Subjective methods for preoperative assessment of functional capacity (2024)

Journal List
BJA Educ
v.22(7); 2022 Jul
PMC9214434

As a library, NLM provides access to scientific literature. Inclusion in an NLM database does not imply endorsem*nt of, or agreement with, the contents by NLM or the National Institutes of Health.
Learn more: PMC Disclaimer | PMC Copyright Notice

BJA Educ. 2022 Jul; 22(7): 249–257.

Published online 2022 May 25. doi:10.1016/j.bjae.2022.02.007

PMCID: PMC9214434

PMID: 35754857

E. Silvapulle^1,^2,^∗ and J. Darvall^1,²

Author information Article notes Copyright and License information PMC Disclaimer

Associated Data

Supplementary Materials

Learning objectives;

By reading this article you should be able to:

•
Explain the principle aims of preoperative assessment of functional capacity.
•
Describe subjective methods for preoperative assessment of functional capacity, including unstructured methods (self-reported stair climbing, assessment of activities of daily living) and structured methods (e.g. the Duke Activity Status Index [DASI]).
•
Clarify the rationale, clinical applications and limitations of each method.
•
Describe the correlation with $\dot{V}$ o₂ peak and anaerobic threshold, and the accuracy in predicting postoperative outcomes for each method.

Key points

For individuals undergoing non-cardiac surgery:

•
The inability to climb two flights of stairs predicts cardiovascular and neurological events, with fair sensitivity (0.71) but poor specificity (0.47).
•
Unstructured assessment of activities of daily living is a poor predictor of cardiac complications (area under the receiver operating characteristic curve [AUROC]=0.66) and mortality (AUROC=0.52).
•
The DASI score inaccurately estimates peak oxygen uptake ( $\dot{V}$ o₂ peak) and should not be used to estimate metabolic equivalents.
•
A DASI score <34 is associated with 30-day mortality and myocardial infarction (odds ratio increases by 1.05 per 1 point below 34). However, when incorporated into a risk model, DASI has poor discriminatory value for these outcomes (AUROC=0.67).

The principal question that functional capacity assessment aims to address is ‘does this patient have the physiological reserve capacity to undergo this specific surgical procedure, without serious perioperative complications’?1 Functional capacity is defined as the maximum (or near-maximum) physical activity that an individual is capable of performing.2 It is frequently expressed in terms of the metabolic equivalent (MET), with 1 MET defined as an oxygen consumption of 3.5 ml kg⁻¹ min⁻¹ (based on the resting oxygen consumption of a 70 kg, 40-yr-old male).2^,3

‘Poor’ functional capacity has been defined as a functional capacity <4 METs. This cut-off was originally established in individuals with suspected coronary artery disease, based on lower survival rates in those able to achieve <4.6 METs on treadmill stress testing.4 Individuals with poor functional capacity are at increased risk of cardiac complications after major non-cardiac surgery.5

Asking a patient about their ability to climb one or two flights of stairs is a time-honoured method for assessing functional capacity. Frequently, clinicians also inquire about activities of daily living (ADLs) and exercise – this may take place in the form of an unstructured assessment (where clinicians devise their own questions) or a structured questionnaire. Understanding the physiological basis, clinical applications, accuracy in risk prediction and limitations of these methods is essential in appreciating their role in preoperative decision-making.

This article addresses both unstructured subjective methods (stair climbing assessment and ADL assessment) and structured subjective methods (Duke Activity Status Index [DASI]). In a forthcoming article, we will detail the objective methods available for preoperative assessment of functional capacity.6

Subjective methods of assessing functional capacity

The subjective methods used most commonly for the preoperative evaluation of functional capacity are stair climbing assessment, ADL assessment and the DASI score. These methods are used to estimate an individual's peak oxygen uptake which, in turn, is considered a surrogate marker for risk of postoperative complications.

The ‘gold standard’ measure of functional capacity is the maximum rate of oxygen uptake ( $\dot{V}$ o₂ max), which is measured by cardiopulmonary exercise testing (CPET).7 However, $\dot{V}$ o₂ max is a physiological endpoint that most individuals do not reach during CPET, whereas $\dot{V}$ o₂ peak reflects an individual's ‘best effort’, and is a clinically useful variable. Therefore, $\dot{V}$ o₂ peak is the measure against which other subjective methods are evaluated. The two most widely used CPET variables for patients undergoing major non-cardiac surgery are $\dot{V}$ o₂ peak and anaerobic threshold (AT). The AT is the $\dot{V}$ o₂ at which metabolic requirements exceed oxygen supply, and anaerobic metabolism takes place.8 A $\dot{V}$ o₂ peak <15 ml kg⁻¹ min⁻¹ and an AT <11 ml kg⁻¹ min⁻¹ have been established as the thresholds that predict significantly higher risk of perioperative morbidity and mortality.9^,10

Studies evaluating methods for assessing functional capacity aim to address two important questions:

(i)
See Also
What Are METs, and How Are They Calculated?Overview of Exercise Stress Testing The Compendium of Physical Activities: METs for All Comparison of Preoperative Assessment of Patient's Metabolic Equivalents (METs) Estimated from History versus Measured by Exercise Cardiac Stress Testing
How accurately does this method reflect the individual's true functional capacity – that is how well does it predict (or correlate with) with CPET-derived $\dot{V}$ o₂ peak and AT?
(ii)
Which thresholds predict postoperative complications, and how accurately do the thresholds predict these outcomes?

This article considers both questions for each subjective method.

Studies evaluating the relationship between subjectively assessed functional capacity and $\dot{V}$ o₂ peak have reported variables of correlation (such as Spearman's rho), agreement (Cohen's kappa) or prediction (positive predictive value [PPV], negative predictive value [NPV], area under the receiver operating characteristic curve [AUROC], positive likelihood ratio). In addition, various predictive measures have been reported to describe the ability of functional capacity assessment to (i) identify individuals at increased risk of postoperative complications (sensitivity, PPV); (ii) identify individuals at low risk of postoperative complications (specificity, NPV); or (iii) discriminate between individuals at high and low risk of perioperative complications (AUROC). With respect to predicting postoperative outcomes, a small number of studies have not reported measures of prediction, but rather measures of association (odds ratio, hazard ratio). Please see Supplementary Table S1 for definitions and interpretations of statistical terminology reported in pertinent studies.

