Heart failure (HF) is a commonly

Transcription

1 ID:3731 O R I G I N A L P A P E R Outcomes and Prognostic Factors of Systolic as Compared With Diastolic Heart Failure in Urban America We sought to describe a large heart failure (HF) population with respect to systolic and diastolic abnormalities in terms of demographics, echocardiographic parameters, and survival. Using data abstracted from the Resource Utilization Among Congestive Heart Failure (REACH) study, a targeted subpopulation of 3471 patients had electrocardiographic, echocardiographic, and clinical data taken from automated sources during the first year of diagnosis. Among the HF population, 1811 (52.2%) had diastolic HF. Prevalence of diastolic HF trended with age, from 46.4% in those less than 45 years to 58.7% in those 85 years or older (p=0.001 for trend). Patients with diastolic HF had a higher mean ejection fraction (55.7% vs. 28.0%), lower left ventricular end-systolic diameter (3.11 vs cm), and lower left atrium: aortic outlet ratio (1.28 vs. 1.38) (p=0.001 for each comparison). Annualized age, sex, and race-adjusted mortality were 11.2% and 13.0% for those with diastolic and systolic HF, respectively (p=0.001). In a large, racially mixed, urban HF population, those with diastolic HF predominate and enjoy better-adjusted survival than counterparts with systolic HF. (CHF. 2005;11:6 11) 2005 CHF, Inc. Heart failure (HF) is a commonly encountered neurohormonal syndrome whose prevalence is increasing as the population ages and medical and surgical therapies improve. 1 Often, HF has been divided pathophysiologically into systolic and diastolic HF as each has independent etiologies, prevalence, treatments, and prognoses. 2 Information about diastolic HF is limited; until now, the largest study has included 623 patients. 3 Furthermore, information regarding minority populations is limited; few studies have specifically addressed blacks with HF, 4 8 despite the trends of higher prevalence and poorer outcomes compared with whites This investigation attempts to define the epidemiology of diastolic HF in a large, urban, academic setting in terms of age, race, and sex. Furthermore, while many studies have compared echocardiographic features of systolic and diastolic HF, 3 7,14 16 we sought to describe these parameters in an HF population derived from automatic sources of data and report survival rates in those patients with diastolic and systolic HF. Methods Setting. The target population was derived from the parent Resource Utilization Among Congestive Heart Failure (REACH) study 17 at Henry Ford Hospital previously described. Briefly, Henry Ford Hospital is a 903- bed tertiary/quaternary care center located in the Detroit metropolitan area and receives patients whose care is primarily within Henry Ford Health System, a vertically integrated, mixedmodel managed care organization that includes urban and suburban satellite clinics in Southeast Michigan. The parent study identified 29,686 outpatient and inpatient HF patients between the years of An index case of HF was based on: 1) at least two outpatient 9th International Classification of Disease Clinical Modification (ICD-9-CM) HF codes (428.XX, , , , , , , , , , , , , ) 18 ; or 2) one inpatient hospitalization under diagnosis related-group 127 or 124, and an ICD-9-CM HF code in the principal position. Hospital admissions received diagnosis-related group and ICD-9-CM codes from professional coders that Peter A. McCullough, MD, MPH; 1 Akshay K. Khandelwal, MD; 2 John E. McKinnon, MD; 2 Heather J. Shenkman, MD; 2 Vijayamalini Pampati, DO; 2 David Nori, MD; 2 Roberta A. Sullivan, BSN, MPH; 3 Keisha R. Sandberg, MPH; 1 Scott Kaatz, DO 2 From the Divisions of Cardiology, Nutrition and Preventive Medicine, William Beaumont Hospital, Royal Oak, MI; 1 Department of Internal Medicine, Henry Ford Hospital, Detroit, MI; 2 Departments of Basic Science and Internal Medicine, Cardiology Section, University of Missouri-Kansas City School of Medicine, Truman Medical Center, Kansas City, MO 3 Address for correspondence: Peter A. McCullough, MD, MPH, Chief, Division of Nutrition and Preventive Medicine, William Beaumont Hospital, 4949 Coolidge Highway, Royal Oak, MI pmc975@yahoo.com Manuscript received June 29, 2004; accepted July 21, systolic vs. diastolic HF in urban America january. february 2005

