Monday 30 May 2016

Surrogate markers of LV systolic function: dP/dt

In a recent blog post we looked at the use of E-point septal separation (EPSS) as a surrogate marker of left ventricular systolic function. In this blog, we're going to look at another surrogate marker, namely dP/dt.

When left ventricular systolic function is normal, there is a rapid rise in left ventricular systolic pressure during systole. The rate at which the ventricular pressure rises is expressed by the term dP/dt, which is the rate of change in pressure (dP) with time (dt). If left ventricular systolic function becomes impaired, the rate of rise in pressure is slower and therefore dP/dt starts to fall.

Measurement of dP/dt using CW Doppler of mitral regurgitation

It's possible to measure dP/dt using echocardiography, but only if mitral regurgitation is present. To make the measurement, we need to obtain a continuous wave (CW) Doppler trace of the mitral regurgitation (MR) from the apical 4-chamber view. It's helpful to set the sweep speed as high as possible, to 'spread out' the trace (this makes it easier to take the measurements).

Freeze the CW Doppler trace, and using the calipers mark the points where the mitral regurgitation jet velocity is 1 m/s and 3 m/s. You need to measure the time interval between these two points, i.e. the time taken for the MR jet velocity to rise from 1 m/s to 3 m/s. The illustration above shows how this measurement is made.

We know from the Bernoulli equation that when the MR jet velocity is 1 m/s, the pressure gradient 'driving' the jet is 4 mmHg (pressure gradient = 4 x velocity squared). Similarly, when the MR jet velocity is 3 m/s, the pressure gradient driving it is 36 mmHg. From this, we can deduce that the change in pressure between the two time points is 32 mmHg (36 mmg minus 4 mmHg).

It's now possible for us to calculate dP/dt, by dividing 32 (the 'dP') by the measured time interval (the 'dt'). The longer it takes for the pressure in the left ventricle to rise during systole, the longer the duration of dt, and therefore the smaller the value of dP/dt.

When left ventricular systolic function is normal, dP/dt is usually >1200 mmHg/s. As left ventricular function deteriorates, dP/dt falls. A severely impaired left ventricle usually has a dP/dt <800 mmHg/s.

So, if your patient has measurable mitral regurgitation, you can use the dP/dt method to assess their left ventricular systolic function. However, there are a couple of pitfalls - the dP/dt method isn't reliable if the mitral regurgitation is acute, or if there is significantly increased afterload (aortic stenosis or systemic hypertension).

You can read more about dP/dt in my textbook, Making Sense of Echocardiography (2nd edition), or in the following paper:

Friday 27 May 2016

Is a left ventricular ejection fraction of 50% 'normal'?

What is a normal left ventricular ejection fraction? A seemingly straightforward question and, if you work in cardiology, you'll probably already have a specific number in mind. But this number probably depends upon which guidelines you use.

There's a surprising variation between guidelines in what is deemed a 'normal' LVEF, and this confusion of reference values has been added to this month with the publication of the ESC's 2016 Heart Failure Guidelines.

Calculating left ventricular end-diastolic volume

If you work in the US, then you probably use the ASE recommendations for cardiac chamber quantification, published in 2015. For 2D echo, there are separate normal ranges for males and females. For males, a normal LVEF is 52-72%, and for females it's 54-74%.

If you work in the UK, then you'll most likely use the BSE guidelines for chamber quantification. According to these, for males and females a normal LVEF is 55% or above. Anything less than 55% is 'impaired'.

So in the US, an LVEF of 54% would be labelled 'normal', whereas in the UK it would be labelled as 'impaired'. A difference of just a few percent might seem like splitting hairs, but the diagnostic labels we give our patients matter - just ask someone who's trying to buy life insurance how much difference a label of 'impaired LV function' can make.

Things have become even more confusing this month with the publication of the European Society of Cardiology's new guidelines on heart failure. These take a new approach to diagnostic labels in heart failure, differentiating between 'reduced' LVEF (<40%) and 'mid-range' LVEF (40-49%). However they clearly identify a 'preserved' LVEF as being 50% or above.

So now we have three different cut-off values for what constitutes a 'normal' LVEF: 50% (ESC), 52%/54% (ASE), and 55% (BSE). Oh, and for one added layer of confusion, the ASE guidelines are endorsed by the European Association of Cardiovascular Imaging, a community of the ESC - so one organisation effectively promotes two different 'normal' values.

One thing that is clear is that there's a grey area around those with 'borderline' LVEF measurements, and a lot of work remains to be done to define which treatments are appropriate within this group. By highlighting those with 'mid-range' LVEFs with their new diagnostic category of HFmrEF, the new ESC guidelines have made a start in encouraging work on this group.

