Eur J Cardiothorac Surg 1999;15:166-172
© 1999 Elsevier Science NL
Predictors of neurological morbidity after coronary artery bypass surgery
Douglas P.B. Janssena,
Luc Noyeza,*,
Johannes A.M. van Drutenb,
Stefan H. Skotnickia,
Leon K. Lacqueta
a Department of Thoracic and Cardiac Surgery 414, University Hospital Nijmegen, Nijmegen, The Netherlands
b Department of Medical Informatics, Epidemiology and Statistics 152, University of Nijmegen, Nijmegen, The Netherlands
Received 31 August 1998;
received in revised form 18 November 1998;
accepted 25 November 1998.
* Corresponding author. Department of Thoracic and Cardiac Surgery 414, University Hospital Nijmegen, P.O. Box 9101, 6500 HB Nijmegen, The Netherlands. Tel: +3-24-361-4744; fax: +31-24-354-0129; e-mail: l.noyez@thchir.azn.nl
 |
Abstract
|
|---|
Objective: The aim of this study was to analyze the postoperative neurological complications after myocardial revascularization. Methods: We analyzed the pre-, peri- and postoperative data of 3834 patients who underwent a primary isolated bypass grafting between January 1987 and December 1995. Postoperative neurological complications (A) were divided into mild complications (B) and major complications (C). Results: The incidence of A increased, from 1.4% to 3.0%. Unifactor risk analysis identified: age>75 years, peripheral vascular atherosclerosis, neurological pathology, aorta-pathology and perioperative myocardial infarction as risk factors for A. Perioperative myocardial infarction and neurological pathology for B; age>75 years, peripheral vascular atherosclerosis, neurological pathology, perioperative myocardial infarction and aorta pathology for C. Multifactor risk regression analysis identified peripheral vascular atherosclerosis, neurological pathology, aorta-pathology, perioperative myocardial infarction and the time cohort 1993–1995 as independent predictors for A; perioperative myocardial infarction and the time cohort 1993–1995 for B; neurological pathology, aorta-pathology and perioperative myocardial infarction for C. Conclusions: Peripheral vascular atherosclerosis, neurological pathology, aorta-pathology, the occurrence of a perioperative myocardial infarction and the time cohort 1993–1995 are identified as independent risk factors for neurological complications.
Key Words: Myocardial revascularization • Comorbidity • Neurological complications • Multifactor analysis
|
Introduction
|
|---|
Neurological morbidity complicates myocardial revascularization in 2–6% of patients. In addition, cognitive impairment occurs in up to 30% of patients [1][2][3]. In a previous report, we described an increase of the postoperative neurological complications from 1.4% to 3.0% between 1987 and 1995 [4]. Although most patients show a good functional recovery within the first half year, these neurological complications results in a prolonged hospital stay, with financial consequences [5].
The purpose of this study was to analyze postoperative neurological complications in relation to pre- and perioperative risk factors, so that we can predict patients at increased risk. The following main questions are addressed: (a) which of the variables measured pre-, and perioperatively can be considered as risk factors for post-operative neurological complications? (b) Which of these factors contribute independently to an increased risk of postoperative neurological complications? (c) Which of these independent predictors change over time?
|
Materials and methods
|
|---|
Patients
With the aid of our database, Coronary Surgery Database Radboud Hospital (CORRAD), a registry that stores pre-, peri-, and postoperative data on all patients undergoing isolated coronary bypass grafting (CABG), we identified a series of 3834 patients undergoing a primary isolated CABG from January 1987 to December 1995. These 9 years are subdivided into three time cohorts of 3 years. January 1987 to December 1989 (1292 patients), January 1990 to December 1992 (1130 patients), and January 1993 to December 1995 (1412 patients). Of these 3834 patients, 85 (2.2%) had postoperative neurological complications. Postoperative neurological complications was defined as when there was a new cerebrovascular accident and/or transient ischaemic attack, so called major complications, but also when there was confusion for more than 12 h or memory disturbances, so called mild complications [6]. Table 1 presents the frequency of neurological complications (A) in the three time cohorts and also a division between mild complications (B) and major complications (C).
