Secondary Logo

Journal Logo

Perioperative medicine

Intraoperative hypotension is associated with myocardial damage in noncardiac surgery

An observational study

Hallqvist, Linn; Mårtensson, Johan; Granath, Fredrik; Sahlén, Anders; Bell, Max

Author Information
European Journal of Anaesthesiology: June 2016 - Volume 33 - Issue 6 - p 450-456
doi: 10.1097/EJA.0000000000000429

Abstract

Introduction

The number of patients undergoing major noncardiac surgery is growing continuously1 with advancements in treating disease and improving quality of life. In Sweden, over 760 000 noncardiac surgeries were performed in 2012.2 Perioperative cardiac complications occur more commonly in aging populations.3 Perioperative myocardial damage is the result of disturbances in myocardial perfusion because of an imbalance of oxygen demand and delivery, causing myocardial ischaemia.4 In the vascular events in noncardiac surgery patients cohort evaluation (VISION) study, the diagnostic definition of myocardial injury after noncardiac surgery was a peak of fourth-generation cardiac troponin-T (TnT) of 0.03 ng ml−1 or greater, caused by myocardial ischaemia. The 30-day mortality was 9.8% among patients suffering myocardial injury after noncardiac surgery.5 In older studies, perioperative myocardial infarction (MI) was associated with in-hospital mortality of 15 to 25%.6–9 Defining perioperative MI is difficult because of absence of ischaemic symptoms.10 Most perioperative MIs occur 24 to 48 h after surgery,11 when patients are usually receiving analgesics and sedatives, limiting the ability to recognise and communicate symptoms. Postoperative signs and symptoms, such as hypotension, tachycardia, shortness of breath or nausea, are not specific for myocardial ischaemia and may be misinterpreted as complications such as atelectasis, pneumonia, hypovolemia or medication side-effects.12 Nevertheless, MI, whether symptomatic or asymptomatic, has an equally poor prognosis.11

MI detection relies on rise and fall of biomarkers.13 Cardiac troponins are the biomarkers of choice in diagnosing myocardial damage and MI.14,15 Recent studies imply that elevated postoperative troponin levels are independently associated with increased mortality after noncardiac surgery.5,16,17 Routine measurement of TnT-levels and, with biomarker elevation, monitoring of 12-lead ECG are needed postoperatively to detect perioperative MI and for long-term prognostic information.11,12,16 The new high-sensitivity cardiac troponin-T (hs-cTnT) assay improves risk assessment and identifies more patients with, or at risk of, myocardial damage and new cardiac events.14

An association between intraoperative hypotension and adverse outcomes remain debatable and there is no consensus on the lowest acceptable intraoperative blood pressure. Data indicate a higher mortality rate in the elderly when there is a decrease in SBP of 40 to 45% from the preoperative baseline.18 An overall causal relation between intraoperative hypotension and 1-year mortality has not been demonstrated.18 However, recent reports indicate intraoperative hypotension contributes to troponin elevation19 and is associated with increased 30-day mortality.20

The aim of our study was to investigate how intraoperative events, especially hypotension, were related to perioperative myocardial damage, as detected by high-sensitivity troponin-T (hs-cTnT) assay. Moreover, we examined if these intraoperative events were related to perioperative MI.

Materials and methods

Ethics approval

The study protocol (2014/1306-31/3) was approved by the Regional Ethics Committee of Stockholm (Chairperson E. Lindblad) on 25 September 2014.

Study population and design

This was an observational cohort study of all adult patients undergoing major elective noncardiac surgery who were scheduled for an overnight admission to the postoperative unit at the Karolinska University Hospital, Stockholm, Sweden, between October 2012 and May 2013. Our exclusion criteria were patients undergoing phaeochromocytoma surgery as these patients commonly have high baseline levels of cardiac biomarkers and extreme blood pressure variability not related to the surgical events. The study protocol was approved by the Regional Ethics Committee of Stockholm, which also waived informed consent.

Data collection

Demographic data and medical history were obtained from medical records at the outpatient preoperative evaluation clinic: age, sex, BMI, smoking status, medical history of cardiovascular disease (hypertension, atrial fibrillation, congestive heart failure and ischaemic heart disease; previous MI or coronary intervention/bypass graft), antihypertensive medication [angiotensin-converting enzyme (ACE) inhibitors, β blockers and calcium channel blockers] and insulin-dependent diabetes mellitus. Patients were classified according to the American Society of Anaesthesiologists (ASA) physical status classification.

