| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Eur J Cardiothorac Surg 2005;28:88-96
© 2005 Elsevier Science NL
a Department of Thoracic and Cardiovascular Surgery, West German Heart Centre, University Hospital Essen, Hufelandstrasse 55, 45122 Essen, Germany
b Department of Neurology, University Hospital Essen, Essen, Germany
c Institute of Diagnostic and Interventional Radiology, University Hospital Essen, Essen, Germany
Received 6 October 2004; received in revised form 23 February 2005; accepted 24 February 2005.
* Corresponding author. Tel.: +49 201 723 4901; fax: +49 201 723 5451. (Email: stephan.knipp{at}uni-essen.de).
 |
Abstract |
|---|
 Top Abstract 1. Introduction2. Materials and methods3. Results4. DiscussionAppendix A. Conference...References |
|---|
3) lesions (range, 1–7). Lesion volume ranged from 50–500mm3 except 1 territorial infarct of 1900mm3. Of a total of 41 lesions, 27 (66%) were located in the right hemisphere and 32 in a subcortical location. By 5 days postoperatively, significant neurocognitive decline was observed in 5 of 13 tests affecting memory, attention and rate of information processing. By 4 months, dysfunction had recovered in all cognitive areas. The presence of new ischemic lesions was not associated with neurocognitive decline at discharge. There was also no significant correlation between clinical and operative variables and the presence of new DW lesions or neuropsychological outcome. Conclusions: Following cardiac valve replacement, new small ischemic brain lesions were detected by diffusion-weighted MRI. Neurocognitive decline was present early after operation, but resolved within 4 months. A correlation of new ischemic lesions to postoperative cognitive dysfunction or clinical variables was not found.
Key Words: Cardiac valve replacement • Magnetic resonance imaging • Neurocognitive function • Diffusion-weighted sequences
|
1. Introduction |
|---|
|
TopAbstract1. Introduction2. Materials and methods3. Results4. DiscussionAppendix A. Conference...References |
|---|
The pathogenesis of neurocognitive decline after heart surgery is still unknown, and many mechanisms have been postulated to be involved including cerebral embolization, hypoperfusion, edema, inflammation and metabolic disturbances [6–8]. Neuroimaging methods such as magnetic resonance imaging (MRI) have been increasingly employed in the study protocols after cardiac surgery since cerebral ischemia secondary to embolization, hypoperfusion, or a combination of these is considered to be a crucial pathophysiological factor underlying postoperative neurocognitive impairment [3]. Using conventional MRI, new ischemic cerebral lesions have been detected in up to a third of patients after coronary revascularization [9–11]. Advanced MRI techniques with diffusion-weighted (DW) sequences are more sensitive than conventional MRI and can detect ischemia within minutes after onset [12]. Patients with postoperative stroke nearly always demonstrate diffusion abnormalities [13]. Some investigators have observed ischemic brain lesions after CABG even in patients without focal neurological deficits [10,14], and in the few studies that systematically investigated the correlation between the appearance of new lesions on MRI and neurocognitive function, results were variable [9–11,14]. As far as cardiac valve surgery is concerned, studies using DW MRI and neurocognitive testing do not exist until present.
The aim of this study was to prospectively evaluate brain injury after cardiac valve replacement by the use of diffusion-weighted MRI and to determine its relationship with neurocognitive function and clinical data.
|
2. Materials and methods |
|---|
|
TopAbstract1. Introduction2. Materials and methods3. Results4. DiscussionAppendix A. Conference...References |
|---|
2.2. Clinical examination
A comprehensive clinical examination, including medical history, physical and detailed neurological examination, was performed at all three points in time during the study. In addition, patients received blood tests, an electrocardiogram and chest X-ray daily during hospitalization and an echocardiogram before surgery, in the operating room and at least once after operation.
