- Email: [email protected]
Carotid artery disease and perioperative stroke risk after surgical aortic valve replacement: A nationwide inpatient sample analysis
Journal of Clinical Neuroscience xxx (2017) xxx–xxx
Contents lists available at ScienceDirect
Journal of Clinical Neuroscience journal homepage: www.elsevier.com/locate/jocn
Clinical commentary
Carotid artery disease and perioperative stroke risk after surgical aortic valve replacement: A nationwide inpatient sample analysis Reshmi Udesh a, Amol Mehta a, Thomas Gleason c, Parthasarathy D. Thirumala a,b,⇑ a
Department of Neurological Surgery, University of Pittsburgh, Pittsburgh, PA, USA University of Pittsburgh School of Medicine, Pittsburgh, PA, USA c Department of Cardiothoracic Surgery, University of Pittsburgh Medical Center, Pittsburgh, PA, USA b
a r t i c l e
i n f o
Article history: Received 12 January 2017 Accepted 3 April 2017 Available online xxxx Keywords: Perioperative stroke Carotid stenosis Aortic valve replacement Carotid artery disease
a b s t r a c t To study the role of carotid stenosis (CS) and cerebrovascular disease as independent risk factors for perioperative stroke following surgical aortic valve replacement (SAVR). The National Inpatient Sample (NIS) database was used for our study. All patients who underwent SAVR from 1999 to 2011 were identified using ICD-9 codes. Univariate and multivariate analysis of baseline characteristics, Elixhauser comorbidities and other covariates were examined to identify independent predictors of perioperative strokes following SAVR. Data on 50,979 patients who underwent SAVR from 1999 to 2011 was obtained. The mean age of the study cohort was 60.5. The study patients were predominantly Caucasian (79.3%) and males (60.01%). The incidence of perioperative stroke was 2.48%. CS (OR 1.8, 95%CI 1.1–2.8, p = 0.009) and cerebral arterial occlusion (OR 3.4, 95% CI 1.3–8.9) significantly increased perioperative stroke risk following SAVR. Infective endocarditis (OR 4.6, 95%CI 3.8–5.6, p = 0.00) and neurological disorders (OR 4.8, 95% CI 4–5.8, p = 0.00) appeared to be the strongest risk factors for strokes. Other risk factors found to be significant predictors of perioperative strokes (p < 0.05) were – age, higher VWR scores, CS, cerebral arterial occlusion, infective endocarditis, DM, HTN, renal failure, neurological disorders, coagulopathy and hypothyroidsm. In conclusion, perioperative stroke risk has remained more or less constant despite advancements in surgical techniques with risk having gone up in patients <65 years of age. CS and cerebral arterial occlusion significantly increase stroke risk following SAVR. Improved patient selection with pre-operative risk stratification and institution of preventive strategies are necessary to improve operative outcomes following SAVR. Ó 2017 Elsevier Ltd. All rights reserved.
1. Introduction Perioperative strokes, defined as neurological deficits which develop within 30 days of the procedure are common complications with an incidence of 2% following surgical aortic valve replacement (SAVR) [1,2]. Their implications on post-operative mortality rates and quality of life are devastating with serious economic ramifications due to longer stays in hospitals and chronic care facilities [3,4]. AVR is the most common cardiac valve surgery performed in the US with 22,000 procedures done annually [5] and reported operative mortality of 2.5–3% [1,2,5]. Previous studies on timing of the strokes show that 55–72% of perioperative strokes occur in the first 24 h and the PARTNER trial reported 51% of
⇑ Corresponding author at: Center for Clinical Neurophysiology, Department of Neurological Surgery, UPMC Presbyterian-Suite B-400, 200 Lothrop Street, Pittsburgh, PA 15213, USA. E-mail address: [email protected] (P.D. Thirumala).
post-operative strokes in the first 10 days following AVR making it important for us to study the periprocedural causes of neurological events in the hospital [6–8]. Beyond the first post-operative week the surgery or its associated risks were no longer implicated to contribute to strokes [6]. Previous studies have shown history of cerebrovascular disease or carotid stenosis (CS) with prior stroke to be the strongest independent risk factor of perioperative stroke following cardiac surgery [9,10]. The common risk factors contributing to perioperative strokes following SAVR include: pre-operatively – age >65 years, female sex, history of cerebrovascular disease (stroke/transient ischemic attacks), cardiac failure, atrial fibrillation, diabetes, hypertension, prior cardiac surgery and endocarditis, etc. [3,9,10]. Intra-operatively, cerebral hypoperfusion or embolic phenomena [4] and post-operatively, new onset atrial fibrillation and low cardiac output syndrome were found culpable [10]. Although several studies have examined the relationship of carotid stenosis with stroke risk following coronary artery bypass
http://dx.doi.org/10.1016/j.jocn.2017.04.006 0967-5868/Ó 2017 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Udesh R et al. Carotid artery disease and perioperative stroke risk after surgical aortic valve replacement: A nationwide inpatient sample analysis. J Clin Neurosci (2017), http://dx.doi.org/10.1016/j.jocn.2017.04.006
2
R. Udesh et al. / Journal of Clinical Neuroscience xxx (2017) xxx–xxx
surgery and cardiac surgeryin general [9,10], a large scale risk analysis of perioperative stroke following aortic valve intervention has not been done before. Some studies also implicate intracranial artery stenosis and vertebral artery stenosis in the etiology of perioperative strokes [11,12]. The primary aim of our study is to identify whether carotid artery disease is an independent risk factor for perioperative stroke following SAVR. We will also evaluate vertebral, basilar and intracranial stenosis and occlusion as a risk factor for perioperative stroke following SAVR. Finally we will evaluate the trends in perioperative stroke and its relationship with carotid disease over a ten year period. Identifying the significance of carotid artery disease in the incidence of perioperative stroke will help us formulate preoperative screening and risk-stratification of patients on the need for carotid revascularization before or after SAVR. Further the need for intra-operative neurophysiological monitoring to cerebral hypoperfusion [13–15] or intense medical management for intracranial stenosis can be evaluated [16]. 2. Methods The National Inpatient Sample (NIS) database was used to select the study population over a period of 11 years from 1999 to 2011 for SAVR. International Classification of Diseases, Ninth RevisionClinical Modification (ICD-9-CM) codes were used to identify diagnoses and risk factors in the NIS database. Patients were identified using ICD-9 codes 35.21 and 35.22 for SAVR and extracted for our analysis. Patients below the age of 18 years were excluded from our study. Baseline characteristics such as age, gender, race, admission status and comorbid conditions commonly observed in the patient demographic undergoing SAVR were studied for our analysis. The NIS database provides 29 Elixhauser comorbidities based on standard ICD-9 codes [17]. Other additional risk factors studied were carotid stenosis, cerebral arterial occlusion, vertebral and basilar artery stenosis, atrial fibrillation, infective endocarditis, left ventricular dysfunction (LVD), previous history of stroke or transient ischemic attack (TIA), previous history of any cardiac surgery and use of cardioplumonary bypass. The ICD-9 codes used to identify the covariates and outcomes are listed in supplement Table 1. The primary outcome studied was perioperative stroke following SAVR. Carotid stenosis was used as the independent variable. Risk stratification was done using Van Walraven (VWR) score. It is a summary score developed for Elixhauser Comorbidities by modeling in-hospital mortality with inpatient admission data [18]. The summary score is a weighted combination of the 29 Elixhauser comorbidities, where a larger comorbidity weight indicates a stronger association between comorbidity and in-hospital mortality. 3. Statistical analysis All data are presented as mean ± SD or percentages. Univariate comparisons between groups were done using unpaired t-tests for continuous variables and a survey-adjusted Wald test for variables that were categorical in nature. In our multivariable analysis, we included variables that were statistically significant in the univariate analysis, that had alarge enough group population, that lacked missing data, and were established as risk factors in previous studies. We elected to exclude each individual Elixhauser Comorbidity and use the van Walraven score as a surrogate, as we did not want to risk overloading our model with variables that may have been either statistically insignificant or not relevant to our outcome of interest. Data extraction was performed using SAS 9.3 (SAS Institute, Inc., Cary, NC), as was the creation of the Elixhauser comorbidity index and the generation of the van Wal-