But for those whose LVEFs are somewhere in the low 50's, I suspect it's going to be some time before we can agree upon a clear definition for those all-important diagnostic labels.

Wednesday 25 May 2016

The ESC's new heart failure terminology: HFmrEF anyone?

In the last few days, the European Society of Cardiology has released its latest guidelines on the management of acute and chronic heart failure. An update of the ESC's last guidelines (which were published in 2012), the new 2016 document contains several changes which will be covered in forthcoming blogs.

One of the most fundamental changes is in the terminology of heart failure. The ESC now recognises three categories of heart failure based upon left ventricular ejection fraction:

New ESC categories in heart failure

The categories of HFpEF and HFrEF have been recognised for several years, but the category of HFmrEF is new. Why has this group of patients been highlighted as a separate group?

Well, the ESC argues that those with a mid-range LVEF (40-49%) are a 'grey area', where the benefits of heart failure therapies on morbidity and mortality have not been so conclusively proven as in the HFrEF (LVEF <40%) group.

The ESC argues that identifying patients with HFmrEF as a new group will help to stimulate research into the characteristics and management of this 'intermediate' category of left ventricular systolic dysfunction.

In its discussion on terminology, the ESC also advises against using the terms 'preserved systolic function' or 'reduced systolic function' This is because patients with HFpEF often have subtle abnormalities of systolic function, and most of those with HFrEF have evidence of diastolic dysfunction. Thus the terms HFpEF, HFrEF (and, of course, the new term HFmrEF) are now preferred by the ESC.

To obtain a copy of the full ESC 2016 guidelines, click the following link:

2016 ESC Guidelines for the diagnosis and treatment of acute and chronic heart failure.

Monday 23 May 2016

Surrogate markers of LV systolic function: EPSS

When we assess left ventricular systolic function with echocardiography, we usually think in terms of left ventricular ejection fraction. But there are many other indicators of left ventricular function, some of which are rarely used. One of these is E-point septal separation, or EPSS.

Increased EPSS in severe dilated cardiomyopathy

EPSS refers to the distance between the anterior leaflet of the mitral valve and the interventricular septum during early diastole. This is easiest to measure using M-mode echo, and the measurement is taken when the anterior mitral leaflet is at its closest to the septum (see image).

Normally, the EPSS is no more than 6mm. The distance gets bigger with worsening left ventricular systolic function, and an EPSS >7mm is a sensitive measure for severely impaired systolic function.

What is the mechanism underlying the EPSS method? The increase in EPSS in left ventricular systolic dysfunction is partly because of ventricular dilatation causing the separation between the anterior mitral leaflet and the septum to increase, but also because of a reduced transmitral flow volume causing the mitral leaflets to open less widely during diastole.

So an increased EPSS is a surrogate marker of impaired left ventricular systolic function. However, you can't use EPSS in mitral stenosis or aortic regurgitation (both of which will falsely increase the EPSS because the anterior leaflet opens less fully), nor should it be used in the presence of septal hypertrophy (which tends to falsely reduce the EPSS).

To read about the value of EPSS in bedside echo, check out the following paper:

Thursday 19 May 2016

What structure are we looking at here?

When we perform cardiac MRI studies, we sometimes make incidental findings outside the heart. It's important to comment on these in our reports, but in order to do so we need to recognise what it is that we're seeing.

Here's an image from a cardiac MRI study. This was one of the first images we obtained, whilst we were performing some initial views (called HASTE images) of the patient's thorax. What is the structure indicated by the red arrow?

What's the structure indicated by the red arrow?
This structure is real (in other words, it's not an artifact). But it's not something we usually see during our cardiac MRI scans. In fact, this is something we see only in around 1% of our scans.

It's easy to mistake this appearance for a part of the aortic arch, but in fact it's part of the lung. This is known as an azygous lobe, and it's a congenital abnormality of the right upper lobe. It forms in utero when an anomalous (laterally displaced) azygous vein penetrates the right upper lobe during development, dividing it into two portions.

The presence of an azygous lobe doesn't cause any clinical symptoms, and the only real significance is that it can make thoracoscopy more challenging than usual.

If you'd like to learn more about cardiac MRI, including how to interpret incidental findings, check out our online Cardiac MRI Essentials course at Medmastery.

