Introduction

It has been clarified that arteriosclerosis progresses before the onset of cardiac disease and sclerosis of the aorta precedes sclerosis of the cerebral artery or coronary artery. To predict a cardiovascular event, quantitative diagnosis of arteriosclerosis is of paramount importance. A simple quantitative method for early diagnosis of arteriosclerosis is required to prevent hemodynamic complications during general anesthesia for oral surgery patients with arteriosclerotic disease. Arterial distensibility is attracting attention as an index for assessing early-stage arteriosclerosis, as it is an independent predictor of all-cause and cardiovascular mortality, fatal and non-fatal coronary events, and fatal strokes in patients with essential hypertension.1, 2

Arterial distensibility can be evaluated by measuring the pulse wave velocity (PWV) between two sites in the arterial tree.3, 4 In particular, brachial-ankle PWV (baPWV), which is a simple non-invasive method of assessing arterial distensibility, has been widely used in Japan.5, 6, 7, 8 BaPWV is closely related to risk factors and organ damage associated with cardiovascular disease.9, 10, 11 This method is calculated from the brachial-ankle distance and brachial-ankle pulse wave spread time via a very simple mechanism that can easily be used for measurement without special techniques. However, the problem with the clinical use of baPWV is that the index itself is closely dependent on blood pressure (BP) levels.12, 13, 14 If the parameter includes functional factors such as BP, it is difficult to evaluate arterial stiffness. To assess so-called ‘true arterial stiffness’ by this simple method, compensation for BP is essential.

To overcome this disadvantage, a novel diagnostic parameter known as the cardio-ankle vascular index (CAVI) has been developed in Japan.15 CAVI is also correlated with coronary stenosis, diabetes mellitus and metabolic syndrome.6, 16, 17, 18, 19, 20 This parameter is reported to be independent of BP levels.4, 12, 13, 15 However, these reports examined the relationship between BP and the parameter for a single measurement at one time point, and it has not been fully clarified whether these parameters are affected in the same patients by short-term changes in hemodynamics, such as those occurring during general anesthesia.

The aim of the present study is to investigate the actual influence of BP changes on CAVI and baPWV examined based on percentage change before and after general anesthesia in oral surgery patients.

Methods

Measurement of CAVI and baPWV

The purpose of this study was explained to the subjects, and consent was obtained for participation in the study and for release of the study data. The study was conducted in accordance with the Declaration of Helsinki.

CAVI and baPWV were measured non-invasively using the VaSera VS-1000 vascular screening system (Fukuda Denshi, Tokyo, Japan).21 CAVI is clinically used as an index reflecting the distensibility of arteries from the aortic valve to the ankle22, 23 or knee (Figure 1). The principles underlying the CAVI13, 15, 22, 23 and baPWV8, 9 measurement methods have been described previously. CAVI is calculated based on the changes in vascular diameter at each beat and PWV.13, 15, 22, 23 Examination can be easily and non-invasively completed in a short period of time by measuring BP at four limbs, phonocardiogram and ECG. To detect the brachial and ankle pulse waves with cuffs, the pressure of the cuffs is kept low at 50 mm Hg to minimize the effects of cuff pressure on hemodynamics.21

Figure 1
figure 1

Measurement of arterial distensibility parameters.

Full size image

CAVI was measured at the following segments: CAVI (heart to ankle) and k-CAVI (heart to knee). Baseline CAVI, k-CAVI and baPWV values were measured between 1600 and 1800 hours, around the same time of hospitalization before surgery. All measurements were conducted in a room kept at a constant temperature with the subject resting in a supine position after resting for 5 min. Ankle-brachial pressure index (ABPI) was also measured at the same time as the distensibility parameters. ABPI represents the ratio of systolic pressure in the ankle over that in the brachium, and is used for the diagnosis of obstructive arteriosclerosis. In this study, subjects with an ABPI <0.9 in the preoperative period were excluded due to suspected obstructive arteriosclerosis or to rule out possible inaccuracies in measured values.6

General anesthesia

With no premedication, general anesthesia was induced intravenously with fentanyl 2 μg kg−1 and propofol 2 mg kg−1 serially at a rate of 10–6 mg kg−1 per h, followed by vecuronium 0.14 mg kg−1 or rocuronium 0.9 mg kg−1 to facilitate endotracheal intubation. General anesthesia was maintained with oxygen (FiO2: 0.4), nitrous oxide and propofol (4–8 mg kg−1 per h). Depth of anesthesia was evaluated with a bispectral index (BIS) monitor (model A-2000, Aspect Medical System, Natick, MA, USA) and was controlled such that BIS values were maintained in a range from 40 to 60.24, 25 Respiration rate and ventilation volume were controlled at 12–14 breaths per min and 8–10 ml kg−1, respectively, such that the partial pressure of end-expiratory carbon dioxide was in the range of 30–35 mm Hg. Left brachial BP, heart rate (HR), arterial oxygen saturation and end-tidal CO2 concentration as an anesthetic monitor were measured using a monitoring system (model BP608, Nippon Colin, Tokyo, Japan).

