Perceptions of Perioperative Stroke Among Chinese Anesthesiologists: Starting a Long March to Eliminate This Underappreciated Complication

In <18 months since the opening of a new teaching hospital in eastern China, 4 patients experienced a stroke after noncardiac, nonneurologic surgery (Table 1). This equates to a 0.03% overall incidence of perioperative stroke based on a total of 13,335 surgical cases performed during this period of time. In all of these cases, the stroke was ischemic in origin and classified as partial anterior circulation infarct.1–3 With patient consent, the imaging studies of 1 patient who experienced a massive left hemispheric stroke are presented (Figure 1).Table 1.: Four Patients Diagnosed With Perioperative Stroke After Noncardiac and Nonneurologic Surgery Within 18 Months of Opening in a New Hospital in Eastern ChinaFigure 1.: A 65-year-old man underwent right total hip arthroplasty under general anesthesia. CT scan 24 h after surgery showed massive cerebral infarction in the left middle cerebral artery territory (A). CT angiography showed multiple cerebrovascular atherosclerotic changes, including moderate stenosis at the origin of right internal carotid artery (white arrow), occlusion at the origin of the left internal carotid artery (yellow arrow), and segmental occlusion of the left vertebral artery (red arrow) (B). CT perfusion imaging showed reduced blood flow in the left middle cerebral artery territory (C) and increased mean transit time in the same region (D). CT indicates computed tomography; PACS, partial anterior circulation stroke syndrome.Perioperative stroke is an uncommon but devastating complication that occurs within 30 days after surgery, with surgery and anesthesia being independent risk factors for stroke.4 The incidence, risk factors, and outcomes of perioperative stroke have been investigated in Caucasian populations based on large databases5–12; however, the literature on perioperative stroke in the Chinese population is scarce and primarily based on the experience of a single hospital.13,14 Given the recent cases of perioperative stroke in this newly opened Chinese hospital, the authors were inspired to look further into the issue of perioperative stroke in China. STROKE IN THE NONPERIOPERATIVE SETTING IN CHINESE VERSUS NON-CHINESE POPULATIONS Compared to perioperative stroke, considerably more research has been conducted to understand the prevalence, incidence, mortality rate, and risk factors in Chinese versus non-Chinese populations in the nonperioperative setting. The age-standardized prevalence and incidence of stroke in China are 1115 per 100,000 people and 247 per 100,000 person-years, respectively.15 Compared with Caucasian populations, the incidence of age-standardized annual first-ever stroke appears higher among Chinese populations between 45 and 74 years of age (range, 205–584 vs 170–335 per 100,000).16 The estimated proportion of hemorrhagic strokes is also higher in Chinese than in Caucasian populations based on community-based studies (28.0% vs 12.4%).16 In hospital-based studies, Chinese populations also appear to have a higher proportion of small-vessel stroke (33.1% vs 19.3%) and atherothrombotic stroke (25.4% vs 14.7%) although cardioembolic stroke risk is lower in Chinese populations (15.8% vs 25.7%).16 Stroke is the leading cause of death in China, with an age-standardized mortality rate of 115 per 100,000 person-years for total stroke (57 for ischemic stroke and 56 for hemorrhagic stroke).17 In the rest of the world, stroke is the second-leading cause of death, with an age-standardized mortality rate of 86 per 100,000 person-years for total stroke (43 for ischemic stroke and 43 for hemorrhagic stroke).18 The differences in the epidemiology of stroke in China may be explained by baseline differences in the prevalence of risk factors. The prevalence of atrial fibrillation, ischemic heart disease, and hypercholesterolemia is significantly lower among Chinese than Caucasian ischemic stroke patients, while the prevalence of hypertension, diabetes, smoking, and alcohol use is similar.19 In China, ischemic stroke patients are also younger and are predominantly male.19 SURVEY OF CHINESE ANESTHESIOLOGISTS Given the devastating consequences of perioperative stroke and the high burden of stroke generally in China, identifying the perceptions and misconceptions regarding perioperative stroke in China would be valuable. Therefore, we conducted a survey of perioperative stroke among Chinese anesthesiologists between July 16, 2017 and August 15, 2017 (30 days). This survey was endorsed by the Committee of Geriatric Anesthesia of the Chinese Society of Anesthesiology, jointly formalized by 4 anesthesiologists (L.M., C.T., T.W., and D.L.M.), and consisted of 24 questions (Supplemental Digital Content, Table, https://links.lww.com/AA/C508). Pilot testing was done with anesthesiologists to improve the flow and ease of reading. The survey was distributed in Chinese in formats compatible with the platforms of WeChat (a Chinese multipurpose messaging