Metabolic and cardiovascular abnormalities in patients with peripheral arterial disease

Metabolic and cardiovascular abnormalities patients with peripheral arterial disease John K. Vyden, M.B. Joanne Thorner, B.A. Koichi Nagasawa, M.D. Teruo Takano, M.D. Marsha F. Groseth-Dittrich, Robyn Perlow, M.D. H. J. C. Swan, M.D., Ph.D. Los Angeles,

B.A.

Calif.

In contrast to ischemic heart disease, little is known about possible metabolic and cardiovascular abnormalities in patients suffering from intermittent claudication occurring as a result of underlying peripheral arterial disease. Previous studies of patients with lower limb arterial disease have suggested that 97.5 per cent of them smoked, and the serum cholesterol level was abnormally high if the patients were nondiabetic.’ While there is good evidence that diabetes mellitus increases the prevalence and severity of arteriosclerosis obliterans, Schadt and his colleagues’ found co-existing diabetes mellitus in only 15 per cent of their patients with atherosclerotic occlusions of the femoral artery. Evidence that hypertension is an important factor in the pathogenesis of arteriosclerosis obliterans is not well documented, although Juergens and his co-workers’ have suggested that blood pressure levels in excess of 150/90 mm. Hg occur in about one quarter of patients with symptomatic lower limb arterial disease against an expected incidence of about 10 per cent. Thus, while some reports suggest that periphFrom the Department of Cardiology, Cedars-Sinai Medical the Department of Medicine, University of California, Calif. Supported in part by Contract No. PH-43-68-1333 Infarction Research Program, National Institutes Heart and Lung Institute, Health Education and Received

for publication

Nov.

1975,

Vol.

Center and Los Angeles,

under Myocardial of Health, National Welfare.

5, 1974.

Reprint requests: John K. Vyden, Cedars-Sinai Medical Center, 4833 wo29.

December,

in

90, No.

M.B., Department Fountain Ave., Los

6, pp.

703-708

of Cardiology, Angeles, Calif.

era1 arterial disease may be associated with some of the risk factors which operate in cerebrovascular and ischemic heart disease, it’s interrelationship with smoking, disturbed glucose tolerance, hyperlipoproteinemia, and concomitant cardiovascular abnormalities is in need of greater study. Hence, a study was undertaken of 28 patients with severe peripheral arterial disease and 28 control subjects in order to define further any metabolic abnormalities in this type of patient. Since it has been reported that all of 15 patients with “Buerger’s disease” had an elevated serum copper level3 the relationship of the anomaly in patients with peripheral arterial disease was also studied. Methods

Twenty-eight consecutive patients of an average age of 63 years (range, 43 to 84 years) with arterial disease, who attended the Peripheral Vascular Out-Patient Department of CedarsSinai Medical Center, were studied. All 28 patients had severe symptoms of lower limb claudication of over 1 year’s duration, so that the limit of walking in all cases was two blocks of ambulation. The pain of claudication was always relieved after several minutes of rest. In all patients the diagnosis was confirmed by standard vascular studies with a 4 limb pneumatic plethysmograph (Electrodiagnostic Instruments, Burbank, Calif.) and temperature-measuring equipment (Leeds and Northrup Speedomax Recorder,

American

Heart

Journal

703

Vyden

et al.

SMOKING % 100

50

4 1 n

-

ELECTROCARDIOGRAM % % 100 100

pco.05 50

50

v) Y

0

uo 0 P.V.D. CONTROL P.V.D. CONTROL

a f14 st3 uI

GLUCOSE TOLERANCE TEST

LIPOPROTEIN ELECTROPHORESIS

-

0 F

I-

i uoEi

‘.V.D. CONTROL

0

a

F‘.V.D. CONTROL

Fig. 1. This figure shows the incidence of smoking, ECG abnormalities, abnormal glucose tolerance tests, and abnormal lipoprotein electrophoresis in patients with peripheral vascular disease (PVD) and control subjects (control). The incidence of smoking, abnormal glucose tolerance, and hyperlipoproteinemia tend to be greater in PVD patients compared to control subjects but the difference was not statistically significant. The incidence of ECG abnormalities in PVD is signif!cantly greater than seen in the control subjects (p < 0.05).