Values for CAVI, k-CAVI, baPWV and BP on the right side, as well as HR, were recorded using VaSera at 10 min after intubation by propofol induction of anesthesia.

Evaluation

Subjects were 30 patients (15 males and 15 females) scheduled to undergo oral surgery for a maximum of 3 h. A strong correlation between age and distensibility parameters (that is, CAVI and baPWV) had already been shown in all 30 patients during the preoperative period19, 20, 21 (Figure 2). Therefore, the influence of BP changes on CAVI and baPWV was evaluated separately in the elderly (⩾65 years) and middle-aged (<65 years) groups to eliminate the effects of age (age-dependence).6

Figure 2
figure 2

Relationship between age and preoperative arterial distensibility parameters in all subjects (n=30).

Full size image

The influence of hemodynamic changes on CAVI, k-CAVI and baPWV from the preoperative resting period to the stage during general anesthesia was evaluated. All parameters (CAVI, k-CAVI, baPWV, ABPI, BP and HR) during general anesthesia 10 min after endotracheal intubation following administration of anesthetic agents (2nd measurement) were compared with baseline data taken before surgery (first measurement). Correlations between the percentage changes in CAVI (ΔCAVI=(2nd CAVI—1st CAVI)/1st CAVI), k-CAVI (Δk-CAVI=(2nd k-CAVI—1st k-CAVI)/1st k-CAVI) or baPWV (ΔbaPWV=(2nd baPWV—1st baPWV)/1st baPWV) and decreases in systolic BP (ΔSBP=(2nd SBP—1st SBP)/1st SBP) were examined.

Statistical analysis

Quantitative data are expressed as mean values±s.d. Statistical analysis included unpaired t-test, paired t-test and simple regression analysis. Comparisons between two groups were performed by unpaired t-test and intra-group comparisons were performed by paired t-test. Associations between ΔSBP and Δdistensibility parameters were assessed by Pearson’s correlation coefficient. A P-value <0.05 was considered to indicate statistical significance.

Results

Table 1 summarizes the clinical characteristics of the subjects. Among the subjects, 14 cases (46.7%) took no anti-hypertensive agents, 12 cases (40.0%) took a calcium antagonist, and nine cases (30.0%) took an angiotensin converting enzyme inhibitor or angiotensin receptor blocker. Oral surgery in the subjects consisted of cystectomy in 18 cases (60.0%), tumorectomy in 11 cases (36.7%) and sagittal split mandibular ramus osteotomy in one case (3.3%).

Table 1 Patient characteristics
Full size table

Figure 3 shows the relationship between preoperative SBP and arterial distensibility parameters. In both groups, baPWV was strongly correlated with SBP, but CAVI and k-CAVI were not significantly correlated at the preoperative stage.

Figure 3
figure 3

Relationship between pre-SBP and arterial distensibility parameters.

Full size image

Table 2 shows the comparison between the elderly and middle-aged groups with regard to parameter values at the preoperative stage and during anesthesia, as well as the percentage change (Δ). On intra-group analysis, SBP decreased significantly from the preoperative stage to during general anesthesia in both groups (elderly group, P<0.001; middle-aged group, P=0.003; unpaired t-test). Mean baPWV during general anesthesia showed a significant decrease (P<0.001 in the elderly group, and P=0.006 in the middle-aged group; paired t-test). In contrast, mean CAVI was only significantly different (P=0.025; paired t-test) in the middle-aged group, and k-CAVI in both groups and CAVI in the elderly group showed no significant differences. ABPI decreased significantly after general anesthesia (elderly group, P<0.002; middle aged group, P<0.009; paired t-test); that is, ankle SBP decreased to a greater degree than brachial SBP under general anesthesia. Although HR decreased significantly only in the elderly group (P=0.020, paired t-test) from the preoperative stage to general anesthesia, there were no significant correlations between ΔCAVI and ΔHR.

