Coronary atherosclerosis and myocardial hypertrophy in relation to body fat distribution in healthy women: an autopsy study on 33 violent deaths

Transcription

1 International Journal of Obesity (1997) 21, 43±49 ß 1997 Stockton Press All rights reserved 0307±0565/97 $12.00 Coronary atherosclerosis and myocardial hypertrophy in relation to body fat distribution in healthy women: an autopsy study on 33 violent deaths ML Kortelainen and T SaÈrkioja University of Oulu, Department of Forensic Medicine, Kajaanintie 52 D, FIN Oulu, Finland OBJECTIVE: To determine the relationship between cardiovascular pathology and body fat distribution in healthy women with no ante mortem clinical evidence of cardiovascular disease. SUBJECTS: Thirty-three female forensic autopsy cases of sudden death from violent causes. METHODS: Body height and weight, the circumferences of the waist and hip and the thicknesses of the subscapular and abdominal subcutaneous fat were measured, and Body Mass Index (BMI) and Waist-to-Hip ratio (WHR) were calculated. Omental, mesenterial and perirenal fat deposits were weighted. Heart weight was indexed to height (2.7), the degree of coronary narrowing was determined in each artery, and myocardial collagen volume fraction and myocyte cross-sectional area were measured. RESULTS: The degree of coronary narrowing, heart weight in absolute terms and indexed to height (2.7), myocyte cross-sectional area and all the measures of obesity were signi cantly positively correlated with age. Regression of coronary narrowing on measures of obesity indicated that a quadratic model tted the data for BMI, waist circumference and intra-abdominal fat better than a linear one. After adjusting for age, the degree of coronary narrowing was related to tertiles of BMI, waist circumference, WHR and intra-abdominal fat, the severity of the narrowings being most marked in the second tertile of BMI (24.0±31.0), waist circumference (80±96 cm) and intraabdominal fat (500±1700 g), but in the third tertile of WHR (over 0.92). Regression on heart weight/height (2.7) on the aforementioned measures of obesity indicated a clearly linear association and heart weight indexed to height (2.7) was related to tertiles of BMI, waist circumference and WHR, and also to tertiles of intra-abdominal fat. CONCLUSIONS: The results suggest that body fatness and abdominal accumulation of fat are associated with the severity of coronary atherosclerosis and myocardial hypertrophy in women with no clinical evidence of cardiovascular disease, but the relationship between coronary lesions and BMI is not linear. Both coronary lesions and myocardial hypertrophy are more advanced as the numerical value for WHR increases in women. Future autopsy studies should be directed at young women with increased WHR in order to determine their risk of developing life-threatening lesions in the atherosclerosis-prone regions of the coronary tree. Keywords: Body fat distribution; atherosclerosis; heart; obesity; WAT Introduction Abdominal or android-type fat accumulation is considered to be an independent risk factor for cardiovascular disease in both men and women 1,2, and it is one of the characteristics of the metabolic syndrome known as the deadly quartet, along with glucose intolerance, lipid disorders and hypertension. 3 The close connection between cardiovascular disease and body fat distribution probably has an important impact on the gap in longevity between men and women. 4 There is a difference between the sexes in Waist-to- Hip ratio (WHR) at the same Body Mass Index (BMI), and men have been found to have more metabolically unfavourable intra-abdominal fat than women. 5,6 Correspondence: Dr M L Kortelainen. Received 27 July 1996; revised 20 September 1996; accepted 23 September 1996 Angiographic studies have revealed an association between abdominal obesity and clinically signi cant coronary narrowings both in men 7,8 and in women. 9,10 The association between atherosclerosis and obesity in general has been evaluated in a number of autopsy series, some of which did not indicate any positive association 11±13 but McGill et al 14 did nd a signi cant association between BMI and the extent of fatty streaks and raised lesions in the right coronary artery, although only in males. The relationship of the abdominal accumulation of fat to coronary atherosclerosis was not studied in the above autopsy surveys, but a previous one carried out at our department 15 indicated signi cant positive associations between BMI, waist circumference, subscapular fat thickness, perirenal fat weight and the severity of coronary arteriosclerosis in men with no antemortem clinical evidence of cardiovascular abnormalities, and also between myocardial hypertrophy and various measures of obesity. In order to

