Gender differences in coronary artery size per 100 g of left ventricular mass in a population without cardiac disease

The outcome of coronary artery disease (CAD) in women is worse than in men [1–8]. Recently, it has been shown in a large registry that early mortality of women <50 years with myocardial infarction is more than twice as high than that of men [9]. This may be due to the presence of a more unfavourable cardiovascular risk profile at the time women need to be treated for CAD. Primarily, women are older, have a higher incidence of diabetes, hypertension and congestive heart failure and more often undergo non-elective procedures than men. Even after multivariate adjustment of these factors, cardiovascular mortality in women is higher than in men [5, 8, 9]. The possibility has been discussed that men are better cared for than women [7, 11]. The role of coronary artery size as a predictor of mortality in patients with myocardial infarction and with coronary artery bypass grafting is controversial [1–5, 7, 10]. In the Coronary Artery Surgery Study, small body size and coronary artery caliber were the strongest predictors of perioperative mortality [3]. The existence of gender differences in coronary artery size has hitherto not been proven with 610 Original article S W I S S M E D W K LY 2 0 0 1 ; 1 3 1 : 6 1 0 – 6 1 5 · w w w. s m w. c h

The outcome of coronary artery disease (CAD) in women is worse than in men [1][2][3][4][5][6][7][8].Recently, it has been shown in a large registry that early mortality of women <50 years with myocardial infarction is more than twice as high than that of men [9].This may be due to the presence of a more unfavourable cardiovascular risk profile at the time women need to be treated for CAD.Pri-marily, women are older, have a higher incidence of diabetes, hypertension and congestive heart failure and more often undergo non-elective procedures than men.Even after multivariate adjustment of these factors, cardiovascular mortality in women is higher than in men [5,8,9].The possibility has been discussed that men are better cared for than women [7,11].
The role of coronary artery size as a predictor of mortality in patients with myocardial infarction and with coronary artery bypass grafting is controversial [1-5, 7, 10].In the Coronary Artery Surgery Study, small body size and coronary artery caliber were the strongest predictors of perioperative mortality [3].
The existence of gender differences in coronary artery size has hitherto not been proven with

Abbreviations
Objectives: To determine whether there is a gender difference in coronary artery size normalised for left ventricular (LV) mass.
Background: Small coronary artery caliber may play a role as a risk factor for coronary artery disease in women.However, the existence of a gender difference in coronary artery size is controversial.Furthermore, coronary artery size ought to be normalised for LV mass, since there is a theoretical relation of coronary artery size to LV mass according to the law of minimum viscous energy loss for the transport of blood in the coronary circulation.
Methods: In 200 individuals (100 women) without cardiac disease and with normal Doppler echocardiography, left main (LCA) and right coronary artery (RCA) size were determined using transoesophageal echocardiography.LV mass was assessed by transgastric M-mode echocardiography.
Results: Age (44 ± 15 years in women; 41 ± 16 years in men), the presence of non-cardiac diseases, cardiovascular risk factors and medication were similar in women and men.LV mass in women was lower than in men (148 ± 36 g, 189 ± 45 g; p <0.0001).LCA and RCA cross-sectional areas in women were smaller than those in men (LCA: 10 ± 3 and 16 ± 5 mm 2 , p <0.0001; RCA: 4 ± 2 and 7 ± 3 mm 2 , p <0.0001, respectively).LCA and RCA cross-sectional areas of women were smaller even after normalisation for LV mass (LCA: 7 ± 3 and 9 ± 3 mm 2 /100 g LV mass, p <0.0001; RCA: 3 ± 1 and 4 ± 1 mm 2 /100g LV mass, p = 0.002, respectively).LCA caliber of women ranged below the theoretically expected size according to the law of minimum viscous energy loss for the transport of blood in the coronary circulation, whereas those of men tended to be above it.
Conclusions: In a population without cardiac disease, women have smaller coronary artery size even after normalisation for left ventricular mass.sufficient certainty.Roberts and Roberts examined proximal portions of the three major epicardial coronary arteries in 98 necropsy patients with different coronary events and did not find a significant gender difference [12].Dodge et al. examined the influence of gender on lumen diameter of normal human coronary arteries using coronary arteriograms and found that women have smaller epicardial arterial diameter than men even after normalisation for body surface area [13].Concerning gender differences in coronary artery size, these studies are limited due to small sample size, to possible underestimation of coronary artery diameter in non-perfused vessels in postmortem analysis, and to the lack of normalisation for left ventricular (LV) mass.
Coronary artery caliber ought to be normalised for LV mass.According to the law of minimum viscous energy loss for the transport of blood in the coronary circulation, coronary artery size is directly determined by LV mass according to a 2 /3 power law [14].The law of minimum viscous energy loss is a physical law which can be derived from Poisewille's law, and which applies to circulatory systems in different biological species.
The purpose of this study was to determine whether there is a gender difference in coronary artery size normalised for LV mass in a population without cardiac disease, and whether coronary artery size is related to LV mass by a 2 / 3 power law as predicted theoretically.

