Diastolic dysfunction Rik Kapila FRCA Ravi P Mahajan DM FRCA

Diastolic dysfunction (DD) is increasingly being recognized as an important cause of heart failure. Often the condition may not be anticipated and difficult to differentiate from systolic dysfunction when symptoms develop. This article describes the pathophysiology, clinical features, diagnosis, and management of the condition.

Pathophysiology Diastole (Fig. 1) is the period of the cardiac cycle between closure of the aortic valve and that of the mitral valve. During this time, the left ventricle (LV) relaxes and refills, ready for the next systolic contraction. Conventionally, diastole is divided into four physiological phases: (i) isovolumetric relaxation: from closure of the aortic valve to opening of the mitral valve; (ii) early rapid filling: transmitral pressure gradient drives LV filling; (iii) diastasis: period of low flow in mid-diastole; (iv) late rapid filling: atrial contraction. In early diastole, the myocardium is unable to generate force, and it returns to its unstressed state. This is an energy-dependent process (lusitropy) requiring the sequestration of calcium from the cytosol. Relaxation allows the LV to fill and accommodate venous return without a significant increase in LV pressure. DD can be evaluated in terms of the onset, rate, and extent of this activity. LV distensibility can be quantified by the position and slope of the LV pressure –volume curve. A tangent drawn to the curve at any point (dp/dv) defines the distensibilty or compliance of the LV. An important attribute of the LV is its ability to accept a wide range of blood volumes without significant elevation in the filling pressure, that is, to become more compliant when faced with the demands of an increase in venous return. This allows the LV to accommodate increases

in venous return without a compromise in stroke volume. DD refers to abnormalities of active myocardial relaxation and passive ventricular filling. It can be attributed to one of the four underlying mechanisms. (i) Slow/incomplete myocardial relaxation: the most common cause of this is myocardial ischaemia, which causes the reduced rate of LV pressure decline. For any given degree of ischaemia, functional impairment of relaxation is greater in the hypertrophied heart; thus, patients with concentric LV hypertrophy (LVH) are particularly susceptible to DD. (ii) Impaired peak LV filling rate: this is because of the failure to generate an adequate trans-mitral pressure gradient, due to either elevated LV pressures or an inability to generate negative LV suction. (iii) Altered elasticity: alterations in cytoskeleton properties lead to passive viscoelastic stiffness of the LV. Neurohumoral responses initiate myocardial fibrosis that decreases LV compliance. (iv) Pericardial constriction: mechanical obstruction to LV expansion will raise LV pressure to transmitral pressure more quickly and thus terminate LV filling earlier. The net effect of all these pathological processes is a higher LV end-diastolic pressure (LVEDP) for any given LV end-diastolic volume (LVEDV)1 (Fig. 2).

Cellular mechanisms Just as the process of excitation –contraction coupling governs ionotropy, lusitropy is governed by repolarization –relaxation coupling. Excitation and contraction are triggered by a rapid increase in intracellular, free, ionized calcium. The termination of this ‘calcium transient’ is the essence of diastolic function; this is controlled by the intracellular regulation of calcium availability and sensitivity.

doi:10.1093/bjaceaccp/mkn046 Continuing Education in Anaesthesia, Critical Care & Pain | Volume 9 Number 1 2009 & The Board of Management and Trustees of the British Journal of Anaesthesia [2009]. All rights reserved. For Permissions, please email: [email protected]

Key points Lusitropy (early diastolic relaxation) is an active, ATPdependent process. Diastolic dysfunction (DD) results from abnormalities in relaxation and filling. DD is often asymptomatic. Doppler echocardiography provides a reliable, noninvasive technique for diagnosis and staging. Measures of flow, pressure, and tissue motion are indices of diastolic function. Distinguishing diastolic heart failure (DHF) from systolic heart failure (SHF) is important because of differences in treatment and prognosis.

Rik Kapila FRCA Specialist Registrar in Anaesthesia Queen’s Medical Centre Nottingham University Hospitals NHS Trust Derby Road Nottingham NG7 2UH UK Ravi P Mahajan DM FRCA Professor of Anaesthesia and Intensive Care University Division of Anaesthesia and Intensive Care Queen’s Medical Centre Nottingham University Hospitals NHS Trust Derby Road Nottingham NG7 2UH UK Tel: þ44 115 8231009 E-mail: [email protected] (for correspondence)

29

Diastolic dysfunction

Table 1 Prevalence of specific symptoms and signs in systolic vs DHF. Data are presented as per cent of patients in each group with the listed symptom or sign of heart failure. There were no statistically significant differences between patients with an ejection fraction .50% vs ,50%8

Fig 1 LV, left atrial, and aortic pressure traces for a single cardiac cycle showing the phases of diastole. A–D: Phases of diastole. A, isovolumetric relaxation; B, early rapid filling; C, diastasis; D, late rapid filling, atrial contraction.