Monterey Park, Calif.).‘, 5 Translumbar arteriography was also performed in those patients considered suitable for surgical repair. A further 28 patients of an average age of 59 years (range, 28 to 80) and suffering from leg pain acted as a control series. The cause of the leg pain in all the control patients was degenerative joint disease of the lumbar spine causing sciatic nerve compression and radiation of pain to the lower limb as confirmed by electromyelography and x-rays of the lumbar spine. In addition, all 28 control patients underwent standard plethysmographic vascular examination and no evidence of peripheral vascular disease (PVD) was found. All 56 patients underwent the following investigations. A standard 12-lead electrocardiogram (ECG), a 5-hour glucose tolerance test (100 Gm. of glucose orally), fasting serum cholesterol, fasting serum triglycerides, serum copper, hematocrit determination, and lipoprotein phenotyping by electrophoresis. These tests were performed as a standard laboratory investigation according to the protocols of the Division of Laboratories and Biochemistry of Cedars-Sinai Medical Center. All patients were prepared for several days with a high carbohydrate intake, prior to testing or glucose tolerance, and were fasted for 14 hours prior to lipoprotein measurements. The glucose tolerance tests were interpreted according to the criteria of Andres.” No patients were receiving

704

estrogen-type compounds. Bra&al arterial blood pressure was determined by auscultation. Statistical evaluation of the results was obtained with multivariant analysis and the Fisher Exact Test for contingency tables. Data relating to incidence of smoking, ECG abnormalities, lipoprotein, and glucose tolerance were compared with the &i-square test; differences in serum cholesterol, triglycerides, copper, and hematocrit were examined by Student’s t-test. Results

Twenty-eight patients with PVD were studied and the results compared with a similar number of control patients who suffered leg pain but were free of vascular disease. Smoking. Fifteen PVD patients had smoked at least five cigarettes daily in the five years prior to the study (Fig. 1). This was a higher prevalence than seen in the control series, in which nine patients smoked, but the difference was not statistically significant. Cardiovascular abnormalities. Sixteen PVD patients showed ECG abnormalities. The most common abnormality seen was evidence of left ventricular hypertrophy, which occurred in nine patients. In seven patients there was evidence of an old myocardial infarction. Two patients had a bundle branch block pattern (Fig. 2). The incidence of ECG abnormalities was great-

December,

1975, Vol. 90, No. 6

Peripheral

arterial

disease

CONTROL SUBJECTS

PERIPHERAL VASCULAR DISEASE

Fig. 2. The incidence of ECG abnormalities in patients with peripheral vascular disease (PVD) and control subjects. The incidence of ECG abnormalities is greater in PVD, particularly in the number of patients who show evidence of left ventricular hypertrophy and an old myocardial infarction. Abbreviations: MI = myocardial infarction, LVH = left ventricular hypertrophy, BBB = bundle branch block, AF = atria1 fibrillation.

SERUM CHOLESTEROL

I mg/lOOml

I

mg/lOOml

250-

.L

SERUM COPPER

SERUM TRIGLYCERIDE JJg/lOOfr

200

II II-

HEMATOCRIT

96 40

150

c

E

r

P
200-

ISO-

11

150 P.V.D.

CONTROL

100

1

P(O.05 l-

100

20

h 50, 1~ P.V.D. CONTROL P.V.D. CONTROL

0 P.V.D. CONTROL

Fig. 3. The level of serum cholesterol, serum triglycerides, serum copper, and hematocrit in patients with peripheral vascular disease (PVD) and control subjects (control). Mean serum triglycerides and copper determinations are higher in PVD patients (p < 0.05). Mean serum cholesterol and hematocrit levels in PVD patients are most identical with those found in control subjects (mean I S.E.M.).

er than that seen in the control subjects (p < 0.05) (Fig. 1). In two thirds of the control group, the ECG was normal. Of the remaining subjects, five had evidence of left ventricular hypertrophy and one each had evidence of an old myocardial infarction, atria1 fibrillation, and two patients had bundle branch block. In PVD patients, mean cuff brachial diastolic blood pressure was 88 mm. Hg, with eight patients having diastolic pressures of 100 mm. Hg or greater. This reading was higher than in control patients, in whom diastolic pressure averaged 83

American

Heart Journal

mm. Hg (p < 0.05). Similarly, mean brachial systolic blood pressure of 156 mm. Hg was higher in PVD patients than the control reading of 144 mm. Hg (p < 0.05). In all, 18 of the 28 PVD patients were hypertensive (systolic pressure of 150 mm. Hg or greater, diastolic pressure of 90 mm. Hg or greater). Metabolic abnormalities. The glucose tolerance tests was abnormal in eight PVD patients, which was not significantly different from the six abnormal tests found in the control group (Fig. 1).

705

Vyden

et al.

TYPE (4%)

PERIPHERAL VASCULAR DISEASE

CONTROL

SUBJECTS

Fig. 4. The incidence of lipoprotein abnormalities in patients with peripheral vascular disease and control subjects. A higher incidence of lipoprotein abnormalities is seen in patients with peripheral vascular disease (32 per cent) when compared to control subjects (12 per cent).

