In this population of aging Australian women, the ICA PI is demonstrated to be associated with cardiovascular disease. It was significantly correlated to aortic PWV and the Framingham 10-year cardiovascular risk. The ICA PI was significantly predicted by cardiovascular risk factors including age, systolic blood pressure, MAP, BMI, smoking and diabetes. The CCA PI was also significantly related to the Framingham risk score and femPWV, though this relationship was not as strong as for the ICA. Neither CCA nor ICA PI was a significant predictor of ischaemic heart disease over age and systolic blood pressure. Reasons for the stronger relationship of the ICA PI to cardiovascular risk over the CCA may include the following: 1 The normal cerebral circulation is maintained in a constant flow due to a well-developed system of autoregulation, where acute falls in perfusion pressure can have potentially disastrous consequences to cerebral function.16 It can be expected that arteriosclerotic changes will increase flow impedance and should be readily detectable in the ICA.17 2 Readings taken in the common carotid will be influenced by the external carotid artery, which supplies the high resistance vascular beds of the muscles and skin of the face and scalp. Assessment of the PI of the internal carotid is noninvasive and relatively easy to acquire. Only one patient (0.6%) could not be assessed due to high positioning of the carotid bifurcation. The intra- and inter-operator repeatability of this potential cardiovascular health measure is still to be established at this site. Reports are varied for the reliability of PI measures in other applications including trans-cranial Doppler assessment of the cerebral vessels19 and intrauterine assessment of foetal vessels,20,21 with both intra-operator repeatability and technician experience having a significant impact on results.
The main limitations of this study include the following: 1 This is a cross-sectional analysis, and it therefore does not imply causality. 2 The delay between evaluation of PWV and PIs approached 2 years in some participants. 3 The rate of ischaemic heart disease was low relative to the population sample size, making estimates of the true effect size and odds ratio or risk measures more unreliable.22 4 We were unable to assess the relationship of PI to the prevalence of stroke as no participants had experienced either an ischaemic or haemorrhagic event.
In summary, the PI, as measured by carotid Doppler ultrasound, was significantly related to the Framingham 10-year cardiovascular risk and aortic stiffening as measured by carotid–femoral PWV. Of the two PIs measured, the ICA had the strongest relationship to cardiovascular risk factors and may relate more closely to cardiovascular disease progression. Neither index significantly contributed to prediction of ischaemic heart disease in this analysis.