You can also read more about the azygous lobe here:

Monday 16 May 2016

Ten key facts about bicuspid aortic valve disease

Bicuspid aortic valve disease (BAVD) is a common problem in which the aortic valve has two, rather than three, cusps. BAVD is frequently picked up as an incidental finding during cardiac imaging. Here are Ten Key Facts about BAVD that you should be aware of in your cardiology clinic:

Transoesophageal echo showing bicuspid aortic valve en face

  1. BAVD affects about 2% of the population
  2. This means that BAVD is the most common congenital cardiac abnormality
  3. BAVD is around 2x commoner in males
  4. BAVD has been associated with an abnormality of the NOTCH1 gene
  5. The most frequent complication of BAVD is aortic stenosis
  6. Around one-third of patients with BAVD will experience a serious complication of the condition (aortic stenosis or regurgitation, infective endocarditis, aortic aneurysm or dissection)
  7. BAVD is not just an abnormality of the aortic valve, but is a disease of the aortic root and ascending aorta as well
  8. Patients with BAVD have a 9x greater risk of aortic dissection than those with a normal aortic valve
  9. BAVD is present in around 50% of patients with coarctation of the aorta
  10. Once diagnosed, patients with BAVD should remain under long-term surveillance which regular imaging assessment of their aortic valve function and aortic root/ascending aortic dimensions
A detailed overview of BAVD, which includes key references, can be accessed from the European Society of Cardiology by clicking here.

Sunday 15 May 2016

Am I okay to go on holiday, doctor?

Around this time of year many of my cardiac patients ask me whether they can arrange a holiday. Some patients are worried about their fitness to fly, others are concerned about arranging travel insurance. In this blog, I'll highlight some useful resources to help answer these questions:

Fitness to fly
The British Cardiovascular Society has published a useful report to help clinicians advise their patients about fitness to fly. Entitled Fitness to fly for passengers with cardiovascular disease, the report can be accessed from the BCS website (click here) and contains a quick-reference guide that covers common cardiac conditions, together with a more detailed review of the evidence.

BCS report on Fitness to Fly (extract)

More generally, the UK Civil Aviation Authority has an online guide on fitness to fly for passengers and healthcare professionals, entitled Am I Fit to Fly?

Travel insurance
Obtaining travel insurance when patients have a cardiovascular condition can sometimes prove challenging. The British Heart Foundation covers this issue on their website (click here), and they also provide a list of sympathetic insurance companies.

Specific conditions
A number of websites provide information on fitness to travel for various cardiovascular conditions. These include:
Finally, NHS Scotland provides a public access website called Fit for Travel which provides travel health information for people travelling abroad from the UK.

Wednesday 11 May 2016

Cabrera's sign and the diagnosis of old MI in LBBB

In a recent blog we looked at the use of the Sgarbossa criteria in the diagnosis of acute myocardial infarction in the setting of pre-existing left bundle branch block.

In this blog we're going to take a look at another useful diagnostic clue: Cabrera's sign, which can be used in the diagnosis of an old myocardial infarction in LBBB.

Cabrera's sign refers to a notch that occurs within the first 40 msec in the ascending portion of the S wave in lead V3 or V4:

Cabrera's sign (arrowed)

This ECG sign was first described by Cabrera & Friedland in 1953.

Although Cabrera's sign has a high specificity for the diagnosis of an old myocardial infarction in LBBB, it has a poor sensitivity.

More details about the original paper by Cabrera and Friedland can be found here:

Tuesday 10 May 2016

Making sense of murmurs: The Levine scale

When we auscultate a systolic heart murmur, we need to describe its loudness or intensity. Some people use descriptive terms like 'soft' or 'loud', but traditionally the loudness of systolic murmurs is graded on a scale of 1 to 6. But what do the numbers mean?

The numeric system is sometimes called the Levine scale, named for Samuel A. Levine who first described it in 1933. The numbers correspond to the following intensities of murmur:

The Levine scale of systolic murmur intensity

In his original paper, Levine said that once physicians had worked with him for a short time, their gradings differed by no more than one gradation, and indeed were in "absolute agreement" in the majority of cases.

You can read more about Levine's grading of murmurs in the following paper:


You can also find Levine's original paper from 1933 here:


Samuel Levine was also, incidentally, the same physician who co-described Lown-Ganong-Levine syndrome, and also described Levine's sign.

Sunday 8 May 2016

What do we mean by 'rate control' in AF?

When we manage patients with atrial fibrillation, we often recommend an approach of 'rate control'. But what exactly is good rate control? When is a patient's ventricular rate well-controlled, and how should we achieve it?

AF with a fast ventricular rate

Acute rate control
In the acute situation, we usually aim for a ventricular rate of 80-100 beats per minute. This is commonly achieved with a beta-blocker or a rate-limiting calcium channel blocker (such as verapamil or diltiazem). Sometimes it is necessary to use amiodarone or digoxin for acute rate control, particular when left ventricular function is significantly impaired.