raven score [19]. All subsequent statistical analyses were performed using Stata version.14 (StataCorp, College Station, TX). 4. Results 4.1. Baseline characteristics The study population comprised of 50,979 patients who underwent surgical aortic valve replacement from 1999 to 2011. The baseline data of the study cohort and univariate analysis of risk factors is provided in Table 1. The mean age of the study cohort was 60.58, with majority of the population (58.13%) under the age of 65. The study patients were predominantly Caucasian (79.3%) and males (60.01%). The perioperative stroke incidence was found to be the same irrespective of gender. The procedure was done electively in 68.47% of the cases. The incidence of perioperative stroke was 2.48% in our study. The patients were stratified based on their Van Walraven (VWR) scores into low/intermediate and high risk groups. The average VWR score was found to be 2.15 with 75.2% of the study patients categorized as low risk. 4.2. Univariate analysis of pre-operative predictors for perioperative strokes Univariate analysis of risk factors (see Tables 1 and 2) suggests the following significant predictors of perioperative stroke (p < 0.05)-advancing age, higher VWR scores, males, African American race, elective procedures, CS, cerebral arterial occlusion, infective endocarditis, CHF, valve disease, cardiopulmonary bypass use, CLD, paralysis and other neurological disorders, complicated DM, RF, coagulopathy, weight loss, fluid and electrolyte abnormalities, hypertension and drug abuse. Carotid stenosis and other precerebral artery stenosis was seen in 1.2% and 0.43% of our study patients and were both significant predictors of strokes (OR 2.01, 95%CI 1.3–3.0, p = 0.001 and OR 2.1, 95%CI 1.1–4.1, p = 0.016 respectively). No patients coded for vertebral or basilar artery stenosis and were not included on our analysis. Infective endocarditis, seen in 4.8% of study patients was another significant predictor (OR 4.8, 95%CI 4.1–5.7, p = 0.00). CBP was used in 88.9% of patients and showed a significantly decreased perioperative stroke risk (OR 0.7, 95%CI 0.6–0.9, p = 0.006). A statistically significant stroke risk was seen in patients with higher VWR scores (p < 0.05). 4.3. Independent predictors of perioperative strokes by multivariate analysis The following risk factors were found to be significant predictors of perioperative strokes (p < 0.05) – age, higher VWR scores, CS, cerebral arterial occlusion, infective endocarditis, DM, HTN, renal failure, neurological disorders, coagulopathy and hypothyroidsm (Table 3). Carotid stenosis (OR 1.8, 95%CI 1.1–2.8, p = 0.009) and cerebral arterial occlusion (OR 3.4, 95% CI 1.3–8.9) were significantly associated with an increased perioperative stroke risk following SAVR. Infective endocarditis (OR 4.6, 95%CI 3.8–5.6, p = 0.00) and neurological disorders (OR 4.8, 95% CI 4– 5.8, p = 0.00) appeared to be the strongest risk factors for strokes. 4.4. Trend analysis An analysis of the trends in perioperative stroke incidence and the average age and VWR scores of patients who underwent SAVR from 1999 to 2011 is depicted in Table 4. The average age of patients undergoing the procedure has remained the same over the years (61.4 years). The VWR score shows an increasing trend over the years from an average score of 1.4 in 1999 to 2.5 in
Please cite this article in press as: Udesh R et al. Carotid artery disease and perioperative stroke risk after surgical aortic valve replacement: A nationwide inpatient sample analysis. J Clin Neurosci (2017), http://dx.doi.org/10.1016/j.jocn.2017.04.006
3
R. Udesh et al. / Journal of Clinical Neuroscience xxx (2017) xxx–xxx Table 1 Baseline characteristics and univariate analysis of pre-operative risk factors. Variables
% of Patients
Perioperative strokes Yes% (n)
No% (n)
Unadjusted Odds Ratio (95% CI)
p-value
Age – M ± SD <65 65–74 75–84 >85 Gender Female Male Race/Ethnicity White African American Hispanic Asian Native American Other/Missing Admission Status Emergent Urgent Elective Risk Category Average Van Walraven Score Low Risk (VWR < 5) Moderate Risk (VWR 5–14) High Risk (VWR > 14) Outcome 1. Perioperative stroke Pre-operative Risk Factors 2. Carotid Stenosis 3. Precerebral artery stenosis (other) 4. Cerebral occlusion 5. Vertebral artery stenosis 6. Basilar artery stenosis 7. LVD and heart failure 8. Atrial Fibrillation 9. Previous h/o stroke or transient ischemic attack (TIA) 10. Infective endocarditis 11. Previous h/o myocardial infarction or angina 12. Previous cardiac surgery 13. Previous CEA/CAS 14. Previous h/o Coronary artery disease or CABG 15. Cardiopulmonary bypass
60.58 ± 0.39 58.13% 22.88% 18.24% 2.16%
2.07% 2.55% 3.29% 4.70%
97.93% 97.45% 96.71% 95.30%
0.671 1.034 1.438 1.974
(0.592–0.761) (0.892–1.200) (1.237–1.671) (1.439–2.707)
0.000 0.649 0.000 0.000
39.98% 60.02%
2.61% (457) 2.41% (635)
97.39% (17,059) 97.59% (25,632)
1.086 (0.954–1.237) 0.664 (0.649–0.681)
0.210 0.000
79.33% 6.91% 8.13% 1.95% 0.53% 3.12%
2.39% 3.47% 2.18% 2.54% 1.23% 3.05%
97.61% 96.53% 97.82% 97.46% 98.77% 96.95%
NA 1.463 0.910 1.063 0.507 1.283
(1.143–1.874) (0.692–1.197) (0.679–1.664) (0.131–1.958) (0.912–1.805)
0.003 0.501 0.788 0.324 0.151
14.92% 16.57% 68.47%
4.42% (258) 3.48% (225) 1.79% (478)
95.58% (5,537) 96.52% (6,232) 98.21% (26,150)
0.779 (0.634–0.958) 0.395 (0.335–0.465)
0.018 0.000
2.15 ± 0.08 75.22% 23.12% 1.64%
1.43% (471) 5.22% (528) 12.42% (92)
98.57% (32,461) 94.78% (9,602) 87.58% (631)
NA 3.805 (3.350–4.321) 9.803 (7.647–12.566)
0.000 0.000
4.84% (25) 5.27% (10) 15.93% (8) 31.46% (1) 32.13% (1) 2.77% (74) 2.67% (402) 1.52% (8) 9.51% (206) 2.27% (205) 1.72% (6) 1.29% (4) 2.48% (46) 2.41% (942)
95.16% (508) 94.73% (181) 84.07% (41) 68.54% (2) 67.87% (2) 97.23% (2,615) 97.33% (14,672) 98.48% (530) 90.49% (1,929) 97.73% (8,897) 98.285 (321) 98.71% (313) 97.52% (1,811) 97.59% (37,971)
2.018 2.194 7.482 18.02 18.58 1.124 1.121 0.603 4.838 0.889 0.683 0.512 0.998 0.780
0.001 0.016 0.000 0.018 0.017 0.358 0.064 0.151 0.000 0.139 0.354 0.181 0.992 0.006
(533) (254) (260) (44)
(635) (83) (60) (16) (2) (33)
(24,921) (9,782) (7,724) (900)
(25,882) (2,253) (2,688) (633) (171) (1,007)
2.48% 1.21% 0.43% 0.11% 0% 0% 6.09% 34.43% 1.21% 4.85% 20.75% 0.74% 0.72% 4.24% 88.91%
(1.338–3.043) (1.159–4.153) (3.456–16.199) (1.63–198.71) (1.67–205.99) (0.875–1.444) (0.993–1.266) (0.302–1.202) (4.096–5.715) (0.761–1.038) (0.306–1.527) (0.192–1.363) (0.736–1.354) (0.654–0.931)
LVD – Left ventricular dysfunction; CEA – carotid endarterectomy; CAS – Carotid artery stenting; CABG – Coronary artery bypass grafting; CI – confidence interval.
2011 (Fig. 1). The perioperative stroke rates have remained more or less in the 2–2.7% range over this decade. When stratified for age and gender, stroke rates have increased over the years in males and patients < 65 years with a decline in stroke incidence in the 65–84 age group. The stroke risk remains the same in females and patients >85 years (Figs. 2 and 3).
5. Discussion Analysis of baseline demographics shows advancing age and increased patient risk profile (VWR scores) to be significant predictors of stroke incidence as with patients undergoing any major cardiac surgery [1,20,21]. Majority of the study patients who underwent SAVR were <65 years (58.1%) and males (60%). Our trend analysis shows a significant increase in stroke rates in patients <65 years which suggest the need for better overall patient selection and assessment of comorbidities. Although prior studies show female gender to be a risk factor for stroke [20,22] our multivariate analysis did not reveal gender to be an independent predictor. The number of patients categorized as medium risk (VWR 5–14) and high risk (VWR > 14) have increased from 1999 to 2011 and our multivariate analysis shows a statistically significant increase in stroke risk with greater VWR scores. Importantly, race
and admission status were not found to be independent predictors of strokes. Carotid stenosis (OR 1.8, 95%CI 1.1–2.8, p = 0.009) and cerebral arterial occlusion (OR 3.4, 95% CI 1.3–8.9) were significantly associated with an increased perioperative stroke risk following SAVR. Cerebrovascular disease or CS are known risk factors of perioperative strokes following cardiac surgery and despite previous studies ruling out pre-operative carotid screening [23] or intervention [8] before cardiac procedures it still remains an important concern with significant perioperative stroke risk. Infective endocarditis (OR 4.6, 95%CI 3.8–5.6, p = 0.00) and previous neurological dysfunction (OR 4.8, 95% CI 4–5.8, p = 0.00) also appeared to be the strong risk factors for strokes. Infective endocarditis is associated with a 12–40% stroke risk due to cerebral septic embolism without the added risk of valve surgery [24]. Several studies show preoperative cerebral imaging with MRI to rule out silent ischemic events and adequate perioperative anticoagulation to reduce perioperative stroke risk [24]. SAMMPRIS [25] and CAVATAS [26] trials signify the role of vertebral, basilar and other intracranial artery stenosis on perioperative strokes. However, we were unable to analyze these risk factors due to inadequate patients being coded for them. Prolonged cardiopulmonary bypass time has been reported to be an important risk factor for stroke during SAVR [9,27]. Our analysis did not reveal the use of CBP to be an
Please cite this article in press as: Udesh R et al. Carotid artery disease and perioperative stroke risk after surgical aortic valve replacement: A nationwide inpatient sample analysis. J Clin Neurosci (2017), http://dx.doi.org/10.1016/j.jocn.2017.04.006
4
R. Udesh et al. / Journal of Clinical Neuroscience xxx (2017) xxx–xxx
Table 2 Univariate analysis of Elixhauser comorbidities. Variables-Elixhauser comorbidities
1. 2. 3. 4. 5. 6. 7. 8. 9.