Perioperative characteristics were collected from electronic medical records. Surgical procedures included gastrointestinal, urological, gynaecological, vascular, head and neck, reconstructive and orthopaedic surgery. The types of anaesthesia were general and/or regional including epidural anaesthesia. The following intraoperative events were recorded by reviewing anaesthetic charts: hypotension (defined as a 50% decrease in SBP relative each patient's baseline lasting >5 min), hypoxemia (SpO2 < 90% for >5 min), tachycardia (increase in heart rate of >30 beats min−1 from baseline for >5 min), blood loss (in ml) and cumulative fluid balance during the first 24 h after surgery. To ensure baseline blood pressures were representative of a patient's habitual values, all blood pressures documented within 2 months prior to surgery were averaged.

Laboratory tests

Blood was obtained for routine laboratory testing on the first postoperative morning on all patients. Levels of hs-cTnT were analysed in serum samples collected 22 h after surgery using the Elecsys 2010 system (Roche Diagnostics GmbH, Mannheim, Germany), which has been available at Karolinska University Hospital since 10 December 2010. This method has a lower detection limit of 2 ng l−1, a 99th-percentile cut-off point of 14 ng l−1, and a coefficient of variation of less than 10% at 13 ng l−1.21 Plasma creatinine was measured the day before and on the first, second and third day after surgery, and levels on the first postoperative day and the highest value within 3 days postoperatively were recorded. Acute kidney injury (AKI) was defined according to the Acute Kidney Injury Network criteria.22

Electrocardiogram

Resting 12-lead ECGs were recorded preoperatively in all patients. Postoperative ECG was performed when clinical findings indicated ongoing myocardial ischaemia.

Outcomes

Incidence of myocardial damage, defined as hs-cTnT more than 14 ng l−1 on the first postoperative day, was evaluated.21 Incidence of perioperative MI and cardiac death within 30 days after surgery were assessed, as were all-cause mortality within 30 days and 6 months.

Perioperative MI was defined according to the joint European Society of Cardiology and American College of Cardiology consensus.23 These criteria include a rise and fall in cardiac troponin levels in association with at least one of the following: ischaemic symptoms, new ECG changes (Q waves, LBBB or altered repolarisation) or regional wall motion abnormalities (or coronary intervention). MI diagnosis was also made post-mortem, based on autopsy findings. All cases were reviewed with independent retrospective adjudication of events by a cardiologist blinded to the hs-cTnT levels measured on postoperative day 1.

Statistical analysis

Data were analysed by STATA version 12.1 (Stata Corp., College Station, Texas, USA) and SAS 9.4; (SAS Institute Inc., Cary, North Carolina, USA). Continuous variables are expressed as median and inter-quartile range and categorical variables as n (%). Associations between baseline and perioperative characteristics and the predefined outcome variables were explored by Mann–Whitney U test and Fisher's exact test for continuous data, and χ2 test and Kruskal–Wallis test for dichotomous variables. In the bivariate analysis, we observed a significant association between a fall in SBP more than 50% from baseline and elevated hs-cTnT. The observed P value for the bivariate association, using an ordinary χ2test, was 0.00002. Even if all 20 preoperative and perioperative factors were considered, a Bonferroni corrected P value would still be highly significant (P = 0.0004), thus unlikely to be a chance finding. Multivariate analyses assessed whether this association was because of confounding from the following factors: as continuous variables: age (categorised as in Table 1), BMI and preoperative creatinine; as dichotomous variables: sex, abnormal ECG, ASA above 2, smoking status, hypertension, congestive heart failure, ischaemic heart disease, atrial fibrillation, antihypertensive medication and insulin-dependent diabetes mellitus. These variables were entered into a logistic regression model with stepwise selection (significance level of 0.05 for both entering and staying in the model). The resulting model yielded essentially unchanged estimates for hypotension and included the following significant adjustment variables: age, preoperative creatinine, abnormal ECG, ASA above 2 and congestive heart failure. Furthermore, the log–linear effect of preoperative creatinine was assessed by categorising it according to quartiles, resulting in a U-shaped effect: elevated risks in first and fourth quartiles [no significant difference between quartile 2 and 3; odds ratio (OR), 0.81]. Preoperative creatinine was therefore included accordingly. The initial model included 15 parameters and a total of 90 events of elevated hs-cTnT. The aim was to assess potential confounding, not to construct a predictive model for elevated hs-cTnT. To investigate whether large but not statistically significant variables were left out from the model, we performed additional analyses where each of the remaining variables was entered one by one into the final model and all yielded ORs close to unity (supplementary Table 3, http://links.lww.com/EJA/A91).

