Brachial-ankle Pulse Wave Velocity and Renal Function can Predict Coronary Artery Disease in Asymptomatic Japanese Patients with Type 2 Diabetes

availability of the 64-slice CT scanner. The latest generation of CT scanners can provide exquisite imaging of cardiac structures, coronary plaque morphology data, and details of atherosclerotic lesions, quickly and with minimal patient discomfort or risk. A recent Japanese observational study demonstrated that serum triglyceride level was a leading predictor of Coronary Heart Disease (CHD), comparable to low-density lipoprotein cholesterol in Japanese patients with type 2 diabetes [2]. The authors suggested that ethnic group-specific strategies for prevention of diabetic macroangiopathy may be indicated, because the serum triglyceride level is not a leading predictor of CHD in diabetic subjects in Western countries [2]. Moreover, the most common reported comorbidities associated with chronic kidney disease are hypertension, retinopathy, CHD, myocardial infarction, and congestive heart failure, according to a study from Kidney Disease Improving Global Outcomes (KDIGO) 2013 [3]. Even mildly reduced estimated glomerular filtration rate (<75 ml/min/1.73 m2) increased the risk of CHD incidence among type 2 diabetes patients [4]. The objectives of this study were to determine the prevalence and severity of coronary atherosclerosis in asymptomatic subjects with type 2 diabetes, since asymptomatic CHD, with ≥ 50% diameter stenosis and myocardial perfusion abnormalities, was reportedly detected in more than half of Japanese patients with type 2 diabetes with vascular complications [5]. Moreover, it was reported that plasma adiponectin levels are an independent predictor of CHD in Caucasian men initially free of CHD [6], and a risk calculator including albuminuria enhances CHD prediction, compared to a calculator using only standard risk factors, in men and women [7]. We also sought to determine which clinical parameter(s), such as pulse wave velocity (PWV), plaque score, and renal function, which are related to diabetic complications, account for the findings on cardiovascular CT angiography.


Introduction
The impact of diabetes on cardiovascular disease is further illustrated by the fact that the risk of developing a myocardial infarction in diabetic patients without known heart disease is equivalent to the risk observed in non-diabetic survivors of a prior infarction [1].Because of the high incidence of coronary artery disease (CAD) among diabetic patients, the need for accurate and early identification is paramount.Computed tomography (CT) angiography received a boost recently with the CT angiography at Yokohama Rosai Hospital between December 2009 and June 2011.All subjects met the following criteria: 1) no symptoms or past history of CAD, 2) not taking insulin, and 3) serum creatinine ≤ 1.0 mg/dl.Thus, type 1 diabetic patients were not included in the study.Patients included in this study were referred to the Heart Center of Yokohama Rosai Hospital by specialists of the Endocrinology & Diabetes Center in our hospital, for diagnostic angiography for unsuspected CAD.
All subjects underwent coronary CT angiography using the Siemens Sensation (Siemens Healthcare, Tokyo) Cardiac 64-Slice CT.Cross-sectional images were reconstructed with a slice thickness of 0.75 mm at 0.4 mm intervals, with the use of an electrocardiogram gated half-scan reconstruction algorithm [8], to obtain an image acquisition window of 164 ms.All vessels > 1.5 mm were evaluated.Patients with a resting heart rate > 60 beats per minute were intravenously medicated with 5 mg metoprolol to enhance image quality and reduce distortion.A heart rate of 60 beats per minute or less critically influences image quality and is considered a desirable threshold to minimize motion artifacts [9].Disease status was categorized as follows: no disease, nonobstructive CAD (one or more coronary lesions with < 50% stenosis), and obstructive CAD (one or more lesions > 50%).Brachial-ankle pulse wave velocity (baPWV) was measured by its apparatus (Kohrin Co, Tokyo) [10].Adiponectin was measured by enzyme-linked immunosorbent assay (Otsuka Pharmaceutical Co, Tokyo, Japan).Levels of HbA1c were determined by using the HA-8170 apparatus (Arkray, Kyoto, Japan), which employs a regular ion exchange HPLC method for quantification of this glycohemoglobin, and laboratory parameters of serum creatinine, cholesterol, triglycerides and other variables were analyzed in a Nippon Denshi Autoanalyzer (JCA-RX20, Tokyo, Japan) as previously reported [11].Each parameter, including visceral and subcutaneous fat area determined by abdominal CT scan, and homeostasis model assessment of insulin resistance (HOMA-IR), was determined, as previously reported elsewhere [11,12].
Descriptive statistics (frequency distributions and mean values) were completed for each patient cohort.All values are presented as the mean ± standard deviation.Statistical significance was evaluated among the three cohorts using oneway analysis of variance.Differences were considered to be statistically significant at a P value of < 0.05.

