The epidemiology and hypertension
of impaired
glucose tolerance
Epidemiologic research indicates that glucose intolerance and hypertension are interrelated phenomena, each powerfully predisposing to atherosclerotic cardiovascular disease. Both diabetic and hypertensive patients have greater amounts of atherogenic risk factors, including dyslipidemia, hyperuricemia, elevated fibrinogen, and left ventricular hypertrophy. Diabetic persons have an increased prevalence of hypertension (50%), and glucose intolerance is more common in hypertension (15% to 18%). Both share a strong relationship to excess weight, but the excess of hypertension In diabetic persons occurs In both lean and obese subjects. Diabetes doubles the risk of hypertension associated with overweight. The risk of coronary disease, stroke, and peripheral arterial disease increases with increasing blood pressure to the same degree in diabetic persons as in nondiabetic persons, but at any level of blood pressure, diabetic persons have a doubled risk of these outcomes. Both diabetic and hypertensive patients are particularly prone to silent or unrecognized myocardial Infarctions. Greater efforts at primary prevention of both hypertenslon and diabetes are clearly needed, including efforts at weight control, exercise, limltation of salt intake, and control of blood lipid levels. In either diabetic or hypertensive candidates for cardiovascular disease, optimization of the chances of avoiding sequelae requires a comprehensive multifactorial approach. Prevention requires more than normalization of either the blood sugar or blood pressure. Rational preventive measures must also include weight reduction, a fat-modified diet, cessation of smoking cigarettes, raising high-density lipoprotein, lowering low-density lipoprotein, and reduction of fibrinogen. Hypertension, obesity, insulin resistance, hyperinsulinemia, hypertriglyceridemia, and low high-density lipoprotein cholesterol tend to coexist. All these factors accelerate atherogenesis and may be responsible for the increased propensity of either diabetic or hypertensive patients to develop coronary heart disease. Treatments that worsen these features in hypertensive or diabetic patients are best avoided. (AM HEART J 1991;121:128&73.)
William
B. Kannel,
Framingham,
MD, MPH:
Peter W. F. Wilson,
Persistent nonphysiologic elevations of blood pressure or blood glucose are a direct cause of cardiovascular sequelae and premature mortality.1-4 Because of their high prevalence throughout the world and their impact on cardiovascular disease, these conditions justify a high priority for detection and treatment. Hypertension and glucose intolerance each play a critical and independent role in atherogenesis, now their chief lethal sequela. This disease has
From tal.b
the
Framingham
Heart
Studya
Supported by National Institutes of Health NOl-92922, NOl-HV 52971, and NOl-HC38038, dation, and Merck, Sharp and Dohme. Reprint Study, 4/O/26968
1268
requests: 5 Thurber
MD,a and Ting-Jie
Zhang, MDb
Mass., and the People$ Republic of China
and
the
Changdu
3rd
Hospi-
grants and contract numbers The Chas. A. Dana Foun-
William B. Kannel, MD, MPH, St., Framingham, MA 01701.
Framingham
Heart
replaced accelerated malignant hypertension and diabetic acidosis as the chief outcome of uncontrolled hypertension and glucose intolerance. This article examines the interrelationship and net and joint impact of these powerful atherogenic risk factors on the occurrence of cardiovascular sequelae using chiefly the Framingham Study data. METHODOLOGY
The Framingham Heart Study was implemented in 1948 in a representative sample of 5209 men and women, ages 30 to 62 years at entry, who were subsequently followed biennially for the occurrence of new cardiovascular events. Every 2 years a history and physical examination were performed and blood chemistries were drawn.5 Blood pressures were obtained with the patient seated using a mercury sphygmomanometer and a cuff wide enough to fit the arm circumference. Blood pressures were obtained by the physician as well as a nurse. Blood was drawn for glucose at all biennial clinic visits except for the fifth, seventh,
"dume
121
Number
4. Part 2
Glucose intolerance
NO glucose intolerance Glucose intolerance
Men
and hypertension
1269
m i
Women
5X.6
d-i 44
33.3
Metropolitan Risk ratio
],76”“‘4
relative weight (% of desirable weight) 1,5’,**”
0 73+
***/I Fig.
1. Incidence
of hypertension
Table I. Comparison of risk factor levels, diabetic versus nondiabetic persons, Framingham cohort 1972, age-adjusted means Men
Women
Diabetes
Diabetes
Present (n = 318)
Absent Present Absent (n = 1495) (n = 328) (n = 2178)
TJ**a
2,, ***
7-.- 7***
< 0.001 +NS by diabetic
status and relative
weight.
Table II. Prevalence of hypertension (subjects, 55 to 64 years, examination study) Age-adjusted
Nondiabetic Diabetic All
by diabetic status 13, Framingham CC, prevalence
Men
Women
26.7 47.6
27.8 61.8
28.9
29.6
Glucose
mg/dl mmol/L SBP mm Hg Body maas index (kg/m?) Hematocrit (‘I( ) Uric acid (mg/dl) LVH (“;) Cigarettes (“i ) SHP, Systolic
147t 8.17t 145t 27.4t
48.87 5.37 14.5* 35.8
blood pressure;
87 4.83 137 26.0 46.2 4.9 7.1 43.0
LVH. left ventricular
134t 7.44t 1497 X3.0t 43.9 4.48t 12.47 32.3
86 4.18 140 25.4 41.9 3.9 5.8 33.6
hypertrophy.