2.3. Neuropsychological assessment
Neurocognitive function was assessed using standardized and well-validated tests. The test battery consisted of 11 psychometric tests covering major cognitive areas known to be vulnerable to organic damage: learning and memory, attention and psychomotor speed, information processing, logical thinking and visuo-spatial perception. The emotional status and depression were rated by two questionnaires. The Reitan Trail making test part B (TMB) and Zimmermann's divided attention test was used to assess attention and psychomotor speed. In the Reitan Trail making test part A (TMA) (examining rate of information processing, psychomotor tracking speed, hand-eye coordination) the task is to connect, by drawing a line, numbers in a sequence (i.e. 1–2–3 etc.), in the TMB test numbers and letters in a sequence (i.e. 1A–2B–3C etc.), as quickly as possible. The Verbal learning test (using Nü;rnberger age inventory) was administered to test for learning and memory of words. The test consisted of an oral presentation of 8 semantically unrelated words that the patient was requested, first, to learn and recall immediately (word list-immediate recall), and then, after a 30-min delay, to recognize those words (out of list of 15), which were presented earlier (word list—delayed recognition). The Digit span subtest of Wechsler memory scale-Revised is a task requiring short-term memory (Digit span forward) and working memory (Digit span backward) for verbal material. The Corsi block-tapping test was used to examine short-term and working memory for visual structures. The investigator tapped with his fingers on a series of blocks, and the patient was requested to recall the sequence, first forwards, then backwards. Visuo-spatial function and recognition of categories and regularities of geometrical objects were tested with Horn's performance test 55+ subtest 9 and 3, respectively. Except for TMA, TMB and divided attention test, a higher score indicates a better function. Mood was rated with von Zersson test (28 items) and depression with the General depression scale—short version (derived from Hospital anxiety and depression scale, 15 items). In these scales, a lower score indicates a lower level of discomfort and depression. All tests were conducted by the same neuropsychologist. Parallel sets of tests were used to minimize the practice effect. Except for TMA, TMB and divided attention test, the individual test scores at each testing point (raw data) were converted into t-scores and ranks of percentage according to reference lists. By this, each patient's performance was comparable with that of healthy subjects of equal age and sex.
2.4. Magnetic resonance imaging of the brain
MR scans of the brain were performed on a standard 1.5-T whole body imaging system (Sonata, Siemens AG, Erlangen, Germany). The protocol included (1) a conventional transaxial T1 (repetition time [TR], 500ms; echo time [TE], 14ms; averages, 2; matrix, 256x256), (2) a transaxial T2 turbo spin-echo (TR, 5120ms; TE, 104ms; averages, 1; matrix, 256x256), and (3) a transaxial fluid-attenuated inversion recovery (FLAIR)- (TR, 9000ms; TE, 115ms; average, 1; matrix, 256x256) weighted sequence. Diffusion-weighted imaging was performed by a single-shot, transaxial spin-echo planar-imaging sequence of the whole brain (TR, 4600ms; TE, 137ms; averages, 2; matrix, 128x128; gradients of b-values, 0, 500 and 1000s/mm2). Field of view was 230mm and slice thickness was 6mm for all sequences. Preexisting brain abnormalities (e.g. microangiopathy, infarctions or atrophy) and the appearance of new DW lesions on postoperative scans were evaluated. Scans were read by two experienced neuroradiologists blinded to the clinical and neuropsychological data of the patient up to the time of examination (before or after surgery). In case of discrepancy, a consensus reading was held. For volume quantification, the images were magnified fourfold, the area of lesion was manually delineated in each image slice by region of interest, and the volume was calculated using standard scanner software (Syngo, Siemens AG, Erlangen, Germany).
