This year marks the 20th anniversary of the completion of data collection for the Revised RCRI,1 which identified 6 risk factors independently associated with the composite outcome of “major cardiac complications.”1 At that time, the ACC/AHA recommendations for the preoperative evaluation of the cardiac patient for noncardiac surgery had identified ischemic heart disease, cardiac failure, diabetes, and stroke as clinical risk predictors.2 Renal insufficiency was added as a risk predictor in the 2002 revision.3 Based on the popularity of the RCRI, its risk predictors were integrated into the 2007 revision of the ACC/AHA guidelines.4 Thus, the RCRI became the “gold standard” for clinical cardiac risk stratification of noncardiac surgery patients.
The RCRI surgical risk category was ignored, in preference for the ACC/AHA surgical risk stratification in the preoperative algorithms of 1996, 2002, and 2007.2–4 In the 2014 guidelines,5 2 additional risk stratification models have been included,6,7 which differ from the RCRI, in that they include a more detailed surgical procedural risk category and other additional clinical risk predictors. Therefore, it appears that the pre-eminence of the RCRI is now in question, and the need to improve upon clinical risk prediction for major adverse cardiac events (MACEs) is being considered. Cardiac morbidity after noncardiac surgery remains a leading cause of postoperative mortality, and thus, cardiac clinical risk stratification remains desirable.8
Risk indices are clinically useful if they provide accurate, well-calibrated risk prediction. This would allow for the application of decision analysis tools,9 for which treatment thresholds are determined to minimize adverse outcomes.10 An example could be determining appropriate perioperative β-blockade administration and balancing preventing myocardial ischemia against potential stroke.5,11
There is a now a need for a new best standard preoperative cardiovascular risk stratification model to replace the RCRI, after its uncertain future in ACC/AHA guidelines.5 A best standard risk prediction model is desirable because it would (1) provide a global standard of cardiovascular risk assessment, (2) be a standard comparator in all cardiovascular risk prediction research, (3) provide comparable data, and (4) allow individual patient data meta-analyses, which should lead to continued progress in clinical risk prediction.
PURPOSE OF THE REVIEW
The purpose of this review was to analyze studies on cardiovascular clinical risk prediction, which had used the previous best standard model (i.e., the RCRI) as a comparator. This review aims to determine whether modification of the risk factors identified in the RCRI or adoption of other risk factors or risk indices would improve upon cardiac risk prediction discrimination when compared with the RCRI.
This review focuses only on the prediction of adverse cardiac outcomes. Although a high-risk category of the RCRI has been associated with both short- and long-term mortality,12–15 a meta-analysis has shown marked heterogeneity around the estimate of mortality (I2 = 95%).12 This is not surprising because deaths after noncardiac surgery attributed to vascular etiologies account for only 45% of perioperative deaths.8 Therefore, it is inappropriate to consider the utility of cardiac risk predictors in predicting all-cause mortality.
On April 1, 2014, a MEDLINE search was conducted using the term “revised cardiac risk index,” which returned 183 abstracts. The abstracts were screened, and 100 full papers were retrieved for potential inclusion in this narrative review. These publications were further supplemented by papers from the author’s own collection and from papers identified in the references of the retrieved papers. This review has focused predominantly on risk prediction studies that used the RCRI as a study comparator. Some large cohort or database studies that have not used the RCRI as a comparator have been included in this article for which an improvement in cardiovascular risk prediction has been suggested.
The data have been presented according to the 7-step process of clinical risk prediction model development, as described by Steyerberg and Vergouwe,16 which follows: (1) the research question and data inspection; (2) coding of clinical risk predictors; (3) model specification (and how to choose risk predictors for inclusion in a model); (4) model estimation of regression coefficients; (5) evaluation of model performance; (6) model validity; and (7) model presentation.16 Therefore, the aim is to present clinically relevant recommendations for model development.
THE RESEARCH QUESTION AND INITIAL DATA INSPECTION
Outcome Definitions for Perioperative Cardiovascular Risk Predictors
All future studies evaluating perioperative clinical risk predictors of cardiovascular outcomes should report on a standard perioperative cardiovascular outcome. This has not happened in most of the studies included in this review. Similar outcomes allow direct comparison and meta-analyses of the performance of potential risk factors. It is recommended that 2 standard outcomes are adopted: (1) myocardial infarction (as per the universal definition)17; and (2) myocardial injury after noncardiac surgery (MINS).18
A perioperative troponin leak is independently associated with 30-day postoperative mortality.8 A recent analysis of the Vascular events In noncardiac Surgery patIents cOhort evaluatioN (VISION) study has shown that an elevated troponin level after noncardiac surgery, which is adjudged to be attributable to ischemia, is independently predictive of 30-day mortality, even in the absence of ischemic symptoms or signs.18 Thus, there is a proposal for a new clinical entity known as MINS, with a diagnostic criterion of a peak troponin T (4th generation) ≥0.03 ng/mL judged to be attributable to myocardial ischemia.18 It is likely that the MINS criteria will be adopted for perioperative medicine because more than 80% of patients who leak troponins perioperatively are clinically asymptomatic for myocardial ischemia.8,19–24 Postoperative troponin surveillance and the adoption of the MINS criteria will identify significantly more patients at increased risk for mortality than perioperative myocardial infarction surveillance alone. Troponin surveillance of high-risk patients is now recommended in the Third Universal Definition of Myocardial Infarction study.17
There are limitations associated with the MINS diagnosis.25 These include: (1) the possibility of preexisting preoperative troponin elevation, which was not evaluated20; (2) a possible selection bias in the exclusion of only symptomatic pulmonary embolism because there were no mandated diagnostic tests for pulmonary embolism; and (3) possible chemotherapy-related troponin elevation because of the prevalence of cancer.26 Indeed, troponin elevation has broadly been classified as a result of cardiac events (coronary and noncoronary) and noncardiac conditions (either existing preoperatively or as a consequence of surgical morbidities).25 As a result, approximately one-third (or less) of 30-day mortality may be explained by MINS, whereas other “nonischemic” etiologies explain the balance.18 However, importantly, optimization of medical therapy in patients with myocardial injury may improve intermediate-term survival.27 To address the uncertainties regarding MINS, it is recommended that preoperative troponins are measured in all future cardiovascular clinical risk prediction research.20,25 Furthermore, the performance of clinical risk predictors for myocardial infarction and MINS may therefore differ. Because troponins are routinely evaluated to fulfill the diagnostic criteria for myocardial infarction, it would be cost-effective to simultaneously determine the utility of clinical risk predictors for MINS and myocardial infarction. The determination of these 2 separate outcomes is important because the etiology, and therefore the management, of MINS and myocardial infarction may differ.
What Is Known About Perioperative Cardiovascular Risk Predictors?
To establish a new best standard cardiovascular risk prediction model, a critical review of our knowledge of cardiovascular risk predictors is necessary. The 6 RCRI risk factors remain independently associated with the original RCRI outcome definition,28 MINS,18 and myocardial infarction and cardiac arrest6 in 2014. To establish a new best standard model with better discrimination than the RCRI,6 one will therefore need to modify the definitions of the current risk factors or add more clinical risk factors.
New Risk Factors That Should Be Included in a Clinical Cardiac Risk Stratification Tool
The models that have shown better discrimination have integrated other strong independent clinical predictors, as well as functional status and surgical categories into the model.6,7 Desirable additions to a new best standard cardiovascular risk prediction model are summarized below.