*p< 0.01. tp < O.lK)l.
and eleventh biennial examinations. The Somogyi-Nelson method was used for determination of glucose in whole blood.‘j Persons were classified as diabetic if casual blood glucose levels exceeded 150 mg/dl (8.33 mmol/L) on two clinic visits, a positive glucose tolerance test was obtained by a local physician, or if insulin or oral hypoglycemic agents were taken.5 Glucose intolerance was designated if
blood sugar levels at the clinic visits were above 120 mg/dl (6.67 mmol/L) but less than 150 mg/dl (8.33 mmol/L) on two separate occasions. Glycosuria was determined in the clinic with Clinitest tablets during the first 10 biennial examinations. Cholesterol determination was obtained at each biennial examination from early in the study. Beginning at examination 11, a full fasting lipid profile was performed with measurement of plasma cholesterol, highdensity lipoprotein (HDL) cholesterol after heparin manganese precipitation,7 ultracentrifugation of specimens with measurement of bottom fraction cholesterol, and calculation of low-density lipoprotein cholesterol and very low-density lipoprotein cholesterol by subtraction7 Fibrinogen was measured at examination 10 by a modification of the Ratnoff method.8 Multiple logistic analysis was used to assess the net and joint effect of impaired glucose tolerance on cardiovascular disease endpoints of coronary heart disease, stroke, pe-
Kannel, Wilson, and Zhang
1270
hmerican
I74
180 j
1 / -
160
i
G
8
April 1991 Heart Journal
Nogl ucose intolerance Glucose intolerance
140 Men
2120’
Women
II’)
g 100’
.liill~-.-~!~~~li~
105
135
165
195
105
135
165
195
Systolic blood pressure (mm Hg) Cholesterol 185 mg/dl, no LVH, nonsmokers 26-year follow up, Framingham Study Fig. 2. Risk of cardiovascular jects, ages 45 years.
Table Ill. Relative by diabetes
risks of cardiovascular Age-
Cardiovascular disease outcome
Cardiac failure Brain infarct Coronary disease Peripheral vascular disease *Adjusted for age, systolic ventricular hypertrophy; 45 to 74 years of age.
ripheral arterial these endpoints
blood ?&year
events imposed
and risk factor-adjusted risk ratios * Men
Women
2.1
2.0 2.2 2.1 5.0
2.2 1.7 4.2 pressure, follow-up,
disease by systolic blood pressure and diabetic
cigarettes, cholesterol, Framingham Study;
EN&left subjects,
disease, and cardiac failure.g Criteria have been detailed elsewhere.5
for
RESULTS
Compared with nondiabetic persons, diabetic persons had greater numbers of atherogenic risk factors including elevated blood pressure, unfavorable total/ HDL cholesterol ratio, elevated uric acid and fibrinogen, and left ventricular hypertrophy (Table I). Diabetic persons have an increased prevalence of hypertension, with as many as 50% having some degree of elevated blood pressure (Table II). Both hypertension and diabetes share a strong relationship to
status; sub-
obesity, but the excess of hypertension in diabetic individuals appears to occur in both lean and obese subjects (Fig. 1). The incidence of both diabetes and hypertension increases progressively with the degree of overweight. The incidence of new onset of hypertension increases with relative weight in both diabetic and nondiabetic women, with diabetic patients twice as hypertensive as nondiabetic patients at any relative weight. In men, diabetics have a higher incidence of hypertension at all relative weights except those exceeding desirable weight by 34% (Fig. 1). In diabetic men no increase in the incidence of hypertension with weight is evident. Thus diabetes doubles the risk of hypertension associated with obesity. Systolic more than the diastolic pressure is related to blood glucose level. CARDIOVASCULAR
SEQIJELAE
The risk of coronary disease,stroke, and peripheral arterial diseaseincreases with blood pressure level to the same degree in diabetic persons as in nondiabetic persons. However, at any level of blood pressure, diabetic patients have a greater risk (Fig. 2). Both hypertension and diabetes predispose to the major cardiovascular diseases(Tables III and IV). As currently defined, the impact of hypertension (twofold to threefold risk ratio) barely exceeds that for diabetes (2.7-fold). However, because hypertension is more
volume121 Number 4, Part 2
Glucose intolerance
IV. Risk of cardiovascular events by hypertension status: Framingham Heart Study, 30-year follow-up
Table
Age-adjusted risk ratio (definite HBP;tfnormal) Age 35-64
yr
Age 65-94
Men
Coronary heart disease Stroke Peripheral arterial disease Cardiac failure Cardiovascular
Women
2.6?