Patient profiles
The clinical characteristics of the subjects, including body weight, waist circumference, visceral and subcutaneous fat area determined by abdominal CT scan, blood pressure, baPWV, serum cholesterol, serum creatinine, fasting glucose, HOMA-IR, and glycosylated hemoglobin (HbA1c), are shown in Table 1a and  b.
Patients with significant stenosis were classified as the obstructive CAD group (n = 21), and the others were included in the non-obstructive group (n = 53), including 17 cases with normal coronary arteries.
Obstructive CAD was found in 28.4% of patients, and 3 cases (4.1%) required percutaneous coronary intervention for left anterior descending disease.

Comparison of each parameter between patients with non-obstructive CAD and those with obstructive CAD
The obstructive CAD group had significantly greater age, and higher serum creatinine, cystatin C, and maximum Intima-Media Thickness (IMT) by ultrasonography of the carotid artery, while estimated glomerular filtration rate was significantly lower in the patients with obstructive CAD (Table 3).HbA1c, waist circumference, visceral fat area, cholesterol, and blood pressure were not significantly different between the groups.There were no significant differences in BMI, urinary albumin excretion, and plasma levels of glycoalbumin, fasting insulin, creatinine, and adiponectin between the two groups.In the multivariate analysis, baPWV, max IMT and serum creatinine proved to be independent factors for estimating the existence of advanced coronary lesions (Table 4).

Variatbles
We also analyzed a Receiver Operating Characteristic (ROC) curve for obstructive CAD, using parameters such as age, maximum IMT, baPWV, and serum Cre level (Figure 1).The cutoff point for predicting obstructive CAD was more than 1550 cm/s of baPWV (sensitivity: 0.90; specificity: 0.68).In patients with more than 1550 cm/s of baPWV, serum Cre level was then the best predictor among the parameters (Cre ± 0.6 mg/ dl, sensitivity: 0.84; specificity: 0.82) (Figure 2).Thus the most useful clinical parameters for predicting obstructive CAD, as determined from the ROC analysis, were as follows: baPWV > 1550 cm/s and serum creatinine > 0.60 mg/dl.