2.5. Anesthesia and surgical procedure
Standard anesthesia techniques were used to perform heart valve replacement with the assistance of cardiopulmonary bypass (CPB). Patients received 1mg flunitrazepam orally for premedication and ethmidate, sufentanil and rocuronuim intravenously to induce general anaesthesia, followed by isofluran inhalation to maintain it. The surgical procedure was performed through a median sternotomy. The CPB circuit consisted of a membrane oxygenator and a 40-µm arterial filter. The lining system was filled with 1600ml Ringer lactate, mannitol, gelafundin and sodiumbicarbonate. Heparin was administered before cannulation to achieve an activated clotting time above 400s during CPB. A curved aortic cannula, a cardioplegia aortic root vent, and a two-stage or two single-stage venous cannula, where appropriate, were used. The operation was performed under moderate hypothermia (30°C) and in cardiac arrest induced by cold Bretschneider solution. During CPB, non-pulsatile pump flow was maintained at 2,4l/minm2. Mean arterial pressure (MAP) was kept above 50mmHg. Arterial partial pressure of carbon dioxide was maintained at 35–40mmHg and arterial pressure of oxygen at 200–250mmHg during perfusion. Acid–base status was monitored with the 
-stat protocol. Hemotocrit was kept above 21% during CPB with packed red blood cells if necessary. De-airing was achieved by (1) flushing the open ascending aorta by slightly releasing the cross-clamp, (2) venting of the left ventricle via the pulmonary vein, (3) venting of the ascending aorta, (4) tilting the operating table to ease release of air, (5) manual decompression of the left atrial appendage and, in some cases, (6) retrograde de-airing via the apex of the left ventricle. After decannulation, protamine sulfate was administered until preoperative activated clotting time was achieved. Postoperatively, patients were weaned from mechanical ventilation under mild analgosedation with propofol. Opioids were administered until postoperative day 2 when chest tubes usually could be removed. Patients were treated postoperatively with unfractionated heparin during the entire course of hospital stay (activated prothrombin time-target range 40–60s).
2.6. Statistical analysis
All statistical analysis were performed with a statistical software package (Statistical Product and Service Solutions, SPSS, version 11.0, SPSS Inc., IL, USA). Differences were considered statistically significant at P<0.05. Normality of distribution was tested with the Kolmogorov–Smirnov test. Neuropsychological test scores were compared between the three testing points by using univariate analysis of variance and the Friedman test, respectively. Differences between preoperative and postoperative test scores were analyzed with paired Student's t-test and the Wilcoxon test, respectively. To analyze the relation between neurocognitive performance after surgery and clinical data, a multivariate analysis with repeated measures was performed. The decline in neuropsychological test scores after surgery was compared between patients with and without new lesions on MRI by means of unpaired Student's t-test, and if being significantly different, by regression analysis. Relations between the presence of new MRI lesions and clinical and surgical data were analyzed by Student's t-test, Wilcoxon test and Fisher's exact test, where appropriate. Data are reported as mean±SD.
|
3. Results |
|---|
|
TopAbstract1. Introduction2. Materials and methods3. Results4. DiscussionAppendix A. Conference...References |
|---|
|
|
|
|
|
|
|
|
|
4. Discussion |
|---|
|
TopAbstract1. Introduction2. Materials and methods3. Results4. DiscussionAppendix A. Conference...References |
|---|
With advances in surgical techniques and anesthesia, the rates of mortality and cardiac morbidity after heart operations have decreased with time, allowing to offer the procedure to patients with more underlying disease and elderly patients. Nevertheless, cardiac surgery is associated with a substantial risk for adverse neurobehavioural outcome. After CABG, the incidence of stroke is described to range from 1.5 to 5.2% in prospective studies [1,2]. A much larger proportion of patients encounter subtle, subclinical postoperative changes in neurocognitive function that are detected by a decrease in neuropsychological test scores. Short-term cognitive decline is reported in 30–80% of patients undergoing CABG with cardiopulmonary bypass [4]. Usually cognitive complaints disappear within 6 months after surgery, but some of the cognitive changes may well persist a year or more [5]. Knowledge about incidence and time course of neurobehavioural outcome following cardiac valve surgery is far less, and the data obtained on CABG patients may not necessarily pertain to valve replacement (VR) patients due to considerable differences between intracardiac and extracardiac procedures.