Study individuals
The study population consisted of 129 patients (48 men), who underwent transoesophageal echocardiography (TEE) for medical reasons and 71 healthy probands (52 men) who had been previously included in a study on diving and cerebral ischaemia in the absence or presence of a patent foramen ovale [15].A history of any cardiac dis-ease as well as an abnormal Doppler echocardiogram were exclusion criteria.In particular, study individuals with LV mass index above the normal range w >134 g/m 2 for men; >109 g/m 2 for women) were not included [16].The study individuals gave informed consent to participate in the study.

Transoesophageal echocardiography (TEE)
TEE was performed in all study individuals in the left lateral supine position using an Acuson Sequoia TM C256 system (Mountain view, California, USA) with a multiplane 3.5-7 MHz TEE probe.The epipharynx was anaesthetised using lidocaine hydrochloride 10% spray.Before intubation of the TEE probe, 2-5 mg Midazolam (Dormicum ® ) was given intravenously.A 3-lead ECG and blood pressure were registered during the procedure.M-mode and 2D echocardiographic images as well as Doppler examinations of valves and pulmonary veins were performed as recommended by the the American Society of Echocardiography [17].M-mode echocardiography of the left ventricle was performed using the transgastric long-axis view, proximal to the papillary muscles.A frequency of 7 MHz was used to image epicardial coronary arteries.The left main coronary artery (LCA) was imaged using the short-axis view, and the right coronary artery (RCA) was examined using both the long-axis and short-axis view (figure 1).For accurate detection of the vessel borders, the region of interest was magnified using the resolution function.Subsequently, the focus was adjusted.Color Doppler flow was not used to measure coronary artery caliber.

Data aquisition
Echocardiographic off-line measurements were performed by reloading data from optical discs to the internal hard disc of the echograph.Two experienced cardiologists from our laboratory performed the measurements of LV mass and coronary artery caliber unaware of gender and clinical data of the study individuals.LV mass was calculated according to the cube formula [16].LCA-and RCA-calibers were measured close to the origin of the vessels at the coronary sinus of the aorta.

Statistical analysis
All data are expressed as mean values ± 1 SD.Nominal data comparison between men and women was performed using the chi square test.Continuous data were analysed by an unpaired t-test.In order to compare measured with the theoretically expected coronary artery cross-sectional area (y = 0.4 ҂ LV mass 0.67 in mm 2 ), power equation fitting was performed (Sigmaplot Version 5.0, SPSS Incorporation) [14].Data were considered significant at p <0.05.

Study population
Age, drug therapy, the presence of non-cardiac diseases, cardiovascular risk factors and medication were similar in men and in women (table 1).

Gender analysis
Haemodynamic and echocardiographic data are listed in table 2. LV mass in women was lower than in men.LCA and RCA cross-sectional areas in women were smaller than those in men.LCA and RCA cross-sectional areas of women were smaller even after normalisation for LV mass.Gender differences in coronary artery cross-sectional areas in relation to LV mass are shown in figure 2 and

men women p
Taking into account the basic structure-function relations of the coronary vascular tree, this study demonstrates that there is a significant gender difference in coronary artery size in a population without cardiac disease, whereby coronary artery cross-sectional areas normalised for LV mass in women are smaller than those in men.

Determinants of coronary artery caliber
To examine the influence of coronary artery size on clinical variables such as outcome of coronary artery disease, one should keep in mind its main determinants.Coronary artery caliber strongly depends on wall shear stress which is related to blood flow, and on myocardial bed size [14,[18][19][20].Using quantitative coronary angiography in patients with and without stenotic lesions, a curvilinear relation ( 2 / 3 power equation) between coronary artery lumen area and regional myocardial mass has been described which is in accordance with the law of minimum viscous energy loss for the transport of blood in the coronary circulation [14].Our data on left main coronary artery size and LV mass seem to comply with this principle of minimal viscous energy dissipation, since the observed relation very closely matches the theoretical prediction following the 2 / 3 power law (y = ax b ; observed data: b = 0.65, predicted data: b = 0.67) (see figure 2).
Coronary artery size should not be assessed in symptomatic patients using angiographically measured cross-sectional lumen area, since it may be underestimated in patients with diffuse coronary artery narrowing, where the luminal area is significantly smaller than would be expected in the absence of CAD [14].In patients with LV hypertrophy and dilated cardiomyopathy, coronary artery caliber is larger than in patients with normal LV mass, but may be too small in relation to myocardial bed size [13,21].Thus, patients with an elevated LV mass index were not included in the present study [16].