Discussion
Our review is the first to evaluate the increased coronary atheromatous burden in patients with type 2 diabetes by MDCT, and provides interesting insights into the characteristics of the disease in these high-risk populations.Our data clearly demonstrated that there is a high prevalence of CAD in type 2 diabetic patients who are asymptomatic and have a negative exercise electrocardiogram.CAD events can be prevented by comprehensive treatment of various risk factors, such as hypertension and dyslipidemia [13].However, the study patients were already well-controlled for hypertension, dyslipidemia, and hyperglycemia by treatment with angiotensin receptor blockers, statins, and oral hypoglycemic agents.
Our data demonstrated that there was no significant difference in BMI and visceral fat area between the obstructive and non-obstructive CAD groups.On the other hand, a study consisting of 2842 men and 3196 women, the FINRISK '92 survey [14], reported that body fat percentage measured by a near-infrared interactance device was a significant predictor of cardiovascular disease (CVD) and CHD events among men and women, but did not provide any additional predictive power over and above simpler measures, such as BMI or waist-to-hip ratio.It was also reported that basal C-peptide is related to cardiovascular predictors (IMT) of type 2 diabetes in Korean patients, suggesting that basal C-peptide does provide a further indication of CVD [15]; however, our data demonstrated that insulin resistance, estimated by HOMA-IR and fasting levels of serum insulin, did not differ between the obstructive and non-obstructive CAD groups.A prospective cohort study of 7067 Chinese type 2 diabetic patients without a history of CHD, enrolled from 1995 to 2005, demonstrated that in type 2 diabetes, albuminuria links conventional risk factors and CHD [16].The onset of chronic kidney disease changes risk associations between lipids and CHD [16], while there was no significant difference in albuminuria between the obstructive and non-obstructive CAD groups in our study.It was previously reported that higher baPWV could be an independent risk factor for future microalbuminuria in patients with essential hypertension [17], suggesting that baPWV may reflect future microalbuminuria and CHD.
Cases with asymptomatic type 2 diabetes were subjected to the Diabetes Audit and Research in Tayside, Scotland, suggesting that age at diagnosis, duration of diabetes, HbA1c, smoking (current, past, never), sex, systolic blood pressure, treated hypertension, total cholesterol, and height seem to be the significant predictors for future cardiovascular events [18].Thus, our data are consistent with the previous findings of the effect of age [18] on obstructive CAD.
It was also reported that baPWV is an independent risk factor for future cardiovascular events in patients with essential hypertension [19].In the present study, baPWV and serum creatinine proved to be the most useful clinical markers for predicting obstructive CAD.MDCT and/or coronary angiography are highly recommended when patients meet the criteria of baPWV > 1550 cm/s and Cre > 0.60 mg/dl.
It was reported that baPWV was significantly and positively associated with age, systolic blood pressure, and the Framingham risk score in hypertensive patients, and also that the areas under the ROCs of PWV can predict the presence of both stroke and CAD [10], while multivariate analysis with the Cox proportional hazards model identified IMT, but not baPWV, as a significant determinant of CVD [10].Another report demonstrated that the combination of the Framingham Risk Score with IMT rather than with baPWV improved the prediction of CVD [20].Moreover, our data clearly suggest that we can predict the risk and possibility of obstructive CAD by examining baPWV and serum creatinine levels in asymptomatic patients with type 2 diabetes.
In conclusion, HbA1c and Glycoalbumin (GA) levels with fasting insulin levels do not always reflect the risk of obstructive CAD.The present data also demonstrated that adiponectin concentration and visceral fat area were not useful markers for predicting obstructive CAD.Finally, the present data clearly demonstrated that we can predict the risk and possibility of obstructive CAD by examining baPWV and serum creatinine levels in asymptomatic patients with type 2 diabetes.On the other hand, the present study was performed in a limited number of asymptomatic type 2 diabetic patients.Moreover, we did not follow up these patients for future cardiovascular events.Thus, further studies are needed on two separate groups, derived from the asymptomatic type 2 diabetic patients according to the values for baPWV and serum creatinine as described above, with prospective monitoring for the presence of obstructive CAD.Receiver operating characteristic (ROC) curve analysis for obstructive coronary artery disease (CAD) in patients with brachial-ankle pulse wave velocity (baPWV) > 1550 cm/s (n = 25 cases) was performed using parameters such as age, maximum intima-media thickness (IMT), and serum creatinine (Cre) levels.

FPG=
Fasting Plasma Glucose, IRI = Immunoreactive Insulin, BP = Blood Pressure, TG = Triglycerides, LDL-Chol = Low-Density Lipoprotein Cholesterol, HDL-Chol = High-Density Lipoprotein Cholesterol, BMI = Body Mass Index.Each parameter was determined as described in detail in the Text.

Table 1a :
Background of the subjects.

Table 2 :
Prevalence of coronary artery disease in asymptomatic type 2 diabetic patients.CAD = Coronary Artery Disease, MDCT = Multidetector Computed Tomography MDCT was performed as described in detail in the Text.

Table 3 :
Comparison of each Parameter between Patients with Non-Obstructive CAD and those with Obstructive CAD.
Results were expressed as mean ± SD. maxIMT; maximum intimamedia thickness, baPWV; brachial pulse wave velocity, eGFR; estimated glomerular filtration rate, U-Alb; urinary albumin excretion rate.Each parameter was determined as described in detail in the Text.

Table 4 :
Multivariate Logistic Analysis of Clinical Characteristics.baPWV, max IMT, and serum creatinine (Cre) were significantly related to the presence of obstructive CAD.Age and systolic blood pressure were not significantly related to obstructive CAD.baPWV = brachial-ankle pulse wave velocity, IMT = Intima-Media Thickness Function can Predict Coronary Artery Disease in Asymptomatic Japanese Patients with Type 2 Diabetes.J Endocrinol Diab 2(3): 5.