In the present study, none of the patients revealed symptoms of a focal neurological deficit on serial postoperative neurological examinations. Neurocognitive testing at discharge compared to baseline performance disclosed impaired cognitive function in 5 of 13 tests. The deterioration was seen in various cognitive areas affecting attention and psychomotor speed (including hand-eye coordination), information processing and memory functions. This observation is in line with numerous other investigators [2,5,15], and particularly complaints related to memory belong to the most common described during the first weeks after cardiac surgery [3]. In contrast to CABG, studies that examine VR patients with the use of neuropsychological tests and with a follow-up beyond two months after operation are scanty. When, instead of psychometric tests, P300 auditory evoked potentials are considered as measure of neurocognitive outcome, it was shown that 7 days after aortic valve replacement, P300 peak latencies were prolonged (impaired) compared with preoperative values [16]. This decrease in P300 latencies was independent of the type of valve prosthesis [17]. Four months postoperatively, neurocognitive decline was reversible in patients with mechanical aortic valve replacement, but not in patients with biological aortic valve replacement [17]. For this contrary development, the authors claimed age to be most important, and cerebral damage related to the type of valve prosthesis may be overestimated. In our cohort, no persistent impairment of neurocognitive function was present 4 months after cardiac valve replacement. In fact, a great deal of evidence exists that advanced age is a crucial variable for the development of postoperative neurocognitive dysfunction [3]. It is left to future studies with elderly patients to elucidate whether the restoration of cognitive function described in our cohort may be ascribed to their relatively young age.
The etiology of neuropsychological impairment after cardiac surgery with CPB is unresolved. To enhance the understanding of the pathogenesis of this disorder, neuroimaging methods including magnetic resonance imaging have been increasingly implied in the study protocols before and after operation. Using conventional MRI, new ischemic brain lesions were detected in up to a third of patients after CABG [9–11,18]. The availability of advanced MRI technology using diffusion-weighted sequences has increased the sensitivity and specificity for the detection of new postoperative brain damage decisively. DW MRI increases the conspicuity of new lesions and allows to differentiate between old and new abnormalities. For this, DW MRI is superior to conventional MRI where the development of new lesions may be underestimated by T2- weighted sequences [9,14]. In a study on 13 patients undergoing CABG, diffusion-weighted lesions were found in 4 patients (31%) with one of them having a focal neurological deficit [19]. In a larger series, Bendszus and colleagues [14] investigated 35 CABG patients and found new ischemic lesions in 9 patients (26%) with none of the patients having neurological deficits. In a very recent study on patients undergoing aortic valve replacement, new DW lesions were reported to be present in 14 of 37 patients (38%) of whom only 3 patients developed focal neurological deficits [20]. In the present study, DW MRI displayed areas of diffusion restriction in 47% of patients after cardiac valve replacement. All lesions were clinically silent, and except for 1 lesion representing an embolic territorial cerebellar infarct, lesions were small ranging from 50 to 500mm3. A similar size of lesions was found in a most recent investigation of our group performed on CABG patients [21]. The incidence of new lesions reported here is essentially in line with reports by other groups, although in our study lesions are slightly more frequent and none of our patients revealed neurological deficits. A possible explanation for this may be that we did not only enrol patients receiving aortic valve replacement but also included patients with double valve surgery resulting in an increased risk for embolization. Comparing the incidence of DW lesions in VR patients with that in CABG patients, the greater ischemic damage following valve surgery is probably due to the operative procedure itself (e.g. opening the ascending aorta, decalcification of the valvular ring, cutting out calcified valves). Consistently, a higher number of intraoperative cerebral microembolic signals were detected in VR as compared to CABG patients [22].
