DR. TRACY: I think differentiating between systolic and diastolic dysfunction is relatively easy because you look primarily at the EF: if a patient has an EF of 50% or more, you most likely have ruled out systolic dysfunction (for this discussion, we cannot discuss normal LVEF with abnormal systolic function). We could discuss further about whether that it is still feasible/appropriate, but for the purpose of this discussion, if an echocardiogram within approximately 12 hours of admission demonstrates an EF of 50% or greater, one should evaluate diastolic dysfunction.
To diagnose diastolic dysfunction, you definitely have to look at many different parameters. You have to look at what is called the mitral valve inflow while looking at the pattern and ratio between early (E) and late (atrial, A) ventricular filling velocities (E/A ratio), tissue Doppler obtained from the mitral valve annuli, also to review the E/e prime ratio, and the pulmonary vein flow, which I have found to be extremely helpful.
Pulmonary vein flow is not easy to measure in patients because you need the sample volume of the echocardiogram at least 1 to 2 cm within the orifice of the pulmonary vein, and I like to try to get more than one pulmonary vein, if possible, as the information that you can get on pulmonary vein flow is really helpful.
I then look at pulmonary pressures and see if they are elevated. I also check the left atrial size and left atrial indexed volume. All this should be obtained on a standard echocardiogram for every patient.
I then examine these parameters to see where one parameter is indicates diastolic dysfunction. So, it’s not just one index that you should consider for diastolic dysfunction. You have to examine these individual parameters for each patient to decide where the patient is fitting. If you consider all of these different indices, you will have a good way of diagnosing a patient with either normal diastolic dysfunction or the range of diastolic dysfunction.
DR. LIEBSON: Dr. Doukky, do you have any further comments about this?
DR. DOUKKY: I agree with Dr. Tracy. You have to look at all indices. I also make an effort to not only specify the stage of diastolic dysfunction but also comment on whether the LV filling pressure is elevated or not, as it may explain the patient’s dyspnea, for example. On the clinical level, knowing whether the patient is volume overloaded or not is most useful for the managing physician.
DR. LIEBSON: Different patients obviously have different issues other than the diastolic pressure, but are there any situations where it may be difficult to assess diastolic dysfunction using echocardiography?
DR. DOUKKY: Yes, there are situations where it is relatively difficult to interpret diastolic function. In atrial fibrillation, for example, the loss of the atrial contraction limits our ability to assess E/A ratio and changes the pulmonary vein flow pattern. In this case, I find the E/e' ratio to be very useful. A high ratio still identifies high LV filling pressure, which is most important clinically. Secondly, the absolute e' velocity provides a useful assessment of LV diastolic relaxation, independent of the loading condition of the patient. In addition to atrial fibrillation, sinus tachycardia represents a similar challenge.
DR. LIEBSON: I think it’s important that the e', or the filling, based upon the mitral annular motion in early diastole, is not heart-rate dependent, and the ratio of E/e' may be very important in atrial fibrillation, where, if you could determine the mitral inflow E and the e' of the mitral annulus at the same time, you could obtain some valuable information about filling pressures.
DR. TRACY: I would like to add to Dr. Doukky’s point. The echocardiogram can also be difficult to interpret in patients who have body habitus issues and chronic obstructive pulmonary disease or those with tachycardia accompanied by extreme shortness of breath. Thus, both atrial fibrillation, where you have to take an average of at least 10 cardiac cycles, and tachycardia, where the E and A wave can be fused, make it difficult to examine and use the echocardiogram parameters.
In order to ensure that our referring physicians do not get frustrated, we have to remember that not every patient has beautiful echocardiographic windows, and we may have to look at other parameters if that patient has a difficult body habitus and bad chronic obstructive pulmonary disease, or if the patient is technically unable to withstand an echocardiogram. I just want to preface that because we still may have technically limited studies.
DR. LIEBSON: Suppose we have the results of the echocardiography and they show that there is a diastolic relaxation abnormality and evidence of diastolic dysfunction, while the systolic function is normal. The next question that arises is what sort of intervention is required. In other words, how does the indication that there is diastolic dysfunction affect treatment?
DR. DOUKKY: I think it comes down to a couple of issues: one is whether there is high LV filling pressure, and second, whether the patient is symptomatic. In symptomatic patients, evidence for elevated LV filling pressure, often seen in grade 2 or 3 diastolic dysfunction, can actually explain the patient’s symptoms. In such a case, treatment with diuretic can relieve the patient’s symptoms. Additional interventions, such as angiotensin II receptor blockers (ARBs), may improve the associated morbidity and lower hospital readmission rate as has been shown in the Effects of Candesartan in Patients with Chronic Heart Failure and Preserved Left-Ventricular Ejection Fraction (CHARM-Preserved) trial.9
Another important piece of information is whether the patient has restrictive filling pattern and whether it is reversible. Irreversible restrictive filling pattern after diuretic treatment or the Valsalva maneuver carry very poor prognostic implications and such a patient may be considered for a heart transplantation.
