FREQUENCY OF ISCHÆMIC EXERCISE E.C.G. CHANGES IN SYMPTOM-FREE MEN WITH VARIOUS FORMS OF PRIMARY HYPERLIPÆMIA

Saturday 5 July 1975

FREQUENCY OF ISCHÆMIC EXERCISE

studied exercise

E.C.G. CHANGES IN SYMPTOM-FREE MEN WITH VARIOUS FORMS OF

lipxmic

PRIMARY HYPERLIPÆMIA

Subjects

LARS GÖRAN EKELUND LARS A. CARLSON ANDERS G. OLSSON King Gustaf V Research Institute and Department of Clinical Physiology and Internal Medicine (Lipid Unit), Karolinska Hospital, Stockholm, Sweden

Serum cholesterol and triglycerides measured in approximately 12,000 men attending a screening centre. 130 symptom-free men (aged thirty-five to sixty-five) were selected from the top 2% with the highest lipid values. They, and 59 normolipæmic controls, were studied by recording electrocardiograms (E.C.G.) before and during exercise. The frequency of so-called ischæmic E.C.G. changes (ST-segment depressions Minnesota code 4·1-4·4) increased with age both in controls and in the hyperlipæmic group. Ischæmic E.C.G. changes were significantly more common in all types of hyperlipæmia (types IIA, IIB, III, and IV) than in controls. The high frequency of the exercise E.C.G. changes in symptom-free hyperlipæmic men reinforces the argument for early treatment of hyperlipæmia to prevent ischæmic heart-disease.

Summary

were

Introduction VARIOUS forms of hyperlipæmia are known to be of importance in the development of ischaemic heartdisease (I.H.D.) 1,2 and are commonly found in patients with I.H.D.3-5 Well-defined manifestations of the disease, including sudden death, myocardial infarction, and angina pectoris, have been used as criteria of I.H.D. I.H.D. is generally associated with advanced coronary atherosclerosis and coronary-artery disease (C.A.D.).6 However, C.A.D. takes many years to develop, and does not produce symptoms for a considerable time. Little is known of the presence of asymptomatic C.A.D. in apparently healthy subjects with various forms of hyperlipaemia. Asymptomatic C.A.D. may give rise to signs of myocardial ischaemia under certain circumstances, and such ischsemia may be detected by recording electrocardiograms (E.C.G.) during and after exercise. Since a demonstration of important myocardial ischaemia in various forms of hyperlipxmia would be of both practical and theoretical interest we 7923

E.c.G.s

in

apparently healthy hyper-

men.

Methods Serum cholesterol and triglyceride (T.G.) levels were measured in fasting subjects attending a health screening centre in Stockholm (Metropol Health Control Centre). This centre examines about 5000 employees per year, about 60% of whom are men, from various companies offering this service free of charge during working hours. 95% of eligible employees attend. The results are available only to the person examined. People who regarded themselves as healthy and who were not found to have any important disease by the physicians at Metropol but had pronounced fasting hyperlipaemia (cholesterol above 350 mg. per 100 ml. and T.G. above 3-5 mmol per litre) were referred to the Lipid Unit, Karolinska Hospital, for further examination. Age-matched controls with serumlipid concentrations below 300 mg. cholesterol per 100 ml. and 2 mmol T.G. per litre were selected randomly from Metropol attenders. A history and clinical examination were taken at the Lipid Unit. Subjects were excluded from the study for the following reasons: history or presence of atherosclerotic vascular disease, hypertension, congestive heart-failure, myocarditis, endocrine and liver disease, anaemia, serum-electrolyte disturbances, and malignant disease. Subjects on the following chronic drug treatments were also excluded: hypolipsemic drugs; digitalis; diuretics; insulin or oral antidiabetics; cortisone; dilators; adrenergic &bgr;-blocking agents; antihypertensives; appetitereducing agents; anticoagulants; and thyroid derivatives. 7% of the men aged more than thirty-five years with hyperlipaemia who were called to the study did not come. 29 were excluded for the reasons given above. In addition 30 men were not included here because they were normolipaemic when examined at the Lipid Unit. Men with normal serum-lipid concentrations were offered the same examination as the hyperlipsemic (H.L.P.) subjects. After exclusions, according to the same criteria as the H.L.P. group, a final control sample of 59 men, 49 aged more ’than thirty-five years, remained. Chemical Methods All cholesterol7 and T.G.8 analyses were performed on frozen serum samples at King Gustaf V Research Institute by semi-automated procedures on an AutoAnalyzer Model 1