If a rapid ventricular rate fails to slow with drug treatment, and if the patient has symptomatic hypotension, ischaemia/angina or heart failure, then immediate DC cardioversion may be appropriate.

Chronic rate control
In chronic AF, the target for rate control in the long term depends upon symptoms. For patients who are asymptomatic, or who have tolerable symptoms, a lenient rate control strategy (resting ventricular rate <110 bpm) is recommended.

For patients with troublesome symptoms despite lenient rate control, or if there is evidence of a tachycardiomyopathy, then a strict rate control strategy should be adopted. This means aiming for a resting heart rate <80 bpm, and <110 bpm on moderate exercise.

When strict rate control approach is taken, a 24h ambulatory ECG should be performed for safety reasons (i.e. to check that there is no evidence of excessive bradycardia/pauses). It may also be appropriate to undertake an exercise test to assess the rate response to exercise.

Chronic rate control is usually achieved with a beta-blocker, rate-limiting calcium channel blocker or digitalis.

For more information on this and other aspects of AF management, take a look at the ESC Atrial Fibrillation Guidelines 2010 (and also the subsequent Focused Update in 2012). It's also worth taking a look at this study from the NEJM:

Friday 6 May 2016

What is the Ashman phenomenon?

Take a look at the ECG rhythm strip below (Fig. 1). How would you describe the broad QRS complex near the middle of the strip? Is this a ventricular ectopic beat? Or something else?

Fig. 1 Rhythm strip

In fact, this broad QRS complex is an aberrantly-conducted beat, and this is an example of the Ashman phenomenon. This phenomenon occurs when you get two beats with a long RR interval followed by a beat with a short RR interval (Fig. 2). In this situation, the beat with the short RR interval is aberrantly conducted - typically with a right bundle branch block morphology.

Fig. 2 Details of the Ashman phenomenon

The Ashman phenomenon occurs because the refractory period of conduction system is proportional to the RR interval of the preceding beat. When you have two beats separated by a long RR interval, the following refractory period will be relatively long. However if the next beat arrives after a short RR interval, the conduction system won't have fully recovered from its refractory period, and so its conduction will be abnormal, giving rise to a broad QRS.

The right bundle has a slightly longer refractory period than the left bundle, so the aberrantly-conducted beat is typically conducted with an RBBB morphology.

The Ashman phenomenon is typically seen in the setting of atrial fibrillation, where the RR interval varies from beat to beat. The phenomenon was first described in 1947.

The significance of the Ashman phenomenon is that the broad QRS complexes are commonly mistaken for ventricular ectopics (also known as premature ventricular complexes or PVCs). However, they're not ectopic beats. Instead, these broad QRS complexes represent the aberrant conduction of a supraventricular beat. The phenomenon itself is harmless, but it's important not to misdiagnose the aberrantly-conducted beats for PVCs.

To learn more about ECG interpretation, check out the ECG Black Belt Workshop at Medmastery. In addition, the following links will take you to some key references:

Gouaux JL, Ashman R. Auricular fibrillation with aberration simulating ventricular paroxysmal tachycardia. Am Heart J 1947; 34: 366-73.

Ashman Phenomenon at the Medscape website

Thursday 5 May 2016

Making sense of the apex beat

Palpation of the apex beat to assess its location and character is one of the key elements of a cardiovascular examination. However the terminology can cause confusion - after all, what exactly is a 'heaving' apex beat, and how does it differ from a 'thrusting' apex beat? Let's see if we can simplify and clarify things a bit.

It's possible to categorize 8 distinct types of apex beat:

The eight types of apex beat

Normal apex beat: This is the easy one. Located in the fifth intercostal space, in the mid-clavicular line, the normal apex beat has a relatively gentle pulsation.

Normal but shifted apex beat: Here the apex beat is normal in character, but its location is displaced because the heart itself has moved as a result of mediastinal shift, for instance due to tension pneumothorax. Remember to check for any tracheal deviation in the suprasternal notch. Another (extreme) example of a 'shifted' apex beat is when the patient has dextrocardia. In this case, the apex beat will be normal in character but impalpable on the patient's left, but it will be palpable on their right.

Impalpable apex beat: The apex beat may be impalpable because of hyperinflated lungs (COPD), high body mass index, or pericardial effusion.

Pressure-loaded apex beat: When there is left ventricular hypertrophy due to pressure overload on the left ventricle (e.g. hypertension, severe aortic stenosis), the apex beat is localized (i.e. not diffuse), undisplaced, and has a forceful character. This is sometimes also called a 'heaving' and 'sustained' apex beat, but I find these terms vague and open to interpretation.

