Introduction. CLINICAL RESEARCH Non-invasive Electrophysiology and Risk Stratification

Transcription

1 Europace (2009) 11, doi: /europace/eup001 CLINICAL RESEARCH Non-invasive Electrophysiology and Risk Stratification Accuracy of manual QRS duration assessment: its importance in patient selection for cardiac resynchronization and implantable cardioverter defibrillator therapy David R. Tomlinson*, Yaver Bashir, Timothy R. Betts, and Kim Rajappan* Department of Cardiology, John Radcliffe Hospital, Headley Way, Headington, Oxford OX3 9DU, UK Received 25 September 2008; accepted after revision 30 December 2008; online publish-ahead-of-print 26 January 2009 Aims Patients with left ventricular systolic dysfunction and electrocardiographic QRS duration (QRSd) 120 ms may obtain symptomatic and prognostic benefits from cardiac resynchronization therapy (CRT). However, clinical trials do not describe the methods used to measure QRSd. We investigated the effect of electrocardiogram (ECG) display format and paper speed on the accuracy of manual QRSd assessment and concordance of manual QRSd with computer-calculated mean and maximal QRSd.... Methods Six cardiologists undertook QRSd measurements on ECGs, with computer-calculated mean QRSd close to 120 ms. and results Display formats were 12-lead, 6-limb, and 6-precordial leads, each at 25 and 50 mm/s. When the computer-calculated mean was used to define QRSd, manual assessment demonstrated 97 and 83% concordance at categorizing QRSd as, and 120 ms, respectively. Using the computer-calculated maximal QRSd, manual assessment demonstrated 83% concordance when QRSd was,120 ms and 19% concordance when QRSd was 120 ms. The six-precordial lead format demonstrated significantly less intra and inter-observer variabilities than the 12-lead, but this did not improve concordance rates.... Conclusion Manual QRSd assessments demonstrate significant variability, and concordance with computer-calculated measurement depends on whether QRSd is defined as the mean or maximal value. Consensus is required both on the most appropriate definition of QRSd and its measurement Keywords Cardiac resynchronization therapy Implantable cardioverter defibrillator Electrocardiogram QRS duration Introduction In patients with left ventricular systolic dysfunction (LVSD), prolongation of the electrocardiographic QRS duration (QRSd) is an independent predictor of all-cause mortality 1,2 and possibly sudden cardiac death. 3 In such patients, an electrocardiographic QRS interval of.120 ms is also commonly associated with delayed ventricular activation, resulting in intra-ventricular dyssynchrony. 4 In this setting, clinical trials of cardiac resynchronization therapy (CRT) principally conducted in patients with a left bundle branch pattern of prolonged ventricular conduction have demonstrated important symptomatic and prognostic benefits. 5,6 Implantable cardioverter defibrillators (ICDs) also confer prognostic benefit as secondary prevention following aborted or resuscitated sudden cardiac death 7 9 and as primary prevention in those patients with severe LVSD but without prior life-threatening ventricular arrhythmias. 10,11 Furthermore, the three most recently published primary prevention ICD trials each demonstrate a trend towards a greater absolute benefit in patients with QRSd 120 ms These data have now been incorporated into UK national guidelines, which recommend consideration of CRT in patients with New York Heart Association (NYHA) class III heart failure despite optimal medical therapy, left ventricular ejection fraction (LVEF),35%, and QRSd * Corresponding authors. Tel: þ , Fax: þ , dave_tomlinson@doctors.org.uk (D.R.T.) or kim.rajappan@orh.nhs.uk (K.R.) Published on behalf of the European Society of Cardiology. All rights reserved. & The Author For permissions please journals.permissions@oxfordjournals.org.