% of Patients
Valve disease CHF Pulmonary circulation disorders Peripheral vascular disease Paralysis Other Neurological disorders Chronic pulmonary disorders DM, uncomplicated DM, complicated 10. Hypothyroidsm 11. Renal failure 12. Liver failure 13. Peptic Ulcer Disease 14. AIDS 15. Lymphoma 16. Metastatic cancer 17. Solid tumor without metastasis 18. Collagen vascular disorders 19. Coagulopathy 20. Obesity 21. Weight loss 22. Fluid and electrolyte disorders 23. Chronic blood loss anemia 24. Nutritional anemia 25. Alcohol abuse 26. Drug abuse 27. Psychoses 28. Depression 29. Hypertension
0.79% 0.90% 0.20% 13.89% 0.97% 2.75% 16.94% 15.17% 2.07% 7.67% 6.25% 1.36% 0.33% 0.08% 0.52% 0.12% 1.82% 2.21% 13.32% 10.02% 1.77% 17.55% 1.19% 13.17% 2.09% 1.45% 1.27% 4.13% 5.01%
Perioperative strokes Yes% (n)
No% (n)
Unadjusted Odds Ratio (95% CI)
p-value
12.65% (44) 10.57% (44) 8.99% (8) 2.44% (149) 48.38% (212) 16.90% (204) 2.33% (174) 2.17% (144) 4.78% (42) 1.74% (58) 3.87% (108) 3.04% (19) 0.44% (1) 0% 1.40% (3) 3.88% (2) 1.82% (14) 1.94% (19) 3.75% (218) 1.91% (85) 9.67% (76) 3.79% (294) 3% (16) 2.19% (130) 2.86% (27) 4.46% (29) 2.32% (13) 1.68% (32) 2.19% (479)
87.35% (299) 89.43% (348) 91.01% (78) 97.56% (5,919) 51.62% (222) 83.10% (998) 97.67% (7,260) 97.83% (6,534) 95.22% (859) 98.26% (3,303) 96.13% (2,637) 96.96% (578) 99.56% (149) 100% (38) 98.60% (223) 96.12% (52) 98.18% (781) 98.06% (947) 96.25% (5,630) 98.09% (4,323) 90.33% (704) 96.21% (7,413) 97% (510) 97.81% (5,658) 97.14% (890) 95.54% (609) 97.68% (546) 98.32% (1,778) 97.81% (21,478)
5.882 4.786 3.897 0.980 45.25 9.594 0.926 0.850 2.013 0.678 1.644 1.234 0.171 NA 0.554 1.587 0.726 0.772 1.663 0.744 4.440 1.747 1.216 0.865 NA 1.855 0.931 0.659 0.784
(4.291–8.063) (3.332–6.875) (1.939–7.833) (0.822–1.169) (36.48–56.12) (8.105–11.356) (0.794–1.080) (0.709–1.019) (1.481–2.736) (0.512–0.899) (1.34–2.017) (0.772–1.971) (0.024–1.180)
0.000 0.000 0.000 0.829 0.000 0.000 0.331 0.080 0.000 0.007 0.000 0.378 0.073
(0.187–1.642) (0.387–6.497) (0.431–1.220) (0.481–1.240) (1.416–1.954) (0.595–0.931) (3.513–5.612 (1.510–2.020) (0.724–2.044) (0.711–1.052)
0.287 0.521 0.227 0.285 0.000 0.010 0.000 0.000 0.459 0.148
(1.272–2.706) (0.537–1.616) (0.462–0.940) (0.692–0.888)
0.001 0.802 0.021 0.000
CHF – Congestive Heart Failure; DM – diabetes mellitus; CI – confidence interval.
Table 3 Multivariate analysis of the predictors of perioperative stroke following aortic valve surgery. Variables
Age VWR Score Low Risk (VWR < 5) Moderate Risk (VWR 5–14) High Risk (VWR > 14) Pre-operative Risk Factors 1. Carotid Stenosis 2. Cerebral Occlusion 3. Infective Endocarditis 4. Cardiopulmonary Bypass Comorbidities 5. Diabetes-complicated 6. Renal Failure 7. Hypertension 8. Neurological Disorders 9. Coagulopathy 10. Hypothyroidsm
Perioperative strokes Odds Ratio (95% CI)
p-value
1.017 (1.012–1.022)
0.000
NA 2.747 (2.363–3.194) 5.157 (3.779–7.037)
0.000 0.000
1.810 3.429 4.652 0.907
(1.159–2.825) (1.323–8.884) (3.875–5.584) (0.758–1.085)
0.009 0.011 0.000 0.288
1.596 0.671 0.813 4.825 1.003 0.629
(1.158–2.202) (0.530–0.850) (0.712–0.927) (4.003–5.816) (0.841–1.188) (0.472–0.838)
0.004 0.001 0.002 0.000 0.000 0.002
Table 4 Trend analysis of perioperative strokes and patient risk scores over the years (1999– 2011). Calendar year
% of Perioperative strokes(n)
Average Van Walraven Score
Average Age
1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 Total
2.55 2.47 2.38 2.96 2.20 2.04 2.85 2.19 2.13 2.74 2.50 2.84 2.72 2.49
1.495 1.469 1.679 1.817 1.798 1.970 1.822 2.329 2.340 2.326 2.735 2.876 2.511
61.40 60.81 61.24 61.24 60.11 59.97 58.95 59.70 60.57 59.87 60.91 61.12 61.44
(86) (97) (87) (93) (83) (66) (94) (69) (61) (83) (86) (87) (100) (1092)
VWR – Von Walraven score; CI – Confidence interval.
independent predictor for stroke as we were unable to determine the duration of its use from the data available. Previous studies show the routine post-operative neurological evaluation of patients with MRI revealed a 38–54% silent stroke rate/TIA rate [28,29]. This subclinical stroke burden which remains grossly underreported raises the immediate need to study prevention strategies and cerebral protection devices [30]. Transcranial Doppler monitoring and DW-MRI have been shown to be efficacious in detecting these perioperative silent ischemic events [29,31,32]. The role of intra-operative neuromonitoring in detecting and preventing strokes during cardiac procedures has remained largely unexamined and requires further attention.
Fig. 1. Risk Stratification of Patients Undergoing SAVR based on VWR Score from 1999 to 2011.
Please cite this article in press as: Udesh R et al. Carotid artery disease and perioperative stroke risk after surgical aortic valve replacement: A nationwide inpatient sample analysis. J Clin Neurosci (2017), http://dx.doi.org/10.1016/j.jocn.2017.04.006
R. Udesh et al. / Journal of Clinical Neuroscience xxx (2017) xxx–xxx
5
rial occlusion, infective endocarditis, DM, HTN, renal failure, neurological disorders, coagulopathy and hypothyroidsm. The trends over the period of 10 years from 1999 to 2011 indicate that perioperative stroke risk has remained more or less constant despite advancements in surgical techniques with risk having gone up in patients <65 years of age. CS and cerebral arterial occlusion significantly increase perioperative stroke risk following SAVR. Improved patient selection with pre-operative risk stratification and institution of preventive strategies are necessary to improve operative outcomes following SAVR. Conflicts of interest None. Fig. 2. Incidence of Perioperative Strokes following SAVR in Males and Females from 1999 to 2011.
Sources of funding None. Acknowledgements None. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.jocn.2017.04.006. References
Fig. 3. Incidence of Perioperative Strokes stratified by Patient Age from 1999 to 2011.