Volume-loaded apex beat: In contrast to a pressure-loaded apex beat, a volume-loaded apex beat is diffuse (palpable over a larger area), displaced inferiorly and laterally, and has a less forceful character. This is sometimes called a 'non-sustained' apex beat. Examples include severe mitral or aortic regurgitation, which place the ventricle under volume overload.

Tapping apex beat: This describes the palpable first heart sound in mitral stenosis.

Double-impulse apex beat: Here there are two beats felt during each systole, and this is found in hypertrophic cardiomyopathy.

Dyskinetic apex beat: This describes an uncoordinated and diffuse apex beat seen with a left ventricular apical aneurysm.

As with all clinical skills, practice makes perfect. So assess the apex beat whenever you perform a cardiovascular system examination, and categorize your findings into one of the eight groups above.

To learn more about examination of the cardiovascular system (and other body systems), check out Chamberlain's Symptoms and Signs in Clinical Medicine, which I co-edited with David Gray.

Monday 2 May 2016

Sgarbossa criteria: How to diagnose myocardial infarction in LBBB

One of the commonest questions I get asked is whether it's possible to diagnose an acute ST-segment elevation myocardial infarction in the presence of pre-existing left bundle branch block (LBBB) on an ECG? Students are often taught that LBBB precludes analysis of the ST segments, but that's not really true. In short, it can be possible to diagnose MI in the setting of LBBB, and to do so we use the Sgarbossa criteria.

Fig. 1 Left bundle branch block

The Sgarbossa criteria use a scoring system to help us identify if an acute MI is present in the presence of LBBB. To use the criteria, we need to look for three features on the ECG and, for each feature we find, we add up the total number of points scored:
  1. The presence of ST segment elevation ≥1 mm in at least one lead with a positive QRS complex scores 5 points
  2. The presence of ST segment depression ≥1 mm in leads V1-V3 scores 3 points
  3. The presence of ≥5 mm ST segment elevation in leads with a negative QRS complex scores 2 points
A score of ≥3 points has a 90% specificity (but a poor sensitivity)
for the diagnosis of myocardial infarction

Fig. 2 Normal 'appropriate'
discordance in LBBB
The Sgarbossa criteria are based upon the principle of 'appropriate discordance'. Normally, in LBBB, any ST segment deviation in a particular lead will be in the opposite direction to the QRS complex in that lead.

In other words, if the QRS complex is negative, then any ST segment deviation will be positive (upright). Conversely, if QRS complex is positive, then the ST segment deviation will usually be negative (depressed). Normally, then, we see 'discordance' between the direction of the QRS complex and any deviation of the ST segment in LBBB (Fig. 2).

The first two Sgarbossa criteria are based on exceptions to this rule - they're looking for ST segment elevation in leads with a positive QRS complex, or ST segment depression in leads with a negative QRS complex. In other words, they're looking for evidence of 'concordance' instead of the normal discordance.

The third criterion is different - this time we're looking for extreme discordance - a highly-elevated ST segment in a lead with a negative QRS complex

Fig. 3 Extreme (>5 mm) ST segment elevation in
leads V1-V3, where the QRS complex is negative
In the example shown in Fig. 3, in leads V1-V3 the ST segments are elevated by >5mm above the isoelectric line - this therefore meets criterion #3 of the Sgarbossa criteria, and this ECG would score 2 points.

A modification to this third criterion has been proposed, in which the criterion is met if the discordant ST segment elevation is ≥25% of the depth of the preceding S wave. Using this modified criterion, instead of the "≥5 mm rule", the sensitivity for detecting acute MI is increased.

The Sgarbossa criteria can also be used in cases of right ventricular pacing (where the paced QRS complexes have an LBBB pattern), but in this situation the criteria are less specific for the diagnosis of acute MI.

To learn more about the Sgarbossa criteria, you can watch my explanatory video at the Medmastery website. In addition, the following links will take you to some key references:

Sgarbossa EB et al. Electrocardiographic Diagnosis of Evolving Acute Myocardial Infarction in the Presence of Left Bundle-Branch Block. N Engl J Med 1996; 334: 481-487.

Smith SW et al. Diagnosis of ST-Elevation Myocardial Infarction in the Presence of Left Bundle Branch Block with the ST-Elevation to S-Wave Ratio in a Modified Sgarbossa Rule. Annals of Emergency Medicine 2012; 60: 766-776.
Sgarbossa Criteria at the always-excellent Life in the Fast Lane website.