2 were subsequently approved by attending physicians. Outpatient encounters had ICD-9-CM codes assigned directly by the attending physician. The parent database provided incident cases of HF, demographic and echocardiography data, and clinical variables. The total number of health system patients during this 10-year period was 5,137,343. Data Management. This subset of the parent study consisted of 3471 inpatients and outpatients with an electronically-documented electrocardiograph stored on the Medical User Systems for Electrocardiography (Marquette Electronics, Rochester, MI) system and two-dimensional M-mode echo data, taken from electronically reported studies obtained during the first year of diagnosis. 19 All 3471 cases had ejection fraction (EF) assessed by echocardiography and calculated by Simpson s rule or visually estimated; the accuracy and reproducibility of this practice has been previously described An average value was obtained if the patient had more than one echocardiogram within the first year of diagnosis. Patients were classified as having diastolic HF based on the presence of HF and an EF of 45%. All-cause mortality as of December 31, 1999, was confirmed based on any death documented within a system facility, reported by State of Michigan Death registry tapes, or listed by the National Social Security Death Index. The Henry Ford Hospital Institutional Review Board granted permission for the study. Data Validation. A case validation was performed to check the positive predictive value of the index case definition of HF by ICD-9-CM coding. We took a 1% random sample of all REACH study subjects (n=271) simultaneously present in the studied population and performed chart abstraction at the time of diagnosis or first patient encounter with explicit criteria applied from the Framingham and First National Health and Nutrition Examination (NHANES-I) 23,24 definitions of congestive heart failure. An important caveat is that the Framingham and NHANES-I definitions were designed for disease-free populations to detect systolic HF. Two physicians, an internist and cardiologist, performed paper chart abstractions for confirmation of HF. Results of echocardiography were also reviewed and analyzed in a similar manner to the overall dataset derived from automatic sources. Statistical Analysis. Baseline characteristics of patients were established, including age, race, and sex, with standard deviation as appropriate. Echocardiography data for each subpopulation of HF are reported as means ± standard deviation where indicated and compared using one-way analysis of variance. Chi-square or analysis of variance was used to describe the age, sex, and race distribution within systolic and diastolic HF populations. Cox regression was performed to evaluate the independent effects of systolic or diastolic HF on race, age and sex-adjusted survival. Independent predictors of mortality (p value for entry=0.25, p value for retention=0.10) were evaluated using multiple logistic regression. Statistical significance was concluded for a two-sided p value Statistical significance was established at a nominal α<0.05 level. Analyses utilized the Statistical Package for Social Sciences for Windows, release 10.0 (SPSS Incorporated, 1999, Chicago, IL). Results Baseline Characteristics. The target population of 3471 patients was followed for an average of 32.4±30.1 months. The average index age was 66.1±14.5 years, with 1723 men (49.6%; 95% confidence interval [CI], 48.0% 51.3%), 1429 whites (41.2%; 95% CI, 40% 42%), 1925 blacks (55.5%; 95% CI, 54% 56%), and 52 other race (3.3%). Overall, there were 1798 deaths (51.8%; 95% CI, 50.1% 53.4%). Based on the specified criteria, 1811 patients (52.2%; 95% CI, 50.5% 53.8%) had diastolic HF. More than half the population was age 65 and above (Table I). The study population was similar to the parent population in terms of age, race, and sex. Those patients with echocardiographic data had longer follow-up times in the database compared with the parent population, 32.4±30.1 vs. 24.1±14.6 months, p< Validation Set Results. From the previously described validation set, 172/271 (63.5%) met the Framingham definition and 151/271 (55.7%) met the NHANES-I definition (score 3) of HF. Both of these standardized definitions have been developed for systolic HF in other datasets. Thus 200 subjects (73.8%) met at least one standardized definition of HF. The physician chart notes indicated HF as a diagnosis in 82.9%. Of those participants with confirmed HF by either the Framingham or NHANES-I definitions, 54.9% and 45.1% had diastolic and