Cardiopulmonary exercise testing—a beginner’s guide to the nine-panel plot

Key points.

• •Cardiopulmonary exercise testing (CPET) is a dynamic, non-invasive assessment of the cardiopulmonary system at rest and during exercise.
• •Deficiencies in the CPET-derived variables anaerobic threshold, peak oxygen consumption, and ventilatory efficiency for carbon dioxide are associated with poor postoperative outcomes.
• •The normal physiological response to exercise is a characteristic increase in heart rate, stroke volume, tidal volume, and ventilatory frequency.
• •Using a systematic approach, the nine-panel plot can be used to identify limitations in cardiac and respiratory capacity.
• •CPET aids risk assessment, identification of comorbidities that may be optimised, and perioperative planning.

Learning objectives.

By reading this article you should be able to:

• •Describe the basic physiology underlying a cardiopulmonary exercise test.
• •Demonstrate a structured approach to evaluating the nine-panel plot of a cardiopulmonary exercise test.
• •Explain the pattern of physiological changes associated with cardiac and respiratory limitation within the nine-panel plot.
• •Show how cardiopulmonary exercise testing can be used to assess patients undergoing major surgery.

It has long been recognised that the physiological stress of major surgery increases an individual’s baseline oxygen consumption ($\dot{V}$O₂), and that patients who are less physically fit are more likely to experience adverse perioperative outcomes. A hypothesis links these two observations: patients with insufficient cardiopulmonary capacity to increase O₂ delivery to match the increase in perioperative O₂ consumption are more likely to experience organ dysfunction.

Cardiopulmonary exercise testing (CPET) is a dynamic, non-invasive assessment of the cardiopulmonary system at rest and during exercise. The objective of CPET is to determine functional capacity in an individual. Deficiencies in CPET-derived variables—specifically ventilatory anaerobic threshold (AT), peak O₂ consumption ($\dot{V}$O_2peak), and ventilatory efficiency for carbon dioxide ($\dot{V}$_E/$\dot{V}$CO₂)—are associated with poor postoperative outcomes (mortality, morbidity, admission to intensive care, and length of hospital stay) after intra-abdominal surgery.1, 2, 3, 4 CPET is being used increasingly as part of a comprehensive perioperative assessment of high-risk patients. This helps to inform patients better of their individual perioperative risks, inform surgical decision-making, and plan both perioperative management and postoperative care.4, 5 CPET is used in 68% of departments performing elective surgery in the UK surgery to help assess elderly patients undergoing major procedures such as gastrointestinal, major vascular, major urology, and thoracic surgeries.⁶ CPET is also used extensively outside the UK, particularly in cardiology, respiratory medicine, and preoperative assessment. CPX International is a society that promotes the use of CPET worldwide.⁷ The indications for and contraindications of CPET, and the use of CPET as part of prehabilitation are beyond the scope of this article but have been detailed elsewhere.8, 9, 10, 11, 12

Cardiopulmonary response to exercise

Sustained exercise requires an increase in O₂ supply to meet the metabolic demands of the muscles. The normal cardiovascular response to exercise includes an increase in systolic blood pressure, a reduction in systemic vascular resistance (which facilitates increased muscle perfusion), and an increase in venous return to the heart, facilitated by the calf muscle pump. Cardiac output increases in proportion to the intensity of exercise, as a result of increases in both HR and stroke volume (SV). Exercise performance may be limited by many pathological factors such as systolic pump failure, diastolic filling abnormalities, and myocardial ischaemia. Minute ventilation also increases in proportion to the intensity of exercise through increases in both tidal volume (V_T) and ventilatory frequency, driven by carbon dioxide (CO₂) production. Patients with pulmonary disease may be unable to increase minute ventilation (${\dot{V}}_{\text{E}}$) sufficiently to keep pace with exercise-induced increases in CO₂ production, resulting in hypercapnia and hypoxaemia.