Stair climbing

During subjective assessment of stair climbing capability, patients are asked if they can climb two flights of stairs or walk four blocks on flat ground, without stopping or experiencing limiting symptoms. Two flights is approximately 44 stairs in most hospitals and four blocks roughly equates to 600 m. Historically, the exertion of climbing two flights of stairs has been used as a surrogate to approximate the physiological stress of major surgery.11 This, in turn, has been said to correlate with a CPET-derived $\dot{V}$ o₂ peak of approximately 14 ml kg⁻¹ min⁻¹, although there is limited evidence for this correlation.12 A recent pilot study of 12 patients suggested the oxygen demand of the surgical stress response may be considerably lower than 4 METs.13

There are few data regarding the expected time frame for completion of a stair climb. In healthy volunteers, both males and females achieve a speed of approximately 1.8 stairs s⁻¹.14 However, studies in surgical patients have established an average speed of roughly 1 step s⁻¹.15 A study of patients scheduled for abdominal surgery evaluated a timed climb of 12 stairs; a stair climb duration of 26.5 s or more was associated with postoperative complications (myocardial ischaemia, wound complications, reoperation and mortality), whereas individuals who completed the stair climb in 15 s or less experienced no postoperative complications.15

Correlation with $\dot{V}$ o₂ peak

Correlating $\dot{V}$ o₂ peak with self-reported stair climbing ability is difficult and so is not widely reported. Studies have evaluated the relationship between witnessed symptom-limited stair climbing and measured $\dot{V}$ o₂ peak, with highly varied results. With each flight consisting of 22 individual stairs, studies in both non-surgical and surgical patients have observed two flights of stairs to correlate with a $\dot{V}$ o₂ peak ranging from 14.0 to 21.3 ml kg⁻¹ min⁻¹ (4.0–6.1 METs).16^,17 A secondary analysis of the Measurement of Exercise Tolerance before Surgery (METS) study, performed by Riedel and colleagues,18 established that asking a patient a single question about the ability to climb one flight of stairs had low predictive ability in identifying individuals with $\dot{V}$ o₂ peak > 16 ml kg⁻¹ min⁻¹ (AUROC=0.55).18 Therefore, there appears to be little consistency across the literature regarding the number of stairs that correlates with 4 METs.

Prediction of postoperative outcomes

A frequently cited study in support of self-reported stair climbing assessment by Reilly and colleagues demonstrated that poor functional capacity, defined as the inability to climb two flights of stairs or walk four blocks on level ground, had fair sensitivity (0.71), poor specificity (0.47) and modest positive likelihood ratio (1.3) for predicting cardiovascular, pulmonary, neurological and infectious complications after elective major non-cardiac surgery.19 Subsequent (and much larger) studies of stair climbing assessment have yielded conflicting results.

A large prospective study by Buse and colleagues20 evaluated 4,560 patients with known coronary, peripheral arterial or cerebrovascular disease undergoing inpatient non-cardiac surgery. Self-reported inability to climb two flights of stairs was associated with 30-day cardiac mortality or cardiac events for low-risk surgery (adjusted hazard ratio=1.44; 95% confidence interval [CI], 0.68–3.05) and high-risk surgery (adjusted hazard ratio=1.59; 95% CI, 1.17–2.16).20 The authors then analysed the incorporation of subjective functional capacity assessment into the Revised Cardiac Risk Index (RCRI). The RCRI uses six independent risk factors (history of ischaemic heart disease, history of congestive cardiac failure, history of cerebrovascular disease, diabetes with insulin preoperatively, preoperative creatinine ≥177 μmol L⁻¹, elevated risk surgery) to predict the risk of in-hospital mortality, myocardial infarction or cardiac arrest.21 Individuals with one or more RCRI risk factors are at increased risk (≥5%) of in-hospital myocardial infarction, cardiac arrest or mortality.22 Adding functional capacity to the RCRI only modestly improved the predictive value for 30-day cardiac events and mortality (AUROC=0.72, compared with AUROC=0.67 for RCRI alone); no threshold value was reported for RCRI. However, adding subjective functional capacity to the RCRI resulted in reclassification of 879 individuals to a more accurate risk category; 35 individuals were incorrectly reclassified.20 Of note, the participants in this study were older and higher risk, with significantly higher mortality rates, compared with patients in the METS study (Table 1).

Table1

Studies evaluating stair climbing assessment. n, number of patients. ∗Mean age 73.2 yrs. 61% of patients ASA Class ≥3. Recruited individuals had a history of cardiovascular disease. †MACE defined as a composite of cardiac mortality, myocardial infarction, acute heart failure, life-threatening arrhythmia. ^‡No threshold value for RCRI reported. ^¶Mean age 64.3 yrs. Fifty-eight percent of patients ASA Class ≥3. ^§Includes myocardial ischaemia or infarction, congestive heart failure, new ventilator support ≥24 h, delirium, stroke, surgical site infection, renal insufficiency, unplanned intensive care admission. AUROC, area under the receiver operating characteristic curve; CI, confidence interval; MACE, major adverse cardiac events. MET, metabolic equivalent; RCRI, Revised Cardiac Risk Index.