The nature of the small lesions on DW MRI is uncertain. There is good evidence, however, to assume that the new lesions correspond to focal cerebral ischemia secondary to embolization. We base this assumption on the radiographic appearance of the lesions (globular, small), their diffuse distribution in almost all vascular territories, their predominance on the right side (66%) and their multiplicity (only 4 patients had a single lesion). Further evidence comes from histological examination of the brain of patients who died after CABG with CPB demonstrating thousands of emboli lodged in the microvessels [23]. Transcranial Doppler can detect many microembolic signals during CPB-assisted cardiac surgery [7,22]. Macroemboli arise from the cardiac chambers or the bypass circuit or from atheromatous plaques in the ascending aorta or aortic arch and are likely to produce clinically relevant infarction [24]. Microemboli, made of gas or solid matter, can frequently be observed during CPB-assisted cardiac surgery and occur more often after cardiac valve replacement than after CABG [22]. Therefore, we hypothesize that the majority of lesions in this study correspond to small brain infarcts caused by thromboemboli or macroscopic air bubbles.
The correlation between the presence of new ischemic brain lesions and postoperative neuropsychological dysfunction is still uncertain. In the few studies that were performed on CABG patients, the relation was variable [9–11,14,19]. In agreement with our previous investigation on CABG patients, also in patients undergoing cardiac valve replacement, we failed to find differences in postoperative neuropsychological performance between patients with and without new DW lesions [21]. Further studies are needed to elucidate the clinical significance of new small brain infarcts for the development of neurobehavioural outcome after cardiac valve surgery.
There are some limitations to this study. First, although diffusion-weighted MRI is superior to conventional MRI, spatial resolution is limited. The method only discloses lesions that extend the pixel size, so that showers of microscopic air emboli are likely to be missed. Our findings may only be the ‘tip of an iceberg’, and the ‘true’ ischemic burden to the brain may be underestimated. DW MRI seems to evolve as a good tool in studies for acute stroke, but whether it is a surrogate marker for brain injury in patients with milder neurobehavioural outcome remains to be awaited. As in all studies using MRI, another limitation is due to the number of patients included (even though only few studies have more patients and studies that examined patients undergoing valve replacement by means of DW MRI and neurocognitive testing do not exist). For this, the absence of correlation between postoperative DW lesions and neurocognitive deficits needs to be confirmed in greater cohorts. The length of follow-up in this study was 4 months. It might be concluded that the decline in function after cardiac valve replacement is only transient and reversible in the majority of patients. However, it is reported that cognitive decline at discharge is a predictor of long-term neurocognitive dysfunction [5]. A reevaluation of patients after one year would help to address this issue. Finally, this study is limited by the lack of a control group. Although results of numerous studies provide evidence of embolization to be a crucial mechanism, the underlying reasons for neurocognitive impairment after cardiac surgery with cardiopulmonary bypass remain uncertain.
In conclusion, in this study we combined the new magnetic resonance technique of diffusion with comprehensive neuropsychological assessment as measures of brain injury after cardiac valve replacement. New DW lesions, consistent with small embolic infarctions, were detected in almost half of the patients. Neurocognitive function was severely impaired early after surgery, but resolved within 4 months. The appearance of new ischemic lesions was not associated with neurocognitive performance and clinical and surgical variables.
|
Appendix A. Conference discussion |
|---|
|
TopAbstract1. Introduction2. Materials and methods3. Results4. DiscussionAppendix A. Conference...References |
|---|
Dr Knipp : No. In this cohort only one cross-clamping was used. However, it is well known that the repositioning of the cross-clamp is significantly associated with a high release of emboli. So this would certainly increase the risk of development of brain lesions.
Dr Revuelta : And in your opinion the aortic wall was free from calcification in all cases?
Dr Knipp : We didn't scan the aorta, the ascending aorta.
Dr Revuelta : I mean just touching the aorta.
Dr Knipp : Yes, of course, we did this, and if there was a hint for gross atherosclerotic disease of the ascending aorta, the clamp was tried to be positioned away from this sclerotic area.
Dr F. Mohr (Leipzig, Germany): How do you explain the disappearance of the infarcts up to three months? Is that an incidence of air embolism, for example, number one? Can you answer that first?
