 |
Paul Older MB BS, LRCP MRCS, FRCA, FANZCA, FFICANZCA,
FJFICM
Adrian Hall MB BS, FANZCA, FJFICM
Western Hospital
Footscray 3011
Melbourne
AUSTRALIA
E-mail paul.older@culham.nex.au
To return to this point - Ctrl+home
Introduction
Are we replacing ignorance by fallacy?
Current preoperative assessment
Myocardial ischaemia
Ventricular function
Cardiopulmonary Exercise Testing: Measurement versus
estimates
Figure 1
The basis for cardiopulmonary exercise testing
That cardiac failure is of more serious prognostic
significance than myocardial ischemia
Cardiopulmonary exercise testing studies and outcome
Conclusion
References
That cardiac disease is a potent cause of postoperative morbidity and mortality is not open to question. The question should be, is myocardial ischemia or cardiac failure the greater threat to the surgical patient? It is our contention that cardiac failure in the postoperative period, is the most important issue and that myocardial ischemia, as an isolated finding, may not be as important as is often thought.
Historically, recent myocardial infarction [1] and congestive cardiac failure [2] were recognized as being associated with high mortality. In 1987 the Confidential Enquiry into Perioperative Deaths [3] highlighted, in a series of over 500,000 patients, that the majority of postoperative deaths occurred in elderly patients, with pre-existing cardiac or pulmonary disease, undergoing major surgery. In 1995 another report from Finland [4] showed the same findings, this time in over 325,000 patients. These articles served to highlight the work of Goldman et al. who published one of the first indices for identification of cardiac risk in non-cardiac surgery in 1977 [2]. Well before this, in 1960, Clowes and Del Guercio had related operative mortality specifically to poor ventricular function [5].
Identification of high-risk patients is of value only if there is a change in the management prompted by this finding. This is important for the effective use of Intensive Care Unit (ICU) beds for post surgical patients. The objective should be to identify preoperatively the patients at risk and triage them for specialist management in the ICU before they develop cardiopulmonary complications. The concept of admitting patients to ICU postoperatively when they have deteriorated on the ward results in poor outcomes due to the high severity of illness at the time of ICU admission. The issues are: How do we identify high-risk patients; what specifically do we look for and what tests do we perform?
ARE WE REPLACING IGNORANCE BY FALLACY?
Much of the literature, particularly that targeting an anesthetic audience, emphasizes myocardial ischemia and myocardial infarction as the main concerns for surgical patients [6,7]. One of the consequences of this approach is that identification of risk factors and preoperative screening tests for cardiac disease tend to concentrate solely on detection of myocardial ischemia and not its effect on ventricular function. In fact, risk factors for cardiac disease in general and for myocardial ischemia in particular, have a common end point - impaired ventricular function. Other important consequences are that perioperative management is stratified only according to risk of ischemia and infarction as identified from these tests. Therapy is then directed to prevention of myocardial ischemia and not optimization of cardiac performance.
Anesthesiologists accept this approach, as their problem is the management of the patient during and immediately after surgery. During this period, global and also myocardial oxygen demand are extremely low with oxygen consumption figures of less than 70 ml/min/m2 [8] (cf resting values 110-140 ml/min/m2). It is unlikely that cardiac failure, in terms of oxygen delivery will manifest itself as a problem at that time. Most anesthesiologists are not directly involved in the postoperative care of these patients and may not be aware of the high oxygen consumption figures that occur in the first 48 hours following major surgery. Major intra-cavity surgery, even in the elderly, is associated with an increase of over 40% in oxygen consumption to 150 ml/min/m2 or higher necessitating a similar increase in cardiac output [9]. This response is neurohumoral and is not ablated by differing anesthetic techniques or adequate postoperative pain relief. It is a response to the imposed stresses of the postoperative period and, even now, is tacitly ignored although Clowes and Del Guercio identified it 40 years ago [5].
Postoperative cardiac failure is difficult to identify in the absence of invasive monitoring. It is a discussion of forward flow rather than elevation of the central venous pressure or rales in the chest, as found with the classical picture of congestive cardiac failure. In 2000 we defined postoperative cardiac failure as the inability of the heart to meet the demand of postoperative stress [10]. It may only be apparent postoperatively when oxygen demand is increased. It may occur independent of both cardiac failure, in the traditional sense, and myocardial ischemia, although all three may coexist.
