NDT Advance Access originally published online on April 16, 2007
Nephrology Dialysis Transplantation 2007 22(7):1815-1819; doi:10.1093/ndt/gfm224
© The Author [2007]. Published by Oxford University Press on behalf of ERA-EDTA. All rights reserved. For Permissions, please email: journals.permissions@oxfordjournals.org
Are we overestimating left ventricular abnormalities in end-stage renal disease?
Patrick B. Mark,
Rajan K. Patel and
Alan G. Jardine
BHF Glasgow Cardiovascular Research Centre, University of Glasgow, Glasgow G12 8TA, Scotland, UK
Correspondence and offprint requests to: Dr Patrick B. Mark, BHF Glasgow Cardiovascular Research Centre, University of Glasgow, Glasgow G12 8TA, Scotland, UK. Email: p.mark{at}clinmed.gla.ac.uk
Keywords: cardiomyopathy; dialysis; echocardiography magnetic resonance imaging; end stage renal failure; left ventricular hypertrophy
 |
Introduction
|
|---|
Uraemic cardiomyopathy, defined by the presence of left ventricular
hypertrophy (LVH), left ventricular (LV) dilatation or LV systolic
dysfunction (LVSD), is reported to be a predictor of premature
cardiovascular mortality in patients with end-stage renal disease
(ESRD) [
1]. Each of these LV abnormalities cumulatively confers
a poorer long term prognosis [
2]. LV function, dimensions and
geometry have conventionally been assessed by echocardiography.
Echocardiography is well established, portable and widely available
and therefore will remain a vital clinical tool. However, echocardiography
has drawbacks, including operator-dependence and difficulty
in obtaining satisfactory acoustic windows. A major technological
limitation of echocardiography is the geometric assumption required
to calculate LV mass and function. Using conventional M-mode
echocardiography to assess LV mass by the Penn [
3] or ASE formulae
[
4], ventricular dimensions are measured in the minor axis and
then volumes estimated from the cube of these values, hence
magnifying initial errors. These techniques assume the left
ventricle to be approximately cubic and, although validated
in normal hearts, patients with ESRD have a high prevalence
of LV abnormalities. As LVH in ESRD is due to a combination
of pressure and volume overload, both concentric and eccentric
remodelling ensue, with approximately 30% of patients exhibiting
eccentric remodelling, that may invalidate the use of these
formulae to calculate LV mass [
5]. Moreover, definition of cardiomyopathy
by echocardiography is difficult, due to the large variation
in intravascular (and hence intraventricular) volume that occurs
in the interdialytic period and during dialysis therapy [
6].
Whilst these changes should not affect haemodynamic assessments
such as measurement of ejection fraction or cardiac output,
any inaccuracy in defining the left ventricular internal dimensions
will lead to errors in detecting the presence of LV dilatation.
Overall, echocardiographic measurements are critically dependent
on phase of the dialysis cycle and ability to assess dry weight.
Alternative techniques, e.g. cardiac magnetic resonance imaging
(CMR) offer a volume-independent assessment.
|
Comparisons between echocardiography and cardiac magnetic resonance imaging for measurement of LV dimensions
|
|---|
Using M-mode echocardiography, a 1D technique, it has been demonstrated
that echocardiography overestimates left ventricular mass in
ESRD compared to CMR [
7]. This disparity between the two methods
has also been demonstrated in other patient populations, such
as patients with LVH secondary to either hypertension [
8] or
aortic stenosis [
9]. Conversely, in one large study of healthy
army recruits, echocardiography (ASE formula) consistently underestimated
LV mass compared to CMR by a mean of 14.3 g [
10]. These studies,
in keeping with those in ESRD, suggest that in patient groups
with a high prevalence of LVH (and therefore distorted LV geometry)
echocardiography tends to overestimate LV mass, but the difference
between the methods is minimal or likely to be reduced or reversed
when larger populations with normal hearts are studied.
|
Uraemic cardiomyopathy as assessed by cardiac magnetic resonance imaging
|
|---|
In incident ESRD patients, previous studies have suggested that
the echocardiographic prevalence of LVH is 50–80%, with
left ventricular dilatation in 20–40% and LVSD present
in approximately 16% of patients [
2]. However in a cohort of
ESRD patients being assessed for renal transplantation assessed
by CMR, LVH was present in 71.6% patients, LV dilatation in
11.2% and LVSD in 8.2% [
11]. The lower prevalence of LVSD may
reflect the fact that potential transplant recipients are a
‘healthier’ ESRD cohort. Contrast-enhanced CMR with
gadolinium-DTPA based in this study identified that the presence
of both LVSD and LV dilatation are essentially restricted to
patients with underlying coronary artery disease. This represents
the only large cohort study using CMR in ESRD and it should
be noted that the cohort of patients is smaller than in previous
echocardiographic studies.