(Technicon Co). Hyperlipoproteinæmia (H.L.P.) was classified three months after diagnosis, according to the principle of Fredrickson et al.9 as modified 10 by the results of quantitative lipoprotein analysis (L.P.).ll The results of the L.P. analysis 12 and the regression analysis of L.P. concentrations on ischsemic exercise E.c.G.13 will be reported in detail elsewhere. There was no difference in mean age between the different H.L.P. groups and control groups within the reported age-groups. A

2 The Exercise Test an

Procedure.—Exercise tests were performed sitting on electrically braked bicycle ergometer (EM 370, Siemens-

Elema) controlled by a heart-rate controller unit (EMT 252, Siemens-Elema) which automatically adjusted the work-load of the bicycle so that a preselected heart-rate could be obtained during work.14 The subjects exercised for six minutes on consecutive loads beginning with a heart-rate of 90 beats per minute and increasing every sixth minute by 20 beats per minute. The exercise was continued until exhaustion or was stopped in a few cases because E.c.G. changes became pronounced. The E.c.G. was recorded, using six limb leads and six chest leads, before and after exercise. During exercise six chest leads were recorded continuously. the ST segment.- The E.C.G. was interof us (L. G. E.) without knowledge of whether it came from a control or H.L.P. subject. A magnifier was used for the visual interpretation and the E.C.G. changes were coded according to the Minnesota criteria.1S The ST level was measured 0-08 seconds after the junction point. There were two main types of sT-segment depressions. The first was either horizontal or downward sloping or straight and very slowly ascending (slope <0-2 mV per second) which we called horizontal ST depression. In this type there were five subgroups: (4’1) an ST depression of ≥0.15 mV; (4-2) an ST depression of 0-1-0-14 mV; (4-3) an ST depression of 0.05-0.09 mV; (4-4) no sT-j depression exceeding 0’05 mV but ST-segment depression of more than 0-05 mV; and (4-5) no ST-J depression but The ST-segment depression of less than 0’05 mV. second main category was junctional depressions, which were coded as 4’6 or 4-7 depending on whether the depression was >0. 15 mV or 0.05-0.14 mV, respectively. If there were any doubts whether an ST depression should be classified as a horizontal or junctional type, it was coded as junctional. E.C.G.s were classified immediately after and two minutes after exercise. Statistical methods.-Fisher’s exact probability test (two-tailed) was used for comparison of proportions. 16

Coding of

preted by

one

Results

Type IIA, IIB, III, and IV H.L.P. was present in 26, 14, 9, and 81 men, respectively. E.C.G.

at

Rest

differences in resting suggestive of myocardial ischemia There

were no

E.C.G. or

changes

infarction

between controls and any H.L.P. group.

Fit;. 2-Frequency of ST-segment depressions in various types of hypertipsemia.

Exercise Test All male H.L.P. groups had a lower physical working capacity than the controls. 14 There were no differences in mean heart-rate at end of exercise or in total time exercised between any H.L.P. group and the controls. The most

striking symptoms during exercise were exhaustion, fatigue, or leg discomfort. These were the reason for stopping exercise in 96% of the control group, 100% of type-IIA subjects, 93% of typeIIB subjects, 89% of type-in subjects, and 98% of type-iv subjects. There were signs of typical angina pectoris in one type-III and one type-iv man. Arrhytmias.—Ventricular premature beats occurred in 15-30% of the men in the different groups, with no significant difference between H.L.P. and control groups. Supraventricular premature beats were rarer and the frequencies did not differ between H.L.P. and control groups. A fifty-six-year-old man (type IIA H.L.P.) had ventricular tachycardia during exercise. H.L.P. and Exercise ST Depressions The frequency of ST depressions increased with age in both the controls and H.L.P. subjects (fig. 1). ST depressions were more commonly found in all than in control groups. In younger subjects frequency of ST depressions 41-44 was increased in type IIA and iv males. In significantly older age-groups significantly more ST depressions were seen in type nB and ill men. When all agegroups were taken together ST depressions 4.1-4.4

types of

H.L.P.

the

AGE - GROUP (yr.) Fit. 1—Freq acnçy of ST-segment depressions (4 1-44) in relation to age in controls and in hyperlipæmic men.