Men
3.3t
VI. Average annual incidence and multivariate analysis of unrecognized myocardisl infarction in subjects with untreated high blood pressure (per 1000, age-adjusted, 30year follow-up, Framingham Study) Incidence
yr
6.0t 2.5t
3.0f 3.0t
3.0f 2.87
Women
2.9t
2.0t
3.1t 1.5)1
3.0t 3.47
3.0t 1.7*
3.8t 3.3t
2.0t l.Bt
diseases
V. Relative and attributable
Age (yr)
Normal
Men 35-64 65-94
1
Diabetes Hypertension
2.0 2.2
Cigarettes ECG-LVH
1.5
2.7
2 7
3 9
2.17
0.5 1
1 3
2 3
2.7t 1.6*
1.6*
35-64 65-94
VII. Average annual incidence and multivariate analysis of unrecognized myocardial infarction in diabetic subjects (per 1000, age-adjusted,30-year follow-up)
Women
3.9 37.7 20.6 6.2
-
Table
Men
Risk factors
Diabetes age-adjusted risk ratio
2
risks for specified risk
Attributable risk (%)
Diabetes and high blood pressure
Women
incidence
factors
Relative risk (%)
High blood pressure
*p < 0.05. tp < 0.001.
HBP, High blood pressure. *p < 0.05. tp < 0.001. IDefinite high blood pressure: zz 160/95 mm Hg. IPersons free of any cardiovascular disease. 11Not significant.
Table
127 1
Table
Cardiovascular event
and hypertension
Relative risk C%J
2.7 3.2 1.1 3.5
Age (yr)
Attributable
risk (%J 5.0 54.6 2.8 6.8
Men 35-64 65-94 Women 35-64 65-94
No diabetes mellitus
Diabetes mellitus
Age-adjusted risk ratio
1 5
4 8
2.9* 1.6
1
1
1.0
2
4
1.5
-
*p < 0.01. ECG-LVH, ECG-left ventricular hypertrophy. Twenty-six-year follow-up, Framingham Study; age.
subjects, 45 to 74 years of
prevalent, the attributable cardiovascular risk for hypertension (38% to 55%) far exceeds that for diabetes (4 % to 5 % ) (Table V). Both diabetes and hypertension independently contribute to the risk of a cardiovascular event. The impact of diabetes on atherosclerotic cardiovascular disease is not entirely attributable to associated hypertension or other cardiovascular risk factors and indicates a unique effect. However, cardiovascular risk in diabetic persons varies widely depending on concomitant cardiovascular risk factors. Hypertension in diabetic persons places them in double jeopardy for cardiovascular sequelae, including coronary disease, stroke, peripheral arterial disease, cardiac failure, left ventricular hypertrophy, retinopathy, and renal damage. The strong relationship between diabetes and cardiac failure, even adjusting for associated hypertension and other atherogenic risk factors, indicates that
diabetes may directly damage the myocardium by producing a cardiomyopathy. Hypertension greatly augments the risk of cardiac failure in diabetic persons. Consequently ,&blockers, which can depress contractility, are not a good choice of antihypertensive therapy in hypertensive diabetic patients. Both diabetic and hypertensive persons are particularly prone to silent or unrecognized myocardial infarctions (Tables VI and VII), necessitating a high degree of vigilance with routine periodic ECG examinations. The diabetic propensity to silent myocardial infarctions appears to be confined to men, whereas hypertensive patients of both sexes are at greater risk of a silent or unrecognized infarction. This increased susceptibility of the hypertensive patient to unrecognized infarction holds even excluding diabetic persons, those under treatment with antihypertensive agents, and those with left ventricular hypertrophy. lo In persons who have already sustained a myocardial infarction, both diabetes and
April
1272
Kannel,
Wilson, and Zhang
~msrican
Females 55-64 years old, Framingham Fig.
3.
Prevalence
hypertension predispose to recurrences and cardiac failure.’ ’ Examination of secular trends in the prevalence of glucose intolerance indicates a threefold increase from the 1950s through the end of the 1970s (Fig. 3). Prevalence of hypertension has failed to decrease if those persons receiving treatment with normalized pressures are included. Clearly, greater efforts at primary prevention of both conditions are needed, and this should include weight control, exercise, limitation of salt intake, and control of blood lipid levels. DISCUSSION
Three decades of epidemiologic research from the Framingham Heart Study indicate that both diabetes and hypertension are powerful predisposing factors for the development of atherosclerotic disease, including coronary heart disease.‘, 4l “-I5 Both conditions increase the propensity to cardiovascular disease twofold to threefold. The increased vulnerability of diabetic women compared with men is unexplained. Diabetes is one of the few conditions that eliminates the female advantage over men. A diminishing impact in women with advancing age suggests that late onset diabetes may be less atherogenic. Impaired glucose tolerance is associated with cardiovascular events whether manifested by a diagnosis of diabetes, hyperglycemia, or glycosuria. Blood glucose levels are a graded independent risk factor for cardiovascular disease in women, even within a relatively normal blood sugar range. Epidemiologic studies indicate an association be-
Heart
1991 Journal
Study
of glucose intolerance.