Age was originally identified as an independent predictor of cardiovascular morbidity.29 Age is an independent predictor of cardiovascular events in the presence of the RCRI risk factors and intraoperative risk factors.18,30,31 In the PeriOperative ISchemic Evaluation II trial, an age of more than 75 years included a population at risk for postoperative myocardial infarction of 23.5% (95% confidence interval [CI], 17.9–30.1).32 The risk for myocardial infarction and cardiac arrest increases 1.8 times between the ages of 50 and 80 years in the National Surgical Quality Improvement Program (NSQIP) model.6 It is possible that increased age controls for the increased risk associated with a longer duration of exposure to a clinical risk factor.33
Peripheral Vascular Disease
Peripheral vascular disease is independently predictive of MINS in the presence of the current RCRI risk factors.18 Peripheral vascular disease is a stronger independent predictor of 30-day mortality than vascular surgery.8 Furthermore, in vascular surgical patients, the severity of the peripheral vascular disease increases independent prognostic information in the presence of the RCRI risk factors.34 In a further study on vascular surgical patients, the arm-ankle index also provided independent predictive information when controlled for the RCRI, with an adjusted odds ratio of 10.16 (95% CI, 2.90–36.02) for cardiac complications.35
Although there are no prospective, observational data on the efficacy of functional status as a risk factor for MACE in the presence of the RCRI, it is an integral component of the ACC/AHA preoperative cardiac evaluation algorithm.4,5 The risk of myocardial infarction and cardiac arrest increases 2.75-fold between patients who were totally functionally independent and patients who were totally functionally dependent.6 A limitation with functional dependence is that it is uncommon, with only 7% of the NSQIP patients having been either partially or totally dependent. Using reclassification statistics,36,37 overall net reclassification improvement (NRI) is 72% (P < 0.00001) when patients are reclassified based on a history of an independent functional status or not, using the data presented in the model by Gupta et al.6 (Table 1). This would suggest that functional independence has clinical utility, but it is driven by the high prevalence of functionally independent patients who do not have adverse events. Reclassification based on functional status significantly increases the proportion of patients with adverse cardiac events who are incorrectly classified at lower risk (NRI −15%, P < 0.0001). Therefore, it is recommended that the ACC/AHA refrain from using good functional capacity to recommend proceeding to surgery but rather consider integrating functional status into a new cardiovascular risk prediction model. This would prevent the significant misclassification of patients at cardiac risk who are advised to proceed with surgery, and it would improve the discrimination of a cardiac morbidity risk prediction model.
Surgical Procedural Risk Stratification
Surgery is a strong modifier of cardiovascular risk.1 The recently published ACC/AHA algorithm5 advocates the NSQIP models,6,7 which integrate surgical risk into preoperative clinical risk prediction. Surgery potentially increases cardiac risk through associated physiologic stress, morbidity associated with blood administration, and the inflammatory response. Indeed, laparoscopic surgery has been associated with less inflammation,38 and this may partly account for the decrease in surgical cardiovascular risk.39
Cardiovascular risk prediction models are unlikely to be able to predict intraoperative physiologic responses, including hypotension, hypertension, bradycardia, and tachycardia,40 and as such, the inclusion of intraoperative risk factors should always be considered. Tachycardia31 and hypotension32 are independently associated with postoperative MACE. Furthermore, the duration of surgery30 and bleeding31,32 are also independently associated with MACE and may also reflect the extent of the physiologic insult of surgery. Transfusion,30,41 which is an extremely strong predictor of postoperative cardiac morbidity, may reflect both the physiologic insult and the morbidities associated with blood administration.
Although these factors are independently predictive of MACE, a surgical procedure risk stratification appears to be preferable to markers of surgical severity such as physiologic derangement, duration of surgery, and transfusion or bleeding. The strongest support for this argument is that there is a large range of risk between specific surgical procedures. For all-cause mortality, there is a 250-fold difference in mortality between the lowest-risk and the highest-risk surgeries.42 When considering cardiac morbidity, there is a 25-fold increase in the risk of postoperative myocardial infarction and cardiac arrest between the lowest- and the highest-risk surgical procedures.6
The strength of a surgery procedure-specific risk factor is underscored by the fact that each surgical procedure would be expected to have its own normal distribution for: (1) the duration of operation; (2) the associated cardiovascular physiologic insult; (3) the amount of bleeding and transfusion required; and (4) the inflammation associated with the operation. Thus, every surgical procedure could be considered to have a characteristic physiologic “fingerprint” of the associated physiologic insult. Using the strategy of integrating a “surgery procedure-specific” risk factor would mean that the utility of the physiologic variables, the duration of surgery, bleeding, and transfusion requirements would only provide additional independent information in the 2.5% of patients whose surgical procedures are at the upper limits of the normal Gaussian distribution for the surgical duration, physiologic stress, inflammatory response, bleeding, and transfusion requirements for each surgical procedure.
Although there are currently no data to illustrate this principle with cardiovascular morbidity, the article by Reynolds et al., which shows the relationship of Surgical Apgar Score43 with mortality for various surgical disciplines, illustrates this point.44 The Surgical Apgar Score comprises a maximum of 10 points, which are calculated postoperatively based on the patient’s estimated blood loss, mean arterial blood pressure, and heart rate, with an assignment of fewer points with increasing physiologic derangement.43 By plotting the mean Surgical Apgar Scores for all the high-risk (defined as a 30-day mortality of >4%), intermediate-risk (1%–4% mortality), and low-risk surgeries (<1% mortality) presented in the article by Reynolds et al.,44 it is possible to see the left shift toward lower points associated with a larger physiologic insult associated with higher-risk surgery (Fig. 1). The Figure is a simplistic graphic representation of the physiologic fingerprints associated with different surgical procedures.
Integration of surgical procedures into cardiovascular risk stratification ensures that risk stratification can be conducted preoperatively, and this can be used to determine the pretest probability for cardiovascular events. Only 5% of the patients outside the 95% confidence intervals would require additional risk restratification intraoperatively. Those patients at the lower spectrum of the physiologic fingerprint would have a significantly decreased perioperative cardiovascular risk, and those at the upper spectrum would have a significantly increased cardiovascular risk. The Surgical Apgar Score43 could then be used to determine a Bayesian “posterior probability” for cardiovascular complications after surgery.
Smaller studies have suggested that some of the traditional RCRI risk factors lose their significance in the presence of intraoperative predictors.30,32,45 However, the largest prospective observational study suggests that all the RCRI risk factors remain independently predictive of MINS, together with a history of peripheral vascular disease and surgical risk.18 Finally, clinical risk factors are most compromised by procedures of higher cardiovascular risk,1,12 and integrating specific high-risk surgical procedures into a new cardiac risk index improves discrimination.6
Preoperative anemia has been associated with 30-day cardiac events.46 However, it is possible that this significant association is lost once “mediating variables” of anemia, such as associated blood transfusion and duration of surgery, are included in a risk prediction model.47 Hence, anemia requires additional investigation before it can be considered for a new best standard clinical risk prediction model.
American Society of Anesthesiologists Grading System
The American Society of Anesthesiologists (ASA) grading has been shown to be a powerful predictor of postoperative myocardial infarction and cardiac arrest, with a 9.9-fold increase in risk from ASA II to ASA IV.6 Unfortunately, the performance of the ASA grading has not been evaluated in the presence of the RCRI in studies of postoperative MACE. It is possible that there may be collinearity between the ASA score and the cardiovascular risk factors. For these reasons, further research on the role of the ASA score in cardiac clinical risk stratification is needed.
Finally, because the majority of studies regarding clinical risk predictors are observational, it is critical that the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE)48 initiative is followed during study and data presentation.
CODING OF CLINICAL RISK PREDICTORS
Coding of continuous risk factors into categorical risk factors results in a loss of discrimination of the risk factor.16 The most undesirable approach is dichotomization of continuous variables,16 such as the preoperative serum creatinine in the RCRI.1 Even adding a second cut point to create 3 categories improves discrimination.18 Therefore, in model development, it is recommended that continuous variables are retained to maximize discrimination.16 Once a continuous clinical risk predictor is included in a prognostic model, then subsequent categorization of the risk predictor can be considered if it results in increased ease of use of a prognostic model without a significant loss of discrimination. This is relevant when developing a new best standard model incorporating the potential continuous risk factors, including age and creatinine, among others.
