This Article Abstract Full Text (PDF) Data Supplement All Versions of this Article: 110/22/3444    most recent 01.CIR.0000148178.19465.11v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Kittleson, M. M. Articles by Hare, J. M. Search for Related Content PubMed PubMed Citation Articles by Kittleson, M. M. Articles by Hare, J. M. Pubmed/NCBI databases Medline Plus Health Information Cardiomyopathy Related Collections Chronic ischemic heart disease Congestive

(Circulation. 2004;110:3444-3451.)
© 2004 American Heart Association, Inc.

Heart Failure

Identification of a Gene Expression Profile That Differentiates Between Ischemic and Nonischemic Cardiomyopathy

Michelle M. Kittleson, MD; Shui Q. Ye, MD, PhD; Rafael A. Irizarry, PhD; Khalid M. Minhas, MD; Gina Edness, RN; John V. Conte, MD; Giovanni Parmigiani, PhD; Leslie W. Miller, MD; Yingjie Chen, MD, PhD; Jennifer L. Hall, PhD; Joe G.N. Garcia, MD; Joshua M. Hare, MD

From Departments of Medicine, Division of Cardiology (M.M.K., K.M.M., G.E., J.M.H.) and Pulmonary and Critical Care Medicine (S.Q.Y., J.G.N.G.), Department of Surgery, Cardiothoracic Division (J.V.C.), Johns Hopkins University School of Medicine and the HOPGENE Applied Genomics in Cardiopulmonary Disease (M.M.K., S.Q.Y., K.M.M., G.E., J.G.N.G., J.M.H.), Baltimore, Md; Department of Biostatistics (R.A.I., G.P.), Johns Hopkins University School of Public Health, Baltimore, Md; and Department of Medicine (L.W.M., Y.C., J.L.H.), University of Minnesota, Minneapolis, Minn.

Correspondence to Joshua M. Hare, MD, Ross 1059, 720 Rutland Ave, Baltimore, MD 21205. E-mail jhare{at}mail.jhmi.edu

Received June 1, 2004; revision received September 15, 2004; accepted September 29, 2004.

	Abstract

Top Abstract Introduction Methods Results Discussion References

 
Background— Gene expression profiling refines diagnostic and prognostic assessment in oncology but has not yet been applied to myocardial diseases. We hypothesized that gene expression differentiates ischemic and nonischemic cardiomyopathy, demonstrating that gene expression profiling by clinical parameters is feasible in cardiology.

Methods and Results— Affymetrix U133A microarrays of 48 myocardial samples from Johns Hopkins Hospital (JHH) and the University of Minnesota (UM) obtained (1) at transplantation or left ventricular assist device (LVAD) placement (end-stage; n=25), (2) after LVAD support (post-LVAD; n=16), and (3) from newly diagnosed patients (biopsy; n=7) were analyzed with prediction analysis of microarrays. A training set was used to develop the profile and test sets to validate the accuracy of the profile. An etiology prediction profile developed in end-stage JHH samples was tested in independent samples from both JHH and UM with 100% sensitivity and 100% specificity in end-stage samples and 33% sensitivity and 100% specificity in both post-LVAD and biopsy samples. The overall sensitivity was 89% (95% CI 75% to 100%), and specificity was 89% (95% CI 60% to 100%) over 210 random partitions of end-stage samples into training and test sets. Age, gender, and hemodynamic differences did not affect the profile’s accuracy in stratified analyses. Select gene expression was confirmed with quantitative polymerase chain reaction.

Conclusions— Gene expression profiling accurately predicts cardiomyopathy etiology, is generalizable to samples from separate institutions, is specific to disease stage, and is unaffected by differences in clinical characteristics. This strongly supports ongoing efforts to incorporate expression profiling-based biomarkers in determining prognosis and response to therapy in heart failure.

Key Words: heart failure • genetics • cardiomyopathy

	Introduction

Top Abstract Introduction Methods Results Discussion References

 
Gene expression profiling has the potential to refine diagnostic and prognostic accuracy in a variety of diseases. This technique has enjoyed widespread success in solid and hematologic malignancies^1–7 and may soon be used in clinical trials. Although the ability to refine diagnosis and predict patient outcome has tremendous importance in myocardial diseases, the application of gene expression profiling is in its earliest stages. Small studies offer novel insights into gene expression in failing and nonfailing hearts,^8–13 dilated and hypertrophic cardiomyopathy,¹⁴ and before and after ventricular assist device placement^15–18; however, gene expression analysis has not yet been used to distinguish clinically relevant cardiovascular disease subtypes. In fact, gene expression prediction analysis for cardiomyopathies is considered controversial because of the contention that unlike tumors, there is a final common pathway independent of etiology for the progression of myocardial disease.¹⁹

The aim of the present study was to test the hypothesis that gene expression profiling could discriminate between the 2 major forms of cardiomyopathy, ischemic (ICM) and nonischemic (NICM). We demonstrate that the methodology is highly generalizable to data obtained in different institutions and is specific to disease stage. This study establishes proof-of-principle that gene expression profiles have the potential to refine the evaluation and treatment of heart failure patients, where management decisions may vary on the basis of disease etiology.^20–25 Our findings strongly support ongoing efforts to incorporate expression profiling-based biomarkers in determining prognosis and response to therapy in heart failure.