CURRENT PREOPERATIVE ASSESSMENT
Various risk indices for non-cardiac surgery have been developed over the last twenty years that emphasize the symptoms of chronic cardiac disease and inadequacy of myocardial perfusion [2,11]. It is noteworthy that preoperative tests for myocardial ischemia are readily and frequently performed, whereas tests for assessment of cardiac failure and evaluation of ventricular function are, in the main, less frequently performed.
Bodenheimer [12] suggests that ‘if the patient has clinically stable angina or has only risk factors for coronary artery disease, non-invasive testing adds little but confusion’. He goes on to say that in the absence of unstable angina the patients’ cardiac status ‘does not warrant evaluation in and of itself’. He bases this argument on the poor positive predictive value of non-invasive testing for postoperative cardiac events. In fact these statements deny the role of cardiac failure as a risk factor. He poses the right question when he asks, "Why do patients experience adverse events and how might these events be prevented?" Unfortunately ‘adverse events’ in this context relates to myocardial ischemia and infarction, not post-operative cardiac failure. He also makes the statement "If postoperative stress is accepted as the etiologic mechanism responsible for postoperative events then the optimal strategy becomes clear" and advocates reduction in postoperative oxygen consumption as a possible management strategy. This latter statement clearly implies that he is discussing an increase in postoperative oxygen demand and the associated increase in cardiac output. The focus should then be on ventricular function under conditions of stress – not just a discussion of myocardial ischemia. Myocardial ischemia may be a cause of ventricular dysfunction but as we have pointed out ventricular dysfunction may exist as a sole entity.
Our concept is that many patients, with or without ischaemia, have impairment of ventricular function that is not clinically detectable and that the preoperative detection of such cardiac failure has major predictive value [13,14,15].
In general, tests of ventricular function are little more than estimates based on clinical history; examples include the American Society of Anesthesiologists Classification of Physical Status published in 1963 [16] or the New York Heart Association (NYHA) Classification of Functional Status as published in 1973 [17]. The latter is a subjective evaluation of limitation of physical activity and was not intended as a preoperative screening test. In 1988 Dunselman et al [18], found in a study using cardiopulmonary exercise testing (CPX), that there was considerable overlap of function between Classes I and II and Classes III and IV. We consider that making distinctions between these classes is crucial to management. The study also revealed a discrepancy in one third of the cases between subjective and objective assessment of severity of heart failure. Dunselman concluded that ‘only data from exercise studies showed differences between the groups’.
To overcome the subjectivity (or lack of objectivity) of assessment in the original document of 1973, a revised version of the NYHA Classification was published in 1994 [19] and included what was termed an ‘Objective Assessment’. This was based on chest radiograph, resting ECG, ECG stress tests, echocardiograms and other radiological images. There was no reference to the use of data from exercise studies. The 1994 document acknowledges, "grading is based on the individual physician’s judgment", and still uses imprecise terms, such as ‘minimal’, ‘moderately severe’ or ‘severe’. In our view these guidelines remain unsatisfactory as they still fail to provide a truly objective assessment of cardiac function.
In 1996 the American College of Cardiology and the American Heart Association (ACC/AHA) published a set of guidelines for preoperative evaluation of patients for non-cardiac surgery [20]. It was pointed out that the presence of coronary artery disease in the presence of a good functional capacity was not a high-risk situation.
In 1993 our group had already published data showing that myocardial ischemia, in the absence of heart failure as defined by CPX tests, had little or no effect on postoperative outcome [13]. In the same study we showed that the combination of cardiac failure and ECG evidence of ischemia at low work rates was associated with a high incidence of postoperative cardiac events. We also showed that the incidence of an abnormal exercise ECG in patients over 65 without any cardiovascular history is about 24% [14]. The 1996 Guidelines support this finding and state that between 20% and 25% of patients with a normal resting ECG will have an abnormal exercise ECG.
