© 2005 The European Society of Cardiology. Published by Elsevier Ltd. All rights reserved.
EDITORIAL
The impact of changing oedematous states on the QRS duration: implications for cardiac resynchronization therapy and implantable cardioverter/defibrillator implantation
aMount Sinai School of Medicine, New York University New York, NY, USA; bDivision of Cardiology, Elmhurst Hospital Center 79-01 Broadway, Elmhurst, NY 11373, USA
Manuscript submitted 9 September 2004. Tel.: +1 718 334 5005; fax: +1 718 334 5990. E-mail address: madiasj{at}nychhc.org
Abstract
Increased ECG QRS duration (QRSd) in patients with dilated cardiomyopathy (DCM) or heart failure (HF) is a well-known phenomenon. The QRSd is not a static ECG measurement but shows fluctuations, and its recent inclusion among the parameters used in referring patients for implantable cardioverter/defibrillators (ICDs) or cardiac resynchronization therapy (CRT) has led to renewed interest in its natural course and its determinants. Although clinical deterioration has been traditionally associated with increasing QRSd, its changes often are left unexplained. Also, the recent description of a decrease in QRSd, well correlated with attenuated amplitude of QRS complexes in patients with peripheral oedema (PERO) in the context of a variety of illnesses, has added complexity to the matter. This communication aims at calling attention to the importance of a few clinical and ECG parameters when documenting changes in the QRSd in serial ECGs. Thus, presence or absence of PERO and change in the patients' weight, along with alteration in the amplitude of QRS complexes and shifts to/from incomplete/complete bundle branch block patterns, all should be considered when assessing changes in QRSd for meaningful follow-up of patients with DCM or CHF, or referral for ICD or CRT. Evaluation of the QRSd as a selection parameter for referring patients suitable for device implantation should continue along with the employment of mechanical analysis of ventricular dyssynchrony. Although reference here is made to QRSd particularly in connection with DCM and HF, the above apply to other oedematous states (i.e. patients with chronic renal failure, or those undergoing haemodialysis).
Key Words: QRS duration, ECG, anasarca, peripheral oedema, congestive heart failure, low voltage ECG, implantable cardioverter/defibrillator, cardiac resynchronization therapy
The ECG QRS duration (QRSd) in milliseconds represents the time required for the two ventricular chambers to be activated, and is measured manually in each ECG lead, or is provided as a mean QRSd (the mean of the QRSd of all, or some of the 12 ECG leads) by automated computer-based ECG interpretation programmes [1]
. In the latter case the mean QRSd is taken as an expression of the QRSd of all 12 leads; furthermore, advanced interpretation programmes provide separate measurement of the QRSd for each ECG lead [1]. The normal QRSd is 
100 ms, and values >100 ms denote an intraventricular conduction delay (IVCD), or bundle branch block (BBB) [2–5].
Increased QRSd often accompanies heart failure (HF) due to dilated cardiomyopathy (DCM) of ischaemic and non-ischaemic variety, and has an impact on prognosis [6–15]. Patients with these conditions are prone to sudden cardiac death, and it has been recently recommended that they receive an implantable cardioverter/defibrillator (ICD) even in the absence of evidence for sustained or non-sustained ventricular tachycardia, and without prior electrophysiological evaluation for induction of such arrhythmias via programmed electrical stimulation [16,17]. In fact preliminary data on the outcomes of ICD recipients showed that a QRSd >120 ms was the strongest, amongst various ECG criteria, predictor of benefit with a 63% reduction in mortality relative to patients treated conventionally (P = 0.004) [17]. HF due to diastolic dysfunction is also frequently manifest as pulmonary venous hypertension, and lung congestion; such patients have preserved or even increased ejection fraction, and as a rule do not have an increased QRSd, although with significant left ventricular hypertrophy, some widening of the QRS may be detected. Moreover QRSd >90 ms confers 1 point when scoring the ECG diagnosis of left ventricular hypertrophy [18]. Patients with HF/DCM are also prone to clinical deterioration, requiring frequent hospitalization even when they are maintained on an optimal drug regimen [19]. For patients caught in this spiral of clinical deterioration, and with IVCD or BBB, particularly of the left bundle branch block (LBBB) variety cardiac resynchronization therapy (CRT) has been advocated [7–12,20]. CRT entails synchronous pacing of both ventricles, but occasionally for patients with LBBB only the left ventricle is paced somewhere in the lateral region, or in the particular site with the most delayed activation at baseline [7–12,20]. CRT is no longer an investigational procedure, since guidelines recommend it for selected patients with ischaemic and non-ischaemic DCM, and HF (New York Heart Association class III and IV), listing it as a Class IIa indication [17,21–23].
