Breathlessness, chest tightness or disproportionate fatigue on exertion is one of the most common reasons a patient is referred, and the diagnostic problem is almost always the same: is this the heart, the lungs, the muscle, or simply deconditioning. The differential is wider than those four, and also includes anaemia, dysfunctional breathing, obesity, medication effects, autonomic dysfunction, metabolic disease and exertional laryngeal obstruction. This page is about choosing and interpreting the tests that answer that question. It covers which investigation suits which clinical question, how an exercise tolerance test is used and when it must be stopped, how to read a cardiopulmonary exercise test report well enough to identify the limiting system, and the simple functional tests that often tell you as much at the bedside. The companion page on exercise testing and physiological assessment covers what these tests measure physiologically; this one is about acting on the result.
The history usually narrows the field before any test is ordered, and the commonest error is running the wrong investigation well rather than the right one adequately.
Exertional wheeze or cough in the absence of cardiac red flags points to the airway. Spirometry is the diagnostic test, distinguishing obstructive from restrictive patterns and assessing reversibility, while peak expiratory flow (PEF) is better suited to serial monitoring and to demonstrating variability than to making the initial diagnosis. Where exercise-induced bronchoconstriction is suspected specifically, a challenge test is more informative than resting spirometry, which is often normal.
Exertional chest discomfort needs care, because UK practice has moved. Current NICE guidance on recent-onset chest pain recommends computed tomography coronary angiography as the first-line investigation where clinical assessment indicates typical or atypical angina, and states explicitly that exercise electrocardiography should not be used to diagnose or exclude stable angina in people without known coronary artery disease. The exercise tolerance test (ETT), which monitors the electrocardiogram, blood pressure and symptoms under progressive load, retains a clear role elsewhere: characterising exertional arrhythmias, assessing the blood pressure response to exercise, measuring functional capacity, and as an alternative to functional imaging in people with established coronary disease. Cardiopulmonary exercise testing (CPET) addresses the harder problem: unexplained exertional breathlessness where integrated exercise physiology is needed to separate overlapping potential limitations or to quantify functional capacity. It is often used once basic cardiac and respiratory tests are unrevealing, but it need not be withheld until every other test is normal. It adds expired gas analysis to the exercise test and so measures the whole oxygen pathway rather than one organ. Its other common use in the UK is preoperative risk assessment. Finally, in frail or older patients, simple functional tests answer the practical question of capacity more cheaply and more relevantly than any laboratory test.
An exercise tolerance test uses a standardised incremental protocol while the electrocardiogram, blood pressure, heart rate and symptoms are monitored. It is used to look for inducible ischaemia, to provoke and characterise exertional arrhythmias, and to assess functional capacity in known coronary disease. Its diagnostic accuracy for coronary disease is modest, which is why it is no longer the UK first-line test for suspected stable angina.
The absolute contraindications overlap with those for exercise generally and should be checked before the patient gets on the treadmill: myocardial infarction within the preceding two days, unstable angina not yet stabilised, uncontrolled symptomatic arrhythmia, symptomatic severe aortic stenosis, acute myocarditis, pericarditis or endocarditis, acute pulmonary embolism, and acute aortic dissection.
The termination criteria are heavily examined, and the distinction between absolute and relative indications matters. Two blood pressure criteria recur. A fall in systolic pressure of more than 10 mmHg from baseline despite an increasing workload should be verified immediately; where it is accompanied by other evidence of myocardial ischaemia it is an absolute indication to terminate, and without such evidence it is generally a relative indication requiring clinical judgement. A hypertensive response beyond roughly 250/115 mmHg is likewise a relative rather than a universal absolute criterion. Symptom-based reasons to terminate include angina or chest pain, signs of poor perfusion such as pallor, cyanosis or cold clammy skin, and central nervous system symptoms including ataxia, dizziness or near-syncope. Add to those significant ST change, sustained ventricular tachycardia, a request from the patient to stop, and any technical failure of monitoring. Of these, the falling systolic pressure is the one candidates most often miss, because instinct expects a rising pressure to be the dangerous finding. Learning the criteria as an undifferentiated list is a mistake; know which are absolute and which are relative.