Preoperatively, diabetes was defined as having a positive glucose tolerance test, oral antidiabetic medication, or insulin dependency. Hypertension was defined as systolic blood pressure >160 mmHg or diastolic pressure >100 mmHg. Peripheral vascular atherosclerosis was registered in patients with either central (carotic) or peripheral vascular disease or operations. Neurological pathology was registered in patients with a cerebrovascular accident, transient ischaemic attack, or a documented history of cognitive disturbances preoperatively. Emergency operation is defined as operation for involving myocardial infarction, ischaemia not responding to medical therapy, or cardiogenic shock. Perioperative, aorta-pathology is noted in patients, with sever calcification in the aorta ascendens. Perioperative myocardial infarction as a new postoperative Q wave or T wave accompanied by increased cardiac enzymes (CPK-MB>10%).
Surgical technique
All patients were operated on using standard cardiopulmonary bypass techniques, aortic and right atrial (two stage) cannulation, and hypothermia (28°–32°C). Myocardial protection during aortic cross-clamping was performed with infusion of cold (4°C) St. Thomas' Hospital cardioplegia until asystole occurred and was maintained by reinfusion of 100 ml/m2 of the solution every 25–30 min or earlier, as needed. Over the time there is no significant difference in bypass and aortic cross-clamp time, neither in number of grafts and distal anastomoses. The only surgical change is the increase in the use of arterial grafts [4].
Statistical analysis
To test which variables can be considered risk factors for neurological complications, Fisher's exact test for 2x2 tables was used (unifactor risk analysis). Changes in the incidence of post-operative neurological complications and changes in the prevalence of risk factors in three time cohorts were examined with the 
2-test. Multiple logistic regression analysis was used to identify risk factors which independently contribute to an increase in risk of post-operative neurological complications (multifactor risk analysis). In the analysis the three time cohorts were represented by two indicator variables, the cohort 1987–1989 was used as reference for the other two cohorts. Statistical significance was assumed at P<0.05.
|
Results
|
|---|
Table 1 shows that the incidence of post-operative neurological complications (A) increases with time, from 1.4% to 3.0% (P=0.02). This increased risk is mainly due to an increased risk for mild complications (B).
Unifactor risk analysis
Table 2 lists the variables tested for post-operative neurological complications. For each variable the risk was calculated for complications A, B and C. Table 3 shows the results of the unifactor risk analysis; per variable the relative risk (RR), ratio of incidence rates, is shown for complication A, B and C. The risk for mild and/or severe neurological problems (A) is not related to sex, diabetes, perfusion time and aortic cross-clamp time. Hypertension and emergency operation just reach the significance level for complication A (P=0.05). Clear risk factors for post-operative neurological complications (A) are: age>75 years, peripheral vascular atherosclerosis, neurological pathology, aorta-pathology and perioperative myocardial infarction. Neurological pathology (P=0.03) and certainly perioperative myocardial infarction (P=0.0005) are associated with a risk for mild neurological complications (B). Age (>75 years) (P=0.008), peripheral vascular atherosclerosis (0.02), perioperative myocardial infarction (P=0.01), but certainly neurological pathology (P=0.003), and most evident aorta-pathology (P<0.0001) are strongly associated with a high relative risk for severe postoperative complications (C).
Multifactor risk analysis
Table 4 shows the results of the multiple logistic regression analysis (for explaining the risk for respectively complication A, B and C). Independent predictors for post-operative complications (A) are peripheral vascular atherosclerosis (P=0.03), neurological pathology (P=0.006), aorta-pathology (P=0.001), perioperative myocardial infarction (p=0.0001), and the time cohort 1993–1995 (P=0.03). Age, diabetes, hypertension and emergency operation are not independent predictors in the multifactor analysis. Only perioperative myocardial infarction (P=0.0002) and the time cohort 1993–1995 (P=0.03) are identified as predictors for mild neurological complications (B). Predictors for severe postoperative complications (C) are neurological pathology (P=0.02) and perioperative myocardial infarction (P=0.04), but in the first place is aorta-pathology (P=0.0001) an independent predictor.