MODEL SPECIFICATION (AND TO CHOOSE RISK PREDICTORS FOR INCLUSION IN A MODEL)
Clinical knowledge is important when considering which specific risk factors to potentially include in a risk prediction model.16 This has been discussed in previous sections. The goal of risk model development must be to derive the most discriminating model. This is ideally achieved when the study sample size is big enough to allow simultaneous entry of all the potential risk factors under investigation with an adequate number of events per variable (EPV) tested.16,49 A model with forced entry of all potential predictors is a more discriminating model than a stepwise entry approach for model development.50 Similarly, a low EPV has the potential to overestimate the strength of the association of a risk factor with an outcome, and occasionally, it can identify an association in the wrong direction.51 Typically, at least 10 EPV should be considered a minimum to prevent this.51 If the sample size is inadequate, to prevent selection bias, it is more acceptable to accept a slightly lower EPV52 than to eliminate potential variables from entry into a derivation model.53 These considerations suggest the need for large collaborative research groups to ensure adequate sample sizes for appropriate cardiovascular risk prediction model development and derivation.
MODEL ESTIMATION OF THE REGRESSION COEFFICIENTS
Even when a new best standard preoperative clinical risk prediction model has been developed, the calibration may be poorer when it is applied to other populations. If the principles highlighted in this review have been followed to develop the new best standard model, it is likely to be secondary to biased model regression coefficients rather than (the almost certain incorrect assumption) that it is the model itself and its associated risk predictors that are at fault. Calibration may be improved by: (1) bootstrapping during derivation and validation50; (2) correction of the model intercept, in which the incidence of the outcome differs between the external population and the derivation population; (3) correction of the model slope, for which there is a common correction to all the coefficients; and (4) correction of the coefficients of each clinical risk predictor.54 We have a relatively good understanding of cardiovascular risk predictors in noncardiac surgery; and as such, a desire to establish new clinical risk prediction models with a large deviation from the principles associated with clinical risk predictors presented here would be undesirable compared with model recalibration.
EVALUATION OF MODEL PERFORMANCE BASED ON DISCRIMINATION, CALIBRATION, AND CLINICAL USEFULNESS
Discrimination is defined as the ability to separate cases (in this case, patients who experience MACE) from controls.55 A clinical risk stratification tool should be able to successfully discriminate high-risk patients (cases) from low-risk patients. It is generally accepted that clinical discrimination is associated with a negative likelihood ratio (LR) of <0.2 and a positive LR ratio of >10.56 For an LR between 0.2 and 10, the clinical usefulness can be determined by the application of the Fagan nomogram, and determination of the posterior probability, based on the pretest probability and an LR of the clinical risk predictors.57 Should the posterior probability pass an accepted treatment threshold, then the risk prediction can be considered clinically useful.
The performance of the 3 risk categories previously advocated by the ACC/AHA4 is shown in Table 2. The lowest risk category has no risk factors, and the RCRI performs well and has clinical utility because it essentially excludes MACE after elective noncardiac surgery.1,28 The clinical utility of the high-risk category (3 or more risk factors) is more controversial, with an LR of 4.9 in the original study,56 and more recently, an LR of 7.9 for MACE in the Clinical Anesthesia Information System (CAIS) study (which is the most contemporaneous evaluation of the RCRI).28 The clinical usefulness of this category is therefore dependent on the pretest probability and the posterior probability in relation to accepted treatment thresholds. An individual patient data meta-analysis of predominantly high-risk noncardiac surgical patients has shown that the ≥3 RCRI risk factors is an independent predictor of death or nonfatal myocardial infarction within 30 postoperative days, even in the presence of elevated preoperative or postoperative B-type natriuretic peptides.58
For clinical cardiovascular risk stratification, the problem remains; however, that the proportion of patients in the intermediate-risk category (1 or 2 risk factors) is unacceptably large, with 56% in the original study,1 and 42% in a recent evaluation of the RCRI,28 together with a clinically useless LR of 1.0 and 1.7, respectively, this poses an excessive clinical burden in trying to identify the 3.6% of the patients in this “gray zone,” who will go on to have a MACE.28 Splitting the intermediate-risk category group into the absolute number of risk factors does not substantially improve the LR, with 1 risk factor having an LR of 0.51 to 1.228 and 2 risk factors with an LR of 2.21 to 3.6.28 Clearly, there is potential here for clinical reclassification from this intermediate-risk category into the low- or high-risk category based on the presence of some additional clinical risk predictor. This has typically been shown with B-type natriuretic peptides.58
For a prognostic model, it is important to demonstrate clinical utility (or usefulness) associated with risk predictor modifications or the addition of new clinical risk predictors.59 This is commonly statistically evaluated with NRI, in which the statistical change in correct and incorrect event reclassification is evaluated against the current best standard model.36,37 A significant increase in patients correctly classified as low and high risk after reclassification according to a new model suggests clinical utility over the standard comparator model. Publication of net reclassification statistics should be considered a prerequisite for any risk prediction model research.
The discrimination of a test may also be described by the area under the receiver operating characteristic curve (AUC), where the percentage reflects the probability that a positive test will experience an event (MACE) when compared with a negative test.36 The CAIS evaluation of the RCRI found moderately good performance of the RCRI, with an AUC of 0.79 (95% CI, 0.76–0.83),28 which was similar to the original study 0.78 (95% CI, 0.73–0.82).
It is possible that the categories originally suggested by the ACC/AHA may not be optimal, and other criteria for categorization of risk based on the RCRI may improve clinical utility. However, separation of the RCRI into low risk (0 or 1 risk factors) and high risk (2 or more risk factors) reported in a meta-analysis of perioperative cardiac complications or death within 30 days of surgery12 showed moderate discrimination (AUC, 0.75; 95% CI, 0.72–0.79) and clinically poor LRs, with a positive LR of 2.78 (95% CI, 1.74–4.45) and a negative LR of 0.45 (95% CI, 0.31–0.67).
In summary, it appears that the absence of any of the RCRI clinical risk factors is clinically useful and essentially excludes perioperative cardiovascular complications. The clinical utility of intermediate- and high-risk RCRI categories is dependent on the posterior probability and associated treatment thresholds.
MODEL VALIDITY BASED ON BOTH INTERNAL AND EXTERNAL VALIDATION
The RCRI has been extensively externally validated. The other 2 risk indices6,7 proposed in the new AHA/ACC preoperative algorithm5 are based on the NSQIP data and are currently validated only in the United States. The NSQIP models provide better discrimination than the RCRI in the United States, although any new best standard model should be globally accessible and applicable. Although the NSQIP models have clinical utility, I would not encourage their adoption for comparative research of cardiac clinical risk predictors unless these NSQIP models can be accurately duplicated globally. The potential risk factors discussed in this article suggest that a new global best standard model for preoperative cardiovascular risk stratification is possible without the complexity of the NSQIP data collection.60
The internal and external validation of the proposed additional clinical risk predictors for cardiovascular outcomes discussed in this article are shown in Table 3.
PRESENTATION OF THE MODEL
Before the new preoperative algorithm of 2014,5 the AHA/ACC followed the principle of a parsimonious model, including a few simple risk predictors. The statistical argument against a parsimonious model is that these models require determination of: (1) which less significant variables to remove; and (2) the coefficients of the remaining variables in the absence of the removed less significant variables.49 This is probably a more difficult task than simply including all potential clinical risk predictions in model development,49 which would also improve discrimination.6,7 The decision by the AHA/ACC to include the NSQIP models6,7 in the 2014 algorithm suggests that: (1) improved discrimination is a priority; (2) these models are now clinically feasible centered on Web-based calculators; and (3) a new best standard model does not necessarily have to be parsimonious.
THE SURGERIES AND PATIENTS FOR WHOM AN ALTERNATIVE MODEL CARDIOVASCULAR RISK PREDICTION MODEL SHOULD BE CONSIDERED
It is likely that most perioperative cardiovascular events follow similar pathophysiologic pathways,64 and as such, it would require exceptional unique circumstances to deviate from a generic cardiovascular risk prediction model. However, there are studies that suggest that it may be inappropriate to apply a generic cardiovascular risk prediction model to all patient groups and surgical specialties. These are discussed.