Cardiopulmonary exercise test

CPET is usually conducted on an electromagnetically braked cycle ergometer, with each test taking approximately 10 min. A wealth of data is gathered from the patient’s expired gases using a rapid gas analyser and a pressure differential pneumotachograph, with additional continuous 12-lead ECG, SpO₂ and non-invasive blood pressure (NIBP) monitoring. Measurements are taken at rest, during unloaded cycling (pedalling without any resistance), pedalling against a continuously increasing resistance (increasing work at a predetermined ramp rate), and in the recovery phase immediately after exercise. By inputting the individual’s height, weight, age, and gender into a CPET computer software package (e.g. BreezeSuite software MCG Diagnostics, St.Paul, MN, USA), predicted normal values can be calculated.

The average CPET records thousands of measurements, including:

• (i)The work rate (in Watts)
• (ii)Metabolic gas exchange measurements: O₂ consumption ($\dot{V}$O₂), CO₂ production ($\dot{V}$CO₂), and respiratory exchange ratio (RER = $\dot{V}$CO₂ divided by $\dot{V}$O₂)
• (iii)Ventilatory measurements: SpO₂, $\dot{V}$_E, V_T, respiratory rate (RR), ventilatory equivalents for O₂ ($\dot{V}$_E/$\dot{V}$O₂), and CO₂ ($\dot{V}$_E/$\dot{V}$CO₂)
• (iv)Cardiovascular variables: HR, ECG ST-segment changes, and NIBP.

These thousands of data points are represented graphically in a standard format called the nine-panel plot (Fig. 1).

Fig 1.

Normal nine-panel plot of a 47-yr-old male weighing 71 kg. ${\dot{V}}_{\text{E}}$ BTPS, minute ventilation at body temperature, ambient pressure saturated with water vapour (L min^–1); time (min); HR, heart rate (beats min^–1); $\dot{V}$O₂/HR, oxygen consumption divided by HR, known as the ‘oxygen pulse’ (ml beat^–1); $\dot{V}$O₂, oxygen consumption (L min^–1); $\dot{V}$CO₂, carbon dioxide production (L min^–1); work (W); ${\dot{V}}_{\text{E}}/\dot{V}{\text{O}}_2$, minute ventilation divided by oxygen consumption (dimensionless); ${\dot{V}}_{\text{E}}/\dot{V}$CO₂, minute ventilation divided by carbon dioxide production (dimensionless); Vt BTPS, tidal volume at body temperature, ambient pressure saturated with water vapour (L min^–1); RER, respiratory exchange ratio $\dot{\text{VC}}$O₂/$\dot{V}$O₂ (dimensionless); PETO₂, end-tidal oxygen tension (mm Hg); PETCO₂, end-tidal carbon dioxide tension (mm Hg); SpO₂, pulse oximeter saturations (%); Rec, recovery; MVV, maximum voluntary ventilation. The burgundy boxes in panels 2 and 5 represent the 80–100% range of maximal predicted HR; the purple box in panel 2 represents the 80–100% range of maximal predicted oxygen pulse; the red boxes in panels 3 and 5 represent predicted 80–100% range of maximal predicted $\dot{V}$O_2peak. The three black vertical lines represent the start of unloaded exercise, the start of loaded exercise and the cessation of loaded exercise. The green vertical dotted line in panel 7 represents the MVV.

Nine-panel plot

A ‘normal’ nine-panel plot is shown in Figure 1. The panels are numbered 1–9 from top left to bottom right. The cardiovascular system is represented by panels 2, 3, and 5; ventilation is represented by panels 1, 4, and 7; whereas panels 6, 8, and 9 show ventilation–perfusion relationships.

At first glance, the nine-panel plot may seem impossibly complex—but so too may the 12-lead ECG seem to a medical student. Most anaesthetists will feel comfortable in grossly interpreting an individual’s ECG, whilst recognising that expert analysis (e.g. by a cardiologist) may diagnose more subtle abnormalities. Similarly, although full interpretation of the nine-panel plot requires expertise, a basic interpretation of the data can be carried out using a systematic approach. The Perioperative Exercise Testing and Training Society (POETTS) provides recommendations on a standardised approach to reporting CPET tests.⁹

Systematic approach to interpreting the nine-panel plot

It is important to recognise what a ‘normal’ CPET looks like, and to understand common patterns of abnormality. We recommend asking a series of nine questions when interpreting a nine-panel plot.