2 44 ascertain whether the fat distribution pattern in healthy women shows any comparable associations with the degree of coronary and myocardial pathology we decided to collect an autopsy series consisting of females who did not have any ante mortem clinical evidence of cardiovascular disease and had died suddenly of unexpected causes. This report follows in its general outlines the same principles as were used in the previous study on males. Material and methods Subjects Thirty-three female cases of sudden unexpected death were collected from medico-legal autopsies performed at the Department of Forensic Medicine, University of Oulu, Finland, during the years 1993± The subjects were not known to have had any ante mortem clinical signs of coronary artery disease, hypertension, lipid disorders or any other cardiovascular abnormalities or chronic diseases. Cases with reports of heavy alcohol consumption, long-term medication or marked ante mortem changes in body weight or with marked crush injuries were likewise excluded. The cases consisted of violent deaths from accidental causes (73%), homicide (3%) or suicide (24%). The data was collected from the medical and police records. The police records included a standard form, in which the results of scene investigations, interviews with family members, information on diseases, alcohol consumption and medication were presented. The medical records were obtained from the hospitals and/ or medical reception centers, and these contained the essential information available from each individual. In addition to the aforementioned criteria, we did not accept cases in which regular smoking, whether in the past or present, was reported. Since no quantitative estimates of smoking habits were available, it seemed reasonable to try to reduce its confounding effect by excluding those who were reported to be regular smokers either in their medical records or accoring to information given by the police (for example interviews with family members, investigations at the scene of the death), although occasional smokers could probably not be excluded by this method nor all the daily smokers. Anthropometric measurements After weighing, the cadavers were placed naked on the autopsy table in the supine position, and the following measurements were made by two autopsy technicians trained for the purpose: height, waist circumference, hip circumference and thickness of the abdominal and subscapular subcutaneous fat. The levels for measuring the circumferences and fat thicknesses were based on the same skeletal reference points as are used for clinical purposes. 16 Body Mass Index (BMI: weight divided by height squared) and Waist-to-Hip ratio (WHR) were calculated from the measurements. Intra-abdominal fat The greater omentum was excized free from intraabdominal tissues, mesenterial fat was excized free from the gut, and perirenal fat was removed from both sides. All fat deposits were weighed fresh. Heart and coronary arteries The heart was removed according to a systematic autopsy protocol, by cutting through the root of the aorta and other great vessels proximal to the atria, and washed externally. Blood clots were removed from the chambers and the heart was then weighed. The heart weight was indexed to height (2.7) (Heart weight/height (2.7)) for each individual by a modi cation of the method presented by de Simone et al 17 in which they indexed the left ventricular mass obtained by echocardiography to height (2.7) (left ventricular mass/height (2.7)) in order to minimize the effect of body size on the evaluation of left ventricular hypertrophy. The coronary arteries were examined by assessing the degree of arteriosclerosis separately the left main coronary artery, left anterior descending artery, left circum ex artery, and right coronary artery. The degree of arteriosclerosis was determined using the degree of luminal narrowing as the criterion for the severity of the disease in each artery. 18 A numerical value from 0±8 was given for each artery according to the degree of stenosis, as presented previously, 15 and the total narrowing score for each individual was obtained from the sum of the scores for the four arteries (maximum ˆ 32). Histology and morphometry Transversely cut slices through the total thickness of the anterior left ventricular wall and septum were taken from the heart, sectioned at 5 mm, and stained with Sirius red which is a speci c collagen stain. 