Despite the introduction of angiotensin-converting

doi:./j.1-.00.000.x 1 1 1 1 1 0 1 0 1 O RIGINAL A RTICLE Sudden Cardiac Arrest in Patients With Severe Nonischemic Heart Failure: Risk Stratification With the Heart Failure Survival Sce Despite the introduction of angiotensin-converting enzyme inhibits, b-blockers, and spironolactone, a substantial number of congestive heart failure (CHF) patients die, many of them suddenly, even after clinical improvement. 1 A shift to sudden death rather than death from pump failure is observed under optimized medical therapy. Sudden cardiac arrests (SCAs) account f me than half of fatal events in stabilized heart transplant candidates. Thus, the control of sudden death in this population becomes increasingly imptant. Although it has been shown to be effective in patients with less advanced disease (New Yk Heart Association [NYHA] stages II III), prophylactic implantation of implantable cardioverter-defibrillats (ICDs) is restricted to select patients due to economic limitations in many countries. Therefe, patients who are most in need of a device have to be identified. Furtherme, in patients with nonischemic cardiomyopathy, there is controversy over the predominant mode of death and whether an ICD would be beneficial in these patients., Risk stratification f SCA is traditionally based on results from many single measures including (but not exclusively) conary angiography, exercise testing, NYHA functional class, left ventricular ejection fraction (LVEF), arrhythmias in longterm ambulaty recdings, signal-averaged electrocardiography, QT-interval duration, dispersion, T-wave alternans, heart rate variability, bareflex sensitivity, programmed electrical stimulation, and brain-natriuretic peptide levels. However, no single parameter is generally useful in identifying patients at very high risk f sudden cardiac death. Furtherme, there are me data available The auths wanted to investigate whether the heart failure survival sce (HFSS) is useful f sudden cardiac arrest (SCA) risk stratification and whether there is a difference with regard to congestive heart failure (CHF) etiology. Data of 1 severe CHF patients (New Yk Heart Association III IV % %, mean age years, mean left ventricular ejection fraction %%) during the period from 1 00 were analyzed. The HFSS was calculated using the Aaronson fmula. SCA-free survival was calculated using Kaplan Meier analysis accding to the etiology of CHF: conary artery disease (CAD) (n=1) vs no CAD (n=). In regard to SCA risk the auths found no significant overall difference between patients with without CAD. Non-CAD HFSS high-risk patients had a significantly higher risk f SCA than all the other groups (P<.001). The HFSS is useful f the detection of a subgroup of non-cad severe CHF patients at a very high risk f SCA. Congest Heart Fail. 00. Ó00 Le Jacq Maria A. Castel, MD; 1 Herbert Nägele, MD From the Thax Institute, Hospital Clinic, Villarroel, Barcelona, Spain; 1 and St Adolfstift, Medical Clinic, Reinbek, Hamburg, Germany Address f crespondence: Maria A. Castel, MD, Thax Institute, Hospital Clinic, Villarroel 10, 00 Barcelona, Spain E-mail: acastel@gmx.de Manuscript received April, 00; revised October 1, 00; accepted November, 00 f patients with conary artery disease (CAD) than f non-cad patients. The heart failure survival sce (HFSS), by using a combination of common clinical variables, provides a me robust risk stratification tool f total mtality in the general population of heart failure patients compared with other measures such as isolated peak oxygen consumption (VO ). It gives a robust measure of prognosis and stratification f heart transplant in patients with and without b-blockers and in men and women. 1 A European study showed that this measure is very useful f the detection of patients who might profit from heart transplant. Therefe, in this present study we investigate whether the HFSS is useful f risk stratification of SCA events in a population of heart transplant candidates with regard to different CHF etiologies. C H F 0 B Methods We analyzed data from 1 severe CHF patients (NYHA III IV % %, mean age years, mean LVEF %%) submitted f evaluation of heart transplant during the period 1 00. The mean observation time was..0 years. The clinical characteristics of the patient population are listed in the Table. The presence of CAD was defined as the occurrence of at least 0% stenosis of at least one maj conary artery. The assessments included physical examination, serum chemistries, and the parameters of the HFSS. Peak VO was determined during maximal treadmill exercise using a modified Naughton protocol and a metabolic cart (Metalyzer II; Ctex, Dresden, Germany). If a patient could not exercise, the resting VO value (Fick method) of the baseline right heart 1 November December 00 1 Dispatch:.1.0 Journal: CHF CE: Balaji Prasad Journal Name Manuscript No. Auth Received: No. of pages: PE: Venkatesh

1 1 1 1 0 1 0 1 Table. Patients Characteristics Stratified by Etiology and Heart Failure Survival Sce CAD NON-CAD HIGH RISK (.1) MEDIUM RISK (.0.0) LOW RISK (.) HIGH RISK (.1) MEDIUM RISK (.0.0) LOW RISK (.) No. 1 Men, % 1 1 0 Mean age, y....1 Body mass index, kg m ( ) ( ) ( ) ( ) ( ) ( ) Peak oxygen consumption, ( ) ( ) (1 0) ( ) ( 1) (1 1) ml kg min a Ejection fraction, % a 1 ( 0) (0 0) (0 ) 0 (1 0) (1 ) ( ) Heart rate, beats min a 0 ( ) (0 ) ( 0) 1 (0 ) 0 ( ) ( ) Mean blood pressure, mm Hg a 0 ( ) (0 ). ( ) 0 ( ) 0 ( ) 0 (0 1) Sodium, meq L a 1 ( 1) 1 (1 1) 1 (1 1) (1 1) 1 (1 1) 1 (1 1) QRS, ms a 1 (0 1) 1 (0 ) 0 (0 1) 1 (0 10) 1 (0 1) (0 1) Abbreviation: CAD, conary artery disease. Nonparametric data are presented as medians (intraquartile range). a Same etiology P<.001 between risk groups. catheterization was used (in 1% of cases). The LVEF was determined using echocardiography. The mean arterial blood pressure was obtained invasively by cannulation of radial arteries during routine right heart catheterization f the determination of pulmonary and peripheral vascular resistance. These procedures were perfmed after intake of chronic heart failure medication. The HFSS was derived f each patient from clinical parameters: the presence of CAD, heart rate, LVEF, mean arterial blood pressure, QRS duration > ms, peak VO, and serum sodium concentration. Dichotomous variables were coded as 1 (present) 0 (absent). Each value f the continuous and dichotomous variables was multiplied by a model coefficient, derived from a proptional hazards model. The products were summed, and the absolute value was taken as the HFSS accding to the following equation: 1 ((0.CAD)+(0.0heart rate)+ ()0.0LVEF)+()0.0mean arterial pressure)+(0.qrs >)+ ()0.0peakVO )+()0.0sodium). F the HFSS, risk strata were defined as low-risk HFSS.1, medium-risk HFSS..0, high-risk HFSS <. using previously described cutoffs. SCA-free survival was calculated with Kaplan Meier analysis accding to the underlying etiology of CHF (ischemic [n=1] vs nonischemic [n=]). Patients with defibrillats already implanted and a class I indication f an ICD at that time (eg, previous documented tachyarrhythmic cardiac arrest) were excluded from the analysis. Also, patients who had to be immediately hospitalized after first evaluation f high-urgent heart transplant were excluded from analysis. The medical therapy of the patients was individually tailed. Only patients who were not improving were processed to heart transplant. A total of % of patients received a heart transplant in long-term follow-up. Heart transplant was a censing event. Patients were either followed prospectively in our outpatient clinic interviewed by regular phone calls; reasons f death events were noted and classified as sudden nonsudden. SCA 1 November December 00

1 1 1 1 0 1 0 1 was defined as death (witnessed unwitnessed) a successful resuscitation within 1 hour of symptoms and documentation of a ventricular tachyarrhythmia without premonity symptoms during a clinically stable period. Death after a period of deteriation with signs and symptoms of CHF despite maximal medical treatment was classified as resulting from progressive CHF. With regard to the underlying arrhythmia, additional infmation was sought through electrocardiography interrogation of pacemaker stage. In case of SCA at home, relatives were asked about the clinical status of the patient in the days befe the event. We used SCA plus sudden death as a combined end point rather than sudden death solely because both situations will, in most cases, result in a class I indication f ICD implantation. Statistical Analysis. Differences in the results of clinical and hemodynamic data were checked f significance by means of the Student t test f matched pairs and are expressed as mean SD. Nonparametric data were checked f significance with Wilcoxon tests and are expressed as medians with intraquartile ranges. Survival rate was calculated by Kaplan Meier analysis (heart transplant and death other than sudden death were censed), and differences between groups were tested f significance by multivariance analysis (with Bonferroni crection) and the log-rank test (Winstat.1, Kalmia Co, Inc, Cambridge, MA, and SPSS f Windows.1, SPSS Inc, Chicago, IL). TheriskofSCAwasexpressedasfreedom from such an event. Results Our patient population was representative f a coht of patients with severe heart failure (NYHA III IV) at study entry (Table). Risk stratification accding to the HFSS showed a typical separation in heart failure parameters. We found a total of 1 SCA events. Of these cases, 1 % were successfully resuscitated. In % of cases there was no infmation about the heart rhythm at COLOUR COLOUR time of death. SCA cases differed in general from non-sca cases by their lower HFSS sce (.0. vs.0.) (P<.01). No significant difference in SCA risk (expressed as freedom from such an event) was found over time between patients with without CAD (Figure 1). On further follow-up, groups of HFSS risk with CAD were separately analyzed and compared. No significant difference was shown in the risk f sudden death - SCA between these groups accding to our definition (Figure ). In contrast, patients with non-cad and a high-risk HFSS sce had a distinctive and significantly greater risk f SCA than the other non-cad groups (Figure ) (P<.001). We further analyzed possible relationshipsofbodymassindex(bmi)tothe HFSS and found a significantly lower BMI in the high-risk group (BMI low risk,. kg m ; BMI medium risk,. kg m ; BMI high risk,. kg m [P=.00 high-risk group vs others]). However, BMI did not separate between etiologies, when they were stratified accding the HFSS groups (Table). BMI showed consistently lower values in the high-risk groups. Patients with cardiac arrest (n=) did not differ from patients without this condition (n=) in terms of BMI (. vs.kg m ; P=not significant). Discussion In regard to patient characteristics accding to HFSS classes (Table), our coht was comparable to the heart transplant candidates studied by Deng and colleagues. Therefe, we think that our results can be extrapolated to other patients with severe heart failure, recommended f heart transplantation. Follow-up data revealed that our Figure 1. Freedom from sudden cardiac arrest (SCA) and disease etiology in conary artery disease (CAD) (n=1) vs non-cad (n=) patients. 0 indicates no CAD; 1, CAD. Figure. Freedom from sudden cardiac arrest (SCA) in conary artery disease (CAD) patients (n=1) stratified by heart failure survival sce (HFSS) P=not significant between the groups. 1 November December 00

1 1 1 1 0 1 0 1 COLOUR Figure. Freedom from sudden cardiac arrest (SCA) in non conary artery disease (CAD) patients (n=) stratified by heart failure survival sce (HFSS). * High-risk class vs others P<.001. REFERENCES 1 Bardy GH, Lee KL, Mark DB, et al. Amiodarone an implantable cardioverter-defibrillat f congestive heart failure. N Engl J Med. 00;:. patients with severe, but stabilized, CHF are at high risk f SCA irrespective of the underlying disease etiology, even after years (in fav of the non-cad patients) (Figure 1). Our data show no significant difference in SCA risk between HFSS groups within the coht of patients with CAD (Figure ). One pointisasomewhatlowerincidenceof SCA in the first observation years in the low-risk patients, but this advantage disappeared after years. Therefe, in this group, the HFSS adds little to our clinical distinction f sudden death risk stratification. CAD and non-cad CHF patients did not differ in their overall SCA risk and should therefe be treated similarly in regard to primary prevention measures (ICD). In our present studywehaveshownfthefirsttime, to our knowledge, that the HFSS is useful f the detection of a subgroup of non-cad severe CHF patients at very high risk f sudden cardiac death SCA (Figure ). In contrast, the BMI did not separate between CAD and non-cad patients, but the high-risk HFSS group in general showed lower BMI values, well fitting to the obesity paradox in CHF. The risk of SCA in this group appears to be substantially higher than observed by others. This may be due to the me advanced stage of disease in our coht. The risk seems to be dependent on data from first evaluation, irrespective of possible later improvement in NYHA classification, abolishing the need f heart transplant. In these patients f whom heart transplant has been deferred due to hemodynamic stabilization, an ICD is, in our opinion, an absolutely necessary safety net. The risk of sudden death is nonlinear over time (Figure and Figure ). The period of highest risk is the first year of observation. Therefe, a quick decision has to be made f early prophylactic ICD implantation. This concept should be evaluated in further studies. When taking into account that approximately 0% to 0% of SCA events are due to tachyarrhythmia, abenefitfrom early prophylactic ICD implantation can be expected. This recommendation is suppted by the results from the Defibrillats in Non-Ischemic Cardiomyopathy Treatment Evaluation (DEFINITE) trial. A causal relationship between a certain critical LVEF and the occurrence of SCA has yet not been demonstrated with respect to methods of risk stratification. That multiple mechanisms may contribute to the occurrence of ventricular tachycardia ventricular fibrillation impliesthatnosingletestislikelyto accurately predict risk of sudden death f all patients. Cardiac patients vary with both the presence and the type of underlying heart disease. In addition, Nägele H, Rödiger W. Sudden death and tailed medical therapy in elective candidates f heart transplantation. J Heart Lung Transplant. 1;1():. mechanisms underlying arrhythmias may depend on the stage of disease. Therefe, in our opinion, the decision f ICD implantation may be based on additional facts to LVEF and NYHA class. 1 Accdingly, in our patients, there was only a modest difference in LVEF between the high- and mediumrisk non-cad patients (Table). We suggest a sce covering different aspects of CHF, such as the HFSS, to aid decision-making at least in patients with non-cad and severe CHF. One could argue that high-risk HFSS patients should receive an ICD anyway accding to guidelines; however, several patients experienced their SCA events after a long follow-up in better NYHA classes and better LVEF (Figure ) and did not fulfill ICD implantation criteria any me. In this regard in our means data of first evaluation should be used irrespective of patient s individual improvement. It is unknown whether the recently published Seattle Heart Failure Model, 1 which incpates continuous and discrete variables to estimate 1-, -, and -year survival, is a better statistical tool to predict SCA. The HFSS is driven mainly by VO, which is not a part of the Seattle Heart Failure Model. In contrast, the Seattle Model is based on the diuretic dosage. Additional studies are needed to determine which model provides the best clinical guideline. Limitations Our study is a retrospective single center analysis and is therefe limited. A Cox proptional hazards model that includes as covariates HFSS groups was not perfmed because of colinearity of variables. Furtherme, our results were obtained befe the wide application of cardiac resynchronization therapy (which itself may alter the risk of sudden cardiac arrest). Therefe, our results should be confirmed in multicenter prospective investigations including modern device therapy. Buxton AE. Not everyone with an ejection fraction <0% should receive an implantable cardioverter-defibrillat. Circulation. 00; 1:. 1 November December 00

1 1 1 1 0 1 0 1 Bänsch D, Antz M, Bocz S, et al. Primary prevention of sudden cardiac death in idiopathic dilated cardiomyopathy: the Cardiomyopathy Trial (CAT). Circulation. 00; : 1. Kadish A, Dyer A, Daubert JP, et al. Investigats prophylactic defibrillat implantation in patients with nonischemic dilated cardiomyopathy. N Engl J Med. 00;0: 1 1. Zipes DP, Wellens HJ. Sudden cardiac death. Circulation. 1;: 1. Buxton AE, Lee KL, Fisher JD, et al. A randomized study of the prevention of sudden death in patients with conary artery disease. N Engl J Med. 1;: 1 10. Moss AJ, Hall WJ, Cannom DS, et al. Improved survival with an implanted defibrillat in patients with conary disease at high risk f ventricular arrhythmia. Multicenter Automatic Defibrillat Implantation Trial Investigats. N Engl J Med. 1;: 1 1. Moss AJ, Zareba W, Hall WJ, et al. Prophylactic implantation of a defibrillat in patients with myocardial infarction and reduced ejection fraction. N Engl J Med. 00;:. Aaronson KD, Schwartz JS, Chen TM, et al. Development and prospective validation of a clinical index to predict survival in ambulaty patients referred f cardiac transplant evaluation. Circulation. 1;:. Lund LH, Aaronson KD, Mancini DM. Predicting survival in ambulaty patients with severe heart failure on beta-blocker therapy. Am J Cardiol. 00;:. 1 Green P, Lund LH, Mancini D. Comparison of peak exercise oxygen consumption and the Heart Failure Survival Sce f predicting prognosis in women versus men. Am J Cardiol. 00;:. Deng MC, De Meester JMJ, Smits JMA, et al. Effect of receiving a heart transplant: analysis of a national coht entered on to a waiting list, stratified by heart failure severity. BMJ. 000;1:. Mancini D, Katz S, Donchez L, et al. Coupling of hemodynamic measurements with oxygen consumption during exercise does not improve risk stratification in patients with heart failure. Circulation. 1;:. Hwich TB, Fonarow GC, Hamilton MA, et al. The relationship between obesity and mtality in patients with heart failure. JAm Coll Cardiol. 001;:. Nägele H, Hashagen S, Azizi M, et al. Analysis of terminal arrhythmias sted in the memy of pacemakers from patients dying suddenly. Europace. 00;:0. 1 Levy W, Mozaffarian D, Linker D, et al. The Seattle Heart Failure model: the prediction of survival in heart failure. Circulation. 00; :1. 1 November December 00

Auth Query Fm Journal: CHF Article: 0 Dear Auth, During the copy-editing of your paper, the following queries arose. Please respond to these by marking up your proofs with the necessary changes/additions. Please write your answers on the query sheet if there is insufficient space on the page proofs. Please write clearly and follow the conventions shown on the attached crections sheet. If returning the proof by fax do not write too close to the paper s edge. Please remember that illegible mark-ups may delay publication. Many thanks f your assistance. Query reference Query Remarks Q1 AUTHOR: Please provide suggested sht title running footer.

MARKED PROOF Please crect and return this set Please use the proof crection marks shown below f all alterations and crections. If you wish to return your proof by fax you should ensure that all amendments are written clearly in dark ink and are made well within the page margins. Instruction to printer Leave unchanged Insert in text the matter indicated in the margin Delete Substitute character substitute part of one me wd(s) Change to italics Change to capitals Change to small capitals Change to bold type Change to bold italic Change to lower case Change italic to upright type Change bold to non-bold type Insert superi character Insert inferi character Insert full stop Insert comma Insert single quotation marks Textual mark under matter to remain through single character, rule underline through all characters to be deleted through letter through characters Encircle matter to be changed through character where required Marginal mark New matter followed by new character new characters under character e.g. over character e.g. and/ Insert double quotation marks and/ Insert hyphen Start new paragraph No new paragraph Transpose Close up linking characters Insert substitute space between characters wds through character where required Reduce space between characters wds between characters wds affected