Focused Bedside Echocardiography: A Working Reference
A hypotensive patient at two in the morning, a probe on the trolley, and nobody else in the hospital who scans. The question is never “please perform an echocardiogram”. It is whether this patient is empty, obstructed, failing or tamponading, and whether the answer changes what you do in the next ten minutes. A focused bedside study answers that class of question extremely well and answers other questions badly, and the difference between a useful scan and a dangerous one is almost entirely about knowing which is which. What follows is a working reference for physicians who scan their own patients without a sonographer or a reporting cardiologist.
On this page
The scope of a focused study
A focused study is a binary instrument applied to a small number of large questions: is the left ventricle severely impaired, is the right ventricle acutely strained, is there a pericardial collection causing physiological compromise, is the inferior vena cava full or empty, and is there gross valvular catastrophe. It is not a substitute for a comprehensive study, and it should never generate a diastolic grade, a valve area or a chamber quantification report. Clinicians get into difficulty in one of two ways: by extending a focused study beyond its remit, or by declining to scan at all and managing a shocked patient on inference.
The decision points
1. Decide the question before you pick up the probe
Scanning without a stated question produces incidental findings and no decisions. Write the question down: undifferentiated hypotension, unexplained breathlessness, suspected tamponade, chest pain with haemodynamic compromise, or assessment of volume responsiveness. The question dictates which windows you need, how long the study takes, and what constitutes a negative result.
2. Decide the minimum dataset for that question
For shock, a subcostal four-chamber view, a parasternal long and short axis, an apical four-chamber view and an inferior caval view will answer most of it. Get the subcostal view first in a supine ventilated patient, because it is frequently the only window available and it answers the tamponade question immediately. If two of the four windows are unobtainable, that is itself a result: state the limitation and escalate rather than reporting an impression built on one image.
3. Decide whether visual assessment is sufficient
For acute management, a graded visual estimate of left ventricular function by an experienced eye is accurate enough and far faster than measurement. Severe impairment, hyperdynamic function and normal function are the three categories that change treatment tonight. Reserve formal biplane measurement for the comprehensive study that follows. What visual estimation cannot do is detect the subtle impairment that matters in cardiotoxicity or infiltrative disease, which is a different clinical setting entirely and covered in our echocardiography masterclass.
4. Decide the volume question from more than the caval diameter
Inferior caval size and collapsibility are useful and frequently over-interpreted, particularly in ventilated patients and in isolated right heart disease. Read the cava alongside the size and behaviour of the left ventricle: a small, vigorously emptying ventricle with a collapsing cava supports fluid administration, while a dilated ventricle with a plethoric cava argues against it. Left atrial size adds the chronic dimension, distinguishing an acutely dry patient from one with long-standing elevated filling pressures.
5. Decide whether the right ventricle is acutely strained
In an acutely breathless or shocked patient this is often the highest-yield question on the study. Look for right ventricular dilatation relative to the left, septal flattening, reduced longitudinal excursion at the tricuspid annulus, and regional preservation of apical contraction with a poorly contracting free wall. That last pattern in a hypotensive patient with clear lungs should redirect the entire working diagnosis. Remember that an acutely strained right ventricle cannot generate a high pulmonary pressure, so a modest tricuspid gradient does not exclude a major obstruction.
6. Decide tamponade on physiology, not on volume
Effusion size correlates poorly with haemodynamic significance. A rapidly accumulating small effusion after intervention or trauma can be lethal while a large chronic collection is tolerated. Look for right atrial systolic collapse, right ventricular diastolic collapse, marked respiratory variation in transvalvular inflow velocities, and a dilated non-collapsing cava. In the post-procedural patient, low threshold and rapid escalation matter more than measurement precision, and the emergency sequences are set out in our cardiac emergency reference for emergency physicians.
7. Decide which valve findings change tonight
Chronic valve grading belongs in a comprehensive study. Three findings do belong in a focused one because they alter immediate management: acute severe regurgitation with a normal-sized ventricle in a shocked patient, a large mobile mass on a valve in a febrile patient, and obstructed or dehisced prosthetic material. Colour Doppler across every valve takes thirty seconds and is worth the time even in a focused protocol.