19 The total collagen volume fraction was determined by computerized image analysis. The areas of connective tissue, muscle and interstitial space were identi ed according to their respective grey levels, and the total collagen volume fraction was calculated as the sum of all connective tissue areas in each section divided by the sum of all connective and muscle tissue areas. 20,21 Myocyte cross-sectional area was measured in transversely cut myocytes in each section divided into four quadrants, measuring at least ten myocytes in each quadrant. Statistical analysis All the analyses were performed using Statistical Package for the Social Sciences software (SPSS). 22 BMI, coronary narrowing score, the weights of the

3 intra-abdominal fat deposits and subscapular fat thickness showed some skewness in their distribution, and therefore logarithmic transformations were used in the calculations when needed. Pearson correlation coef cients were calculated between age and other variables, and between two variables after controlling for age or age and BMI together as covariates. The nature of the relationship between cardiac parameters and the various measures of obesity was investigated by tting linear and quadratic models with each cardiac parameter as a dependent variable and each measure of obesity as an independent variable. Tertiles of BMI, waist circumference, WHR and intra-abdominal fat were calculated, the cut-off points being 24.0 and 31.0 for BMI, 80 cm and 96 cm for waist circumference, 0.81 and 0.92 for WHR, and 500 g and 1700 g for intra-abdominal fat (sum of the weights of mesenterial, omental and perirenal fat). An analysis of variance with age as a covariate was used to compare the means for coronary narrowing across tertiles of BMI, waist circumference, WHR and intra-abdominal fat, and the means for heart weight/height (2.7) across the tertiles. Results The descriptive statistics, including means, standard deviations and ranges, are presented in Table 1. The ages of the women ranged from 19±63 y and all the indicators of obesity varied greatly, the lowest values of BMI, for example, being under 20.0 and the highest above The degree of coronary lesions was generally mild, the narrowing of the vessels never exceeding 50% in any of the individuals. Age had signi cant positive correlations with all the indicators of obesity, the degree of coronary narrowing, absolute heart weight and heart weight indexed to height (2.7) and myocyte cross-sectional area, but not with the myocardial collagen volume fraction (Table 2). When adjusted for age, waist circumference, WHR, and the thicknesses of abdominal and subscapular subcutaneous fat were all signi cantly positively correlated with the weights of mesenterial, omental and perirenal fat and with the sum of these weights (all P values 0.001). When adjusted for age and BMI together, only WHR was signi cantly positively correlated with the weights of all intra-abdominal fat deposits. The thickness of abdominal fat was also signi cantly correlated with the sum of the intra-abdominal fat deposits, but waist circumference was no more associated with any of the weights of the intra-abdominal fat deposits or with their sum value (Table 3). Testing of the linearity between the cardiac parameters and the indicators of obesity showed the linear model to t between heart weight/height (2.7) and BMI, body weight, waist circumference, WHR and intra-abdominal fat weight, while a quadratic regression model generally tted better in the case of the relation between coronary narrowing and the indicators of obesity, with the exception of WHR. After 45 Table 1 Descriptive statistics for the series of 33 autopsy cases Mean s.d. Range Table 2 Pearson correlation coef cients between age and other parameters Age (y) ±63 Height (cm) ±174 Weight (kg) ±115 BMI (kg/m2) ±43.3 Waist circumference (cm) ±129 WHR ±1.02 Abdominal subcutaneous fat (mm) ±75 Subscapular subcutaneous fat (mm) ±35 Intra-abdominal fat (g) ±2605 Perirenal fat (g) ±1176 Mesenterial fat (g) ±1120 Omental fat (g) ±1285 Coronary narrowing (score 0±32) 3 3 0±15 Heart weight (g) ±483 Heart weight/height (2.7) (g/m (2.7)) ±131 Myocyte cross-sectional area (mm 2 ) ±341 Collagen volume fraction (%) ±11.2 Height Weight 0.366* BMI 0.455** Waist circumference 0.529** WHR 0.575** Abdominal subcutaneous fat 0.415* Subscapular subcutaneous fat 0.484** Intra-abdominal fat 0.616*** Perirenal fat 0.567** Mesenteril fat 0.668*** Omental fat 0.440** Coronary narrowing 0.638*** Heart weight 0.564** Heart weight/height (2.7) 0.625*** Myocyte cross-sectional area 0.458** Collagen volume fraction *P < 0.05; **P < 0.01; ***P < Table 3 Partial correlation coef cients, adjusted for age and BMI, for waist circumference, WHR and subcutaneous fat in relation to intra-abdominal fat n33 Mesenterial fat Perirenal fat Omental fat Visceral fat Waist circumference WHR 0.368* 0.364* 0.337* 0.461** Subcutaneous fat Abdominal 0.356* ** 0.437** Subscapular *P < 0.05; **P < 0.01

