Cerebral complications after coronary artery bypass and heart valve surgery: Risk factors and onset of symptoms

Cerebral Complications After Coronary Artery Bypass and Heart Valve Surgery: Risk Factors and Onset of Symptoms Ewa A h l g r e n , MD and Claes Ar~n, MD, PhD

Objectives: Cerebral complications continue to be a major cause of morbidity after cardiac surgery. Earlier studies have mainly focused on intraoperative events, but symptoms may also occur later in the postoperative period. The purpose of this study was to determine the incidence and risk factors of focal neurologic complications and timing of cerebral symptoms. Design: A retrospective study. Setting: Link6ping University Hospital. Participants: Two thousand four hundred eighty patients who underwent cardiac surgery from 1992 to 1995. Interventions: Standard cardiopulmonary bypass (CPB) technique was used in all patients. Anticoagulant treatment included heparin and patients with coronary artery surgery were also administered acetylsalicylic acid and valvesurgery patients received warfarin or dicumarol. Measurements and Main Results: Seventy-five patients (3%) had focal neurologic deficits and/or confusion postoperatively. In 32 patients (43%), the onset was not intraoperative but occurred later in the postoperative period. The lowest incidence of cerebral complications was found in patients who underwent single-valve replacement (1.2%)

and the highest incidence was found in patients who underwent combined procedures (valve and coronary artery surgery; 7.6%). Patients greater than 70 years of age had a complication rate of 4.1% compared w i t h 2.5% in patients aged 70 years and less (p < 0.05). The incidence of diabetes mellitus was 11.4% in the entire series, but was more common (18.7%; p < 0.05) in patients w i t h cerebral symptoms. Also, 5.9% of all patients had a history of cerebrovascular disease compared with 14.7% (p < 0.01) of patients with cerebral complications. Conclusion: Cerebral complications may be delayed after cardiac surgery, suggesting causes of cerebral damage other than intraoperative events. Valve-surgery patients had the lowest incidence and patients with combined procedures had the highest incidence of cerebral complications. Advanced age, diabetes mellitus, and preexisting cerebrovascular disease increased the risk. Copyright© 1998by W.B. Saunders Company

HE INCIDENCE OF cerebral complications after cardiac surgery varies widely, according to the type of surgery and differences in patient selection. The rate for focal neurologic deficits varies between 0.7% and 3.8% in retrospective studies, but is generally higher (4.8% to 5.2%) in prospective studies. 1,2 Preoperative factors reported to increase the incidence of cerebral dysfunction after cardiac surgery included diabetes mellitus, advanced age, and a history of cerebrovascular disease. 3-5 Historically, intracardiac operations such as valve replacements have typically carried a higher risk of adverse neurologic outcome compared with coronary artery bypass grafting (CABG). 6-8 This was likely because of the increased risk of macroembolization of air or particulate matter from the surgical field during open-chamber surgery.6,9 However, the risk is increasing in patients who undergo CABG because changes have occurred in the selection of patients. Refinements in medical management and the introduction of percutaneous transluminal coronary angioplasty have resulted in a tendency to operate on older patients. These patients have more extensive coronary and peripheral vascular atherosclerosis, including cerebral vessels and the ascending aorta, and require more grafts and longer cardiopulmonary bypass (CPB) time. 3,6,1°-~3 These are all considered risk factors for cerebral complications.

A group of cardiac patients who may be at particularly high risk for postoperative cerebral dysfunction are those undergoing an intracardiac procedure combined with a CABG procedure. 7,14,15 Previous reports on cerebral complications have mainly focused on intraoperative events in which two mechanisms are considered responsible for adverse neurologic outcome. These are the embolization of gaseous and solid forms originating from the surgical field, the ascending aorta, and the CPB circuit and reduced cerebral perfusion as a result of reduced cerebral blood flow. 16 Data on the incidence and etiology of such complications occurring in the immediate postoperative period are sparse compared with data on intraoperative events. The purpose of the current retrospective study was to determine the incidence of focal neurologic complications and risk factors and record the time span from surgery to the first observation of neurologic symptoms.

T

From the Departments of Cardiothoracic Anaesthesia and Intensive Care and Cardiothoracie Surgery, University Hospital, LinkOping, Sweden. Presented at the Eleventh Annual Meeting of the European Association of Cardiothoracic Anesthesiologists, held in Leuven, Belgium, May 22-25, 1996. Address reprint requests to Ewa Ahlgren, MD, Department of Cardiothoracic Anaesthesia and Intensive Care, University Hospital, LinkOping, Sweden. Copyright © 1998 by W.B. Saunders Company 1053-0770/98/1203-000558.00/0 270

KEY WORDS: cardiac surgery, cerebral complications, neurologic complications, cerebral damage

METHODS

At the LinkOping Heart Center, Link6ping, Sweden, 2,480 patients underwent cardiac surgery from March 1992 to February 1995. The mean age was 64.6 years (range, 18 to 87 years). The male-female ratio was 1,825:655 (73.6%:26.4%). The distribution of procedures was 1,744 (70.3%) CABGs, 400 (16.1%) single-valve replacements, 198 (8%) combined procedures with both valves and CABGs, and 138 (5.6%) miscellaneous operations, including atrium septum defect, ventricular septum defect, myxoma, and graft repair of the ascending aorta. The patients were premedicated with oxicone, 4 to 10 mg, and scopolamine, 0.2 to 0.5 mg, by intramuscular injection. Anesthesia was induced with intravenous fentanyl, 10 to 20 ~g/kg, and thiopental, 100 to 250 mg, or midazolam, 2 to 10 rag, and muscle paralysis was achieved with pancuronium or vercuronium, 0.1 mg/kg. Isoflurane, 1% to 1.5%, was added for maintenance of anesthesia, and additional doses of fentanyl were administered on the decision of the anesthesiologist.

Journal of Cardiothoracic and VascularAnesthesia, Vo112, No 3 (June), 1998: pp 270-273

CEREBRAL COMPLICATIONS AND CARDIAC SURGERY

271

Standard CPB technique was used in all patients with moderate hypothermia (28 ° to 32°C). Membrane oxygenators and roller pumps were used. The CPB flow was approximately 2.2 L/min/m2 during normothermia, and the mean arterial blood pressure was maintained at greater than 50 mmHg during CPB. The CABG patients were administered acetylsalicylic acid (ASA), 500 rag, rectally 8 hours after surgery. On the following day, oral therapy with ASA, 160 mg dally, was initiated. Patients with intolerance to ASA were administered dipyridamole, 75 mg three times a day. In addition, heparin, 5,000 IU, was administered subcutaneously three times a day until the patients were fully mobilized. Patients who had undergone valve replacement received heparin, 7,500 IU, subcutaneously 3 times a day starting 12 hours postoperatively. Warfarin or dicumarol treatment was started on the first postoperative day. Heparin was discontinued when the prothrombin value reached therapeutic level. Anticoagulation was modified in selected cases mainly because of prolonged postoperative bleeding. Cerebral complications were defined as focal neurologic deficits and/or confusion not related to anesthesia or postoperative analgesia. The patients' charts were reviewed after discharge by a research nurse, supervised by a surgeon. Records of patients with postoperative cerebral complications were again reviewed by the authors. Risk factors analyzed were diabetes mellitus, preexisting cerebrovascular disease, and age greater than 70 years. Time from surgery to the time of presentation of cerebral symptoms was recorded. The chi-square method was used for statistical analyses. RESULTS

The 30-day mortality rate in the total series of 2,480 patients was 2.1% and 1.5% in the patients who underwent CABG and the patients who underwent single-valve surgery, respectively. The mortality rate was 5.5% in patients who underwent combined procedures and 6.5% in the miscellaneous group. Seventy-five patients (3%), 24 women and 51 men, had cerebral complications after surgery. The mean age of this group was 67.3 years (range, 54 to 81 years) compared with 64.6 years (range, 16 to 87 years) in the entire series (p < 0.05). The lowest incidence was found in the patients who underwent single-valve surgery, and the highest incidences were in the groups of patients with miscellaneous surgery and combined procedures, respectively (Table 1). Ten of the 75 patients (13.3%) died within 30 days. Four of these underwent surgery for acute dissection of the ascending aorta. In the total series, patients greater than 70 years of age had a cerebral complication rate of 4.1% compared with 2.5% in patients of 70 years or less (p < 0.05); in the CABG group, 3.7% compared with 2.0% in corresponding age groups (p < 0.05). Table 2 shows that

Table 2. Risk Factors and Cerebral Complications Total Series PatientsWith Cerebral (n - 2,480) Complications (n = 75)

Age >70 years Diabetes Previous cerebrovascular disease

No.