Human Immunodeficiency Virus
A study of human immunodeficiency virus (HIV)–infected vascular surgical patients has shown a similar incidence of postoperative MACE to traditional atherosclerotic vascular surgical patients, although the HIV-positive patients had significantly fewer RCRI risk factors.65 It appears that these risk factors are unlikely to perform well in HIV patients, and the development of a specific HIV cardiovascular risk model should be developed. It is likely that cardiovascular risk in HIV-positive patients is modified both positively66 and negatively with highly antiretroviral therapy.67 Statin therapy may offer cardiovascular protection.68 The CD4 may also modify outcome.69,70 The effects of medical therapy for HIV and the associated CD4 counts on postoperative cardiac morbidity need to be investigated. This could be considered a unique pathophysiologic circumstance that demands an independent cardiovascular risk prediction model.
The RCRI has consistently shown inferior discrimination in vascular surgical patients when compared with its performance in other types of noncardiac surgery.6,12,28,71 In a systematic review, the performance of the RCRI was significantly poorer in vascular surgical patients compared with other noncardiac surgery.12 In the NSQIP data, the RCRI had an AUC of 0.591 for vascular surgery compared with the entire cohort of 0.747 for the RCRI.6 The calibration of the RCRI appears to be poorer in vascular surgical patients because vascular surgical procedures of increasing cardiovascular risk have been associated with a progressive underestimation of the cardiovascular complications by the RCRI.72
All the RCRI risk factors appear to be predictive in vascular surgery,73 consistent with a common pathophysiology for perioperative cardiac events across all types of noncardiac surgery. As in other noncardiac surgeries, similar modifications to the RCRI in vascular surgery would improve cardiac clinical risk stratification, including age72–75 and the severity of specific vascular surgical procedures.74 However, important modifiers of risk in vascular surgery include a history of coronary revascularization, which has been associated with an independent reduction in postoperative MACE.72,74 It appears that previous coronary revascularization that was medically indicated provides independent coronary protection in vascular surgical patients because this benefit has not been realized when vascular patients have been randomized to preoperative coronary revascularization.76 Second, medications appear to modify risk in these patients, too, with an increased risk of MACE associated with chronic β-blockade.72,74 Third, chronic obstructive pulmonary disease appears to be an important predictor of cardiovascular risk in vascular surgical patients.72
The identification of these 3 factors as independent predictors of MACE in vascular surgical patients may be a result of the high prevalence of these factors in vascular surgical patients as opposed to other noncardiac surgeries, which would increase their positive or negative predictive values.77 Other predictors have also been identified, such as smoking72 and hypertension,72,73 although their importance is debatable because they have been less consistently identified.
There are data to suggest that for elective aortic aneurysm surgery, a totally new risk model may be needed because the RCRI has been shown to have significantly poorer performance for MACE than the Physiological and Operative Severity Score for enUmeration of Mortality (V(p)-POSSUM) and Preoperative Risk Score of the Estimation of Physiological Ability and Surgical Stress Score (PRS of E-PASS).78 However, currently, there does not appear to be an advantageous clinical risk stratification tool for elective aortic surgery.79
In summary, although the RCRI could be modified and improved in vascular surgical patients using the approaches discussed previously, which are applicable to all noncardiac surgeries, a vascular-specific clinical risk stratification tool is likely to be preferable. It is likely that aortic surgery may require another independent risk model.
In thoracic surgery, a “Thoracic RCRI” has been proposed, which includes a pneumonectomy score to improve performance in predicting cardiovascular complications after thoracic surgery.80 Subsequent evaluations have been inconsistent, with either no difference between the RCRI and the Thoracic RCRI for elective lung resection in predicting MACE (AUC, 0.59; 95% CI, 0.51–0.67 and 0.57; 95% CI, 0.49–0.66, respectively),81 or other external cohorts showing significantly better discrimination with the Thoracic RCRI with an AUC of 0.74 (95% CI, 0.71–0.76).82 It appears that modification of the RCRI may be appropriate for thoracic surgery, provided an appropriate surgical procedure is added to the risk prediction model.
Although the predictive role of congestive cardiac failure and diabetes has been questioned in thoracic surgery,80 the small data sets and lack of a sound indication that the pathophysiology of cardiovascular events should be different in thoracic surgical patients would demand reservations at this point before drawing any conclusions regarding congestive cardiac failure or diabetes in these patients.
THE FUTURE ROLE OF BIOMARKERS TO MODIFY AND IMPROVE UPON THE RCRI
Because routine preoperative creatinine screening is currently required for cardiovascular risk stratification, it is likely that preoperative biomarker estimation will also be integrated into routine preoperative cardiovascular risk stratification in the future. There are strong individual patient data that show that preoperative B-type natriuretic peptides significantly increase the number of patients correctly reclassified as being at risk of MACE even in the presence of the RCRI.58 In vascular surgical patients, the RCRI has been shown to be redundant in the presence of B-type natriuretic peptides serum estimation.83 It has been recommended that a demonstration of B-type natriuretic peptide–directed therapy improves postoperative cardiovascular outcomes before it can be integrated into preoperative risk stratification.59,84 Once this is achieved, this will result again in a significant change in the characteristics and constituents of a preoperative clinical risk stratification tool.
Preoperative risk stratification with troponins is currently controversial. Preoperative troponin estimation has been shown to result in a significant increase in misclassification of patients who subsequently have cardiac complications,85,86 although this may be a function of less-sensitive troponin assays used85 or the cut points used.86 However, if one conducts an overall net reclassification of the only preoperative data of high-sensitive troponin T presented in the article by Nagele et al.20 on patients who subsequently develop myocardial infarction, preoperative troponin elevation is associated with a significantly improved risk stratification (overall NRI 61%, P = 0.001) and no misclassification of patients who develop myocardial infarction (NRI 40%, P = 0.03). Based on these conflicting data, the role of preoperative risk stratification with troponins is not yet clear. However, an elevated preoperative troponin is strongly associated with both short- and long-term cardiac morbidity and mortality, and these patients are at a significantly increased risk.20,86 Finally, it is recommended that preoperative troponin estimation should be considered in all studies of perioperative cardiovascular outcomes, as discussed in the previous section of standardized outcomes.
Recommendations to Improve Preoperative Cardiac Risk Stratification
To improve the preoperative clinical risk stratification for perioperative MACE, it is recommended that a new best standard model is developed that maintains the clinical risk factors identified in the RCRI, with the following modifications to include: (1) additional glomerular filtration rate cut points (as opposed to a single creatinine cut point); (2) age; (3) a history of peripheral vascular disease; (4) functional capacity; and (5) a specific surgical procedural category. One would expect a substantial improvement in the discrimination of the RCRI with this approach.39 Furthermore, it is time to move from a simple parsimonious model1 to a more discriminating model.5 This is easily achieved with a calculator on a handheld device or online.6 The ACC/AHA should consider removal of the functional capacity risk stratification at step 5 in the algorithm.5
This new best standard risk model proposal could be realized currently because the VISION study8 has prospectively collected data on 40,000 patients with objective cardiovascular outcomes for the entire cohort. All the data that have been recommended in this article to improve upon clinical cardiovascular risk stratification have been prospectively collected in the VISION data set. Finally, a model should be specifically developed for HIV patients and possibly vascular surgical patients.
The RCRI risk factors continue to be clinically relevant in 2014. Incorporation of age, peripheral vascular disease, functional capacity, and a specific surgery procedural risk factor would significantly improve model discrimination.
Name: Bruce Biccard, MBChB, FCA(SA), FFARCSI, MMedSci, PhD.
Contribution: This author wrote the manuscript.
Attestation: Bruce Biccard approved this manuscript.
This manuscript was handled by: Martin J. London, MD.