Pre-operative carotid Doppler screening of CS is being studied increasingly for patients undergoing transcatheter aortic valve implantation [33], however; its utility in candidates for SAVR has not been examined comprehensively [23]. Further studies examining the role of standardized perioperative neurological evaluation and other preventive strategies will give us valuable insight into reducing perioperative stroke risk following SAVR. Study limitations Despite the valuable information gained from this large scale risk analysis of perioperative strokes on 50,000 patients our study is not without limitations. As the data is pooled from different hospitals and institutions there could be considerable variability in pre-operative patient selection and management which could have affected the outcome rates. Being an administrative database, the NIS relies heavily on correct coding of diagnoses and procedures and any coding errors or inconsistencies could result in inaccurate data. Inadequate coding of patients with vertebral/basilar artery disease which are important risk factors for strokes could be an indication of patients not being screened for them or a coding error. Information on strokes which could have occurred after the patients’ hospital discharge but within the 30-day post-operative limit and post-operative management of stroke patients is unavailable. Conclusion Risk factors found to be significant predictors of perioperative strokes (p < 0.05) were – age, higher VWR scores, CS, cerebral arte-
[1] Agarwal S, Garg A, Parashar A, et al. In-hospital mortality and stroke after surgical aortic valve replacement: A nationwide perspective. J Thorac Cardiovasc Surg 2015;150(571–8):e8. [2] Biancari F, Martin M, Bordin G, et al. Basic data from 176 studies on the immediate outcome after aortic valve replacement with or without coronary artery bypass surgery. J Cardiothorac Vasc Anesth 2014;28:1251–6. [3] Almassi GH, Sommers T, Moritz TE, et al. Stroke in cardiac surgical patients: determinants and outcome. Ann Thorac Surg 1999;68:391–7 [discussion 7–8]. [4] Wolman RL, Nussmeier NA, Aggarwal A, et al. Cerebral injury after cardiac surgery: identification of a group at extraordinary risk. Multicenter Study of Perioperative Ischemia Research Group (McSPI) and the Ischemia Research Education Foundation (IREF) Investigators. Stroke 1999;30:514–22. [5] Society of Thoracic Surgeons Adult Cardiac Surgery Database-Executive Summary. Available at: http://wwwstsorg/sites/default/files/documents/ 2015Harvest4ExecutiveSummarypdf. [6] Miller DC, Blackstone EH, Mack MJ, et al. Transcatheter (TAVR) versus surgical (AVR) aortic valve replacement: occurrence, hazard, risk factors, and consequences of neurologic events in the PARTNER trial. J Thorac Cardiovasc Surg 2012;143(832–43):e13. [7] Filsoufi F, Rahmanian PB, Castillo JG, et al. Incidence, imaging analysis, and early and late outcomes of stroke after cardiac valve operation. Am J Cardiol 2008;101:1472–8. [8] Li Y, Walicki D, Mathiesen C, et al. Strokes after cardiac surgery and relationship to carotid stenosis. Arch Neurol 2009;66:1091–6. [9] Bucerius J, Gummert JF, Borger MA, et al. Stroke after cardiac surgery: a risk factor analysis of 16,184 consecutive adult patients. Ann Thorac Surg 2003;75:472–8. [10] Hogue Jr CW, De Wet CJ, Schechtman KB, et al. The importance of prior stroke for the adjusted risk of neurologic injury after cardiac surgery for women and men. Anesthesiology 2003;98:823–9. [11] Coward LJ, McCabe DJ, Ederle J, et al. Long-term outcome after angioplasty and stenting for symptomatic vertebral artery stenosis compared with medical treatment in the Carotid And Vertebral Artery Transluminal Angioplasty Study (CAVATAS): a randomized trial. Stroke 2007;38:1526–30. [12] Derdeyn CP, Fiorella D, Lynn MJ, et al. Mechanisms of stroke after intracranial angioplasty and stenting in the SAMMPRIS trial. Neurosurgery 2013;72:777–95 [discussion 95]. [13] Nwachuku EL, Balzer JR, Yabes JG, et al. Diagnostic value of somatosensory evoked potential changes during carotid endarterectomy: a systematic review and meta-analysis. JAMA Neurol 2015;72:73–80. [14] Thiagarajan K, Cheng HL, Huang JE, et al. Is two really better than one? Examining the superiority of dual modality neurophysiological monitoring during carotid endarterectomy: A meta-analysis. World Neurosurg 2015;84 (1941–9):e1. [15] Thirumala PD, Udesh R, Muralidharan A, et al. Diagnostic value of somatosensory-evoked potential monitoring during cerebral aneurysm
Please cite this article in press as: Udesh R et al. Carotid artery disease and perioperative stroke risk after surgical aortic valve replacement: A nationwide inpatient sample analysis. J Clin Neurosci (2017), http://dx.doi.org/10.1016/j.jocn.2017.04.006
6
[16]
[17] [18]
[19]
[20]
[21]
[22]
[23]
[24]
R. Udesh et al. / Journal of Clinical Neuroscience xxx (2017) xxx–xxx clipping: a systematic review. World Neurosurg 2016;89:672–80. http://dx. doi.org/10.1016/j.wneu.2015.12.008. Derdeyn CP, Chimowitz MI, Lynn MJ, Fiorella D, Turan TN, Janis LS, et al. Aggressive medical treatment with or without stenting in high-risk patients with intracranial artery stenosis (SAMMPRIS): the final results of a randomised trial. Lancet 2014;383:333–41. http://dx.doi.org/10.1016/S0140-6736(13) 62038-3. Elixhauser A, Steiner C, Harris DR, et al. Comorbidity measures for use with administrative data. Med Care 1998;36:8–27. van Walraven C, Austin PC, Jennings A, et al. A modification of the Elixhauser comorbidity measures into a point system for hospital death using administrative data. Med Care 2009;47:626–33. Databases HCUP. Healthcare Cost and Utilization Project (HCUP). Rockville, MD: Agency for Healthcare Research and Quality; 2003. www.hcup-us. ahrq.gov/nisoverview.jsp. Idrees JJ, Schiltz NK, Johnston DR, et al. Trends, predictors, and outcomes of stroke after surgical aortic valve replacement in the United States. Ann Thorac Surg 2016;101:927–35. Mortasawi A, Arnrich B, Rosendahl U, et al. Is age an independent determinant of mortality in cardiac surgery as suggested by the EuroSCORE? BMC Surg 2002;2:8. Hogue Jr CW, Barzilai B, Pieper KS, et al. Sex differences in neurological outcomes and mortality after cardiac surgery: a society of thoracic surgery national database report. Circulation 2001;103:2133–7. Condado JF, Jensen HA, Maini A, et al. Should we perform carotid doppler screening before surgical or transcatheter aortic valve replacement? Ann Thorac Surg 2016. http://dx.doi.org/10.1016/j.athoracsur.2016.06.076. pii: S0003-4975(16)30792-5. Behrouz R. Preoperative cerebrovascular evaluation in patients with infective endocarditis. Clin Cardiol 2015;38:439–42.
[25] Kwon HM, Lynn MJ, Turan TN, et al. Frequency, risk factors, and outcome of coexistent small vessel disease and intracranial arterial stenosis: results from the stenting and Aggressive Medical Management for Preventing Recurrent Stroke in Intracranial Stenosis (SAMMPRIS) trial. JAMA Neurol 2016;73:36–42. [26] Ederle J, Bonati LH, Dobson J, et al. Endovascular treatment with angioplasty or stenting versus endarterectomy in patients with carotid artery stenosis in the Carotid And Vertebral Artery Transluminal Angioplasty Study (CAVATAS): long-term follow-up of a randomised trial. Lancet Neurol 2009;8:898–907. [27] Brown WR, Moody DM, Challa VR, et al. Longer duration of cardiopulmonary bypass is associated with greater numbers of cerebral microemboli. Stroke 2000;31:707–13. [28] Messe SR, Acker MA, Kasner SE, et al. Stroke after aortic valve surgery: results from a prospective cohort. Circulation 2014;129:2253–61. [29] Stolz E, Gerriets T, Kluge A, et al. Diffusion-weighted magnetic resonance imaging and neurobiochemical markers after aortic valve replacement: implications for future neuroprotective trials? Stroke 2004;35:888–92. [30] Floyd TF, Shah PN, Price CC, et al. Clinically silent cerebral ischemic events after cardiac surgery: their incidence, regional vascular occurrence, and procedural dependence. Ann Thorac Surg 2006;81:2160–6. [31] Knipp SC, Matatko N, Schlamann M, et al. Small ischemic brain lesions after cardiac valve replacement detected by diffusion-weighted magnetic resonance imaging: relation to neurocognitive function. Eur J Cradiothorac Surg 2005;28:88–96. [32] Spaziano M, Francese DP, Leon MB, et al. Imaging and functional testing to assess clinical and subclinical neurological events after transcatheter or surgical aortic valve replacement: a comprehensive review. J Am Coll Cardiol 2014;64:1950–63. [33] Ravindra VM, Mazur MD, Kumpati GS, et al. Carotid artery stenosis in the setting of transcatheter aortic valve replacement: clinical and technical considerations of carotid stenting. World Neruosurg 2016;86:194–8.