tween hyperinsulinemia, insulin resistance, and hypertension, even in nonobese people.16 Also, insulin resistance has been found in hypertensive persons who are neither obese nor diabetic, which suggests a basic relationship between insulin and blood pressure elevation. Insulin is known to enhance reabsorption of sodium by the kidney impeding sodium diuresis, resulting in higher pressures. Whatever their pathogenetic or etiologic relationships, both diabetes and hypertension each enhance atherogenesis. Diabetes and hypertension share in common predisposing obesity and insulin resistance. The combination of impaired glucose tolerance, hypertension, insulin resistance, obesity, lipoprotein lipase deficiency, hypertriglyceridemia, and low HDL cholesterol often occurs jointly, and when present in combination, powerfully predisposes to cardiovascular sequelae. It appears important to treat hypertension selectively in diabetic persons, because this appears to delay the cardiovascular sequelae of diabetes. Furthermore, with the increasing prevalence of glucose intolerance, greater attention should be given to the metabolic effects of antihypertensive drugs, because the multiple risk factors are not readily evident. The antihypertensive agent used in these patients should not adversely affect atherogenic blood lipid levels, glucose tolerance, uric acid, and magnesium or potassium balance. Agents used must also not induce orthostatic hypotension, renal hypoperfusion, or sexual dysfunction, to which the diabetic patient is prone. Although traditional diabetic therapy has emphasized blood glucose control and its regulation to avoid
Volume Number
121 4, Part 2
large- or small-vessel disease,17 the modern approach clearly requires a broader approach to diabetes and borderline diabetic conditions.1a-20 Individuals with adult onset diabetes are more likely to develop premature cardiovascular disease and to die of its complications or consequent cardiac failure. A multifactorial intervention to treat blood pressure, obesity, and lipids, in addition to managing glucose levels, is definitely needed. REFERENCES
1. Kannel WB, McGee DL. Diabetes and cardiovascular risk factors: the Framingham Study. Circulation 1979;59:8-13. 2. Ruderman NB. Haudenshild C. Diabetes as an atherogenic factor. Prog Cardiovasc Dis 1984;26:373-400. 3. Fuller JH, Shipley MJ, Rose G, Jarrett RJ, Keen H. Coronary heart disease risk and impaired glucose tolerance: the Whitehall Study. Lancet 198O;i:1373-6. 4. Kannel WB. Risk factors in hypertension. J Cirdiovasc Pharmacol 1989;13 (suppl 1):4-10. 5. Kannel WB, Wolf PA, Garrison RJ. The Framingham Study Section 35: Survival following initial cardiovascular events. U.S. Dept Commerce 1988; PB 88-204029. 6. Nelson N. A photometric adaptation of the Somogyi method for the determination of glucose. J Biol Chem 1944;153:375-80. 7. Lipid Research Clinics Program. Manual of Laboratory Operations, vol 1: lipid and lipoprotein analysis. Bethesda, Md.: National Institutes of Health, 1974; DHEW publication no. (NIH) 75-628. 8. Swain WR, Feders MB. Fibrinogen assay. Clin Chem 1967; 13:1026-8.
Glucose intolerance
and hypertension
1273
9. Walker SH, Duncan DB. Estimation of the probability of an event as a function of several independent variables. Biometrika 1967;54:167-79. 10. Kannel WB, Dannenberg AL, Abbott RD. Unrecognized myocardial infarction and hypertension: the Framingham Study. AM HEART J 1985;109:581-5. 11. Abbott RD, Donahue RP, Kannel WB, Wilson PWF. The impact of diabetes on survival following myocardial infarction in men versus women: the Framingham Study. JAMA 1988; 260:3456-60. 12. Wilson PWF, Kannel WB, Anderson KM. Lipids, glucose intolerance and vascular disease: the Framingham Study. In: James C, Pometta D, eds. Monographs in atherosclerosis: vol 13. Basel: S Karger, 1985:1-11. 13. Keen H. Jarrett RJ. Fuller JH. McCartnev P. Hvoerelvcemia and arterial disease. Diabetes ‘1981;30 (suppl 2)149-53: 14. Kannel WB, McGee DL. Diabetes and cardiovascular disease: the Framingham Study. JAMA 1979;241:2035. 15. Kannel WB, Dawber TR, McGee DL. Perspectives of systolic hypertension: the Framingham Study. Circulation 1980;61: 1179-82. 16. Reaven GM, Greenfield MS. Diabetic hypertriglyceridemia, evidence for the clinical syndromes. Diabetes 1981;30 (suppl 2):66-75. 17. Schor S. The University Group Diabetes Program: a statistician looks at the mortality results. JAMA 1971;217:1671-5. 18. Haffner SM, Stern MP, Hazuda HP, Mitchell BD, Patterson JK. Cardiovascular risk factors in confirmed prediabetic individuals: does the clock for coronary heart disease start ticking before the onset of clinical diabetes? JAMA 1990;263:2893-8. 19. Stout RW. Insulin and atheroma: 20.year perspective. Diabetes Care 1990;13:631-54. 20. Zavaroni I, Bonora E, Pagliara M, et al. Risk factors for coronary artery disease in healthy persons with hyperinsulinemia and normal glucose tolerance. N Engl J Med 1989;320:702-6.
