|
|
||||||||
Ann Thorac Surg 2006;81:1738-1744
© 2006 The Society of Thoracic Surgeons
a College of Medicine, Drexel University, Philadelphia, Pennsylvania
b Department of Clinical and Health Psychology, Drexel University, Philadelphia, Pennsylvania
c Behavioral Medicine, Christiana Care Health System, Wilmington, Delaware
d Voorhees, New Jersey
e Neuropsychology Department, Kessler Rehabilitation Research and Educational Corporation, West Orange, New Jersey
f Psychology Department, Bryant College, Smithfield, Rhode Island
g Cardio-Thoracic Surgery, Lankenau Hospital, Wynnewood, Pennsylvania
h College of Medicine, University of California-Irvine, Irvine, California
Accepted for publication December 2, 2005.
* Address correspondence to Dr Petrucci, Drexel University College of Medicine, Psychiatry and Medicine, 245 N 15th Street, MS 115, Philadelphia, PA 19102 (Email: ralph.petrucci{at}drexelmed.edu).
| Abstract |
|---|
|
|
|---|
METHODS: This prospective cross-sectional cognitive preliminary study evaluated 252 potential heart transplant candidates assessing functions in memory, motor, and processing speed. Patients were divided into three HF groups based on severity of disease: group 1 outpatients (n = 113), group 2 in-patients requiring inotropic infusion (n = 83), and group 3 inpatients likely requiring MCAD support (n = 56). Aggregate z-scores for memory, motor, and processing speed and independent samples t tests assessed intergroup differences on 13 cognitive measures.
RESULTS: A broad pattern of cognitive impairment was observed within the advanced HF group; fewer deficits were found in group 1 outpatients and more severe deficits in group 3 MCAD subjects. A difference in motor functions was observed as the earliest abnormality, with group 3 showing significant changes compared with group 1. The most dramatic changes were seen in domain mental processing speed along with specific verbal and visual memory functions, which were slower in group 3 compared with groups 1 and 2.
CONCLUSIONS: Cognitive deficits are common in advanced HF and worsen with increasing severity of HF. Appropriately designed and randomized studies will be needed to demonstrate if earlier MCAD implantation is warranted to arrest cognitive dysfunction and better postimplantation adaptation.
| Introduction |
|---|
|
|
|---|
It is assumed that low cardiac output in advanced HF leads to renal and hepatic dysfunction. The impact of long-term cardiac insufficiency on the function of end organs highlights the need to intervene before the end-organ damage sets in. While it is now increasingly recognized that low cardiac output also adversely affects neurobehavioral functions through decreased blood flow, this may also need to be evaluated in advanced HF. The quality of neurobehavioral performance significantly influences functional independence after MCAD placement [4, 5]; and therefore, it is critical to assess the cognitive ability of the patients to manage a bridge or destination device [6]. The cognitive performance of device aspirants or recipients has not been adequately studied and compared with other modes of management strategies in advanced HF [7]
Neuropsychological assessment has been commonly performed and has revealed frequent cognitive impairment in coronary bypass and transplant recipients [8, 9]. Preprocedure and postprocedure measures have shown a range of cognitive deficits [1013], and the postoperative behavioral adjustment in such patients is significantly influenced by neurobehavioral function before surgery [1418]. Earlier studies of pretransplant cognitive assessment have demonstrated moderate to severe impairment in up to 60% of patients with heart failure [1922]. Ischemic stress, as a consequence of heart failure, has been shown to contribute to cognitive deficits including attention deficit and total and delayed memory recall impairment [23, 24].
The present prospective investigation involved a baseline comparison of cognitive functions obtained during initial transplant evaluation in a large patient population with variable degrees of advanced HF. The cohort included transplant evaluations in outpatients or during hospitalization in patients requiring intravenous inotropic therapy or prospective MCAD placement.