Please cite this article in press as: Udesh R et al. Carotid artery disease and perioperative stroke risk after surgical aortic valve replacement: A nationwide inpatient sample analysis. J Clin Neurosci (2017), http://dx.doi.org/10.1016/j.jocn.2017.04.006
Contents lists available at ScienceDirect
Journal of Clinical Neuroscience journal homepage: www.elsevier.com/locate/jocn
Clinical commentary
Carotid artery disease and perioperative stroke risk after surgical aortic valve replacement: A nationwide inpatient sample analysis Reshmi Udesh a, Amol Mehta a, Thomas Gleason c, Parthasarathy D. Thirumala a,b,⇑ a
Department of Neurological Surgery, University of Pittsburgh, Pittsburgh, PA, USA University of Pittsburgh School of Medicine, Pittsburgh, PA, USA c Department of Cardiothoracic Surgery, University of Pittsburgh Medical Center, Pittsburgh, PA, USA b
a r t i c l e
i n f o
Article history: Received 12 January 2017 Accepted 3 April 2017 Available online xxxx Keywords: Perioperative stroke Carotid stenosis Aortic valve replacement Carotid artery disease
a b s t r a c t To study the role of carotid stenosis (CS) and cerebrovascular disease as independent risk factors for perioperative stroke following surgical aortic valve replacement (SAVR). The National Inpatient Sample (NIS) database was used for our study. All patients who underwent SAVR from 1999 to 2011 were identified using ICD-9 codes. Univariate and multivariate analysis of baseline characteristics, Elixhauser comorbidities and other covariates were examined to identify independent predictors of perioperative strokes following SAVR. Data on 50,979 patients who underwent SAVR from 1999 to 2011 was obtained. The mean age of the study cohort was 60.5. The study patients were predominantly Caucasian (79.3%) and males (60.01%). The incidence of perioperative stroke was 2.48%. CS (OR 1.8, 95%CI 1.1–2.8, p = 0.009) and cerebral arterial occlusion (OR 3.4, 95% CI 1.3–8.9) significantly increased perioperative stroke risk following SAVR. Infective endocarditis (OR 4.6, 95%CI 3.8–5.6, p = 0.00) and neurological disorders (OR 4.8, 95% CI 4–5.8, p = 0.00) appeared to be the strongest risk factors for strokes. Other risk factors found to be significant predictors of perioperative strokes (p < 0.05) were – age, higher VWR scores, CS, cerebral arterial occlusion, infective endocarditis, DM, HTN, renal failure, neurological disorders, coagulopathy and hypothyroidsm. In conclusion, perioperative stroke risk has remained more or less constant despite advancements in surgical techniques with risk having gone up in patients <65 years of age. CS and cerebral arterial occlusion significantly increase stroke risk following SAVR. Improved patient selection with pre-operative risk stratification and institution of preventive strategies are necessary to improve operative outcomes following SAVR. Ó 2017 Elsevier Ltd. All rights reserved.
1. Introduction Perioperative strokes, defined as neurological deficits which develop within 30 days of the procedure are common complications with an incidence of 2% following surgical aortic valve replacement (SAVR) [1,2]. Their implications on post-operative mortality rates and quality of life are devastating with serious economic ramifications due to longer stays in hospitals and chronic care facilities [3,4]. AVR is the most common cardiac valve surgery performed in the US with 22,000 procedures done annually [5] and reported operative mortality of 2.5–3% [1,2,5]. Previous studies on timing of the strokes show that 55–72% of perioperative strokes occur in the first 24 h and the PARTNER trial reported 51% of
⇑ Corresponding author at: Center for Clinical Neurophysiology, Department of Neurological Surgery, UPMC Presbyterian-Suite B-400, 200 Lothrop Street, Pittsburgh, PA 15213, USA. E-mail address: [email protected] (P.D. Thirumala).
post-operative strokes in the first 10 days following AVR making it important for us to study the periprocedural causes of neurological events in the hospital [6–8]. Beyond the first post-operative week the surgery or its associated risks were no longer implicated to contribute to strokes [6]. Previous studies have shown history of cerebrovascular disease or carotid stenosis (CS) with prior stroke to be the strongest independent risk factor of perioperative stroke following cardiac surgery [9,10]. The common risk factors contributing to perioperative strokes following SAVR include: pre-operatively – age >65 years, female sex, history of cerebrovascular disease (stroke/transient ischemic attacks), cardiac failure, atrial fibrillation, diabetes, hypertension, prior cardiac surgery and endocarditis, etc. [3,9,10]. Intra-operatively, cerebral hypoperfusion or embolic phenomena [4] and post-operatively, new onset atrial fibrillation and low cardiac output syndrome were found culpable [10]. Although several studies have examined the relationship of carotid stenosis with stroke risk following coronary artery bypass
http://dx.doi.org/10.1016/j.jocn.2017.04.006 0967-5868/Ó 2017 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Udesh R et al. Carotid artery disease and perioperative stroke risk after surgical aortic valve replacement: A nationwide inpatient sample analysis. J Clin Neurosci (2017), http://dx.doi.org/10.1016/j.jocn.2017.04.006
2
R. Udesh et al. / Journal of Clinical Neuroscience xxx (2017) xxx–xxx
surgery and cardiac surgeryin general [9,10], a large scale risk analysis of perioperative stroke following aortic valve intervention has not been done before. Some studies also implicate intracranial artery stenosis and vertebral artery stenosis in the etiology of perioperative strokes [11,12]. The primary aim of our study is to identify whether carotid artery disease is an independent risk factor for perioperative stroke following SAVR. We will also evaluate vertebral, basilar and intracranial stenosis and occlusion as a risk factor for perioperative stroke following SAVR. Finally we will evaluate the trends in perioperative stroke and its relationship with carotid disease over a ten year period. Identifying the significance of carotid artery disease in the incidence of perioperative stroke will help us formulate preoperative screening and risk-stratification of patients on the need for carotid revascularization before or after SAVR. Further the need for intra-operative neurophysiological monitoring to cerebral hypoperfusion [13–15] or intense medical management for intracranial stenosis can be evaluated [16]. 2. Methods The National Inpatient Sample (NIS) database was used to select the study population over a period of 11 years from 1999 to 2011 for SAVR. International Classification of Diseases, Ninth RevisionClinical Modification (ICD-9-CM) codes were used to identify diagnoses and risk factors in the NIS database. Patients were identified using ICD-9 codes 35.21 and 35.22 for SAVR and extracted for our analysis. Patients below the age of 18 years were excluded from our study. Baseline characteristics such as age, gender, race, admission status and comorbid conditions commonly observed in the patient demographic undergoing SAVR were studied for our analysis. The NIS database provides 29 Elixhauser comorbidities based on standard ICD-9 codes [17]. Other additional risk factors studied were carotid stenosis, cerebral arterial occlusion, vertebral and basilar artery stenosis, atrial fibrillation, infective endocarditis, left ventricular dysfunction (LVD), previous history of stroke or transient ischemic attack (TIA), previous history of any cardiac surgery and use of cardioplumonary bypass. The ICD-9 codes used to identify the covariates and outcomes are listed in supplement Table 1. The primary outcome studied was perioperative stroke following SAVR. Carotid stenosis was used as the independent variable. Risk stratification was done using Van Walraven (VWR) score. It is a summary score developed for Elixhauser Comorbidities by modeling in-hospital mortality with inpatient admission data [18]. The summary score is a weighted combination of the 29 Elixhauser comorbidities, where a larger comorbidity weight indicates a stronger association between comorbidity and in-hospital mortality. 3. Statistical analysis All data are presented as mean ± SD or percentages. Univariate comparisons between groups were done using unpaired t-tests for continuous variables and a survey-adjusted Wald test for variables that were categorical in nature. In our multivariable analysis, we included variables that were statistically significant in the univariate analysis, that had alarge enough group population, that lacked missing data, and were established as risk factors in previous studies. We elected to exclude each individual Elixhauser Comorbidity and use the van Walraven score as a surrogate, as we did not want to risk overloading our model with variables that may have been either statistically insignificant or not relevant to our outcome of interest. Data extraction was performed using SAS 9.3 (SAS Institute, Inc., Cary, NC), as was the creation of the Elixhauser comorbidity index and the generation of the van Wal-