Discussion
Gender differences of left main coronary artery cross-sectional area (vertical axis) in relation to left ventricular mass (horizontal axis).Thin solid curve: theoretically expected left coronary artery caliber according to the law of minimum viscous energy loss for the transport of blood in the coronary circulation (y = 0.40 x 0.67 ).Thick solid curve: power equation fitting to compare measurements in all study individuals with theroretically expected left coronary artery size (y = 0.47 x 0.65 , r = 0.4, p <0.0001, SEE [standard error of estimate] = 4.8 mm 2 ).Dashed curve: power equation fitting to compare measurements in men with theoretically expected coronary artery size (y = 5.8 x 0.20 , r = 0.17, p = 0.1, SEE = 4.6 mm 2 ).Dashed-dotted curve: power equation fitting for measurements in women (y = 2.2 x 0.29 , r = 0.20, p = 0.04, SEE = 2.97 mm 2 ).
Left Ventricular Mass (g) energy loss for the transport of blood in the coronary circulation, LCA sizes of women tended to range below the theoretically expected size, whereas those of men tended to be above it (see the dashed and dashed-dotted curves in figure 2).

Literature data on gender-related coronary artery size
To our knowledge there is no study which has examined gender-related coronary artery size normalised for LV mass in a sizeable population without cardiac disease.In 98 patients with CAD, there was no gender difference in coronary artery size; however, those necropsy findings were presented without normalisation for LV mass [12].Dodge et al. found significantly smaller coronary artery caliber in women even after normalisation for body surface area using carefully selected coronary angiograms without arteriosclerosis, but the sample size was limited (10 women) [13].Studies including larger sample sizes provided mortality data of patients undergoing coronary artery bypass surgery [1,3,22].The Northern New England Cardiovascular Disease Study Group found a significantly smaller mid left anterior descending coronary artery after correction for body surface area [22].The main limitation of this study was that LV mass data were not available.

Study limitations
TEE has been shown to be reliable for the examination of proximal coronary arteries [23].However, the follwing limitation of this approach for measuring coronary artery caliber should be mentioned: although the problem with reduced lateral resolution for the measurement of horizontal distances can be managed with optimal echograph settings (ie, high transducer frequency, lowest possible gain), this may still be relevant for the measurement of coronary artery caliber using TEE [24].However, it is not relevant in left coronary artery caliber measurement since this is obtained in a perpendicular projection.Conversely, the origin of the right coronary artery is directed parallel to the ultrasonic beam causing reduced image quality of the vessel wall due to deminished lateral resolution.Thus, the measurement of the right coronary artery caliber in our study population was probably less precise than left coronary artery measure-ments.Moreover, for calculation of the cross-sectional area, errors in the measurement of dimensions are increased by a power of two.However, the intima-media border of the left main coronary artery was precisely identified in all cases due to the orthogonal direction of the ultrasound beam relative to the imaged artery.Coronary arteries were measured by image magnification.However, although the frequency of image aquisition is increased when the region of interest is made smaller, the number of pixels remains the same.Thus, in study individuals with poor vessel delineation, the accuracy of coronary artery caliber measurement was not improved by image magnification.
In the present study, gender analysis was performed in a population without known cardiac disease.The incidence of asymptomatic coronary artery lesions could have been larger than expected in a normal population due to the relatively high presence of cardiovascular risk factors and cerebrovascular disease (table 1).However, there were no gender-related differences in cardiovascular risk profile, drug therapy, nor in the prevalence of non-cardiac diseases among our study individuals.

Conclusions
Coronary artery caliber is related to LV mass by a 2 /3 power law as predicted by the law of viscous energy loss for the transport of blood in the coronary circulation.In a population without cardiac disease, women have smaller coronary artery caliber normalised for LV mass than men.Left main coronary artery sizes of women tended to range below the theoretically expected size, whereas those of men tended to be above it.

Figure 1
Figure 1 Transoesophageal short-axis view of the ascending aorta shows the origin of the left main coronary artery (A and C), and transoesophageal long-axis view shows the origin of the right coronary artery (B and D) at the aortic sinus.Arrows in C and D demonstrate the measurement of coronary artery diameters.
Figure 3 Gender differences of right coronary artery cross-sectional area in relation to left ventricular mass.Thick solid curve: power equation fitting for measured RCA sizes (y = 0.07 x 0.84 , r = 0.41, p <0.0001, SEE [standard error of estimate] = 2.5 mm 2 ).
Correspondence:Prof. Dr.Christian  Seiler, MD Cardiology Swiss Cardiovascular Center Bern University Hospital CH-3010 Bern e-mail: christian.seiler.cardio@insel.chSwiss Medical Weekly: Call for papers Swiss Medical Weekly The many reasons why you should choose SMW to publish your research Official journal of the Swiss Society of Infectious disease the Swiss Society of Internal Medicine the Swiss Respiratory Society Med Wochenschr (1871-2000) Swiss Med Wkly (continues Schweiz Med Wochenschr from 2001) Editores Medicorum Helveticorum
3. Power equation fitting of measured LCA cross-sectional areas compares well with the theoretically expected LCA size (0.4 ҂ LV mass 0.67 in mm 2 ).According to the law of minimum viscous * Patients with previously diagnosed paradoxical embolism in whom a residual shunt of a device occluded PFO was ruled out using TEE.Abbreviations: ACE = angiotensin converting enzyme; NS = not significant; PFO = patent foramen ovale; TEE = transoesophageal echocardiography.

Table 2
Haemodynamic and echocardiographic data.