CRT is currently implemented in a small fraction of patients with DCM and HF, but not all receiving such device show amelioration of their condition [14]. Thus, it is imperative to identify the subgroup(s) of patients who will benefit from CRT, prior to its wide application. Accordingly, it has been suggested that, along with the clinical evaluation and laboratory testing showing advanced left ventricular or biventricular systolic dysfunction, a QRSd >120 ms or >130 ms, particularly with LBBB morphology identifies patients who may benefit from CRT [20,22,23]. By "biventricular" herein, it is implied that CRT is implemented for left ventricular systolic dysfunction, since there is little evidence available that CRT is advisable for right ventricular systolic dysfunction. It is not always clear from the literature that the candidates for CRT had their QRSd measured by computer or manually [8,20], although some authors [24] indicate that automation was employed. The literature lacks precise statements concerning how many ECG leads were measured in order to calculate the mean QRSd, rarely is it stated that all 12 leads were employed [8]. Although automated measurements in general are expected to be more reliable and reproducible, candidates for CRT have BBB, and thus manual measurements of QRSd probably are less prone to error by virtue of the wider QRS intervals. The significance of baseline QRSd is shown by its association with mechanical improvement (%
dP/dtmax), consequent upon biventricular, left ventricular, and right septal pacing [8]. However, the QRSd is not the prime or the only determinant of response to CRT in HF patients, but the pattern(s) of ventricular activation and intra-, and interventricular conduction delay; this is why normalization of the QRSd is not always associated with clinical improvement. Also, it is intriguing that haemodynamic response to CRT has been found without QRSd shortening, or even with some QRS widening [8]. Finally, the association of incremental changes over time in the QRDd with worsening of HF and increasing mortality [15] might be useful in identifying the subgroup of patients for whom ICD and CRT should be targeted. The above should be considered in the context of the current implementation of mechanical analysis of left ventricular dyssynchrony using radionuclide angiography, and echocardiographic and tissue Doppler imaging. Clinicians and researchers are trying to reconcile evidence that patients with normal QRSd improve after CRT, and the frequent dissociation between QRSd and left ventricular systolic synchrony [25–27].
This reliance on the QRSd in management decisions has enhanced its utility, and requires a re-evaluation of its stability or possible changes over time in both health and disease. In a comprehensive treatment of the topic, provided in a classic textbook [2], different methodologies for calculating the QRSd are discussed as they apply to single and multiple leads; the QRSd depends on the method used to detect the onset and offset of the QRS complexes, and pattern recognition applied to several vectocardiographic and ECG programmes. The same book provides normal values of QRSd for different ages and different populations [2]. Finally, an exposition is provided in the same source on the importance of identification of the different waveforms, the "thickness" of the baseline, and the frequent employment of averaging a number of QRS complexes with similar configuration for both manual and automated measurements. It appears that the "field" of how the QRSd is determined is still evolving with many automated algorithms employed by various manufacturers, measurements applied to single leads, some, or all of the 12 ECG leads, superimposition of many QRS complexes of the same lead, utilization of a single or a "representative" lead, and thus it has not been concluded that a particular method or approach can be considered as the "gold standard" [2].
There is great variation in the approach employed in the calculation of QRSd by different manufacturers of ECG equipment. Automated interpretation ECG programmes rely on different algorithms for the calculation of QRSd, such as the electrocardiograph manufactured by Hewlett Packard (acquired by Philips) which employs the sum of the longest 5 of the 12 QRSd measurements from the individual leads divided by 5. Marquette (acquired by General Electric) uses all 12 leads for the QRSd; their method is based on the superimposition of a set of "representative" beats of all 12 ECG leads, each of which is a "median" of many complexes of the individual lead. The QRSd is calculated from the onset of the lead with the earlier onset to the offset of the lead with the latest offset; they have found in their earlier work that the QRSd calculated with this algorithm correlates well with the vectorcardiography-calculated QRSd. Mortara Instruments employs all 12 leads to determine the QRSd. The first process determines the dominant beat's morphology; once this is established, an average beat is calculated and on this resulting beat (the average of the dominant beats) the measurement of QRSd is carried out.