• 1.Is the test maximal in terms of effort?
• It is important to know why an individual’s CPET was stopped. Did the patient achieve a ‘maximal effort’, that is, will the CPET-derived values of AT and $\dot{V}$O_2peak be valid, or was the test terminated for some other reason—musculoskeletal pain, hypotension, arrhythmias, leg claudication, ST-segment depression, or cardiac chest pain? Maximal effort is defined in a number of ways, for example:

  • (i)
    Achieving more than 80% of the predicted work. On the basis of the patient’s height, weight, age, and sex, the CPET software calculates the predicted maximum work and this can be compared with the work achieved at the end of the test. The increase in work during the test is seen as the black line with crosses increasing linearly in Figure 1, panel 3. In this example, the CPET software predicted a maximum work of 178 W and the patient achieved a maximum of 221 W during the test (124%).
  • (ii)
    Achieving a HR of >80% of the predicted maximum, calculated using the formula: Predicted maximum = 220 beats min^–1 – age. For ease of interpretation, the CPET software displays a box as shown in panel 2 representing the 80–100% range of predicted maximum HR. A maximal effort can then be identified as burgundy-coloured diamonds entering the burgundy box in panel 2 of Figure 1. In this example, the patient is 47 yrs old and thus the maximum predicted heart rate is 173 beats min^–1. During the CPET, his maximum recorded heart rate was 166 beats min^–1 and therefore deemed a maximal effort (96%).
  • (iii)
    Achieving a RER ($\dot{V}$CO₂/$\dot{V}$O₂) of more than 1.15, as seen in panel 8 of Figure 1. In this example, the maximum RER achieved was 1.32.

• 2.What is the $\dot{V}$O_2peak in panel 3?

• During progressively increasing exercise intensity, $\dot{V}$O₂ normally increases linearly in proportion to work. $\dot{V}$O_2max is the maximum $\dot{V}$O₂ achievable by an individual performing a specific type of exercise, and is usually reached by young, physically fit individuals during CPET. $\dot{V}$O_2max is often not achieved by elderly individuals who are physically deconditioned, have comorbidities, or both. Instead $\dot{V}$O_2peak, the highest $\dot{V}$O₂ measured, is recorded, and this is usually the $\dot{V}$O₂ at the point when the test is terminated. In Figure 1, the highest $\dot{V}$O₂ achieved (panel 3) is 2.7 L min^–1 which, given that the patient undertaking this test weighed 71 kg, equates to a $\dot{V}$O_2peak of 38 ml O₂ kg^–1 min^–1. Patients whose $\dot{V}$O_2peak is less than 15 ml O₂ kg^–1 min^–1 are at greater risk of perioperative complications.2, 3

• 3.Is the $\dot{V}$O₂–work relationship normal?

• As resistance is added to the cycle ergometer, an individual’s oxygen consumption normally increases at 10 ml O₂ min^–1 W^–1. The increase in $\dot{V}$O₂ (red dots) should parallel the increase in work (black crosses) as seen in panel 3 of Figure 1.

• 4.Can I determine AT in this test?

• This an important question. The AT is the point at which the O₂ demand of the muscles exceeds the ability of the cardiopulmonary system to supply O₂. Muscle cells will begin to generate ATP through anaerobic metabolism, a process which produces lactic acid. In turn, lactic acid is buffered by circulating bicarbonate, generating further CO₂. In most individuals this will be evident in panel 5 as the point where $\dot{V}$CO₂ increases disproportionately when compared with $\dot{V}$O₂. This is most easily seen by using the ‘V slope method’, in which two lines of best fit are drawn through the $\dot{V}$CO₂ data points plotted against $\dot{V}$O₂ in panel 5 (see Fig. 2). Initially $\dot{V}$O₂ and $\dot{V}$CO₂ increase at the same rate, but after AT the rate of CO₂ production increases, resulting in a second line with a steeper gradient. The $\dot{V}$O₂ at the point of intersection of the two lines of best fit is the AT.