systolic HF, respectively. Of those participants who did not meet one of these definitions, the rates of diastolic and systolic dysfunction by echocardiography were 59.1% and 40.9%, respectively. Epidemiology of Systolic and Diastolic Heart Failure. In terms of sex distribution, 751 men (41.5%) and 1060 women (58.5%) had diastolic HF (Χ 2 =101, p=0.001). In terms of race distribution, 986 blacks (54.4%) and 773 whites (42.7%) had diastolic HF (Χ 2 =2.7, p=0.10). There were relatively more patients with diastolic HF within each succeeding age category (p=0.001 for trend) (Figure 1). Echocardiographic Parameters. As expected, based on the study definitions of systolic and diastolic HF, there was a significant difference in mean EF between the two HF subpopulations (Table II). Patients with diastolic HF had statistically significant lower left ventricular end systolic diameter (LVESD) and a lower left atrial to aortic outflow ratio when compared with their counterparts with systolic HF. The left atrial to aortic outflow ratio served as an index of relative left atrial enlargement with ratios >1.0 indicating left atrial enlargement. The septum to posterior free wall ratio of measured thickness was similar for those with systolic and diastolic HF (Table II). systolic vs. diastolic HF in urban America january. february

3 Percentage 60% 50% 40% 30% 20% 10% 0% <45 years (N=330) Figure 1. Distribution of diastolic heart failure according to age group (p=0.001 for trend). Number for each age subgroup is presented within the data bars. Mortality. There were 958 deaths (57.7% [95% CI] 55.3% 60.1%) among patients with systolic HF and 850 deaths (46.9% [95% CI] 44.1% 48.7%) among patients with diastolic HF (p=0.001). Patients with diastolic HF demonstrated better age, sex, and race-adjusted survival rates than those with systolic HF (Figure 2). The annualized age, sex, and race-adjusted mortality rate for the first 72 months was 11.2% for diastolic HF and 13.0% for systolic HF (p=0.001). Male sex and older age were significant covariates of diastolic HF mortality, while male sex, black race, and older age were significant covariates of systolic HF mortality (Table III). Of note, longer QRS duration was associated with higher mortality among those with diastolic and systolic HF, p=0.05 and p=0.01, respectively. Discussion The principal REACH study demonstrated that the HF epidemic has worsened throughout the 1990s 17 when years (N=1114) years (N=1780) Age category (total) years (N=247) ICD-9-CM codes are used for case identification. The present study indicates that in a large sample, with adequate representation of African Americans, that diastolic HF is more common than systolic HF. As prior studies have shown, the rates of diastolic HF increase with age. In addition, considerably more women (58.5%) have diastolic heart failure than men (41.5%). Furthermore, the mortality rate is lower for those with diastolic HF even after adjusting for age, sex, and race. Several features of this study distinguish it from previous ones. This study represents data on 3471 HF patients; specifically, 1811 patients with diastolic HF. The prevalence of diastolic HF within HF was 50% and similar to previous reports. 15,25 28 Over four epidemiological studies, the total number of patients with diastolic congestive heart failure and mortality data was 384 patients. 15,25 27 If only patients with HF and preserved EF are included from a recent study by Redfield et al., 29 the total number studied is just above 420. While our study is not epidemiological in the strictest sense, it does represent a population comprised of both sick and well individuals, all participants in an urban health care system that numbered more than 5 million over a 10-year period. Other results from this study are generally consistent with previous studies that have reported on diastolic HF. 15,25 26 Similar to our results, a predominance of women with diastolic HF was seen. 4,7,16,25,27,30,31 While our trend of increasing rates of diastolic HF with successive age groups was seen previously, 15,26,31 33 the magnitude of diastolic HF in various age categories of HF has not been reported, probably due to a smaller number of patients. This study also incorporated a large number of American blacks (1925) with HF, 55% of the study population. Studies have shown approximately similar racial distributions within diastolic HF, 4,6 7 although Kinney 5 described a population of two thirds whites