- Email: [email protected]
Life expectancy of elderly and very elderly patients with chronic heart failure
Life expectancy of elderly and very elderly patients with chronic heart failure Andrew Owen, FRCP, PhD Canterbury, UK
Introduction
The survival of patients with chronic heart failure is typically reported as a comparison of different groups of patients using the hazard ratio from a Cox proportional hazards analysis. The absolute survival is generally neglected. Furthermore, attention is often focused on relatively young patients although chronic heart failure largely affects older patients. The present study was undertaken to determine the life expectancy (a measure of absolute survival) of older patients with chronic heart failure.
Methods Patients N75 years with chronic heart failure caused by impaired left ventricular systolic function who attended an outpatient clinic were included in the study. Follow-up commenced on August 1, 1993, and continued until September 30, 2005, when vital status was ascertained. Mean survival time was calculated as a measure of life expectancy. Results There were 210 patients included in the study. Male patients of mean age 80 years had a life expectancy of 3.9 years (95% CI 3.2-4.5), compared with that of 7 years for men in the general population of the same age. For female patients of mean age 80 years, the life expectancy was 4.5 years (95% CI 3.6-5.7), compared with 8.5 years for the general population of women of the same age. Conclusion The presence of chronic heart failure in older patients results in an approximately 50% reduction in life expectancy. (Am Heart J 2006;151:1322.e121322.e4.) Chronic heart failure is a condition that largely affects older persons. The median age at its presentation is 75 years,1 and most patients are N65 years.2 Clinical trials in heart failure have largely excluded patients N75 years or have only included small numbers. Our knowledge of the survival experience of older patients with heart failure has therefore been derived from observational studies. These have included patients after admission to hospital with heart failure3 - 6 and those from a heart failure clinic in which admission to hospital was not an inclusion criterion.7 Other reports8-10 have been based on observations of communities. In all these studies, patients who do not have reduced left ventricular systolic function or who may not even have had an assessment of left ventricular function at all have been included. For patients without impaired left ventricular systolic function, the left ventricular abnormality responsible for their symptoms and signs is invariably not reported. Many patients with the signs and symptoms of heart failure have noncardiac causes (eg, obesity or lung disease) as the underlying pathology,11 hence the importance of
From the Department of Cardiology, Kent and Canterbury Hospital, Canterbury, UK. Submitted November 6, 2005; revised , ; accepted March 20, 2006. Reprint requests: Andrew Owen, FRCP, PhD, Department of Cardiology, Kent and Canterbury Hospital, Ethelbert Rd, CT1 4NG Canterbury, UK. 0002-8703/$ - see front matter n 2006, Mosby, Inc. All rights reserved. doi:10.1016/j.ahj.2006.03.017
establishing a cardiac abnormality in addition to the presence of the signs and symptoms of heart failure as recommended by contemporary guidelines.12 Thus, it is not clear as to what extent such studies have included patients who do not have heart failure. This uncertainty makes it difficult to interpret the reported findings. Reports of the survival of patients with chronic heart failure typically have limited follow-up, making it impossible to estimate life expectancy. Prolonged followup has been reported10 with small numbers of patients, but the diagnosis was based on clinical evaluation without echocardiography, making it impossible to determine the extent to which patients without left ventricular dysfunction were included. Life expectancy was not reported. The emphasis has typically been on the comparison of the survival of different groups using the hazard ratio derived from a Cox proportional hazards analysis rather than on the life expectancy per se of the groups. Life expectancy at a given age is the additional years of life expected (mean extra years of life). The purposes of the present study were to estimate the life expectancy of older patients with chronic heart failure and to compare this with that of the general population of the same age.