• Sometimes, data scatter means that any sudden increase in $\dot{V}$CO₂ is less obvious, but AT can often be identified by using other parameters, such as the ventilatory equivalents for oxygen ($\dot{V}$_E/$\dot{V}$O₂) and carbon dioxide (${\dot{V}}_{\text{E}}/\dot{V}{\text{CO}}_2$). In panel 6, AT is the point at which ${\dot{V}}_{\text{E}}/\dot{V}{\text{O}}_2$ starts to increase whereas ${\dot{V}}_{\text{E}}/\dot{V}{\text{CO}}_2$ remains relatively constant or decreases slightly (see Fig. 2). This is because ${\dot{V}}_{\text{E}}$ increases disproportionately to $\dot{V}$O₂ but proportionally to $\dot{V}$CO₂ after AT. In up to 10% of tests however, it may not be possible to determine the AT.

Fig 2.

Panels 5 and 6 demonstrating anaerobic threshold (AT). For abbreviations, see legend to Figure 1. In panel 5 the burgundy diamonds demonstrate an increase in HR during the test. The blue squares in panel 5 represent the V slope method of determining AT with $\dot{V}$CO₂ (y axis) plotted against $\dot{V}$O₂ (x-axis). The sudden change to a steeper gradient indicates the onset of anaerobic metabolism, that is the AT. In panel 6 AT can also be determined as the point at which V_E/$\dot{V}$O₂ (red dots) starts to increase whereas V_E/$\dot{V}$CO₂ (blue squares) remains relatively constant.

• 5.If so, what is the $\dot{V}$O₂ at AT?

• AT is at the point at which the $\dot{V}$CO₂ data points in panel 5 form a new trajectory along a steeper gradient. CPET computer software packages allow you to draw a line of best fit along these data points—AT is the VO₂ at the point of intersection of these lines of best fit (see Fig. 2). In this example, AT occurs at $\dot{V}$O₂ of 1.45 L min^–1. In this 71 kg patient, AT therefore occurs at 20.4 ml O₂ kg^–1 min^–1. Patients whose AT is less than 10.2 ml O₂ kg^–1 min^–1 are at greater risk of perioperative complications.³

• 6.Was the HR response normal?

• A normal response to increasing exercise intensity is a linear increase in HR, as seen in Figure 1, panel 2. Immediately after the cessation of exercise, a rapid decrease in HR is seen. As HR is mainly responsible for the increase in cardiac output, rate-limiting medication (e.g. β-adrenergic receptor antagonists) may blunt this HR response and cause chronotropic incompetence.

• 7.Does the oxygen pulse increase with exercise?

• The oxygen pulse is the $\dot{V}$O₂ divided by HR, and represents the product of the stroke volume and the arterial-venous oxygen difference. It can be seen in panel 2 and can be viewed as a surrogate for stroke volume, and as such should increase at the start of exercise before slowly reaching a plateau at its highest predicted value.

• 8.Is there any ventilatory limitation?

• The normal ventilatory response to exercise is a linear increase in ${\dot{V}}_{\text{E}}$ with exercise intensity up to AT, followed by a disproportionately higher increase in ${\dot{V}}_{\text{E}}$, driven by increased CO₂ production. Healthy individuals have ample ventilatory capacity, and exercise is never limited by ventilation. However, exercise capacity may be limited in patients with obstructive or restrictive lung disease. Before the start of the CPET, forced expiratory volume in first second (FEV₁) and forced vital capacity (FVC) are measured using static spirometry. Maximum voluntary ventilation (MVV) is a measure of the maximum volume of air that can be inhaled and exhaled within 1 min. MVV can either be measured directly, or more commonly it is calculated indirectly by multiplying the FEV₁ by 40. In a normal patient, maximum ${\dot{V}}_{\text{E}}$ does not exceed 80% of MVV—the measured MVV is marked as a vertical green dotted line to the right-hand side of the x-axis in Figure 1, panel 7. In some CPET software packages, a second vertical line representing 80% of the MVV is also present. In addition, SpO₂ should remain above 95% throughout the test, and is recorded in Figure 1, panel 9.