| Patients and Methods |
|---|
|
|
|---|
Inclusion for the neuropsychological examination required patients to sit up in bed, maintain an arterial oxygen saturation of greater than 90%, and provide verbal consent. These patients were not considered for randomized medication versus device trials. Criteria were met by 113 group 1, 83 group 2, and 56 group 3 patients (Table 1), and they form the basis of the present report. Although all potential transplant candidates undergo neurobehavioral evaluation, only 37% patients could complete the tests for enrollment; this study, therefore, does not fulfill the criteria for consecutive recruitment. Patients of groups 2 and 3 were examined in the hospital while on telemetry monitoring. Twenty-nine of the 56 group 3 patients received MCAD support (Thoratec-11, Thoratec Corp, Pleasanton, CA; Abiomed-8, Abiomed Inc, Danvers, MA; Novacor-9, World Heart Inc, Oakland, CA; and AbioCor-1, Abiomed Inc) before transplant. Seventeen of the 29 device patients went on to receive transplantation, and 1 patient received a totally implantable artificial heart; 11 patients died before transplant. Patients remained in their initial respective groups and were not reexamined or moved into another group with the progression of their cardiac disease. There were no controls established for medications, hemodynamics, or concomitant medical disease. Over the 18 years of recruitment for the study, major changes have occurred in management of heart failure that may have inadvertently impacted the outcomes of this study. A retrospective data review with individual consent waived was approved by the Institutional Review Board (University Project 1000547) on May 18, 2000. Individual verbal consent was obtained at time of testing.
|
Neuropsychological Examination
The neuropsychological measures were selected for ease of administration with these advanced HF patients. A 45- to 60-minute bedside testing session was designed to survey 13 cognitive functions and was administered in standard fashion to all groups. Testing was conducted once during the initial presentation within each group. The measures were divided into three broad domains: memory, motor functions, and processing speed.
The memory evaluation included the administration of the Wechsler Memory Scale (WMS) [25], the Benton Visual Retention Test (Benton VRT; Administration AForm C) [26], and the Rey Auditory Verbal Learning Test (Rey AVLT) [27]. The WMS is composed of subtests measuring verbal and visual memory ability with immediate and 30-minute delayed recall procedures. Verbal memory was assessed by asking the patient to recall two short stories verbatim. Visual memory tests required the patient to recall and reproduce three designs. The WMS yields separate normally distributed scores for verbal and visual memory. The Benton VRT has single or multiple designs for recall that are scored for the number correct and error quality. Fifteen words are presented verbally in the Rey AVLT over five trials, followed by an additional trial in which a nonsense list is introduced for distraction. The total number of original words recalled from 6 trials is scored. Motor function was evaluated with a finger-tapper and hand-grip dynamometer [28]. The mean number of finger taps in a predefined time period and the mean grip strength in kilograms were separately recorded for dominant and nondominant hands. Mental processing speed was measured through administration of the Trail Making Tests A and B [28]. Trail Making A requires the patient to draw a line between sequential numbers, and Trail Making B requires the patient to draw a line alternating sequentially between circled numbers and letters.
Statistical Analysis
Aggregate z-scores were created for three major cognitive domains: memory (with verbal and visual subsets), motor function, and processing speed. The z-scores were used to standardize data for analyses. Independent sample t tests were performed to assess intergroup differences between groups 1, 2, and 3 HF patients. Controlling for individual variables was not included in this design. The confidence level was established at p less than 0.05 for all functions with equality of variances considered for each analysis. Under these three major categories, 13 individual cognitive measures were evaluated. Independent samples t tests were performed for the 13 individual cognitive measures between groups 1, 2, and 3 (Table 2). Bonferroni corrections were applied for the three domains (p < 0.017) and the 13 cognitive measures (p < 0.004). The severity of the cardiac illness prevented some patients from completing all 13 cognitive measures; consequently sample size may differ across analyses.
|
| Results |
|---|
|
|
|---|
|
|
Mental Processing Speed
Mental processing speed was seen to decline across the three groups with increasing severity (Fig 2D). Processing speed was slower for group 3 than groups 1 and 2. Although group 2 scored below group 1 on Trail Making B, the difference was not significant. Processing speed, as measured by Trail Making A and B, appears to get slower in each group. The performance of group 3 was slower than the other groups in this cognitive domain; however, it only approached significance after the Bonferroni correction. The time for the MCAD group Trail Making A and B performance was more than 1 and 2 minutes, respectively, differentiating this group from other patients.
| Comment |
|---|
|
|
|---|
Possible Mechanisms of Cognitive Dysfunction
A number of factors may contribute alone or in combination to neurobehavioral deficits in the advanced HF population. Whereas the low cardiac output is known to contribute to the cognitive abnormalities [29, 30], it is conceivable that systolic hypotension and ischemia in the absence of cerebrovascular events [31] may add to neurobehavioral dysfunction, with the role of hypotension particularly important in women [3235]. Patients with HF have been shown to experience losses in cerebral grey matter in the mesial temporal lobe, including the hippocampus and parahippocampus, regions particularly sensitive to hypoxia and accountable for memory changes. Brain volumetric analysis shows associated grey matter loss in regions involving the insula, basal ganglia, cerebellum, dorsal midbrain, and frontal regions [36]. Further, the use of intravenous inotropic agents may complicate a cognitive disturbance; these agents have significant apoptogenic and necrotogenic potential and may inflict direct damage. Inotropic agents such as dobutamine are known to lead to myocellular toxicity [37, 38]; however, any parallel effects on neurobehavioral functioning in this study in the absence of controls is speculative.