1. Lee TH, Marcantonio ER, Mangione CM, Thomas EJ, Polanczyk CA, Cook EF, Sugarbaker DJ, Donaldson MC, Poss R, Ho KK, Ludwig LE, Pedan A, Goldman L. Derivation and prospective validation of a simple index for prediction of cardiac risk of major noncardiac surgery. Circulation. 1999;100:1043–9
2. Eagle KA, Brundage BH, Chaitman BR, Ewy GA, Fleisher LA, Hertzer NR, Leppo JA, Ryan T, Schlant RC, Spencer WH III, Spittell JA Jr, Twiss RD, Ritchie JL, Cheitlin MD, Gardner TJ, Garson A Jr, Lewis RP, Gibbons RJ, O’Rourke RA, Ryan TJ. Guidelines for perioperative cardiovascular evaluation for noncardiac surgery. Report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines. Committee on Perioperative Cardiovascular Evaluation for Noncardiac Surgery. Circulation. 1996;93:1278–317
3. Eagle KA, Berger PB, Calkins H, Chaitman BR, Ewy GA, Fleischmann KE, Fleisher LA, Froehlich JB, Gusberg RJ, Leppo JA, Ryan T, Schlant RC, Winters WL Jr, Gibbons RJ, Antman EM, Alpert JS, Faxon DP, Fuster V, Gregoratos G, Jacobs AK, Hiratzka LF, Russell RO, Smith SC Jr. ACC/AHA guideline update for perioperative cardiovascular evaluation for noncardiac surgery—executive summary: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Committee to Update the 1996 Guidelines on Perioperative Cardiovascular Evaluation for Noncardiac Surgery). J Am Coll Cardiol. 2002;39:542–53
4. Fleisher LA, Beckman JA, Brown KA, Calkins H, Chaikof E, Fleischmann KE, Freeman WK, Froehlich JB, Kasper EK, Kersten JR, Riegel B, Robb JF, Smith SC Jr, Jacobs AK, Adams CD, Anderson JL, Antman EM, Buller CE, Creager MA, Ettinger SM, Faxon DP, Fuster V, Halperin JL, Hiratzka LF, Hunt SA, Lytle BW, Nishimura R, Ornato JP, Page RL, Riegel B, Tarkington LG, Yancy CWACC/AHA TASK FORCE MEMBERS. . ACC/AHA 2007 Guidelines on Perioperative Cardiovascular Evaluation and Care for Noncardiac Surgery: Executive Summary: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Writing Committee to Revise the 2002 Guidelines on Perioperative Cardiovascular Evaluation for Noncardiac Surgery): developed in collaboration with the American Society of Echocardiography, American Society of Nuclear Cardiology, Heart Rhythm Society, Society of Cardiovascular Anesthesiologists, Society for Cardiovascular Angiography and Interventions, Society for Vascular Medicine and Biology, and Society for Vascular Surgery. Circulation. 2007;116:1971–96
5. Fleisher LA, Linde-Zwirble WT. Incidence, outcome, and attributable resource use associated with pulmonary and cardiac complications after major small and large bowel procedures. Perioper Med (Lond). 2014;3:7
6. Gupta PK, Gupta H, Sundaram A, Kaushik M, Fang X, Miller WJ, Esterbrooks DJ, Hunter CB, Pipinos II, Johanning JM, Lynch TG, Forse RA, Mohiuddin SM, Mooss AN. Development and validation of a risk calculator for prediction of cardiac risk after surgery. Circulation. 2011;124:381–7
7. Cohen ME, Ko CY, Bilimoria KY, Zhou L, Huffman K, Wang X, Liu Y, Kraemer K, Meng X, Merkow R, Chow W, Matel B, Richards K, Hart AJ, Dimick JB, Hall BL. Optimizing ACS NSQIP modeling for evaluation of surgical quality and risk: patient risk adjustment, procedure mix adjustment, shrinkage adjustment, and surgical focus. J Am Coll Surg. 2013;217:336–46.e1
8. Devereaux PJ, Chan MT, Alonso-Coello P, Walsh M, Berwanger O, Villar JC, Wang CY, Garutti RI, Jacka MJ, Sigamani A, Srinathan S, Biccard BM, Chow CK, Abraham V, Tiboni M, Pettit S, Szczeklik W, Lurati Buse G, Botto F, Guyatt G, Heels-Ansdell D, Sessler DI, Thorlund K, Garg AX, Mrkobrada M, Thomas S, Rodseth RN, Pearse RM, Thabane L, McQueen MJ, VanHelder T, Bhandari M, Bosch J, Kurz A, Polanczyk C, Malaga G, Nagele P, Le Manach Y, Leuwer M, Yusuf S. Association between postoperative troponin levels and 30-day mortality among patients undergoing noncardiac surgery. JAMA. 2012;307:2295–304
9. Vickers AJ, Elkin EB. Decision curve analysis: a novel method for evaluating prediction models. Med Decis Making. 2006;26:565–74
10. Vickers AJ. Decision analysis for the evaluation of diagnostic tests, prediction models and molecular markers. Am Stat. 2008;62:314–20
11. Wijeysundera DN, Duncan D, Nkonde-Price C, Virani SS, Washam JB, Fleischmann KE, Fleisher LA. Perioperative beta-blockade in noncardiac surgery: a systematic review for the 2014 ACC/AHA guideline on perioperative cardiovascular evaluation and management of patients undergoing noncardiac surgery: a report of the American College of Cardiology/American Heart Association Task Force on practice guidelines. J Am Coll Cardiol. 2014;64:2406–25
12. Ford MK, Beattie WS, Wijeysundera DN. Systematic review: prediction of perioperative cardiac complications and mortality by the revised cardiac risk index. Ann Intern Med. 2010;152:26–35
13. Archan S, Roscher CR, Fairman RM, Fleisher LA. Revised Cardiac Risk Index (Lee) and perioperative cardiac events as predictors of long-term mortality in patients undergoing endovascular abdominal aortic aneurysm repair. J Cardiothorac Vasc Anesth. 2010;24:84–90
14. London MJ, Hur K, Schwartz GG, Henderson WG. Association of perioperative β-blockade with mortality and cardiovascular morbidity following major noncardiac surgery. JAMA. 2013;309:1704–13
15. Lindenauer PK, Pekow P, Wang K, Mamidi DK, Gutierrez B, Benjamin EM. Perioperative beta-blocker therapy and mortality after major noncardiac surgery. N Engl J Med. 2005;353:349–61
16. Steyerberg EW, Vergouwe Y. Towards better clinical prediction models: seven steps for development and an ABCD for validation. Eur Heart J. 2014;35:1925–31
17. Thygesen K, Alpert JS, Jaffe AS, Simoons ML, Chaitman BR, White HD, Thygesen K, Alpert JS, White HD, Jaffe AS, Katus HA, Apple FS, Lindahl B, Morrow DA, Chaitman BR, Clemmensen PM, Johanson P, Hod H, Underwood R, Bax JJ, Bonow JJ, Pinto F, Gibbons RJ, Fox KA, Atar D, Newby LK, Galvani M, Hamm CW, Uretsky BF, Steg PG, Wijns W, Bassand JP, Menasche P, Ravkilde J, Ohman EM, Antman EM, Wallentin LC, Armstrong PW, Simoons ML, Januzzi JL, Nieminen MS, Gheorghiade M, Filippatos G, Luepker RV, Fortmann SP, Rosamond WD, Levy D, Wood D, Smith SC, Hu D, Lopez-Sendon JL, Robertson RM, Weaver D, Tendera M, Bove AA, Parkhomenko AN, Vasilieva EJ, Mendis S, Bax JJ, Baumgartner H, Ceconi C, Dean V, Deaton C, Fagard R, Funck-Brentano C, Hasdai D, Hoes A, Kirchhof P, Knuuti J, Kolh P, McDonagh T, Moulin C, Popescu BA, Reiner Z, Sechtem U, Sirnes PA, Tendera M, Torbicki A, Vahanian A, Windecker S, Morais J, Aguiar C, Almahmeed W, Arnar DO, Barili F, Bloch KD, Bolger AF, Botker HE, Bozkurt B, Bugiardini R, Cannon C, de Lemos J, Eberli FR, Escobar E, Hlatky M, James S, Kern KB, Moliterno DJ, Mueller C, Neskovic AN, Pieske BM, Schulman SP, Storey RF, Taubert KA, Vranckx P, Wagner DRJoint ESC/ACCF/AHA/WHF Task Force for Universal Definition of Myocardial Infarction; Authors/Task Force Members Chairpersons; Biomarker Subcommittee; ECG Subcommittee; Imaging Subcommittee; Classification Subcommittee; Intervention Subcommittee; Trials & Registries Subcommittee; ESC Committee for Practice Guidelines (CPG); Document Reviewers. . Third universal definition of myocardial infarction. J Am Coll Cardiol. 2012;60:1581–98