2 Accuracy of manual QRSd assessment ms, 13 as well as prophylactic ICD implantation in patients with ischaemic cardiomyopathy, an LVEF of,30%, and again QRSd 120 ms. 14 However, the specific methods for QRSd measurement were not described in these publications, and international guidelines do not specify a preferred measurement technique This is of concern because when the QRSd is close to 120 ms, the decision to recommend de novo ICD implantation or undertake further risk stratification investigations is critically dependent on the ECG interpretation. If QRSd is,120 ms, this may include Holter monitoring and electrophysiological testing. 17 Thus, inaccurate QRSd classification may increase costs by exposing patients to otherwise unnecessary invasive procedures or deny some patients the beneficial effects of CRT and prophylactic ICD. Finally, the lack of a clear guideline on methods for QRSd measurement may create intra- and inter-individual inconsistencies in device prescription. Methods Study design The effect of ECG display format and paper speed on manual QRSd measurement was investigated in a selection of ECGs, each with a computer-calculated mean QRSd close to 120 ms. Chosen paper formats were standard 12-lead, 6-limb, or 6-precordial lead tracings, all separately recorded at 25 and 50 mm/s paper speeds. FX-4101U or FX-4010 ECG machines (Fukuda Denshi Co., Japan) were used in all subjects, and on line computer-extended analysis calculated both the mean and individual lead QRSd. Electrocardiograms The study ECG series comprised the following: (i) five subjects ECGs, each with a mean calculated QRSd,120 ms, but with maximal QRSd 120 ms in at least one lead; (ii) one subject s ECG with both mean and maximal calculated QRSd 120 ms, positive control ; (iii) one subject s ECG with both mean and maximal calculated QRSd,120 ms, negative control. Electrocardiogram analysis Six cardiologists undertook manual QRSd analysis. All study ECGs were anonymized, printed on standard ECG paper, and presented in a random order. Clinicians were asked to undertake measurement of the QRSd using their usual technique and also to record the lead used for final measurement. None was asked to specifically record whether the QRSd was,120 or 120 ms, nor were they aware that this particular interval was the subject of the study. ECG measurements were then classified according to this interval for study purposes, following a complete return of analysed ECGs. Intra- and inter-observer variabilities for QRSd measurement were calculated using an additional subject s ECG with mean QRSd,120 ms, but maximal calculated QRSd 120 ms; four copies in each different format were randomly distributed throughout the study ECG series. Study outcomes The primary study outcome was concordance between manually assessed and computer-calculated QRSd, with the latter defined in two ways both as the mean and maximum value. Secondary outcomes were the intra- and inter-observer variabilities of QRSd measurement and concordance in categorizing QRSd as,120 or 120 ms, according to the ECG format. None of the clinicians was aware of the primary or secondary study outcomes. Statistical analysis The primary outcome was expressed as raw concordance rates. Intraand inter-observer variabilities in absolute QRSd measurement between ECG formats were analysed using the Kruskal Wallis statistic with Dunn s multiple comparison test. Intra- and inter-observer concordance at defining QRSd as,120 or 120 ms according to the ECG format was analysed using x 2 test for trend. A P-value less than 0.05 was considered to represent statistical significance. Results Electrocardiogram characteristics Details of study ECGs with computer-calculated mean and range of QRSd are shown in Figure 1. The lead with maximal calculated QRSd was precordial leads V1 V4 in all cases. Manual QRSd measurements were undertaken with a ruler in one study participant and by directly measuring from the paper visually (without calipers) in the remaining five. Concordance between manual and calculated QRSd When the computer-calculated QRSd was defined according to the mean value and was,120 ms, manual assessment demonstrated good concordance with no significant effect of changing ECG format (Table 1). When the computer-calculated mean QRSd was 120 ms, manual assessment demonstrated lower concordance and was dependent on ECG format. Concordance for the 12-lead format at 25 mm/s was 83%; however, both six-limb and six-precordial formats had poorer concordance (0 67%), with the greatest effect evident at 50 mm/s paper speed. When the computer-calculated QRSd was defined according to the maximal value and was,120 ms, manual assessment demonstrated concordance rates of % for all ECG formats. However, when the computer-calculated maximal QRSd was 120 ms, manual assessment demonstrated concordance rates of 0 19%, and greatest concordance occurred using the 12-lead format at 25 mm/s