Fig 2 Pressure– volume relationship for the LV during diastole. In DD, LV pressure is higher for any given LV volume. In addition, any given change in left ventricular volume will induce a greater change in LV pressure.

The ATP-dependent processes that allow restoration of resting tension are the following: (i) Dissociation of calcium from tropomyosin C. This is determined by the relative affinity of calcium for its tropomyosinbinding site. (ii) Lysis of actin–myosin tension bonds and release of cross-bridges. (iii) Phosphorylation of phospholamban: this myocyte protein regulates the activity of sarcoplasmic reticulum ATPase pumps. Its phosphorylation increases the rate of calcium sequestration. (iv) Clearance of calcium from the cytosol by sarcoplasmic reticulum and sarcolemmal ATPases, which transport calcium into the sarcoplasmic reticulum and across the sarcolemma, respectively. (v) Sodium/calcium exchange by secondary active transport. (vi) Restoration of sarcomere resting length.2 In DD, the calcium transient is prolonged as a result of dysfunction of any of the processes mentioned above. A decrease in

30

Symptoms Dyspnoea on exertion Paroxysmal nocturnal dyspnoea Orthopnoea Physical examination Jugular venous distension Displaced apical impulse S3 S4 Hepatomegaly Oedema Chest X-ray Cardiomegaly Pulmonary venous hypertension

Diastolic heart failure (EF .50%)

Systolic heart failure (EF ,50%)

85 55 60

96 50 73

35 50 45 45 15 30

46 60 65 66 16 40

90 75

96

ATP levels below a critical level, as is seen in myocardial ischaemia, can lead to both incomplete sequestration of calcium and incomplete actin– myosin dissociation, leading to a state of ‘partial persistent systole’. A small depletion in ATP tends to effect diastole more than systole, because the myosin heads have a higher ATP affinity and so dissociation requires greater ATP availability. The function of titin is another cellular mechanism of importance. Titin is a sarcomere spanning myocyte protein that is compressed during actin –myosin cross-bridge formation. Its function is to forcefully re-expand during early diastole, like a recoiling spring, generating a negative LV pressure that ‘sucks’ blood across the mitral valve. The magnitude of this negative force is matched to the force of LV contraction, so that greater systolic myocyte shortening facilitates greater LV filling without an increase in left atrial pressure (LAP). In its compressed state, titin also desensitizes the actin–myosin complex to calcium, which encourages actin– myosin dissociation to facilitate cellular relengthening.3

Clinical manifestations Asymptomatic DD is more prevalent than symptomatic disease. When present, the symptoms of DHF are indistinguishable from those of SHF (Table 1). These include: decreased exercise capacity; neurohumoral activation with sodium and water retention; paroxysmal nocturnal dyspnoea; and orthopnoea. Because of the increased chamber stiffness, patients with DD cannot increase LVEDV on exertion, preventing the necessary increase in stroke volume. Therefore, exercise intolerance is often an early symptom. Forty per cent of patients with heart failure have a preserved ejection fraction. The prevalence of DHF relative to that of SHF increases with age and hypertensive disease and is more common in women.4 Patients with DD have a particular intolerance of certain kinds of haemodynamic stress.

Continuing Education in Anaesthesia, Critical Care & Pain j Volume 9 Number 1 2009

Diastolic dysfunction

(i) Atrial fibrillation: with progression of DD, the atrial contribution to LV filling becomes progressively more important. The loss of atrial contraction in atrial fibrillation significantly compromises LV filling. Unfortunately, due to a high incidence of left atrial dilatation, atrial fibrillation is common in DD. (ii) Tachycardia: increasing heart rate truncates the late phase of diastolic filling, which becomes progressively more important with advancing disease to achieve an adequate LVEDV. (iii) Hypertension: increasing LV wall tension worsens myocardial relaxation. (iv) Ischaemia: The associated increase in LAP and PVP manifest clinically as respiratory symptoms of wheeze, shortness of breath, overt pulmonary oedema, and inability to take deep breaths. These symptoms are collectively referred to as ‘anginal equivalents’ in DD and often provide the earliest clinical indicators of the disease.5