Experience in the general medical wards, or the Coronary Care Unit provides the opportunity to evaluate serial ECGs, and this has taught us that the QRSd is not fixed, but a difference by a few ms (e.g. 5–10 ms) (as indicated in the automated ECG interpretation report) is commonplace even in patients with normal QRSd. For serial measurements in patients with QRSd <100 ms, and at paper speed of 25 mm/s, it is not feasible to rely on manual measurements, and thus all these comments refer to automated measurements [1]. Changes in QRSd are also encountered in the office or cardiac clinic, where ECGs are obtained at longer time intervals. Such differences in the QRSd may be due to physiological reasons, or machine- or manual-based measurement variability, and occur even when serial ECGs are obtained sequentially a few minutes apart, without removing and reapplying the chest leads [2]. Also, it is not meaningful to compare measurements of QRSd from different studies, unless identical measurement methods have been used [2]. Another factor influencing the mean QRSd is the variability of the precordial leads due to employment of different thoracic landmarks [4] in serial ECGs; since the ECG algorithm electronically calculates the mean QRSd, based on QRSd measurements of the individual 12 leads [1], the mean QRSd in serial ECGs is expected to be variable. Larger differences in QRSd are seen in serial ECGs of patients with change of incomplete to complete BBB or IVCD. When such shifts are associated with alterations in QRS complexes diagnostic of BBB, the explanation is obvious; however, often only changes in the QRSd are documented with unchanged morphologies. Not much work has been focused on differentiating these two phenomena.
In general, when clinicians care for their patients they rely on a "hybrid" approach for the QRSd information, i.e. automated measurements, and their overall visual evaluation of the ECG errors can be anticipated. Assessment of the QRSd is practically based on measurements included on the ECG hard-copy report, although a visual estimation or even a measurement of a few of the wider leads is advisable. If one considers the difference in the detection resolution by ECG machines ({small tilde}5 µV) and what a group of cardiologists could recognize on enlarged conventional ECG recordings ({small tilde}20 mV) [2], reliance on the automated QRSd measurements is acceptable. Small-amplitude (<20 µV) waves at the onset and offset of QRS complexes may be difficult to characterize whether they represent part of the QRS complex or are due to noise; consistency in many complexes from the same lead by using the majority rule, and across different leads should be employed [2]. As a rule there are but a few complexes (2–5, depending on the heart rate) in the standard ECG, and thus checking the intra-lead consistency is at best limited. Relying on automation for QRSd is inevitable, is currently practised by most clinicians, is being used in research, and will be the rule in the future.
An observation was recently reported on the shortening of QRSd in patients with HF undergoing treatment; such changes were not seen in patients without HF [24]. It is important that this shortening of QRSd changed the qualification of certain patients who were previously deemed to be candidates for CRT, on account among other parameters of a QRSd >130 ms, but who did not qualify after successful therapy of HF. The mechanism of this shortening of QRSd in patients treated for HF has not been elucidated, but intuitively improved ventricular function and/or decrease in ventricular volumes with therapy could be expected to account for this. This speculation is based on assumption that reduction of heart volumes consequent upon effective management of HF shortens the length of the depolarization pathway, thus leading to shortening of QRSd. The noted shortening of QRSd is plausible, since it is the inverse of what has been traditionally believed, and recently confirmed [15], for patients clinically deteriorating, whose ECGs showed increasing QRSd. The issue becomes even more complex mechanistically when considering the effect of a reduced intraventricular blood volume in patients effectively treated for HF and its attenuating influence on the voltage of surface QRS complexes ("Brody effect") [28,29]. Speculation is justified that reduction of the QRS amplitude would be associated with shortening of QRS width, as noted before in patients with peripheral oedema (PERO) [30] (vide infra). The "Brody effect" refers to the effect on the transmission only of the heart's potentials to the body surface, and does not invoke changes of the cardiac muscle per se. However, it is possible that HF with PERO is also associated with an oedematous myocardial state that leads to decreased conduction and resultant conduction defects (leading to increased QRSd), which resolves once myocardial oedema is reduced.
Myocardial oedema has not been assessed in animal models or patients with PERO by histopathological studies or by diagnostic techniques, e.g. magnetic resonance imaging. Indirect evidence that the changes in patients with PERO are due to alterations in the transfer of cardiac potentials to the body surface (external to the heart) is that intracardiac electrograms remained stable after resolution of PERO, while the voltage of the standard ECG was markedly augmented [31]. The phenomenon of QRSd variability over time in patients managed for HF must be complex, since underlying ischaemia, intervening myocardial infarction (possibly silent), metabolic disturbances, and drugs with variable effects on the QRSd conspire to produce an environment of uncertainty when a causative mechanism is sought [24].
