Predictive value of resting heart rate for cardiovascular and all-cause mortality

Abstracts

FUNDAMENTO: A frequência cardíaca em repouso (cuja média está entre 60 e 80 bpm) é uma das mais simples variáveis cardiovasculares e tem sido considerada como um preditor de mortalidade cardiovascular e geral. OBJETIVO: Avaliar o valor preditivo da frequência cardíaca em repouso (FCR), antes do teste ergométrico (TE), na mortalidade cardiovascular (CV) e geral. MÉTODOS: Estudo de caso-controle, que utilizou informações contidas nos bancos de dados do laboratório de ergometria de um hospital especializado em cardiologia e os registros de óbitos da Secretaria da Saúde em uma cidade do sul do Brasil, de janeiro de 1995 a junho de 2007. Foram analisados 7.055 pacientes, sendo 1.645 (23,3%) do grupo caso (óbitos) e 5.410 (76,7%) do grupo controle (vivos). Foi calculado o ponto de corte da FCR para mortalidade, através da curva ROC e realizada a análise multivariada para as variáveis selecionadas. Os desfechos foram mortalidade CV e geral. RESULTADOS: A incidência de mortalidade CV foi de 674 casos (9,5%); a FCR > 78 bpm foi o ponto de corte. Após ajustado para as variáveis selecionadas, o odds ratio (OR) para FCR > 78 bpm foi de 3,5 (IC 95% = 2,9 - 4,2) para mortalidade CV e 3,6 (IC 95% = 3,2 - 4,0) para mortalidade geral. CONCLUSÃO: A FCR > 78 bpm é um preditor independente de mortalidade cardiovascular e geral.

Valor preditivo dos testes; frequência cardíaca; teste de esforço

FUNDAMENTO: La frecuencia cardíaca en reposo (cuya media está entre 60 y 80 lpm) es una de las más simples variables cardiovasculares y ha sido considerada como un predictor de mortalidad cardiovascular y general.' OBJETIVO: Evaluar el valor predictivo de la frecuencia cardíaca en reposo (FCR), antes del test ergométrico (TE), en la mortalidad cardiovascular (CV) y general. MÉTODOS: Estudio de caso-control, que utilizó informaciones contenidas en los bancos de datos del laboratorio de ergometría de un hospital especializado en cardiología y los registros de óbitos de la Secretaría de Salud en una ciudad del sur del Brasil, de enero de 1995 a junio de 2007. Fueron analizados 7.055 pacientes, siendo 1.645 (23,3%) del grupo caso (óbitos) y 5.410 (76,7%) del grupo control (vivos). Fue calculado el punto de corte de la FCR para mortalidad, a través de la curva ROC y realizado el análisis multivariado para las variables seleccionadas. Los desenlaces fueron mortalidad CV y general. RESULTADOS: La incidencia de mortalidad CV fue de 674 casos (9,5%); la FCR > 78 lpm fue el punto de corte. Después de ajustado para las variables seleccionadas, el odds ratio (OR) para FCR > 78 lpm fue de 3,5 (IC 95% = 2,9 - 4,2) para mortalidad CV y 3,6 (IC 95% = 3,2 - 4,0) para mortalidad general. CONCLUSIÓN: La FCR > 78 lpm es un predictor independiente de mortalidad cardiovascular y general.

Valor predictivo de los tests; frecuencia cardíaca; prueba de esfuerzo

ORIGINAL ARTICLE

Instituto de Cardiologia do Rio Grande do Sul/Fundação Universitária de Cardiologia, Porto Alegre, RS - Brazil

Mailing address

ABSTRACT

BACKGROUND: Resting heart rate (which ranges from 60 to 80 bpm) is one of the simplest cardiovascular parameters, and has been considered as a predictor of cardiovascular and all-cause mortality.

OBJECTIVE: To determine the predictive value of resting heart rate (RHR) before exercise stress testing (ET) for cardiovascular (CV) and all-cause mortality.

METHODS: This was a case-control study using data from the database of the Exercise Testing Laboratory of a cardiac hospital and the death records of the Health Department of a city located in the South of Brazil from January 1995 to June 2007. A total of 7,055 patients were studied; 1,645 (23.3%) in the case group (deceased) and 5,410 (76.7%) in the control group (alive). The cut-off value of RHR for mortality was derived from the ROC curve, and a multivariate analysis was performed for the selected variables. The study's outcome measures were cardiovascular and all-cause mortality.