Diagnosis Differentiation between DHF and SHF cannot be made on the basis of history, physical examination, ECG, and chest X-ray alone. The diagnosis of DD is often presumed from the presence of symptomatic heart failure with a preserved ejection fraction of .50%. Alternatively, it can be inferred from a raised serum brain natriuretic peptide and an ejection fraction of .50%. Studies of LV flow, volume, or pressure serve to confirm what is clinically suspected. The gold standard is the direct measurement of LV pressures and LV volumes by micromanometer and conductance catheters, respectively. The rate of LV pressure decline, the increase in LV pressure with atrial contraction, and the LVEDP can then be calculated; however, these techniques are both invasive and impractical. Echocardiography provides an acceptable, reliable, and non-invasive alternative.6

Echocardiography In addition to providing fundamental information on the chamber size, systolic function, and valvular integrity, 2D echocardiography can be used to analyse characteristics of diastolic filling. Left atrial enlargement with associated atrial fibrillation provides an easily identifiable indicator of DD and the need for further echocardiographic evaluation. (i) Transmitral flow velocity: the early diastolic peak filling velocity when the transmitral pressure gradient is greatest generates the E wave velocity on the echocardiogram (Fig. 3). The late diastolic peak filling velocity associated with atrial contraction generates the A wave. Because the normal atrial contribution to total diastolic filling is only 30%, a normal A wave is smaller than the mitral E wave, with an E/A ratio .1. DD initially produces a low E wave and a high A wave

Fig 3 Normal left ventricular haemodynamics. The upper panel shows the change in LV volume through a single cardiac cycle. The lower panel shows the rate of change of the LV volume (dV/dt). The cardiac cycle begins at end-diastole. With the onset or systole, LV volume decreases until end-systole. dV/dt reaches its maximum during early diastole and obtains a secondary peak filling rate with atrial systole. It is these two peak filling rates, E and A, that we use for echocardiographic evaluation of DD.

velocity, with reversal of the E:A ratio. As disease progresses and LV compliance is reduced further, LA pressure progressively increases to maintain a transmitral pressure gradient. The E wave increases until E/A ratios are .1.5. During the process of this transition, the E/A ratio will temporarily normalize, despite the presence of moderately severe disease. This is referred to as pseudonormalization and highlights a limitation to the sole use of E/A ratios for diagnosis. This problem can be overcome by altering the loading conditions on the myocardium, for example, with Valsalva or glycerine trinitrate administration during echocardiography. (ii) Pulmonary venous flow (PVF): the pulsatile PVF pattern is generated by the x and y descents of the LAP tracing. Atrial relaxation (x-descent) and LV diastole (y-descent) cause forward PVF. During atrial systole, there is normally a small amount of retrograde PVF. In DD, PVF reversal associated with atrial contraction becomes progressively more pronounced as LAP increases. (iii) Isovolumetric relaxation time: this is the time between aortic valve closure and mitral valve opening (Fig. 4). It reflects the myocardial relaxation time and is normally 70 (12) ms. With DD, poor relaxation prolongs IRT and a value of .110 ms is

Continuing Education in Anaesthesia, Critical Care & Pain j Volume 9 Number 1 2009

31

Diastolic dysfunction

Fig 4 Schematic representation of diastolic transmitral Doppler assessment. When compared with the normal pattern of diastolic transmitral flow, mild DD will show E:A reversal with a decrease in early diastolic flow velocity (E), a prolonged E wave DT, and an increased peak diastolic flow velocity with atrial contraction. As disease progresses, there is a compensatory increase in E wave with shortening of the DT and a decreasing contribution of A in ventricular filling.

considered significant. Pseudonormalization of this value also occurs with advancing disease because IRT becomes progressively shortened. (iv) Deceleration time (DT): the rate of dissipation of the transmitral pressure gradient is also a function of LV compliance (Fig. 4). The faster the LV pressure decreases the shorter the DT. Normal DT is 180 –240 ms. Again, the prolongation of the DT seen in early DD is reversed in moderate to severe disease, as there is a progressive compensatory increase in LAP. (v) Tissue Doppler: this uses Doppler shifts of ultrasound waves to calculate the velocity of myocardial tissue movement in a similar way to that of blood flow (Fig. 5). It can be used to assess the extent and timing of diastolic wall motion. As the cardiac apex is relatively fixed, the mitral valve ring moves towards the apex during systole. During diastole, the annulus initially moves away from the apex (E0 ) and then back towards the apex during atrial contraction (A0 ). These values are comparable with those of transmitral flow. The most clinically robust index of diastolic function is the combined assessment of transmitral flow and mitral annulus velocity. The E:E0 ratio most confidently separates normal filling pressures (,8) from elevated filling pressures (.12).7

Treatment Treatment of DD aims to control specific underlying pathology, such as hypertension or coronary artery disease, or it is targeted at

32

Fig 5 Transmitral diastolic Doppler in flow patterns are obtained across the open mitral valve leaflets. Here, we see a normal E:A ratio (A) associated with pseudonormalization and the subsequent demonstration of moderate DD using a Valsalva manoeuvre in the same patient (B). The third trace shows how the E and A waves change with advancing disease (C).

the signs and symptoms of DHF. Whether or not these interventions reverse the pathophysiological changes of DD is difficult to evaluate, as these interventions also alter the loading conditions on the heart. Treatment can be pharmacological or non-pharmacological.