without specifying the remainder. Only a second study by Ghali et al. 8 of 78 black patients with diastolic impairment, based on fractional shortening, fared better than those with systolic impairment. The current study includes representation from a large, American, urban-based, often understudied HF subpopulation, which proves crucial as the medical community struggles to define race in the understanding and treatment of HF As for echocardiographic parameters, most studies showed similarly smaller LVESD and left atrial measurement in those patients with diastolic HF, although many demonstrated a significantly greater incidence of ventricular hypertrophy as measured by posterior wall thickness. 3 6,14 15 Our study cannot address hypertrophy since it is not directly reported or derived from the variables available in their database. The study of black patients by Ghali et al. 7 again found significantly decreased LVESD and increased posterior wall thickness among subjects with preserved 8 systolic vs. diastolic HF in urban America january. february 2005

4 systolic function, although it showed no significant difference in left atrial dimension. In general, those with diastolic HF have less ventricular and atrial distention than those with systolic HF. While we do not have diagnostic echocardiographic data in this analysis, their findings are supportive of the contention that diastolic dysfunction was likely present in those patients with a preserved EF. Annualized adjusted mortality with diastolic HF (11.2%) was lower than systolic HF (13.0%) over 72 months. Importantly, these rates are higher than those often observed in clinical trials, which, because of selection bias, inherently recruit lower risk populations than community studies. Older age and male race were adverse predictors for diastolic HF mortality. This survival advantage found in diastolic HF patients is consistent with previously published reports, including the all-black study by Ghali et al, 8 even though annualized mortality in studies with similar mean ages varied significantly (3% 33%). 28,36 This outcome may reflect longer follow-up, greater number of patients, and the status of Henry Ford Hospital as an urban referral center. Senni and Redfield 37 reported that mortality between systolic and diastolic HF did not appear to differ significantly in contemporary studies, however, the studies that supported their contention had significantly older populations and much smaller sample sizes. Furthermore, they did not have studies with a large numbers of blacks. We found that black race was an independent predictor of mortality for those participants with systolic, but not diastolic, HF. The natural history of diastolic HF certainly deserves more study; however, we can conclude that, on average, patients with diastolic HF have a better prognosis than those with systolic HF. This study has multiple limitations. Misclassification bias of systolic and diastolic HF by a simple EF cutpoint may have occurred. This limitation could have been overcome by more complex Doppler evaluation, which was not done in this study. By excluding the possibility of combinations of systolic and diastolic HF, and by ignoring implications of altered hemodynamics (including effect of baseline medications) at the time of Survival Figure 2. Age, sex, and race-adjusted survival of systolic vs. diastolic heart failure. CHF=congestive heart failure echocardiography, we have underestimated the prevalence of diastolic HF. Characterizing systolic HF has been easier than diastolic HF, whose definitions range from various degrees of left ventricular filling patterns 37 to more recently proposed catheterization-derived left ventricular relaxation, filling, or distensibility indices We did not have other common echocardiographic variables including other chamber dimensions and measurements in the database that could have helped considerably. Our use of a simple EF cutpoint of 45% has precedence in the epidemiological literature, as it is practical when working with large databases. 