Methods This was a retrospective study on data generated during the course of routine clinical care. Patients were included if
American Heart Journal June 2006
1322.e2 Owen
they had attended an outpatient clinic at the Kent and Canterbury Hospital with the signs and symptoms of heart failure, had had an echocardiogram demonstrating reduced left ventricular systolic function (ejection fraction b40%), and were at least 75 years old at the time of presentation. Patients with primary valvular disease were not included. Patients were drawn from the city of Canterbury and the surrounding area. This is a relatively affluent area and a popular retirement destination, 20% of the population is N65 years. All patients were registered with a general practitioner in this area. Follow-up commenced on August 1, 1993, and continued until September 30, 2005. Patient inclusion ceased on December 31, 2000. Vital status was determined on September 30, 2005, by interrogation of the hospital activity system. All patients who are registered with a local general practitioner at the time of death have the date of death registered on this system. In addition, all outpatient attendances are logged. Patients who had no date of death were allocated a censored time corresponding to the last date they were known to be alive. Patients who met the inclusion criteria were identified by reviewing a register of echocardiograms. This register provided limited patient information, the indication for the echocardiogram, and a summary of the findings. Patients were included if the indication for their echocardiogram was clinically suspected heart failure (signs and symptoms) and the echocardiogram summary indicated impaired left ventricular systolic function. It was not possible to obtain the full medical records of all patients, many of whom had died many years previously. Thus, the study was based on the limited patient information available from the register. The data were divided into male and female groups and age groups of between 75 and 85 years (henceforth referred to as the elderly group) and of z85 years (henceforth referred to as the very elderly group) to facilitate the comparison with published life expectancy for elderly subjects in the general population. This investigation conforms to the principles outlined in the Declaration of Helsinki and was approved by the local research ethics committee.
Statistical analysis Life expectancy was estimated as the mean survival time and is represented by the area under the Kaplan-Meier survival curve. This was determined to be the longest event time. The sample mean survival time is asymptotically normally distributed, allowing us to use the standard normal distribution to obtain approximate 95% CIs.13
Results All patients who could be identified as fulfilling the inclusion criteria were included. Data were available for 210 patients, 124 (59%) of whom were male, giving a male-to-female ratio of approximately 3:2. The KaplanMeier estimated survival functions for the elderly and very elderly patients are plotted in Figure 1. The survival of the very elderly patients is substantially worse than that for the elderly patients (log rank test P = .0003), as would be expected.
Figure 1
Kaplan-Meier survival curves for elderly (continuous line) and very elderly (broken line) patients with chronic heart failure.
For the elderly group, 122 deaths were observed, leaving 42 patients with censored survival times. For the very elderly group, 40 deaths were observed, leaving 6 patients with censored survival times. For both groups, the survival curves declined to zero because the longest period of follow-up happens to relate to patients who have an observed death time rather than a censored survival time. The life expectancy and the median survival time for the 4 groups are shown in Table I. The life expectancy for the very elderly patients is less than that for the elderly patients, as expected, but the small number of patients leads to wide CIs. Table I also shows the life expectancy and median survival for the elderly and very elderly patients, with male and female patients combined. The life expectancy for the very elderly patients is some 2 years less than that for the elderly patients (t test, P b .0001). The approximate life expectancy over the period of the study of male subjects of age 80 years in the general population in England is 7 years; that for female subjects, 8.5 years.14 We therefore see that the presence of chronic heart failure in elderly patients leads to an approximately 50% reduction in life expectancy. No information is available for the life expectancy of the general population of very elderly subjects.