The majority of patients with ESRD also had LVH, the mean LV mass index was approximately 65 g m–2 less than in similar large echocardiographic studies [1]. The implication is that whilst LVH remains the prevalent form of cardiomyopathy in patients with ESRD, by using CMR, the proportion of patients with LV dilation is reduced, as accurate definition of the ventricular chamber is allowed and systematic overestimation of LV mass is avoided. Studies performing echocardiography pre- and post-dialysis suggest that the measured LV mass index differs by 26 g m–2, (equivalent to an approximate change in absolute LV mass of 45 g) dependent on the timing of study and the associated changes in intravascular volume [12]. With CMR, the change in LV mass before and after dialysis has been shown to be of the order of 10 g [6]. Whilst some of the reduction in LV mass post dialysis is due to changes in the water content of the interstitial tissues of the heart in volume overload, the disparity between CMR and echocardiography suggests that any error in calculation of LV mass due to changes in chamber dimension is greatly magnified when echocardiography is used. Additionally, these findings suggest that LV chamber dimensions are dependent on hydration status and have been shown by CMR to significantly vary during the inter-dialytic period in patients receiving haemodialysis [6]. Accurate definition is therefore required to discriminate between the presence of LV dilatation due to hypervolaemia and that due to underlying ischaemic, rather than uraemic, cardiomyopathy.
It is worth stressing that initial studies comparing CMR to echocardiography were performed at 1.0 Tesla providing lower resolution CMR images [7,9,10] compared to more recent studies performed using a 1.5 Tesla scanner using a fast imaging with steady state in free precession sequence (the current convention for CMR studies) [6,9,11] allowing better contrast between the ventricular myocardium and blood pool.
Although post-mortem validation comparing in vivo CMR images with post mortem specimens has not been performed in humans, ex vivo imaging studies show close agreement between CMR measured ventricular mass and true LV mass [13]. A number of studies have shown CMR to be reproducible with low inter- and intra-observer variability (typically approximately 5%), both in normal volunteers as well as in patients with left ventricular hypertrophy, heart failure, myocardial infarction and dilated cardiomyopathy [14].
|
Can CMR allow us to improve echocardiographic methods in patients with ESRD?
|
|---|
Comparison of these methods and development of algorithms to
improve echocardiography methods is dependent on the fundamental
premise that CMR does truly represent a ‘gold standard’
for measurement of LV dimensions, i.e. a reproducible, accurate
method, independent of geometric assumptions and ideally validated
with post mortem specimens. No attempt has yet been made to
derive a correction factor to optimise M-mode echocardiographic
measures of left ventricular mass in the ESRD population. Derivation
of a revised echocardiographic formula for estimation of LV
mass in ESRD is an attractive concept. Data from a pilot study
(
Figure 1) illustrates how such a formula may be developed.
The relationship between the two methods to measure LV mass
is approximately linear; therefore, a revised formula may only
require a simple modification to the constant in the conventional
Penn cube formula. Additionally,
Figure 1 demonstrates that
patients with LVMI >200 g m
–2 by echocardiography are
likely to have LVH confirmed by CMR. It is only patients with
lesser degrees of LVH who may be mislabelled as having ‘uraemic
cardiomyopathy’ (in the form of LVH), who may in fact
have normal left ventricles. A larger comparative study is required
to investigate both these notions, as well as to analyse whether
gender difference or remodelling pattern is of importance in
revising echocardiographic formulae.
Additionally, there are fundamental differences in the methods
of analysis. LV mass is calculated by planimetry using a modified
Simpson's rule (
Figure 2). Conventionally, the papillary muscles
are included in drawing endocardial borders to perform CMR analysis
of LV mass [
15]. With echocardiography, the papillary muscles
are avoided for M-mode measurements and excluded from the endocardial
border for bi-plane measurements. This would appear to be a
crucial issue but perhaps, as both methods, echocardiography
and CMR, have evolved essentially separately and been validated
against post-mortem specimens, animal models and latex casts,
this issue is underplayed in the literature.