paper speed. Intra-observer variability There was no significant difference in intra-observer variability of manual QRSd assessment between ECG formats, although there was slightly greater variability for the 12-lead, 25 mm/s format (Table 2). Furthermore, in this format, only four of six clinicians provided manual QRSd measurements, resulting in 100% concordance at categorizing ECGs as either,120 or 120 ms. In contrast, when using a six-precordial ECG format, all manual QRSd measurements had 100% concordance, regardless of the paper speed (x 2 test for trend, P ¼ 0.03). Inter-observer variability There was significantly greater inter-observer variability in measured QRSd from the 12-lead format compared with the precordial lead format at 25 mm/s paper speed (Table 3). Furthermore, the inter-observer agreement in categorizing ECGs

3 640 D.R. Tomlinson et al. Figure 1 Study electrocardiograms (ECGS) with mean calculated and ranges of QRS duration for each display format (ms, milliseconds); four copies of ECG 1 were used to determine intra- and inter-observer observer variabilities, whereas ECGs 7 and 8 were positive and negative controls, respectively. Table 1 Raw concordance between measured and calculated QRS duration (QRSd) in categorizing the QRSd as either <120 or 120 ms, when the gold standard QRSd was defined as either the mean or maximum calculated value ECG format Raw concordance between measured and calculated QRSd (%)... Calculated mean Calculated QRSd... maximum QRSd... <120 ms 120 ms <120 ms 120 ms lead 25 mm/s lead 50 mm/s Limb 25 mm/s Limb 50 mm/s V1 V6 25 mm/s V1 V6 50 mm/s Table 2 Intra-observer variability of manually assessed QRS duration ECG format Median Clinicians with intra-observer concordance for variability (range) QRSd (ms) categorization: n (%) lead 25 mm/s 25 (10 50) 4 (67)* 12-lead 50 mm/s 10 (0 25) 5 (83) Limb 25 mm/s 20 (5 40) 5 (83) Limb 50 mm/s 20 (10 30) 6 (100) V1 V6 25 mm/s 20 (0 25) 6 (100) V1 V6 50 mm/s 12.5 (10 35) 6 (100) The number and percentage of clinicians with manual assessments resulting in 100% concordance for categorizing QRSd as either,120 or 120 ms are also shown, according to the ECG format. *P ¼ 0.03, for trend. according to the QRSd was also dependent on the format, with a significant trend observed against a 12-lead, 25 mm/s display. Overall, complete inter-observer concordance for QRSd categorization as,120 or 120 ms was only achieved in two of four ECGs in a 12-lead format compared with four of four ECGs using the six-precordial lead format at 25 mm/s paper speed. Variation in choice of electrocardiogram leads for manual QRSd assessment For analysis of ECGs in a 12-lead format, no study participant consistently measured QRSd from the same lead; however, the majority of measurements were taken from leads V1 V4. Using the four identical ECGs from which intra- and inter-observer variabilities were determined and at 25 and 50 mm/s paper speed,

4 Accuracy of manual QRSd assessment 641 Table 3 Inter-observer variability of manually assessed QRS duration ECG format Median inter-observer QRSd assessments: n (%) Number of ECGs with concordance variability (range) (ms)... for QRSd categorization: n (%) <120 ms 120 ms lead 25 mm/s 35 (20 50) 21 (88) 3 (12) 2 (50) 12-lead 50 mm/s 22.5 (20 40) 23 (96) 1 (4) 3 (75) Limb 25 mm/s 35 (20 40) 23 (96) 1 (4) 3 (75) Limb 50 mm/s 40 (30 40) 24 (100) 0 4 (100) V1 V6 25 mm/s 12.5 (10 20)* 24 (100) 0 4 (100) V1 V6 50 mm/s 12.5 (0 30) 24 (100) 0 4 (100) Inter-observer variation in QRSd categorization as,120 or 120 ms is also shown, according to the ECG format (six observers each analysed four identical ECGs in each format, hence 24 observations were made). *P ¼ 0.04, inter-observer variability V1 V6 compared with 12-lead, at 25 mm/s. P ¼ 0.02, for trend. QRSd was assessed from one of leads V1 V4 in 58 and 75% of measurements, a limb lead in 29 and 21%, and lead V5 or V6 in 13 and 4%, respectively. Notably, inter-observer variability in the choice of ECG leads for QRSd assessment resulted in no instances where all participants used the same lead for any individual ECG. Discussion These data demonstrate important differences in the concordance rates between manual and computer-calculated QRSd, depending on whether QRSd was defined according to the mean or maximal computer-calculated value. Manual assessment showed good concordance with computer-calculated measurements (.90%) when the QRSd was,120 ms. However, when the computer-calculated QRSd was 120 ms, manual assessment demonstrated lower concordance, particularly when QRSd was defined according to the maximal value. Furthermore, there was also significant intra- and inter-observer variabilities in absolute QRSd measurement. This finding may partly be explained by the variation in the choice of ECG lead used for manual assessments, as changing to a six-precordial lead format improved both intra- and inter-observer variabilities and concordance