of impaired
glucose tolerance
Epidemiologic research indicates that glucose intolerance and hypertension are interrelated phenomena, each powerfully predisposing to atherosclerotic cardiovascular disease. Both diabetic and hypertensive patients have greater amounts of atherogenic risk factors, including dyslipidemia, hyperuricemia, elevated fibrinogen, and left ventricular hypertrophy. Diabetic persons have an increased prevalence of hypertension (50%), and glucose intolerance is more common in hypertension (15% to 18%). Both share a strong relationship to excess weight, but the excess of hypertension In diabetic persons occurs In both lean and obese subjects. Diabetes doubles the risk of hypertension associated with overweight. The risk of coronary disease, stroke, and peripheral arterial disease increases with increasing blood pressure to the same degree in diabetic persons as in nondiabetic persons, but at any level of blood pressure, diabetic persons have a doubled risk of these outcomes. Both diabetic and hypertensive patients are particularly prone to silent or unrecognized myocardial Infarctions. Greater efforts at primary prevention of both hypertenslon and diabetes are clearly needed, including efforts at weight control, exercise, limltation of salt intake, and control of blood lipid levels. In either diabetic or hypertensive candidates for cardiovascular disease, optimization of the chances of avoiding sequelae requires a comprehensive multifactorial approach. Prevention requires more than normalization of either the blood sugar or blood pressure. Rational preventive measures must also include weight reduction, a fat-modified diet, cessation of smoking cigarettes, raising high-density lipoprotein, lowering low-density lipoprotein, and reduction of fibrinogen. Hypertension, obesity, insulin resistance, hyperinsulinemia, hypertriglyceridemia, and low high-density lipoprotein cholesterol tend to coexist. All these factors accelerate atherogenesis and may be responsible for the increased propensity of either diabetic or hypertensive patients to develop coronary heart disease. Treatments that worsen these features in hypertensive or diabetic patients are best avoided. (AM HEART J 1991;121:128&73.)
William
B. Kannel,
Framingham,
MD, MPH:
Peter W. F. Wilson,
Persistent nonphysiologic elevations of blood pressure or blood glucose are a direct cause of cardiovascular sequelae and premature mortality.1-4 Because of their high prevalence throughout the world and their impact on cardiovascular disease, these conditions justify a high priority for detection and treatment. Hypertension and glucose intolerance each play a critical and independent role in atherogenesis, now their chief lethal sequela. This disease has
From tal.b
the
Framingham
Heart
Studya
Supported by National Institutes of Health NOl-92922, NOl-HV 52971, and NOl-HC38038, dation, and Merck, Sharp and Dohme. Reprint Study, 4/O/26968
1268
requests: 5 Thurber
MD,a and Ting-Jie
Zhang, MDb
Mass., and the People$ Republic of China
and
the
Changdu
3rd
Hospi-
grants and contract numbers The Chas. A. Dana Foun-
William B. Kannel, MD, MPH, St., Framingham, MA 01701.
Framingham
Heart
replaced accelerated malignant hypertension and diabetic acidosis as the chief outcome of uncontrolled hypertension and glucose intolerance. This article examines the interrelationship and net and joint impact of these powerful atherogenic risk factors on the occurrence of cardiovascular sequelae using chiefly the Framingham Study data. METHODOLOGY
The Framingham Heart Study was implemented in 1948 in a representative sample of 5209 men and women, ages 30 to 62 years at entry, who were subsequently followed biennially for the occurrence of new cardiovascular events. Every 2 years a history and physical examination were performed and blood chemistries were drawn.5 Blood pressures were obtained with the patient seated using a mercury sphygmomanometer and a cuff wide enough to fit the arm circumference. Blood pressures were obtained by the physician as well as a nurse. Blood was drawn for glucose at all biennial clinic visits except for the fifth, seventh,
"dume
121
Number
4. Part 2
Glucose intolerance
NO glucose intolerance Glucose intolerance
Men
and hypertension
1269
m i
Women
5X.6
d-i 44
33.3
Metropolitan Risk ratio
],76”“‘4
relative weight (% of desirable weight) 1,5’,**”
0 73+
***/I Fig.
1. Incidence
of hypertension
Table I. Comparison of risk factor levels, diabetic versus nondiabetic persons, Framingham cohort 1972, age-adjusted means Men
Women
Diabetes
Diabetes
Present (n = 318)
Absent Present Absent (n = 1495) (n = 328) (n = 2178)
TJ**a
2,, ***
7-.- 7***
< 0.001 +NS by diabetic
status and relative
weight.
Table II. Prevalence of hypertension (subjects, 55 to 64 years, examination study) Age-adjusted
Nondiabetic Diabetic All
by diabetic status 13, Framingham CC, prevalence
Men
Women
26.7 47.6
27.8 61.8
28.9
29.6
Glucose
mg/dl mmol/L SBP mm Hg Body maas index (kg/m?) Hematocrit (‘I( ) Uric acid (mg/dl) LVH (“;) Cigarettes (“i ) SHP, Systolic
147t 8.17t 145t 27.4t
48.87 5.37 14.5* 35.8
blood pressure;
87 4.83 137 26.0 46.2 4.9 7.1 43.0
LVH. left ventricular
134t 7.44t 1497 X3.0t 43.9 4.48t 12.47 32.3
86 4.18 140 25.4 41.9 3.9 5.8 33.6
hypertrophy.