Discussion This study examined the mortality experience of 210 patients N75 years with chronic heart failure caused by impaired left ventricular systolic function. The maleto-female ratio was approximately 3:2. The male-tofemale ratio for individuals of this age group in the Canterbury area is approximately 2:3.15 If we assume that this
American Heart Journal Volume 151, Number 6
Owen 1322.e3
Table I. Life expectancy and median survival of elderly and very elderly patients with chronic heart failure
Male elderly patients Male very elderly patients Female elderly patients Female very elderly patients All elderly patients All very elderly patients
No. of patients
Mean age at diagnosis (y)
102 22 62 24 164 46
79.8 88.8 80.6 88.9 80.2 88.9
ratio has not greatly changed over the previous decade and that there was no sex referral bias of patients with chronic heart failure, then the male-to-female incidence ratio is approximately 2:1. There is a general anecdotal perception that chronic heart failure in older patients is more common in women, which is not supported by the current study. This perception, at least in part, arises from considering patients with the signs and symptoms of heart failure but without left ventricular systolic dysfunction to have heart failure. Such patients are more likely to be female.3 It is clearly important to be precise and not combine groups of patients that may have different male-to-female ratios, a different underlying pathology, and a different prognosis. A preponderance of male to female patients has also been found in a study on hospital admissions of elderly patients with heart failure.16 Studies examining the survival of patients with chronic heart failure tend to concentrate on the relative survival between various groups. The findings are typically presented in terms of the hazard ratio determined from a Cox proportional hazards analysis. In the present study, the absolute survival of older patients with chronic heart failure in terms of life expectancy, with up to 12 years of follow-up, was examined. The findings suggest that the presence of chronic heart failure leads to an approximately 50% reduction in life expectancy, as compared with that for a general population of elderly subjects. Very elderly patients have a particularly poor prognosis, with a median survival of only 1 year and a life expectancy of 2.2 years, approximately half that of the elderly patients. The point estimates of life expectancy given in Table I might suggest that the life expectancy for female elderly patients is better than that for male elderly patients. Conversely, for very elderly patients, the life expectancy appears to be better for male patients. The confidence limits, however, are wide and overlap, so it is not reasonable to make this deduction. This should be considered a hypothesis-generating observation that would require a much larger study to be tested. For younger patients with chronic heart failure caused by left ventricular systolic dysfunction, previous studies have not suggested a difference in survival between men and women.3,5,6,15
Life expectancy and 95% CI (y) 3.9 2.6 4.5 1.9 4.2 2.2
(3.2-4.5) (1.4-3.9) (3.6-5.6) (1.0-2.5) (3.7-4.8) (1.5-3.0)
Median survival (y) 3.3 1.0 4.9 0.9 3.5 1.0
The life expectancy of heart failure patients was compared with that of the general population of England. This population includes well subjects, patients with other diseases, and patients with heart failure (diagnosed and undiagnosed). This life expectancy has increased over the years of the study. Thus, in 1991, the life expectancy of men and that of women aged 80 years in the general population were 6.4 and 8.3 years, respectively. In 2002, these had increased to 7.2 and 8.7 years, respectively.14 Approximate life expectation rates for men and women of 7 and 8.5 years, respectively, were used to represent the life expectancy pertaining during the period of study.
Study limitations The prolonged follow-up necessary to enable a reasonable estimate of life expectancy to be made admits the possibility that the natural history of the condition may have changed over time (eg, the use of more effective treatments such as h-blockers in more recent times). Nevertheless, the findings presented provide the best estimate of life expectancy currently available for real world elderly and very elderly patients with chronic heart failure. It was not possible to obtain complete medical records for most patients who had died. It was therefore not possible to carry out any form of regression analysis to identify prognostic factors. To undertake such an analysis using the records that could be obtained would lead to bias (eg, in favor of those patients who were still alive) and would therefore be of no value. The patients studied were not randomly recruited but were referred by their general practitioner to an outpatient clinic. It is therefore likely that they may have had a more advanced disease than those not referred and therefore not included. Conversely, general practitioners would have been unlikely to have referred very sick patients in nursing homes who would have been more likely to have had a more advanced disease. The nonrandom patient selection is clearly a limitation of the study. The patient group studied is representative of older patients with heart failure caused by left ventricular systolic dysfunction who are referred to hospital outpatient clinics. Population studies on somewhat younger
1322.e4 Owen
patients have been reported, but these have often not used echocardiography to confirm the diagnosis and often excluded residents of care homes.8,10 Different approaches have different limitations. It is felt that the approach taken here complements previous studies that have taken different approaches.
References 1. Purcell IF, Poole-Wilson PA. Heart failure: why and how to define it? Eur J Heart Fail 1999;1:7 - 10. 2. Senni M, Tribouilloy CM, Rodeheffer RJ. Congestive heart failure in the community, a study of all incidence cases in Olmstead County, Minnesota, in 1991. Circulation 1998;98:2282 - 9. 3. Kerzner R, Gage BF, Freedland KE, et al. Predictors of mortality in younger and older patients with heart failure and preserved or reduced left ventricular ejection fraction. Am Heart J 2003;146:286 - 90. 4. Feinglass J, Martin GJ, Lin E, et al. Is heart failure survival improving? Evidence from 2323 elderly patients hospitalised between 1989-2000. Am Heart J 2003;146:111 - 4. 5. Vaccarino V, Chen Y, Wang Y, et al. Sex differences in the clinical care and outcomes of congestive heart failure in the elderly. Am Heart J 1999;138:835 - 42. 6. Cowie MR, Wood DA, Coates AJS, et al. Survival of patients with a new diagnosis of heart failure: a population based study. Heart 2000;83:505 - 10.