First author, year (n)	Surgical cohort	Outcomes	Thresholds for prediction	Accuracy of prediction
Buse, 202120 (4,560)	Inpatient, non-cardiac surgery∗	30-day cardiac mortality, MACE, 30-day and 1-yr mortality^†	Self-reported ability to climb two flights of stairs (achieve 4 METs)	Prediction of 30-day cardiac mortality and MACE^†: RCRI alone: AUROC=0.67 (95% CI, 0.65–0.71)^‡ Functional capacity plus RCRI: AUROC=0.72 (95% CI, 0.69–0.75)^‡
Reilly, 199919 (600)	Elective, major non-cardiac surgery^¶	Cardiovascular, pulmonary, neurological, infectious, miscellaneous complications^§	Poor exercise tolerance defined as self-reported inability to climb two flights of stairs or walk four blocks	Poor exercise tolerance: Sensitivity=0.71 Specificity=0.47 Positive likelihood ratio=1.3 for predicting outcomes

Open in a separate window

In summary, there remains considerable uncertainty over the number of stairs that correlates with 4 METs in surgical patients. Self-reported stair climbing assessment has only a modest ability to predict postoperative complications; further research is required to evaluate the predictive performance of stair climbing assessment in combination with other risk factors, such as those comprising the RCRI.

Assessment of ADLs

During preoperative assessment, patients are assessed on their ability to carry out ADLs. This informs clinicians of the most physically demanding activity a patient can perform, from which METS can be estimated. Functional capacity can then be classified as poor (<4 METs), moderate (4–6 METS), good (7–10 METS) or excellent (>10 METS).4

Many detailed compendia are available, which list activities and their corresponding METs. Box 1 contains an example of a brief compendium used commonly in perioperative medicine.23

Box 1

Estimated metabolic equivalents (METs) from activities of daily living.23

1
MET= Eating, getting dressed, working at a desk
2
METs= Showering, walking down eight steps
3
METs= Walking on a flat surface for one or two blocks
4
METs= Raking leaves, weeding, pushing a power mower, walking up 2 flights of stairs
5
METs= Walking 6.4 km h⁻¹, social dancing, washing a car
6
METs= Nine holes of golf carrying clubs, heavy carpentry, using a push mower
7
METs= Digging, spading soil, singles tennis, carrying 27 kg
8
METs= Moving heavy furniture, jogging slowly, carrying 9 kg upstairs
9
METs= Bicycling at a moderate pace, sawing wood, slow jump rope
10
METs= Brisk swimming, bicycling uphill, walking briskly, uphill jogs 9.7 km h⁻¹
11
METs= Cross-country skiing, full-court basketball
12
METs= Running continuously at 12.8 km h⁻¹

Alt-text: Box 1

Correlation with $\dot{V}$ o₂ peak

To date, there are no studies evaluating the relationship between self-reported ADLs and CPET-derived $\dot{V}$ o₂ peak. However, Weinstein and colleagues3 performed a retrospective study of 170 patients undergoing non-cardiac surgery who had an indication for cardiac stress testing. The authors compared the METs estimated from ADLs assessment with METs achieved on exercise stress testing using the Bruce protocol.4 The Bruce protocol (Box 2) is a common treadmill protocol for exercise stress testing, and permits an approximate estimation of METs achieved.24 The METs estimated by ADLs assessment were, on average, 3.3 METs lower than that achieved on cardiac stress testing. There was no agreement between the two methods (Cohen's kappa=0.02).4 It is interesting to note that, in 91% of patients, clinicians underestimated the METs that could be achieved on exercise stress testing (Table 2).

Box 2

Estimated metabolic equivalents (METs) during different stages of the Bruce exercise testing protocol.24

Stage 1 (2.7 km h⁻¹)= 4 METs ( $\dot{V}$ o₂ peak 14 ml kg⁻¹ min⁻¹)
See Also
Association of metabolic equivalent of task (MET) score in length of stay in hospital following radical cystectomy with urinary diversion: a multi-institutional study
Stage 2 (4 km h⁻¹)= 7 METs ( $\dot{V}$ o₂ peak 24.5 ml kg⁻¹ min⁻¹)
Stage 3 (5.4 km h⁻¹)= 10 METs ( $\dot{V}$ o₂ peak 35 ml kg⁻¹ min⁻¹)
Stage 4 (6.7 km h⁻¹) = 13 METs ( $\dot{V}$ o₂ peak 45.5 ml kg⁻¹ min⁻¹)
Stage 5 (8 km h⁻¹)= 16 METs $\dot{(V}$ o₂ peak 56 ml kg⁻¹ min⁻¹)

Alt-text: Box 2

Table2

Studies evaluating activities of daily living assessment. n, number of patients. ∗All individuals had an indication for cardiac stress testing; indications included unknown or suspected poor functional capacity, exertional angina. ^†No predictive variables reported in this study. ^‡Myocardial infarction, acute congestive heart failure, cardiac arrest, acute myocardial ischaemia, acute renal failure, stroke, respiratory failure, peripheral vascular occlusion. ADLs, activities of daily living; AUROC, area under the receiver operating characteristic curve; MET, metabolic equivalent.

First author, year (n)	Surgical cohort	Study design	Outcomes	Accuracy of prediction
Weinstein, 20183 (170)	Non-cardiac surgery∗	Retrospective	Compared METs achieved on history (ADLs assessment) vs METs achieved on cardiac stress testing	METs estimated by ADLs: 3.3 METs lower than that achieved on cardiac stress testing Agreement between the two methods: Cohen's kappa=0.02^†
Wiklund, 200125 (5,939)	Elective, inpatient non-cardiac surgery	Retrospective	Cardiovascular, respiratory, neurological or renal complications; mortality^‡	Functional capacity <4 METs (as estimated by ADLs): Predicted cardiac complications (AUROC=0.66) Predicted mortality (AUROC=0.52)

Open in a separate window

Prediction of postoperative outcomes

Use of the compendium of ADLs described (or various iterations thereof) to identify patients at increased risk of perioperative complications has not been widely investigated. In the largest study to date, Wiklund and colleagues retrospectively evaluated 5,939 hospital in-patients undergoing elective, non-cardiac surgery.25 As might be expected, age and ASA physical status were found to be significant predictors of cardiac complications (AUROC=0.81 and 0.74, respectively) and mortality (AUROC=0.78 and 0.80, respectively); threshold values for age and ASA physical status were not reported. However, low functional capacity (defined as the inability to achieve 4 METs on ADLs assessment) poorly predicted cardiac complications (AUROC=0.66) and mortality (AUROC=0.52) (Table 2).25