RESULTS: Six hundred and seventy-four patients died of cardiovascular diseases (9.5%); the cut-off value was RHR > 78 bpm). After adjusting for selected variables, the odds ratio (OR) of RHR > 78 bpm was 3.5 (95% CI 2.9 to 4.2) for CV mortality and 3.6 (95% CI 3.2 to 4.0) for all-cause mortality.

CONCLUSION: Resting heart rate > 78 is an independent predictor of cardiovascular and all-cause mortality.

Key words: Predictive value of tests; heart rate; exercise test/mortality.

Introduction

Intrinsic heart rate (HR), in the absence of any neurohumoral influence, is approximately 100 to 120 bpm. In the healthy individual, heart rate at rest reflects a balance between the tonically active sympathetic and parasympathetic systems, with a predominance of the latter. Such predominance results in a resting heart rate (RHR) that is lower than the intrinsic heart rate, one of the simplest cardiovascular parameters that usually ranges from 60 to 80 bpm¹. Resting HR, or baseline HR, refers to the number of heartbeats per minute at rest².

Since 1980, it is known that resting heart rate is a prognostic factor for coronary artery disease (CAD) and that it is associated with cardiovascular³ and all-cause⁴ mortality.

Several past long-term follow-up studies have demonstrated the impact of elevated resting HR in the general population⁵; elderly subjects⁶; and hypertensive⁷ as well as acute myocardial infarction pacients⁶; on both cardiovascular^8,9 and all-cause^8,10. This finding has been confirmed by more recent studies. ^9,11,12. In multivariate analyses, many studies have found that the relative risk (RR) for elevated RHR remained high even after being adjusted for several cardiovascular risk factors. These studies also found that increased RHR is a predictor of cardiovascular and all-cause mortality^11,13,14.

This study sought to evaluate the predictive value of resting (pre-exercise) heart rate for cardiovascular and all-cause mortality, establishing a cut-off value in a sample of patients undergoing exercise stress testing (ET). Investigating the relationship between RHR and mortality is extremely important. The findings of our study both support and add new evidence to the existing literature regarding the predictive value of RHR for cardiovascular and all-cause mortality.

Methods

This was a case-control study using data from the database of exercise tests of a cardiac hospital and the death records of a city located in the South of Brazil from January 1995 to June 2007. The case group (deceased) consisted of patients selected from the Death Registry who had undergone exercise testing, and the control group (alive) consisted of patients extracted from the database of the Exercise Testing Laboratory (aged 25 or older). Patients who had the same name in the database were excluded from the study when there was insufficient data to determine if they were one and the same person. The Probabilistic Record Linkage Software (Link Plus)¹⁵ developed by the U.S Centers for Disease Control (CDC), a sophisticated linkage mechanism, was used to mach the names from the database of the Exercise Testing Laboratory with those of the Death Registry. Matching was done by gender and the same year at which the exercise test had been performed. After data analysis, a total of 7,055 patients were selected: 1,645 for the case group (deceased) and 5,410 for the control group (the number of patients was greater in this group to minimize any potential bias resulting from deaths away from home).

The outcome measures selected were cardiovascular and all-cause mortality. Death was considered as due to cardiovascular disease if the death certificate listed acute myocardial infarction (AMI), cardiorespiratory arrest, angina pectoris, ischemic heart disease, cardiogenic shock, sudden death, congestive heart failure, arrhythmias, or other causes of death included in chapter IX of the International Classification of Diseases, 10th Revision (ICD-10)¹⁶. Before beginning the test, a seated resting heart rate was measured with a heart rate monitor and blood pressure (BP) was taken manually. The Bruce protocol¹⁷ was used for most patients, and the indication for the test (about 98%) was based on the diagnosis.