8. Decide when the focused study has reached its limit
The moment your management depends on a number rather than a category, the focused study is over and a comprehensive one is required. The same applies when transoesophageal imaging is likely to be needed, as in suspected endocarditis with negative transthoracic images, prosthetic valve dysfunction, aortic dissection and unexplained embolic stroke. Knowing this boundary is what separates competent bedside imaging from overreach, and our practical echocardiography reference sets out where each modality takes over.
When to escalate, transfer or call for help
- Immediately, during the scan: tamponade physiology, acute severe regurgitation, a mechanical complication after infarction, or a mobile intracardiac mass. These findings need a cardiologist and often a surgeon, not a follow-up appointment.
- Same day: newly severe ventricular impairment, acute right ventricular strain, or a suspected prosthetic valve problem. Treat while arranging the definitive study; the two are not sequential.
- When the windows fail: escalate rather than reassure. An inadequate study in a deteriorating patient is a reason to transfer, not a reason to observe overnight.
- For chronic findings: route to a full study and outpatient assessment, using the diagnosis and imaging hub and the heart failure guide for general physicians to decide what needs to happen before the appointment.
What to document
- The clinical question the study was performed to answer, stated at the top.
- That the study was focused rather than comprehensive, in explicit words.
- Which windows were obtained and which were not.
- Categorical findings rather than measurements: severely impaired, normal, dilated, effusion with or without compromise.
- The action taken as a result, and whether a comprehensive study has been requested and when.
- Images stored and retrievable, since the next clinician will want to compare rather than take your word for it.
Bench card
| Clinical question | Look at | Finding that changes management |
|---|---|---|
| Undifferentiated shock | Subcostal, parasternal, apical, caval | Severe impairment, tamponade, right strain, empty hyperdynamic ventricle |
| Fluid or no fluid | Cava plus left ventricular size and behaviour | Small vigorous ventricle with collapsing cava favours volume |
| Acute breathlessness | Ventricular function, septal shape, valves | Right dilatation with septal flattening redirects the diagnosis |
| Post-procedural deterioration | Pericardium first, then function | Any new effusion with chamber collapse |
| Fever with a murmur | Colour Doppler across all valves | Mobile mass or new regurgitation |
| Chest pain with compromise | Regional wall motion, aortic root, pericardium | New regional abnormality or a dilated root |
Three habits that keep bedside scanning safe: state the question in writing before scanning, report in categories rather than numbers, and name every window you could not obtain. Short-form revision notes for physicians who scan occasionally rather than daily are collected in cardiology quick notes for the non-cardiologist and across the echocardiography reference collection. A note-based companion to this material is published as these practical echocardiography quick notes.
Questions from the floor
How much training is needed before scanning unsupervised in an emergency?
Enough supervised studies to recognise a normal heart reliably from several windows, which is a larger number than most short courses provide. Competence in a focused protocol is defined by knowing what you cannot answer, and that judgement develops later than image acquisition does.
Can a focused study exclude pulmonary embolism?
No. It can support the diagnosis in a shocked patient by demonstrating acute right ventricular strain, and it can redirect management immediately when it does. A normal right ventricle does not exclude embolism in a haemodynamically stable patient, and the definitive investigation remains cross-sectional imaging.
Should I report an incidental finding picked up on a focused scan?
Yes, but as an observation requiring confirmation rather than as a diagnosis. Write that it was seen on a focused study, describe it plainly, and request a comprehensive assessment. Silently ignoring it and formally diagnosing it are both errors.
Is handheld equipment adequate for these decisions?
For the categorical questions above, modern handheld devices are largely adequate and vastly better than no imaging. They are not adequate for quantification, for spectral Doppler assessment in most cases, or for anything requiring measurement precision, and reports should name the device used.
Adult Echocardiography Normal Values: A Reporting Bench Card
You have scanned the patient yourself, there is no reporting cardiologist on site, and the referring surgeon wants an answer before the list starts. The question is rarely “what is normal” in the abstract. It is whether this particular number, measured on this particular study, should change what happens to the patient today. This reference groups adult transthoracic values around that question, with the measurement traps that most often corrupt an otherwise competent study, and it assumes you are the one holding the probe as well as writing the report.