Table 1
Table 1:
Predictors and odds ratios of myocardial damage (high-sensitivity cardiac troponin T >14 ng l−1) in relation to an intraoperative hypotensive eventa

The strong risk factors found in the multivariate model suggest that the risk of an elevated hs-cTnT varies substantially between individuals. To illustrate the effect of hypotension in different strata of underlying risk, we used the significant parameters to create three levels of risk:

  1. Low risk = no risk factor or only abnormal ECG or ASA above 2 or preoperative creatinine less than 59 μmol l−1.
  2. Medium risk = only age more than 70 years and/or preoperative creatinine at least 79 μmol l−1, combinations of two risk factors: ASA above 2 and (preoperative creatinine <59 μmol l−1or >79 μmol l−1or age 70 to 79 years or abnormal ECG) or preoperative creatinine less than 59 μmol l−1and abnormal ECG.
  3. High risk = remaining combinations.

The receiver operating characteristics curve for the association between hs-cTnT and MI was plotted, and the area under the curve was calculated. A P value less than 0.05 was considered statistically significant.

Results

Between October 2012 and May 2013, 307 patients fulfilled the inclusion criteria and completed the 6-month follow-up. Patients with phaeochromocytoma (n = 7) were excluded. Of the final cohort of 300 patients, 90 patients (30%) had myocardial damage on postoperative day 1 and 15 patients (5%) developed MI within 30 days. Two-thirds had their MI within the first 4 postoperative days and the majority (87%) within a week (supplementary Table 4, http://links.lww.com/EJA/A91). The 15 patients with MI had significantly higher hs-cTnT levels, 73% had myocardial damage vs. 28% of the patients who did not have a perioperative MI (P < 0.001). All-cause mortality at 30 days and 6 months was 1.7 and 4%, respectively.

Baseline and perioperative characteristics of patients, with and without myocardial damage, are presented in Table 2. For the entire cohort, average age was 67 years and 53% were women. The most prevalent cardiovascular risk factor was hypertension (43%) and the most common surgery was gastrointestinal surgery (40%), followed by urological (29%) and gynaecological (17%) surgery.

Table 2
Table 2:
Characteristics of the cohort and the proportion of myocardial damage, defined as high-sensitivity cardiac troponin T more than 14 ng l−1 after surgery

Patients with myocardial damage on postoperative day 1 were older with more cardiovascular risk factors than patients without myocardial damage. One-third of patients with myocardial damage had an abnormal preoperative ECG, compared with 8% with normal hs-cTnT levels (P < 0.001). More than twice as many patients with myocardial damage, 38 vs. 17%, had chronic treatment with β blockers (P < 0.001). Moreover, they had more intraoperative adverse events and worse outcome, with significantly more MIs.

In supplementary Table 1, http://links.lww.com/EJA/A91, baseline and perioperative characteristics of patients with and without MI within 30 days are presented. Patients developing perioperative MI were older, more often smokers and had more cardiovascular risk factors. Less than 50% of patients with perioperative MI had a normal preoperative ECG as compared with 87% of patients without MI (P < 0.001). AKI within three postoperative days was more frequent in patients with (60%) than without (21%) perioperative MI (P < 0.001).

Patients with elevated hs-cTnT and patients diagnosed with MI had an intraoperative fall in SBP more than 50% from baseline more frequently. Of all 300 patients, 34 (12%) had that intraoperative hypotensive event, 21 of those had myocardial damage and eight patients were diagnosed with perioperative MI (P < 0.001).