View larger version (95K):
[in this window]
[in a new window]
[Download PowerPoint slide]
|
Fig. 2. Schematic view of analysis of LV dimensions in a patient with severe LVH by CMR from horizontal long axis view (above) with short axis slices indicated. The end diastolic contours are demonstrated on the short axis slices (below)
|
|
|
Improvements to echocardiographic technique for assessment of left ventricular mass
|
|---|
Although less widely used to estimate LV mass than the M-mode
method, 2D echocardiography is more accurate and reproducible
than M-mode methods. However, one study using these methods
in patients with hypertension suggests that this technique also
suffers from intra-observer variability and wide limits of agreement
compared to CMR [
16]. Using intravenous contrast combined with
harmonic Doppler imaging has been shown to improve the accuracy
of LV mass measurements with echocardiography [
17]. Finally,
3D echocardiography measures of LV mass shows close correlation
with CMR measures, although there are relatively few comparative
studies [
18]. It should therefore be remembered that the deficiencies
in using echocardiography to define LVH reflect the assumptions
used to calculate LV mass rather than the technique itself.
The more complex echocardiographic techniques compare much more
closely with CMR for measurements. These benefits may come at
the expense of the ease and convenience of conventional echocardiography.
|
Limitations of CMR
|
|---|
CMR is obviously less widely available than echocardiography
and is more expensive. To develop from a research tool into
a more widely used clinical investigation, reduction in scan
time is required, at least to perform a routine study of cardiac
mass and function. Automated analysis software will allow the
results of such studies to be rapidly processed. A number of
patients will be unable to undergo scanning, due to standard
contraindications to MRI such as claustrophobia, presence of
a pacemaker or implantable cardiac defibrillator or other implanted
ferromagnetic objects. Additionally, although contrast-enhanced
CMR with gadolinium-based contrast agents offered some promising
insights into myocardial tissue abnormalities in ESRD [
11],
the recent association between use of these contrast agents
and development of nephrogenic systemic fibrosis will makes
further study using these agents impossible, until this safety
issue is resolved [
20].
|
Implications for treatment of uraemic cardiomyopathy
|
|---|
The prognostic implications of uraemic cardiomyopathy defined
by CMR have not yet been reported. Although echocardiography
may overestimate the prevalence of LVH, survival studies suggest
that normal LV dimensions measured by echocardiography are associated
with good long-term outcome [
2]. As LVH is associated with poor
outcome in ESRD, regression of LVH remains a therapeutic target.
The ability to accurately detect small non-artefactual changes
in LV mass by CMR allows smaller sample size in clinical trials
of regression of LVH. One prospective study, comparing the effect
of conventional versus nocturnal haemodialysis on LV mass, estimates
that a total sample size of 38 patients (19 per limb allowing
for patient dropout) will be required to detect a 10 g reduction
in LV mass [
19]. Potentially, such studies may be conducted
in a single centre, reducing the costs and organisational support.
Ultimately to reduce cardiovascular risk in patients with ESRD,
better definition of cardiac dimensions is required to facilitate
identification of targets for intervention. Whether this is
by expanded use of CMR, 3D echocardiography or optimization
of conventional echocardiography remains to be seen. Hopefully,
these novel developments in cardiac imaging will translate into
detection of superior methods of regression of LVH in appropriately
powered clinical trials, either in patients with ESRD, or preferably,
in patients with less severe chronic kidney disease, when regression
of LVH is likely to be a more achievable goal.
|
Acknowledgements
|
|---|
PBM is funded by the British Heart Foundation.
Conflict of interest statement. None declared.