The ACC/AHA guidelines of 2002 [21] specifically state that myocardial ischemia at high-levels of exercise (greater than 85% of age predicted) is a low risk situation. However we take issue with the statement in "Recommendations: When and Which Test", that "in most ambulatory patients, the test of choice is exercise ECG testing, which can provide an estimate of functional capacity and detect myocardial ischemia through changes in the ECG and hemodynamic response". An ECG stress test provides a very poor estimate of functional capacity; in addition the instantaneous estimate of actual work rate is very inaccurate [22]. Measurement of blood pressure and pulse rate change during exercise does not constitute measurement of the hemodynamic response, i.e. adequacy of forward flow (cardiac output) and oxygen delivery. In a study comparing invasive and non-invasive blood pressure measurement during exercise, we showed that there was poor correlation between the two techniques. In particular, the maximum values obtained were very much higher with the invasive method. Blood pressure measurements made only 20 seconds after exercise ceased, showed systolic readings 20-30 mmHg less than that achieved at end exercise (unpublished data, McGrath BP, Newman R, Older P 1989). ECG stress tests should not be used for assessment of hemodynamic response.
In 1999 Lee et al [23] published a validation of a clinical assessment for prediction of cardiac risk in non-cardiac surgery. The Revised Cardiac Risk Index is an enhancement of the original concept of Goldman et al of 1977 [2]; Professor Goldman is a co-author of the revised index. The index is based on historical or simple laboratory data and assigns scores to various factors, including history of congestive cardiac failure and surgery specific risk - the latter was included in the 1996 ACC/AHA Guidelines [20]. According to the definitions of the Revised Cardiac Risk Index, patients undergoing repair of abdominal aortic aneurysm, and thoracic and abdominal procedures were excluded from Class 1 (low risk). In addition, there was no relationship between risk class and major cardiac complications among patients who underwent aortic aneurysm surgery. Although claiming superiority over other published risk prediction indices, Lee acknowledges that the use of the Revised Cardiac Risk Index ‘remains to be defined’.
Other so-called objective measurements of ventricular systolic function including radionuclide angiography, transthoracic echocardiography or dobutamine stress echocardiography have been shown not to be reliable as screening tests for detection of operative risk in major surgery [10]. Both Higginbotham [24] and Cohen-Solal [25] have shown that ejection fraction does not correlate with cardiac failure as defined by aerobic capacity in patients with either coronary artery disease or heart failure. In 1981 Franciosa reported that no estimate of cardiac function including left ventricular end-diastolic dimensions, ejection fraction or treadmill time correlated well with measured exercised capacity [26]. This is not to decry the potential value of these tests in the appropriate situation. It is, however, to say that they are of doubtful value as screening tests for assessment of cardiac risk for non-cardiac surgery.
Preoperative assessment of ventricular function is frequently made in terms of the metabolic equivalent of common activities of daily living as described by the Duke Activity Status Index published in 1989 [27]. A metabolic equivalent, or MET, is the amount of oxygen consumed whilst sitting at rest by a 40 year old 70 kg male and equates to 3.5 ml O2/min/kg. The energy costs of various activities, in METs, are tabulated by many authorities [28].
In fact all of these ‘tests of ventricular function’ are actually surrogates of a true CPX test as described by Sue and Wasserman in 1991 [29].
CARDIOPULMONARY EXERCISE TESTING: MEASUREMENT VERSUS ESTIMATES.
As objective measurement of ventricular function is readily performed it seems pointless to use estimates. The most reliable and objective screening test for both myocardial ischemia and ventricular function is the cardiopulmonary exercise test [30,31].
A cardiopulmonary exercise test using respiratory gas analysis and, of choice, a bicycle ergometer, will give exactly the same ECG information as the treadmill test and does not rely on estimates of METs. The workload of a bicycle ergometer is known precisely and VO2 is measured directly; thus estimates of METs are rendered obsolete.
The ACC/AHA Guidelines suggest that a patient unable to perform an estimated 4 METs has poor functional capacity and is likely to need further investigation. They also suggest that a moderate functional capacity equates to 4-7 METs. They urge that every effort must be made to detect unsuspected heart failure by careful clinical history and examination. These statements themselves need examination.
In 1985 Weber and Janicki classified cardiac failure in terms of AT and peak aerobic capacity [32]. In this classification those patients with an AT greater than 14 ml/min/kg have no cardiac failure; those with AT between 11-14 ml/min/kg have mild cardiac failure; those with AT between 8-11ml/min/kg have moderate cardiac failure and those with AT less than 8 ml/min/kg have severe cardiac failure. The 4 METs exercise capacity quoted in the 2002 ACC/AHA Guidelines [21] equates to a sustained work rate of 14 ml/min/kg (4 x 3.5 ml/min/kg). Thus, using these guidelines, any patient defined objectively with any degree of heart failure by Weber and Janicki criteria, is high risk.