*p< 0.01. tp < O.lK)l.
and eleventh biennial examinations. The Somogyi-Nelson method was used for determination of glucose in whole blood.‘j Persons were classified as diabetic if casual blood glucose levels exceeded 150 mg/dl (8.33 mmol/L) on two clinic visits, a positive glucose tolerance test was obtained by a local physician, or if insulin or oral hypoglycemic agents were taken.5 Glucose intolerance was designated if
blood sugar levels at the clinic visits were above 120 mg/dl (6.67 mmol/L) but less than 150 mg/dl (8.33 mmol/L) on two separate occasions. Glycosuria was determined in the clinic with Clinitest tablets during the first 10 biennial examinations. Cholesterol determination was obtained at each biennial examination from early in the study. Beginning at examination 11, a full fasting lipid profile was performed with measurement of plasma cholesterol, highdensity lipoprotein (HDL) cholesterol after heparin manganese precipitation,7 ultracentrifugation of specimens with measurement of bottom fraction cholesterol, and calculation of low-density lipoprotein cholesterol and very low-density lipoprotein cholesterol by subtraction7 Fibrinogen was measured at examination 10 by a modification of the Ratnoff method.8 Multiple logistic analysis was used to assess the net and joint effect of impaired glucose tolerance on cardiovascular disease endpoints of coronary heart disease, stroke, pe-
Kannel, Wilson, and Zhang
1270
hmerican
I74
180 j
1 / -
160
i
G
8
April 1991 Heart Journal
Nogl ucose intolerance Glucose intolerance
140 Men
2120’
Women
II’)
g 100’
.liill~-.-~!~~~li~
105
135
165
195
105
135
165
195
Systolic blood pressure (mm Hg) Cholesterol 185 mg/dl, no LVH, nonsmokers 26-year follow up, Framingham Study Fig. 2. Risk of cardiovascular jects, ages 45 years.
Table Ill. Relative by diabetes
risks of cardiovascular Age-
Cardiovascular disease outcome
Cardiac failure Brain infarct Coronary disease Peripheral vascular disease *Adjusted for age, systolic ventricular hypertrophy; 45 to 74 years of age.
ripheral arterial these endpoints
blood ?&year
events imposed
and risk factor-adjusted risk ratios * Men
Women
2.1
2.0 2.2 2.1 5.0
2.2 1.7 4.2 pressure, follow-up,
disease by systolic blood pressure and diabetic
cigarettes, cholesterol, Framingham Study;
EN&left subjects,
disease, and cardiac failure.g Criteria have been detailed elsewhere.5
for
RESULTS
Compared with nondiabetic persons, diabetic persons had greater numbers of atherogenic risk factors including elevated blood pressure, unfavorable total/ HDL cholesterol ratio, elevated uric acid and fibrinogen, and left ventricular hypertrophy (Table I). Diabetic persons have an increased prevalence of hypertension, with as many as 50% having some degree of elevated blood pressure (Table II). Both hypertension and diabetes share a strong relationship to
status; sub-
obesity, but the excess of hypertension in diabetic individuals appears to occur in both lean and obese subjects (Fig. 1). The incidence of both diabetes and hypertension increases progressively with the degree of overweight. The incidence of new onset of hypertension increases with relative weight in both diabetic and nondiabetic women, with diabetic patients twice as hypertensive as nondiabetic patients at any relative weight. In men, diabetics have a higher incidence of hypertension at all relative weights except those exceeding desirable weight by 34% (Fig. 1). In diabetic men no increase in the incidence of hypertension with weight is evident. Thus diabetes doubles the risk of hypertension associated with obesity. Systolic more than the diastolic pressure is related to blood glucose level. CARDIOVASCULAR
SEQIJELAE
The risk of coronary disease,stroke, and peripheral arterial diseaseincreases with blood pressure level to the same degree in diabetic persons as in nondiabetic persons. However, at any level of blood pressure, diabetic patients have a greater risk (Fig. 2). Both hypertension and diabetes predispose to the major cardiovascular diseases(Tables III and IV). As currently defined, the impact of hypertension (twofold to threefold risk ratio) barely exceeds that for diabetes (2.7-fold). However, because hypertension is more
volume121 Number 4, Part 2
Glucose intolerance
IV. Risk of cardiovascular events by hypertension status: Framingham Heart Study, 30-year follow-up
Table
Age-adjusted risk ratio (definite HBP;tfnormal) Age 35-64
yr
Age 65-94
Men
Coronary heart disease Stroke Peripheral arterial disease Cardiac failure Cardiovascular
Women
2.6?