3

significantly more common in all male H.L.P. groups (fig. 2). The distributions of ST depressions 41-42 and 41-44 were similar-i.e., they were most common were

found

in type not

HA

to

and

be

IIB men.

However, differences

were

statistically significant. Discussion

ST Depressions During Exercise recorded during or after exercise are a sensitive method of detecting C.A.D. and predicting its outcome."-" The horizontal ST depressions (Minnesota code 4’1-4’3) occurring during exercise are generally taken as indicating ischaemia, since their frequency correlates with that of I.H.D.20-23 Ischxmic s.T. depressions are, however, not specific to C.A.D. and could be interpreted as indicating increased sympathetic tone, anaemia, electrolyte disturbances, or treatment with cardiotropic drugs. However, in the absence of these conditions, ischxmic ST depressions can be regarded as an indicator of myocardial ischemia which is often due to c.A.D. The reliability of horizontal ST depressions in predicting I.H.D. varies between 60 and 85 % and increases if one only accepts ST depressions of 0.1 mV (Minnesota code 41-42) or more.24-26 However, in this instance the sensitivity decreases especially if the work-load is far from maximum, as in Master’s double stepteSt.21 One way to increase the sensitivity is to use a near maximum work-load.23 The knowledge of this type of exercise testing in detecting silent C.A.D. is not yet as extensive as that from earlier submaximal tests, but several reports 23,27-29 indicate that sensitivity and specificity are high. inclusion of horizontal ST depressions of the magnitude of 005 mV (Minnesota code 4-3-4-4) will increase the sensitivity, but there is a risk of a lower degree of specificity. Available evidence and the conditions of the study thus indicate that the sT-segment depressions we determined were largely due to myocardial ischaemia, which in turn was most probably due to coronary atherosclerosis. In healthy subjects the frequency of horizontal ST depressions during exercise varies with sex and age. 30 At age forty the frequency in males is about 10% and 30% at between fifty-five and sixty.30 The frequency of ST depressions in the control groups of the present study accords with previous results in healthy men.30-32 The frequency of ST depressions was increased in all types of H.L.P. Since the 4 types of H.L.P. were due to increases of V.L.D.L. and/or L.D.L., it seems reasonable to assume that V.L.D.L. and L.D.L. were the cause of the myocardial ischaemia. This hypothesis was further substantiated by the results of regression analysis performed on the material of this study and reported elsewhere 13 which has shown a direct, positive relation between the concentration of both V.L.D.L. and L.D.L. and ST depressions. Furthermore, when smoking, bloodpressure (hypertensives being excluded), and glucose tolerance were taken into account in the multiple regression analysis, the frequency of ST depressions was not increased." V.L.D.L., and particularly L.D.L., may filter into the arterial wall, as suggested in the L.P. filtration hypothesis of atherosclerosis. 33 L.D.L. 35 and V.L.D.L. 36 are

Significance of E.c.G.s

be present in the arterial wall. The amount of L.D.L. recovered at necropsy in aortic intimas correlated with the serum levels of cholesterol some time before death.36 Also the glucose aminoglucans of the aortic wall interact with and bind L.D.L. and V.L.D.L .37 In the light of these studies it seems possible that our finding of increased frequency of ST depressions with increasing V.L.D.L. and L.D.L. concentrations may be the result of an increased atherosclerosis due to raised said

to

concentrations in the coronary arteries. One of several important clinical aspects emerging from this study is thus the combined importance of both V.L.D.L. and L.D.L. The finding of E.C.G. changes suggesting myocardial ischaemia in healthy men with H.L.P. reinforces the argument for early and intensive detection and treatment of H.L.P. in the prevention of l.H.D. This work was supported by grants from the Petrus and Augusta Hedlunds Foundation, the Swedish National Association against Heart and Chest Diseases, and the Swedish

L.P.