The following variables were analyzed: gender, age, smoking status, use of cardiovascular drugs, resting BP > 140/90 mmHg, VO₂max < 28 ml/kg/min, body mass index (BMI) > 25 kg/m²; hypercholesterolemia, and diabetes. The latter two were validated based on the analysis of a subsample of 1,198 Registration Forms for Exercise Testing (on cardiovascular risk factors) and compared with laboratory tests using the Kappa Index of Agreement¹⁸ (considered acceptable when > 0.40): hypercholesterolemia and cholesterol > 200 mg/dl (K = 0.510); diabetes and blood glucose > 126 mg/dl (K = 0.621).

In this study, VO₂ max (maximal oxygen uptake) was determined indirectly, taking into account age, gender, and tolerated workload, as described in the Brazilian Consensus on Ergometry¹⁷.

Statistical analyses were performed using the SPSS™ software package (version 15.0). The following tests were applied: the Student's t-test, Pearson's chi-square test, ROC curve, confidence intervals, and multivariate analysis with odds ratio. P values < 0.05 were considered statistically significant, and the confidence interval was 95%.

This study protocol was approved by the Research Ethics Committees of the local Health Department and the Institutional Review Board.

Results

A total of 7,055 patients were evaluated, 1,645 (23.3%) of whom were in the case group (deceased) and 5,410 (76.7%) in the control group (alive). The mean age of the study population was 55.43 ± 10.48 years, and 61.3% of the patients were male. The follow-up period was 12 years. Cardiovascular mortality rate was 9.5% (674 patients).

The distribution of the remaining variables between groups is shown in Table 1. Mean resting HR of the case group (deceased), for both cardiovascular and all-cause death, was similar (83.18 ± 15.50 and 83.19 ± 15.50), but when it was compared with the control group (72.13 ± 12.64), there was a significant difference (p < 0.001). Smoking, diabetes, resting BP > 140/90 mmHg and VO₂max < 28 ml/kg/min were more prevalent in the case group (CV and all-cause death) than in the control group (p < 0.001); cardiovascular drugs were more frequently used in the case group (CV death) (69%) than in the control group (57.5%), p < 0.001. Hypercholesterolemia and BMI > 25 kg/m² were equally distributed among groups (p > 0.005).

Regarding age distribution, Table 2 shows that approximately 30% of the patients who died were in the 51 to 60 and 61 to 70 age groups; 6.4% (CV death) and 6.9% (all-cause death) were in the 25 to 40 age group; and 16.8% (CV death) and 15.3% (all-cause death) were in the 71 to 88 age group, greater than in the control group (3.5%). The age difference between groups was statistically significant (p < 0.001).

Figure 1 depicts the ROC curves for the case group, both for CV and all-cause death, which were similar to one another; minimal resting HR was 41 bpm, and maximum resting HR was 134 bpm. The ROC curve analysis showed the optimal sensitivity (Se) and specificity (Sp) of the cut-off value of resting heart rate for all-cause death (sensitivity of 0.618 and 1 and specificity of 0. 315) and CV death (sensitivity of 0.614 and 1 and specificity of 0.315), resulting in 61% and 68%, respectively. The cut-off value for mortality, therefore, was RHR > 78 bpm.

Tables 3 and 4 show the correlation between the cut-off value for mortality and the significant variables of the study.

Table 4

Table 3 shows that more women than men had RHR > 78 bpm (OR 1.32, CI 1.19 -1.46). Smoking (OR 1.21, CI 1.08 -1.36), diabetes (OR 1.65, CI 1.41-1.95), VO₂max < 28 ml/kg/min (OR 2.04, CI 1.71-2.43), and resting BP > 140/90 mmHg (OR 1.36, CI 1.24 -1.50) were associated with resting HR > 78 bpm. Cardiovascular drugs were found to have a protective effect, since only 37.5% of the patients who were using them had resting HR > 78 bpm.

Table 5 shows the logistic regression analysis for the case group (CV death). Male gender (OR 1.36, CI 1.12-1.65), smoking (OR 1,92, CI 1.58-2.35), diabetes (OR 1.91; CI 1.50-2.43), use of cardiovascular drugs (OR 1.27, CI 1.04-1.55), resting BP > 140/90 mmHg (OR 1.20, CI 1.003-1.440), and VO₂max < 28 ml/kg/min (OR 1.35, CI 1.003-1.820) were associated with cardiovascular mortality. Age was significant only in the oldest age group (71 to 88), with odds ratio of 6.38 (CI 4.11-9.88). When evaluated separately, resting HR > 78 bpm was found to be an independent variable (OR 3.56, CI 2.99-4.24).