Potential Clinical Implications of Cognitive Dysfunction
The present study raises an important question. Should assist devices be offered to patients sooner, that is, before they develop the progressive neurobehavioral compromise? Assessing presurgical cognitive compromise as outlined by the Food and Drug Administration mandate [39] is especially important as devices take on a more prominent role in destination therapy and with irreversible cognitive decline a likely relative contraindication [40]. Since there is growing evidence to show left ventricular mechanical support offers greater improvement with heart failure symptoms compared with inotropic therapy, and as observed in the present study if early mechanical support spares further cognitive decline, it may be prudent to recommend MCAD implantation as soon as there is relative evidence of deficits in mental processing speed and memory function deterioration begin to appear. Although MCAD implantation may offer an increased degree of cognitive stability, microemboli will continue to remain a threat. A longitudinal design would assist with the understanding of important variables, such as monitoring microembolic events and whether early MCAD placement leads to cognitive decline, stabilization, or improvement. The question of whether avoiding prolonged inotropic support by earlier MCAD implantation and thereby preventing further cognitive compromise needs consideration.
Limitations of the Study
Many of the sicker HF patients could not complete the entire neuropsychological battery for a number of medically related reasons. Testing was completed by the most able of this sick population. Only those with complete data could be included in the analysis, thus decreasing sample size and inducing selection bias. The second limitation of the study involves the lack of controls for medications, hemodynamics, or comorbid medical conditions. Any and all of these factors can influence neurobehavioral outcomes. The inclusion criteria were limited to NYHA class III to IV symptoms, an ejection fraction less than 20%, heart transplant candidacy, and without controls. The third limitation results from having less than 20% female patients in our study population, albeit this roughly represents the number of women with heart failure. Fourth, heart failure treatment regimes have changed over the 18-year period, further confounding the design beyond our control. Fifth, there was no premorbid cognitive parameter available for comparison at the time of the neuropsychological testing except for the educational history. Sixth, given the prevalence of ischemic heart disease in this group, the findings from transesophageal echocardiography and vascular studies to detect cerebral emboli would be appropriate. Lastly, we have not provided another study group for cognitive comparison such as a cardiac bypass population or patients with another chronic noncardiac disease process. Without a cross-group comparison, findings from the present study cannot be interpreted as unique to the advanced HF population. This limitation also may not allow exclusion of the confounding effects of use of sedatives, physical deconditioning, and transitional metabolic alterations. Despite the limitations, this is a cross-sectional cognitive study comparing prospective MCAD patients with other HF cohorts.
Suggested Cognitive Screening With This Cardiac Group
As a result of this research, it is clear that modifications to standard cognitive protocols should be implemented with this fragile population. A brief screening battery is suggested because of the patient's physical condition and time limitations. We suggest that auditory and visual memory functions with delayed procedures (WMS) in combination with processing speed and executive abilities (Trail Making) are essential components. Isolated motor functions should be eliminated. Reviewing language functions and spatial processing should be added.
Does Cognitive Dysfunction Represent Yet Another End-Organ Dysfunction?
Despite these limitations, we consider these results substantially informative, adding to our understanding of neurobehavioral perturbations in patients with advanced HF. Our ongoing research involves a larger sample size with control groups, serial cognitive testing before and after implant, corroborative family information, and longitudinal follow-up. It is possible that the cognitive dysfunction represents another end-organ dysfunction in advanced HF. If it is demonstrated that MCAD implantation can arrest progressive neurobehavioral decline, decisions for earlier intervention would become logical.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
R. L.C. Vogels, P. Scheltens, J. M. Schroeder-Tanka, and H. C. Weinstein Cognitive impairment in heart failure: A systematic review of the literature Eur J Heart Fail, May 1, 2007; 9(5): 440 - 449. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |