Neurocrit Care DOI 10.1007/s12028-014-9967-x

ORIGINAL ARTICLE

A Prospective Observational Study of Seizures After Cardiac Surgery Using Continuous EEG Monitoring Teneille E. Gofton • Michael W. A. Chu • Loretta Norton • Stephanie A. Fox Lindsay Chase • John M. Murkin • G. Bryan Young



Ó Springer Science+Business Media New York 2014

Abstract Introduction Recently, there have been several retrospective reports suggesting an increased frequency in seizures after cardiopulmonary bypass, associated with increased patient morbidity. We sought to prospectively investigate the incidence of electrographic seizures without clear convulsive clinical correlates and subsequent neurologic injury following cardiac surgery. Methods This single-center, prospective, observational study used continuous subhairline electroencephalographic (cEEG) monitoring in the intensive care unit following routine cardiac surgery, ranging from coronary bypass surgery to complex aortic arch reconstruction. The primary outcome was the proportion of patients developing postoperative seizures, as confirmed on cEEG monitoring. Secondary outcomes included neurologic injury, postoperative complications, mortality, and ICU and hospital lengths of stay. Results 101 consenting patients were included and 3 patients had seizures (2 focal and convulsive, 1 generalized and electrographic). All three patients with seizures were

T. E. Gofton (&)  L. Norton  G. B. Young Department of Clinical Neurological Sciences, London Health Sciences Centre, University of Western Ontario, 339 Windermere Road, London, ON N6A 5A5, Canada e-mail: [email protected] M. W. A. Chu  S. A. Fox  L. Chase Division of Cardiac Surgery, Department of Surgery, London Health Sciences Centre, University of Western Ontario, 339 Windermere Road, London, ON N6A 5A5, Canada J. M. Murkin Department of Anaesthesia and Perioperative Medicine, London Health Sciences Centre, University of Western Ontario, 339 Windermere Road, London, ON N6A 5A5, Canada

C65 years old, had ‘‘open-chamber’’ procedures, and had cardiopulmonary bypass times >120 min. One of the 3 patients with seizures was exposed to higher doses of tranexamic acid. None of the patients with seizures had permanent neurologic sequelae and all were doing well at 1-year follow-up. There was no increased morbidity or mortality in patients with seizures. Conclusions Electrographic seizures occur infrequently after cardiac surgery and are generally associated with a good prognosis. Prophylactic cEEG monitoring is unlikely to be cost-effective in this population. (ClinicalTrials.gov Identifier: NCT01291992). Keywords Seizure  Tranexamic acid  Cefazolin  Cardiopulmonary bypass  Electroencephalography

Introduction Seizures after cardiac surgery are usually diagnosed clinically and are thought to be frequently associated with neurologic injury [1]. The etiology is multifactorial including cardioembolic strokes, air emboli, atherosclerotic debris, release of inflammatory mediators during surgery, and exposure to potentially epileptogenic medications such as the antifibrinolytics [2] and cephalosporin antibiotics [1, 3, 4]. Recently, there have been reports of increased frequency of seizures after cardiac surgery and retrospective evidence suggests that exposure to high-dose tranexamic acid (>100 mg/kg) may be an independent risk factor for seizures post-cardiac surgery [5–10]. Other risk factors include pre-operative renal dysfunction, open-chamber cardiac surgery, advanced age, prolonged cardiopulmonary bypass time, prior cardiac surgery, peripheral vascular disease, and pre-operative cardiac arrest [7, 8, 11, 12].

123

Neurocrit Care

Seizures are also known to occur in a wide variety of critically ill patients. In general ICU patients the prevalence of seizures is 8–11 % [13, 14], up to 34 % in neuroICU patients [15], and in patients at higher risk for seizures the risk of seizures is as high as 48 % (patients failing to recover consciousness after a convulsive seizure) [16]. A very high proportion of seizures in critically ill patients is non-convulsive in nature and can only be detected using continuous video-electroencephalographic (cEEG) monitoring. Non-convulsive seizures can be detrimental to cerebral function and may cause injury [17]. Non-convulsive seizures may also be associated with altered cognition and altered cognition occurs commonly after cardiac surgery. While it has been described that clinically evident seizures occur in 1–4 % of patients after cardiac surgery [11], it is unknown whether non-convulsive seizures also occur at a higher frequency in this population. It has also been suggested that post-cardiac surgery seizures are associated with longer mechanical ventilation, longer length of stay in hospital, and overall increased morbidity and mortality [18]. The following study is a prospective observational study designed to investigate the incidence of convulsive and electrographic (no clinically identified convulsive event) seizures and subsequent neurologic injury following cardiac surgery using cEEG monitoring.

Patient and Methods The study protocol was approved by the Health Sciences Research Ethics Board at Western University prior to patient recruitment. All study patients provided informed written consent prior to inclusion in the study. The study was registered with ClinicalTrials.gov-Identifier NCT01291992. Estimated Sample Size Since our study was preliminary, exploratory, prospective, and original, it was difficult to make a power calculation. In general ICUs, over 90 % of seizures are non-convulsive [19] and with EEG monitoring, the overall incidence of seizures is about 28 % [20]. With 4-channel subhairline EEG monitoring the sensitivity for detecting seizures is 68 %. With 126 patients enrolled we estimated a yield of 126 9 28 9 68 % = 24 patients with seizures. We considered this to be sufficient for our hypothesis generating, exploratory study. Inclusion and Exclusion Criteria (Fig. 1) Eligible patients were over 18 years of age, able to provide informed consent pre-operatively, underwent all types of

123

Fig. 1 Flow diagram detailing eligible patients in the study

cardiac surgery (elective, urgent, emergent, or salvage), and were planned to be admitted to the Cardiac Surgery Recovery Unit (CSRU) post-operatively. We elected to include a broad range of cardiac procedures to provide the best ‘‘panoramic’’ view of seizure activity after cardiac surgery. Exclusion criteria included patients unable to provide informed consent, patients with a history of epilepsy, or patients not planning admission to the CSRU post-operatively. Patients for whom inadequate duration of cEEG recordings were acquired were excluded from the analysis (n = 23). Study Procedure This investigation was a single-center prospective observational study performed at University Hospital, London Health Sciences Centre (LHSC) at Western University in London, Ontario, Canada between September 2010 and October 2011. Consecutive consenting patients were included in the analysis. All patients underwent routine cardiac surgery including coronary artery bypass; valve repair or replacement (minimally invasive and sternotomy); atrial septal defect repair; and aortic root, ascending aorta, and aortic arch reconstructive surgery. Patients received the usual standard of care during their surgery and post-operative hospitalization. Routine cardiopulmonary bypass techniques were utilized including roller or centrifugal pumps, return of cardiotomy suction, and mild hypothermia (35°C). Central cannulation was employed for all sternotomy-based cases, with the exception of axillary cannulation (8 mm Dacron side graft) for aortic arch repair.

Neurocrit Care

Peripheral cannulation with femoral or axillary approaches was employed for minimally invasive operations. The operative field was flooded with CO2 during open-chamber operations. The use of antifibrinolytic agents (tranexamic acid), antibiotics (usually cefazolin, except in the case of allergy), and anesthetic agents was at the discretion of the treating anesthetist and surgeon during the surgery. Upon completion of surgery, all patients were cared for in the CSRU. Immediately following admission to the CSRU, a subhairline array of EEG electrodes was placed on all patients for 24 h of prospective continuous monitoring of the cEEG [21]. Briefly, the cEEG setup consists of a 9 sticker electrode system that is applied to the frontal and temporal areas resulting in 4 electrode derivations (General Electric EEG System) [21]. The cEEG is displayed at the bedside, remotely on a monitor, and is stored for further analysis and review. A detailed description of the cEEG used in this study has been previously published [21] and is known to be 68 % sensitive and 98 % specific for seizure detection. Continuous EEGs were reviewed by a qualified electroencephalographer (GBY or TEG) following acquisition of data. EEGs were classified based on previously described methods [22]. The presence or absence of seizures and whether onset was focal or generalized were noted. Any patient identified as having a seizure received a formal neurologic evaluation and an appropriate workup was undertaken based on the clinical situation. The total dose of tranexamic acid was calculated as a combination of the initial loading dose and the infusion dose divided by each patient’s weight in kg (total dose in mg/kg). The total peri-operative dose of cefazolin was calculated as the total mg/kg including the dose at the beginning of the surgery and at regular intervals thereafter. Data to reflect renal function included measurement of serum creatinine (preoperative) and calculation of the estimated glomerular filtration rate (eGFR) using the MDRD (long form) equation representing eGFR in ml/min/ 1.73 m2 [23]. Statistical Analysis All demographic and perioperative clinical data were collected prospectively. All categorical variables are presented as proportions and continuous variables are presented as mean ± standard deviation.

Results One hundred and twenty-six patients were screened for the study and 25 were excluded because they did not require admission to CSRU (1), they did not undergo surgery (1),

Table 1 Demographic participants

and

Characteristic Age Gender

surgical Average (stdev)

information

65.4 (10.6) – –

73 male (72.3), 29 female (28.7)

84.0 (16.3) –

BMI

29.2 (5.2)

Creatinine

94.8 (69.7) –

eGFR

study

Number of patients (%)

Weight (kg)

Peak postoperative creatinine

for



101.0 (60.4) – 82.1 (28.7) –

Pre-operative comorbidities Stroke/transient ischemic attack



8 (7.9)

Atrial fibrillation Peripheral vascular disease

– –

17 (16.8) 22 (21.8)

Recent myocardial infarction



11 (10.9)

Congestive heart failure



21 (20.8)

Body mass index > 30



34 (33.7)

Chronic obstructive pulmonary disease



22 (21.8)

Diabetes mellitus



27 (26.7)

Creatinine > 120



9 (8.9)

Re-operation



6 (5.9)

CABG



52 (51.5)

Valvular CABG and valvular

– –

35 (34.7) 4 (4.0)

Procedure type

ASD repair



2 (1.9)

Aortic



2 (1.9)

Aortic and valvular



4 (4.0)

Elective



73 (72.3)

Urgent



27 (26.7)

Emergency



1 (0.9)

Salvage



1 (0.9)

Urgency of procedure

Cross clamp time, mean (min)

77.7 (47.1) –

Cardiopulmonary bypass time, 120.0 (64.6) – mean (min) Days in CSRU, mean/median 1.8/1 (1.7) – Days in hospital, mean/ median

7.8/6 (8.3)



GFR glomerular filtration rate, BMI body mass index, CABG coronary artery bypass graft, ASD atrial septal defect, CSRU cardiac surgery recovery unit

or there was an inadequate duration of cEEG data (100 mg/kg), the second patient was exposed to low-dose tranexamic acid and the third patient had no exposure to tranexamic acid [5, 8, 10]. One-year follow-up confirmed no long-term sequelae of the seizures in these three patients. The anesthetic agents used in cardiac surgery at our institution include fentanyl, sufentanyl, propofol, and sevoflurane.

Discussion Incidence of Seizures Our finding of a 3 % incidence (two convulsive and one non-convulsive) of seizures (if we consider that our EEG monitoring system is 68 % sensitive, we could well have

missed one patient with non-convulsive seizures, yielding an incidence of 4 %) is much lower than would be expected in a general ICU [13, 14] and much less than that seen in neurocritical care populations [13]. In the general ICU population, seizures are often seen in patients with cerebral injury, focal ischemia, or metabolic disturbances [13]. The vast majority of patients post cardiac surgery do not have pre- or post-operative cerebral pathology or severe metabolic disturbances that would predispose them to seizures. Thus, it is reassuring that post-cardiac surgery patients have fewer seizures than other ICU patient populations. While only a small number of patients had seizures, this single-center prospective observational study demonstrated a higher incidence of seizures following surgery in patients specifically having open-chamber cardiac surgery; confirming a similar observation in our previous retrospective

123

Neurocrit Care

Fig. 3 a Intra-operative exposure to tranexamic acid (TXA) and cefazolin in mg/kg. The data are shown as the average and the individual data for each patient having seizures. b eGFR presented as the average and the individual data for each patient having seizures. c Length of stay in the CSRU and length of stay in hospital in days shown as the average and the individual data for each patient having seizures

123

study [8]. Patients with seizures were older than 64 years of age and had more prolonged duration of surgery and cardiopulmonary bypass times. These results are consistent with prior reports based on retrospective studies of seizures associated with cardiac surgery [7, 8, 11]. While no statistical association between the total exposure to tranexamic acid and the occurrence of seizures can be made (based on the low number of post-operative seizures encountered in this study), only one of the 3 patient with seizures had exposure to higher doses of tranexamic acid; which has previously been reported as a potential risk factor for seizures after cardiac surgery [6–8, 10–12]. The incidence of seizures after cardiac surgery and exposure to tranexamic acid vary from 1 to 4 % [6, 7]. We do not feel that exposure to anesthetic agents was a potential contributor to seizure etiology in this study. During the study, patients were exposed to fentanyl, sufentanyl, propofol, or sevoflurane. Of these agents only sevoflurane administration has been associated with seizures in pediatric patients [24]. A previous study has demonstrated that fentanyl induction of anesthesia using high-dose fentanyl is not associated with EEG changes [25]. All anesthetic agents were discontinued at the conclusion of surgery, and while still in the operating room, several hours prior to the onset of seizures. All seizures in our prospective study were self-limited and no occurrences of status epilepticus were found. Previous studies show that 30 % of seizures could be missed using the subhairline EEG montage used in this study. Since we should have captured 68 % of the seizures, we may have missed a maximum of 1 patient with seizures in the study population. These data should be interpreted with caution due to the low number of patients with recorded seizures in our cohort. Continuous and extended EEG monitoring is costly and laborious in the ICU. Our results suggest that a high number of potentially detrimental non-convulsive seizures are not occurring in patients after cardiac surgery. In this case, it is unlikely that cEEG after cardiac surgery would be costeffective or improve long-term patient outcomes. Recently reported risk factors associated with seizures after cardiac surgery include open-chamber cardiac surgery [5, 7, 11], deep hypothermic circulatory arrest, [11] aortic calcification or atheroma, [11] critical preoperative state [11] or high APACHE II score, [7] tranexamic acid exposure, [7, 8, 26] preoperative cardiac arrest [7], long CPB time [7], previous cardiac surgery [7], poor renal function [5, 6], age C 75 years [5], and peripheral vascular disease [5]. In our study group, the majority of patients were recruited from a pre-operative clinic, were clinically stable pre-operatively and few had had previous surgery. This may be one reason that the rate of seizures seen in our study was quite low.

Neurocrit Care

On its own, cefazolin is known to have epileptogenic properties [4, 27]. Likewise, tranexamic acid is known to be epileptogenic in experimental models [2, 28]. Experimental evidence suggests that concomitant exposure to tranexamic acid and cefazolin may increase the risk of post-operative seizures [27]. It is unknown whether this phenomenon is borne out clinically or if the potential risk is additive or synergistic. This would an important consideration in patients undergoing cardiac surgery who also have impaired renal function. Both tranexamic acid and cefazolin are renally cleared and, therefore, the effective circulating dose and the duration of exposure to both agents is likely to be higher if careful dose adjustments are not made. Patient Outcomes In our study population, patients presenting with a seizure within 24 h of cardiac surgery was not shown to significantly affect patient outcomes. There was no significant increase in ICU or total hospital lengths of stay. There was no significant increase in post-operative complications and there was no effect on mortality. These results differ from previous reports in the literature from retrospective database studies and case–control studies, which have shown increases in duration of mechanical ventilation [5], increases in ICU [5] or hospital length of stay [6], and increases in mortality associated with seizures after cardiac surgery [7, 11]. Study Limitations The following study was a single-center prospective observational study with a lower than expected yield of patients with seizures. While observational study designs have inherent limitations, this study design has a high degree of external validity and reflects daily clinical practice well. Observational studies may be subject to selection bias, which we attempted to minimize by including all consecutive consenting patients undergoing cardiac surgery at our institution. The prospective design of this study also provides strength to the results. Despite the single-center study design, we feel that the data obtained are generalizable to most patients undergoing cardiac surgery because of the prospective study design, the diversity of patients enrolled, and the cEEG monitoring applied to all included patients. Furthermore, we did not exclude patients who had cerebral ischemic complications from surgery. Peri-operative stroke will certainly increase the risk of post-operative seizures [11], but exposure to epileptogenic medications

will potentially further decrease the seizure threshold and, therefore, may continue to put these patients at higher risk of seizures. Of note, none of the patients with post-operative seizures had significant associated cerebral abnormalities on neuroimaging or physical examination. In this prospective study fewer patients had seizures than predicted. Thus, the study was inadequately powered for detailed statistical analysis. Given this fact, the data should be interpreted with caution and no firm conclusions can be drawn. We hypothesize that the reason for the difference in the incidence of seizures within 24 h of cardiac surgery being lower in this prospective study (2010–2011) when compared to the recent retrospective (2007–2009) study at our institution [8], reflects a change in clinical practice. In between the two studies, the dosing regimens for tranexamic acid used during cardiac surgery were reduced based on our institutional findings and findings in the literature suggesting an association between high-dose exposure to tranexamic acid and an increased risk of seizures following cardiac surgery [5, 8].

Conclusion While non-convulsive seizures do occur after cardiac surgery, the incidence of electrographic seizures in this population is low (1 %) and all electrographic seizures recorded were self-limited with no deleterious effects on outcomes. Our findings are in agreement with previous studies with respect to potential risk factors for seizures after cardiac surgery, but not supportive of the previously reported deleterious effects of seizures on patient outcomes and mortality. The cardiac surgery patient population appears to be qualitatively different from ICU patients who have more serious, acute/subacute systemic illnesses or primary central nervous system insults. Thus, the prevalence of seizures in ICUs is considerably higher than in CSRUs. Further study on the effects of exposure to multiple epileptogenic medications in the peri-operative period in the context of impaired drug metabolism will be of interest in an effort to reduce the risk of post-operative seizures in patients after cardiac surgery. Acknowledgments Many thanks to Drs. Bob Kiaii, Ray Guo, Richard Novick, Neil McKenzie, and Mackenzie Quantz who had patients under their care participate in this study. Funding Sources No external funding was received for this study. None of the study authors received compensation, financial or otherwise, for completion of the study. Sources of Support

None.

123

Neurocrit Care Conflict of interest/Disclosures

None.

References 1. Roach GW, Kanchuger M, Mangano CM, et al. Adverse cerebral outcomes after coronary bypass surgery. Multicenter Study of Perioperative Ischemia Research Group and the Ischemia Research and Education Foundation Investigators. N Engl J Med. 1996;335:1857–63. 2. Schlag MG, Hopf R, Zifko U, Redl H. Epileptic seizures following cortical application of fibrin sealants containing tranexamic acid in rats. Acta Neurochir. 2002;144:63–9. 3. Coffey CE, Massey EW, Roberts KB, Curtis S, Jones RH, Pryor DB. Natural history of cerebral complications of coronary artery bypass graft surgery. Neurology. 1983;33:1416–21. 4. Sugimoto M, Uchida I, Mashimo T, et al. Evidence for the involvement of GABA(A) receptor blockade in convulsions induced by cephalosporins. Neuropharmacology. 2003;45:304–14. 5. Kalavrouziotis D, Voisine P, Mohammadi S, Dionne S, Dagenais F. High-dose tranexamic acid is an independent predictor of early seizure after cardiopulmonary bypass. Ann Thorac Surg. 2012;93:148–54. 6. Montes FR, Pardo DF, Carreno M, Arciniegas C, Dennis RJ, Umana JP. Risk factors associated with postoperative seizures in patients undergoing cardiac surgery who received tranexamic acid: a case-control study. Ann Card Anaesth. 2012;15:6–12. 7. Manji RA, Grocott HP, Leake J, et al. Seizures following cardiac surgery: the impact of tranexamic acid and other risk factors. Can J Anaesth. 2012;59:6–13. 8. Murkin JM, Falter F, Granton J, Young GB, Burt C, Chu M. High-dose tranexamic acid is associated with nonischemic clinical seizures in cardiac surgical patients. Anesth Analg. 2010;110:350–3. 9. Martin K, Knorr J, Breuer T, et al. Seizures after open heart surgery: comparison of epsilon-aminocaproic acid and tranexamic acid. J Cardiothorac Vasc Anesth. 2011;25:20–5. 10. Keyl C, Uhl R, Beyersdorf F, et al. High-dose tranexamic acid is related to increased risk of generalized seizures after aortic valve replacement. Eur J Cardiothorac Surg. 2011;39:e114–21. 11. Goldstone AB, Bronster DJ, Anyanwu AC, et al. Predictors and outcomes of seizures after cardiac surgery: a multivariable analysis of 2,578 patients. Ann Thorac Surg. 2011;91:514–8. 12. Bell D, Marasco S, Almeida A, Rowland M. Tranexamic acid in cardiac surgery and postoperative seizures: a case report series. Heart Surg Forum. 2010;13(E257–9):13.