The estimated ‘crude’ OR for the association between an intraoperative hypotensive event and increased hs-cTnT was 4.6 [95% confidence interval (CI), 2.2 to 9.9, not shown in table]. Multivariate logistic regression analysis demonstrated that an intraoperative hypotensive event more than 50% (OR 4.4; 95% CI, 1.8 to11.1), age groups: 60 to 69 years (OR 1.5; 95% CI, 0.6 to 3.7), age 70 to 79 years (OR 3.5; 95% CI, 1.4 to 8.9) and age at least 80 years (OR 8.4; 95% CI, 2.7 to 28.6), preoperative abnormal ECG (OR 2.9; 95% CI, 1.3 to 6.6), ASA classification above 2 (OR 2.4; 95% CI, 1.3 to 4.6), congestive heart failure (OR 8.7; 95% CI, 1.4 to 73.6) and preoperative creatinine less than 59 μmol l−1 (OR 3.1; 95% CI, 1.3 to 7.4) and more than 79 μmol l−1 (OR 3.8; 95% CI, 1.9 to 7.9) were independent predictors of hs-cTnT elevation on postoperative day 1 (Table 1). As the chronic antihypertensive medications shown in Table 2 did not independently predict myocardial damage or change the OR of intraoperative hypotensive events more than 50%, they are not shown in Table 1 [ACE inhibitors (OR 0.9; 95% CI, 0.4 to 2), β blockers (OR 1.0; 95% CI, 0.4 to 2.1), calcium channel blockers (OR 1.1; 95% CI, 0.4 to 2.6)].

To further illustrate the risk of myocardial damage after an intraoperative hypotensive event, a risk score (detailed in methods) was created from the covariates significantly associated with hs-cTnT more than 14 ng l−1. The patients were divided into three equally sized groups based on risk estimates: a low, a median and a high-risk group. As seen in the supplement, in Figure 1, http://links.lww.com/EJA/A91 (and further detailed in supplementary Table 2, http://links.lww.com/EJA/A91), the risk of elevated levels of hs-cTnT after surgery increases considerably in the presence of intraoperative hypotension in all three risk groups. Specifically, we see relative risks of 4.8, 2.6 and 1.3 in the low, medium and high-risk groups, respectively.

Table 3 details the association between the combination of a hypotensive event, hs-cTnT elevation and MI. The combination of several factors lead to a higher likelihood of the event, indicating an incidence of MI of 24% when the aforementioned risk factors are present compared with 0.5% when absent.

Table 3
Table 3:
Association between the combination of hypotensive eventa, myocardial damage (high-sensitivity cardiac troponin T >14 ng l−1) and myocardial infarction less than 30 days

In supplementary Table 4, http://links.lww.com/EJA/A91, all 15 MI cases within 30 days postoperatively are detailed.

Discussion

This study found that the incidence of myocardial damage on postoperative day 1 and MI within 30 days after major elective noncardiac surgery was 30 and 5%, respectively. Intraoperative hypotension, defined as a fall in SBP more than 50% from baseline for more than 5 min, was independently associated with both myocardial damage on postoperative day 1 and MI within 30 days after surgery. Interestingly, the association between intraoperative hypotension and myocardial damage was consistent and independent of baseline comorbidity. Chronic antihypertensive medications were not independently associated with myocardial damage or intraoperative hypotensive events. As ACE inhibitors are known to increase intraoperative blood pressure instability, national guidelines recommend that patients should discontinue this treatment on the day of surgery. Furthermore, preoperative comorbidity risk factors were more prevalent in cases with high hs-cTnT (>14 ng.l−1) after surgery as well as in patients with perioperative MI. Notably, almost one in four of the patients with perioperative MI had a negative hs-cTnT in the early postoperative phase, possibly indicating later adverse events during hospitalization. In line with the findings from Devereaux and co-workers,11 most MIs occurred within 48 h of surgery (supplementary Table 4, http://links.lww.com/EJA/A91).

The VISION study found that one in 10 patients with myocardial injury after noncardiac surgery will die within 30 days and that more than 80% of these patients go unnoticed without the postoperative troponin measurement.5 Previous studies detailed associations between peak postoperative fourth-generation TnT and 30-day mortality.17 In a small study of 140 patients, sensitive cardiac troponin I was associated with major adverse cardiac events but no differences during the perioperative course were found.24 Two meta-analyses in recent years have detailed the relationship between postoperative leakage of troponin and mortality. Redfern et al.25 used nine studies to investigate how elevation of troponin below the diagnostic threshold for perioperative MI, without symptoms or ischaemic electrocardiography changes or echocardiography signs, was predictive of all-cause mortality at 30 days after vascular surgery. Levy and co-workers16 analysed 14 studies of a total of 3318 patients (of whom 459 died): they demonstrated that increased troponin postoperatively was an independent predictor of all-cause mortality 1 year after noncardiac surgery. At the time of writing, we are unaware of any other studies using hs-cTnT in the perioperative setting.