and social media app) and New Youth Anesthesia (the largest, web-based anesthesia forum in China). Consent for participating this survey was not required. A total of 5147 anesthesiologists completed the survey, giving an approximate survey response rate of 9.8% based on an estimated 53,000 practicing anesthesiologists in China.20 The survey response rate may be higher because it is not clear whether all anesthesiologists practicing in China received this survey. SURVEY RESULTS: PERCEPTIONS OF PERIOPERATIVE STROKE AMONG CHINESE ANESTHESIOLOGISTS Details of the responses are presented in Supplemental Digital Content, Table, https://links.lww.com/AA/C508. The survey showed that 86% of respondents had either experienced or were aware of at least 1 case of perioperative stroke in their hospitals. About one-third (34%) of respondents were unsure about the reported incidence of perioperative stroke in noncardiac, nonneurologic surgeries. The largest proportion of respondents (21%) chose carotid endarterectomy as the type of surgery with the greatest risk of perioperative stroke. History of hypertension, diabetes, stroke, transient ischemic attack, atrial fibrillation, and age were identified as risk factors for perioperative stroke by >80% of respondents. About half (52%) of respondents chose hypoperfusion as the most common cause of perioperative stroke, while 37% chose thromboembolic stroke as the most common etiology. Opinions on strategies to prevent cerebral ischemia included maintenance of cerebral perfusion pressure (97%), maintenance of cardiac output (75%), avoidance of hypocapnia (61%), and avoidance of cerebral vasoconstrictive drugs (50%). Finally, more than two-thirds (69%) of respondents agreed that the survey was relevant to their clinical practice, and the majority (87%) indicated that the survey made them pay more attention to perioperative stroke. THE UNDERAPPRECIATION OF PERIOPERATIVE STROKE Our survey demonstrated that the majority of anesthesiologists in China have anecdotally experienced or heard about cases of perioperative stroke in their practice setting. Despite this level of awareness, we identified several knowledge gaps related to the incidence, risk factors, mechanisms, and prevention of perioperative stroke, suggesting an underappreciation of this complication among Chinese anesthesiologists. The incidence of perioperative stroke varies depending on the surgical population.6,21 The overall incidence in noncardiac, nonneurological surgical patients in the United States is approximately 0.1% based on multiple cohort studies.5,6 In contrast, the incidence of perioperative stroke in China is poorly studied, with only 1 previous published study.13 This study reported an incidence of 0.1% in noncardiac, nonneurological surgical patients based on the experience of a single hospital.13 Although the estimated prevalence of stroke is higher among Chinese versus Caucasian populations in the nonperioperative setting,16 the current literature is not sufficient to determine whether the incidence of perioperative stroke is also higher in Chinese compared to Caucasian populations. Multiple independent predictors of perioperative stoke have been identified by several large epidemiological studies in the United States.22 In contrast, the risk factors for perioperative stroke in China are inadequately studied. One study based on the database of a single hospital identified American Society of Anesthesiologists physical status, history of stroke or transient ischemic attack, diabetes mellitus requiring insulin therapy, carotid artery stenosis, and preoperative anemia as risk factors in Chinese patients.13 Most of the risk factors for perioperative stroke chosen by the survey respondents are consistent with previously published studies from the United States. However, evidence suggests a lower prevalence of atrial fibrillation, a prominent cause of cardioembolic stroke, in Chinese populations.16 Although the survey respondents were overall well informed on stroke risk factors, we identified several knowledge gaps. More respondents chose carotid endarterectomy over stenting (21% vs 3%) as the type of surgery bearing the greatest risk of perioperative stroke, which is not consistent with the literature. Stenting has been shown to carry a higher risk of perioperative stroke than carotid endarterectomy (4.1% vs 2.3%, respectively).23 Although 12% of respondents chose craniotomy as the most common cause of perioperative stroke, the incidence of stroke as a result of cerebrovascular complications unrelated to the cranial surgery per se is unknown. In contrast with the current literature, more than half (52%) of respondents regarded hypoperfusion as the most common cause of perioperative stroke. In cardiac surgery, embolic stroke accounts for 62% of perioperative strokes, hypoperfusion 9%, and thrombosis 1%.21 In noncardiac surgery, embolic stroke remains most common (34%), followed by hypoperfusion (19%), thrombotic (12%), and hemorrhagic (4%) stroke based on pooled data of 106 patients from 5 