123

13. Young GB, Doig GS. Continuous EEG monitoring in comatose intensive care patients: epileptiform activity in etiologically distinct groups. Neurocrit Care. 2005;2:5–10. 14. Towne AR, Waterhouse EJ, Boggs JG, et al. Prevalence of nonconvulsive status epilepticus in comatose patients. Neurology. 2000;54:340–5. 15. Gupta SK, Parihar A. Seizures in the Intesive Care Unit. JK Sci. 2000;2:81. 16. DeLorenzo RJ, Waterhouse EJ, Towne AR, et al. Persistent nonconvulsive status epilepticus after the control of convulsive status epilepticus. Epilepsia. 1998;39:833–40. 17. Young GB, Jordan KG. Do nonconvulsive seizures damage the brain?–Yes. Arch Neurol. 1998;55:117–9. 18. Goldstone AB, Bronster DJ, Chikwe J. Seizures after adult cardiac surgery. J Cardiothorac Vasc Anesth. 2011;25:e25–6. 19. Claassen J, Mayer SA, Kowalski RG. Detection of electrographic seizures with continuous EEG monitoring in critically ill patients. Neurology. 2004;62:1743–8. 20. Bleck TP, Smith MC, Pierre SJ, Jares JJ, Murray J, Hansen CA. Neurologic complications of critical medical illnesses. Crit Care Med. 1993;21:98–103. 21. Young GB, Sharpe MD, Savard M, Al Thenayan E, Norton L, Davies-Schinkel C. Seizure detection with a commercially available bedside EEG monitor and the subhairline montage. Neurocrit Care. 2009;11:411–6. 22. Young GB, McLachlan RS, Kreeft JH, Demelo JD. An electroencephalographic classification for coma. Can J Neurol Sci. 1997;24:320–5. 23. Levey AS, Greene T. A simplified equation to predict glomerular filtration rate from serum creatinine (Abstr). J Am Soc Nephrol. 2000;11:155A. 24. Constant I, Seeman R, Murat I. Sevoflurane and epileptiform EEG changes. Paediatr Anaesth. 2005;15(4):266–74. 25. Murkin JM, Moldenhauer CC, Hug CC Jr, Epstein CM. Absence of seizures during induction of anesthesia with high-dose fentanyl. Anesth Analg. 1984;63(5):489–94. 26. Kalavrouziotis D, Voisine P, Mohammadi S, Dionne S, Dagenais F. Reply. Ann Thorac Surg. 2012;94:1038–9. 27. Akula KK, Dhir A, Kulkarni SK. Pro-convulsant effect of cefazolin sodium against pentylenetetrazol- or picrotoxin-induced convulsions in mice. Indian J Exp Biol. 2007;45:720–5. 28. Furtmuller R, Schlag MG, Berger M, et al. Tranexamic acid, a widely used antifibrinolytic agent, causes convulsions by a gamma-aminobutyric acid(A) receptor antagonistic effect. J Pharmacol Exp Ther. 2002;301:168–73.

A prospective observational study of seizures after cardiac surgery using continuous EEG monitoring.

Recently, there have been several retrospective reports suggesting an increased frequency in seizures after cardiopulmonary bypass, associated with in...
555KB Sizes 0 Downloads 3 Views