Mean serum cholesterol level was 252 mg. per 100 ml. in PVD patients; in control subjects the average level was 247 mg. per 100 ml. When these values were compared to the normal range adjusted for age,’ only one PVD patient had an abnormally high serum cholesterol value, which was identical to one subject seen in the control group. In contradistinction, mean serum triglyceride levels showed a marked contrast between the two groups of patients. In PVD patients, mean serum triglyceride level was 198 mg. per 100 ml., which was higher than the level of 151 mg. per 100 ml. (p < 0.025) seen in the control group (Fig. 3). When these values were compared to the normal range adjusted for age, exactly half of the PVD patients had abnormally higher serum triglyceride levels, in contrast to five abnormal readings in the normal group. Type IV hyperlipoproteinemia was found in seven and Type V in one PVD patient. In the control group Type IV was found in only three subjects (Fig. 4). The presence of Type IV hyperlipoproteinemia showed a positive correlation with increased serum triglyceride levels. Hematocrit levels were nearly identical in both groups of subjects (Fig. 3). Statistical examination by multivariant analysis showed a significant correlation (p < 0.01) between a diabetic glucose tolerance test, elevated triglycerides, hyperlipoproteinemia, and high hematocrit levels. Serum copper levels were elevated in seven of the 28 PVD patients. The average level of 133 pg per 100 ml. in PVD patients was higher than the 706

mean value of 118 pg per 100 ml. seen in control subjects (p < 0.05). Analysis showed a statistical relationship (p < 0.05) between an elevation of serum copper levels and a history of current cigarette smoking. An elevated serum copper level was not related to any other of the variables examined. Discussion

The natural history of peripheral vascular disease needs to be better understood. One important reason for this is that, while symptoms and signs of peripheral arterial disease usually become manifest late in life, significant impairment of limb flow is often detectable for a decade or more before the onset of symptoms.s Another reason for the further need for identification of factors which may be contributing to the development of peripheral vascular disease is the relative accessibility of the peripheral vessels for diagnostic study which may allow the early identification of patients with asymptomatic arteriosclerotic disease before the more serious consequenes of the same disease process involve the myocardium or central nervous system. In the present study the number of patients studied is small by some standards, but analysis of the results obtained shows that the correlative trends are strong and that the further enlargement of the series by a doubling or tripling of the number of patients studied does not appear to be justified. The Framingham study showed an increased risk of intermittent claudication in patients with angina and coronary heart disease, suggesting a December,

1975, Vol. 90, No. 6

Peripheral

common underlying basis for claudication and coronary disease.9 This study showed that the principal hazard for subjects with claudication appeared to derive from an increased propensity to cardiovascular morbidity and death rather than from the consequences of impaired circulation to the limb. These findings are again supported in the present study in that, whereas the mean cuff brachial diastolic blood pressure was 88 mm. Hg, which may be normal for this age group, eight of the 28 PVD patients had brachial diastolic pressures of 100 mm. Hg or greater. In all, 18 of the 28 PVD patients were hypertensive (brachial systolic pressure of 150 mm. Hg or greater; diastolic pressure of 90 mm. Hg or greater). This is in contrast to the study of Juergens and his colleagues, who found this level of blood pressure in only 25 per cent of patients with lower limb arterial disease. Patients with arterial disease of the legs usually have lower blood pressure in the legs than in the arms, which may be related to the stenotic pressures in their lower limbs, but further investigation into the responsible mechanisms is needed. The incidence of ECG abnormality encountered was 63 per cent, with seven patients showing evidence of an old myocardial infarction. While no direct relationship was seen between diastolic pressure levels and the type of ECG abnormalities detected, 25 of 28 PVD patients (89 per cent) had either an ECG abnormality or a cuff pressure of 90 mm. Hg or greater. This incidence was greater than the finding of the same abnormalities in 19 of the 28 control subjects (p < 0.05). Since this incidence of cardiovascular abnormalities is so high, it appears essential that a patient presenting symptomatology of PVD must be thoroughly examined for other signs of abnormality in the cardiovascular system. This would be particularly important if major peripheral vascular surgery is contemplated. The high incidence of multisystem illnesses in patients with PVD is again demonstrated if metabolic abnormalities are considered. Twenty-one of twenty-eight PVD patients (75 per cent) had metabolic abnormalities. If both cardiovascular and metabolic abnormalities are considered, then 26 of 28 PVD patients (93 per cent) had abnormalities in addition to their PVD. In the present study there exists a strong correlation between abnormalities of glucose tolerance, elevated triglycerides, and hyperlipoAmerican Heart Journal