• 9.Were there any ECG changes?

• During the CPET, continuous ECG and intermittent NIBP are recorded, and any abnormalities will be included within the CPET report. Cardiac ischaemia during the exercise test may result in characteristic ECG changes: ST depression of 1 mm in two adjacent leads or 2 mm in a single lead, T-wave inversion, left- or right-bundle branch block, or exercise-induced arrhythmias.

Cardiac limitation

Figure 3 shows the nine-panel plot of an 80-yr-old patient weighing 64 kg who underwent CPET as part of the preoperative assessment for repair of a 7 cm abdominal aortic aneurysm. His medical history included ischaemic heart disease, with a coronary artery bypass performed 20 yrs previously. On the basis of the patient’s height, weight, age, and sex, the CPET software predicted a maximum work of 112 W and a maximum HR of 140 beats min^–1. The patient terminated the CPET at 8 min because of fatigue, at which point he had reached a peak work of 84 W. His pre-test blood pressure was 158/84 mm Hg, and his post-test blood pressure was 139/96 mm Hg. Although the patient did not experience chest pain, there was significant ST depression on the ECG at peak exercise, but these resolved rapidly in the recovery period.

Fig 3.

Cardiac limitation. For abbreviations, see legend to Figure 1. The main abnormalities seen are: in panel 2 no discernible increase in oxygen pulse $\dot{V}$O₂ /HR (purple circles) with increasing work and a fall in late exercise after anaerobic threshold (AT); in panel 3 the gradient of the red dot $\dot{V}$O₂ data points falls away from paralleling the black crosses of the work slope after AT and the peak value is low.

Following the systematic approach, his nine-panel plot can be interpreted:

• (i)Although the patient only achieved 75% of his predicted maximum work, his peak HR was greater than 80% of the maximum predicted, as demonstrated by burgundy data points within the burgundy shaded box in panel 2. In addition, the RER in panel 8 was 1.4, and therefore greater than the required value of 1.15. The test was therefore maximal in terms of effort.
• (ii)The maximum recorded $\dot{V}$O₂ in panel 3 was 0.9 L O₂ min^–1, which for a 64 kg patient equates to 14.1 ml O₂ kg^–1 min^–1.
• (iii)The $\dot{V}$O₂–work relationship is abnormal in panel 3: the increase in $\dot{V}$O₂ (the gradient of the line) is low and does not parallel the increase in work.
• (iv)The AT can be determined using the V-slope method in panel 5 and is shown on the nine-panel plot. AT can be confirmed in panel 6, where V_E/$\dot{V}$O₂ clearly separates from ${\dot{V}}_{\text{E}}/\dot{V}{\text{CO}}_2$ at AT.
• (v)The $\dot{V}$O₂ at AT is 0.7 L O₂ min^–1. In a 64 kg patient, this equates to 10.9 ml O₂ kg^–1 min^–1.
• (vi)Panel 2 shows a normal linear increase of HR with work.
• (vii)The oxygen pulse in panel 2 is very abnormal, with no discernible increase with increasing work.
• (viii)There was no ventilatory limitation—the data points were well to the left of the dotted green line representing MVV and therefore will also be less than 80% of MVV.
• (ix)Ischaemic ECG changes were seen immediately before test termination, which rapidly resolved in the recovery phase.

This test demonstrates a patient who, despite a normal HR response to exercise, had an oxygen pulse (a surrogate for stroke volume) which did not increase and a decrease in blood pressure after exercise, both of which suggest left ventricular myocardial ischaemia. Both AT and $\dot{V}$O_2peak were around the level at which greater perioperative complications are seen. For these reasons, the patient was offered an endovascular repair of his aortic aneurysm rather than open surgery.

Respiratory limitation

Figure 4 shows the nine-panel plot of a 78-yr-old patient weighing 70 kg who underwent CPET as part of the preoperative assessment for repair of a 7.5 cm abdominal aortic aneurysm. His medical history included chronic obstructive pulmonary disease and hypertension. His FEV₁ was 1.5 L, resulting in a calculated MVV of 60 L min^–1. On the basis of the patient’s height, weight, age, and gender, the predicted maximum work was 124 W, and the predicted maximum HR was 142 beats min^–1. The patient terminated the CPET at 7 min because of breathlessness, at which point he had reached a peak work of 113 W. SpO₂ immediately after the test was 97%.