Medical Research Council. Requests for reprints should be addressed to L. A. C., King Gustaf V Research Institute, Karolinska Hospital, S-104 01 Stockholm 60, Sweden. REFERENCES 1. Carlson, L. A., Böttiger, L. E. Lancet, 1972, i, 865. 2. Kannel, W. B., Dawber, T., Friedman, G., Glennon, W., McNamara, P. Ann. intern. Med. 1964, 61, 888. 3. Carlson, L. A. Acta med. scand. 1960, 167, 399. 4. Carlson, L. A., Ericsson, M. Atherosclerosis, 1975, 21, 435. 5. Goldstein, J. L., Hazzard, W. R., Schrott, H. G., Bierman, E. L., Motulsky, A. G. J. clin. Invest. 1973, 52, 1533. 6. Mitchell, J. R. A., Schwartz, C. J. Arterial Disease. Philadelphia, 1965. 7. Block, W. D., Jarrett, K. J., Leoine, B. in Automation in Analytical Chemistry (edited by L. T. Skeggs); vol. I, p. 345. New York, 1965. 8. Kessler, G., Lederer, H. ibid. p. 341. 9. Fredrickson, D. S., Levy, R. I., Lees, R. S. New Engl. J. Med. 1967, 276, 34. 10. Beaumont, J. L., Carlson, L. A., Cooper, G. R., Fejfar, Z., Fredrickson, D. S., Strasser, T. Bull. Wld Hlth Org. 1970, 43, 891. 11. Carlson, K. J. clin. Path. 1973, 26, suppl. 5, 32. 12. Olsson, A. G., Carlson, L. A. Acta med. scand. 1975, suppl. 580. 13. Olsson, A. G., Ekelund, L. G., Carlson, L. A. ibid. (in the press). 14. Ekelund, L. G., Olsson, A. G. ibid. (in the press). 15. Scandinavian Committee on E.C.G. Classification. ibid. 1967, suppl. 481. 16. Siegel, S. Nonparametric Statistics for the Behavioural Sciences. Tokyo, 1956. 17. Åstrand, I. Scand. J. clin. Lab. Invest. 1969, 23, 271. 18. Beard, E. F., Garcia, E., Burke, G. E., Dear, W. E. Dis. Chest, 1969, 56, 405. 19. Blackburn, H., Taylor, H. L., Keys, A. Am. J. Cardiol. 1970, 25, 85. 20. Call, R. W., Clyman, B., Kaserman, D. R., Eckardt, D. J. occup. Med. 1974, 16, 9. 21. Doan, A. E., Peterson, D. R., Blackmon, J. R., Bruce, R. A. Am. Heart J. 1965, 69, 11. 22. Doyle, J. T., Kinch, S. H. Circulation, 1970, 41, 545. 23. Kattus, A. A., Jorgensen, C. R., Worden, R. E., Alvaro, A. B. ibid. 1971, 44, 585. 24. Froelicher, V. F. Prev. Med. 1973, 2, 592. 25. Most, A. S., Hornsten, T. R., Hofer, V., Bruce, R. A. Archs intern. Med. 1968, 121, 225. 26. Robb, G. P., Marks, H. H. J. Am. med. Ass. 1967, 220, 918. 27. Bruce, R. A. in The Heart (edited by J. W. Hurst); p. 314. New York, 1974. 28. Cumming, G. R., Dufresne, C. Can. med. Ass. J. 1972, 106, 649. 29. Detry, J.-M. in Exercise Testing and Training in Coronary Heart Disease (edited by S. A. Arscia); p. 83. Bruxelles, 1972. 30. Åstrand, I. in Measurement in Exercise Electrocardiography (edited by H. Blackburn); p. 69. Springfield, Illinois, 1969. 31. Bruce, R. A., Gey, G. O., Cooper, M. N., Fisher, I. D., Peterson, D. R. Am. J. Cardiol. 1974, 33, 459. 32. Strandell, T. Acta med. scand. 1963, 174, 479. 33. Page, I. H. Circulation, 1954, 10, 1. 34. Smith, E. B., Slater, R. S. in Atherosclerosis (edited by R. J.

Jones); p. 42. Berlin,

1970.

35. Walton, K. W. in Atherosclerosis (edited by G. Schettler and A.

Weizel); p. 93. Berlin, 1974. Smith, E. B., Slater, R. S. Lancet, 1972, i, 463. 37. Iverius, P. H. Acta Univ. Upsal. 1971, 113. 36.