%

No.

%

p

851 283

34 11.4

34 14

45 18.7

<0.05 <0.05

146

5.9

11

14.7

<0.01

diabetes mellitus was more common in patients with cerebral symptoms (p < 0.05), as was a history of cerebrovascular disease (p < 0.01). Sixty-two of 75 patients with cerebral dysfunction had focal neurologic deficits and among these, 23 patients had hemiparesis. Four patients had confusion as a single symptom and another four patients had both confusion and focal neurologic deficits (Table 3). Figure 1 shows the time from surgery to the first appearance of neurologic symptoms. In 27 patients, the condition was detected when the patients regained consciousness from anesthesia and the damage was considered to have occurred intraoperatively. In 32 patients, the onset was not intraoperative but occurred later, and in 16 patients, it was difficult to decide the time of onset of symptoms. Table 4 lists the type of operation and tinting of cerebral symptoms. Postoperative cerebral computed tomography (CT) scans were performed in 48 of 75 patients with cerebral dysfunction. Recent brain infarctions were found in 40 patients, white matter changes suggesting chronic ischemia in two patients, and five patients had normal CT scans. All patients with normal CT scans had symptoms compatible with a focal neurologic deficit. No bleeding was found except for one subdural hematoma in an 80-year-old man who underwent uncomplicated surgery with CABG and aortic valve replacement. Immediately after surgery, slight hemiparesis was noticed and neurosurgery was required. He recovered and, at the time of discharge, he could manage with some personal assistance. DISCUSSION

The cause of cerebral complications after cardiac surgery is often difficult to determine because the etiology is multifactorial and complex. The most interest has focused on intraoperative factors, such as macroembolizations (atheromatous debris, air bubbles, fat) from the heart or great vessels, and microemboli from the extracorporeal circulation circuit, 6,9,17 as well as hypoperfusion of the brain during surgery and especially during CPB. Little, however, is reported on the incidence of cerebral complications in the immediate postoperative period compared with intraoperative events, although recent studies 8,18,19 have focused on the development of a deficit later in the hospital

Table 1. Distribution of Cerebral Complications Total

CABG Valves Combined procedures (valves and CABG) Miscellaneous procedures Total

Cerebral Complications

No.

No.

%

1,744 400 198 138 2,480

44 5 15 11 75

2.5 1.2 7.6 7.8 3.0

Table 3. Cerebral Symptoms Symptoms

No. of Patients

Visual defects Confusion Hemiparesis Focal deficit other than hemiparesis Coma Total

8 8 23 31 11 81

272

AHLGREN AND ARIAN

3O 26 24

I[]Valve l~ M ~ I I ~

V ~_

m-

ImCAla

1--

18"161412lOg642O-

--i

0

1

2

3

4

5

6

7

8 days

Fig 1. Onset of symptoms.

course. In the series of the authors, 43% of the patients with cerebral complications had a free interval between surgery and the onset of cerebral symptoms. This free interval was more common in patients who underwent CABG; the valvereplacement patients had a comparatively earlier onset of symptoms, but in 16 patients, it was difficult to determine the first appearance of the neurologic symptoms. These 16 patients had a prolonged stay in the intensive care unit and it was unknown if the cerebral damage was actually intraoperative and the deficits were only detected later (because of confounding medical conditions). In the six other patients, data were not conclusive whether the onset of symptoms was delayed. Although there are limitations because the study is retrospecfive, these data suggest that events other than intraoperative may be the cause of cerebral complications after cardiac surgery. Ricotta et al8 reported that the majority of the patients with cerebral complications after CABG developed the neurologic deficits later in the immediate postoperative period, and therefore suggested that these are more important than the intraoperative events. Possible mechanisms not analyzed in the present study may be supraventricular tachyarrhythmias with embolizations, hypoperfusion, and the state of hypercoagulability after surgery. In the present series, 3% of patients had cerebral dysfunction after cardiac surgery, and this incidence is similar to that reported by other centers. 1,2The lowest incidence was found in the single-valve replacement group, which conflicts with the previous and most recent data reporting a higher risk for cerebral dysfunction in intracardiac operations compared with extracardiac procedures. 6-8 In a recent study, Kuroda et al2° reported a higher incidence of cerebral complications in patients who underwent isolated CABG surgery compared with those who underwent isolated cardiac-valve procedures. A difference in preexisting comorbidity between the patients undergoing Table 4. Onset of Cerebral Symptoms Surgery Valve CAB G Combined Miscellaneous Total