Men
3.3t
VI. Average annual incidence and multivariate analysis of unrecognized myocardisl infarction in subjects with untreated high blood pressure (per 1000, age-adjusted, 30year follow-up, Framingham Study) Incidence
yr
6.0t 2.5t
3.0f 3.0t
3.0f 2.87
Women
2.9t
2.0t
3.1t 1.5)1
3.0t 3.47
3.0t 1.7*
3.8t 3.3t
2.0t l.Bt
diseases
V. Relative and attributable
Age (yr)
Normal
Men 35-64 65-94
1
Diabetes Hypertension
2.0 2.2
Cigarettes ECG-LVH
1.5
2.7
2 7
3 9
2.17
0.5 1
1 3
2 3
2.7t 1.6*
1.6*
35-64 65-94
VII. Average annual incidence and multivariate analysis of unrecognized myocardial infarction in diabetic subjects (per 1000, age-adjusted,30-year follow-up)
Women
3.9 37.7 20.6 6.2
-
Table
Men
Risk factors
Diabetes age-adjusted risk ratio
2
risks for specified risk
Attributable risk (%)
Diabetes and high blood pressure
Women
incidence
factors
Relative risk (%)
High blood pressure
*p < 0.05. tp < 0.001.
HBP, High blood pressure. *p < 0.05. tp < 0.001. IDefinite high blood pressure: zz 160/95 mm Hg. IPersons free of any cardiovascular disease. 11Not significant.
Table
127 1
Table
Cardiovascular event
and hypertension
Relative risk C%J
2.7 3.2 1.1 3.5
Age (yr)
Attributable
risk (%J 5.0 54.6 2.8 6.8
Men 35-64 65-94 Women 35-64 65-94
No diabetes mellitus
Diabetes mellitus
Age-adjusted risk ratio
1 5
4 8
2.9* 1.6
1
1
1.0
2
4
1.5
-
*p < 0.01. ECG-LVH, ECG-left ventricular hypertrophy. Twenty-six-year follow-up, Framingham Study; age.
subjects, 45 to 74 years of
prevalent, the attributable cardiovascular risk for hypertension (38% to 55%) far exceeds that for diabetes (4 % to 5 % ) (Table V). Both diabetes and hypertension independently contribute to the risk of a cardiovascular event. The impact of diabetes on atherosclerotic cardiovascular disease is not entirely attributable to associated hypertension or other cardiovascular risk factors and indicates a unique effect. However, cardiovascular risk in diabetic persons varies widely depending on concomitant cardiovascular risk factors. Hypertension in diabetic persons places them in double jeopardy for cardiovascular sequelae, including coronary disease, stroke, peripheral arterial disease, cardiac failure, left ventricular hypertrophy, retinopathy, and renal damage. The strong relationship between diabetes and cardiac failure, even adjusting for associated hypertension and other atherogenic risk factors, indicates that
diabetes may directly damage the myocardium by producing a cardiomyopathy. Hypertension greatly augments the risk of cardiac failure in diabetic persons. Consequently ,&blockers, which can depress contractility, are not a good choice of antihypertensive therapy in hypertensive diabetic patients. Both diabetic and hypertensive persons are particularly prone to silent or unrecognized myocardial infarctions (Tables VI and VII), necessitating a high degree of vigilance with routine periodic ECG examinations. The diabetic propensity to silent myocardial infarctions appears to be confined to men, whereas hypertensive patients of both sexes are at greater risk of a silent or unrecognized infarction. This increased susceptibility of the hypertensive patient to unrecognized infarction holds even excluding diabetic persons, those under treatment with antihypertensive agents, and those with left ventricular hypertrophy. lo In persons who have already sustained a myocardial infarction, both diabetes and
April
1272
Kannel,
Wilson, and Zhang
~msrican
Females 55-64 years old, Framingham Fig.
3.
Prevalence
hypertension predispose to recurrences and cardiac failure.’ ’ Examination of secular trends in the prevalence of glucose intolerance indicates a threefold increase from the 1950s through the end of the 1970s (Fig. 3). Prevalence of hypertension has failed to decrease if those persons receiving treatment with normalized pressures are included. Clearly, greater efforts at primary prevention of both conditions are needed, and this should include weight control, exercise, limitation of salt intake, and control of blood lipid levels. DISCUSSION
Three decades of epidemiologic research from the Framingham Heart Study indicate that both diabetes and hypertension are powerful predisposing factors for the development of atherosclerotic disease, including coronary heart disease.‘, 4l “-I5 Both conditions increase the propensity to cardiovascular disease twofold to threefold. The increased vulnerability of diabetic women compared with men is unexplained. Diabetes is one of the few conditions that eliminates the female advantage over men. A diminishing impact in women with advancing age suggests that late onset diabetes may be less atherogenic. Impaired glucose tolerance is associated with cardiovascular events whether manifested by a diagnosis of diabetes, hyperglycemia, or glycosuria. Blood glucose levels are a graded independent risk factor for cardiovascular disease in women, even within a relatively normal blood sugar range. Epidemiologic studies indicate an association be-
Heart
1991 Journal
Study
of glucose intolerance.