Our work measuring oxygen consumption directly via CPX suggests a different situation. In our current series of 1240 consecutive patients over 50 years of age scheduled for major surgery, 1004 cases had an AT of less than 14 ml/min/kg i.e. more than 80% (unpublished data Older P, Hall A). We have found that elderly patients undergoing major surgery who have an anerobic threshold (AT) of more than 11 ml/min/kg do not require ICU admission and do not have postoperative cardiac events i.e. postoperative myocardial infarction or cardiac failure, even if they have ischemia as detected during the CPX test [15]. This AT of 11 ml/min/kg equates to 3.1 METs or an average external workload of 50 watts. Only 383 cases of the 1240, i.e. 32%, had an AT of less than 11 ml/min/kg and using our criteria were classified as high risk. We do not believe it possible to make a clinical differentiation between patients with an AT in the range 11 ml/min/kg to 14 ml/min/kg. In our experience, validated by 2 published consecutive series totaling over 735 patients [13,15], this distinction is of major and pivotal importance in preoperative assessment and perioperative management.
The average AT of the study population of 1240 consecutive patients is 12.4 ml/min/kg (3.5 METs), i.e. between 11 ml/min/kg and 14 ml/min/kg. A sustained work rate of 7 METs, which is described as good functional capacity, would be equivalent to an AT of 25 ml/min/kg. This level of aerobic capacity equates to an external workload on bicycle ergometry of over 140 watts and would be unlikely in any patient over 60 years of age. In the series of 1240 patients to which we have referred there were two patients with an AT of greater than 25 ml/min/kg. It is clear that we need something much more accurate than estimates of functional capacity.
The Guidelines maintain that extremes of age are a minor clinical predictor of perioperative risk. With this we concur. Figure 1 shows AT versus age for 1240 patients. Age is clearly not an adequate discriminator of ventricular function in that one standard deviation of any age group embraces all the means of the entire cohort! In 1993 Paty et al [33] showed, in patients having surgery for resection of abdominal aortic aneurysms, that mortality was 3% in patients over 80 (n=116) and 2% in patients under 80 (n= 622).
Comparison of anerobic threshold (mean and standard deviation) with age for 1240 elderly surgical patients.
|
Professor Wasserman elegantly summarized this issue when he described the inherent dangers of using age as a discriminator of surgical risk. He pointed out that ‘we all age at different rates… using age will deny surgery to potential survivors and place younger patients with occult cardiopulmonary disease at great risk.’ [34]. We believe that measurement of physiologic function is more important than chronological age.
During exercise up to the AT, aerobic metabolism supplies the majority of ATP at a cellular level; after this there is an ever increasing need for supplementation by anerobic means, with a consequent lactic acidosis. This situation occurs not only in exercising muscle but also in any tissue where oxygen delivery is inadequate to meet energy requirement.
It was shown by Clowes and Del Guercio in 1960 [5], that non-survivors of major surgery failed to increase their cardiac output in the postoperative period and succumbed with an increasing metabolic acidosis; whilst survivors did increase their cardiac output. This was well before the era in which inotrope support was used. Although the means of management have now changed, i.e. postoperative monitoring with attention to optimization of oxygen delivery, the underlying physiology relating to major surgery remains the same.
In 1996 we postulated the concept of a ‘surgical anerobic threshold’ [14], this being the point where tissue oxygen delivery is unable to support aerobic generation of ATP. If aerobic metabolism is inadequate due to impaired ventricular function with consequent impairment of oxygen delivery, anerobic metabolism must supplement ATP production, but at the cost of lactic acid production. This will result in acidosis and impaired cellular function that is manifested as postoperative cardiac failure as described by Clowes and Del Guercio [5]. In our opinion the presence of a rising lactate in postoperative patients should prompt immediate investigation of the adequacy of oxygen delivery and global oxygen extraction. We have found that an elevated lactate is generally associated with inadequate cardiac output and oxygen delivery in the presence of high oxygen consumption and an oxygen extraction ratio of 30% or more.