4 46 Table 4 Partial correlation coef cients, adjusted for age, between heart weight/height (2.7) and BMI, waist circumference, WHR, subcutaneous fat and intra-abdominal fat n33 Heart weight/height (2.7) Myocyte cross-sectional area Collagen volume fraction BMI 0.471** Waist circumference WHR Subcutaneous fat Abdominal Subscapular Intra-abdominal fat 0.536** 0.517** Perirenal fat 0.588** 0.682*** Mesenterial fat 0.502** 0.436* Omental fat *P < 0.05; **P < 0.01; ***P < Figure 1 Means of coronary narrowing (score 0±32) across tertiles of Body Mass Index. F ˆ 8.026, P ˆ Analysis of variance, age adjusted Figure 3 Means of coronary narrowing (score 0±32) across tertiles of Waist-to-Hip ratio. F ˆ 6.596, P ˆ Analysis of variance, age adjusted adjusting for age, the degree of coronary narrowing varied signi cantly over the tertiles of BMI, waist circumference, WHR and intra-abdominal fat. The degree of coronary narrowing was most severe in the third tertile of WHR (Figure 3), but in the second tertile of BMI (Figure 1), waist circumference (Figure 5) and intra-abdominal fat (Figure 7). Heart weight varied from 218±483 g, or from 53± 131 when indexed to height (2.7) (Table 1). When adjusted for age, heart weight indexed to height (2.7) was signi cantly positely correlated with BMI and with the weights of mesenterial and perirenal fat and the sum of the intra-abdominal fat deposits (Table 4). Myocyte cross-sectional area had also signi cant positive correlations with all the other intra-abdominal fat weights except omental fat (Table 4). Heart weight indexed to height (2.7) varied signi cantly over the tertiles of BMI (Figure 2), WHR (Figure 4), waist circumference (Figure 6) and intra-abdominal fat (Figure 8), being lowest in the rst tertile of each of these measures of obesity and highest in the third tertile. Figure 2 Means of heart weight/height (2.7) across tertiles of Body Mass Index. F ˆ 9.911, P ˆ Analysis of variance, age adjusted Figure 4 Means of heart weight/height (2.7) across tertiles of Waist-to-Hip ratio. F ˆ 9.209, P ˆ Analysis of variance, age adjusted

5 47 Figure 5 Means of coronary narrowing (score 0±32) across tertiles of waist circumference. F ˆ , P ˆ Analysis of variance, age adjusted Figure 8 Means of heart weight/height (2.7) across tertiles of intra-abdominal fat. F ˆ , P ˆ Analysis of variance, age adjusted Figure 6 Means of heart weight/height (2.7) across tertiles of waist circumference. F ˆ 8.787, P ˆ Analysis of variance, age adjusted Figure 7 Means of coronary narrowing (score 0±32) across tertiles of intra-abdominal fat. F ˆ 8.823, P ˆ Analysis of variance, age adjusted Discussion The autopsy ndings presented here suggest that obesity is associated with the severity of coronary atherosclerosis and myocardial hypertrophy in women without any ante mortem clinical evidence of cardiovascular disease. An abdominal pattern of fat accumulation proved to be associated with subclinical coronary and myocardial pathlogy, as also observed earlier 15 in ante mortem healthy males, but there were some differences between the results. The women in the highest tertiles of BMI, of waist circumference and of intra-abdominal fat did not have the most severe overall degree of coronary arteriosclerosis as did the males as the most severe changes were seen in the second tertile of BMI (24.0±31.0), of waist circumference (80±96 cm) and of intra-abdominal fat (500±1700 g). This may result from a different distribution of BMI in these two autopsy surveys, because the cutoff point for the highest tertile of BMI in the male group was only 26.0, 15 and consequently the most obese male individuals were in fact only moderately overweight. The results of Adams- Campbell et al 23 based on coronary angiograms also showed an upside-down U-shaped relationship between BMI and coronary artery disease in an African-American population consisting of both males and females. The curvilinearity of the relation between the severity of coronary changes and BMI needs further autopsy-based investigation in a larger group of women with a similar distribution