Intraoperative D e l a y e d 3 10 6 8 27

1 24 6 1 32

Unknown

Total

1 10 3 2 16

5 44 15 11 75

extracardiac and intracardiac surgery may partly explain the results in that study. Similarly, Cernaianu et al, 19in a retrospective review of 2,455 patients, found the lowest rate of cerebral complications in patients who underwent aortic valve replacement (0.9%). However, the risk of cerebral complications in CABG surgery is increasing because of the type of patients now operated on. The tendency to operate on older and sicker patients has not affected the outcome from a cardiovascular standpoint; in fact, these patients often do very well postoperatively, but neurologic morbidity has increased. 21 In the authors' series, the highest incidence of cerebral dysfunction was found in patients who underwent combined procedures (7.6%) and in the miscellaneous group (7.8%), and these patients also had the highest mortality rate. The high incidence of cerebral complications in patients with combined procedures agrees with the results from other recent reports. 7 In a multi-institutionalstudy by Wolman et al,lS it was noted that the incidence of overt adverse neurologic outcome was approximately doubled in patients undergoing CABG that included an intracardiac procedure compared with CABG alone. Inclusion of an intracardiac procedure was an independent risk factor. 15 Such patients have all the risks inherent in both CABG and open-chamber surgery. In the authors' study, diabetes mellitus, age greater than 70 years, and a previous history of cerebrovascular disease increased the risk of cerebral complications. This agrees with the findings of others. 3-6 In the present study, CT of the brain in patients with postoperative cerebral dysfunction revealed pathologic findings in 43 of 48 patients, and brain infarctions and white matter changes were common. No bleeding was found except for one subdural hematoma. This agrees with an earlier study in this center of 1,400 patients who had undergone cardiac surgery and suggests that the dominating etiology of cerebral dysfunction after cardiac surgery is infarction.22 These findings suggest that there is no need for CT scans to differentiate between infarction and bleeding for decision-making on postoperative anticoagulant treatment. There are several limitations of the present study. As with all retrospective studies, the frequency of cerebral complications may have been underestimated if inadequately documented. Risk factors and other variables may not have been systematically recorded, resulting in incomplete data. This problem would be mitigated somewhat by the fact that all discharge summaries at the authors' departments are carefully reviewed by a research nurse, and all intraoperative and postoperative complications are recorded and classified according to a standard procedure. The incidence of adverse neurologic outcome may be further underestimated by not examining neuropsychologic outcome. The follow-up was not long term and it is possible that some of these adverse outcomes may have been temporary and may have resolved over a short period of time. In an earlier study at this center, one third of the patients with cerebral complications after cardiac surgery reported complete recovery from the neurologic deficits in a telephone interview 10 to 43 months after the operation. 22 It is apparent that concomitant neurologic morbidity associated with cardiac surgery has increased, particularly in CABG patients and in patients with combined procedures. Possible mechanisms of neurologic damage and strategies for cerebral protection have been discussed in several earlier studies.

CEREBRAL COMPLICATIONS AND CARDIAC SURGERY

273

Atheromatous plaques in the ascending aorta are a strong and independent risk factor for postoperative cerebral dysfunction, which may be reduced by intraoperative epiaortic echocardiographic scanning and modification of surgical technique. 6,H Mean arterial perfusion pressure greater than 80 mmI-Ig during CPB has been proven to reduce cerebral complications in patients at risk. 23 It is also important to provide the brain with sufficient blood flow during the rewarming phase of CPB to match oxygen delivery to the increased metabolic demands during this period, to avoid cerebral ischemia. Hypercoagulability in the postoperative period is a reason for aggressive anticoagulant therapy, which of course must be weighed against the risk of bleeding from the surgical field. Is Postoperative supraventricular tachyarrhythmias have been reported by others as possible causes of cerebral embolization and cerebral dysfunc-

tion. 6 Aggressive antiarrhythmic treatment during the early postoperative period may reduce the risk of such emboli formation. In conclusion, intraoperative strategies, including careful manipulation of the ascending aorta and the perfusion technique during CPB, are undoubtedly crucial f o r neurologic outcome after cardiac surgery. The results of the present study suggest that the immediate postoperative period is also of importance and management of the patients during this phase needs to be further evaluated. Patients undergoing combined procedures were at particularly high risk for cerebral complications, especially those with advanced age and concomitant cerebrovascular disease, whereas patients undergoing single-vaNe replacement surgery had the lowest incidence of cerebral complications.