THAT CARDIAC FAILURE IS OF MORE SERIOUS PROGNOSTIC SIGNIFICANCE THAN MYOCARDIAL ISCHEMIA.
Del Guercio, in 1980 [35], showed that it is possible to predict operative mortality by preoperative hemodynamic studies using a pulmonary artery catheter to evaluate ventricular performance. He found that elevation of pulmonary artery occlusion pressure at rest; intrapulmonary shunt greater than 20% and low cardiac index all had serious prognostic significance. He also found that inadequate oxygen transport is more likely to be detected by desaturation of mixed venous gas samples than by arterial desaturation. Patients with large arterio-venous oxygen differences at rest were classed as having moderate or advanced functional deficits because they were already using their mixed venous oxygen reserve. Changes in blood flow or oxygen demand postoperatively were thus more likely to result in tissue hypoxia in such patients.
Further, he showed that clinical evaluation was unable to identify high-risk patients. All patients in the study had been cleared for surgery by standard assessment but only 13.5% had normal measured hemodynamic, respiratory and oxygen transport function.
In a study in 1988 [9] we showed that 13% of patients evaluated in a similar fashion to those of Del Guercio had serious cardiac problems with 11% having a resting cardiac index of less than 2.2 l/min/m2. These patients had already been clinically evaluated and had been scheduled for major surgery. Seven of the one hundred study patients had their operation postponed due to problems that could not be rectified in the time available. Another six had their operations cancelled. One patient opted for surgery despite the risks and died on day two; three others identified as being in the high-risk group were in need of essential vascular surgery but died following surgery of cardiovascular complications.
These studies also imply that ventricular function is of major importance in determining postoperative morbidity and mortality and that methods relying solely on clinical evaluation are doomed to failure. These lessons may well have been forgotten.
In a study of 548 consecutive elderly patients scheduled for major intracavity surgery, published by our group in 1999 [15] we related outcome to cardiopulmonary function as assessed preoperatively by CPX. The overall mortality rate for major intra-abdominal surgery in that study was 3.9% (21 of 548). The incidence of myocardial ischemia, on exercise ECG criteria, was 24% (132 patients). Of the 21 deaths, twelve were unrelated to cardiopulmonary disease. The remaining nine deaths were attributable solely to poor cardiopulmonary function, with only two of these patients having myocardial ischemia identified preoperatively. Only one of these nine patients died following a myocardial infarction. Seven of the nine cardiovascular deaths had poor ventricular function identified preoperatively i.e. poor ventricular function was a more potent discriminator of death than myocardial ischemia. There were no deaths related to cardiopulmonary causes in any patient with adequate ventricular function as previously defined on CPX, i.e. an AT greater than 11ml/min/kg; even if myocardial ischemia had been detected.
If myocardial ischemia were the dominant factor causing perioperative morbidity then one would expect to see morbidity predominantly in the group with ischemia. This is clearly not the case.
Myocardial ischemia is caused by a failure of oxygen delivery to support the regeneration of high-energy phosphate in the myocardium needed to allow normal ventricular contraction. In 1996 [14] and in 2002 [36] we published data showing the relevance of the time of onset of myocardial ischemia during exercise testing. The majority of our patients have normal resting ECG’s and there appear to be two major patterns of ST segment change during the exercise test. Ischemia with onset early in exercise, i.e. before the AT, is associated with a low AT. In other patients the ischemia occurs late in exercise i.e. after the AT. We have found that the average AT for patients who developed ischemia at low work rates was 10.4 ml/min/kg whereas the AT averaged 13.9 ml/min/kg in those who developed ischemia late in exercise at higher work rates [14].
We contend that patients will reach their ‘surgical AT’ when oxygen demand and cardiac output rise postoperatively to the equivalent level where ‘cardiac failure’ requiring anerobic metabolism occurred during exercise, i.e. the exercise AT. This will result in impaired ventricular function manifesting as postoperative cardiac failure as defined above.
The clinical relevance of this is enormous. In patients with a lower AT, this postoperative ventricular dysfunction will be more likely and will occur at lower levels of postoperative stress, with consequently higher risk of morbidity and mortality. If myocardial ischemia develops and further impairs already poor ventricular function then morbidity and mortality will be even higher.