of BMI in order to see whether this phenomenon is reproducible and whether it might be explained by some factor not yet discovered. The fact that the coronary changes were most severe in the third tertile of WHR (over 0.92), supports clinical observations that women with a more masculine type of body structure have a greater cardiovascular risk. 4 In women WHR values over 0.92 clearly indicate abdominal obesity as the average female value of WHR has been reported to be under The signi cant positive correlations found here between WHR and all the measures of intra-abdominal fat, even when adjusted for BMI and age together, parallel the close relationship between WHR and intra-abdominal fat established by computed tomography. 24 Waist circumference proved to be a weaker indicator of intra-abdominal fat in women in this material. Cardiac size, when adjusted for age and indexed to body size showed a linear correlation with the measures of obesity, the relation being statistically sig-

6 48 ni cant with BMI and intra-abdominal fat. This is similar to the results obtained earlier for the males. 15 Slightly elevated blood pressure, representing one feature of the metabolic syndrome, may have been the underlying factor for the mild myocardial hypertrophy in spite of the absence of any clinically veri ed hypertension, since no other causes of hypertrophy such as valve diseases or cardiomyopathies were found. 25,26 An obesity-related increase in cardiac output and stroke volume could also be a stimulus to increase heart weight. The effect of stroke volume has been shown to be independent of that of blood pressure. 27 Myocyte cellular hypertrophy, which has also been reported in hypertensive individuals, 28 was seen here, but there was no signi cant increase in collagen, which is described as characteristic of hypertrophied hearts. 21 It is known, however, that the appearance of brous tissue can either accompany myocyte hypertrophy or be delayed until the process of myocyte growth. 29 The present results were probably in uenced by the accelerating effect of the menopause on the atherosclerotic process in the older subjects. There are also certain well-known general dif culties inherent in post mortem studies: the reliability of the ante mortem data, early autolytic changes during the post mortem period and their effects on tissue processing, and various aspects concerning proper adjustment for many possible confounding factors such as serum lipids, smoking (regular, past, present, passive etc.), alcohol consumption, physical activity, dietary habits and emotional stress. There are some earlier studies in which lipid analyses have been carried out on cadaver blood samples, 30±32 but considerable uctuations have been reported to take place in plasma lipid and lipoprotein values even during the rst 24 h after death. 33 In general, the concentrations of natural chemical substances are distorted rapidly by post mortem autolysis, 26 and therefore the time elapsing from death to blood sampling should be as short as possible and should be well standardized in order to obtain comparable results. Smoking has been checked by means of thiocyanate measurements in cadaver blood, 14 but the data on cigarette consumption during the subject's lifetime unfortunately still remain very unreliable. Conclusions The present autopsy ndings on female subjects indicate that body fatnesss is associated with coronary atherosclerosis and myocardial hypertrophy in women with no lifetime clinical diagnosis of any cardiovascular disease. Increased WHR is associated with more advanced coronary lesions, but their relationship to BMI is curvilinear, with the most advanced coronary narrowings found in the group of only moderately overweight women. On the other hand, BMI and the size of intra-abdominal fat deposits seem to have an apparent linear correlation with myocardial hypertrophy. Recent research has shown that the lipid-rich core originates early in the development of an atherosclerotic lesion, 34 and certain types of plaque are considered more vulnerable than the others. 