In an effort to unravel this puzzle it is important first to differentiate between changes in the QRSd with and without a shift to and from IVCD or BBB. There are no details in the above-cited retrospective report [24] about the types of IVCD that these patients showed in their ECGs. Also there are no data on whether the HF was manifest by dyspnoea and fatigue, or by significant PERO, and in such a case the extent of weight reduction with therapy.
The opposite to the above observation [24] has been recently described, that of a decrease in QRSd in patients developing PERO in the context of a variety of illnesses [30,31] (Fig. 1); such shortening of QRSd correlated with the corresponding decrease in the amplitude of QRS complexes, expressed as the sum of the highest peak to nadir of QRS of all 12 ECG leads [30,31]. When patients with PERO subsequently lost weight, their QRS complexes "recovered" a significant portion of their lost amplitude and the corresponding QRSd also increased (Fig. 1). A limited experience with patients with DCM treated for HF [32,33] did not reveal conclusively an increase in the QRSd commensurate with increase in the amplitude of QRS complexes, following marked weight loss resulting from intense diuresis. The ECGs of these patients showed IVCD, or BBB. Response of QRSd in patients with IVCD, BBB, and normal QRSd treated for HF needs to be investigated.
|
QRS), and 12-lead ECGs obtained at admission (A), at peak weight (B), and at subsequent lowest weight (C), respectively. The corresponding QRSd at A, B, and C were 83, 58, and 75 ms. Reprinted here with the permission of the American College of Cardiology (Ref. [31]The decreases/increases in the QRSd seen in patients with corresponding worsening/improving oedematous states probably do not denote alterations in the conduction times (i.e. a real physiological change), but an apparent change (i.e. a measurement effect) related to the overall (not only peak) attenuation/augmentation of the amplitude of QRS complexes. Accordingly, portions at the onset and offset of QRS complexes change in amplitude and are under-represented/over-represented in the electrocardiograph's measurement algorithm, operating with a fixed resolution for signal detection [30]
.
These changes in QRSd also affect patients with changing oedematous states in the context of haemodialysis for chronic renal failure; during 26 sessions of haemodialysis a statistically significant mean increase of {small tilde}5 ms in the QRSd was noted in one patient, as assessed from automated measurements [34]. Thus, it appears that this experience [30–34] contradicts that discussed earlier [24], according to which management of HF may effect a decrease in QRSd. It is conceivable that both these opposing mechanisms (one lengthening and the other shortening the QRSd) operate simultaneously to influence this measurement in patients undergoing therapy for HF. Thus, the diverse values of the QRSd after successful therapy may reflect the varying contribution of these two mechanisms. Along this line of thought it would have been contributory to evaluate the extent of weight loss in response to therapy for HF in the patients who showed a decrease in the QRSd and the ones who did not in the work referred to above [24].
What is then to be done for future clarification of this matter and current practice application? (1) Obviously more research is needed on the natural course of QRSd in patients with DCM and HF, focusing on patients improving as a result of therapy, the ones who remain stable, and the others who deteriorate in spite of treatment. (2) When referring to changes in QRSd from comparison of 
2 ECGs, the morphology of the QRS complexes need to be included so that it becomes clear in the outset whether the change in the QRSd is due to plain shortening/widening of the QRS or change to/from normal/abnormal intraventricular conduction (IVCD, or BBB); these details may provide insight as to the pathophysiological mechanisms underlying the change in QRSd. (3) A reference to the changes in the amplitudes of QRS complexes should also be made when evaluating alterations in QRSd. (4) A description of the state of the patient(s) with DCM and HF regarding the presence/absence of PERO and weight(s) should always be considered with the corresponding ECG(s).
Thus it appears that the QRSd is a dynamic component of the ECG. In using it in the context of merely following our patients over time, or referring them for new therapies (ICD, CRT), we should evaluate it along the lines stated above. All these factors are necessary for accurate evaluation of changes in the QRSd. Since reference to cut-off points in QRSd (>120, or >130 ms) is often made when recommending ICD or CRT, it is prudent to consider the QRSd of the patient under treatment, after the state of maximal alleviation of PERO and/or the lowest possible weight has been achieved. Accordingly, nephrologists often refer to this state as "dry weight", and a certain number, specific for each patient undergoing haemodialysis, is employed. Finally, it is exciting that the QRSd, a component in the ECG interpretation heretofore attracting casual attention, has been thrust to centre stage among the other clinical data [25–27] employed in management decisions of patients with DCM and HF. We should all take the challenge to advance further its utility.
