The study was conducted from August 2005 to September 2006. We included 101 consecutive patients who complained of chest pain but showed no significant coronary stenosis on coronary angiography. A complete history, as well as physical and laboratory examinations, were obtained from all patients before performing the coronary angiography. Smokers were defined as subjects who had smoked regularly for more than 1 year. Hypertension was defined as systolic blood pressure ≥140 mm Hg or diastolic blood pressure ≥90 mm Hg or current use of antihypertensive agents. Diabetes was defined as the use of insulin, oral hypoglycemic agents, or a fasting plasma glucose level of ≥126 mg/dL. After the coronary angiogram, we confirmed the clinical diagnosis and categorized the patient as follows: acute myocardial infarction, unstable angina, microvascular angina, noncardiac chest pain, or congestive heart failure. The definition of acute myocardial infarct was a typical rise and/or gradual fall (troponin) or a more rapid rise and fall (creatine kinase-MB fraction) of biochemical markers of myocardial necrosis, with at least one ischemic symptom (chest pain) or electrocardiogram (ECG) changes indicative of ischemia. Unstable angina was considered as ischemic symptoms (chest pain) and no elevation in bio-chemical markers with ECG changes (new 1 mm or greater ST-segment depression or T-wave inversion in multiple precordial leads). The definition of microvascular angina required ischemic symptoms (chest pain) but no elevation in biochemical markers with the following ECG changes: ST-segment depression of 0.5-1 mm, T-wave flattening, or inver-sion <1 mm in the lead with a dominant R wave. The non-cardiac chest pain group complained of ischemic symptoms (chest pain) but no elevation in biochemical markers and a normal or unchanged ECG. Congestive heart failure was diagnosed clinically using the Framingham criteria.

Coronary angiography was performed with the radial or femoral approach using the standard Judkins technique. Coronary arteries were shown in the left and right oblique planes and in cranial and caudal angulations. All angiograms were filmed at 15 frames per second. The Thrombolysis in Myocardial Infarction frame count (TIMI frame count, TFC) was obtained by the method of Gibson et al.13) The first frame used for counting was the first frame in which dye fully entered the artery. The last frame counted was the one in which dye first entered the end point branch off the target artery. The follow-ing distal landmark branches of the target artery were used for the analysis: the distal bifurcation in the left anterior descending artery (LAD), the distal bifurcation of the segment with the longest total distance in the left circumflex artery (LCX), and the first branch of the posterolateral artery in the right coronary artery (RCA). Because the length of the LAD is anatomically longer than that of the LCX and RCA, the LAD frame counts were corrected by dividing by 1.7 to obtain a corrected TFC, as previously reported.13) We calculated the mean TFC for the LAD, LCX, and RCA. All participants with a mean TFC greater than two standard deviations from mean TFC in the noncardiac chest pain subgroup were considered to have SCF, whereas those with a value below this threshold were considered to have NCF. The threshold TFC value was 25 frames.

The carotid IMT was calculated by ultrasound equipment with a linear transducer (12-MHz, Vivid 7 dimension, GE Healthcare, Piscataway, NJ, USA). The right and left common carotid arteries were scanned in a longitudinal plane with loss of the parallel configuration of the near and far walls of the common carotid artery. The reference point for measuring IMT was the beginning of the dilatation of the carotid bulb in the bifurcation area. The IMT was measured manually as the thickest point of the transition zone between the lumenintima and media-adventitia, 1 cm proximal to the reference point along at least 1 cm of axial length.14) The values for each carotid artery were averaged to obtain the mean carotid IMT. Carotid IMT was assessed by two cardiologists blinded to other clinical data. Disagreements were resolved by a third observer.

We measured PWV on the left and right side using a Vasera VS-1000 instrument (Fukuda Denshi Co. Ltd, Tokyo, Japan), which simultaneously records PWV, blood pressure, ECG and heart sounds. The PWV is theoretically calculated as the distance from two sites over the pulse transit time. The time interval between the brachium and the ankle (ΔTba; pulse wave transit time) was defined as the time interval between the wave front of the brachial waveform and that of the ankle wave form. The distance between sampling points was calculated automatically according to subject height. The path length from the heart to the brachium (Lb) was expressed using the following equation: Lb=0.2195×height of the patient (cm)-2.0734.15) The path length from the heart to the ankle (La) was expressed using the following equation: La=0.8129×height of the patient (cm)+12.328.15) Finally, the PWV was calculated from the following equation: PWV=(La-Lb)/ΔTba.15) The mean PWV value was calculated as the average of both PWV values.

The statistical analysis was performed using the general linear model in JMP7.0 software (SAS Inc., Cary, NC, USA). The categorical variables were expressed as counts and percents. The categorical variables were compared with Pearson's chi-square test and an analysis of variance. Continuous variables are expressed as the mean±standard deviation (SD), and the independent t-test was used to compare the two groups. The Pearson's correlation test was used to determine the correlation between the TFC and the carotid IMT.

The clinical characteristics of the 101 patients in this study are summarized in Table 1. Overall, 54% of the subjects had hypertension, 28% had diabetes mellitus (DM), and 42% were smokers. Eighty-six patients were categorized into the NCF group, and 15 patients in the SCF group. Male patients (n= 11, 73.3%) were significantly more common in the SCF group than those in the NCF group (n=37, 43.0%, p<0.05). However, age, presence of DM, hypertension, and lowdensity lipoprotein levels did not differ between the NCF and SCF groups. Smoking tended to be more common in the SCF group, but the difference was not statistically significant (Table 2).

A significant difference was observed in the clinical diagnoses between the SCF and NCF groups. Microvascular angina was the most common clinical diagnosis (n=11, 73.3%) in the SCF group, whereas non-cardiac chest pain was the most common diagnosis in the NCF group (n=41, 47.7%). The TFC was 28.8±3.5 frames in the SCF group and 15.7±4.5 frames in the NCF group (Table 2). In particular, the TFC increased significantly in a subgroup with microvascular angina (21.8± 7.8 frame count), compared to that in the other clinical subgroups (Fig. 1). We also compared the TFC according to risk factors. The TFC in male patients tended to be higher than that of female patients (18.9±7.0 vs. 16.6±5.5, respectively), but the result was not statistically significant. The TFC was not different with the presence of other risk factors, such as DM, hypertension, or smoking.

Carotid IMT was positively correlated with TFC (r=0.40, p<0.001), but no significant correlation was found between PWV and TFC (Fig. 2). Carotid IMT increased significantly in the SCF group compared to that in the NCF group (1.2± 0.3 mm and 0.8±0.1 mm, respectively, p<0.01). However, no difference was found in the PWV between the SCF and NCF groups (SCF; 14.1±2.4 m/s, NCF; 14.5±2.8 m/s) (Fig. 3).