A recent study by Marsman and colleagues evaluated 4,879 patients aged 60 yrs and older undergoing elective non-cardiac surgery.26 Functional capacity was assessed with a questionnaire that also ascertained the reason for exercise limitation. Overall, 17% of patients were assessed as having poor functional capacity; 83% were assessed as having normal functional capacity. Subjectively assessed poor functional capacity (<4 METs, or unknown functional capacity) had very poor discriminatory ability (AUROC=0.55) and poor PPV (0.22), but good NPV (0.87) for predicting postoperative myocardial injury. A regression model comprising age and RCRI had fair predictive value for postoperative myocardial infarction (AUROC=0.71); the performance of this model was not improved with the addition of functional capacity assessment (AUROC=0.71). Threshold values were not reported for age or RCRI.26

In summary, limited data suggest that assessment of ADLs does not provide an accurate estimation of METs. Furthermore, the inability to achieve 4 METs on subjectively assessed ADLs has poor predictive value for postoperative cardiac complications.

Duke Activity Status Index

The DASI is a 12-item, self-reported questionnaire in which individuals report their ability to perform common daily physical activities and exercises of varying intensity.

Correlation with $\dot{V}$ o₂ peak and AT

Each question is assigned a weighting (provided in parentheses below), based on the known metabolic cost of that activity.

(i) Can you take care of yourself (eating, dressing, bathing or using the toilet)? (2.75)
(ii) Can you walk indoors, such as around your house? (1.75)
(iii) Can you walk a block or two on level ground? (2.75)
(iv) Can you climb a flight of stairs or walk up a hill? (5.50)
(v) Can you run a short distance? (8.00)
(vi) Can you do light work around the house such as dusting or washing dishes? (2.70)
(vii) Can you do moderate work around the house such as vacuuming, sweeping floors or carrying in groceries? (3.50)
(viii) Can you do heavy work around the house such as scrubbing floors, or lifting and moving heavy furniture? (8.00)
(ix) Can you do yard work such as raking leaves, weeding or pushing a power mower? (4.50)
(x) Can you have sexual relations? (5.25)
(xi) Can you participate in moderate recreational activities such as golf, bowling, dancing, doubles tennis, or throwing a baseball or football? (6.00)
(xii) Can you participate in strenuous sports such as swimming, singles tennis, football, basketball or skiing? (7.50)

The DASI score equals the sum of weightings from ‘Yes’ replies. A formula converts the final DASI score to estimate $\dot{V}$ o₂ peak and then METs27:

$\dot{V}$ o₂ peak (ml kg⁻¹ min⁻¹)=(0.43×DASI)+9.6.

METs= $\dot{V}$ o₂ peak/(3.5 ml kg⁻¹ min⁻¹)

This gives a minimum score of 0 points (2.74 METs) and a maximum score of 58.2 points (9.89 METs).

The DASI was developed and validated by Hlatky and colleagues in 1989. The development cohort consisted of 50 cardiology patients in whom the questionnaire was administered by an interviewer; in contrast, the validation cohort (50 cardiology patients) completed a self-administered questionnaire.27 Within the development cohort, there was strong correlation between DASI-estimated $\dot{V}$ o₂ peak and CPET-derived $\dot{V}$ o₂ peak (Spearman ρ=0.81, p<0.0001). However, there was much weaker correlation between the two variables in the validation cohort (Spearman ρ=0.58), suggesting the mode of administration potentially affected DASI scores.27

The METS study demonstrated that individuals with a DASI score of 34 (which converts to 6.92 METs) were only able to achieve $\dot{V}$ o₂ peak 17–18 ml kg⁻¹ min⁻¹ (approximately 5 METs) on CPET. The inaccuracy of the DASI conversion formula has been suggested in previous studies, and the cause is likely to be multifactorial. Firstly, DASI was self-administered in the METs study, whereas Hlatky and colleagues considered an interviewer-administered questionnaire to be more accurate.27 In addition, the results obtained by Hlatky and colleagues showed poorer accuracy of DASI-estimated $\dot{V}$ o₂ peak when CPET-derived $\dot{V}$ o₂ peak was less than 17.5 ml kg⁻¹ min⁻¹ (5 METs).27

A secondary cross-sectional analysis of the METS study developed a five-item modified DASI score (M-DASI), based on the five questions with the greatest ability to predict $\dot{V}$ o₂ peak > 16 ml kg⁻¹ min⁻¹ and AT > 11 ml kg⁻¹ min⁻¹.19 These questions were: (i) Are you able to climb a flight of stairs or walk up a hill? (ii) Are you able to do heavy work around the house? (iii) Are you able to do yard work? (iv) Are you able to have sexual relations? and (v) Are you able to participate in strenuous sports? Each question was assigned an equal weighting of 1 point. If a patient scored 0, the probability of achieving $\dot{V}$ o₂ peak > 16 ml kg⁻¹ min⁻¹ or AT >11 ml kg⁻¹ min⁻¹ on CPET was 23% and 20%, respectively. A score of 4 out of 5 predicted 76.1% and 58.7% likelihood of achieving $\dot{V}$ o₂ peak > 16 ml kg⁻¹ min⁻¹ and AT > 11 ml kg⁻¹ min⁻¹ on CPET, respectively. Using only five questions from the original questionnaire did not appear to compromise the predictive value of the DASI score; there was, in fact, a modest improvement in performance. However, the overall ability of M-DASI to predict $\dot{V}$ o₂ peak >16 ml kg⁻¹ min⁻¹ was only fair (AUROC=0.73), with poor ability to predict AT >11 ml kg⁻¹ min⁻¹ (AUROC=0.67).19