As minor surgery does not induce this extent of stress response the same patient may undergo lesser surgery without cardiovascular morbidity. In 1980 Backer et al [37] reported no cardiac complications in 288 ophthalmologic operations in patients with a previous myocardial infarction. This compared to a reinfarction rate of 6.1% in other surgery at the same hospital.
Significantly, in our studies [13,14,15], there has been no cardiovascular morbidity in patients with adequate ventricular function as defined by an AT of greater than 11 ml/min/kg. These patients are managed without need for high dependency or intensive care facilities. As a precautionary measure we have admitted patients to a high dependency unit who have late myocardial ischemia on CPX but with an AT of 11 ml/min/kg or greater. Currently we are running a trial to allow all such patients with or without ischemia to go to the ward. We still admit patients scheduled for intracavity surgery where such surgery is likely to be prolonged or result in high oxygen demand stress. This includes cases such as pancreatico-duodenectomy procedure, total gastrectomy or esophagectomy, this equates to Surgery Specific Risk as defined in the ACC/AHA Guidelines [21].
Postoperative outcome is mainly influenced by ventricular function. Tests to identify myocardial ischaemia will fail to detect cardiac failure and are inadequate as a screening test for identification of cardiac risk in noncardiac surgical patients. We use CPX testing as the sole test to evaluate cardiopulmonary function and myocardial ischemia, and modify perioperative management according to the result. We find that the degree of cardiac failure is the most important predictor of morbidity and mortality. This is independent of myocardial ischemia, however myocardial ischemia combined with moderate to severe cardiac failure (AT < 11ml/min/kg) is predictive of the highest morbidity and mortality.
1. Rao TK, Jacobs KH, El-Etr AA (1983) Reinfarction following anesthesia in patients with myocardial infarction. Anesthesiology 59:499-5065
2. Goldman L, Caldera DL, Nussbaum SR et al (1977) Multifactorial index of cardiac risk in non-cardiac surgical procedures. N Engl J Med 297:845-850
3. Buck N, Devlin HB, Lunn JN (1987) The report of a confidential enquiry into perioperative deaths. London: The Nuffield Provincial Hospitals Trust and the Kings Fund
4. Tikkanen J, Hovi-Viander H (1995) Deaths associated with anaesthesia and surgery in Finland in 1986 compared to 1995. Acta Anaesthesiol Scand 39:262-267
5. Clowes GHA, Del Guercio LRM (1960) Circulatory response to trauma of surgical operations. Metabolism 9:67-81
6. Fleisher LA (1992) Perioperative Myocardial Ischemia and Infarction. In : Beattie C, Fleisher LA (ed) International Anesthesiology Clinics, Volume 30 Number 1. Little, Brown and Company, Boston, pp 1-17
7. Aitkenhead AR (1993) Postoperative ischemia: cardiac morbidity after non-cardiac surgery. Lancet 341:731-732
8. Shoemaker WC, Thangathurai D, Wo CJC et al (1999) Intraoperative evaluation of tissue perfusion in high-risk patients by invasive and noninvasive hemodynamic monitoring. Crit Care Med 27:2147-2152
9. Older PO, Smith R (1988) Experience with the preoperative invasive measurement of haemodynamic, respiratory and renal function in 100 elderly patients scheduled for major abdominal surgery. Anaesth Intensive Care 16:389-95
10. Older P, Smith R, Hall A, French C (2000) Preoperative cardiopulmonary risk assessment by cardiopulmonary exercise testing. Critical Care and Resuscitation. 2:55-65
11. Detsky AS, Abrams HB, Forbath N, Scott JG, Hilliard JR (1986) Cardiac assessment in patients undergoing non-cardiac surgery. A multifactorial clinical risk index. Arch Intern Med 146:2131-2134
12. Bodenheimer MM (1996) Noncardiac surgery in the cardiac patient: What is the question? Ann Intern Med 124:763-766
13. Older P, Smith R, Courtney P, Hone R (1993) Preoperative evaluation of cardiac failure and ischaemia in elderly patients by cardiopulmonary exercise testing. Chest 104:701-704
14. Older PO, Hall AC (1996) The role of cardiopulmonary exercise testing for preoperative evaluation of the elderly. In: Wasserman K (ed) Exercise gas exchange in heart disease. Futura Publishing Company, Armonk New York, pp 287-297
15. Older P, Hall A, Hader R (1999) Cardiopulmonary exercise testing as a screening test for perioperative management of major surgery in the elderly. Chest 116:355-362
16. American Society of Anesthesiologists. (1963) New classification of physical status. Anesthesiology 24:111
17. The Criteria Committee of the New York Heart Association. (1973) Diseases of the heart and blood vessels. In : Nomenclature and criteria for diagnosis, 7th edn. Little, Brown and Company, Boston, pp 286
18. Dunselman PH, Kuntze CE, Van Bruggen A et al (1988) Value of New York Heart Association classification, radionuclide ventriculography, and cardiopulmonary exercise tests for selection of patients for congestive heart failure studies. Am Heart J 116:1475-1482
19. The Criteria Committee of the New York Heart Association. (1994) Diseases of the heart and blood vessels. In : Nomenclature and criteria for diagnosis, 9th edn. Little, Brown and Company, Boston, pp 253-256