References
[1] Hewlett-Packard Interpretive Cardiograph. Physician's Guide. HP Part No. M1700-92908, Edition 4, Printed in USA September 1994. p. 7-1, C-1.
[2] Macfarlane PW and Vetch Lawrie TD. Comprehensive electrocardiology 1989; New York Pergamon pp. 140 417, 440, 450, 1147, 1442, 1603, 1651.
[3] Mirvis DM. Electrocardiography. A physiologic approach 1993; St Louis Mosby pp. 138 283, 304.
[4] Wagner GS. Marriott's practical electrocardiography 1994; 9th ed Baltimore Williams and Wilkins pp. 15 41, 42.
[5] In Dunn MI and Lipman BS (Eds.). Lipman-Massie clinical electrocardiography 1984; 7th ed Chicago Year Book Medical Publishers pp. 146.
[6] Silverman ME, Pressel MD, Brackett JC, et al. Prognostic value of the signal-averaged electrocardiogram and a prolonged QRS in ischemic and nonischemic cardiomyopathy. Am J Cardiol 1995; 75: 460–464.[CrossRef][Web of Science][Medline]
[7] Bristow MR, Feldman AM, Saxon LA. for the COMPANION Steering Committee and Companion Clinical Investigators. Heart failure management using implantable devices for ventricular resynchronization: Comparison of medical therapy, pacing, and defibrillation in chronic heart failure (COMPANION) trial. J Cardiac Fail 2000; 6: 276–285.[CrossRef][Web of Science][Medline]
[8] Kass DA, Chen C-H, Curry C, et al. Improved left ventricular mechanics from acute VDD pacing in patients with dilated cardiomyopathy and ventricular conduction delay. Circulation 1999; 99: 1567–1573.
[9] Saxon LA, Kumar UN, De Marco T. Heart failure and cardiac resynchronization therapies: U.S. experience in the year 2000. Ann Noninv Electrocardiol 2000; 5: 188–194.[CrossRef]
[10] Stellbrink C, Auricchio A, Diem B, et al. Potential benefit of biventricular pacing in patients with congestive heart failure and ventricular tachyarrhythmia [Abstract]. Am J Cardiol 1999; 83: 143D–150D.[Web of Science][Medline]
[11] Auricchio A, Stellbrink C, Sack S, et al. The Pacing Therapies for Congestive Heart Failure (PATH-CHF) study: rationale, design and endpoints of a prospective randomized multicenter study [Abstract]. Am J Cardiol 1999; 83: 130D–135D.[CrossRef][Web of Science][Medline]
[12] Auricchio A, Klein H, Tockman B, et al. Transvenous biventricular pacing for heart failure: Can the obstacles be overcome? [Abstract]. Am J Cardiol 1999; 83: 136D–142D.[CrossRef][Web of Science][Medline]
[13] Shamim W, Francis DP, Yousuffuddin M, et al. Intraventricular conduction delay: a prognostic marker in chronic heart failure. Int J Cardiol 1999; 70: 171–178.[CrossRef][Web of Science][Medline]
[14] Aranda JM, Schofield RS, Leach D, et al. Ventricular dyssynchrony in dilated cardiomyopathy: the role of biventricular pacing in the treatment of congestive heart failure. Clin Cardiol 2002; 25: 357–362.[Web of Science][Medline]
[15] Shamim W, Yousufuddin M, Cicoria M, et al. Incremental changes in QRS duration in serial ECGs over time identify high risk elderly patients with heart failure. Heart 2002; 88: 47–51.
[16] Moss A, Zareba W, Hall WJ, et al. Prophylactic implantation of a defibrillator in patients with myocardial infarction and reduced ejection fraction. N Engl J Med 2002; 346: 877–883.
[17] Gregoratos G, Abrams J, Epstein AE, et al. ACC/AHA/NASPE2002 Guideline update for implantation of cardiac pacemakers and antiarrhythmia devices: summary article. A report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (ACC/AHA/NASPE Committee to Update the 1998 Pacemaker Guidelines). Circulation 2002; 106: 2145–2161.
[18] Romhilt DW, Bove KE, Norris RJ, et al. A critical appraisal of the electrocardiographic criteria for the diagnosis of left ventricular hypertrophy. Circulation 1969; 40: 185–195.