Studies have shown that the DASI is relatively poor at predicting CPET variables, and that a significant portion of individuals with low DASI scores are, in fact, able to achieve satisfactory $\dot{V}$ o₂ peak and AT on CPET.28 A small study of 50 patients found that a DASI score >37.45 predicted $\dot{V}$ o₂ peak >15 ml kg⁻¹ min⁻¹ with PPV=1.0 and NPV=0.32, whereas a DASI score of >46 predicted AT > 11 ml kg⁻¹ min⁻¹ with PPV=1.0 and NPV=0.40. Overall, DASI had fair predictive power for $\dot{V}$ o₂ peak > 15 ml kg⁻¹ min⁻¹ (AUROC=0.77; 95% CI, 0.62–0.90) and AT ≥ 11 ml kg⁻¹ min⁻¹ (AUROC=0.77; 95% CI, 0.63–0.99), although no specific DASI thresholds were cited for these performance data.29 Therefore, it may be that DASI is better at identifying individuals with moderate-to-good functional capacity, who are at lower risk of postoperative complications.

Patients undergoing cancer surgery are a different cohort, in that they have often experienced deconditioning recently because of treatment for cancer. Neoadjuvant chemotherapy can reduce $\dot{V}$ o₂ peak by up to 30% and, therefore, it is likely that patients report their functional capacity before any treatment. As a result, DASI-estimated $\dot{V}$ o₂ peak has been found to overestimate CPET-derived $\dot{V}$ o₂ peak by 8 ml kg⁻¹ min⁻¹.30 Further studies are required to determine the accuracy and utility of DASI in these patients.

Prediction of postoperative outcomes

The METS study was a landmark prospective cohort study that compared the accuracy of subjective and objective methods of functional capacity assessment, with regard to ability to predict CPET-derived $\dot{V}$ o₂ peak and AT, and postoperative myocardial infarction and mortality.31 Functional capacity was assessed using an unstructured subjective method (although the details of clinicians' estimates were not provided). Structured subjective methods (DASI) and objective measurements (CPET-derived variables and N-terminal pro-B-type natriuretic peptide [NT-proBNP]) were also obtained. The predictive performance of the DASI alone (without other covariates) was not reported in this study. The authors found that, compared with a baseline risk model (comprising RCRI), addition of the DASI only minimally improved predictive value for 30-day mortality or myocardial infarction (AUROC=0.59 and 0.67, respectively). Compared with a baseline risk model comprising age, sex and RCRI, addition of DASI only marginally improved predictive value for 1-yr mortality (AUROC=0.65 and AUROC=0.69, respectively).

An interesting finding from the METS study was that, when combined with the RCRI, DASI scores were associated with 30-day mortality or myocardial infarction (adjusted odds ratio=0.91; 95% CI, 0.83–0.99), but $\dot{V}$ o₂ peak combined with RCRI was not associated with such complications (adjusted odds ratio=0.90; 95% CI, 0.71–1.16). The adjusted odds ratio was expressed with respect to 1 MET increase in $\dot{V}$ o₂ peak or DASI score. The basis for this finding may be that, compared with objective tests of cardiorespiratory function, the DASI indirectly measures other domains essential to postoperative recovery, such as musculoskeletal strength, cognition and frailty.31 Whether an individual develops cardiovascular complications postoperatively is likely to be determined by more than the ability of the patient's cardiorespiratory system to match oxygen supply with demand.32

A nested cohort analysis of the METS study demonstrated that incorporating DASI into a multivariate regression model (comprising RCRI and N-terminal pro-BNP) poorly predicted 30-day mortality or myocardial infarction (AUROC=0.67). However, a DASI score <34 was associated with increased odds of 30-day mortality or myocardial infarction (odds ratio increased by 1.05 per 1 point decrease below 34; 95% CI, 1.00–1.09) and moderate-to-severe complications (odds ratio increased by 1.03 per 1 point decrease below 34; 95% CI, 1.01–1.05).33 However, given the lower confidence limits approach 1.00, this threshold needs to be confirmed in future studies (Table 3).

Table3

Studies evaluating the Duke Activity Status Index (DASI). n, number of patients. ∗No M-DASI thresholds reported. ^†No predictive variables reported. ^‡Median age 65 yrs; 34% of patients ASA Class ≥3. ^¶Baseline model comprised age, sex and RCRI. ^§Mean age 79 yrs, mean DASI score 30.2 (6.45 METs – calculated using original conversion formula, by Hlatky and colleagues27). AUROC, area under the receiver operating characteristic curve; MET, metabolic equivalent; M-DASI, modified DASI; NPV, negative predictive value; OR, odds ratio; PPV, positive predictive value; RCRI, Revised Cardiac Risk Index.

First author, year (n)	Surgical cohort	Outcomes	Thresholds for prediction	Accuracy of prediction
Riedel, 202118 (1,455)	Elective non-cardiac surgery	Ability of a 5-item modified DASI score (M-DASI) to predict CPET-derived $\dot{V}$ o₂ peak and AT, compared with original (12-item) DASI	$\dot{V}$ o₂ peak > 16 ml kg⁻¹ min⁻¹, AT > 11 ml kg⁻¹ min⁻¹	Prediction of $\dot{V}$ o₂ peak > 16 ml kg⁻¹ min⁻¹: M-DASI: AUROC=0.73∗ Original DASI: AUROC=0.71 Prediction of AT > 11 ml kg⁻¹ min⁻¹: M-DASI: AUROC=0.67∗ Original DASI: AUROC=0.66
Wijeysundera, 202033 (1,546)	Elective non-cardiac surgery	30-day mortality, myocardial injury, myocardial infarction, moderate-to-severe complications	DASI score of 34	DASI < 34 associated with increased risk of 30-day mortality or myocardial infarction: OR=1.05 per 1 point decrease below 34 (95% CI, 1.00–1.09)^†
Wijeysundera, 201831 (1,401) METS study	Elective non-cardiac surgery^‡	30-day myocardial infarction, 30-day and 1-yr mortality	–	Correlation between $\dot{V}$ o₂ peak and DASI: Spearman ρ=0.43, p<0.0001^† Prediction of 30-day mortality or myocardial infarction: Baseline model: AUROC 0.59^¶ Baseline model + DASI: AUROC 0.67^¶
Struthers, 200829 (50)	Intra-abdominal surgery^§	Ability of DASI to predict CPET-derived values	$\dot{V}$ o₂ peak > 15 ml kg⁻¹ min⁻¹, AT > 11 ml kg⁻¹ min⁻¹	DASI > 37.45 predicted $\dot{V}$ o₂ peak > 15 ml kg⁻¹ min⁻¹: PPV 1.0 NPV 0.32 DASI > 46 predicted AT > 11 ml kg⁻¹ min⁻¹: PPV 1.0 NPV 0.40