On this page
Before you measure anything
Two habits determine whether the rest of the study means anything. Index chamber dimensions and volumes to body surface area, because an unindexed left atrial diameter is close to uninterpretable at either extreme of body size. And apply sex-specific ranges, since using male thresholds in women systematically under-diagnoses chamber enlargement. Both are trivial to do and both are commonly skipped when the department is busy.
The decision points
1. Decide whether the study is good enough to answer the question asked
Apical foreshortening is the commonest and most expensive error in adult echocardiography. It shrinks volumes and inflates ejection fraction, occasionally by enough to move a patient across a device eligibility threshold. Move the patient rather than the probe to find the true apex. If the endocardium is not seen across two contiguous segments, say so in the report and consider contrast rather than reporting a number you do not believe.
2. Establish left ventricular size and systolic function on volumes, not linear dimensions
Linear measurements remain useful for serial comparison in the same patient, but biplane volumetric assessment is what should drive decisions about valve intervention and device eligibility. Reference ranges for the adult left ventricle sit in the bench card below. Inter-observer variability for ejection fraction is of the order of five to ten percentage points, so a report stating an ejection fraction to one decimal place claims a precision the technique does not possess. Report a category or a range when image quality is imperfect.
3. Ask whether “normal ejection fraction” is actually normal
Global longitudinal strain detects impairment while ejection fraction remains untouched, and it has become the deciding measurement in cardiotoxicity surveillance, infiltrative disease and early hypertensive heart disease. Values more negative than minus twenty per cent are conventionally normal, with a healthy band roughly between minus sixteen and minus twenty-two per cent depending on vendor and software version. That vendor dependence is clinically important: serial comparisons must be made on the same platform, and the report should name the system used. Pattern recognition in strain mapping is developed further in the echocardiography atlas for clinicians.
4. Use the left atrium as your barometer of chronic filling pressure
Left atrial volume index is the single most informative atrial measurement and an independent predictor of atrial fibrillation, stroke and mortality. Linear atrial diameter survives in reports out of habit and should not drive interpretation. Left atrial reservoir strain has now entered mainstream diastolic assessment, with values at or below eighteen per cent supporting elevated left atrial pressure; it is less load-dependent than the ratio of early filling velocity to annular velocity and performs well when that ratio is equivocal.
5. Report at least two right ventricular indices
The right ventricle is geometrically uncooperative and no single number suffices. Longitudinal indices such as annular plane systolic excursion and tissue Doppler systolic velocity interrogate the base only, and they fall after cardiac surgery even when global right ventricular performance is preserved. Pair a longitudinal index with fractional area change or free wall strain. Estimate right atrial pressure from inferior caval diameter and its inspiratory collapse, then add it to the peak tricuspid regurgitant gradient for a systolic pulmonary artery pressure. Recent right heart guidance grades these parameters across four tiers rather than as simply normal or abnormal, which matters when you are tracking a patient with pulmonary hypertension over time.
6. Grade diastolic function on the primary variables, then adjudicate
Current practice evaluates annular early diastolic velocity, the ratio of early filling to annular velocity, and tricuspid regurgitant velocity together rather than through a stepwise cascade with separate entry points for preserved and reduced ejection fraction. When the primaries disagree or are unobtainable, supporting measures carry the decision: left atrial reservoir strain, pulmonary venous systolic to diastolic ratio, left atrial volume index and isovolumic relaxation time. Grade one shows impaired relaxation with normal filling pressure, grade two shows two or more abnormal primaries with intermediate filling pressures, and grade three shows all three abnormal with markedly elevated pressures, subdivided by whether the pattern reverses with Valsalva. Irreversibility carries the worse prognosis.
7. Apply valve thresholds, then look for the situations that defeat them
Low-flow, low-gradient aortic stenosis is the recurring trap. When the calculated valve area is in the severe range but the mean gradient is not, measure stroke volume index; a value below 35 mL/m squared should prompt dobutamine stress echocardiography or calcium scoring rather than a reflex report of moderate disease. Doppler angle above roughly twenty degrees materially underestimates velocity, so interrogate aortic stenosis from multiple windows, including right parasternal, before concluding that gradients are modest. Structured practice in these decisions is the focus of our echocardiography masterclass.
When to escalate, transfer or call for help
- Same day: new severe left ventricular impairment with clinical decompensation, a large pericardial effusion with respiratory variation and chamber collapse, suspected endocarditis with a mobile mass, or a mechanical complication after infarction.