Table 5 shows the change of prevalence of the independent risk factors in the three time cohorts. Only the prevalence of preoperative neurological pathology is clearly increased in the time cohort 1993–1995 (P=0.001). Table 6 shows the relation between the independent risk factors, peripheral vascular atherosclerosis, neurological pathology, aorta-pathology and perioperative myocardial infarction, and the risk factors, emergency operation and hypertension, identified in the unifactor analysis. Emergency operation is very strongly related to an increased risk of perioperative myocardial infarction (elective 1.7%; emergency 15.4%). Hypertension is associated with an increased risk for peripheral vascular atherosclerosis, aorta-pathology and perioperative myocardial infarction.
View this table:
[in this window]
[in a new window]
|
Table 6. Relationship between the risk factors emergency operation, hypertension and independent risk factors for postoperative neurological complications
|
|
|
4. Discussion
|
|---|
Neurological complications or problems represent one of the most costly complications of cardiac surgery. This morbidity is attended with prolonged ICU and/or hospital stay, and eventually rehabilitation programs [1][2][3][5][7]. Harrison showed that, severe postoperative neurological problems, such as stroke, is mostly due to macroembolism and that microembolism can cause mild neurological problems, such as confusion and memory disturbances [6]. We differentiated in our analysis between three groups of neurological complications, first the total number (severe and/or mild), second, the severe complications, which may hide mild complications, and third, only the mild complications.
This study demonstrates that the significant increase of neurological complications over the years is mainly due to the increase of the mild neurological complications (Table 1). The unifactor risk analysis shows no predictive value of the variables sex, perfusion time>100 min and aortic cross-clamp time>60 min (Table 3). The question of a possible influence of bypass time on neurological complications is interesting, because of the evolution of coronary artery bypass surgery without pump. In a study of patients with a history of stroke, patients with postoperative neurological deficit had a significant longer bypass time than patients without deficit. However, those with a deficit were older, received more distal anastomoses, and more patients had a valve replacement. It is known that valve operations had a higher incidence of cerebral injuries compared with isolated myocardial revascularizations [5]. A recent study, comparing the serum S-100ß levels, which are increasing with neurological injury, showed a significant greater elevation in intracardiac procedures as in CABG and a correlation with increasing patients-age, but not with bypass time [8]. However, Westaby and colleagues [9] demonstrates that the elevation in serum levels S-100ß did not occur in patients who underwent CABG without bypass. A recent study of Browne et al., (Browne S, Westaby S, Taggart DP. Neuropsychological and respiratory injury following CABG surgery with and without cardiopulmonary bypass. Presented at the 11th Annual meeting of the European Association for Cardio-Thoracic Surgery, Copenhagen, Denmark, September 1997), comparing neuropsychological and respiratory injury following CAGB with and without cardiopulmonary bypass showed that bypass does contribute to respiratory injury, but does not appear to be responsible for neuropsychological problems. The association between age and neurological complications is described in other studies, several mechanisms are therefore responsible, the increasing prevalence of atherosclerosis as well as an increased risk of plaque embolization is more likely [10][11]. The fact that age is a significant predictor for sever neurological complications and not for confusion is supporting this theory. Hypertension and elective/emergency operation show a weak, significant association with neurological complications, however, this association disposes when severe and mild complications are studied separately.
Neurological pathology and perioperative myocardial infarction were significant predictors for neurological problems, as good for severe and/or mild problems. The vulnerability of patients with a previous history of cerebrovascular accident and/or ischemic attack, is supported by several studies [5]. The recent study of Taggart [8], however, suggests that the period between the cerebral injury and the operation may be important. Perioperative myocardial infarction shows only a weak significant correlation with severe neurological problems, but a strong correlation with confusion. Peripheral vascular atherosclerosis, significant for predictor for neurological problems, tend to be significant for confusion, but is clear significant for cerebrovascular accident and/or transischemic attack. Also here we see a support of the theory of Harrison [6]. Aorta-pathology, eventually resulting in disrupter of calcified plaques during aortic cross-clamping, macroembolisms, in the aortic wall, is a significant predictor for neurological problems, but extremely for postoperative stroke. In patients with aorta-pathology, we noted no isolated mild neurological complications. In the case of a severe stroke, mild complications may be masked by the severity of the major complication. Therefore we must take care with the interpretation of `zero' patients in the group of mild neurological complications and we can only conclude that if patients have neurological complications due to aorta-pathology, this results in severe neurological complications. Therefore, but also because the power to detect a significant relationship between an independent risk factor and a neurological complication (A, B or C) decreases if the number of persons with this risk factor is low. Only 32 patients (0.8%) had aorta-pathology risk, no mild complications were noted; because of this we excluded aorta-pathology as a possible risk factor for mild neurological problems in the logistic regression analysis.