20. Eagle KA, Brunage BH, Chaitman BR et al (1996) 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 93:1278-1317.
21. Eagle KA, Berger PB, Calkins H et al (2002) ACC/AHA Guideline update for perioperative cardiovascular evaluation for noncardiac surgery. Circulation 105:1257-1267
22. Wasserman K, Hansen JE, Sue DY, Whipp BJ, Casaburi R (1994) Protocols for exercise testing. In : Principles of exercise testing and interpretation, 2nd edn. Lea and Febiger, Malvern Pennsylvania, pp 103
23. Lee TH, Marcantoniao ER, Mangione CM et al (1999) Derivation and prospective validation of a simple index for prediction of cardiac risk of major noncardiac surgery. Circulation 100:1043-1049
24. Higginbotham MB (1996) Diastolic dysfunction and exercise gas exchange. In: Wasserman K, ed. Exercise gas exchange in heart disease. Futura Publishing Company, Armonk New York, pp 39-54
25. Cohen-Solal A (1996) Cardiopulmonary exercise testing in chronic heart failure. In: Wasserman K (ed) Exercise gas exchange in heart disease. Futura Publishing Company, Armonk New York, pp 17-38
26. Franciosa JA, Park M, Levine TB (1981) Lack of correlation between exercise capacity and indexes of resting left ventricular performance in heart failure. Am J Cardiol 47:33-39
27. Hlatky MA, Boineau RE, Higginbotham MB et al (1989) A brief self administered questionnaire to determine functional capacity (the Duke Activity Status Index). Am J Cardiol 64:651-654
28. Park KW (2001) Tests of myocardial function. In : Park KW (ed) International Anesthesiology Clinics, Volume 39 Number 4. Little, Brown and Company, Boston, pp 11-19
29. Sue DY, Wasserman K (1991) Impact of integrative cardiopulmonary exercise testing on clinical decision making. Chest 99:981-992
30. Kleber FX, Sabin GV, Winter UJ, et al (1997) Angiotensin-converting enzyme inhibitors in preventing in preventing remodeling and development of heart failure acute myocardial infarction: results of the German multicenter study of the effects of captopril on cardiopulmonary exercise parameters (ECCE). Am J Cardiol 80:162A-167A
31. Lipkin DP (1987) The role of exercise testing in chronic heart failure. Br Heart J 58:559-566
32. Weber KT, Janicki JS (1985) Cardiopulmonary exercise testing for evaluation of chronic cardiac failure. Am J Cardiol 55:22A-31A.
33. Paty PS, Lloyd WE, Chang BB, Darling RC, Leather RP, Shah DM (1993) Aortic replacement for abdominal aortic aneurysm in elderly patients. Am J Surg 166:191-193
34. Wasserman K (1993) Preoperative evaluation of cardiovascular reserve in the elderly. Chest 104:663-664
35. Del Guercio LRM, Cohn JD (1980) Monitoring operative risk in the elderly. JAMA 243:1350-55
36. Older PO, Hall AC (2002) Preoperative assessment of elderly surgical patients. In : Wasserman K (ed) Cardiopulmonary exercise testing and cardiovascular health. Futura Publishing Company, Armonk New York pp 119-133
37. Backer CL, Tinker JH, Robertson DM, Vlietstra RE (1980) Myocardial reinfarction following local anesthesia for ophthalmic surgery. Anesth Analg 59:267-262