18. Botto F, Alonso-Coello P, Chan MT, Villar JC, Xavier D, Srinathan S, Guyatt G, Cruz P, Graham M, Wang CY, Berwanger O, Pearse RM, Biccard BM, Abraham V, Malaga G, Hillis GS, Rodseth RN, Cook D, Polanczyk CA, Szczeklik W, Sessler DI, Sheth T, Ackland GL, Leuwer M, Garg AX, Lemanach Y, Pettit S, Heels-Ansdell D, Luratibuse G, Walsh M, Sapsford R, Schünemann HJ, Kurz A, Thomas S, Mrkobrada M, Thabane L, Gerstein H, Paniagua P, Nagele P, Raina P, Yusuf S, Devereaux PJ, Devereaux PJ, Sessler DI, Walsh M, Guyatt G, McQueen MJ, Bhandari M, Cook D, Bosch J, Buckley N, Yusuf S, Chow CK, Hillis GS, Halliwell R, Li S, Lee VW, Mooney J, Polanczyk CA, Furtado MV, Berwanger O, Suzumura E, Santucci E, Leite K, Santo JA, Jardim CA, Cavalcanti AB, Guimaraes HP, Jacka MJ, Graham M, McAlister F, McMurtry S, Townsend D, Pannu N, Bagshaw S, Bessissow A, Bhandari M, Duceppe E, Eikelboom J, Ganame J, Hankinson J, Hill S, Jolly S, Lamy A, Ling E, Magloire P, Pare G, Reddy D, Szalay D, Tittley J, Weitz J, Whitlock R, Darvish-Kazim S, Debeer J, Kavsak P, Kearon C, Mizera R, O’Donnell M, McQueen M, Pinthus J, Ribas S, Simunovic M, Tandon V, Vanhelder T, Winemaker M, Gerstein H, McDonald S, O’Bryne P, Patel A, Paul J, Punthakee Z, Raymer K, Salehian O, Spencer F, Walter S, Worster A, Adili A, Clase C, Cook D, Crowther M, Douketis J, Gangji A, Jackson P, Lim W, Lovrics P, Mazzadi S, Orovan W, Rudkowski J, Soth M, Tiboni M, Acedillo R, Garg A, Hildebrand A, Lam N, Macneil D, Mrkobrada M, Roshanov PS, Srinathan SK, Ramsey C, John PS, Thorlacius L, Siddiqui FS, Grocott HP, McKay A, Lee TW, Amadeo R, Funk D, McDonald H, Zacharias J, Villar JC, Cortés OL, Chaparro MS, Vásquez S, Castañeda A, Ferreira S, Coriat P, Monneret D, Goarin JP, Esteve CI, Royer C, Daas G, Chan MT, Choi GY, Gin T, Lit LC, Xavier D, Sigamani A, Faruqui A, Dhanpal R, Almeida S, Cherian J, Furruqh S, Abraham V, Afzal L, George P, Mala S, Schünemann H, Muti P, Vizza E, Wang CY, Ong GS, Mansor M, Tan AS, Shariffuddin II, Vasanthan V, Hashim NH, Undok AW, Ki U, Lai HY, Ahmad WA, Razack AH, Malaga G, Valderrama-Victoria V, Loza-Herrera JD, De Los Angeles Lazo M, Rotta-Rotta A, Szczeklik W, Sokolowska B, Musial J, Gorka J, Iwaszczuk P, Kozka M, Chwala M, Raczek M, Mrowiecki T, Kaczmarek B, Biccard B, Cassimjee H, Gopalan D, Kisten T, Mugabi A, Naidoo P, Naidoo R, Rodseth R, Skinner D, Torborg A, Paniagua P, Urrutia G, Maestre ML, Santaló M, Gonzalez R, Font A, Martínez C, Pelaez X, De Antonio M, Villamor JM, García JA, Ferré MJ, Popova E, Alonso-Coello P, Garutti I, Cruz P, Fernández C, Palencia M, Díaz S, Del Castillo T, Varela A, de Miguel A, Muñoz M, Piñeiro P, Cusati G, Del Barrio M, Membrillo MJ, Orozco D, Reyes F, Sapsford RJ, Barth J, Scott J, Hall A, Howell S, Lobley M, Woods J, Howard S, Fletcher J, Dewhirst N, Williams C, Rushton A, Welters I, Leuwer M, Pearse R, Ackland G, Khan A, Niebrzegowska E, Benton S, Wragg A, Archbold A, Smith A, McAlees E, Ramballi C, Macdonald N, Januszewska M, Stephens R, Reyes A, Paredes LG, Sultan P, Cain D, Whittle J, Del Arroyo AG, Sessler DI, Kurz A, Sun Z, Finnegan PS, Egan C, Honar H, Shahinyan A, Panjasawatwong K, Fu AY, Wang S, Reineks E, Nagele P, Blood J, Kalin M, Gibson D, Wildes TVascular events In noncardiac Surgery patIents cOhort evaluatioN (VISION) Writing Group, on behalf of The Vascular events In noncardiac Surgery patIents cOhort evaluatioN (VISION) Investigators; Appendix 1. The Vascular events In noncardiac Surgery patIents cOhort evaluatioN (VISION) Study Investigators Writing Group; Appendix 2. The Vascular events In noncardiac Surgery patIents cOhort evaluatioN Operations Committee; Vascular events In noncardiac Surgery patIents cOhort evaluatioN VISION Study Investigators. . Myocardial injury after noncardiac surgery: a large, international, prospective cohort study establishing diagnostic criteria, characteristics, predictors, and 30-day outcomes. Anesthesiology. 2015;120:564–78
19. van Waes JA, Nathoe HM, de Graaff JC, Kemperman H, de Borst GJ, Peelen LM, van Klei WACardiac Health After Surgery (CHASE) Investigators. . Myocardial injury after noncardiac surgery and its association with short-term mortality. Circulation. 2013;127:2264–71
20. Nagele P, Brown F, Gage BF, Gibson DW, Miller JP, Jaffe AS, Apple FS, Scott MG. High-sensitivity cardiac troponin T in prediction and diagnosis of myocardial infarction and long-term mortality after noncardiac surgery. Am Heart J. 2013;166:325–32.e1
21. Chong CP, van Gaal WJ, Profitis K, Ryan JE, Savige J, Lim WK. Electrocardiograph changes, troponin levels and cardiac complications after orthopaedic surgery. Ann Acad Med Singapore. 2013;42:24–32
22. Hietala P, Strandberg M, Strandberg N, Gullichsen E, Airaksinen KE. Perioperative myocardial infarctions are common and often unrecognized in patients undergoing hip fracture surgery. J Trauma Acute Care Surg. 2013;74:1087–91
23. Marston N, Sandoval Y, Zakharova M, Brenes-Salazar J, Santili S, Adabag S, McFalls EO, Garcia S. Troponin elevations following vascular surgery in patients without preoperative myocardial ischemia. South Med J. 2013;106:612–7
24. Garcia S, Marston N, Sandoval Y, Pierpont G, Adabag S, Brenes J, Santilli S, McFalls EO. Prognostic value of 12-lead electrocardiogram and peak troponin I level after vascular surgery. J Vasc Surg. 2013;57:166–72
25. Nathoe HM, van Klei WA, Beattie WS. Perioperative troponin elevation: always myocardial injury, but not always myocardial infarction. Anesth Analg. 2014;119:1014–6
26. Cardinale D, Bacchiani G, Beggiato M, Colombo A, Cipolla CM. Strategies to prevent and treat cardiovascular risk in cancer patients. Semin Oncol. 2013;40:186–98
27. Foucrier A, Rodseth R, Aissaoui M, Ibanes C, Goarin JP, Landais P, Coriat P, Le Manach Y. The long-term impact of early cardiovascular therapy intensification for postoperative troponin elevation after major vascular surgery. Anesth Analg. 2014;119:1053–63
28. Davis C, Tait G, Carroll J, Wijeysundera DN, Beattie WS. The Revised Cardiac Risk Index in the new millennium: a single-centre prospective cohort re-evaluation of the original variables in 9,519 consecutive elective surgical patients. Can J Anesth. 2013;60:855–63