raven score [19]. All subsequent statistical analyses were performed using Stata version.14 (StataCorp, College Station, TX). 4. Results 4.1. Baseline characteristics The study population comprised of 50,979 patients who underwent surgical aortic valve replacement from 1999 to 2011. The baseline data of the study cohort and univariate analysis of risk factors is provided in Table 1. The mean age of the study cohort was 60.58, with majority of the population (58.13%) under the age of 65. The study patients were predominantly Caucasian (79.3%) and males (60.01%). The perioperative stroke incidence was found to be the same irrespective of gender. The procedure was done electively in 68.47% of the cases. The incidence of perioperative stroke was 2.48% in our study. The patients were stratified based on their Van Walraven (VWR) scores into low/intermediate and high risk groups. The average VWR score was found to be 2.15 with 75.2% of the study patients categorized as low risk. 4.2. Univariate analysis of pre-operative predictors for perioperative strokes Univariate analysis of risk factors (see Tables 1 and 2) suggests the following significant predictors of perioperative stroke (p < 0.05)-advancing age, higher VWR scores, males, African American race, elective procedures, CS, cerebral arterial occlusion, infective endocarditis, CHF, valve disease, cardiopulmonary bypass use, CLD, paralysis and other neurological disorders, complicated DM, RF, coagulopathy, weight loss, fluid and electrolyte abnormalities, hypertension and drug abuse. Carotid stenosis and other precerebral artery stenosis was seen in 1.2% and 0.43% of our study patients and were both significant predictors of strokes (OR 2.01, 95%CI 1.3–3.0, p = 0.001 and OR 2.1, 95%CI 1.1–4.1, p = 0.016 respectively). No patients coded for vertebral or basilar artery stenosis and were not included on our analysis. Infective endocarditis, seen in 4.8% of study patients was another significant predictor (OR 4.8, 95%CI 4.1–5.7, p = 0.00). CBP was used in 88.9% of patients and showed a significantly decreased perioperative stroke risk (OR 0.7, 95%CI 0.6–0.9, p = 0.006). A statistically significant stroke risk was seen in patients with higher VWR scores (p < 0.05). 4.3. Independent predictors of perioperative strokes by multivariate analysis The following risk factors were found to be significant predictors of perioperative strokes (p < 0.05) – age, higher VWR scores, CS, cerebral arterial occlusion, infective endocarditis, DM, HTN, renal failure, neurological disorders, coagulopathy and hypothyroidsm (Table 3). Carotid stenosis (OR 1.8, 95%CI 1.1–2.8, p = 0.009) and cerebral arterial occlusion (OR 3.4, 95% CI 1.3–8.9) were significantly associated with an increased perioperative stroke risk following SAVR. Infective endocarditis (OR 4.6, 95%CI 3.8–5.6, p = 0.00) and neurological disorders (OR 4.8, 95% CI 4– 5.8, p = 0.00) appeared to be the strongest risk factors for strokes. 4.4. Trend analysis An analysis of the trends in perioperative stroke incidence and the average age and VWR scores of patients who underwent SAVR from 1999 to 2011 is depicted in Table 4. The average age of patients undergoing the procedure has remained the same over the years (61.4 years). The VWR score shows an increasing trend over the years from an average score of 1.4 in 1999 to 2.5 in
Please cite this article in press as: Udesh R et al. Carotid artery disease and perioperative stroke risk after surgical aortic valve replacement: A nationwide inpatient sample analysis. J Clin Neurosci (2017), http://dx.doi.org/10.1016/j.jocn.2017.04.006
3
R. Udesh et al. / Journal of Clinical Neuroscience xxx (2017) xxx–xxx Table 1 Baseline characteristics and univariate analysis of pre-operative risk factors. Variables
% of Patients
Perioperative strokes Yes% (n)
No% (n)
Unadjusted Odds Ratio (95% CI)
p-value
Age – M ± SD <65 65–74 75–84 >85 Gender Female Male Race/Ethnicity White African American Hispanic Asian Native American Other/Missing Admission Status Emergent Urgent Elective Risk Category Average Van Walraven Score Low Risk (VWR < 5) Moderate Risk (VWR 5–14) High Risk (VWR > 14) Outcome 1. Perioperative stroke Pre-operative Risk Factors 2. Carotid Stenosis 3. Precerebral artery stenosis (other) 4. Cerebral occlusion 5. Vertebral artery stenosis 6. Basilar artery stenosis 7. LVD and heart failure 8. Atrial Fibrillation 9. Previous h/o stroke or transient ischemic attack (TIA) 10. Infective endocarditis 11. Previous h/o myocardial infarction or angina 12. Previous cardiac surgery 13. Previous CEA/CAS 14. Previous h/o Coronary artery disease or CABG 15. Cardiopulmonary bypass
60.58 ± 0.39 58.13% 22.88% 18.24% 2.16%
2.07% 2.55% 3.29% 4.70%
97.93% 97.45% 96.71% 95.30%
0.671 1.034 1.438 1.974
(0.592–0.761) (0.892–1.200) (1.237–1.671) (1.439–2.707)
0.000 0.649 0.000 0.000
39.98% 60.02%
2.61% (457) 2.41% (635)
97.39% (17,059) 97.59% (25,632)
1.086 (0.954–1.237) 0.664 (0.649–0.681)
0.210 0.000
79.33% 6.91% 8.13% 1.95% 0.53% 3.12%
2.39% 3.47% 2.18% 2.54% 1.23% 3.05%
97.61% 96.53% 97.82% 97.46% 98.77% 96.95%
NA 1.463 0.910 1.063 0.507 1.283
(1.143–1.874) (0.692–1.197) (0.679–1.664) (0.131–1.958) (0.912–1.805)
0.003 0.501 0.788 0.324 0.151
14.92% 16.57% 68.47%
4.42% (258) 3.48% (225) 1.79% (478)
95.58% (5,537) 96.52% (6,232) 98.21% (26,150)
0.779 (0.634–0.958) 0.395 (0.335–0.465)
0.018 0.000
2.15 ± 0.08 75.22% 23.12% 1.64%
1.43% (471) 5.22% (528) 12.42% (92)
98.57% (32,461) 94.78% (9,602) 87.58% (631)
NA 3.805 (3.350–4.321) 9.803 (7.647–12.566)
0.000 0.000
4.84% (25) 5.27% (10) 15.93% (8) 31.46% (1) 32.13% (1) 2.77% (74) 2.67% (402) 1.52% (8) 9.51% (206) 2.27% (205) 1.72% (6) 1.29% (4) 2.48% (46) 2.41% (942)
95.16% (508) 94.73% (181) 84.07% (41) 68.54% (2) 67.87% (2) 97.23% (2,615) 97.33% (14,672) 98.48% (530) 90.49% (1,929) 97.73% (8,897) 98.285 (321) 98.71% (313) 97.52% (1,811) 97.59% (37,971)
2.018 2.194 7.482 18.02 18.58 1.124 1.121 0.603 4.838 0.889 0.683 0.512 0.998 0.780
0.001 0.016 0.000 0.018 0.017 0.358 0.064 0.151 0.000 0.139 0.354 0.181 0.992 0.006
(533) (254) (260) (44)
(635) (83) (60) (16) (2) (33)
(24,921) (9,782) (7,724) (900)
(25,882) (2,253) (2,688) (633) (171) (1,007)
2.48% 1.21% 0.43% 0.11% 0% 0% 6.09% 34.43% 1.21% 4.85% 20.75% 0.74% 0.72% 4.24% 88.91%
(1.338–3.043) (1.159–4.153) (3.456–16.199) (1.63–198.71) (1.67–205.99) (0.875–1.444) (0.993–1.266) (0.302–1.202) (4.096–5.715) (0.761–1.038) (0.306–1.527) (0.192–1.363) (0.736–1.354) (0.654–0.931)
LVD – Left ventricular dysfunction; CEA – carotid endarterectomy; CAS – Carotid artery stenting; CABG – Coronary artery bypass grafting; CI – confidence interval.
2011 (Fig. 1). The perioperative stroke rates have remained more or less in the 2–2.7% range over this decade. When stratified for age and gender, stroke rates have increased over the years in males and patients < 65 years with a decline in stroke incidence in the 65–84 age group. The stroke risk remains the same in females and patients >85 years (Figs. 2 and 3).
5. Discussion Analysis of baseline demographics shows advancing age and increased patient risk profile (VWR scores) to be significant predictors of stroke incidence as with patients undergoing any major cardiac surgery [1,20,21]. Majority of the study patients who underwent SAVR were <65 years (58.1%) and males (60%). Our trend analysis shows a significant increase in stroke rates in patients <65 years which suggest the need for better overall patient selection and assessment of comorbidities. Although prior studies show female gender to be a risk factor for stroke [20,22] our multivariate analysis did not reveal gender to be an independent predictor. The number of patients categorized as medium risk (VWR 5–14) and high risk (VWR > 14) have increased from 1999 to 2011 and our multivariate analysis shows a statistically significant increase in stroke risk with greater VWR scores. Importantly, race
and admission status were not found to be independent predictors of strokes. Carotid stenosis (OR 1.8, 95%CI 1.1–2.8, p = 0.009) and cerebral arterial occlusion (OR 3.4, 95% CI 1.3–8.9) were significantly associated with an increased perioperative stroke risk following SAVR. Cerebrovascular disease or CS are known risk factors of perioperative strokes following cardiac surgery and despite previous studies ruling out pre-operative carotid screening [23] or intervention [8] before cardiac procedures it still remains an important concern with significant perioperative stroke risk. Infective endocarditis (OR 4.6, 95%CI 3.8–5.6, p = 0.00) and previous neurological dysfunction (OR 4.8, 95% CI 4–5.8, p = 0.00) also appeared to be the strong risk factors for strokes. Infective endocarditis is associated with a 12–40% stroke risk due to cerebral septic embolism without the added risk of valve surgery [24]. Several studies show preoperative cerebral imaging with MRI to rule out silent ischemic events and adequate perioperative anticoagulation to reduce perioperative stroke risk [24]. SAMMPRIS [25] and CAVATAS [26] trials signify the role of vertebral, basilar and other intracranial artery stenosis on perioperative strokes. However, we were unable to analyze these risk factors due to inadequate patients being coded for them. Prolonged cardiopulmonary bypass time has been reported to be an important risk factor for stroke during SAVR [9,27]. Our analysis did not reveal the use of CBP to be an
Please cite this article in press as: Udesh R et al. Carotid artery disease and perioperative stroke risk after surgical aortic valve replacement: A nationwide inpatient sample analysis. J Clin Neurosci (2017), http://dx.doi.org/10.1016/j.jocn.2017.04.006
4
R. Udesh et al. / Journal of Clinical Neuroscience xxx (2017) xxx–xxx
Table 2 Univariate analysis of Elixhauser comorbidities. Variables-Elixhauser comorbidities
1. 2. 3. 4. 5. 6. 7. 8. 9.