tween hyperinsulinemia, insulin resistance, and hypertension, even in nonobese people.16 Also, insulin resistance has been found in hypertensive persons who are neither obese nor diabetic, which suggests a basic relationship between insulin and blood pressure elevation. Insulin is known to enhance reabsorption of sodium by the kidney impeding sodium diuresis, resulting in higher pressures. Whatever their pathogenetic or etiologic relationships, both diabetes and hypertension each enhance atherogenesis. Diabetes and hypertension share in common predisposing obesity and insulin resistance. The combination of impaired glucose tolerance, hypertension, insulin resistance, obesity, lipoprotein lipase deficiency, hypertriglyceridemia, and low HDL cholesterol often occurs jointly, and when present in combination, powerfully predisposes to cardiovascular sequelae. It appears important to treat hypertension selectively in diabetic persons, because this appears to delay the cardiovascular sequelae of diabetes. Furthermore, with the increasing prevalence of glucose intolerance, greater attention should be given to the metabolic effects of antihypertensive drugs, because the multiple risk factors are not readily evident. The antihypertensive agent used in these patients should not adversely affect atherogenic blood lipid levels, glucose tolerance, uric acid, and magnesium or potassium balance. Agents used must also not induce orthostatic hypotension, renal hypoperfusion, or sexual dysfunction, to which the diabetic patient is prone. Although traditional diabetic therapy has emphasized blood glucose control and its regulation to avoid
Volume Number
121 4, Part 2
large- or small-vessel disease,17 the modern approach clearly requires a broader approach to diabetes and borderline diabetic conditions.1a-20 Individuals with adult onset diabetes are more likely to develop premature cardiovascular disease and to die of its complications or consequent cardiac failure. A multifactorial intervention to treat blood pressure, obesity, and lipids, in addition to managing glucose levels, is definitely needed. REFERENCES
1. Kannel WB, McGee DL. Diabetes and cardiovascular risk factors: the Framingham Study. Circulation 1979;59:8-13. 2. Ruderman NB. Haudenshild C. Diabetes as an atherogenic factor. Prog Cardiovasc Dis 1984;26:373-400. 3. Fuller JH, Shipley MJ, Rose G, Jarrett RJ, Keen H. Coronary heart disease risk and impaired glucose tolerance: the Whitehall Study. Lancet 198O;i:1373-6. 4. Kannel WB. Risk factors in hypertension. J Cirdiovasc Pharmacol 1989;13 (suppl 1):4-10. 5. Kannel WB, Wolf PA, Garrison RJ. The Framingham Study Section 35: Survival following initial cardiovascular events. U.S. Dept Commerce 1988; PB 88-204029. 6. Nelson N. A photometric adaptation of the Somogyi method for the determination of glucose. J Biol Chem 1944;153:375-80. 7. Lipid Research Clinics Program. Manual of Laboratory Operations, vol 1: lipid and lipoprotein analysis. Bethesda, Md.: National Institutes of Health, 1974; DHEW publication no. (NIH) 75-628. 8. Swain WR, Feders MB. Fibrinogen assay. Clin Chem 1967; 13:1026-8.
Glucose intolerance
and hypertension
1273
9. Walker SH, Duncan DB. Estimation of the probability of an event as a function of several independent variables. Biometrika 1967;54:167-79. 10. Kannel WB, Dannenberg AL, Abbott RD. Unrecognized myocardial infarction and hypertension: the Framingham Study. AM HEART J 1985;109:581-5. 11. Abbott RD, Donahue RP, Kannel WB, Wilson PWF. The impact of diabetes on survival following myocardial infarction in men versus women: the Framingham Study. JAMA 1988; 260:3456-60. 12. Wilson PWF, Kannel WB, Anderson KM. Lipids, glucose intolerance and vascular disease: the Framingham Study. In: James C, Pometta D, eds. Monographs in atherosclerosis: vol 13. Basel: S Karger, 1985:1-11. 13. Keen H. Jarrett RJ. Fuller JH. McCartnev P. Hvoerelvcemia and arterial disease. Diabetes ‘1981;30 (suppl 2)149-53: 14. Kannel WB, McGee DL. Diabetes and cardiovascular disease: the Framingham Study. JAMA 1979;241:2035. 15. Kannel WB, Dawber TR, McGee DL. Perspectives of systolic hypertension: the Framingham Study. Circulation 1980;61: 1179-82. 16. Reaven GM, Greenfield MS. Diabetic hypertriglyceridemia, evidence for the clinical syndromes. Diabetes 1981;30 (suppl 2):66-75. 17. Schor S. The University Group Diabetes Program: a statistician looks at the mortality results. JAMA 1971;217:1671-5. 18. Haffner SM, Stern MP, Hazuda HP, Mitchell BD, Patterson JK. Cardiovascular risk factors in confirmed prediabetic individuals: does the clock for coronary heart disease start ticking before the onset of clinical diabetes? JAMA 1990;263:2893-8. 19. Stout RW. Insulin and atheroma: 20.year perspective. Diabetes Care 1990;13:631-54. 20. Zavaroni I, Bonora E, Pagliara M, et al. Risk factors for coronary artery disease in healthy persons with hyperinsulinemia and normal glucose tolerance. N Engl J Med 1989;320:702-6.