arterial

disease

proteinemia, as is also commonly encountered in patients with coronary heart disease. The most common disturbance of hyperlipoproteinemia encountered was of the pre-beta type IV, which confirms the findings of Greenhalgh and colleagues’” with respect to PVD. An interesting negative finding in this group of patients with profound atherosclerotic vascular disease is that at the time of measurement, the serum cholesterol level was elevated in only one of them. This is in contrast to the earlier study of Juergens and co-workers,’ who found that in a series of nondiabetic patients with arteriosclerosis obliterans, mean cholesterol level was 50 mg. per 100 ml. higher than that of the control patients. In the present study the mean cholesterol level of the nondiabetic patients was 244 mg. per 100 ml. which was nearly identical to that of the present control series. Juergens also found in his study the incidence of smoking to be 97.5 per cent whereas in the present study only 15 of the 28 patients smoked. When other factors known to predispose to accelerated vascular disease (such as smoking, the presence of hypertension, diabetes mellitus, or hyperlipidemia) are considered, then every patient in the present study had such a risk factor evident, with the majority of them having two or more risk factors present. Khandekar and co-workers’ have reported that 100 per cent of the patients with Buerger’s disease had elevated serum copper levels and the suggestion has been made by others that possibly this abnormality may be useful as a screening test for PVD. In this study a positive correlation (p < 0.05) was found between a history of current smoking and an abnormally high serum copper level. No relationship was found between serum copper and any other of the measurements obtained in this group of patients. The fact that serum copper was normal in 21 of the 28 PVD patients (75 per cent), suggests that an estimation of serum copper used routinely as a screening test for arteriosclerotic PVD is not indicated. Whether or not this test would be of value in differentiating “Buerger’s disease” from nonsmoking patients with arteriosclerotic PVD needs to be determined. Summary

Twenty-eight consecutive patients of an average age of 63 years with intermittent claudication secondary to underlying peripheral arterial 707

Vyden et al.

disease were studied for evidence of metabolic or other cardiovascular abnormalities and the results obtained were compared with those of 28 matched control subjects free of vascular disease. Patients with peripheral arterial disease had significantly higher levels of systolic and diastolic blood pressure, a greater incidence of ECG abnormalities, lipoprotein abnormalities, elevated serum triglycerides, and serum copper. The incidence of smoking and abnormal glucose tolerance, while higher in peripheral arterial disease patients, was not statistically significant. Hematocrit and serum cholesterol levels were nearly identical in both groups of patients. Twenty-six of the 28 patiens with peripheral arterial disease had either a cardiovascular or a metabolic abnormality, indicating the high incidence of multisystem illness in this disorder. The epidemiologic data in peripheral arterial disease are similar to those in coronary artery disease but some measurements contrast sharply, such as the apparent normal level of serum cholesterol in patients with peripheral arterial disease. The authors wish to acknowledge the statistical assistance of Mr. Alex Dubehnan and the editorial assistance of Mrs. Sharman Jar&on. Computing aeaistance was obtained from

708

the Health Sciencea Computing Facility, University of California, Los Angeles, sponsored by NIH Special Research Resources Grant RR-3. REFERENCES

1. Juergene, J. L., Barker, N. W., and Hines, E. A.: Atherosclerosis obliterans: Review of 520 casea with special reference to pathogenic and prognostic factors, Circulation 21:1&3.1960. 2. Schadt, D. C., Hines, E. A., Jr., Juergens, J. L., and Barker, N. W.: Chronic atherosclerotic occlusion of the femoral artery, J. A. M. A. 175:937, 1961. 3. Khandekar, J. D., Sepaha, G. C., and Mukerji, D. P.: Serum copper in peripheral vascular disease, N. Engl. J. Med. 283:319, 1970. 4. Strandnesa, D. E.: Peripheral arterial disease, Boston, 1969, Little, Brown & Company, pp. 92-113. 5. Winaor, T., and Hyman, C.: A primer of peripheral vascular diseases, Philadelphia, 1965, Lea & Febiger, Publishers, pp. 66-160. 6. Andres, R.: Angina and diabetes, Med. Clin. North Am. 55:635, 1971. 7. Jones, R. J.: The hyperlipoproteinemias: Detection, diagnosis, and management, Med. Clin. North Am. 57:61, 1973. 8. Widmer, L. K., Breensher, A., and Kannel, W. B.: Occlusion of peripheral arteries. A study of 6,406 working subjects, Circulation 30:636, 1964. 9. Kannel, W. B., Skinner, J. J., Schwartz, M. J., et al.: Intermittent claudication incidence in the Framingham study, Circulation 4 1: 875, 1970. 10. Greenhalgh, R. M., Rosengarten, D. S., Mervart, I., et al.: Serum lipids and lipoproteins in peripheral vascular disease, Lancet 2:947, 1971.

December,

1975, Vol. 90, No. 6