Four independent risk factors are found for mild and/or severe complications (Table 4A): Peripheral vascular atherosclerosis, neurological pathology, aorta-pathology and perioperative myocardial infarction. Further analysis identified aorta-pathology, perioperative myocardial infarction and neurological pathology as independent predictors for sever neurological complications, what is again supporting the `macroembolism'-theory of Harrison [6]. Peripheral vascular atherosclerosis matures as predictor for either severe or mild neurological complications. Only perioperative myocardial infarction was an independent predictor for as well severe and mild neurological complications. Perioperative myocardial infarction can result in a low cardiac output syndrome, eventually a period of hypotension, inducing cerebral ischemia and mild neurological problems. On the other site, the occurrence of a perioperative myocardial infarction may also be the result of the severity of the atherosclerosis, in combination with the difficulties of the myocardial revascularization. And thus being an exponent of severe atherosclerosis, identified by the unifactor analysis as a predictor of severe neurological complications.
The changes in prevalence of pre-operative neurological pathology (Table 5) is reason for concern. The prevalence in the 1993–1995 cohort is 11.6%. this percentage is substantially larger than the prevalence observed in the time cohorts 1987–1989, 1990–1992. Furthermore the cohort 1993–1995 appears to be an independent predictor for increased risk for neurological complications (Table 4). Therefore, factors not included in this risk analysis, related to the time cohort, influences the risk for neurological complications. Possibly the tendency to accept lower perfusion pressure (50–60 mmHg) during cardiopulmonary bypass, can be a reason for the increased risk of neurological morbidity, as described by Gold et al. [12]. An other explanation, is related to the definition and registration of `mild neurological' complications. Mild complications are described by confusion or memory disturbances for more than 12 h, without formal neurological or neuro-psychological examination. This is of course a serious deficit of our study, however, inevitable in a retrospective analysis. Therefore it is possible that mild neurological problems are better registered during this period. Also, during this last period, patients are discharged faster from intensive care to intermediate care or the ward. Soft neurological complications may be more registered, because on the intensive care these mild complications are more `accepted' or `hidden' in the whole of intensive care problems. In the unifactor analysis, neurological pathology was a predictor of mild neurological problems. So it is good possible that the higher risk for especially mild neurological complications during the last time cohort is a consequence of the increasing number of patients with preoperative neurological pathology, who are more at risk for mild neurological complications in combination with our fast discharge policy. This is important, because on the ward, these patients need more care, what can be translated into more nursing costs. Furthermore, these patients are more likely to aspirate, to require reintubation because of hypoxia due to their confusion or depressed consciousness. [5]. Concerning these two remarks, a logical question is, are all patients benefited with the maintenance of low perfusion pressure during bypass and with the fast-track policy? As it is known that patients who develop even minor postoperative complications have excess costs in comparison to patients who have no complications [12][13].
The relation between elective/emergency operation and neurological problems is especially determined by the occurrence of a perioperative myocardial infarction and also, however lesser, by the presence of aorta-pathology. We known that patients operated in emergency, but in which we can prevent or limiting a myocardial infarction, are mostly not causing any problem. It is the occurrence of a perioperative myocardial infarction that is responsible for the frequency of neurological problems after emergency operations. Hypertension, resulting in atherosclerotic degeneration of the vessels, lead to more neurological problems possible because the high frequency of peripheral vascular atherosclerosis and the higher incidence of perioperative myocardial infarction due to distal coronary pathology.
In this study, we analyzed neuro-and psychological disorders after coronary bypass surgery, and we accept the criticism of methodological limitations, as the definition of the `mild neurological complications'. However, the importance of such as studies is that they accentuate the problem, and serve as a source of ideas for design of prospective controlled studies [14].