	Methods

Top Abstract Introduction Methods Results Discussion References

 
Patients
The study comprised 48 samples. Myocardial tissue from different disease stages was obtained (1) at left ventricular assist device (LVAD) placement or cardiac transplantation (n=25, end stage), (2) after LVAD support (n=16; post-LVAD), and (3) from prospectively collected newly diagnosed patients at endomyocardial biopsy (n=7; biopsy). Samples were from 2 institutions: the Johns Hopkins Hospital (n=34) and the University of Minnesota (n=14). Samples from the latter institution were collected and prepared independently,¹⁵ and the gene expression data files were kindly provided.

Written informed consent was obtained from all patients undergoing endomyocardial biopsy for sample collection and medical chart abstraction. Myocardial tissue obtained at LVAD placement, after LVAD support, or at cardiac transplantation, however, is considered discarded tissue. Therefore, we obtained an exemption from the Johns Hopkins Institution Review Board for its collection and medical chart abstraction without written informed consent.

ICM was defined as histological evidence of ischemic injury, and all ICM patients exhibited severe coronary artery disease (>75% stenosis of the left anterior descending artery and at least one other proximal epicardial artery) and/or a documented history of myocardial infarction.²⁶ NICM patients had no history of myocardial infarction, revascularization, or coronary artery disease. Newly diagnosed patients were those presenting or referred to Johns Hopkins Hospital with a new diagnosis of cardiomyopathy and symptoms for 6 months or less²⁰ for further diagnostic evaluation, which included endomyocardial biopsy.

Prediction Analysis
Sample collection and preparation, microarray hybridization, data normalization, and quantitative polymerase chain reaction (PCR) are fully described in the Appendix in the online-only Data Supplement. To develop a gene expression profile that distinguished ICM from NICM, we used Prediction Analysis of Microarrays (PAM)³ implemented in the R package for statistical computing (available at www.R-project.org).

A number of prediction analyses were performed (Figure 1). Sixteen end-stage cardiomyopathy samples (6 ICM and 10 NICM) from Johns Hopkins Hospital formed a training set to develop the etiology prediction profile. There were 3 test sets to validate the profile: (1) the remaining 9 end-stage cardiomyopathy samples, including 7 from the University of Minnesota; (2) 16 post-LVAD samples; and (3) 7 biopsy samples.

View larger version (24K): [in this window] [in a new window]   Figure 1. Overall study design. Samples were first divided into training set of end-stage samples from Johns Hopkins Hospital (JHH) used to develop etiology prediction profile. It then was validated by determining its sensitivity and specificity in 3 test sets: end-stage samples from University of Minnesota (UM), post-LVAD samples from both JHH and UM, and endomyocardial biopsy samples from newly diagnosed patients at JHH. Then, end-stage samples were randomly partitioned into training and test sets to identify representative etiology prediction profile and overall sensitivity and specificity.

Because the accuracy of the profile could differ on the basis of the random division of samples into training and test tests, the above analysis was repeated with 210 random partitions of the samples into a 16-sample training set and 9-sample test set to determine the overall accuracy. Each random partition identified different, overlapping sets of genes, but a 90-gene profile repeatedly minimized the cross-validation error. We applied PAM to the entire set of 25 end-stage samples to identify the 90-gene profile as the representative etiology prediction profile. The genes in the etiology prediction profile were classified by the Gene Ontology Consortium system, and the profile was visualized by hierarchical clustering and a heat map²⁷ that used euclidean distance with complete linkage.

	Results

Top Abstract Introduction Methods Results Discussion References

 
Patient Characteristics
All end-stage patients exhibited severely reduced ejection fraction, left ventricular dilation, elevated pulmonary arterial and wedge pressures, and reduced cardiac index (Table 1). Compared with end-stage NICM patients, end-stage and post-LVAD ICM patients were older, and all were male. In addition, compared with end-stage NICM patients, end-stage ICM patients were all taking ACE inhibitors and less often were taking intravenous inotropes. Hemodynamic and remodeling indices were similar between end-stage ICM and NICM patients. Newly diagnosed ICM patients were older than their NICM counterparts and had better remodeling indices.