29. Goldman L, Caldera DL, Nussbaum SR, Southwick FS, Krogstad D, Murray B, Burke DS, O’Malley TA, Goroll AH, Caplan CH, Nolan J, Carabello B, Slater EE. Multifactorial index of cardiac risk in noncardiac surgical procedures. N Engl J Med. 1977;297:845–50
30. Bae MH, Lee JH, Yang DH, Park HS, Cho Y, Chae SC. Usefulness of surgical parameters as predictors of postoperative cardiac events in patients undergoing non-cardiac surgery. Circ J. 2014;78:718–23
31. Devereaux PJ, Xavier D, Pogue J, Guyatt G, Sigamani A, Garutti I, Leslie K, Rao-Melacini P, Chrolavicius S, Yang H, Macdonald C, Avezum A, Lanthier L, Hu W, Yusuf SPOISE (PeriOperative ISchemic Evaluation) Investigators. . Characteristics and short-term prognosis of perioperative myocardial infarction in patients undergoing noncardiac surgery: a cohort study. Ann Intern Med. 2011;154:523–8
32. Devereaux PJ, Sessler DI, Leslie K, Kurz A, Mrkobrada M, Alonso-Coello P, Villar JC, Sigamani A, Biccard BM, Meyhoff CS, Parlow JL, Guyatt G, Robinson A, Garg AX, Rodseth RN, Botto F, Lurati Buse G, Xavier D, Chan MT, Tiboni M, Cook D, Kumar PA, Forget P, Malaga G, Fleischmann E, Amir M, Eikelboom J, Mizera R, Torres D, Wang CY, Vanhelder T, Paniagua P, Berwanger O, Srinathan S, Graham M, Pasin L, Le Manach Y, Gao P, Pogue J, Whitlock R, Lamy A, Kearon C, Chow C, Pettit S, Chrolavicius S, Yusuf SPOISE-2 Investigators. . Clonidine in patients undergoing noncardiac surgery. N Engl J Med. 2014;370:1504–13
33. Biccard BM, Rodseth RN. Utility of clinical risk predictors for preoperative cardiovascular risk prediction. Br J Anaesth. 2011;107:133–43
34. Flu WJ, van Kuijk JP, Voûte MT, Kuiper R, Verhagen HJ, Bax JJ, Poldermans D. Asymptomatic low ankle-brachial index in vascular surgery patients: a predictor of perioperative myocardial damage. Eur J Vasc Endovasc Surg. 2010;39:62–9
35. Fisher BW, Ramsay G, Majumdar SR, Hrazdil CT, Finegan BA, Padwal RS, McAlister FA. The ankle-to-arm blood pressure index predicts risk of cardiac complications after noncardiac surgery. Anesth Analg. 2008;107:149–54
36. Pencina MJ, D’Agostino RB Sr, D’Agostino RB Jr, Vasan RS. Evaluating the added predictive ability of a new marker: from area under the ROC curve to reclassification and beyond. Stat Med. 2008;27:157–72
37. Pencina MJ, D’Agostino RB Sr, Steyerberg EW. Extensions of net reclassification improvement calculations to measure usefulness of new biomarkers. Stat Med. 2011;30:11–21
38. Sammour T, Kahokehr A, Chan S, Booth RJ, Hill AG. The humoral response after laparoscopic versus open colorectal surgery: a meta-analysis. J Surg Res. 2010;164:28–37
39. Boersma E, Kertai MD, Schouten O, Bax JJ, Noordzij P, Steyerberg EW, Schinkel AF, van Santen M, Simoons ML, Thomson IR, Klein J, van Urk H, Poldermans D. Perioperative cardiovascular mortality in noncardiac surgery: validation of the Lee cardiac risk index. Am J Med. 2005;118:1134–41
40. Röhrig R, Junger A, Hartmann B, Klasen J, Quinzio L, Jost A, Benson M, Hempelmann G. The incidence and prediction of automatically detected intraoperative cardiovascular events in noncardiac surgery. Anesth Analg. 2004;98:569–77
41. Biccard BM, Rodseth RN. What evidence is there for intraoperative predictors of perioperative cardiac outcomes? A systematic review. Perioper Med (Lond). 2013;2:14
42. Noordzij PG, Poldermans D, Schouten O, Bax JJ, Schreiner FA, Boersma E. Postoperative mortality in The Netherlands: a population-based analysis of surgery-specific risk in adults. Anesthesiology. 2010;112:1105–15
43. Gawande AA, Kwaan MR, Regenbogen SE, Lipsitz SA, Zinner MJ. An Apgar score for surgery. J Am Coll Surg. 2007;204:201–8
44. Reynolds PQ, Sanders NW, Schildcrout JS, Mercaldo ND, St Jacques PJ. Expansion of the surgical Apgar score across all surgical subspecialties as a means to predict postoperative mortality. Anesthesiology. 2011;114:1305–12
45. Dillon JK, Liu SY, Patel CM, Schmidt BL. Identifying risk factors for postoperative cardiovascular and respiratory complications after major oral cancer surgery. Head Neck. 2011;33:112–6
46. Wu WC, Schifftner TL, Henderson WG, Eaton CB, Poses RM, Uttley G, Sharma SC, Vezeridis M, Khuri SF, Friedmann PD. Preoperative hematocrit levels and postoperative outcomes in older patients undergoing noncardiac surgery. JAMA. 2007;297:2481–8
47. Saager L, Turan A, Reynolds LF, Dalton JE, Mascha EJ, Kurz A. The association between preoperative anemia and 30-day mortality and morbidity in noncardiac surgical patients. Anesth Analg. 2013;117:909–15
48. von Elm E, Altman DG, Egger M, Pocock SJ, Gøtzsche PC, Vandenbroucke JPSTROBE Initiative. . The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. PLoS Med. 2007;4:e296
49. Steyerberg EW Clinical Prediction Models. 2010 New York, NY Springer
50. Steyerberg EW, Eijkemans MJ, Harrell FE Jr, Habbema JD. Prognostic modeling with logistic regression analysis: in search of a sensible strategy in small data sets. Med Decis Making. 2001;21:45–56
51. Peduzzi P, Concato J, Kemper E, Holford TR, Feinstein AR. A simulation study of the number of events per variable in logistic regression analysis. J Clin Epidemiol. 1996;49:1373–9
52. Steyerberg EW, Schemper M, Harrell FE. Logistic regression modeling and the number of events per variable: selection bias dominates. J Clin Epidemiol. 2011;64:1464–5
53. Steyerberg EW, Eijkemans MJ, Habbema JD. Stepwise selection in small data sets: a simulation study of bias in logistic regression analysis. J Clin Epidemiol. 1999;52:935–42
54. Brasher PM, Beattie WS. Adjusting clinical prediction rules: an academic exercise or the potential for real world clinical applications in perioperative medicine? Can J Anesth. 2009;56:190–3
55. Cook NR, Ridker PM. Advances in measuring the effect of individual predictors of cardiovascular risk: the role of reclassification measures. Ann Intern Med. 2009;150:795–802
56. Ridley S. Cardiac scoring systems—what is their value? Anaesthesia. 2003;58:985–91
57. Fagan TJ. Letter: nomogram for Bayes theorem. N Engl J Med. 1975;293:257
58. Rodseth RN, Biccard BM, Le Manach Y, Sessler DI, Lurati Buse GA, Thabane L, Schutt RC, Bolliger D, Cagini L, Cardinale D, Chong CP, Chu R, Cnotliwy M, Di Somma S, Fahrner R, Lim WK, Mahla E, Manikandan R, Puma F, Pyun WB, Radović M, Rajagopalan S, Suttie S, Vanniyasingam T, van Gaal WJ, Waliszek M, Devereaux PJ. The prognostic value of pre-operative and post-operative B-type natriuretic peptides in patients undergoing noncardiac surgery: B-type natriuretic peptide and N-terminal fragment of pro-B-type natriuretic peptide: a systematic review and individual patient data meta-analysis. J Am Coll Cardiol. 2014;63:170–80