Open in a separate window

Application of subjective methods in clinical practice

Owing to its simplicity and ease of execution, subjective reporting of stair climbing has remained common in preoperative anaesthetic assessment. However, the accuracy of this assessment relies on the patient's recall and self-reported ability (or estimation thereof) to climb stairs. There is also likely to be variability in responses, as there is currently no standardised stair climbing questionnaire for surgical patients.

The method of functional capacity assessment and risk stratification based on self-reported stair climbing is dichotomous – fewer than two flights of stairs denoting high perioperative risk, two or more flights of stairs denoting low risk. There are few data for accurate risk assessment based on an ordinal scale. It is also important to note that stair climbing is not, per se, a standardised measure – on average, each flight consists of 18–22 stairs, but the number and height of the stairs can vary significantly.12 Furthermore, clinicians rarely obtain information regarding the speed of ascent – an individual who takes 1 min to climb two flights of stairs (for cardiorespiratory reasons) clearly has significantly reduced functional capacity, compared with an individual who can climb two flights of stairs in 15 s.

For those patients in whom non-cardiorespiratory factors limit exercise (such as arthritis, back pain), stair climbing is not an appropriate assessment of physical fitness. It should also be noted that current guidelines for preoperative cardiovascular evaluation do not recommend stair climbing assessment as a sole screening tool for identifying patients who require exercise or stress testing.11^,34

Self-reported inability to climb two flights of stairs is associated with cardiovascular complications in patients with known cardiovascular disease undergoing high-risk (rather than low-risk) surgery.20 This may account for the poor ability of subjective assessment to predict 30-day mortality or myocardial infarction in the METS study (AUROC=0.57), as two thirds of participants were of ASA grade 1 or 2. Furthermore, there was an unexpectedly low incidence of complications (2% for 30-day myocardial infarction and <1% for 30-day mortality).31 Therefore, although stair climbing assessment is a poor predictor of true functional capacity and $\dot{V}$ o₂ peak, it may be useful in identifying individuals at increased risk of postoperative complications, particularly before high-risk surgery.

The assessment of ADLs has the advantage of being simple and inexpensive, and is often used to estimate functional capacity in patients who do not climb stairs or walk on a frequent basis. Although ADLs assessment would be expected to be reflective of peak functional capacity (compared with the binary assessment of stair climbing), there is currently no evidence for any correlation or agreement between subjectively estimated ADLs and CPET-derived $\dot{V}$ o₂ peak.4 Furthermore, studies have failed to demonstrate an ability of assessing ADLs to predict postoperative complications.25 This may be a result of poor inter-rater reliability amongst anaesthetists evaluating ADLs, as this method does not include a standardised (or validated) questionnaire. Interestingly, the European Society of Cardiology/European Society of Anaesthesiology Guidelines on non-cardiac surgery recommend the use of RCRI risk factors, plus an abridged compendium of ADLs, for cardiac risk assessment.34 The estimated METs (based on ADLs) are derived from DASI and a detailed compendium by Fletcher and colleagues.27^,35 However, current evidence does not support the use of unstructured ADLs assessment to evaluate functional capacity preoperatively.

The DASI is a simple, brief (<5 min) questionnaire with construct validity.31 The METS study found the DASI conversion formula (for estimating $\dot{V}$ o₂ peak) to be rather inaccurate, which corroborated findings from other studies.27^,33 At a DASI score of 34, the conversion formula overestimates functional capacity by 2 METs.31 The American College of Cardiology/American Heart Association Guideline on Perioperative Cardiovascular Evaluation recommends considering further cardiac testing if functional capacity <4 METs (as determined by DASI). However, this may result in some patients with poor functional capacity being categorised erroneously as having moderate–good functional capacity.11 As individuals achieving a DASI score below 34 may be at increased risk of postoperative cardiovascular complications, this may be a more appropriate threshold for considering further cardiac investigations.31 Although this needs substantiation in future studies, this is the most evidence-based threshold we have to date.33

Overall, functional capacity assessment is useful for guiding preoperative cardiovascular risk assessment. This, in turn, assists identification of patients who should be considered for preoperative cardiac testing, and potentially those who may benefit from optimisation of medical therapy, coronary revascularisation, or both, before high-risk surgery. It should be noted, however, that it may be difficult to establish a universal cut-off (or minimum level of function) that predicts complications accurately, as perioperative risk is highly variable and influenced by a patient's age and comorbidities; the surgical technique used; the operating surgeon; and the anaesthetic and analgesic strategies adopted.

Conclusions

Functional capacity assessment and risk stratification are integral components of preoperative evaluation. Subjective methods for assessing functional capacity, such as assessment of stair climbing and ADLs, have been in use for several decades. However, DASI is the only structured, validated method for preoperative assessment of functional capacity.

Assessment of stair climbing, ADLs and DASI all correlate poorly with CPET-derived variables. The inability to climb two flights of stairs may be useful to identify individuals at increased risk of postoperative cardiac complications, in those with known cardiovascular disease undergoing high-risk surgery. In low-risk patients, this method appears to be less predictive of postoperative sequelae. Overall, stair climbing assessment has not consistently shown good sensitivity for predicting complications after major non-cardiac surgery, emphasising the need for a standardised questionnaire.