28 There may have been another source of misclassification bias, that is, patients without HF, but who were coded as having HF and then getting into the database and falling into the p=0.001 Months Diastolic CHF Systolic CHF No. at risk Diastolic Systolic category of diastolic HF. Of patients who did not meet standardized definitions of HF or have HF in the chart notes, 59.1% had an EF of 45% while 40.9% had an EF of <45% suggesting that underlying HF was present but not the major clinical issue at the time. Importantly, when we analyzed subsets of those patients who met standardized definitions of HF, or those with or without portions of missing data, we consistently found that diastolic HF was predominant. The false-negative rate of the ICD-9-CM code strategy used in REACH is unknown, although in another study where ICD-9-CM code 428.XX was used alone the rate was 87.4%. 41 Hence, it is possible that if we had sampled the healthy population, we would have found higher rates of diastolic dysfunction without clinical HF, as was recently reported. 29 systolic vs. diastolic HF in urban America january. february

5 Table I. Baseline Characteristics of the Target Population and of the Parent Resource Utilization Among Congestive Heart Failure (REACH) Study TARGET POPULATION WITH ECHOCARDIOGRAPHIC DATA (N=3471) PARENT POPULATION FROM ICD-9-CM CODES (N=29,686) Average index age (yr) 66.1± ±14.5 Sex Male 1723 (49.6%) 14,060 (47.4%) Female 1748 (50.4%) 15,626 (52.6%) Race White 1429 (41.2%) 14,945 (50.3%) Black 1925 (55.5%) 13,121 (44.2%) Other 117 (3.4%) 1455 (4.9%) Average follow-up (mo) 32.4± ±14.6 ICD-9-CM=International Classification of Disease, Ninth Revision, Clinical Modification Table II. Echocardiographic Parameters of Systolic Dysfunction and Diastolic Heart Failure (HF) Subpopulations ECHOCARDIOGRAPHIC PARAMETERS SYSTOLIC HF (N=1660) DIASTOLIC HF (N=1811) p VALUE Left ventricular ejection fraction (%) 28.0± ± Left ventricular end-systolic dimension (cm)* 4.74± ± Left atrium:aortic outflow tract ratio 1.38± ± Septum:posterior free wall ratio 1.04± ± *Left ventricular end-systolic diameter was only obtainable on 1326 patients. Table III. Significant Predictors of Mortality of Systolic and Diastolic Heart Failure (HF) DIASTOLIC HF SYSTOLIC HF OR 95% CI (p VALUE) OR 95% CI (p VALUE) Age (each year) (<0.0001) (<0.0001) Male (0.03) (0.04) Black race (0.51) (0.06) Ejection fraction (%) (0.79) (0.08) QRS Duration (each ms) (0.05) (0.01) Variables excluded from the model due to missing data were left ventricular end-systolic dimension and left atrial/aortic ratio. OR=odds ratio; CI=confidence interval Conclusions In this American urban setting, diastolic HF constituted slightly over 50% of HF cases. Determinants appear to be older age and female gender, but not race. There are significant echocardiographic differences that separate diastolic HF from systolic HF, including smaller left atrial and ventricular chamber dimensions. Patients with diastolic HF had better age, sex, and race-adjusted survival rates when compared with those with systolic HF. Our results indicate that diastolic HF has a different natural history than systolic HF; focused studies on the determinants, prevention, and treatment of diastolic HF are warranted. REFERENCES 1 Cowie MR, Mosterd A, Wood DW, et al. The epidemiology of heart failure. Eur Heart J. 1997;18: Gaasch WH. Diagnosis and treatment of heart failure based on left ventricular systolic or diastolic dysfunction. JAMA. 1994;271: Cohn JN, Johnson G. Heart failure with normal ejection fraction: The V-HEFT study. Veterans Administrative Cooperative Study Group. Circulation. 1990;81(2 suppl):iii-48 III Dougherty AH, Naccarelli GV, Gray EL, et al. Congestive heart failure with normal systolic function. Am J Cardiol. 1984;54: Kinney EL, Wright RJ. Survival in patients with heart failure and normal basal systolic wall motion. Angiology. 1989;40: Aguirre FV, Pearson AC, Lewen MK, et al. Usefulness of Doppler echocardiography in the diagnosis of congestive heart failure. Am J Cardiol. 1989;63: Ghali JK, Kadakia S, Cooper RS, et al. Bedside diagnosis of preserved versus impaired left ventricular systolic function in heart failure. Am J Cardiol. 1991;67: Ghali JK, Kadakia S, Bhatt A, et al. Survival of heart failure patients with preserved versus impaired systolic function: the prognostication implication of blood pressure. Am Heart J. 1992;123: Yusuf S, Thom T, Abbott RD. Changes in hypertension treatment and in congestive heart failure mortality in the United States. Hypertension. 1989;13(5 suppl):i-74 I Alexander M, Grumbach K, Selby J, et al. Hospitalization for congestive heart failure. Explaining racial differences. JAMA. 1995;274: May DS, Kelly JJ, Mendlein JM, et al. Surveillance of major causes of hospitalization among the elderly, MMWR CDC Surveill Summ. 1991;40(1): Gillum RF. Epidemiology of heart failure in the United States. Am Heart J. 1993;126: Ghali J, Cooper R, Ford E. Trends in 10 systolic vs. diastolic HF in urban America january. february 2005