View this table: [in this window] [in a new window]   TABLE 1. Clinical Characteristics of the Study Subjects

Diagnostic Accuracy
We developed an etiology prediction profile using a training set of samples that demonstrated 100% sensitivity and 100% specificity when applied to independent end-stage ICM and NICM samples. This perfect accuracy was also achieved in a test set in which the majority of samples were from an institution distinct from that used to create the prediction profile.

We assessed whether the etiology prediction profile was affected by disease stage. In post-LVAD samples, the gene expression profile correctly classified all NICM samples (n=13; specificity 100%) but only classified 1 of 3 ICM samples correctly (sensitivity 33%). In biopsy samples from patients with newly diagnosed cardiomyopathy, the profile again correctly classified all NICM samples (n=4; specificity 100%) but only classified 1 of 3 ICM samples correctly (sensitivity 33%). The overall accuracy over 210 random partitions of training and test sets was sensitivity 89% (95% CI 75% to 100%) and specificity 89% (95% CI 60% to 100%).

Effect of Clinical Characteristics
We examined the predictive accuracy of the profile in strata based on each clinical characteristic (Table 2). All ICM patients were male and taking ACE inhibitors, so we could not ascertain whether the profile would apply to ICM women not taking ACE inhibitors. However, within each stratum, the sensitivity and specificity were similar, and all were comparable to the overall sensitivity and specificity.

View this table: [in this window] [in a new window]   TABLE 2. Sensitivity and Specificity of Etiology Prediction Profile in Strata Defined by Clinical Covariates

Characterization of the Etiology Prediction Profile
In the 210 combinations of training and test set samples, the greatest accuracy was achieved with profiles that contained 90 genes. A 90-gene expression profile exhibited perfect accuracy 30% of the time. The majority of genes in the representative etiology prediction profile were involved in signal transduction, metabolism, and cell growth/maintenance (Figure 2). Most were upregulated in ICM, with an average fold change of 1.9±0.9 (Table 3).

View larger version (12K): [in this window] [in a new window]   Figure 2. Number of genes upregulated and downregulated in ICM relative to NICM classified by functional group. CGM indicates cell growth/maintenance; CYT, cytoskeleton; DEV, development; IMM, immune response; MET, metabolism; OTH, other; SIG, signal transduction; and TRA, transport.

View this table: [in this window] [in a new window]   TABLE 3. Gene Expression Prediction Profile

View this table: [in this window] [in a new window]   TABLE 3. Continued

In a hierarchical clustering algorithm, 11 of the 16 ICM samples and 30 of the 32 NICM samples formed a distinct cluster (Figure 3). Whereas the biopsy samples clustered together, the samples did not cluster by pre- or post-LVAD status or by institution of origin.

View larger version (67K): [in this window] [in a new window]   Figure 3. Hierarchical clustering of 90 genes in 48 samples based on similarity in gene expression and relatedness of samples. Each row represents 1 gene and each column represents 1 sample. Color in each cell reflects level of expression of corresponding gene in corresponding sample, relative to its mean level of expression in entire set of samples. Expression levels greater than mean are shaded in blue, and those below mean are shaded in red. Samples form 2 distinct clusters based on etiology. Arrows denote samples that do not appear in their etiology cluster. ICMB denotes newly diagnosed ischemic cardiomyopathy and NICMB denotes newly diagnosed nonischemic cardiomyopathy.

Quantitative PCR
Levels of transcript for 4 genes in the prediction profile were confirmed with quantitative PCR (Figure 4). In all 4 cases, the direction of the fold change by microarray and quantitative PCR was the same.

View larger version (12K): [in this window] [in a new window]   Figure 4. Independent assessment of gene-expression levels. Quantitative PCR (qPCR) was used to determine transcript abundance of 4 genes. Fold change in expression is in ischemic relative to nonischemic hearts according to qPCR (solid bars) and microarrays (open bars) with standard errors. Gene symbols are those from Table 3.

	Discussion

Top Abstract Introduction Methods Results Discussion References

 
The major new finding of this report is the identification of a gene expression profile that accurately distinguishes ischemic and nonischemic cardiomyopathy. The prediction profile was generalizable to samples from different laboratories and was independent of disease stage for nonischemic samples. Gene expression profiles have been successfully correlated with etiology and outcome in oncology.^1–7 There is an equal need to refine diagnostic and prognostic techniques in myocardial diseases, but advances have been restricted by limited tissue access. The findings of the present study demonstrate that gene expression profiling can accurately identify etiology in cardiovascular disease and support ongoing efforts to incorporate expression profiling-based biomarkers in determining prognosis and response to therapy.