On the other hand, echocardiographic findings of preserved systolic function and grade 1 diastolic dysfunction without evidence of elevated filling pressure carries a good prognosis and may simply represent relaxation abnormality associated with the normal aging process. However, grade 1 diastolic dysfunction in patients younger than 60 years of age may represent impaired relaxation most commonly caused by long-standing hypertension and may be a precursor to more advanced diastolic impairment. In the latter case, no specific intervention other than addressing the underlying condition, such as hypertension, is needed.
Occasionally, we encounter what some call grade 1B diastolic dysfunction, which is when you have a relaxation abnormality manifesting as reversal of E/A ratio, but along with that, you have evidence for elevated LV filling pressure demonstrated by an increased E/e' ratio. You can think of this as a transitional phase between diastolic dysfunction grades 1 and 2. This may be associated with heart failure symptoms, which may improve with diuretics.
DR. LIEBSON: Dr. Tracy, do you feel that there are certain classes of pharmacologic agents that might be more helpful in patients with diastolic dysfunction?
DR. TRACY: The data at this point do not support one specific form of a medication that is going to be the most beneficial. Obviously, angiotensin-converting enzyme (ACE) and ARBs, beta-blockers, and diuretics are paramount. The problem with diastolic dysfunction is that we have actually not been able to treat diastolic dysfunction as well as we would like to.
Patients diagnosed with diastolic heart failure actually portend a poor prognosis. One reason may be that we do not treat it as aggressively and as early as we should. It could also be that we think we have to focus on patients with systolic dysfunction and make sure that their blood pressure is better and under control and that their fluid status is tighter.
But the diastolic dysfunction patient actually portends a worse prognosis. I think it is important that when we encounter a patient with grades 1, 1A, or 1B diastolic heart failure, the referring physician must establish that the patient does not have a previous history of diastolic dysfunction and that they are not currently being treated for blood pressure. For this patient, the mild blood pressure must be monitored and the physician must check if maybe diet, exercise, and losing weight would control the blood pressure better.
If the condition has taken a toll on the heart muscle and is no longer just isolated elevated blood pressure, but also shows signs of early diastolic dysfunction, then you need to be aggressive with that patient’s treatment to be able to control the blood pressure.
I have had female patients who I have performed echocardiography for while they were pregnant. I would perform an echocardiography for them for tachycardia and for diagnosing grade 1 diastolic dysfunction. I have actually been performing an echocardiography on such patients 6 months post-partum because the diastolic dysfunction showing on a baseline echocardiogram may be a marker of diastolic dysfunction that is going to develop in these patients.
So, I repeat the echocardiography to see if the earlier findings are just a normal physiologic response to pregnancy or an early marker of what’s to come. With diastolic dysfunction, as Dr. Doukky said, I also use ACE inhibitors, ARBs, and diuretics. Those have been shown to be our best therapeutic interventions for diastolic dysfunction.
DR. LIEBSON: I would just like to add an overview on various types of agents. With regards to diastolic stiffness, especially involving the LV, the only current therapy with a salutary effect on vascular and ventricular stiffness that leads to reduced smooth muscle growth, reduced growth factor expression, and regression of myocardial fibrosis consists of the class of ACE inhibitors. There is evidence that ARBs and aldosterone receptor antagonists may also have efficacy in decreasing myocardial fibrosis and thus diastolic dysfunction. Beta-blockers and certain calcium channel blockers (verapamil and diltiazem) may also be helpful by prolonging diastolic filling time.10–15
ACE inhibitor therapy, as we know, may result in favorable remodeling of the atrium independent of blood pressure.16
There have been several clinical trials involving patients with heart failure having normal EFs. These include the CHARM-Preserved perindopril trial9 and the Irbesartan in Heart Failure with Preserved Systolic Function (I-PRESERVE) trial.17 All of these patients supposedly had normal EF, but the actual EFs were ≥40%. Now, we know that in echocardiography, we consider the normal EF to be >50%, so some of these patients certainly had EFs below what we considered normal.
The diastolic indices in these studies included the intraventricular relaxation time, the E/A ratio, the deceleration time, and the left atrial dimension. The problem was that in all 3 of these studies, there was no evidence that diastolic function had any prognostic significance with the primary outcome, which in these cases, were deaths or heart failure hospitalization. However, there is some evidence in the CHARM-Preserved trial that candesartan, an ARB, significantly reduced hospitalization for congestive heart failure and in the Perindopril in Elderly People with Chronic Heart Failure (PEP-CHF) study,18 there was some improvement in symptoms and exercise capacity with perindopril administration.