59. Hlatky MA, Greenland P, Arnett DK, Ballantyne CM, Criqui MH, Elkind MS, Go AS, Harrell FE Jr, Hong Y, Howard BV, Howard VJ, Hsue PY, Kramer CM, McConnell JP, Normand SL, O’Donnell CJ, Smith SC Jr, Wilson PWAmerican Heart Association Expert Panel on Subclinical Atherosclerotic Diseases and Emerging Risk Factors and the Stroke Council. . Criteria for evaluation of novel markers of cardiovascular risk: a scientific statement from the American Heart Association. Circulation. 2009;119:2408–16
60. Khuri SF, Henderson WG, Daley J, Jonasson O, Jones RS, Campbell DA Jr, Fink AS, Mentzer RM Jr, Steeger JEPrincipal Site Investigators of the Patient Safety in Surgery Study. . The patient safety in surgery study: background, study design, and patient populations. J Am Coll Surg. 2007;204:1089–102
61. Detsky AS, Abrams HB, Forbath N, Scott JG, Hilliard JR. Cardiac assessment for patients undergoing noncardiac surgery. A multifactorial clinical risk index. Arch Intern Med. 1986;146:2131–4
62. Bartels C, Bechtel JF, Hossmann V, Horsch S. Cardiac risk stratification for high-risk vascular surgery. Circulation. 1997;95:2473–5
63. Haynes AB, Regenbogen SE, Weiser TG, Lipsitz SR, Dziekan G, Berry WR, Gawande AA. Surgical outcome measurement for a global patient population: validation of the Surgical Apgar Score in 8 countries. Surgery. 2011;149:519–24
64. Devereaux PJ, Goldman L, Cook DJ, Gilbert K, Leslie K, Guyatt GH. Perioperative cardiac events in patients undergoing noncardiac surgery: a review of the magnitude of the problem, the pathophysiology of the events and methods to estimate and communicate risk. CMAJ. 2005;173:627–34
65. Redman LA, Naidoo P, Biccard BM. HIV, vascular surgery and cardiovascular outcomes: a South African cohort study. Anaesthesia. 2014;69:208–13
66. Johnson LF, Mossong J, Dorrington RE, Schomaker M, Hoffmann CJ, Keiser O, Fox MP, Wood R, Prozesky H, Giddy J, Garone DB, Cornell M, Egger M, Boulle AInternational Epidemiologic Databases to Evaluate AIDS Southern Africa Collaboration. . Life expectancies of South African adults starting antiretroviral treatment: collaborative analysis of cohort studies. PLoS Med. 2013;10:e1001418
67. Monsuez JJ, Charniot JC, Escaut L, Teicher E, Wyplosz B, Couzigou C, Vignat N, Vittecoq D. HIV-associated vascular diseases: structural and functional changes, clinical implications. Int J Cardiol. 2009;133:293–306
68. Chetty R, Batitang S, Nair R. Large artery vasculopathy in HIV-positive patients: another vasculitic enigma. Hum Pathol. 2000;31:374–9
69. Madiba TE, Muckart DJ, Thomson SR. Human immunodeficiency disease: how should it affect surgical decision making? World J Surg. 2009;33:899–909
70. Lin PH, Bush RL, Yao Q, Lam R, Paladugu R, Zhou W, Chen C, Lumsden AB. Abdominal aortic surgery in patients with human immunodeficiency virus infection. Am J Surg. 2004;188:690–7
71. Payne CJ, Bryce GJ, Gibson SC, Kingsmore DB. The Revised Cardiac Risk Index performs poorly in patients undergoing major vascular surgery: a prospective observational study. Eur J Anaesthesiol. 2013;30:713–5
72. Bertges DJ, Goodney PP, Zhao Y, Schanzer A, Nolan BW, Likosky DS, Eldrup-Jorgensen J, Cronenwett JLVascular Study Group of New England. . The Vascular Study Group of New England Cardiac Risk Index (VSG-CRI) predicts cardiac complications more accurately than the Revised Cardiac Risk Index in vascular surgery patients. J Vasc Surg. 2010;52:674–83, 683.e1–683.e3
73. Welten GM, Schouten O, van Domburg RT, Feringa HH, Hoeks SE, Dunkelgrün M, van Gestel YR, Goei D, Bax JJ, Poldermans D. The influence of aging on the prognostic value of the revised cardiac risk index for postoperative cardiac complications in vascular surgery patients. Eur J Vasc Endovasc Surg. 2007;34:632–8
74. Moodley Y, Naidoo P, Biccard BM. The South African Vascular Surgical Cardiac Risk Index (SAVS-CRI): a prospective observational study. S Afr Med J. 2013;103:746–50
75. Schouten O, Kok NF, Hoedt MT, van Laanen JH, Poldermans D. The influence of aneurysm size on perioperative cardiac outcome in elective open infrarenal aortic aneurysm repair. J Vasc Surg. 2006;44:435–41
76. Garcia S, Moritz TE, Goldman S, Littooy F, Pierpont G, Larsen GC, Reda DJ, Ward HB, McFalls EO. Perioperative complications after vascular surgery are predicted by the revised cardiac risk index but are not reduced in high-risk subsets with preoperative revascularization. Circ Cardiovasc Qual Outcomes. 2009;2:73–7
77. Ray P, Le Manach Y, Riou B, Houle TT. Statistical evaluation of a biomarker. Anesthesiology. 2010;112:1023–40
78. Bryce GJ, Payne CJ, Gibson SC, Kingsmore DB, Byrne DS, Delles C. Risk stratification scores in elective open abdominal aortic aneurysm repair: are they suitable for preoperative decision making? Eur J Vasc Endovasc Surg. 2012;44:55–61
79. Patterson BO, Holt PJ, Hinchliffe R, Loftus IM, Thompson MM. Predicting risk in elective abdominal aortic aneurysm repair: a systematic review of current evidence. Eur J Vasc Endovasc Surg. 2008;36:637–45
80. Brunelli A, Varela G, Salati M, Jimenez MF, Pompili C, Novoa N, Sabbatini A. Recalibration of the revised cardiac risk index in lung resection candidates. Ann Thorac Surg. 2010;90:199–203
81. Wotton R, Marshall A, Kerr A, Bishay E, Kalkat M, Rajesh P, Steyn R, Naidu B. Does the revised cardiac risk index predict cardiac complications following elective lung resection? J Cardiothorac Surg. 2013;8:220
82. Ferguson MK, Celauro AD, Vigneswaran WT. Validation of a modified scoring system for cardiovascular risk associated with major lung resection. Eur J Cardiothorac Surg. 2012;41:598–602
83. Biccard BM, Lurati Buse GA, Burkhart C, Cuthbertson BH, Filipovic M, Gibson SC, Mahla E, Leibowitz DW, Rodseth RN. The influence of clinical risk factors on pre-operative B-type natriuretic peptide risk stratification of vascular surgical patients. Anaesthesia. 2012;67:55–9
84. Biccard BM, Devereaux PJ, Rodseth RN. Cardiac biomarkers in the prediction of risk in the non-cardiac surgery setting. Anaesthesia. 2014;69:484–93
85. Biccard BM, Naidoo P, de Vasconcellos K. What is the best pre-operative risk stratification tool for major adverse cardiac events following elective vascular surgery? A prospective observational cohort study evaluating pre-operative myocardial ischaemia monitoring and biomarker analysis. Anaesthesia. 2012;67:389–95
86. Gillmann HJ, Meinders A, Grohennig A, Larmann J, Bünte C, Calmer S, Sahlmann B, Rustum S, Aper T, Lichtinghagen R, Koch A, Teebken OE, Theilmeier G. Perioperative levels and changes of high-sensitivity troponin T are associated with cardiovascular events in vascular surgery patients. Crit Care Med. 2014;42:1498–506