Although the main focus of the present study was proof-of-principle, a gene expression profile that distinguishes ICM and NICM could provide a valuable adjunct to diagnostic imaging and metabolic tools. ICM and NICM are distinct diseases; patients with ICM have decreased survival compared with their NICM counterparts^20,21 and respond differently to therapies.^22–25 An etiology prediction profile would offer diagnostic insight, especially in patients with heart failure out of proportion to their coronary artery disease, up to 11% in one observational study.²⁶

Although we have demonstrated that end-stage cardiomyopathy can be accurately classified by gene expression, a more relevant prediction profile would focus on newly diagnosed patients. Therefore, we also tested the end-stage etiology prediction profile in 7 endomyocardial biopsy samples prospectively collected from patients with newly diagnosed cardiomyopathy. The profile performed perfectly in NICM, whereas only 1 of 3 ICM samples was classified correctly. This suggests that compared with NICM patients, those with ICM exhibit greater changes in gene expression as disease progresses. These results parallel those from post-LVAD patients and emphasize the need for stage-specific prediction profiles. To the best of our knowledge, this is the first evidence that microarray hybridization from endomyocardial biopsies is feasible; this success encourages future gene expression profiling studies using endomyocardial biopsies with RNA amplification.

Prior studies have shown that cardiomyopathy of different etiologies exhibits different patterns of gene expression^9,18; however, neither study developed or prospectively validated a gene expression prediction profile. In fact, one study comparing the gene expression of ICM and NICM found no differentially expressed genes.¹¹ That study used pooled samples from only 2 ICM and 2 NICM patients and likely did not have adequate resolution to detect changes in gene expression.²⁸

Although the differential gene expression between failing and nonfailing hearts has been attributed to age and gender differences,¹³ this analysis has not been extended to ICM and NICM; however, we addressed this possibility by stratifying our analysis by clinical characteristics. The sensitivity and specificity were not affected, which indicates that the accuracy of the etiology prediction profile is not an artifact of differences in baseline characteristics.

The majority of the genes in the etiology prediction profile were not observed in prior microarray analyses. This supports the validity of those studies because they compared failing and nonfailing hearts^9–13 or hearts after LVAD support,^15–18 rather than ICM and NICM. The present analysis also focused on prediction and thus targeted different genes than one that investigates differential gene expression.³

Many of the genes in the etiology prediction profile are not known to be expressed in myocardial tissue. This discrepancy has been observed in prior microarray experiments in cardiomyopathy^9–18 and stems from the gap between the number of genes on the microarray platform and our ability to define their functions.²⁹ However, the inability to justify the biological validity of every gene in the profile does not invalidate its clinical utility. In prediction analysis, it is the pattern of gene expression rather than the individual genes themselves that serves as a biomarker of disease.^3,30,31

Nevertheless, there is biological plausibility for a number of genes in the prediction profile. The upregulation of signal transduction genes in ischemic hearts, including several protein phosphatases and a mitogen-activate protein kinase, is supported by evidence that their gene products may protect against ischemic injury.^32–34 The upregulation of endothelin-converting enzyme in ICM over NICM has also been described.³⁵ One would expect upregulation of genes involved in cell growth/maintenance, including ribosomal and cell division cycle proteins, because the myocyte proliferation rate is higher in ICM than NICM.³⁶ However, although these findings support the biological validity of the etiology prediction profile, the changes bear further investigation with a study focused on differential gene expression. If confirmed, these genes could provide insight into new cause-specific therapies for heart failure patients.

Gene expression analysis is considered hypothesis generating until validated by another technique. Unlike the majority of studies in cardiology, in which microarray analysis focuses on the discovery of novel genetic pathways, the present analysis concentrates on prediction. Thus, validation involves evaluating the predictive accuracy of the profile in independent, blinded samples.^30,31,37 This is an established approach among studies in the cancer literature.^5–7 However, the level of transcript abundance should also be confirmed with quantitative PCR to determine whether the prediction profile offers utility independent of the microarray platform. We confirmed the expression level of 4 genes in the prediction profile using quantitative PCR, and the fold changes agreed in all cases.

Several methodological aspects of the present study warrant mention. There is little information regarding sample-size requirements in microarray analysis. One study determined that for accurate class prediction of etiology, a training set of 10 to 20 samples is required.²⁸ Thus, the sample size in the present study was adequate for prediction. We were also able to maximize the amount of information obtained by random partitioning of samples. Furthermore, to the best of our knowledge, the present study includes the largest number of samples in a cardiovascular microarray study to date.

Finally, it may be argued that a gene expression profile that identifies patients on the basis of prognosis would be more clinically valuable than one based on etiology, which is determinable by other methods. The present findings are valuable proof-of-concept that other predictions will be possible in the future. Indeed, the transition from gene profiling of etiology^1–3 to gene profiling of prognosis^4–7 represents the path taken in the oncology experience.

The present study represents the first use of gene expression profiling in cardiovascular disease and the first evidence that microarray hybridization from endomyocardial biopsies is feasible. Microarray analysis has the potential to optimize the diagnosis and management of patients with myocardial diseases. These results form the basis for future studies using molecular profiling to distinguish cardiomyopathy patients by other relevant clinical parameters. Studies are currently under way to develop gene expression profiles that distinguish ischemic and nonischemic cardiomyopathy in newly diagnosed patients and to differentiate these patients by prognosis and response to therapy.

	Acknowledgments

 
This research was supported by NIH grants 1U01-HL-066583 (Dr Garcia) and 5RO1-HL-065455 (Dr Hare). Dr Hare is a recipient of the Paul Beeson Physician Faculty Scholars in Aging Research Award. The authors are indebted to Maryann Albaugh, RN, Stacey Nash, RN, Elayne Breton, RN, and Thomas Brown, BA, for their valuable assistance in sample collection and to Katharine Gutman and Elizabeth Schifano for technical support.

	Footnotes

 
The online-only Data Supplement, which contains an Appendix, can be found with this article at http://www.circulationaha.org.

Presented at the Council on Clinical Cardiology Samuel A. Levine Young Clinical Investigator Award competition at the 77th Scientific Sessions of the American Heart Association, New Orleans, La, November 7 to 10, 2004, and published in abstract form (Circulation. 2004;110[suppl III]:III-335).

	References

Top Abstract Introduction Methods Results Discussion References

 

1. Alizadeh AA, Eisen MB, Davis RE, Ma C, Lossos IS, Rosenwald A, Boldrick JC, Sabet H, Tran T, Yu X, Powell JI, Yang L, Marti GE, Moore T, Hudson J Jr, Lu L, Lewis DB, Tibshirani R, Sherlock G, Chan WC, Greiner TC, Weisenburger DD, Armitage JO, Warnke R, Levy R, Wilson W, Grever MR, Byrd JC, Botstein D, Brown PO, Staudt LM. Distinct types of diffuse large B-cell lymphoma identified by gene expression profiling. Nature. 2000; 403: 503–511.[CrossRef][Medline] [Order article via Infotrieve]
   
2. Golub TR, Slonim DK, Tamayo P, Huard C, Gaasenbeek M, Mesirov JP, Coller H, Loh ML, Downing JR, Caligiuri MA, Bloomfield CD, Lander ES. Molecular classification of cancer: class discovery and class prediction by gene expression monitoring. Science. 1999; 286: 531–537.[Abstract/Free Full Text]
   
3. Tibshirani R, Hastie T, Narasimhan B, Chu G. Diagnosis of multiple cancer types by shrunken centroids of gene expression. Proc Natl Acad Sci U S A. 2002; 99: 6567–6572.[Abstract/Free Full Text]
   
4. Rosenwald A, Wright G, Chan WC, Connors JM, Campo E, Fisher RI, Gascoyne RD, Muller-Hermelink HK, Smeland EB, Giltnane JM, Hurt EM, Zhao H, Averett L, Yang L, Wilson WH, Jaffe ES, Simon R, Klausner RD, Powell J, Duffey PL, Longo DL, Greiner TC, Weisenburger DD, Sanger WG, Dave BJ, Lynch JC, Vose J, Armitage JO, Montserrat E, Lopez-Guillermo A, Grogan TM, Miller TP, LeBlanc M, Ott G, Kvaloy S, Delabie J, Holte H, Krajci P, Stokke T, Staudt LM. The use of molecular profiling to predict survival after chemotherapy for diffuse large-B-cell lymphoma. N Engl J Med. 2002; 346: 1937–1947.[Abstract/Free Full Text]
   
5. van de Vijver MJ, He YD, van’t Veer LJ, Dai H, Hart AA, Voskuil DW, Schreiber GJ, Peterse JL, Roberts C, Marton MJ, Parrish M, Atsma D, Witteveen A, Glas A, Delahaye L, van der Velde T, Bartelink H, Rodenhuis S, Rutgers ET, Friend SH, Bernards R. A gene-expression signature as a predictor of survival in breast cancer. N Engl J Med. 2002; 347: 1999–2009.[Abstract/Free Full Text]
   
6. Valk PJ, Verhaak RG, Beijen MA, Erpelinck CA, Barjesteh van Waalwijk van Doorn-Khosrovani S, Boer JM, Beverloo HB, Moorhouse MJ, van der Spek PJ, Lowenberg B, Delwel R. Prognostically useful gene-expression profiles in acute myeloid leukemia. N Engl J Med. 2004; 350: 1617–1628.[Abstract/Free Full Text]
   
7. Bullinger L, Dohner K, Bair E, Frohling S, Schlenk RF, Tibshirani R, Dohner H, Pollack JR. Use of gene-expression profiling to identify prognostic subclasses in adult acute myeloid leukemia. N Engl J Med. 2004; 350: 1605–1616.[Abstract/Free Full Text]
   
8. Barrans JD, Allen PD, Stamatiou D, Dzau VJ, Liew CC. Global gene expression profiling of end-stage dilated cardiomyopathy using a human cardiovascular-based cDNA microarray. Am J Pathol. 2002; 160: 2035–2043.[Abstract/Free Full Text]
   
9. Tan FL, Moravec CS, Li J, Apperson-Hansen C, McCarthy PM, Young JB, Bond M. The gene expression fingerprint of human heart failure. Proc Natl Acad Sci U S A. 2002; 99: 11387–11392.[Abstract/Free Full Text]
   
10. Yung CK, Halperin VL, Tomaselli GF, Winslow RL. Gene expression profiles in end-stage human idiopathic dilated cardiomyopathy: altered expression of apoptotic and cytoskeletal genes. Genomics. 2004; 83: 281–297.[CrossRef][Medline] [Order article via Infotrieve]
    
11. Steenman M, Chen YW, Le Cunff M, Lamirault G, Varro A, Hoffman E, Leger JJ. Transcriptomal analysis of failing and nonfailing human hearts. Physiol Genomics. 2003; 12: 97–112.[Abstract/Free Full Text]
    
12. Kaab S, Barth AS, Margerie D, Dugas M, Gebauer M, Zwermann L, Merk S, Pfeufer A, Steinmeyer K, Bleich M, Kreuzer E, Steinbeck G, Nabauer M. Global gene expression in human myocardium-oligonucleotide microarray analysis of regional diversity and transcriptional regulation in heart failure. J Mol Med. 2004; 82: 308–316.[CrossRef][Medline] [Order article via Infotrieve]
    
13. Boheler KR, Volkova M, Morrell C, Garg R, Zhu Y, Margulies K, Seymour AM, Lakatta EG. Sex- and age-dependent human transcriptome variability: implications for chronic heart failure. Proc Natl Acad Sci U S A. 2003; 100: 2754–2759.[Abstract/Free Full Text]
    
14. Hwang JJ, Allen PD, Tseng GC, Lam CW, Fananapazir L, Dzau VJ, Liew CC. Microarray gene expression profiles in dilated and hypertrophic cardiomyopathic end-stage heart failure. Physiol Genomics. 2002; 10: 31–44.[Abstract/Free Full Text]
    
15. Chen Y, Park S, Li Y, Missov E, Hou M, Han X, Hall JL, Miller LW, Bache RJ. Alterations of gene expression in failing myocardium following left ventricular assist device support. Physiol Genomics. 2003; 14: 251–260.[Abstract/Free Full Text]
    
16. Hall JL, Grindle S, Han X, Fermin D, Park S, Chen Y, Bache RJ, Mariash A, Guan Z, Ormaza S, Thompson J, Graziano J, Sam Lazaro SE, Pan S, Simari RD, Miller LW. Genomic profiling of the human heart before and after mechanical support with a ventricular assist device reveals alterations in vascular signaling networks. Physiol Genomics. 2004; 17: 283–291.[Abstract/Free Full Text]
    
17. Chen MM, Ashley EA, Deng DX, Tsalenko A, Deng A, Tabibiazar R, Ben Dor A, Fenster B, Yang E, King JY, Fowler M, Robbins R, Johnson FL, Bruhn L, McDonagh T, Dargie H, Yakhini Z, Tsao PS, Quertermous T. Novel role for the potent endogenous inotrope apelin in human cardiac dysfunction. Circulation. 2003; 108: 1432–1439.[Abstract/Free Full Text]
    
18. Blaxall BC, Tschannen-Moran BM, Milano CA, Koch WJ. Differential gene expression and genomic patient stratification following left ventricular assist device support. J Am Coll Cardiol. 2003; 41: 1096–1106.[Abstract/Free Full Text]
    
19. Towbin JA, Bowles NE. Molecular genetics of left ventricular dysfunction. Curr Mol Med. 2001; 1: 81–90.[CrossRef][Medline] [Order article via Infotrieve]
    
20. Felker GM, Thompson RE, Hare JM, Hruban RH, Clemetson DE, Howard DL, Baughman KL, Kasper EK. Underlying causes and long-term survival in patients with initially unexplained cardiomyopathy. N Engl J Med. 2000; 342: 1077–1084.[Abstract/Free Full Text]
    
21. Dries DL, Sweitzer NK, Drazner MH, Stevenson LW, Gersh BJ. Prognostic impact of diabetes mellitus in patients with heart failure according to the etiology of left ventricular systolic dysfunction. J Am Coll Cardiol. 2001; 38: 421–428.[Abstract/Free Full Text]
    
22. Felker GM, Benza RL, Chandler AB, Leimberger JD, Cuffe MS, Califf RM, Gheorghiade M, O’Connor CM. Heart failure etiology and response to milrinone in decompensated heart failure: results from the OPTIME-CHF study. J Am Coll Cardiol. 2003; 41: 997–1003.[Abstract/Free Full Text]
    
23. Kittleson M, Hurwitz S, Shah MR, Nohria A, Lewis E, Givertz M, Fang J, Jarcho J, Mudge G, Stevenson LW. Development of circulatory-renal limitations to angiotensin-converting enzyme inhibitors identifies patients with severe heart failure and early mortality. J Am Coll Cardiol. 2003; 41: 2029–2035.[Abstract/Free Full Text]
    
24. Follath F, Cleland JG, Klein W, Murphy R. Etiology and response to drug treatment in heart failure. J Am Coll Cardiol. 1998; 32: 1167–1172.[Abstract/Free Full Text]
    
25. Reynolds MR, Josephson ME. MADIT II(second Multicenter Automated Defibrillator Implantation Trial) debate: risk stratification, costs, and public policy. Circulation. 2003; 108: 1779–1783.[Free Full Text]
    
26. Felker GM, Shaw LK, O’Connor CM. A standardized definition of ischemic cardiomyopathy for use in clinical research. J Am Coll Cardiol. 2002; 39: 210–218.[Abstract/Free Full Text]
    
27. Eisen MB, Spellman PT, Brown PO, Botstein D. Cluster analysis and display of genome-wide expression patterns. Proc Natl Acad Sci U S A. 1998; 95: 14863–14868.[Abstract/Free Full Text]
    
28. Mukherjee S, Tamayo P, Rogers S, Rifkin R, Engle A, Campbell C, Golub TR, Mesirov JP. Estimating dataset size requirements for classifying DNA microarray data. J Comput Biol. 2003; 10: 119–142.[CrossRef][Medline] [Order article via Infotrieve]
    
29. Butte A. The use and analysis of microarray data. Nat Rev Drug Discov. 2002; 1: 951–960.[CrossRef][Medline] [Order article via Infotrieve]
    
30. Liu ET, Karuturi KR. Microarrays and clinical investigations. N Engl J Med. 2004; 350: 1595–1597.[Free Full Text]
    
31. Cook SA, Rosenzweig A. DNA microarrays: implications for cardiovascular medicine. Circ Res. 2002; 91: 559–564.[Abstract/Free Full Text]
    
32. Ladilov Y, Maxeiner H, Wolf C, Schafer C, Meuter K, Piper HM. Role of protein phosphatases in hypoxic preconditioning. Am J Physiol Heart Circ Physiol. 2002; 283: H1092–H1098.[Abstract/Free Full Text]
    
33. Ng DC, Court NW, dos Remedios CG, Bogoyevitch MA. Activation of signal transducer and activator of transcription (STAT) pathways in failing human hearts. Cardiovasc Res. 2003; 57: 333–346.[Abstract/Free Full Text]
    
34. Schulman D, Latchman DS, Yellon DM. Urocortin protects the heart from reperfusion injury via upregulation of p42/p44 MAPK signaling pathway. Am J Physiol Heart Circ Physiol. 2002; 283: H1481–H1488.[Abstract/Free Full Text]
    
35. Serneri GG, Cecioni I, Vanni S, Paniccia R, Bandinelli B, Vetere A, Janming X, Bertolozzi I, Boddi M, Lisi GF, Sani G, Modesti PA. Selective upregulation of cardiac endothelin system in patients with ischemic but not idiopathic dilated cardiomyopathy: endothelin-1 system in the human failing heart. Circ Res. 2000; 86: 377–385.[Abstract/Free Full Text]
    
36. Kajstura J, Leri A, Finato N, Di Loreto C, Beltrami CA, Anversa P. Myocyte proliferation in end-stage cardiac failure in humans. Proc Natl Acad Sci U S A. 1998; 95: 8801–8805.[Abstract/Free Full Text]
    
37. Simon R, Radmacher MD, Dobbin K, McShane LM. Pitfalls in the use of DNA microarray data for diagnostic and prognostic classification. J Natl Cancer Inst. 2003; 95: 14–18.[Free Full Text]
    

This article has been cited by other articles:

				G. W. Dorn II and S. J. Matkovich Put Your Chips on Transcriptomics Circulation, July 15, 2008; 118(3): 216 - 218. [Full Text] [PDF]

				B. Heidecker, E. K. Kasper, I. S. Wittstein, H. C. Champion, E. Breton, S. D. Russell, M. M. Kittleson, K. L. Baughman, and J. M. Hare Transcriptomic Biomarkers for Individual Risk Assessment in New-Onset Heart Failure Circulation, July 15, 2008; 118(3): 238 - 246. [Abstract] [Full Text] [PDF]

				R. Kuner, A. S. Barth, M. Ruschhaupt, A. Buness, L. Zwermann, E. Kreuzer, G. Steinbeck, A. Poustka, H. Sultmann, and M. Nabauer Genomic analysis reveals poor separation of human cardiomyopathies of ischemic and nonischemic etiologies Physiol Genomics, June 10, 2008; 34(1): 88 - 94. [Abstract] [Full Text] [PDF]

				T. A. Bullard, T. L. Protack, F. Aguilar, S. Bagwe, H. T. Massey, and B. C. Blaxall Identification of Nogo as a novel indicator of heart failure Physiol Genomics, January 17, 2008; 32(2): 182 - 189. [Abstract] [Full Text] [PDF]

				T. P. Cappola Molecular Remodeling in Human Heart Failure J. Am. Coll. Cardiol., January 15, 2008; 51(2): 137 - 138. [Full Text] [PDF]

				S. Ikeda, S. W. Kong, J. Lu, E. Bisping, H. Zhang, P. D. Allen, T. R. Golub, B. Pieske, and W. T. Pu Altered microRNA expression in human heart disease Physiol Genomics, November 14, 2007; 31(3): 367 - 373. [Abstract] [Full Text] [PDF]

				G. Kitsios and E. Zintzaras Genetic Variation associated with Ischemic Heart Failure: A HuGE Review and Meta-Analysis Am. J. Epidemiol., September 15, 2007; 166(6): 619 - 633. [Abstract] [Full Text] [PDF]

				H. Ashrafian and H. Watkins Reviews of Translational Medicine and Genomics in Cardiovascular Disease: New Disease Taxonomy and Therapeutic Implications: Cardiomyopathies: Therapeutics Based on Molecular Phenotype J. Am. Coll. Cardiol., March 27, 2007; 49(12): 1251 - 1264. [Abstract] [Full Text] [PDF]

				G. S. Ginsburg, D. Seo, and C. Frazier Microarrays Coming of Age in Cardiovascular Medicine: Standards, Predictions, and Biology J. Am. Coll. Cardiol., October 17, 2006; 48(8): 1618 - 1620. [Full Text] [PDF]

				M. P. Donahue, D. A. Marchuk, and H. A. Rockman Redefining Heart Failure: The Utility of Genomics J. Am. Coll. Cardiol., October 3, 2006; 48(7): 1289 - 1298. [Abstract] [Full Text] [PDF]

				D. Seo, G. S. Ginsburg, and P. J. Goldschmidt-Clermont Gene Expression Analysis of Cardiovascular Diseases: Novel Insights Into Biology and Clinical Applications J. Am. Coll. Cardiol., July 18, 2006; 48(2): 227 - 235. [Abstract] [Full Text] [PDF]

				A. S. Barth and J. M. Hare The Potential for the Transcriptome to Serve as a Clinical Biomarker for Cardiovascular Diseases Circ. Res., June 23, 2006; 98(12): 1459 - 1461. [Full Text] [PDF]

				W.H. W. Tang and G. S. Francis The Year in Heart Failure J. Am. Coll. Cardiol., December 6, 2005; 46(11): 2125 - 2133. [Full Text] [PDF]

				M. M. Kittleson, K. M. Minhas, R. A. Irizarry, S. Q. Ye, G. Edness, E. Breton, J. V. Conte, G. Tomaselli, J. G. N. Garcia, and J. M. Hare Gene expression analysis of ischemic and nonischemic cardiomyopathy: shared and distinct genes in the development of heart failure Physiol Genomics, May 11, 2005; 21(3): 299 - 307. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Data Supplement All Versions of this Article: 110/22/3444    most recent 01.CIR.0000148178.19465.11v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Kittleson, M. M. Articles by Hare, J. M. Search for Related Content PubMed PubMed Citation Articles by Kittleson, M. M. Articles by Hare, J. M. Pubmed/NCBI databases Medline Plus Health Information Cardiomyopathy Related Collections Chronic ischemic heart disease Congestive