- Urgent referral: severe aortic stenosis with any symptom, severe primary mitral regurgitation, or newly severe tricuspid regurgitation with right ventricular dilatation. Landmark data from the PARTNER programme in aortic stenosis and COAPT in secondary mitral regurgitation have moved these thresholds well below what many general clinics still assume.
- Refer for a second study rather than a decision: discordant valve grading, unexplained pulmonary hypertension, or a strain result that conflicts with a visually normal ventricle. Transoesophageal or cross-sectional imaging usually resolves it, and the diagnosis and imaging reference hub sets out when each modality adds value.
- Flag for heart failure follow-up: elevated filling pressures with preserved ejection fraction. These patients are frequently discharged as normal studies, and the heart failure, rhythm and risk hub is the natural next stop.
What to document
- The method used for each headline measurement, particularly for ejection fraction.
- Indexation to body surface area, with height and weight recorded.
- The vendor and software platform when strain is reported, so that serial studies remain comparable.
- An explicit statement of image quality and which segments were not adequately visualised.
- Whether measurements in atrial fibrillation were averaged across at least five cycles or obtained by the index-beat method.
- A one-line interpretive conclusion that answers the referral question rather than restating the numbers.
Bench card
| Parameter | Men | Women |
|---|---|---|
| End-diastolic diameter | 42 to 58 mm | 38 to 52 mm |
| End-diastolic volume index | 34 to 74 mL/m2 | 29 to 61 mL/m2 |
| End-systolic volume index | 11 to 31 mL/m2 | 8 to 24 mL/m2 |
| Ejection fraction, biplane | 52 to 72 per cent | 54 to 74 per cent |
| Wall thickness | 10 mm or less | 9 mm or less |
| Mass index | 115 g/m2 or less | 95 g/m2 or less |
| Parameter | Normal | Note |
|---|---|---|
| Left atrial volume index | 16 to 34 mL/m2 | Above 34 is enlarged; severity bands follow at 41 and 48 |
| Left atrial reservoir strain | Above 18 per cent | At or below 18 supports raised left atrial pressure |
| Right atrial volume index | Below 30 mL/m2 | Now preferred over area |
| Right ventricular basal diameter | Below 41 mm | Four-chamber view, focused on the right ventricle |
| Annular plane systolic excursion | 17 mm or more | Unreliable after cardiac surgery |
| Fractional area change | Above 35 per cent | Pair with a longitudinal index |
| Global longitudinal strain | More negative than minus 20 per cent | Vendor dependent; keep the platform constant |
| Tricuspid regurgitant velocity | Below 2.8 m/s | Threshold embedded in diastolic grading |
Severity thresholds worth keeping to hand: aortic stenosis is severe at a peak velocity of 4.0 m/s, a mean gradient of 40 mmHg or a valve area of 1.0 cm2; mitral stenosis at a valve area of 1.5 cm2; mitral regurgitation at an effective orifice of 0.40 cm2 or a regurgitant volume of 60 mL. The full set, with acquisition protocols, sits in the practical echocardiography reference and across the echocardiography collection. A companion chart version of these values is published as an adult echocardiography normal values chart.
Questions from the floor
How should chamber size be indexed in severe obesity?
Body surface area indexing under-detects enlargement at high body weight because the denominator scales with the excess mass. Height-based allometric indexing performs better and is worth adopting where your software supports it. At minimum, state which method was used.
Can I report diastolic function in atrial fibrillation?
Partially. Annular velocities, the ratio to early filling velocity and tricuspid regurgitant velocity remain usable if averaged across enough cycles, but the E to A ratio and atrial volume lose their meaning. Report what is interpretable and name what is not.
Does a normal ejection fraction exclude significant disease?
No, and this is the most consequential limitation in routine practice. Amyloid, early cardiotoxicity, aortic stenosis with a small hypertrophied ventricle and heart failure with preserved ejection fraction all present with an ejection fraction inside the reference range. Strain and filling pressure assessment are what separate them.
My laboratory has an older machine. Are these ranges still applicable?
Chamber and Doppler ranges are broadly robust across platforms. Strain is not, and older systems may not support it at all. Where strain is unavailable, lean harder on atrial size, filling pressure indices and careful serial comparison rather than treating ejection fraction as sufficient.