Inquiring about ADLs has poor predictive value for all postoperative complications, and unstructured questionnaires should not be used for preoperative evaluation. Instead, clinicians should use the DASI, which provides a structured, validated method for assessing ADLs and functional capacity. The threshold established thus far is a DASI score of 34; this may be useful for identifying individuals at increased risk of cardiac complications, who require further cardiorespiratory investigations. However, clinicians should be cautious about estimating METs (and, in particular, deciding the need for further investigation) using the DASI conversion formula, as it can significantly overestimate functional capacity.

Finally, several studies have evaluated the incorporation of subjective functional capacity assessment (including DASI) into risk prediction models. However, this addition has not been shown to improve model performance. Therefore, there is currently insufficient evidence for using these ‘hybrid’ risk models for quantitative risk stratification.

Declaration of interests

The authors declare that they have no conflicts of interest.

Biographies

•

Earlene Silvapulle MMed(Periop) FANZCA is an anaesthetist at the Royal Melbourne Hospital, Australia, and a research fellow at the University of Melbourne. She sits on the Perioperative Medicine Committee, and has research and clinical interests in improving functional capacity assessment and risk prediction. Her PhD thesis is examining the relationship between functional capacity, health status and postoperative myocardial injury.

•

Associate Professor Jai Darvall MEpid PhD FANZCA FCICM is an anaesthetist and intensive care specialist and co-lead of the perioperative medicine service within the Department of Anaesthesia and Pain Management, Royal Melbourne Hospital. A senior lecturer at the University of Melbourne, he has particular research interests in the identification and management of frailty in critical care and perioperative medicine.

Notes

Matrix codes: 1H02, 2A03, 3I00

Footnotes

MCQs

The associated MCQs (to support CME/CPD activity) will be accessible at www.bjaed.org/cme/home by subscribers to BJA Education.

Supplementary data to this article can be found online at https://doi.org/10.1016/j.bjae.2022.02.007.

Supplementary material

The following is the Supplementary data to this article:

Multimedia component 1:

Click here to view.^{(22K, docx)}Multimedia component 1

References

1. Chand M., Armstrong T., Britton G., Nash G. How and why do we measure surgical risk? JR Soc Med. 2007;100:508–512. [PMC free article] [PubMed] [Google Scholar]

2. Arena R., Myers J., Williams M., et al. Assessment of functional capacity in clinical and research settings. A scientific statement from the American heart association committee on exercise, rehabilitation, and prevention of the council on clinical cardiology and the council on cardiovascular nursing. Circulation. 2007;116:329–343. [PubMed] [Google Scholar]

3. Weinstein A., Sigurdsson M., Bader A. Comparison of preoperative assessment of patient’s metabolic equivalents (METs) estimated from history versus measured by exercise cardiac stress testing. Anesthesiol Res Pract. 2018;2018:5912726. [PMC free article] [PubMed] [Google Scholar]

4. Mark D., Hlatky M., Harrell F., et al. Exercise treadmill score for predicting prognosis in coronary artery disease. Ann Intern Med. 1987;106:793–800. [PubMed] [Google Scholar]

5. James S., Jhanji S., Smith A., et al. Comparison of the prognostic accuracy of scoring systems, cardiopulmonary exercise testing, and plasma biomarkers: a single-centre observational pilot study. Br J Anaesth. 2014;112:491–497. [PubMed] [Google Scholar]

6. Silvapulle E., Darvall J. Objective methods for preoperative assessment of functional capacity. BJA Educ. 2022;22 [PMC free article] [PubMed] [Google Scholar]

7. Tran D. Cardiopulmonary exercise testing. Methods Mol Biol. 2018;1735:285–295. [PubMed] [Google Scholar]

8. Albouaini K., Egred M., Alahmar A., Wright D. Cardiopulmonary exercise testing and its application. Heart. 2007;93:1285–1292. [PMC free article] [PubMed] [Google Scholar]

9. Loewen G., Watson D., Kohman L., et al. Preoperative exercise VO₂ measurement for lung resection candidates: results of Cancer and Leukemia Group B Protocol 9238. JThorac Oncol. 2007;2:619–625. [PubMed] [Google Scholar]

10. Older P., Hall A., Hader R. Cardiopulmonary exercise testing as a screening test for perioperative management of major surgery in the elderly. Chest. 1999;116:355–362. [PubMed] [Google Scholar]

11. Fleisher L., Fleischmann K., Auerbach A., et al. 2014 ACC/AHA guideline on perioperative cardiovascular evaluation and management of patients undergoing non-cardiac surgery: a report of the American College of Cardiology/American Heart Association Task Force on Practice guidelines. Circulation. 2014;130:e278–e333. [PubMed] [Google Scholar]

12. Biccard B. Relationship between the inability to climb two flights of stairs and outcome after major non-cardiac surgery: implications for the pre-operative assessment of functional capacity. Anaesthesia. 2005;60:588–593. [PubMed] [Google Scholar]

13. Taylor H., Simons K., Wilmott C., Smith R., Bramley D. Afeasibility study to investigate post-operative oxygen consumption (POpOC) after colorectal surgery requiring bowel resection. Pilot Feasibility Stud. 2019;22:1–6. [PMC free article] [PubMed] [Google Scholar]

14. Halder A., Gao C., Miller M., Kuklane K. Oxygen uptake and muscle activity limitations during stepping on a stair machine at three different climbing speeds. Ergonomics. 2018;61:1382–1394. [PubMed] [Google Scholar]

15. Baker S., Waldrop M., Swords J., et al. Timed stair-climbing as a surrogate marker for sarcopenia measurements in predicting surgical outcomes. JGastrointest Surg. 2019;23:2459–2465. [PubMed] [Google Scholar]