REFERENCES 1. Lynn GM, Stefanko K, Reed JF: Risk factors for stroke after coronary artery bypass. J Thorac Cardiovasc Surg 104:1518-1523, 1992 2. Mills SA: Cerebral injury and cardiac operations. Ann Thorac Surg 56:86-91, 1993 3. Gardner TJ, Horneffer PJ, Manolio TA: Stroke following coronary artery bypass grafting: A ten-year study. Ann Thorac Surg 40:574-581, 1985 4. Remond JM, Greene PS, Goldsborough MA, et al: Neurologic injury in cardiac surgical patients with a history of stroke. Ann Thorac Surg 61:42-47, 1996 5. Wareing TH, Davila-Roman VG, Daily BB, et al: Strategy for the reduction of stroke incidence in cardiac surgical patients. Ann Thorac Surg 55:1400-1408, 1993 6. Mills SA: Risk factors for cerebral injury and cardiac surgery. Ann Thorac Surg 59:1296-1299, 1995 7. Nussmeier NA: Adverse neurologic events: Risks of intracardiac versus extracardiac surgery. J Cardiothorac Vasc Anesth 10:31-37, 1996 8. Ricotta JJ, Faggioli GL, Castilone A: Risk factors for stroke after cardiac surgery. J Vasc Surg 21:359-364, 1995 9. Doblar DD: Cerebrovascular assessment of the high-risk patient: The -ole of transcranial Doppler ultrasound. J Cardiothorac Vasc Anesth 10:4-14, 1996 10. Amarenco R Duyckaerts C, Tzouria C: The prevalence of ulcerated plaques in the aortic arch in patients with stroke. N Engl J Med 326:221-225, 1992 11. Barbut D, Gold JP: Aortic atheromatosis and risks of cerebral embolization. J Cardiothorac Vasc Anesth 10:24-30, 1996 12. Loop FD, Lytle BV, Cosgrove DM: Coronary artery bypass surgery in the elderly. Cleve Clin J Med 55:23-24, 1988 13. Tunick PA, Rosenzweig BR Katz ES: High risk for vascular

events in patients with protuding aortic atheromas: A prospective study. JAm Coll Cardio123:1085-1090, 1994 14. Wolman RL, Aggarwal A, Kanchuger M: Risk factors for adverse neurologic outcome following intracardiac surgery. Anesth Analg 80:SCA58, 1995 15. Wolman RL, Kanchuger MS, Newman MF: Adverse neurologic outcome following cardiac surgery. Anesth Analg 78:484, 1994 16. Hornic R Smith PL, Taylor KM: Cerebral complications after coronary bypass grafts. CUlTOpin Cardiol 9:670-679, 1994 17. Blauth CI: Macroemboli and microemboli during cardiopulmonary bypass. Ann Thorac Surg 59:1300-1303, 1995 18. Libman RB, Wirkowski E, Neystat M, et al: Stroke associated with cardiac surgery. Arch Ne~arol 54:83-87, 1997 19. Cernaianu AC, Vassilidze YV, Flum DR, et al: Predictors of stroke after cardiac surgery. J Card Surg 10:334-339, 1995 20. Kuroda Y, Uchimoto R, Kaieda R: Central nervous system complications after cardiac surgery: A comparison between coronary artery bypass grafting and valve surgery. Anesth Analg 76:222-227, 1993 21. Tuman KJ, McCarthy RJ, Najafi H, et al: Differential effects of advanced age on neurologic and cardiac risks of coronary artery operations. J Thorac Cardiovasc Surg 104:1510-1517, 1992 22. Johansson T, Aren C, Fransson SG, et al: Intra- and postoperative cerebral complications of open-heart surgery. Scand J Thor Cardiovasc Surg 29:17-22, 1995 23. Gold JR Charlson ME, Williams-Russo P, et al: Improvement of outcomes after coronary artery bypass: A randomized trial comparing intraoperative high versus low mean arterial pressure. J Thorac Cardiovasc Surg 110:1302-1304, 1995