Table 6 shows the logistic regression analysis for all-cause mortality. Male gender (OR 1.27, CI 1.11-1.45), smoking (OR 1.68, CI 1.46-1.94), diabetes (OR 1.43, CI 1.18-1.73), VO₂max < 28 ml/kg/min (OR 1.28, CI 1.03-1.59), plus age groups 61 to 70 (OR: 1.52, CI: 1.17-1.97) and age 71 to 88 (OR 6.31, CI 4.62-8.61) were associated with all-cause mortality. When resting HR > 78 bpm (OR 3.62, CI 3.21-4.08) was evaluated separately, it showed the same behavior.

Discussion

In our study, elevated RHR was found to be associated with mortality, as reported by other authors^9,14,19,20.

In earlier studies, such as the National Health and Nutrition Examination Survey (NHANES), Spandau Health Test (SHT), and The Three Chicago Epidemiologic Studies, heart rate was associated with cardiovascular and noncardiovascular mortality²¹.

Other studies have confirmed these data, such as the CORDIS Trial and MATISS: Kristal-Boneh et al¹¹; in the CORDIS trial, found that elevated RHR was strongly associated with both all-cause (RR 2.23, CI 1.4-3.6; RHR > 90 bpm) and cardiovascular mortality, even after controlling for many known risk factors. Seccareccia et al¹³; in the MATISS study, found that in a low-risk Italian population heart rate increment was associated with a RR increase of 1.52 (CI 1.29-1.78) for all-cause mortality and 1.63 (CI: 1.26-2.10) for cardiovascular mortality.

It is known that individuals with elevated resting heart rate have less HR variability (reflecting an imbalance in the autonomic nervous system), which predisposes to arrhythmias, heart failure, atherosclerosis and increased risk of death²². In this study, hypercholesterolemia and BMI > 25 kg/m² were equally distributed between the case and control groups. Although no variable alone was significant, the association with other risk factors, such as systemic hypertension, cigarette smoking, and diabetes may increase mortality rates²³.

All the remaining variables were associated with resting HR > 78 bpm, as evidenced by several previous studies demonstrating that smoking, hypertension, diabetes, and sedentary lifestyle are associated with increased HR and mortality risk^7,24,25. Only in the case group (CV death) was resting BP > 140/90 mmHg significant, and the patients who died were 1.20 times more likely to have increased blood pressure. The impact of elevated heart rate on hypertensive as well as diabetic patients, increasing by three-to-four-fold the risk of cardiovascular events and twice the risk of dying from such an event, compared to the general population, is already well documented in the literature^4,26,27. The adrenergic hyperactivity involved in these risk factors and the imbalance between the sympathetic and parasympathetic activities explain these findings².

Table 3 shows that patients with RHR > 78 bpm tended to be women (42.8%), although cardiovascular death (65.9%) and all-cause death (64.3%) were more prevalent in men. Even though RHR was higher in women, the association with the development of hypertension, atherosclerosis, and CV morbidity and mortality is lower or absent in this group²⁸. It is possible that women, especially premenopausal women, are protected from the deleterious effects of elevated HR due to an estrogen-associated increase in their serum HDL-cholesterol levels²⁹. As shown in Table 4, patients who died from cardiovascular disease and from all causes, and who belonged to the age groups 51 to 60 and 61 to 70 years, were four times more likely to have increased RHR. In healthy individuals, HR is inversely related to age². Cardiovascular risk factors are known to be associated with increased HR, therefore the effects of elevated HR in said age groups are even more deleterious^30,31.

Evaluated separately, RHR > 78 bpm was found to have an impact on cardiovascular and all-cause mortality. In our study, the patients who died were more than three times as likely to have increased RHR than those who did not (OR 3.56 and 3.62, respectively). This finding supports the association of increased RHR with mortality. The deleterious effects of elevated HR have already been demonstrated in the Framingham study²¹. In this study, in a cohort of 5,070 subjects free of cardiovascular diseases at baseline, cardiovascular mortality increased progressively with resting heart rate.

The large number of patients included in our study makes it possible to confirm the hypothesis that the association of elevated RHR with mortality is not accidental.

Concluding, our findings suggest a relationship between RHR and mortality, from a cut-off value of RHR > 78 bpm, found both in CV and all-case deaths, with greater risk in the 71 to 88 age group. In our study, RHR > 78 bpm, even after adjustment for gender, diabetes, smoking, use of cardiovascular drugs, resting BP > 140/90 mmHg, and VO₂max < 28 ml/kg/min, was found to be an independent predictor of cardiovascular and all-cause mortality.

Study limitations

Possible limitations to our results include the following: the origin, not described, of all patients undergoing exercise test (some may have come from the ER and have increased RHR); the method used to consider patients as alive in the study (i. e., the fact that they were not included in the death records of the Health Department); the reliability of the information abstracted from the Death Certificates; and the fact that the Kappa test was the only statistical method used to validate the study variables.

Potential Conflict of Interest

No potential conflict of interest relevant to this article was reported.

Sources of Funding

There were no external funding sources for this study.

Study Association

This study is not associated with any post-graduation program.

References

• 1. Woods SL, Froelicher ESS, Motzer SU. Enfermagem em cardiologia. 4Ş ed. São Paulo: Manole; 2005.
• 2. Dyer AR, Persky V, Stamler J, Paul O, Shekelle RB, Berkson DM, et al. Heart rate as a prognostic factor for coronary heart disease and mortality: findings in three Chicago epidemiologic studies. Am J Epidemiol. 1980; 112 (6): 736-49.
• 3. Diaz A, Bourassa MG, Guertin MC, Tardif JC. Long-term prognostic value of resting heart rate in patients with suspected or proven coronary artery disease. Eur Heart J. 2005; 26 (10): 967-74.
• 4. Benetos A, Rudnichi A, Thomas F, Safar M, Guize L. Influence of heart rate on mortality in a French population: role of age, gender, and blood pressure. Hypertension. 1999; 33 (1): 44-52.
• 5. Disegni E, Goldbourt U, Reicher-Reiss H, Kaplinsky E, Zion M, Boyko V, et al. The predictive value of admission heart rate on mortality in patients with acute myocardial infarction. SPRINT Study Group. Secondary Prevention Reinfarction Israeli Nifedipine Trial. J Clin Epidemiol. 1995; 48 (10): 1197-205.
• 6. Palatini P, Thijs L, Staessen JA, Fagard RH, Bulpitt CJ, Clement DL, et al. Predictive value of clinic and ambulatory heart rate for mortality in elderly subjects with systolic hypertension. Arch Intern Med. 2002; 162 (20): 2313-21.
• 7. Gillman MW, Kennel WB, Belanger A. Influence of heart rate on mortality among persons with hypertension: The Framingham Study. Am Heart J. 1993; 125: 1148-54.
• 8. Medalie JH, Kahn HA, Neufeld HN. Five-year myocardial infarction evidence II: Association of single variables to age and birthplace. J Chronic Dis. 1973; 26 (6): 325-49.
• 9. Goldberg RJ, Larson M, Levy D. Factors associated with survival to 75 year of age in middle-aged men and women: The Framingham Study. Arch Intern Med. 1996; 156 (5): 505-9.
• 10. Kristal-Boneh E, Silber H, Harari G, Froom P. The association of resting heart rate with cardiovascular, cancer and all-cause mortality. Eight year follow-up of 3527 male Israeli employees (the CORDIS Study). Eur Heart J. 2000; 21 (2): 116-24.
• 11. Cucherat M. Quantitative relationship between resting heart rate reduction and magnitude of clinical benefits in post-myocardial infarction. Eur Heart J. 2007; 28 (24): 3012-9.