Pharmacological Hypertension Regression of LVH is an important therapeutic goal, with angiotensin II inhibitors, calcium channel blockers, and ACE inhibitors affording the most significant regression. Choice of antihypertensive, however, must be within the context of other co-morbidities such as diabetes mellitus.

Continuing Education in Anaesthesia, Critical Care & Pain j Volume 9 Number 1 2009

Diastolic dysfunction

Chronic AF As LV filling in DD occurs predominantly in late diastole and is more dependent on atrial contraction, restoration and maintenance of sinus rhythm is preferred. When this cannot be achieved, rate control to preserve late diastolic time is also important. Digoxin, although useful in SHF, is not commonly used therapy for DHF, where the inotropic enhancement is not required and calcium sequestration may be further impaired.

Coronary artery disease Ischaemia induces an increase in LVEDP and an upward shift of the diastolic pressure –volume curve. Myocardial oxygen supply is key to the energy using processes of active relaxation, so reduction in myocardial oxygen demand can improve ventricular compliance and attenuate myocardial remodelling.

Volume overload Activation of the renin–angiotensin system serves to increase circulating volume and induces a progressive, compensatory increase in LAP to drive LV filling. Direct autocrine effects of angiotensin II and aldosterone on cardiac myocytes also perpetuate cardiac remodelling and progression of cardiomyopathy. However, ACE inhibitors, nitrates, and diuretics must be used with extreme caution in DD as these patients are exquisitely sensitive to the peripheral vasodilatation and reduction in circulatory volume that maintain a supra normal LAP for an adequate stroke volume.

Other therapies Direct alterations in ionic flux using calcium channel blockers and calcium sensitizers increase the rate of myocardial relaxation and intensive lowering of lipid levels with statins also shows promise as a future therapeutic strategy.

Prognosis The prognosis of DD depends on whether or not the patient is symptomatic. For symptomatic disease, the prognosis for DHF is more favourable than that of SHF. Annual mortality rates and hospitalization rates are lower. Independent predictors of mortality in DHF include older age, male gender, lower LV ejection fraction, peripheral vascular disease, and diabetes mellitus. In asymptomatic patients, all cause mortality does exceed that of age-matched controls even in mild disease (E:A ,0.6). As age increases above 70 yr, mortality rates for systolic and DD become nearly equivalent.6

References 1. Grossman W. Defining diastolic dysfunction. Circulation 2000; 101: 2020– 1 2. Villars PS, Hamlin SK, Shaw AD, Kanusky JT. Role of diastole in left ventricular function, 1: biochemical and biomechanical events. Am J Crit Care 2004; 13: 394 –405 3. Kass DA, Bronzwaer JG, Paulus WJ. What mechanisms underlie diastolic dysfunction in heart failure? Circ Res 2004; 94: 1533–42 4. Vasan RS, Benjamin EJ, Levy D. Prevalence, clinical features and prognosis of diastolic heart failure. J Am Coll Cardiol 2005; 26: 1565–74 5. Zile MR, Brutsaert DL. New concepts in diastolic dysfunction and diastolic heart failure. Circulation 2002; 105: 1387–93 6. European Study Group on Diastolic Heart failure. How to diagnose diastolic heart failure. Eur Heart J 1998; 19: 990–1003 7. Hamlin SK, Villars PS, Kanusky JT, Shaw AD. Role of diastole in left ventricular function, II: diagnosis and treatment. Am J Crit Care 2004; 13: 453–66

Non-pharmacological During exercise in healthy enhanced so that LV input output. This is achieved by during early diastole due to

enhances the transmitral pressure gradient without increasing LAP. This effect is lost in DD, so the aim of dynamic, isotonic exercise conditioning is to reduce resting heart rate, improve calcium uptake rate by sarcoplasmic reticulum, and induce physiological hypertrophy.7

individuals, diastolic function is remains precisely matched to LV a marked decrease in LV pressure a greater LV suction effect, which

8. Echeverria HH, Bilsker MS, Myerburg RJ, Kessler KM. Congestive heart failure: echocardiographic insights. Am J Med 1983; 75: 750– 5

Please see multiple choice questions 26 –30

Continuing Education in Anaesthesia, Critical Care & Pain j Volume 9 Number 1 2009

33