Fig 4.

Respiratory limitation. For abbreviations, see legend to Figure 1. The main abnormality seen is in panel 7, where there is no breathing reserve, that is there are purple diamond data points up to the maximum voluntary ventilation (MVV), the vertical dashed line at the right-hand side of the panel at 60 L min^–1, indicating that breathing reserve has been exhausted at the end of exercise.

Following the systematic approach, his nine-panel plot can be interpreted:

• (i)The test was maximal in terms of effort: the patient only achieved an HR of 109 beats min^–1 (77% of predicted) but did manage 91% of predicted work and an RER in excess of 1.15 (Fig. 4, panel 8).
• (ii)The $\dot{V}$O_2peak in panel 3 was 1.25 L O₂ min^–1, which for a 70 kg patient equates to 17.9 ml O₂ kg^–1 min^–1.
• (iii)The $\dot{V}$O₂–work relationship is normal, as $\dot{V}$O₂ can be seen increasing in parallel with work in panel 3.
• (iv)There is a clear increase in the gradient of the $\dot{V}$CO₂ data points in panel 5, so AT can be determined using the V-slope method. The point at which ${\dot{V}}_{\text{E}}/\dot{V}{\text{O}}_2$ and V_E/$\dot{V}$CO₂ separate in panel 6 is less clearly seen.
• (v)The $\dot{V}$O₂ at AT in panel 5 is 0.8 L O₂ min^–1. For a 70 kg patient, this equates to11.4 ml O₂ kg^–1 min^–1.
• (vi)Panel 2 shows a normal linear increase in HR with increasing work.
• (vii)The oxygen pulse in panel 2 shows a steady increase with increasing work, a normal response to exercise.
• (viii)There is respiratory limitation. There are data points beyond 80% of MVV, the vertical dashed line at the right-hand side of panel 7, indicating that breathing reserve has been exhausted at the end of exercise.
• (ix)There were no ECG abnormalities reported on this test.

This test was terminated because of dyspnoea. The CPET determined the cause of breathlessness as respiratory limitation rather than cardiac limitation. Despite the AT and $\dot{V}$O_2peak placing the patient in a lower-risk group for open abdominal aortic aneurysm repair, because the patient had respiratory limitation of exercise it was felt that the risk of postoperative pulmonary complications outweighed the benefit of open surgery, and the patient was offered an endovascular repair of his aneurysm.

Conclusions

CPET is a useful preoperative test to assess and risk stratify patients undergoing major surgery. The absolute values of AT, $\dot{V}$O_2peak, and ${\dot{V}}_{\text{E}}/\dot{V}{\text{CO}}_2$ are important to determine, but this article aims to equip the reader with a systematic approach to the nine-panel plot to give further insight into the patient’s respiratory and cardiac physiological reserve. By understanding the physiological principles of CPET, the anaesthetist may gain additional information into potential perioperative complications, which may in turn influence management decisions. Research is ongoing into the feasibility of improving cardiopulmonary fitness with preoperative exercise training (‘prehabilitation’) and to determine whether these fitter patients are at lower perioperative risk of morbidity and mortality.¹⁰

Declaration of interest

The authors declare that they have no conflicts of interest.

MCQs

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

Biographies

David Chambers DPhil MRCP FRCA is a consultant anaesthetist at Salford Royal NHS Foundation Trust, with interests in neuroanaesthesia, perioperative medicine and medical education. He has written several educational articles and also a textbook for the Primary FRCA.

Nick Wisely FRCA is a consultant anaesthetist at Manchester University NHS Foundation Trust and an honorary senior lecturer at the University of Manchester, with interests in perioperative medicine and vascular anaesthesia. He is the lead for cardiopulmonary exercise testing within the trust, has written several journal articles on CPET, and teaches on a regional CPET course.

Matrix codes: 1A01, 2A03, 3I00