Few investigations specify the impact of hypotension on perioperative cardiac events but in a nested case–control study, intraoperative hypotension was found to be associated with postoperative stroke.26 In a retrospective analysis from the Cleveland Clinic of patients undergoing noncardiac surgery, mean arterial pressure less than 55 mmHg was associated with adverse outcomes, such as AKI and myocardial injury.19 However, in that study intraoperative blood pressure levels were not related to baseline blood pressure and the study did not report data on patients with preoperative hypertension, a comorbidity likely affecting the impact of intraoperative hypotensive events.

A systematic review on intraoperative hypotension identified 140 definitions used in 130 studies.27 A single blood pressure target derived from group summary data is unlikely to be optimal for individual patients which is why we used the change in SBP from baseline to define intraoperative hypotension in our study. Patients with chronic arterial hypertension may be affected more by hypotensive episodes with respect to multiple organ systems. In a large multicentre randomised study, testing high vs. low blood pressure targets in septic shock patients, benefits of higher blood pressure were only seen among hypertensive patients, and strictly with respect to their renal outcome.28 Despite uncertainties regarding blood pressure we do know that over 10 of the 200 million adults undergoing major noncardiac surgery annually will suffer elevation of troponin within 30 days after surgery.3,17 In the POISE placebo group, a randomised control trial of over 8000 patients, 1.4% suffered vascular death, 0.5% suffered stroke, 0.5% had nonfatal cardiac arrest and 5.7% suffered myocardial injury in the first 30 postoperative days.29 Most researchers have focused on the association between elevated biomarkers in the postoperative period and adverse outcomes, and it may be that the systematic use of biomarkers, such as cardiac troponins, could identify patients at risk. These patients could, in theory, be medicated with statin therapy which has been shown to modify prognosis in nonsurgical patient groups but our results suggest that avoidance of hypotension could also be of value.

Our study has strengths and limitations. We have high-resolution data on all 300 cases, both regarding co-morbidities and perioperative events. Instead of mean arterial pressure levels, we have information on blood pressure decline from patient baseline. Furthermore, we used hs-cTnT, shown to improve risk assessment in acute coronary syndromes.14 Causality between preoperative risks, perioperative events like hypotension, biomarker hs-cTnT-signal indicating myocardial damage and hard endpoints such as MI and death cannot be proven. Actually, it is likely that an effect modifier between these pre, intra and perioperative events is in play, but larger cohorts are needed to determine this. Moreover, we lack preoperative hs-cTnT, which would have been helpful in determining the impact of preoperative risk factors. More than half of the patients had combined general and epidural anaesthesia, known to increase the risk of hypotension. Lastly, as this is a single-centre study, there are inherent limitations in the ability to generalise our findings.

Conclusion

Intraoperative hypotension, defined as a 50% reduction in SBP from baseline for more than 5 min, may be an important event occuring in a subset of individuals who suffer cardiac damage and are at high risk of developing MI in the perioperative period. Studies focussing on avoiding intraoperative hypotension may improve outcomes in future.

Acknowledgements relating to this article

Assistance with the study: none.

Financial support and sponsorship: none.

Conflicts of interest: none.

Presentation: none.