rates. In this regard, it is worth noting the significant variation in the computer-calculated QRSd measurements between individual ECG leads. Although the mean QRSd was,120 ms, the computer-calculated QRSd was 120 ms in at least one lead (usually from leads V1 V4). Thus, manual assessment of QRSd from a standard 12-lead ECG format may be subject to significant variability, simply due to a lack of standardization of the approach to measuring this interval. Clinical implications The findings of the present study are of great relevance to current guidelines on patient selection for both CRT and ICD. Recent published data on CRT for patients with narrow QRS (,120 ms) have reported conflicting results. Some case series have demonstrated improvement in both LV dimensions and symptoms in unselected patients 18 or those selected on the basis of echocardiographic dyssynchrony. 19,20 However, in the RethinQ study, 21 subgroup analysis of patients with narrow QRS demonstrated no difference between controls and those receiving CRT in the hard endpoints of peak oxygen consumption, LVEF, and dimensions. Consequently, present guidelines define an ECG QRS interval of 120 ms as the lower cut-off criterion for undertaking echocardiographic dyssynchrony assessments. In those patients with a QRSd between 120 and 150 ms demonstrating dyssynchrony, significant LVSD, and with NYHA III/IV symptoms, CRT implantation is recommended. 13,16,22 For prophylactic ICD implantation, ACC/ AHA/ESC guidelines do not specify an ECG QRS interval. 23 However, UK National Institute for Health and Clinical Excellence (NICE) guidelines for patients with ischaemic heart disease reserve ICD for those patients with LVEF,30% and QRS duration 120 ms. 14 Notably, no guideline explicitly recommends a specific method for QRSd assessment. Importantly, clinical trials of both the ICD and CRT have utilized entry criteria of a measured QRS interval 5 or any two leads of a 12-lead ECG with a QRSd 120 ms. 6 Although there is evidence for QRSd assessment to be based on the maximal QRSd in one or two leads, this was precisely the area where manual assessment demonstrated poorest concordance with the computer-calculated QRSd. In fact, apart from the negative control, all ECGs in this present study were eligible for entry into either COMPANION 6 or CARE-HF; 5 however, most manual assessments resulted in an ECG QRSd classification of,120 ms. Study limitations The calculated QRSd was assessed using a single ECG machine manufacturer (Fukuda Denshi Co.); so possible differences in QRSd calculations between different manufacturers algorithms were not investigated. This issue is worthy of further study, as one possible recommendation from this present study is that all patients with a QRSd close to 120 ms and under consideration for CRT should undergo computer-calculated ECG QRSd assessment. However, this would only be appropriate if there were no important inter-manufacturer differences in calculated QRSd measurements.

5 642 D.R. Tomlinson et al. Conclusions Manual QRSd assessments demonstrate significant variability, and concordance with computer-calculated measurement depends on whether QRSd is defined as the mean or maximal value. We propose that guidelines should be updated to reflect a consensus view on how best to measure QRSd (manual vs. computercalculated) and whether to use the mean or maximal QRSd to better standardize decision-making in cardiac rhythm management device implantation. Without this initiative, our data suggest that based on inaccurate and inconsistent manual QRSd measurements patients may be denied device therapy for which they may otherwise have been eligible, potentially resulting in adverse outcomes on both morbidity and mortality. Acknowledgements We thank Lyn Barrett, Cardiac Physiologist at the ECG Department, John Radcliffe Hospital, Oxford, UK, for performing the ECGs used in this study. Conflict of interest: none declared. References 1. Shenkman HJ, Pampati V, Khandelwal AK, McKinnon J, Nori D, Kaatz S et al. Congestive heart failure and QRS duration: establishing prognosis study. Chest 2002; 122: Kalra PR, Sharma R, Shamim W, Doehner W, Wensel R, Bolger AP et al. Clinical characteristics and survival of patients with chronic heart failure and prolonged QRS duration. Int J Cardiol 2002;86: Iuliano S, Fisher SG, Karasik PE, Fletcher RD, Singh SN. QRS duration and mortality in patients with congestive heart failure. Am Heart J 2002;143: Cazeau S, Leclercq C, Lavergne T, Walker S, Varma C, Linde C et al. Effects of multisite biventricular pacing in patients with heart failure and intraventricular conduction delay. N Engl J Med 2001;344: Cleland JG, Daubert JC, Erdmann E, Freemantle N, Gras D, Kappenberger L et al. The effect of cardiac resynchronization on morbidity and mortality in heart failure. N Engl J Med 2005;352: Bristow MR, Saxon LA, Boehmer J, Krueger S, Kass DA, De Marco T et al. Cardiac-resynchronization therapy with or without an implantable defibrillator in advanced chronic heart failure. N Engl J Med 2004;350: The Antiarrhythmics Versus Implantable Defibrillators (AVID) Investigators. A comparison of antiarrhythmic-drug therapy with implantable defibrillators in patients resuscitated from near-fatal ventricular arrhythmias. N Engl J Med 1997; 337: Kuck KH, Cappato R, Siebels J, Ruppel R. Randomized comparison of antiarrhythmic drug therapy with implantable defibrillators in patients resuscitated from cardiac arrest: the Cardiac Arrest Study Hamburg (CASH). Circulation 2000;102: Connolly SJ, Gent M, Roberts RS, Dorian P, Roy D, Sheldon RS et al. Canadian Implantable Defibrillator Study (CIDS): a randomized trial of the implantable cardioverter defibrillator against amiodarone. Circulation 2000;101: Moss AJ, Zareba W, Hall WJ, Klein H, Wilber DJ, Cannom DS et al. Prophylactic implantation of a defibrillator in patients with myocardial infarction and reduced ejection fraction. N Engl J Med 2002;346: Bardy GH, Lee KL, Mark DB, Poole JE, Packer DL, Boineau R et al. Amiodarone or an implantable cardioverter-defibrillator for congestive heart failure. N Engl J Med 2005;352: Kadish A, Dyer A, Daubert JP, Quigg R, Estes NA, Anderson KP et al. Prophylactic defibrillator implantation in patients with nonischemic dilated cardiomyopathy. N Engl J Med 2004;350: Cardiac resynchronisation therapy for the treatment of heart failure: NICE Technology appraisal guidance (2007). 14. Implantable Cardioverter Defibrillators for Arrhythmias: NICE Technology appraisal guidance (January 2006). 15. Radford MJ, Arnold JM, Bennett SJ, Cinquegrani MP, Cleland JG, Havranek EP et al. ACC/AHA key data elements definitions for measuring the clinical management outcomes of patients with chronic heart failure: a report of the American College of Cardiology/American Heart Association Task Force on Clinical Data Standards (Writing Committee to Develop Heart Failure Clinical Data Standards): developed in collaboration with the American College of Chest Physicians the International Society for Heart Lung Transplantation: endorsed by the Heart Failure Society of America. Circulation 2005;112: Epstein AE, DiMarco JP, Ellenbogen KA, Estes NA III, Freedman RA, Gettes LS et al. ACC/AHA/HRS 2008 Guidelines for device-based therapy of cardiac rhythm abnormalities: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Writing Committee to Revise the ACC/AHA/NASPE 2002 Guideline Update for Implantation of Cardiac Pacemakers and Antiarrhythmia Devices): developed in collaboration with the American Association for Thoracic Surgery and Society of Thoracic Surgeons. Circulation 2008;117:e Moss AJ, Hall WJ, Cannom DS, Daubert JP, Higgins SL, Klein H et al. Improved survival with an implanted defibrillator in patients with coronary disease at high risk for ventricular arrhythmia. Multicenter Automatic Defibrillator Implantation Trial Investigators. N Engl J Med 1996;335: Gasparini M, Regoli F, Galimberti P, Ceriotti C, Bonadies M, Mangiavacchi M et al. Three years of cardiac resynchronization therapy: could superior benefits be obtained in patients with heart failure and narrow QRS? Pacing Clin Electrophysiol 2007;30:S Bleeker GB, Holman ER, Steendijk P, Boersma E, van der Wall EE, Schalij MJ et al. Cardiac resynchronization therapy in patients with a narrow QRS complex. JAm Coll Cardiol 2006;48: Yu CM, Chan YS, Zhang Q, Yip GW, Chan CK, Kum LC et al. Benefits of cardiac resynchronization therapy for heart failure patients with narrow QRS complexes and coexisting systolic asynchrony by echocardiography. J Am Coll Cardiol 2006;48: Beshai JF, Grimm RA, Nagueh SF, Baker JH III, Beau SL, Greenberg SM et al. Cardiac-resynchronization therapy in heart failure with narrow QRS complexes. N Engl J Med 2007;357: Hunt SA, Abraham WT, Chin MH, Feldman AM, Francis GS, Ganiats TG et al. ACC/AHA 2005 Guideline Update for the Diagnosis and Management of Chronic Heart Failure in the Adult: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Writing Committee to Update the 2001 Guidelines for the Evaluation and Management of Heart Failure): developed in collaboration with the American College of Chest Physicians and the International Society for Heart and Lung Transplantation: endorsed by the Heart Rhythm Society. Circulation 2005;112: e Zipes DP, Camm AJ, Borggrefe M, Buxton AE, Chaitman B, Fromer M et al. ACC/ AHA/ESC 2006 Guidelines for management of patients with ventricular arrhythmias and the prevention of sudden cardiac death: a report of the American College of Cardiology/American Heart Association Task Force and the European Society of Cardiology Committee for Practice Guidelines (writing committee to develop guidelines for management of patients with ventricular arrhythmias and the prevention of sudden cardiac death): developed in collaboration with the European Heart Rhythm Association and the Heart Rhythm Society. Circulation 2006;114:e