American Heart Journal June 2006
7. Cicoira M, Davos CH, Florea V, et al. Chronic heart failure in the very elderly: clinical status, survival and prognostic factors in 188 patients more than 70 years old. Am Heart J 2001;142:174 - 80. 8. Mosterd A, Cost B, Hoes AW, et al. The prognosis of heart failure in the general population. The Rotterdam Study. Eur Heart J 2001; 22:1318 - 27. 9. Levy D, Kenchaiah S, Larson MG, et al. Longterm trends in the incidence and survival with heart failure. N Engl J Med 2002; 347:1397 - 402. 10. Nielsen OW, Hilden J, McDonagh T, et al. Survival differences between heart failure in general practices and in hospitals. Heart 2003;89:1298 - 302. 11. Cowie MR, Struthers AD, Wood DA, et al. Value of natriuretic peptides in assessment of patients with possible new heart failure in primary care. Lancet 1997;350:1349 - 53. 12. Remme WJ, Swedberg K, Task Force for the Diagnosis and Treatment of Chronic Heart Failure K, European Society of Cardiology. Guidelines for the diagnosis and treatment of chronic heart failure. Eur Heart J 2001;22:1527 - 60. 13. Hosmer DW, Lemeshow S. Applied survival analysis. New York: J Wiley; 1999. 14. www.statistics.gov.uk/statBase/Expodata/spreadsheets/d9026.xls [Accessed 16 October 2005]. 15. www.publications.doh.gov.uk/stats/population/ads2002-residentpopulations.xls [Accessed 16 October 2005]. 16. Cohen N, Ilgiyaev E, Almoznino-Sarafian D, et al. Sex related bedside clinical variables associated with survival of older inpatients with heart failure. Eur J Heart Fail 2004;6:781 - 6.
Introduction
The survival of patients with chronic heart failure is typically reported as a comparison of different groups of patients using the hazard ratio from a Cox proportional hazards analysis. The absolute survival is generally neglected. Furthermore, attention is often focused on relatively young patients although chronic heart failure largely affects older patients. The present study was undertaken to determine the life expectancy (a measure of absolute survival) of older patients with chronic heart failure.
Methods Patients N75 years with chronic heart failure caused by impaired left ventricular systolic function who attended an outpatient clinic were included in the study. Follow-up commenced on August 1, 1993, and continued until September 30, 2005, when vital status was ascertained. Mean survival time was calculated as a measure of life expectancy. Results There were 210 patients included in the study. Male patients of mean age 80 years had a life expectancy of 3.9 years (95% CI 3.2-4.5), compared with that of 7 years for men in the general population of the same age. For female patients of mean age 80 years, the life expectancy was 4.5 years (95% CI 3.6-5.7), compared with 8.5 years for the general population of women of the same age. Conclusion The presence of chronic heart failure in older patients results in an approximately 50% reduction in life expectancy. (Am Heart J 2006;151:1322.e121322.e4.) Chronic heart failure is a condition that largely affects older persons. The median age at its presentation is 75 years,1 and most patients are N65 years.2 Clinical trials in heart failure have largely excluded patients N75 years or have only included small numbers. Our knowledge of the survival experience of older patients with heart failure has therefore been derived from observational studies. These have included patients after admission to hospital with heart failure3 - 6 and those from a heart failure clinic in which admission to hospital was not an inclusion criterion.7 Other reports8-10 have been based on observations of communities. In all these studies, patients who do not have reduced left ventricular systolic function or who may not even have had an assessment of left ventricular function at all have been included. For patients without impaired left ventricular systolic function, the left ventricular abnormality responsible for their symptoms and signs is invariably not reported. Many patients with the signs and symptoms of heart failure have noncardiac causes (eg, obesity or lung disease) as the underlying pathology,11 hence the importance of
From the Department of Cardiology, Kent and Canterbury Hospital, Canterbury, UK. Submitted November 6, 2005; revised , ; accepted March 20, 2006. Reprint requests: Andrew Owen, FRCP, PhD, Department of Cardiology, Kent and Canterbury Hospital, Ethelbert Rd, CT1 4NG Canterbury, UK. 0002-8703/$ - see front matter n 2006, Mosby, Inc. All rights reserved. doi:10.1016/j.ahj.2006.03.017
establishing a cardiac abnormality in addition to the presence of the signs and symptoms of heart failure as recommended by contemporary guidelines.12 Thus, it is not clear as to what extent such studies have included patients who do not have heart failure. This uncertainty makes it difficult to interpret the reported findings. Reports of the survival of patients with chronic heart failure typically have limited follow-up, making it impossible to estimate life expectancy. Prolonged followup has been reported10 with small numbers of patients, but the diagnosis was based on clinical evaluation without echocardiography, making it impossible to determine the extent to which patients without left ventricular dysfunction were included. Life expectancy was not reported. The emphasis has typically been on the comparison of the survival of different groups using the hazard ratio derived from a Cox proportional hazards analysis rather than on the life expectancy per se of the groups. Life expectancy at a given age is the additional years of life expected (mean extra years of life). The purposes of the present study were to estimate the life expectancy of older patients with chronic heart failure and to compare this with that of the general population of the same age.
Methods This was a retrospective study on data generated during the course of routine clinical care. Patients were included if
American Heart Journal June 2006
1322.e2 Owen
they had attended an outpatient clinic at the Kent and Canterbury Hospital with the signs and symptoms of heart failure, had had an echocardiogram demonstrating reduced left ventricular systolic function (ejection fraction b40%), and were at least 75 years old at the time of presentation. Patients with primary valvular disease were not included. Patients were drawn from the city of Canterbury and the surrounding area. This is a relatively affluent area and a popular retirement destination, 20% of the population is N65 years. All patients were registered with a general practitioner in this area. Follow-up commenced on August 1, 1993, and continued until September 30, 2005. Patient inclusion ceased on December 31, 2000. Vital status was determined on September 30, 2005, by interrogation of the hospital activity system. All patients who are registered with a local general practitioner at the time of death have the date of death registered on this system. In addition, all outpatient attendances are logged. Patients who had no date of death were allocated a censored time corresponding to the last date they were known to be alive. Patients who met the inclusion criteria were identified by reviewing a register of echocardiograms. This register provided limited patient information, the indication for the echocardiogram, and a summary of the findings. Patients were included if the indication for their echocardiogram was clinically suspected heart failure (signs and symptoms) and the echocardiogram summary indicated impaired left ventricular systolic function. It was not possible to obtain the full medical records of all patients, many of whom had died many years previously. Thus, the study was based on the limited patient information available from the register. The data were divided into male and female groups and age groups of between 75 and 85 years (henceforth referred to as the elderly group) and of z85 years (henceforth referred to as the very elderly group) to facilitate the comparison with published life expectancy for elderly subjects in the general population. This investigation conforms to the principles outlined in the Declaration of Helsinki and was approved by the local research ethics committee.
Statistical analysis Life expectancy was estimated as the mean survival time and is represented by the area under the Kaplan-Meier survival curve. This was determined to be the longest event time. The sample mean survival time is asymptotically normally distributed, allowing us to use the standard normal distribution to obtain approximate 95% CIs.13
Results All patients who could be identified as fulfilling the inclusion criteria were included. Data were available for 210 patients, 124 (59%) of whom were male, giving a male-to-female ratio of approximately 3:2. The KaplanMeier estimated survival functions for the elderly and very elderly patients are plotted in Figure 1. The survival of the very elderly patients is substantially worse than that for the elderly patients (log rank test P = .0003), as would be expected.
Figure 1
Kaplan-Meier survival curves for elderly (continuous line) and very elderly (broken line) patients with chronic heart failure.
For the elderly group, 122 deaths were observed, leaving 42 patients with censored survival times. For the very elderly group, 40 deaths were observed, leaving 6 patients with censored survival times. For both groups, the survival curves declined to zero because the longest period of follow-up happens to relate to patients who have an observed death time rather than a censored survival time. The life expectancy and the median survival time for the 4 groups are shown in Table I. The life expectancy for the very elderly patients is less than that for the elderly patients, as expected, but the small number of patients leads to wide CIs. Table I also shows the life expectancy and median survival for the elderly and very elderly patients, with male and female patients combined. The life expectancy for the very elderly patients is some 2 years less than that for the elderly patients (t test, P b .0001). The approximate life expectancy over the period of the study of male subjects of age 80 years in the general population in England is 7 years; that for female subjects, 8.5 years.14 We therefore see that the presence of chronic heart failure in elderly patients leads to an approximately 50% reduction in life expectancy. No information is available for the life expectancy of the general population of very elderly subjects.
Discussion This study examined the mortality experience of 210 patients N75 years with chronic heart failure caused by impaired left ventricular systolic function. The maleto-female ratio was approximately 3:2. The male-tofemale ratio for individuals of this age group in the Canterbury area is approximately 2:3.15 If we assume that this
American Heart Journal Volume 151, Number 6
Owen 1322.e3
Table I. Life expectancy and median survival of elderly and very elderly patients with chronic heart failure
Male elderly patients Male very elderly patients Female elderly patients Female very elderly patients All elderly patients All very elderly patients
No. of patients
Mean age at diagnosis (y)
102 22 62 24 164 46
79.8 88.8 80.6 88.9 80.2 88.9
ratio has not greatly changed over the previous decade and that there was no sex referral bias of patients with chronic heart failure, then the male-to-female incidence ratio is approximately 2:1. There is a general anecdotal perception that chronic heart failure in older patients is more common in women, which is not supported by the current study. This perception, at least in part, arises from considering patients with the signs and symptoms of heart failure but without left ventricular systolic dysfunction to have heart failure. Such patients are more likely to be female.3 It is clearly important to be precise and not combine groups of patients that may have different male-to-female ratios, a different underlying pathology, and a different prognosis. A preponderance of male to female patients has also been found in a study on hospital admissions of elderly patients with heart failure.16 Studies examining the survival of patients with chronic heart failure tend to concentrate on the relative survival between various groups. The findings are typically presented in terms of the hazard ratio determined from a Cox proportional hazards analysis. In the present study, the absolute survival of older patients with chronic heart failure in terms of life expectancy, with up to 12 years of follow-up, was examined. The findings suggest that the presence of chronic heart failure leads to an approximately 50% reduction in life expectancy, as compared with that for a general population of elderly subjects. Very elderly patients have a particularly poor prognosis, with a median survival of only 1 year and a life expectancy of 2.2 years, approximately half that of the elderly patients. The point estimates of life expectancy given in Table I might suggest that the life expectancy for female elderly patients is better than that for male elderly patients. Conversely, for very elderly patients, the life expectancy appears to be better for male patients. The confidence limits, however, are wide and overlap, so it is not reasonable to make this deduction. This should be considered a hypothesis-generating observation that would require a much larger study to be tested. For younger patients with chronic heart failure caused by left ventricular systolic dysfunction, previous studies have not suggested a difference in survival between men and women.3,5,6,15
Life expectancy and 95% CI (y) 3.9 2.6 4.5 1.9 4.2 2.2
(3.2-4.5) (1.4-3.9) (3.6-5.6) (1.0-2.5) (3.7-4.8) (1.5-3.0)
Median survival (y) 3.3 1.0 4.9 0.9 3.5 1.0
The life expectancy of heart failure patients was compared with that of the general population of England. This population includes well subjects, patients with other diseases, and patients with heart failure (diagnosed and undiagnosed). This life expectancy has increased over the years of the study. Thus, in 1991, the life expectancy of men and that of women aged 80 years in the general population were 6.4 and 8.3 years, respectively. In 2002, these had increased to 7.2 and 8.7 years, respectively.14 Approximate life expectation rates for men and women of 7 and 8.5 years, respectively, were used to represent the life expectancy pertaining during the period of study.
Study limitations The prolonged follow-up necessary to enable a reasonable estimate of life expectancy to be made admits the possibility that the natural history of the condition may have changed over time (eg, the use of more effective treatments such as h-blockers in more recent times). Nevertheless, the findings presented provide the best estimate of life expectancy currently available for real world elderly and very elderly patients with chronic heart failure. It was not possible to obtain complete medical records for most patients who had died. It was therefore not possible to carry out any form of regression analysis to identify prognostic factors. To undertake such an analysis using the records that could be obtained would lead to bias (eg, in favor of those patients who were still alive) and would therefore be of no value. The patients studied were not randomly recruited but were referred by their general practitioner to an outpatient clinic. It is therefore likely that they may have had a more advanced disease than those not referred and therefore not included. Conversely, general practitioners would have been unlikely to have referred very sick patients in nursing homes who would have been more likely to have had a more advanced disease. The nonrandom patient selection is clearly a limitation of the study. The patient group studied is representative of older patients with heart failure caused by left ventricular systolic dysfunction who are referred to hospital outpatient clinics. Population studies on somewhat younger
1322.e4 Owen
patients have been reported, but these have often not used echocardiography to confirm the diagnosis and often excluded residents of care homes.8,10 Different approaches have different limitations. It is felt that the approach taken here complements previous studies that have taken different approaches.
References 1. Purcell IF, Poole-Wilson PA. Heart failure: why and how to define it? Eur J Heart Fail 1999;1:7 - 10. 2. Senni M, Tribouilloy CM, Rodeheffer RJ. Congestive heart failure in the community, a study of all incidence cases in Olmstead County, Minnesota, in 1991. Circulation 1998;98:2282 - 9. 3. Kerzner R, Gage BF, Freedland KE, et al. Predictors of mortality in younger and older patients with heart failure and preserved or reduced left ventricular ejection fraction. Am Heart J 2003;146:286 - 90. 4. Feinglass J, Martin GJ, Lin E, et al. Is heart failure survival improving? Evidence from 2323 elderly patients hospitalised between 1989-2000. Am Heart J 2003;146:111 - 4. 5. Vaccarino V, Chen Y, Wang Y, et al. Sex differences in the clinical care and outcomes of congestive heart failure in the elderly. Am Heart J 1999;138:835 - 42. 6. Cowie MR, Wood DA, Coates AJS, et al. Survival of patients with a new diagnosis of heart failure: a population based study. Heart 2000;83:505 - 10.
American Heart Journal June 2006
7. Cicoira M, Davos CH, Florea V, et al. Chronic heart failure in the very elderly: clinical status, survival and prognostic factors in 188 patients more than 70 years old. Am Heart J 2001;142:174 - 80. 8. Mosterd A, Cost B, Hoes AW, et al. The prognosis of heart failure in the general population. The Rotterdam Study. Eur Heart J 2001; 22:1318 - 27. 9. Levy D, Kenchaiah S, Larson MG, et al. Longterm trends in the incidence and survival with heart failure. N Engl J Med 2002; 347:1397 - 402. 10. Nielsen OW, Hilden J, McDonagh T, et al. Survival differences between heart failure in general practices and in hospitals. Heart 2003;89:1298 - 302. 11. Cowie MR, Struthers AD, Wood DA, et al. Value of natriuretic peptides in assessment of patients with possible new heart failure in primary care. Lancet 1997;350:1349 - 53. 12. Remme WJ, Swedberg K, Task Force for the Diagnosis and Treatment of Chronic Heart Failure K, European Society of Cardiology. Guidelines for the diagnosis and treatment of chronic heart failure. Eur Heart J 2001;22:1527 - 60. 13. Hosmer DW, Lemeshow S. Applied survival analysis. New York: J Wiley; 1999. 14. www.statistics.gov.uk/statBase/Expodata/spreadsheets/d9026.xls [Accessed 16 October 2005]. 15. www.publications.doh.gov.uk/stats/population/ads2002-residentpopulations.xls [Accessed 16 October 2005]. 16. Cohen N, Ilgiyaev E, Almoznino-Sarafian D, et al. Sex related bedside clinical variables associated with survival of older inpatients with heart failure. Eur J Heart Fail 2004;6:781 - 6.