Diagnostic Value of Ankle Brachial Index for Myocardial Ischemia in Asymptomatic Diabetic Patients and Comparison with Myocardial Perfusion Imaging

Objective: The ankle-brachial index (ABI) as a simple test which can detect peripheral arterial disease (PAD) . Therefore in this study we try to evaluate the diagnostic value of ABI for silent myocardial ischemia in diabetic patients and compare the results with myocardial perfusion imaging (MPI) results. Materials and Methods: All 149 diabetic in this study were to different sex, cholesterol, familial history, and high blood pressure, level of ischemia, myocardial infraction (MI), left ventricle (LV) volume, ejection fraction (EF), and wall motion. Then the relationship of ABI index and these parameters were investigated. Results: According to the calculated ABI the data was investigated based on ABI lower and higher than 0.9 .The frequency of ABI> 0.9 was 16 (11%) and< 0.9 was 133 (89%). There was no significant relationship between all the mentioned parameters and ABI index ( P -value: 0.05). Conclusion: This study suggests the ABI sensitivity and specificity for diagnose of silent ischemia in asymptomatic diabetic patients is very low and in this case ABI cannot replace MPI by any means.


Introduction
he ankle-brachial index (ABI) as a simple test can detect peripheral arterial disease (PAD) especially in the lower extremities (1). ABI is equal to the ratio of the ankle systolic pressure to the brachial systolic pressure. In a normal person the ankle pressure is slightly more or close to the brachial pressure; hence ABI is around 1-1.4 (1)(2)(3). In PAD patients, the ankle pressure lowers due to the upstream hemodynamic lesions which results in a lower ABI (ABI<1) (4,5). The ABI depends on qualitative and quantitative aspects of other diseases such as atherosclerotic disease. In atherosclerotic disease the serially located lesions additively contribute to decrease distal pressure which enables the ABI to measure the severity and number of atherosclerotic lesions located in the lower extremity (6,7). On the contrary, the ABI is particularly unreliable in diabetes disease, which is attributed to the stiffness of arterial vessel walls. Theses rigid arteries cause a false positive read by the sphygmomanometer (8,9). Diabetic patients are susceptible to a wide range of other disease including retinopathy (10), obesity (11), resistance to insulin (12), infections (13, 14), antonym disorders and most importantly coronary artery disease (CAD) and PAD (15)(16)(17). The incidence of silent myocardial ischemia (SMI) and stroke in diabetic patients is 2-7 times higher than non-diabetic patients. Therefore continuous monitoring of diabetic patients could prevent the SMI (18,19). Myocardial perfusion imaging (MPI) as the gold standard for prognosis and diagnosis of CAD. However, due to ionizing radiation of MPI method encourages finding different and safer methods. In this study we are trying to evaluate the diagnostic value of ABI for SMI in diabetic patients and compare the results with MPI results.

Settings and Patients
This study was organized in Fatemeh Zahra Hospital of Sari located in the north of Iran. The subjects of this study were 149 type 2 diabetes who referred to the hospital for MPI. These patients were selected by convenient method during 2015-2016. All the patients were 30-70 years old. We excluded patients who had heart disease from this study. All the demographic information including smoking, type and duration of diabetes, sex, age, blood pressure and etc. were registered for the selected patients.

MPI test
First, a standard dose of 99m Tc-MIBI was injected to the patient and the rest phase image was acquired by a gamma camera (Zemence Company, dual headed gamma camera, Germany 2011). The following day the stress phase image was acquired.

ABI measurement
Before any measurement, the patient should be in the supine position for 5 min. then, the systolic blood pressure of one arm and one ankle were measured. The ABI value was calculated using the following equation: systolic blood pressure of the ankle brachial arterial systolic pressure

Statistical Analysis
P-value <0.05 is considered as a significant difference. In addition, SPSS version 14 software was used for data analysis. Negative predictive value (NPV), positive predictive value (PPV) and overall accuracy (OA) were also calculated.

Ethical considerations
The subjects in this study agreed to contribute in this study and their health were not endanger in this study by any means. All the patients' information will stay confidential. This study was approved by Mazandaran University of Medical Sciences ethical committee. The ethical code for this study is IR.MAZUMS.REC.94-1843.

Descriptive statistics
All 149 patients in this study were statistically analyzes and categorized according to different parameters ( According to the calculated ABI the data was investigated based on ABI lower and higher than 0.9 .The frequency of ABI> 0.9 was 16 (11%) and< 0.9 was 133 (89%) In order to evaluate the relationship of ABI with ischemia, MI, LV volume, EF and wall motion first we categorized the patients based on ABI values (above 0.9 and under 0.9) for each parameter. Then, we performed Chisquare test to determine whether there is a significant relationship between or not. Based on the data present in table 3 there is no significant relationship between ABI and other parameters (table2).

A receiver operating characteristic (ROC) curve
ROC curve can determine the diagnostic value of ABI. By using this carve different parameter including the area under carve (AUC), p-value and 95% percent confidence interval are provided in table 6. Moreover, diagnostic indicators including sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), true negative rate (TNR) and true positive rate (TPR) are calculated (Table 3). This data indicate the low diagnostic value of ABI for prediction of silent ischemia in diabetic patients in comparison with MPI.

Discussion
Most of the patients (71.14%) in this study are   (22) female. The administration of insulin is the most common treatment for these patients, and most of them possess high level of cholesterol which is very common for diabetic patients. The calculation of ABI for this papulation indicates a normal distribution for the most part. However, different studies suggest that diabetic patients have low ABI and they are very susceptible to CAD and PAD (20)(21)(22)(23)(24). The relationship between ABI and different level of ischemia was also investigated. Even though the findings suggested that mild, moderate and sever ischemic diabetic patients ABI are lower than normal ABI, but these difference were not statistically significant. It is noteworthy that the increment in the population of patients may change the results. Likewise the relationship of ABI and MI was also not statistically significant which agrees with other studies (22). However some studies suggested that patients with low ABI are 2-3 times more susceptible to heart stroke (25). Diabetic patients with low ABI are highly at risk of CAD in previous studies (26)(27)(28). This contradictory can be due to the small patient population of our study. For good measures we also investigated the relationship between heart function (EF, wall motion and LV function) and ABI. The data indicates that there was no significant relationship between ABI and neither EF nor wall motion and LV function which is in parallel with other studies (29). Finally, ROC curve displayed a very low diagnostic value for ABI in ischemic diabetic patients in comparison with MPI. According

Conclusions
This study suggests the ABI sensitivity and specificity for diagnose of silent ischemia in asymptomatic diabetic patients is very low and in this case ABI cannot replace MPI by any means. However it is noteworthy that more profound studies with a larger patient population (especially above 55 years old) is needed to shed light upon this topic.

Funding
This work was financially supported by Mazandaran University of medical sciences. of an oscillometric ankle-brachial index in the diagnosis of lower limb peripheral arterial disease: