idiopathic dilated cardiomyopathy is a disease of urn&a; etiology. One of the earliest studies of the natural history d

disease (1) reported n 31% l-year mortality Me and a atients with this disease

identified at the Mayo Clinic between and fi973. cently published studies from this same i ution exami a populafion-based cohort with idiopathic dilated cardiomy- opathy identified in Olmsted County between 1975 and 1984, and reported an increasing incidence of the disease and a mortality rate that was significantly Lower (5% at 1 yem and 20% at 5 years) than that reported in the earlier referral cohort (2,3). Olmsted county is the county in which Roch- ester, Minnesota and the Mayo Clinic are located. Ah

Feorn the Division of Cardiovascular Dise~s and Internal Med% Mayo Cliaic and Mayo Foundation, Rochester, Minnesota. This study was funded in part by grants from the Mayo Foundation and from the Joseph P and Jeantne Slnllivan Foundation, Chicago, IfSino~~

Manu!;cript received April 7, 19%; revised tnanusc;ipl reccivcd July 6,

: Dr. Margarei M. kdfietd. Division of Cardiovas:ulw Dkase, Mayo Clinic, 200 First Skeet Southwest, Rochester. Minnesok 55905.

01993 by ki; htasrican College of Cardiology

medical care for tbe

matural history of ~dio~at

each other and wit

outcome of the current referral cohorts will also he: corn-

1922 REDFWXDETAL. NAIWRAE HISTORYOF IDlOPATWIC DILATEDCARIDIOMYOPATfIY

d with that of the popuIation_based cohort diagnosed at institution between 1975 and 1984 (2) to determine

the effkzt of referral bias on studies of the natural history of idiopathic dilated cardiomyopathy.

!Patients with a diagnosis of dilated cardiomyopathy at the aq Clinic between 1976 and 1987 were identified retrca-

istory of heart failure, this

abIe from the record and

&as at the time of diagnosis were obtained

from the medical record of that the recorded history, patients Heart Association functio was not complete enough

the ba

ncas in survival between these previously published referral

orts were assessed by a ~~ut~~s~ comparison of survival curves based on ~~~~w~dq~ for- mula. Statistical significance implies a p value < 0.05 unless othe~~s~ stated.

Folllow-up data were obtained on 222 patients with idio- thic dilated ~~diomyo~thy identified between 1976 and

t&d between 1976 between 1982 and

1987. The clinical characteristics of the two referral groups are

represented in Tabk 1. The mean and median age of the two cohorts were not different. Tbe 19764981 referra! cohort was “sicker” in that the patients were more symE;_umatic and had more signs of decompensated heart fake when f&t identified than did the patients in the 1982-1987 referral cohort. The: 1976-1981 cohort was also more likely to have

.6&X Vol. 2%. No. i

*p 4 0.05. 6VCD = intraventricular conductim defect; SW = jugular venom pressure; NYMA class = New York &art Association funclionai

chss: WCs = premature rentricuhr complexes.

cardiomegnly an and median ejection . More patients in

a serum sodium level < %37 mEq/liter, The ECG characteristics of the two groups were similar.

The prevalence of risk factors for ventricular dysfunction such as alcohol abuse (5% to 7%), history of symptoms suggestive of myocarditis (16% to ES%), family history of idiopathic dilated cardiomyopathy (4% to 7%), history of rheumatic fever and diabetes mellitus was similar in the two cohorts.

The pattern of medication use was assessed at the time the patients were first identified and (w able) at the last subsequent visit to t the time the study questionnaire was distributed. Approxi- mately 65% of patients were taking digoxin amd diuretic drugs when first identified. No patient in the 1976-1981 cohort and 20% of patients in the 1982-1987 cohort were receiving angiotensin-converting enzyme inhibitor therapy when first seen. ApprOXhmitely 50% of patients in each cohort

those in the 1982-8987 cohort were beiag treated ane~otensin-cotasre~~n~ enzyme inhibitors ~~~~w~~~ questionnaire was returned, 50%

rts were being treated with these agents. oronary artery disease was

co~~~~ at cardiac beterizat~on in 156 patieats. TRe

described in the current study (referral cohort 1976-1981 and referial cohort ~9$2-~~~7~ is depicted in Figure B. AB depicted are the survival curves from tkl: previolssly pu lisbed MO-1973 referral cohort and Bished population-bas dilated cardiomyopat tween 1975 and 1981. S referral cohort was n3t di described 1960-197: efe follow-up (p > O.ai_C!. Survival in the c’srrent ~9~~-~~~7 referral cohort was better than that of the 1976-198 B rekrra[l cohort (p = 0.0002) and was significantly better than that of the previously published f960-1975 re points in 6 years of follow-up (p < OSX) in the previously published populatio r,ignificantly better than that of the 1982-8987 referral cohort at years 2 to 6 of foilow-up (p < 0.05).

Figure 2 depicts the survival in the current referral cohorts in each New York I-kart Association functiona! class. As this figure shows, the improved survival in the more recent l982-1987 referral cohort when compared to tie: 1976-1981 referral cohort appears to be strongest ia those patients who were in functiona! class I or II1 at the time of initial evaltiation.

Several clinical, radi aphic and EC33 variables wet-k: examined as possible p cm5 of morta~ity~ Tbosc vati* ablf s that were ificant predictors of mortality by univari- ate analysis alo ith the hazard r&s and confidence lewk

are listed in Table 2. Using multivar&c analysis ofthcae and other selected variables, independent predictors ~~~~Or~~~~~

WCFC defined and are listed in Table 3. ~~~~ct~~~~~ &fist ejection fraction, diqaosis during the earlier stt@ (197&49$1), and the presence of a third heart soMi’id wWe independent predictors of mortality. Hmportant!y age, sex,

1924 REDFIELD ET AL. NATURAL )s1[STORY OF BBKWATHIC DILATED CARDtOMYOPATMY

JACC Vol. 22, No. 7 eeembcs PWkI92P-6

nt with idiopathic dilated cardiomyopathy

a a

Figure I. Survival curves fur the four cohorts of parients with idiapath’ previously published previously reported population the ~9~~-~~$~ referral cob ral cohort, The numbers

Figure 2. Survival curves for the 19764981 versus 19824987 refwral cohorts by New York Heart Associ- ation (NYWA) functional class. The improved survival evident in the 19824P87 referral cohort was strongest for patients in functional classes I and II.

arm% natural histwy ,Pf A recenti study by Co:id

et al. (2) indicaled an increasing incidence of idiopathic dilated cardi0~~y~pat~y over two successive time periorls from 1975 to 1984. This finding was interpreted as indicating increasing recognitian and pathic dilated csrdiomys ~ch~card~Qgra~~y and r vide objective evidence af systolic dys with indeterminate symptoms, a~I~~i~g

thic dilated cardiomyopathy. ile ~ati~~lt§ in the 1982- 8‘7 cohort lived longer be- they were identified e er in the CCWS~ d their

disease, the fact that diagnosis in the 1976-1981 period was an in&q~~dent predictor of mortality indicates that other than early diagnosis are responsible f0r the im survival in the more recent cohort. ~rn~r~~~m~flts in treat-

TabIe 3. Independent Predictors of Mortality by Multivariate Analysis

P Hazard Confidence Chi-Square Value Ratio Limits

NY HA class” 17.63 < O.H!Ql 1.63 1.29-2.06 Ejection fraction? 13.74 0.0002 0.68 0.66-0.69 Diagnosis in 1976 to 1981 11.93 0.0006 I .92 I .32-2.30

S3 10.83 0.001 I.91 1.29-2.84

*Hazard ratio reflects risk with an increase of one functional class. tHazard ratio reflects risk with an increase of IO percentage points. Abbre- viations as in Tables I and 2.

24% 0f the I982-1198’7 referral co

rny~~at~y c0~d~~ctcd in a referral populatio r&ml bias on studies of the natural histor conditions is increasingly recognized (14,

in, part to the k&hood that patients

patients with a more fslvorable ~ro~~0~i examined in the currem study may be a ~~~~er pr~al~~~ce of

0ther diseases in the referral population. Most large studies 0f the natural history of i

ts im diagnosis and the

failure (&IO,1 1,12,23) provide US@

sis in patients with ischcmic sad nonisc

1926 REDFI.ELD ET AL. NATURAL HlSTORY OF IDIOPATHIC DILATED CARDIOMYOPATHY

cared for at referral institutions as part of a randomized trial, extrapolation of these data to patients with idiopathic dilated cdomyopathy not participating in clinical trials is not always appropriate, as suggested by the current study.

ts Extensive data concerning cl c data were recorded in each patient and were examined as possible predictors of mrtality. Using multivariate analysis, only era of diagnosis (being in the earlier referral cohort), functional class, ejec-

md the presence of a third heart sound were predictors of mortality. Despite the subjective ~~~t~o~~ classificwtion and the inherent limie

Not til p&Ms underwent coronary ~ngiog~~hy and some pWMs may have had occult significant coronary artery disease. Patients who did not undergo corosnstry

y were older, tended to be more symptomatic and signs of decompensated heti ftilurc than dad

those who underwent coronztry wertz corrected for the severity their survivd WBS simile to th

Thus, while one cannat CX-

those with idi* thy 6123, their similu prognosis

impressive that even though ct and subjective variable, it

of clinical course in idiopathic

I. Fuster Q. Gersh BJ. Giuliani ER, Tajik AJ, Brandenburg R6, Frye RL. The natural history of idiopathic dilated cardiomyopathy. Am J Cardio! 1981:47:525-31.

