG042331), the Jahnigen Career Development Award, the Research Career Development Core of the Johns Hopkins University Claude D. Pepper Older Americans Independence Center, NIA P30AG021334; the International Anesthesia Research Society, and the Johns Hopkins Clinician Scientist Award (Dr. Brown); NIH grant (RO1 HL092259) (Dr. Hogue); and an institutional NIH KL2 grant from the Johns Hopkins Institute for Clinical and Translational Research (KL2TR001077) (Dr. Faigle).
The authors thank Gayane Yenokyan, PhD, for statistical support.
Appendix IIPatient and surgical characteristics according to magnetic resonance imaging (MRI) completion. Unless otherwise indicated, data are given as no. (%) or median (interquartile range).
Composite complications include stroke, inotropic drug >24 hours, new intra-aortic balloon pump, mechanical ventilation >48 hours, sepsis, new dialysis requirement, cardiac arrest, or reoperation because of bleeding.
COPD = chronic obstructive pulmonary disease; EuroSCORE = European System for Cardiac Operative Risk Evaluation; CABG = coronary artery bypass graft surgery.
The χ2 test was used for significance.
† Fisher’s exact test was used for significance.
‡ Composite complications include stroke, inotropic drug >24 hours, new intra-aortic balloon pump, mechanical ventilation >48 hours, sepsis, new dialysis requirement, cardiac arrest, or reoperation because of bleeding.
Appendix IIIPatient and surgical characteristics according to median sulcal grade (larger grade indicating larger sulcal size). Unless otherwise indicated, data are given as no. (%) or median (interquartile range).
Composite complications include stroke, inotropic drug >24 hours, new intra-aortic balloon pump, mechanical ventilation >48 hours, sepsis, new dialysis requirement, cardiac arrest, or reoperation because of bleeding.
COPD = chronic obstructive pulmonary disease; EuroSCORE = European System for Cardiac Operative Risk Evaluation; CABG = coronary artery bypass graft surgery; NA = not applicable.
The χ2 test was used for significance.
† Fisher’s exact test was used for significance.
‡ Composite complications include stroke, inotropic drug >24 hours, new intra-aortic balloon pump, mechanical ventilation >48 hours, sepsis, new dialysis requirement, cardiac arrest, or reoperation because of bleeding.
Appendix IVPatient and surgical characteristics according to median grade of white matter hyperintensities. Unless otherwise indicated, values are given as no. (%) or median (interquartile range).
Composite complications include stroke, inotropic drug>24 hours, new intra-aortic balloon pump, mechanical ventilation>48 hours, sepsis, new dialysis requirement, cardiac arrest, or reoperation because of bleeding.
COPD = chronic obstructive pulmonary disease; EuroSCORE = European System for Cardiac Operative Risk Evaluation; CABG = coronary artery bypass graft surgery; NA = not applicable.
The χ2 test was used for significance.
† Fisher’s exact test was used for significance.
‡ Composite complications include stroke, inotropic drug>24 hours, new intra-aortic balloon pump, mechanical ventilation>48 hours, sepsis, new dialysis requirement, cardiac arrest, or reoperation because of bleeding.
Appendix VPatient and surgical characteristics according to the presence of new ischemic lesions. Unless otherwise indicated, values are given as no. (%) or median (interquartile range).
Composite complications include stroke, inotropic drug>24 hours, new intra-aortic balloon pump, mechanical ventilation>48 hours, sepsis, new dialysis requirement, cardiac arrest, or reoperation because of bleeding.
COPD = chronic obstructive pulmonary disease; EuroSCORE = European System for Cardiac Operative Risk Evaluation; CABG = coronary artery bypass graft surgery; NA = not applicable.
The χ2 test was used for significance.
† Fisher’s exact test was used for significance.
‡ Composite complications include stroke, inotropic drug>24 hours, new intra-aortic balloon pump, mechanical ventilation>48 hours, sepsis, new dialysis requirement, cardiac arrest, or reoperation because of bleeding.
Association of pre-operative brain pathology with post-operative delirium in a cohort of non-small cell lung cancer patients undergoing surgical resection.
Brain atrophy and white-matter hyperintensities are not significantly associated with incidence and severity of postoperative delirium in older persons without dementia.
Duration and magnitude of blood pressure below cerebral autoregulation threshold during cardiopulmonary bypass is associated with major morbidity and operative mortality.
Cognitive outcomes three years after coronary artery bypass surgery: a comparison of on-pump coronary artery bypass graft surgery and nonsurgical controls.
The χ2 test was used for significance.
01211-4&id=gr1.jpg){kind=link}
01211-4&id=fx1.jpg){kind=link}