References

1. Etzioni DA, Liu JH, Maggard MA, et al. The aging population and its impact on the surgery workforce. Ann Surg 2003; 238:170–177.
2. Socialstyrelsens statistikdatabas. Operationer i sluten vård 2012. Available from: http://www.socialstyrelsen.se/statistik/statistikdatabas. [Accessed 1 April 2014]
3. Weiser TG, Regenbogen SE, Thompson KD, et al. An estimation of the global volume of surgery: a modelling strategy based on available data. Lancet 2008; 372:139–144.
4. Landesberg G, Beattie WS, Mosseri M, et al. Perioperative myocardial infarction. Circulation 2009; 119:2936–2944.
5. Botto F, Alonso-Coello P, Chan MT, et al. Myocardial injury after noncardiac surgery: a large, international, prospective cohort study establishing diagnostic criteria, characteristics, predictors, and 30-day outcomes. Anesthesiology 2014; 120:564–578.
6. Ashton CM, Petersen NJ, Wray NP, et al. The incidence of perioperative myocardial infarction in men undergoing noncardiac surgery. Ann Inter Med 1993; 118:504–510.
7. Badner NH, Gelb AW. Postoperative myocardial infarction (PMI) after noncardiac surgery. Anesthesiology 1999; 90:644.
8. Kumar R, McKinney WP, Raj G, et al. Adverse cardiac events after surgery: assessing risk in a veteran population. J Gen Inter Med 2001; 16:507–518.
9. Shah KB, Kleinman BS, Rao TL, et al. Angina and other risk factors in patients with cardiac diseases undergoing noncardiac operations. Anesth Analg 1990; 70:240–247.
10. Mangano DT, Wong MG, London MJ, et al. Perioperative myocardial ischemia in patients undergoing noncardiac surgery--II: Incidence and severity during the 1st week after surgery. The Study of Perioperative Ischemia (SPI) Research Group. J Am Coll Cardiol 1991; 17:851–857.
11. Devereaux PJ, Xavier D, Pogue J, et al. Characteristics and short-term prognosis of perioperative myocardial infarction in patients undergoing noncardiac surgery: a cohort study. Ann Inter Med 2011; 154:523–528.
12. Devereaux PJ, Goldman L, Yusuf S, et al. Surveillance and prevention of major perioperative ischemic cardiac events in patients undergoing noncardiac surgery: a review. CMAJ 2005; 173:779–788.
13. Landesberg G, Shatz V, Akopnik I, et al. Association of cardiac troponin, CK-MB, and postoperative myocardial ischemia with long-term survival after major vascular surgery. J Am Coll Cardiol 2003; 42:1547–1554.
14. Lindahl B, Venge P, James S. The new high-sensitivity cardiac troponin T assay improves risk assessment in acute coronary syndromes. Am Heart J 2010; 160:224–229.
15. Mair J. High-sensitivity cardiac troponins in everyday clinical practice. World J Cardiol 2014; 6:175–182.
16. Levy M, Heels-Ansdell D, Hiralal R, et al. Prognostic value of troponin and creatine kinase muscle and brain isoenzyme measurement after noncardiac surgery: a systematic review and meta-analysis. Anesthesiology 2011; 114:796–806.
17. Devereaux PJ, Chan MT, Alonso-Coello P, et al. Association between postoperative troponin levels and 30-day mortality among patients undergoing noncardiac surgery. JAMA 2012; 307:2295–2304.
18. Bijker JB, van Klei WA, Vergouwe Y, et al. Intraoperative hypotension and 1-year mortality after noncardiac surgery. Anesthesiology 2009; 111:1217–1226.
19. Walsh M, Devereaux PJ, Garg AX, et al. Relationship between intraoperative mean arterial pressure and clinical outcomes after noncardiac surgery: toward an empirical definition of hypotension. Anesthesiology 2013; 119:507–515.
20. Monk TG, Bronsert MR, Henderson WG, et al. Association between intraoperative hypotension and hypertension and 30-day postoperative mortality in noncardiac surgery. Anesthesiology 2015; 123:307–319.
21. Giannitsis E, Kurz K, Hallermayer K, et al. Analytical validation of a high-sensitivity cardiac troponin T assay. Clin Chem 2010; 56:254–261.
22. Kellum JA, Lameire N. for the KAKIGWG. Diagnosis, evaluation, and management of acute kidney injury: a KDIGO summary (Part 1). Crit Care 2013; 17:204.
23. Thygesen K, Alpert JS, Jaffe AS, et al. Third universal definition of myocardial infarction. Circulation 2012; 126:2020–2035.
24. Borges FK, Furtado MV, Rossini AP, et al. Clinical use of ultrasensitive cardiac troponin I assay in intermediate- and high-risk surgery patients. Dis Markers 2013; 35:945–953.
25. Redfern G, Rodseth RN, Biccard BM. Outcomes in vascular surgical patients with isolated postoperative troponin leak: a meta-analysis. Anaesthesia 2011; 66:604–610.
26. Bijker JB, Persoon S, Peelen LM, et al. Intraoperative hypotension and perioperative ischemic stroke after general surgery: a nested case-control study. Anesthesiology 2012; 116:658–664.
27. Bijker JB, van Klei WA, Kappen TH, et al. Incidence of intraoperative hypotension as a function of the chosen definition: literature definitions applied to a retrospective cohort using automated data collection. Anesthesiology 2007; 107:213–220.
28. Asfar P, Meziani F, Hamel JF, et al. High versus low blood-pressure target in patients with septic shock. N Engl J Med 2014; 370:1583–1593.
29. Devereaux PJ, Yang H, Yusuf S, et al. Effects of extended-release metoprolol succinate in patients undergoing noncardiac surgery (POISE trial): a randomised controlled trial. Lancet 2008; 371:1839–1847.
© 2016 European Society of Anaesthesiology