studies (Table 2). The etiology was unclear in 31% of patients diagnosed with stroke among these studies. Importantly, the majority of these studies were based on Caucasian populations and included a high proportion of patients with atrial fibrillation. Data specific to the Chinese population are lacking.Table 2.: Patients With Different Subtypes of Perioperative Stroke After Noncardiac Surgery Based on Studies With Details of EtiologyNotably, 42% of respondents indicated a desire to use cerebral autoregulation to determine ideal cerebral perfusion pressure. Currently, there is no validated, intraoperative monitor of cerebral autoregulation. More than 50% of respondents chose an absolute blood pressure value as the intervention threshold despite the recent evidence suggesting that individualized blood pressure management based on “relative” change results in more favorable outcomes.28 Half of the respondents chose to avoid cerebral vasoconstrictive drugs despite the fact that most clinically used vasopressors have no direct action on the cerebral vasculature.29,30 This concern could potentially result in avoidance of vasopressors and possible hypotension. Most respondents (59%) chose 10 g/dL as a minimum hemoglobin threshold. However, the current recommended hemoglobin threshold is >9 g/dL in noncardiac and nonneurological surgical patients.22 There is currently no consensus on the optimal cardiac output, inspired oxygen fraction, or arterial blood carbon dioxide level in patients at risk of perioperative stroke, although these modifiable variables may impact outcomes. PERIOPERATIVE STROKE IN CHINA AND BEYOND: THE LONG MARCHFigure 2.: The long march from where we are to where we need to be for perioperative stroke. The risk factors associated with different stages of perioperative care need to be identified and modified. Currently, not every risk factor being identified or speculated has a validated solution (marked in red font and with a question mark). For example, 1 consensus statement recommended avoiding intraoperative hypotension in patients at high risk of perioperative stroke22; however, the explicit BP target for an individual patient, after taking into account the baseline measurement and the risk of organ ischemia versus BP-related hemorrhage, remains to be elucidated. BP indicates blood pressure; Etco 2, end-tidal carbon dioxide.In summary, perioperative stroke is inadequately studied in Chinese populations, and there is a lack of understanding of the differences between Chinese and Caucasian populations. Our survey identifies knowledge gaps and highlights the underappreciation of this complication in China. These findings are consistent with a recent survey conducted in Canadian anesthesiologists,31 and the Society of Neuroscience in Anesthesiology and Critical Care consensus statement on perioperative stroke raised similar concerns.22 Although there is increasing interest in expanding our understanding of perioperative stroke, many steps remain toward reaching this goal (Figure 2). Each stage of the perioperative period needs examination, particularly to identify modifiable risk factors and potential targets for intervention. At this time, evidence on which to base practice recommendations is lacking in some aspects that are deemed clinically relevant. The relative rarity of this complication is an impediment to high-quality clinical studies. Finally, our survey illustrates the need to identify differences in the incidence and risk factors for perioperative stroke among different populations, including the Chinese population. A logical next step would be a Chinese or international registry for perioperative stroke to better define the epidemiology and risk factors, as well as targeted educational initiatives to increase awareness and vigilance among anesthesiologists. DISCLOSURES Name: Lingzhong Meng, MD. Contribution: This author helped design the survey and write the manuscript. Conflicts of Interest: L. Meng is a consultant to CAS Medical Systems, Inc. Name: Jianjun Li, MD. Contribution: This author helped provide the case series and write the manuscript. Conflicts of Interest: None. Name: Alana M. Flexman, MD. Contribution: This author helped write the manuscript. Conflicts of Interest: None. Name: Chuanyao Tong, MD, PhD. Contribution: This author helped design the survey and write the manuscript. Conflicts of Interest: None. Name: Xiangyong Zhou, MD. Contribution: This author helped conduct the survey and write the manuscript. Conflicts of Interest: None. Name: Adrian W. Gelb, MBChB. Contribution: This author helped write the manuscript. Conflicts of Interest: A. W. Gelb is a consultant to Masimo Corporation. Name: Tianlong Wang, MD. Contribution: This author helped conduct the survey and write the manuscript. Conflicts of Interest: None. Name: David L. McDonagh, MD. Contribution: This author helped design the survey and write the manuscript. Conflicts of Interest: D. L. McDonagh has received research funding from Lungpacer Medical Inc. This manuscript was handled by: Tong J. Gan, MD.