In conclusion, the present study emphasizes that peripheral vascular atherosclerosis, preoperative neurological pathology, aorta-pathology and the occurrence of a perioperative myocardial infarction, are predictors for neurological postoperative complications. In addition we suggest that different strategies in patient care for patients with or without these risk factors can be useful in reduction of postoperative (mild) neurological complications and eventually cost reduction.
|
References
|
|---|
- Lynn G.N., Stefanko K., Reed J.F., Gee W., Nicholas G. Risk factors for stroke after coronary artery bypass. J Thorac Cardiovasc Surg 1992;104:1518-1523.[Abstract]
- Newman S., Smith P., Treasure T., Joseph P., Ell P., Harrison M. Acute neuropsychological consequences of coronary artery bypass grafting. Curr Psychol Res Rev 1887;6:115-124.
- Reed G.L., Singer D., Picard E. Stroke following coronary artery bypass grafting. N Engl J Med 1988;319:1246-1250.[Abstract]
- Noyez L., Janssen D.P.B., van Druten J.A.M., Skotnicki S.H., Lacquet L.K. Coronary bypass surgery: what is changing? Analysis of 3834 patients undergoing primary isolated myocardial revascularization. Eur J Cardio-thoracic Surg 1998;13:365-369.[Abstract/Free Full Text]
- Redmond J.M., Greene P.S., Goldsborough M.A., Cameron D.E., Stuart R.S., Sussman M.S., Watkins L., Lashinger J.C., McKhann G.M., Johnston M.V., Baumgartner W.A. Neurological injury in cardiac surgical patients with a history of stroke. Ann Thorac Surg 1996;61:42-47.[Abstract/Free Full Text]
- Harrison M.J.G. Neurological complications of coronary artery bypass grafting: diffuse or focal ischemia?. Ann Thorac Surg 1995;59:1356-1358.[Abstract/Free Full Text]
- Christakis G.T., Fremes S.E., Naylor C.D., Chen E., Roa V., Goldman B.S. Impact of preoperative risk and perioperative morbidity on ICU stay following coronary bypass surgery. Cardiovasc Surg 1996;4:29-35.[Medline]
- Taggart D.T., Mazel J.W., Bhattacharya K., Meston N., Standing S.J., Kay J.D.S., Pillai R., Johnsson P., Westaby S. Comparison of serum S-100ß levels during CABG and intracardiac operations. Ann Thorac Surg 1997;63:492-496.[Abstract/Free Full Text]
- Westaby S., Johnsson P., Parry A. Serum S100 protein: a potential marker for cerebral events during cardiopulmonary bypass. Ann Thorac Surg 1996;61:88-92.[Abstract/Free Full Text]
- Tuman K.J., McCarthy R.J., Najafi H., Ivankovich A.D. Differential effects of advanced age on neurological and cardiac risks of coronary artery operations. J Thorac Cardiovasc Surg 1992;104:1510-1517.[Abstract]
- Heyer E.J., Delphin E., Adams D.C. Cerebral dysfunction after cardiac operations in elderly patients. Ann Thorac Surg 1995;60:1716-1722.[Abstract/Free Full Text]
- Gold J.P., Charlson M.E., Williams-Russo P., Szatrowski T.P., Peterson J.C., Pirraglia P.A., Hartmann G.S., Yao F.S.F., Hollenberg J.P., Barbut D., Hayes J.G., Thomas S.J., Purcell M.H., Mattis S., Gorkin L., Post M., Krieger K.H., Isom O.W. Improvement of outcomes after coronary bypass. A randomized trial comparing intraoperative high versus low mean arterial pressure. J Thorac Cardiovasc Surg 1995;110:1302-1314.[Abstract/Free Full Text]
- Magovern J.A., Sakert T., Magovern G.J., Jr., Benckart D.H., Burkhoder J.A., Liebler G.A., Magovern G.J., Sr. A model that predicts morbidity and mortality after coronary artery bypass graft surgery. J Am Coll Cardiol 1996;28:1147-1153.[Abstract]
- Engels E.A., Spitz M.R. PACE – setting research. Lancet 1997;350:677-678.[Medline]
This article has been cited by other articles:
|
|
|
|
 
F. M. van Eck, L. Noyez, F. W.A. Verheugt, and R. M.H.J. Brouwer
Changing profile of patients undergoing redo-coronary artery surgery
Eur. J. Cardiothorac. Surg.,
February 1, 2002;
21(2):
205 - 211.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