% of Patients
Valve disease CHF Pulmonary circulation disorders Peripheral vascular disease Paralysis Other Neurological disorders Chronic pulmonary disorders DM, uncomplicated DM, complicated 10. Hypothyroidsm 11. Renal failure 12. Liver failure 13. Peptic Ulcer Disease 14. AIDS 15. Lymphoma 16. Metastatic cancer 17. Solid tumor without metastasis 18. Collagen vascular disorders 19. Coagulopathy 20. Obesity 21. Weight loss 22. Fluid and electrolyte disorders 23. Chronic blood loss anemia 24. Nutritional anemia 25. Alcohol abuse 26. Drug abuse 27. Psychoses 28. Depression 29. Hypertension
0.79% 0.90% 0.20% 13.89% 0.97% 2.75% 16.94% 15.17% 2.07% 7.67% 6.25% 1.36% 0.33% 0.08% 0.52% 0.12% 1.82% 2.21% 13.32% 10.02% 1.77% 17.55% 1.19% 13.17% 2.09% 1.45% 1.27% 4.13% 5.01%
Perioperative strokes Yes% (n)
No% (n)
Unadjusted Odds Ratio (95% CI)
p-value
12.65% (44) 10.57% (44) 8.99% (8) 2.44% (149) 48.38% (212) 16.90% (204) 2.33% (174) 2.17% (144) 4.78% (42) 1.74% (58) 3.87% (108) 3.04% (19) 0.44% (1) 0% 1.40% (3) 3.88% (2) 1.82% (14) 1.94% (19) 3.75% (218) 1.91% (85) 9.67% (76) 3.79% (294) 3% (16) 2.19% (130) 2.86% (27) 4.46% (29) 2.32% (13) 1.68% (32) 2.19% (479)
87.35% (299) 89.43% (348) 91.01% (78) 97.56% (5,919) 51.62% (222) 83.10% (998) 97.67% (7,260) 97.83% (6,534) 95.22% (859) 98.26% (3,303) 96.13% (2,637) 96.96% (578) 99.56% (149) 100% (38) 98.60% (223) 96.12% (52) 98.18% (781) 98.06% (947) 96.25% (5,630) 98.09% (4,323) 90.33% (704) 96.21% (7,413) 97% (510) 97.81% (5,658) 97.14% (890) 95.54% (609) 97.68% (546) 98.32% (1,778) 97.81% (21,478)
5.882 4.786 3.897 0.980 45.25 9.594 0.926 0.850 2.013 0.678 1.644 1.234 0.171 NA 0.554 1.587 0.726 0.772 1.663 0.744 4.440 1.747 1.216 0.865 NA 1.855 0.931 0.659 0.784
(4.291–8.063) (3.332–6.875) (1.939–7.833) (0.822–1.169) (36.48–56.12) (8.105–11.356) (0.794–1.080) (0.709–1.019) (1.481–2.736) (0.512–0.899) (1.34–2.017) (0.772–1.971) (0.024–1.180)
0.000 0.000 0.000 0.829 0.000 0.000 0.331 0.080 0.000 0.007 0.000 0.378 0.073
(0.187–1.642) (0.387–6.497) (0.431–1.220) (0.481–1.240) (1.416–1.954) (0.595–0.931) (3.513–5.612 (1.510–2.020) (0.724–2.044) (0.711–1.052)
0.287 0.521 0.227 0.285 0.000 0.010 0.000 0.000 0.459 0.148
(1.272–2.706) (0.537–1.616) (0.462–0.940) (0.692–0.888)
0.001 0.802 0.021 0.000
CHF – Congestive Heart Failure; DM – diabetes mellitus; CI – confidence interval.
Table 3 Multivariate analysis of the predictors of perioperative stroke following aortic valve surgery. Variables
Age VWR Score Low Risk (VWR < 5) Moderate Risk (VWR 5–14) High Risk (VWR > 14) Pre-operative Risk Factors 1. Carotid Stenosis 2. Cerebral Occlusion 3. Infective Endocarditis 4. Cardiopulmonary Bypass Comorbidities 5. Diabetes-complicated 6. Renal Failure 7. Hypertension 8. Neurological Disorders 9. Coagulopathy 10. Hypothyroidsm
Perioperative strokes Odds Ratio (95% CI)
p-value
1.017 (1.012–1.022)
0.000
NA 2.747 (2.363–3.194) 5.157 (3.779–7.037)
0.000 0.000
1.810 3.429 4.652 0.907
(1.159–2.825) (1.323–8.884) (3.875–5.584) (0.758–1.085)
0.009 0.011 0.000 0.288
1.596 0.671 0.813 4.825 1.003 0.629
(1.158–2.202) (0.530–0.850) (0.712–0.927) (4.003–5.816) (0.841–1.188) (0.472–0.838)
0.004 0.001 0.002 0.000 0.000 0.002
Table 4 Trend analysis of perioperative strokes and patient risk scores over the years (1999– 2011). Calendar year
% of Perioperative strokes(n)
Average Van Walraven Score
Average Age
1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 Total
2.55 2.47 2.38 2.96 2.20 2.04 2.85 2.19 2.13 2.74 2.50 2.84 2.72 2.49
1.495 1.469 1.679 1.817 1.798 1.970 1.822 2.329 2.340 2.326 2.735 2.876 2.511
61.40 60.81 61.24 61.24 60.11 59.97 58.95 59.70 60.57 59.87 60.91 61.12 61.44
(86) (97) (87) (93) (83) (66) (94) (69) (61) (83) (86) (87) (100) (1092)
VWR – Von Walraven score; CI – Confidence interval.
independent predictor for stroke as we were unable to determine the duration of its use from the data available. Previous studies show the routine post-operative neurological evaluation of patients with MRI revealed a 38–54% silent stroke rate/TIA rate [28,29]. This subclinical stroke burden which remains grossly underreported raises the immediate need to study prevention strategies and cerebral protection devices [30]. Transcranial Doppler monitoring and DW-MRI have been shown to be efficacious in detecting these perioperative silent ischemic events [29,31,32]. The role of intra-operative neuromonitoring in detecting and preventing strokes during cardiac procedures has remained largely unexamined and requires further attention.
Fig. 1. Risk Stratification of Patients Undergoing SAVR based on VWR Score from 1999 to 2011.
Please cite this article in press as: Udesh R et al. Carotid artery disease and perioperative stroke risk after surgical aortic valve replacement: A nationwide inpatient sample analysis. J Clin Neurosci (2017), http://dx.doi.org/10.1016/j.jocn.2017.04.006
R. Udesh et al. / Journal of Clinical Neuroscience xxx (2017) xxx–xxx
5
rial occlusion, infective endocarditis, DM, HTN, renal failure, neurological disorders, coagulopathy and hypothyroidsm. The trends over the period of 10 years from 1999 to 2011 indicate that perioperative stroke risk has remained more or less constant despite advancements in surgical techniques with risk having gone up in patients <65 years of age. CS and cerebral arterial occlusion significantly increase perioperative stroke risk following SAVR. Improved patient selection with pre-operative risk stratification and institution of preventive strategies are necessary to improve operative outcomes following SAVR. Conflicts of interest None. Fig. 2. Incidence of Perioperative Strokes following SAVR in Males and Females from 1999 to 2011.
Sources of funding None. Acknowledgements None. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.jocn.2017.04.006. References
Fig. 3. Incidence of Perioperative Strokes stratified by Patient Age from 1999 to 2011.