[19] Pitt B, Zannad F, Remme WJ, et al. The effect of spironolactone on morbidity and mortality in patients with severe heart failure. Randomized Aldactone Evaluation Study investigators. N Engl J Med 1999; 341: 709–717.
[20] St John Sutton MG, Plappert T, Abraham WT, et al. Effect of cardiac resynchronization therapy on left ventricular size and function in chronic heart failure. Circulation 2003; 107: 1985–1990.
[21] Hunt SA, Baker DW, Chin MH, et al. ACC/AHA guidelines for the evaluation and management of chronic heart failure in the adult: executive summary. A report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Committee to Revise the 1995 Guidelines for the Evaluation and Management of Heart Failure) developed in collaboration with the International Society for Heart and Lung Transplantation endorsed by the Heart Failure Society of America. J Am Coll Cardiol 2001; 38: 2101–2113.
[22] Cazeau S, Leclercq C, Lavergne T, et al. Multisite Stimulation in Cardiomyopathies (MUSTIC) Study Investigators. Effects of multisite biventricular pacing in patients with heart failure and intraventricular conduction delay. N Engl J Med 2001; 344: 873–880.
[23] Abraham WT, Fisher WG, Smith AL, et al. for MIRACLE study group. Multicenter InSync Randomized Clinical Evaluation. Cardiac resynchronization in chronic heart failure. N Engl J Med 2002; 346: 1845–1853.
[24] Aranda JM, Carlson ER, Pauly DF, et al. QRS duration variability in patients with heart failure. Am J Cardiol 2002; 90: 335–337.[CrossRef][Web of Science][Medline]
[25] Notabartolo D, Merlino JD, Smith AL, et al. Usefulness of the peak velocity difference by tissue Doppler imaging technique as an effective predictor of response to cardiac resynchronization therapy. Am J Cardiol 2004; 94: 817–820.[CrossRef][Web of Science][Medline]
[26] Philippon F. Cardiac resynchronization therapy: device-based medicine for heart failure. J Card Surg 2004; 19: 270–274.[CrossRef][Web of Science][Medline]
[27] Sade LE, Kanzaki H, Severyn D, et al. Quantification of radial mechanical dyssynchrony in patients with left bundle branch block and idiopathic dilated cardiomyopathy without conduction delay by tissue displacement imaging. Am J Cardiol 2004; 94: 514–518.[CrossRef][Web of Science][Medline]
[28] Brody DA. A theoretical analysis of intracavitary blood mass influence on the electrocardiogram. Circ Res 1956; 4: 731–738.
[29] Rudy Y and Plonsey R. Comments on the effects of variations in the size of the heart on the magnitude of ECG potentials. J Electrocardiol 1980; 13: 79–82.[CrossRef][Web of Science][Medline]
[30] Madias JE. Significance of shortening of the mean QRS duration of the standard electrocardiogram in patients developing peripheral edema. Am J Cardiol 2002; 89: 1444–1446.[CrossRef][Web of Science][Medline]
[31] Madias JE, Bazaz R, Agarwal H, et al. Anasarca-mediated attenuation of the amplitude of ECG complexes: a description of a heretofore unrecognized phenomenon. J Am Coll Cardiol 2001; 38: 756–764.
[32] Madias JE, Agarwal H, Win M, et al. Effect of weight loss in congestive heart failure from idiopathic dilated cardiomyopathy on electrocardiographic QRS voltage. Am J Cardiol 2002; 89: 86–88.[CrossRef][Web of Science][Medline]
[33] Madias JE, Attanti S, Narayan V. Relationship among electrocardiographic potential amplitude, weight, and resistance/reactance/impedance in a patient with peripheral edema treated for congestive heart failure. J Electrocardiol 2003; 36: 167–171.[Web of Science][Medline]
[34] Madias JE and Narayan V. Augmentation of the amplitude of electrocardiographic QRS complexes immediately following hemodialysis: a study of 26 hemodialysis sessions of a single patient, aided by measurements of resistance/reactance/impedance. J Electrocardiol 2003; 36: 263–271.[CrossRef][Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  M. Chen, C. Huo, and D.-j. Huang QRS Duration in Patients Hospitalized for Worsening Heart Failure JAMA, October 22, 2008; 300(16): 1879 - 1879. [Full Text] [PDF]
|
|
|
|
|
|
A. Hakeem, S. Bhatti, and Z. Samad QRS Duration in Patients Hospitalized for Worsening Heart Failure JAMA, October 22, 2008; 300(16): 1880 - 1880. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||