A cardiopulmonary exercise test report looks intimidating but only a handful of variables carry the diagnosis. Peak oxygen uptake is the headline measure of exercise capacity, and it should be compared against reference values appropriate to age, sex, body size, exercise modality and the reporting laboratory rather than against a single universal cut-off. The respiratory exchange ratio, the ratio of carbon dioxide production to oxygen uptake, contributes to judging effort: European respiratory guidance treats a peak value above roughly 1.05 as supportive of high effort rather than as proof, and it is interpreted alongside symptoms, heart rate, any oxygen uptake plateau, achieved workload and the ventilatory response.
Four variables then localise the problem. The ventilatory or gas-exchange threshold is the point at which carbon dioxide production rises disproportionately to oxygen uptake, identified most reliably by the V-slope method of plotting one against the other; a low threshold indicates that oxygen delivery fails early. The ventilatory efficiency slope, the relationship between minute ventilation and carbon dioxide production, rises as efficiency worsens, and although a value under about 30 is often quoted as normal the threshold varies with age, protocol, disease and laboratory reference. Oxygen pulse is oxygen uptake divided by heart rate; it reflects oxygen uptake per heartbeat and is influenced by stroke volume, peripheral oxygen extraction, haemoglobin and the chronotropic response, so it is a useful but impure surrogate for stroke volume. It should climb steadily through exercise, and early flattening indicates an impaired cardiovascular response which may reflect myocardial ischaemia, though it is not specific for it. Breathing reserve is the ventilation left unused at peak exercise; exhaustion supports a ventilatory constraint, but interpretation also requires flow-volume behaviour, inspiratory capacity, oxygen saturation, symptoms and gas exchange.
Three broad patterns are described, and they are aids to reasoning rather than proof. Cardiac limitation tends to give a low peak oxygen uptake, a low threshold, a raised ventilatory efficiency slope, a flattening oxygen pulse and a preserved breathing reserve. Pulmonary limitation tends to give a low peak oxygen uptake with the breathing reserve exhausted. Deconditioning is a possible conclusion where exercise capacity is reduced, effort was adequate, and no cardiovascular, ventilatory, gas exchange, metabolic or musculoskeletal limitation is demonstrated. In practice limitation is frequently mixed, and no single arrow pattern establishes a diagnosis on its own. Trained athletes are the exception that proves the rule, since they can exhaust their breathing reserve while achieving a supranormal peak oxygen uptake.
Laboratory testing is not always the right answer, and in older or frailer patients a functional test often gives more useful information in less time. The five times sit-to-stand test asks the patient to rise from a chair and sit back down five times as quickly as possible without using their arms. Chair height and the exact timing endpoint vary between protocols, commonly a seat of about 43 to 45 cm, so follow the protocol the service has validated rather than assuming one universal standard. It assesses lower limb function and is directly relevant to transfers, but it is a composite measure influenced by strength, power, balance, pain, cognition and technique rather than a pure strength test, and it is one component of falls risk assessment rather than a standalone assessment.
Handgrip strength measured with a dynamometer is the other workhorse. Low grip strength is associated with mortality, longer hospital stay and disability, and it is used to identify probable sarcopenia before any measurement of muscle mass. It is effort dependent and confounded by hand pathology such as arthritis or carpal tunnel syndrome, and by hand size relative to the dynamometer. Gait speed over a short measured distance is the third, and it is used as a marker of severity rather than of strength. The general principle is the same as with any test: know what the result is confounded by before you act on it.
Sign up to get full access to 10 topics of your choice, including all sections, clinical pearls, and exam tips.
Sign up free10 free topics included with your account. Full access from £24.17/month.
Sections included with full access