2. Codd MB, Sugrue DD, Gersh BJ. Melton LJ II!. ~~ide~~~~~~y of idiopathic dilated and hypertrophic cardiomyopathy; a ~~~la~i~n~b~s~~ study in Olmsred County, Minnesota, 197S-1984. Circulation 1989;80: W-72.

3. Sugrue DD, Rodeheffer RJ, Codd MB, Bulhud DJ, Fuster V, Gersh BJ. The clinical course of idiopat myopathy; a population- based study. Ann Intern Med

4, Goldman L, Hashimoto 3, bok EF, Los&o A. Com~n~tiv~ repr@ ducibility LI alidity of sys for assessing cardiovascular funct class: adva @S Of I new iRe activity se&. C~~~n~~~~~~n 198 1 m&%

5. Cox DR. Regression mndcls and lif&tnb!es. J R Stat Sot t 22u.

6, &IS hsrilutr Inc, Supplcmanrrcl L SCP’S Guide. Version 5 Edition. Gary, NC ~nst~tnt~; l9R6d

tier P. ~on~~r~rnet~~c estimation from incomplete obser- vations. J Am Stat Assw 1~~$~~3~4~7~$l~

8. The &wxws T&d severe congestive heu

9. Cohn JN, Johnson G, hyd~l~ine~i$oso~~id~ heurt failure. N En@ J

10, The SOLVD Iavesrigataw, reduced IeR v nlprr ejection fractions and congestive heart failure. N En@ J Med

Il. The SQLVD Investigators. Effect of enuh-qxil on ~~~~~~~t~ and the develo~nt~nt of heart ftilure ia asyn~~tont~t~c p&rats with reduced left ventricul~ ejection frnctions. M En81 J Med l~2;~27~~~-~l~

I%+ Nekr MA, B~nnwn~d B, MoyC LA, et nl., an Uehnlf of the SAVE Investigators. Effect of cnptapril on mortality and morbidity in pat

r dyflmction after myocnrdinl ~nfa~ti~n, N En@ J

13, vention of heart f&W$--a new agenda. N Engl 3 Med

14, lectian bias in the referral of patients and the natural history of surgical conditions. Mayo Clin Proc 1985;6WtW%.

15, Shapiro L&I, Zezulkip A. Hypertrophic cardiomyopathy: a ciommon disease with a proflosis. Five year experience of a district general hospital, Br Heart J 1%3;5053Q-3.

16. Fmnciosu JA, Wilen M, Ziesche St Cohn JN. Survivali in mee with severe chronic left ventricular failure due to either coronary heart disease or

c dilated cardiomyopathy. Am J Cardiol 1983;51:831-6. 17. A, Lanzarini L, CornaIbu C, et al. Dilated (congestive) cardio-

y: Follow~up study of I37 patients. G Ital Cardiol 1984;14:492-8. 18. kxz RA. Clbasohan A, Gakley CM. Prediction of outcome in dilated

~~diomyopathy. Br Heart J 1987:58:3X%-9. 19. ICa;bn H, Becker R, Fischer J, et ul. Studies on the etiology, the clinical

courseI and the proBncrsis of patients with dilated cardiomyopathy @CM)+ Z Kardiol 1982;7 I :497-508.

20. Fig& HR. Rahlf G. Nieger M, Luig H, Kreuzer H. Spontaneous hemodynamic improvement of stabilization and associated biopsy find- ,+;-; ill ~&ciUs with congestive cardiomyopathy. Circulation 1985;71: IQ%104.

21. Sshwarz F, Mull G, Zebe H, et al. Determinants of survival in patients with congestive cardiomyopathy: quantitative morphologic findings and left venlrieular hemodynumics. Circulation 1984;70:923-8.

22, Unverferth DV, Ma@xien RD, Moeschberger ML, Baker PB. Fetters JK, Feier CV. Factors influencing the one-year mortality of dilated cardiomy- oputhy. Am J Cardiol 1%4$4: 147-52.

23. Cohn JN, Archibald DG, Ziesche S, et al. Wect of vasodilator therapy on mortality in chronic congestive heart failure. N Engl I Med 1986;314: 1547-52.