6 hospitalization rates for heart failure in the United States, Arch Intern Med. 1990;150: Andersson B, Caidahl K, Waagstein F. An echocardiographic evaluation of patients with idiopathic heart failure. Chest. 1995;107: Kitzman DW, Gardin JM, Gottdiener JS, et al. Importance of heart failure with preserved systolic function in patients 65 years of age. Am J Cardiol. 2001;87: Echeverria HH, Bilsker HS, Myerburg RJ, et al. Congestive heart failure: echocardiographic insights. Am J Med. 1983;75: McCullough PA, Philbin EF, Spertus JA, et al. Confirmation of a heart failure epidemic: findings from the Resource Utilization Among Congestive Heart Failure (REACH) study. J Am Coll Cardiol. 2002;39: U.S. Health Care Financing Administration. Diagnosis Related Groups Definitions Manual. Wallingford, Conn: 3M Health Information Services; Shenkman HJ, Pampati V, Khandelwal AK, et al. Congestive heart failure and QRS duration: establishing prognosis (CONQUEST) study. Chest. 2002;122: Choy AM, Darbar D, Lang CC, et al. Detection of left ventricular dysfunction after acute myocardial infarction: comparison of clinical, echocardiographic, and neurohormonal methods. Br Heart J. 1994;72: Amico AF, Lichtenberg GS, Reisner SA, et al. Superiority of visual versus computerized echocardiographic estimation of radionuclide left ventricular ejection fraction. Am Heart J. 1989;118(6): Van Royen N, Jaffe CC, Krumholz HM, et al. Comparison and reproducibility of visual echocardiographic and quantitative radionuclide left ventricular ejection fractions. Am J Cardiol. 1996;77: Ho KK, Pinsky JL, Kannel WB, et al. The epidemiology of heart failure: the Framingham Study. J Am Coll Cardiol. 1993;22:6A 13A. 24 Schocken DD, Arrieta MI, Leaverton PE, et al. Prevalence and mortality rate of congestive heart failure in the United States. J Am Coll Cardiol. 1992;20(2): Vasan RS, Larson MG, Benjamin EJ, et al. Congestive heart failure in subjects with normal versus reduced left ventricular ejection fraction. J Am Coll Cardiol. 1999;33: Senni M, Tribouilloy CM, Rodeheffer RJ, et al. Congestive heart failure in the community: a study of all incident cases in Olmsted County, Minnesota, in Circulation. 1998;98: Devereux RB, Roman MJ, Liu JE, et al. Congestive heart failure despite normal left ventricular systolic function in a populationbased sample: the Strong Heart Study. Am J Cardiol. 2000;86: Vasan RS, Benjamin EJ, Levy D. Prevalence, clinical features, and prognosis of diastolic heart failure: an epidemiological perspective. J Am Coll Cardiol. 1995;26: Redfield MM, Jacobsen SJ, Burnett JC, et al. Burden of systolic and diastolic ventricular dysfunction in the community. JAMA. 2003;289: Soufer R, Wohlgelernter D, Vita NA, et al. Intact systolic left ventricular function in clinical congestive heart failure. Am J Cardiol. 1985;55: Wong WF, Gold S, Fukuyama O, et al. Diastolic dysfunction in elderly patients with congestive heart failure. Am J Cardiol. 1989;63: Diller PM, Smucker DR, Graham RJ. Congestive heart failure due to diastolic or systolic dysfunction. Frequency and patient characteristics in an ambulatory setting. Arch Fam Med. 1999;8: Takarada A, Kurogane H, Minamiji K, et al. Congestive heart failure in the elderly echocardiographic insights. Jpn Circ J. 1992;56: Wood AJ. Racial differences in the response to drugs pointers to genetic differences. N Engl J Med. 2001;344: Schwartz RS. Racial profiling in medical research. N Engl J Med. 2001;344: Mandinov L, Eberli FR, Seiler C, et al. Diastolic heart failure. Cardiovasc Res. 2000;45: Senni M, Redfield MM. Heart failure with preserved systolic function: a different natural history? J Am Coll Cardiol. 2001;38: Vasan RS, Levy D. Defining diastolic heart failure: a call for standardized diagnostic criteria. Circulation. 2000;101: Vasan RS, Benjamin EJ. Diastolic heart failure no time to relax. N Engl J Med. 2001;344: Zile MR, Gaasch WH, Carroll JD, et al. Heart failure with a normal ejection fraction: is measurement of diastolic function necessary to make the diagnosis of diastolic heart failure? Circulation. 2001;104: Goff DC, Pandey DK, Chan FA, et al. Congestive heart failure in the United States: is there more than meets the (ICD) code? The Corpus Christi Heart Project. Arch Intern Med. 2000;160: systolic vs. diastolic HF in urban America january. february