Current issues of ACP Journal Club are published in Annals of Internal Medicine


Hemoglobin A1c levels were associated with increased cardiovascular disease and all-cause mortality in persons with and without diabetes


ACP J Club. 2005 Mar-Apr;142:52. doi:10.7326/ACPJC-2005-142-2-052

Clinical Impact Ratings

GIM/FP/GP: 6 stars

Cardiology: 6 stars

Endocrinology: 6 stars

Source Citation

Khaw KT, Wareham N, Bingham S, et al. Association of hemoglobin A1c with cardiovascular disease and mortality in adults: the European Prospective Investigation into Cancer in Norfolk. Ann Intern Med. 2004;141:413-20. [PubMed ID: 15381514]



What is the relation between hemoglobin (Hb) A1c levels and coronary heart disease (CHD) events, cardiovascular disease (CVD) events, and all-cause mortality?


Design: Cohort study (European Prospective Investigation into Cancer in Norfolk [EPIC-Norfolk]) with a mean follow-up of 6 years.

Setting: Norfolk, England, United Kingdom.

Patients: 10 232 patients 45 to 79 years of age (54% women, 2.4% with diabetes) who were recruited from general practice registers and had baseline data on a health and lifestyle questionnaire and HbA1c levels.

Risk factors: HbA1clevels, known diabetes, age, body mass index, waist-to-hip ratio, systolic blood pressure, cholesterol levels, triglyceride levels, cigarette smoking, and history of heart attack or stroke.

Outcomes: CHD events (hospital admission or death from CHD), CVD events (hospital admission or death from CHD, stroke, or other vascular causes), and all-cause mortality.

Main results

Persons with known or undiagnosed diabetes had a greater risk for all-cause mortality and CVD or CHD events than did those without diabetes. A gradient of increasing rates of all-cause mortality, CHD, and CVD was found for the entire distribution of HbA1c levels in men and women (P < 0.001 for linear trend).

Regression analyses adjusted for age and other risk factors (except for HbA1c level) showed that compared with persons without diabetes, men with diabetes had a higher risk for CHD events, CVD events, and all-cause mortality, and women with diabetes had an increased risk for CHD events and CVD events (Table). In a regression analysis that did not adjust for diabetes, HbA1c levels predicted an increased risk for CHD, CVD, and all-cause mortality in both men and women. A 1% increase in HbA1c level was associated with a 20% to 30% increase in event rates. When the regression model included both diabetes and HbA1c levels, similar relative risks for an increase in HBA1c were noted, but diabetes was no longer a significant risk factor.


A 1% increase in hemoglobin A1c levels was associated with a 20% to 30% increase in cardiovascular events and all-cause mortality in men and women 45 to 79 years of age. This relation was independent of diabetes status.

Sources of funding: Medical Research Council United Kingdom; Cancer Research United Kingdom; European Union; Stroke Association; British Heart Foundation.

For correspondence: Dr. K.T. Khaw, University of Cambridge, Cambridge, England, UK. E-mail

Table. Adjusted relative risk (RR) for coronary heart disease events, cardiovascular disease events, and all-cause mortality by hemoglobin (Hb) A1c levels and history of diabetes*

Outcomes at mean 6 y Adjusted RR (95% CI) per 1% increase in HbA1c level Adjusted RR (CI) for diabetes history (yes vs no)
Men Women Men Women
Coronary heart disease 1.25 (1.16 to 1.34) 1.20 (1.07 to 1.34) 1.87 (1.30 to 2.71) 2.71 (1.48 to 4.98)
Cardiovascular disease 1.21 (1.13 to 1.29) 1.21 (1.11 to 1.31) 1.81 (1.31 to 2.49) 1.89 (1.09 to 3.23)
All-cause mortality 1.24 (1.14 to 1.34) 1.28 (1.06 to 1.32) 1.94 (1.31 to 2.87) 0.71 (0.26 to 1.93)†

*Adjusted for age and cardiovascular risk factors (systolic blood pressure, serum cholesterol level, body mass index, waist-to-hip ratio, cigarette smoking, and history of myocardial infarction or stroke).
†Not significant.


Diabetes mellitus is a major risk factor for CVD, and unlike hypertension, smoking, and dyslipidemia, it is becoming more common over time.

The diagnostic criteria for diabetes include fasting plasma glucose levels ≥ 7.0 mmol/L or 2-hour postload plasma glucose levels ≥ 11.1 mmol/L, values above which the risk for microvascular complications of diabetes, such as retinopathy, neuropathy, and nephropathy, increases. These values correspond approximately to an HbA1c level of 7%. However, there is considerable epidemiologic evidence that the risk for CVD begins to increase at lower glycemic levels than would be considered “abnormal”—levels that would not be associated with increased microvascular disease risk (1). The study by Khaw and colleagues adds to this evidence because of its large study population and particularly large number of female participants.

In this study, 72% of the excess CVD risk that was attributable to higher HbA1c levels occurred in patients with HbA1c levels of 5.0% to 6.9%. In light of this evidence, perhaps the cutpoint for a “normal” HbA1c level should be revised downward, as has been done for cholesterol and blood pressure. It would also be desirable to develop and validate cardiovascular risk calculators that include HbA1c level as a predictor variable, as has been done for patients with type 2 diabetes in the UKPDS Risk Engine (2). In the meantime, HbA1c levels provide an additional measure of an individual patient's CVD risk.

Ronald J. Sigal, MD
Ottawa Health Research Institute and University of Ottawa
Ottawa, Ontario, Canada


1. Selvin E, Marinopoulos S, Berkenblit G, et al. Meta-analysis: glycosylated hemoglobin and cardiovascular disease in diabetes mellitus. Ann Intern Med. 2004;141:421-31. [PubMed ID: 15381515]

2. Stevens RJ, Kothari V, Adler AI, Stratton IM. The UKPDS risk engine: a model for the risk of coronary heart disease in Type II diabetes (UKPDS 56). Clin Sci (Lond). 2001;101:671-9. [PubMed ID: 11724655]