16. Brunelli A., Xiume F., Refai M., et al. Peak oxygen consumption measured during the stair-climbing test in lung resection candidates. Respiration. 2010;80:207–211. [PubMed] [Google Scholar]

17. Pollock M., Roa J., Benditt J., Celli B. Estimation of ventilatory reserve by stair climbing. A study in patients with chronic airflow obstruction. Chest. 1993;104:1378–1383. [PubMed] [Google Scholar]

18. Riedel B., Li M.H., Lee C.H.A., et al.METS Study Investigators A simplified (modified) Duke Activity Status Index (M-DASI) to characterise functional capacity: a secondary analysis of the Measurement of Exercise Tolerance before Surgery (METS) study. Br J Anaesth. 2021;126:181–190. [PubMed] [Google Scholar]

19. Reilly D., McNeely M., Doerner D., et al. Self-reported exercise tolerance and the risk of serious perioperative complications. Arch Intern Med. 1999;159:2185–2192. [PubMed] [Google Scholar]

20. Buse G., Puelacher C., Gualandro D., et al. Association between self-reported functional capacity and major adverse cardiac events in patients at elevated risk undergoing non-cardiac surgery: a prospective diagnostic cohort study. Br J Anaesth. 2021;126:102–110. [PubMed] [Google Scholar]

21. Lee T., Marcantonio E., Mangione C., et al. Derivation and prospective validation of a simple index for prediction of cardiac risk of major non-cardiac surgery. Circulation. 1999;100:1043–1049. [PubMed] [Google Scholar]

22. Duceppe E., Parlow J., MacDonald P., et al. Canadian Cardiovascular Society guidelines on perioperative cardiac risk assessment and management for patients who undergo non-cardiac surgery. Can J Cardiol. 2017;33:17–32. [PubMed] [Google Scholar]

23. McAuley P., Myers J., Abella J., Froelicher V. Evaluation of a specific activity questionnaire to predict mortality in men referred for exercise testing. Am Heart J. 2006;151:890e1–8907. [PubMed] [Google Scholar]

24. Garner K., Pomeroy W., Arnold J. Exercise stress testing: indications and common questions. Am Fam Physician. 2017;96:293. 9A. [PubMed] [Google Scholar]

25. Wiklund R., Stein H., Rosenbaum S. Activities of daily living and cardiovascular complications following elective, non-cardiac surgery. Yale J Biol Medi. 2001;74:75–87. [PMC free article] [PubMed] [Google Scholar]

26. Marsman M., Jv Waes, Grobben R., Weersink C., Wv Klei. Added value of subjective assessed functional capacity before non-cardiac surgery in predicting postoperative myocardial injury. Eur J Prev Cardiol. 2020;11 2047487320906918. [PubMed] [Google Scholar]

27. Hlatky M., Boineau R., Higginbotham M., et al. Abrief self-administered questionnaire to determine functional capacity (the Duke Activity Status Index) Am J Cardiol. 1989;64:651–654. [PubMed] [Google Scholar]

28. Ainsworth B., Haskell W., Herrmann S., et al. 2011 Compendium of Physical Activities: a second update of codes and MET values. Med Sci Sports Exerc. 2011;43:1575–1581. [PubMed] [Google Scholar]

29. Struthers E., Erasmus P., Holmes K., et al. Assessing fitness for surgery: a comparison of questionnaire, incremental shuttle walk, and cardiopulmonary exercise testing in general surgical patients. Br J Anaesth. 2008;101:774–780. [PubMed] [Google Scholar]

30. Li M., Bolshinsky V., Ismail H., et al. Comparison of Duke Activity Status Index with cardiopulmonary exercise testing in cancer patients. JAnesth. 2018;32:576–584. [PubMed] [Google Scholar]

31. Wijeysundera D., Pearse R., Shulman M., et al. Assessment of functional capacity before major non-cardiac surgery: an international, prospective cohort study. Lancet. 2018;391:2631–2640. [PubMed] [Google Scholar]

32. Hanley C., Wijeysundera D. Perioperative risk assessment – focus on functional capacity. Curr Opin Anaesthesiol. 2021;34:309–316. [PubMed] [Google Scholar]

33. Wijeysundera D., Beattie W., Hillis G., et al. Integration of the Duke Activity Status Index into preoperative risk evaluation: a multicentre prospective cohort study. Br J Anaesth. 2020;124:261–270. [PubMed] [Google Scholar]

34. Kristensen S., Knuuti J., Saraste A., et al. 2014 ESC/ESA Guidelines on non-cardiac surgery: cardiovascular assessment and management. The Joint Task Force on non-cardiac surgery: cardiovascular assessment and management of the European Society of Cardiology (ESC) and the European Society of Anaesthesiology (ESA) Eur Heart J. 2014;35:2383–2431. [PubMed] [Google Scholar]

35. Fletcher G., Ades P., Kligfield P., et al. Exercise standards for testing and training: a scientific statement from the American Heart Association. Circulation. 2013;128:873–934. [PubMed] [Google Scholar]

Articles from BJA Education are provided here courtesy of Elsevier

Subjective methods for preoperative assessment of functional capacity (2024)

Associated Data

Learning objectives;

Key points

Subjective methods of assessing functional capacity

Stair climbing

Correlation with V˙o2 peak

Prediction of postoperative outcomes

Table1

Assessment of ADLs

Box 1

Estimated metabolic equivalents (METs) from activities of daily living.23

Correlation with V˙o2 peak

Box 2

Estimated metabolic equivalents (METs) during different stages of the Bruce exercise testing protocol.24

Table2

Prediction of postoperative outcomes

Duke Activity Status Index

Correlation with V˙o2 peak and AT

Prediction of postoperative outcomes

Table3

Application of subjective methods in clinical practice

Conclusions

Declaration of interests

Biographies

Notes

Footnotes

Supplementary material

References

Correlation with $\dot{V}$ o₂ peak

Correlation with $\dot{V}$ o₂ peak

Correlation with $\dot{V}$ o₂ peak and AT