|
References
|
|---|
- Foley RN, Parfrey PS, Harnett JD, Kent GM, Murray DC, Barre PE. The prognostic importance of left ventricular geometry in uremic cardiomyopathy. J Am Soc Nephrol (1995) 5:2024–2031.[Abstract]
- Parfrey PS, Foley RN, Harnett JD, Kent GM, Murray DC, Barre PE. Outcome and risk factors for left ventricular disorders in chronic uraemia. Nephrol Dial Transplant (1996) 11:1277–1285.[Abstract/Free Full Text]
- Devereux RB, Reichek N. Echocardiographic determination of left ventricular mass in man: anatomic validation of the method. Circulation (1977) 55:613–618.[Abstract/Free Full Text]
- Devereux RB, Alonso DR, Lutas EM, et al. Echocardiographic assessment of left ventricular hypertrophy: comparison to necropsy findings. Am J Cardiol (1986) 57:450–458.[CrossRef][ISI][Medline]
- Zoccali C, Benedetto FA, Mallamaci F, et al. Prognostic impact of the indexation of left ventricular mass in patients undergoing dialysis. J Am Soc Nephrol (2001) 12:2768–2774.[Abstract/Free Full Text]
- Hunold P, Vogt FM, Heemann UW, et al. Myocardial mass and volume measurement of hypertrophic left ventricles by MRI–study in dialysis patients examined before and after dialysis. J Cardiovasc Magn Reson (2003) 5:553–561.[CrossRef][ISI][Medline]
- Stewart GA, Foster J, Cowan M, et al. Echocardiography overestimates left ventricular mass in hemodialysis patients relative to magnetic resonance imaging. Kidney Int (1999) 56:2248–2253.[CrossRef][ISI][Medline]
- Missouris CG, Forbat SM, Singer DR, et al. Echocardiography overestimates left ventricular mass: a comparative study with magnetic resonance imaging in patients with hypertension. J Hypertens (1996) 14:1005–1010.[ISI][Medline]
- Rajappan K, Bellenger NG, Melina G, et al. Assessment of left ventricular mass regression after aortic valve replacement: cardiovascular magnetic resonance versus M-mode echocardiography. Eur J Cardiothorac Surg (2003) 24:59–65.[Abstract/Free Full Text]
- Myerson SG, Montgomery HE, World MJ, et al. Left ventricular mass: reliability of M-mode and 2-dimensional echocardiographic formulas. Hypertension (2002) 40:673–678.[Abstract/Free Full Text]
- Mark PB, Johnston N, Groenning BA, et al. Redefinition of uremic cardiomyopathy by contrast-enhanced cardiac magnetic resonance imaging. Kidney Int (2006) 69:1839–1845.[CrossRef][ISI][Medline]
- Harnett JD, Murphy B, Collingwood P, et al. The reliability and validity of echocardiographic measurement of left ventricular mass index in hemodialysis patients. Nephron (1993) 65:212–214.[ISI][Medline]
- Bottini PB, Carr AA, Prisant LM, et al. Magnetic resonance imaging compared to echocardiography to assess left ventricular mass in the hypertensive patient. Am J Hypertens (1995) 8:221–228.[CrossRef][ISI][Medline]
- Myerson SG, Bellenger NG, Pennell DJ. Assessment of left ventricular mass by cardiovascular magnetic resonance. Hypertension (2002) 39:750–755.[Abstract/Free Full Text]
- Alfakih K, Plein S, Thiele H, Jones T, Ridgway JP, Sivananthan MU. Normal human left and right ventricular dimensions for MRI as assessed by turbo gradient echo and steady-state free precession imaging sequences. J Magn Reson Imaging (2003) 17:323–329.[CrossRef][ISI][Medline]
- Alfakih K, Bloomer T, Bainbridge S, et al. A comparison of left ventricular mass between two-dimensional echocardiography, using fundamental and tissue harmonic imaging and cardiac MRI in patients with hypertension. Eur J Radiol (2004) 52:103–109.[CrossRef][ISI][Medline]
- Bezante GP, Chen X, Molinari G, et al. Left ventricular myocardial mass determination by contrast enhanced colour Doppler compared with magnetic resonance imaging. Heart (2005) 91:38–43.[Abstract/Free Full Text]
- Sugeng L, Mor-Avi V, Weinert L, et al. Quantitative assessment of left ventricular size and function: side-by-side comparison of real-time three-dimensional echocardiography and computed tomography with magnetic resonance reference. Circulation (2006) 114:654–661.[Abstract/Free Full Text]
- Walsh M, Manns BJ, Klarenbach S, Quinn R, Tonelli M, Culleton BF. The effects of nocturnal hemodialysis compared to conventional hemodialysis on change in left ventricular mass: rationale and study design of a randomized controlled pilot study. BMC Nephrol (2006) 7:2.[CrossRef][Medline]
- Marckmann P, Skov L, Rossen K, et al. Nephrogenic systemic fibrosis: suspected causative role of gadodiamide used for contrast-enhanced magnetic resonance imaging. J Am Soc Nephrol. (2006) 17:2359–2362.[Abstract/Free Full Text]
Received for publication: 24.10.06
Accepted in revised form: 22. 3.07
CiteULike 
Connotea 
Del.icio.us What's this?
Top
Introduction
Comparisons between...