• 12. Seccareccia F, Pannozzo F, Dima F, Minoprio A, Menditto A, Lo Noce C, et al. Heart rate as a predictor of mortality: the MATISS project. Am J Public Health. 2001; 91 (8): 1258-63.
• 13. Okamura T, Hayakawa T, Kadowaki T, Kita Y, Okayama A, Elliott P, et al. Resting heart rate and cause-specific death in a 16.5-year cohort study of the Japanese general population. Am Heart J. 2004; 147 (6): 1024-32.
• ¹⁴
  Ministério da Saúde. Datasus. Classificação Internacional de Doenças. CID. [Acesso em 2009 dez 13]. Disponível em http://www.datasus.gov.br/cid10/webhelp/cid10.htm
• 15. Sociedade Brasileira de Cardiologia. II Diretrizes sobre teste ergométrico. Arq Bras Cardiol. 2002; 78 (supl 2): 1-17.
• 16. Topol EJ. Tratado de cardiologia. 2Ş ed. Rio de Janeiro: Guanabara Koogan; 2005.
• 17. Araújo HG. Bioestatítstica teórica e computacional. Rio de Janeiro: Guanabara Koogan; 2001.
• 18. Levy S, Guize L. Heart rate, a major prognostic factor of cardiovascular risk. Therapie. 2006; 61 (2): 115-9.
• 19. Okamura T, Hayakawa T, Kadowaki T, Kita Y, Okayama A, Elliott P, et al. Resting heart rate and cause-specific death in a 16.5-year cohort study of the Japanese general population. Am Heart J. 2004; 147 (6): 1024-32.
• 20. Cook S, Togni M, Schaub MC, Wenaweser P, Hess OM. High heart rate: a cardiovascular risk factor? Eur Heart J. 2006; 27 (20): 2387-93.
• 21. Gillum RF, Makuc DM, Feldman JJ. Pulse rate, coronary heart disease, and death: the NHANES I Epidemiologic Follow-up Study. Am Heart J. 1991; 121 (1 Pt 1): 172-7.
• 22. Fujiura Y, Adachi H, Tsuruta M, Jacobs Jr DR, Hirai Y, Imaizumi T. Heart rate and mortality in a Japanese general population: an 18-year follow-up study. J Clin Epidemiol. 2001; 54 (5): 495-500.
• 23. Sociedade Brasielira de Cardiologia. IV Diretriz brasileira sobre dislipidemia e prevenção da aterosclerose. Arq Bras Cardiol. 2007; 88 (supl. 1): 2-19.
• 24. Duprez D. Heart rate: an independent predictor for hypertension? Hypertension. 2006; 24: 1711-3.
• 25. Inoue T, Iseki K, Iseki C, Ohya Y, Kinjo K, Takishita S. The impact of heart rate on the development of metabolic syndrome - a longitudinal analysis of screened cohort in Okinawa Japan [abstract]. Eur Heart J. 2006; 27 (Suppl): 383.
• 26. Tiernay WM, Brunt M, Kesterson J, Zhou XH, L'Italien G, Lapuerta P. Quantifying risk of adverse clinical events with primary care patients with hypertension. Ann Fam Med. 2004; 2 (3): 209-17.
• 27. Selvin E, Marinopoulos S, Berkenblit G, Rami T, Brancati FL, Powe NR, et al. Meta-analysis: glycosylated hemoglobin and cardiovascular disease in diabetes mellitus. Ann Intern Med. 2004;141 (6): 421-31.
• 28. Chang M, Havlik RJ, Corti MC, Chaves PH, Fried LP, Guralnik JM. Relation of heart rate at rest and mortality in the Women's Health and Aging Study. Am J Cardiol. 2003; 92 (11): 1294-9.
• 29. Benetos A, Thomas F, Bean K, Albaladejo, Palatini P, Guize L. Resting heart rate in older people: a predictor of survival to age 85. J Am Geriatr Soc. 2003; 51: 284-5.
• 30. Palatini P, Casiglia E, Julius S, Pessina AC. High heart rate: a risk factor for cardiovascular death in elderly men. Arch Intern Med. 1999; 159 (6): 585-92.
• 31. Jouven X, Zureik M, Desnos M, Guerot C, Ducimetiere P. Resting heart rate as a predictive risk factor for sudden death in middle-aged men. Cardiovasc Res. 2001; 50 (2): 373-8.

Predictive value of resting heart rate for cardiovascular and all-cause mortality

Jaqueline Eilert Fagundes; Iran Castro

Publication Dates

• Publication in this collection
  12 Nov 2010
• Date of issue
  Dec 2010

History

• Received
  2090
• Reviewed
  08 Jan 2010
• Accepted
  24 Feb 2010