Dr Knipp : The appearance of the lesions early after surgery on diffusion-weighted imaging suggests that these lesions are ischemic lesions secondary to embolization, and because of the volume of the lesions, it is very likely that they are thromboemboli or macroscopic air bubbles, but you can't differentiate from transcranial Doppler or from radiography the exact nature of the lesions. And typical for diffusion-weighted imaging is that the lesions come up in this technique early after surgery, much earlier than on conventional MR imaging, but then disappear, at least on diffusion-weighted images, often also on T2-weighted images, late after surgery. If there is a persistent area of signal hyperintensity on T2-weighted images three months after surgery, it is very much suggestive of persistent neuronal damage.
Dr Mohr : Second question. Your early neurocognitive dysfunction has usually been evaluated on day 7, 8, that is what I would expect, postsurgery. Is that right?
Dr Knipp : Yes, about six days after surgery.
Dr Mohr : So how do you distinguish the influence of sedative medication, because many of these patients take some drugs overnight and sleep, which will have an impact on their neurocognitive function? Did you look for that?
Dr Knipp : Yes. The examination was performed by experienced neurologists and neuropsychologists, and all of the patients had these examinations when they were free of the influence of any opioids and sedation and so on. Often they could be brought to the examination by walking.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Presented at the joint 18th Annual Meeting of the European Association for Cardio-thoracic Surgery and the 12th Annual Meeting of the European Society of Thoracic Surgeons, Leipzig, Germany, September 12-15, 2004. |
References |
|---|
|
TopAbstract1. Introduction2. Materials and methods3. Results4. DiscussionAppendix A. Conference...References |
|---|
This article has been cited by other articles:
|
|
 |
|
  E. F.M. Wijdicks, N. Campeau, and T. Sundt Reversible unilateral brain edema presenting with major neurologic deficit after valve repair. Ann. Thorac. Surg., August 1, 2008; 86(2): 634 - 637. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Palombo, V. Faraglia, N. Stella, E. Giugni, A. Bozzao, and M. Taurino Late Evaluation of Silent Cerebral Ischemia Detected by Diffusion-Weighted MR Imaging after Filter-Protected Carotid Artery Stenting AJNR Am. J. Neuroradiol., August 1, 2008; 29(7): 1340 - 1343. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Yamada, Y. Nagakane, H. Sasajima, M. Nakagawa, K. Mineura, T. Masunami, K. Akazawa, and T. Nishimura Incidental Acute Infarcts Identified on Diffusion-Weighted Images: A University Hospital-Based Study AJNR Am. J. Neuroradiol., May 1, 2008; 29(5): 937 - 940. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. A. Barber, S. Hach, L. J. Tippett, L. Ross, A. F. Merry, and P. Milsom Cerebral Ischemic Lesions on Diffusion-Weighted Imaging Are Associated With Neurocognitive Decline After Cardiac Surgery Stroke, May 1, 2008; 39(5): 1427 - 1433. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Kunihara, D. Tscholl, F. Langer, G. Heinz, F. Sata, and H.-J. Schafers Cognitive brain function after hypothermic circulatory arrest assessed by cognitive P300 evoked potentials Eur. J. Cardiothorac. Surg., September 1, 2007; 32(3): 507 - 513. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. F. Gottesman, A. E. Hillis, M. A. Grega, L. M. Borowicz Jr, O. A. Selnes, W. A. Baumgartner, and G. M. McKhann Early Postoperative Cognitive Dysfunction and Blood Pressure During Coronary Artery Bypass Graft Operation Arch Neurol, August 1, 2007; 64(8): 1111 - 1114. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Kastrup, K. Groschel, and U. Ernemann Response to Letter by Cohen Stroke, May 1, 2007; 38(5): e19 - e20. [Full Text] [PDF]
|
|
|
|
|
|
T. F. Floyd, P. N. Shah, C. C. Price, F. Harris, S. J. Ratcliffe, M. A. Acker, J. E. Bavaria, H. Rahmouni, B. Kuersten, S. Wiegers, et al. Clinically Silent Cerebral Ischemic Events After Cardiac Surgery: Their Incidence, Regional Vascular Occurrence, and Procedural Dependence Ann. Thorac. Surg., June 1, 2006; 81(6): 2160 - 2166. [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 |