35 Future investigations should be directed at young women with increased WHR in order to see whether their body structure carries any risk of developing lifethreatening lesions in the arteriosclerosis-prone regions of the coronary tree. References 1 Larsson B et al. Abdominal adipose tissue distribution, obesity, and risk of cardiovascular disease and death: 13 year follow up of participants in the study of men born in Br Med J 1984; 288: 1401± Lapidus L et al. Distribution of adipose tissue and risk of cardiovascular disease and death: a 12 year follow up of participants in the population study of women in Gothenburg, Sweden. Br Med J 1984; 289: 1257± Kaplan NM. The deadly quartet: upper body obesity, glucose intolerance, hypertriglyceridemia, and hypertension. Arch Intern Med 1989; 149: 1514± Kissebah AH, Krakower GR. Regional adiposity and morbidity. Physiol Rev 1994; 74 (4): 761± Lemieux S et al. Sex differences in the relation of visceral adipose tissue accumulation to total body fatness. Am J Clin Nutr 1993; 58: 463± Matsuzawa Y et al. Pathophysiology and pathogenesis of visceral fat obesity. Obes Res 1995; 3 (Suppl): 187± Hauner H et al. Body fat distribution in men with angiographically con rmed coronary artery disease. Atherosclerosis 1990; 85: 203± Zamboni M et al. Relation of body fat distribution in men and degree of coronary narrowings in coronary artery disease. Am J Cardiol 1992; 70: 1135± Hauner H, Bognar E, Blum A. Body fat distribution and its association with metabolic and hormonal risk factors in women with angiographically assessed coronary artery disease. Evidence for the presence of a metabolic syndrome. Atherosclerosis 1994; 105: 209± Clark LT et al. Obesity, distribution of body fat and coronary artery disease in black women. Amer J Cardiol 1994; 73: 895± Amad KH, Brennan JC, Alexander JK. The cardiac pathology of chronic exogenous obesity. Circulation 1965; 32: 740± Feinleb M et al. The relation of antemortem characteristics to cardiovascular ndings at necropsyðthe Framingham study. Atherosclerosis 1979; 34: 145± Joseph A et al. Manifestations of coronary atherosclerosis in young trauma victimsðan autopsy study. J Am Coll Cardiol 1993; 22: 459± McGill HC et al. Relation of glycohemoglobin and adiposity to atherosclerosis in youth. Arterioscler Thromb Vasc Biol 1995; 15: 431± Kortelainen M-L. Association between cardiac pathology and fat tissue distribution in an autopsy series of men without premortem evidence of cardiovascular disease. Int J Obesity 1996; 20: 245± Van der Kooy K, Seidel JC. Techniques for the measurement of visceral fat: a practical guide. Int J Obesity 1993; 17: 187± 196.

7 17 de Simone G et al. Left ventricular mass and body size in normotensive children and adults: assessing of allometric relations and impact of overweight. J Am Coll Cardiol 1992; 20: 1251± Champ CS, Coghill SB. Visual aid for quick assessment of coronary artery stenosis at necropsy. Clin Pathol 1989; 49: 887± Junqueira LCU, Bignolas G, Brentani RR. Picrosirius staining plus polarization microscopy, a speci c method for collagen detection in tissue sections. Histochem J 1979; 11: 447± Jalil JE et al. Fibrillar collagen and myocardial stiffness in the intact hypertrophied rat left ventricle. Circ Res 1989; 64: 1041± Brilla CG, Weber KT. Reactive and reparative myocardial brosis in arterial hypertension in the rat. Cardiovasc Res 1992; 26: 671± SPSS Inc. SPSS for Windows 6.1 SPSS Inc: Chicago, Adams-Campbell LL, Peniston RL, Kim KS, Mensah E. Body Mass Index and coronary artery disease in African-Americans. Obes Res 1995; 3: 215± Peiris AN et al. Relationship of anthropometric measurements of body fat distribution to metabolic pro le in premenopausal women. Acta Med Scand 1988; 723 (Suppl): 179± Silver MD (ed). Cardiac Pathology. 2nd edn. Churchill Livingstone: New York, Knight B. Forensic Pathology. 1st edn. Edward Arnold: London, Ganau A et al. Relation of left ventricular hemodynamic load and contractile performance to left ventricular mass in hypertension. Circulation 1990; 81: 25± Olivetti G et al. Myocyte cellular hypertrophy is responsible for ventricular remodelling in the hypertrophied heart of middle aged individuals in the absence of cardiac failure. Cardiovasc Res 1994; 28: 1199± Weber KT, Brilla CG, Janicki JS. Myocardial brosis: funcitonal signi cance and regulatory factors. Cardiovasc Res 1993; 27: 341± Enticknap JP. Serum fatty acids after fatal heart attacks. Lancet 1960; 1: Enticknap JP. Lipids in cadaveric sera after fatal heart attacks. J Clin Pathol 1961; 14: 496± Enticknap JP. Fatty acid content of cadaveric sera in fatal ischaemic heart disease. Clin Sci 1962; 23: 425± SaÈrkioja T et al. Stability of plasma total cholesterol, triglycerides, and apolipoproteins B and A-I during the early postmortem period. J Forensic Sci 1988; 33 (6): 1432± Guyton JR, Klemp KF. Development of the lipid-rich core in human atherosclerosis. Arterioscler Thromb Vasc Biol 1996; 16: 4± Schroeder AP, Falk E. Vulnerable and dangerous coronary plaques. Atherosclerosis 1995; 118 (Suppl): S141±S