Pre-operative carotid Doppler screening of CS is being studied increasingly for patients undergoing transcatheter aortic valve implantation [33], however; its utility in candidates for SAVR has not been examined comprehensively [23]. Further studies examining the role of standardized perioperative neurological evaluation and other preventive strategies will give us valuable insight into reducing perioperative stroke risk following SAVR. Study limitations Despite the valuable information gained from this large scale risk analysis of perioperative strokes on 50,000 patients our study is not without limitations. As the data is pooled from different hospitals and institutions there could be considerable variability in pre-operative patient selection and management which could have affected the outcome rates. Being an administrative database, the NIS relies heavily on correct coding of diagnoses and procedures and any coding errors or inconsistencies could result in inaccurate data. Inadequate coding of patients with vertebral/basilar artery disease which are important risk factors for strokes could be an indication of patients not being screened for them or a coding error. Information on strokes which could have occurred after the patients’ hospital discharge but within the 30-day post-operative limit and post-operative management of stroke patients is unavailable. Conclusion Risk factors found to be significant predictors of perioperative strokes (p < 0.05) were – age, higher VWR scores, CS, cerebral arte-
[1] Agarwal S, Garg A, Parashar A, et al. In-hospital mortality and stroke after surgical aortic valve replacement: A nationwide perspective. J Thorac Cardiovasc Surg 2015;150(571–8):e8. [2] Biancari F, Martin M, Bordin G, et al. Basic data from 176 studies on the immediate outcome after aortic valve replacement with or without coronary artery bypass surgery. J Cardiothorac Vasc Anesth 2014;28:1251–6. [3] Almassi GH, Sommers T, Moritz TE, et al. Stroke in cardiac surgical patients: determinants and outcome. Ann Thorac Surg 1999;68:391–7 [discussion 7–8]. [4] Wolman RL, Nussmeier NA, Aggarwal A, et al. Cerebral injury after cardiac surgery: identification of a group at extraordinary risk. Multicenter Study of Perioperative Ischemia Research Group (McSPI) and the Ischemia Research Education Foundation (IREF) Investigators. Stroke 1999;30:514–22. [5] Society of Thoracic Surgeons Adult Cardiac Surgery Database-Executive Summary. Available at: http://wwwstsorg/sites/default/files/documents/ 2015Harvest4ExecutiveSummarypdf. [6] Miller DC, Blackstone EH, Mack MJ, et al. Transcatheter (TAVR) versus surgical (AVR) aortic valve replacement: occurrence, hazard, risk factors, and consequences of neurologic events in the PARTNER trial. J Thorac Cardiovasc Surg 2012;143(832–43):e13. [7] Filsoufi F, Rahmanian PB, Castillo JG, et al. Incidence, imaging analysis, and early and late outcomes of stroke after cardiac valve operation. Am J Cardiol 2008;101:1472–8. [8] Li Y, Walicki D, Mathiesen C, et al. Strokes after cardiac surgery and relationship to carotid stenosis. Arch Neurol 2009;66:1091–6. [9] Bucerius J, Gummert JF, Borger MA, et al. Stroke after cardiac surgery: a risk factor analysis of 16,184 consecutive adult patients. Ann Thorac Surg 2003;75:472–8. [10] Hogue Jr CW, De Wet CJ, Schechtman KB, et al. The importance of prior stroke for the adjusted risk of neurologic injury after cardiac surgery for women and men. Anesthesiology 2003;98:823–9. [11] Coward LJ, McCabe DJ, Ederle J, et al. Long-term outcome after angioplasty and stenting for symptomatic vertebral artery stenosis compared with medical treatment in the Carotid And Vertebral Artery Transluminal Angioplasty Study (CAVATAS): a randomized trial. Stroke 2007;38:1526–30. [12] Derdeyn CP, Fiorella D, Lynn MJ, et al. Mechanisms of stroke after intracranial angioplasty and stenting in the SAMMPRIS trial. Neurosurgery 2013;72:777–95 [discussion 95]. [13] Nwachuku EL, Balzer JR, Yabes JG, et al. Diagnostic value of somatosensory evoked potential changes during carotid endarterectomy: a systematic review and meta-analysis. JAMA Neurol 2015;72:73–80. [14] Thiagarajan K, Cheng HL, Huang JE, et al. Is two really better than one? Examining the superiority of dual modality neurophysiological monitoring during carotid endarterectomy: A meta-analysis. World Neurosurg 2015;84 (1941–9):e1. [15] Thirumala PD, Udesh R, Muralidharan A, et al. Diagnostic value of somatosensory-evoked potential monitoring during cerebral aneurysm
Please cite this article in press as: Udesh R et al. Carotid artery disease and perioperative stroke risk after surgical aortic valve replacement: A nationwide inpatient sample analysis. J Clin Neurosci (2017), http://dx.doi.org/10.1016/j.jocn.2017.04.006
6
[16]
[17] [18]
[19]
[20]
[21]
[22]
[23]
[24]
R. Udesh et al. / Journal of Clinical Neuroscience xxx (2017) xxx–xxx clipping: a systematic review. World Neurosurg 2016;89:672–80. http://dx. doi.org/10.1016/j.wneu.2015.12.008. Derdeyn CP, Chimowitz MI, Lynn MJ, Fiorella D, Turan TN, Janis LS, et al. Aggressive medical treatment with or without stenting in high-risk patients with intracranial artery stenosis (SAMMPRIS): the final results of a randomised trial. Lancet 2014;383:333–41. http://dx.doi.org/10.1016/S0140-6736(13) 62038-3. Elixhauser A, Steiner C, Harris DR, et al. Comorbidity measures for use with administrative data. Med Care 1998;36:8–27. van Walraven C, Austin PC, Jennings A, et al. A modification of the Elixhauser comorbidity measures into a point system for hospital death using administrative data. Med Care 2009;47:626–33. Databases HCUP. Healthcare Cost and Utilization Project (HCUP). Rockville, MD: Agency for Healthcare Research and Quality; 2003. www.hcup-us. ahrq.gov/nisoverview.jsp. Idrees JJ, Schiltz NK, Johnston DR, et al. Trends, predictors, and outcomes of stroke after surgical aortic valve replacement in the United States. Ann Thorac Surg 2016;101:927–35. Mortasawi A, Arnrich B, Rosendahl U, et al. Is age an independent determinant of mortality in cardiac surgery as suggested by the EuroSCORE? BMC Surg 2002;2:8. Hogue Jr CW, Barzilai B, Pieper KS, et al. Sex differences in neurological outcomes and mortality after cardiac surgery: a society of thoracic surgery national database report. Circulation 2001;103:2133–7. Condado JF, Jensen HA, Maini A, et al. Should we perform carotid doppler screening before surgical or transcatheter aortic valve replacement? Ann Thorac Surg 2016. http://dx.doi.org/10.1016/j.athoracsur.2016.06.076. pii: S0003-4975(16)30792-5. Behrouz R. Preoperative cerebrovascular evaluation in patients with infective endocarditis. Clin Cardiol 2015;38:439–42.
[25] Kwon HM, Lynn MJ, Turan TN, et al. Frequency, risk factors, and outcome of coexistent small vessel disease and intracranial arterial stenosis: results from the stenting and Aggressive Medical Management for Preventing Recurrent Stroke in Intracranial Stenosis (SAMMPRIS) trial. JAMA Neurol 2016;73:36–42. [26] Ederle J, Bonati LH, Dobson J, et al. Endovascular treatment with angioplasty or stenting versus endarterectomy in patients with carotid artery stenosis in the Carotid And Vertebral Artery Transluminal Angioplasty Study (CAVATAS): long-term follow-up of a randomised trial. Lancet Neurol 2009;8:898–907. [27] Brown WR, Moody DM, Challa VR, et al. Longer duration of cardiopulmonary bypass is associated with greater numbers of cerebral microemboli. Stroke 2000;31:707–13. [28] Messe SR, Acker MA, Kasner SE, et al. Stroke after aortic valve surgery: results from a prospective cohort. Circulation 2014;129:2253–61. [29] Stolz E, Gerriets T, Kluge A, et al. Diffusion-weighted magnetic resonance imaging and neurobiochemical markers after aortic valve replacement: implications for future neuroprotective trials? Stroke 2004;35:888–92. [30] Floyd TF, Shah PN, Price CC, et al. Clinically silent cerebral ischemic events after cardiac surgery: their incidence, regional vascular occurrence, and procedural dependence. Ann Thorac Surg 2006;81:2160–6. [31] Knipp SC, Matatko N, Schlamann M, et al. Small ischemic brain lesions after cardiac valve replacement detected by diffusion-weighted magnetic resonance imaging: relation to neurocognitive function. Eur J Cradiothorac Surg 2005;28:88–96. [32] Spaziano M, Francese DP, Leon MB, et al. Imaging and functional testing to assess clinical and subclinical neurological events after transcatheter or surgical aortic valve replacement: a comprehensive review. J Am Coll Cardiol 2014;64:1950–63. [33] Ravindra VM, Mazur MD, Kumpati GS, et al. Carotid artery stenosis in the setting of transcatheter aortic valve replacement: clinical and technical considerations of carotid stenting. World Neruosurg 2016;86:194–8.
Please cite this article in press as: Udesh R et al. Carotid artery disease and perioperative stroke risk after surgical aortic valve replacement: A nationwide inpatient sample analysis. J Clin Neurosci (2017), http://dx.doi.org/10.1016/j.jocn.2017.04.006