Table 2 Comparison of arterial distensibility and hemodynamic parameters between elderly and middle-aged groups
Full size table

On analysis between the two groups, CAVI, k-CAVI and baPWV in the elderly group were significantly higher than those in the middle-aged group at the preoperative period and during anesthesia. ΔCAVI or Δk-CAVI did not show any significant differences between the two groups, although ΔbaPWV in the elderly group was much higher than in the middle-aged group.

Figure 4 depicts the relationship between ΔSBP and Δdistensibility parameters. ΔCAVI and Δk-CAVI did not show any significant correlations with ΔSBP (ΔCAVI: P=0.327, r=0.253 in the elderly group, P=0.608, r=0.157 in the middle-aged group; Δk-CAVI: P=0.983, r=0.005 in the elderly group, P=0.252, r=0.406 in the middle-aged group), whereas ΔbaPWV showed a significant decrease in parallel with ΔSBP (P=0.009, r=0.614 in the elderly group, P=0.015, r=0.657 in the middle-aged group).

Figure 4
figure 4

Relationship between ΔSBP and Δarterial distensibility parameters.

Full size image

Discussion

In this study, CAVI showed no relationship with SBP at the preoperative stage. In addition, baPWV changed significantly in parallel with percentage changes in BP under general anesthesia, whereas CAVI and k-CAVI did not significantly change. General anesthesia is unlikely to improve arteriosclerosis itself, and thus, CAVI and k-CAVI are apparently less influenced by short-term changes in BP. Because of its independence from BP, CAVI may be a more sensitive and purer indicator of arterial wall stiffness than baPWV. This makes CAVI useful for serial evaluation of arterial stiffness within the same patient. Yambe et al.26 reported that baPWV was positively correlated with both intima-media thickness and plaque score in hypertensive patients, whereas Masugata et al.27 reported that baPWV was associated with plaque score in type 2 diabetes, and Okura et al.4 found that CAVI was related to intima-media thickness, but not to plaque score.4 Thus, CAVI was confirmed to be an index of arteriosclerosis rather than atherosclerosis, as CAVI had no correlation with plaque score in hypertensive patients.

BaPWV has been developed as a more convenient assessment of arterial distensibility. However, the problem with the clinical use of baPWV is that the index itself is closely dependent on BP.4 In addition to morphological or pathophysiological changes, environmental and psychological factors may increase BP and affect baPWV measurement. Therefore, baPWV is less useful as a functional stiffness parameter, as assessment of so-called ‘true arterial stiffness’ requires compensation for BP.13

The present results also confirmed a preoperative relationship between SBP and distensibility parameters, in agreement with several other papers.16 The serial effects of short-term changes in BP on CAVI in the same patient have not been reported previously, although it has been reported that CAVI has a much lower correlation with BP than baPWV, but this was based on single measurements in patients.6 As shown in Table 2, BP and baPWV decreased significantly, and CAVI decreased slightly, but not significantly, during general anesthesia.

Propofol has a sedative effect and a direct vasodilator action on vessel smooth muscle.28, 29 The vascular distensibility of the muscular vessels is affected by sympathetic nerve activity. It is very likely that the influence of anesthesia, whereby the vascular smooth muscle of the muscular vessels relaxes and expands, is reflected in baPWV, as ABPI decreased uniformly despite being measured at 10 min after tracheal intubation under general anesthesia (that is, ankle SBP decreased to a greater degree than brachial SBP). As baPWV is affected by BP and sympathetic tonus, the decrease in baPWV paralleled that in BP due to the vasodilator effects of propofol. In contrast, the relationships between ΔSBP and ΔCAVI or Δk-CAVI during general anesthesia did not show significant parallels. CAVI may only be affected by sympathetic tonus, and thus CAVI varied slightly in this study.

Very recently, Shirai et al.30 suggested that CAVI was not affected by BP at the time of measurement, but was affected by changes in the contractility of smooth muscle cells after administration of β-1 and α-1 adrenergic receptor blockers. They concluded that CAVI may be composed of both organic stiffness and functional stiffness factors based on smooth muscle contraction. Thus, the slight decrease in CAVI by anesthesia in this study cannot be ignored, and may be consistent with the effects of an α-1 adrenergic blocking vasodilator.

In addition, there have been no reports on the utility of k-CAVI as an arterial stiffness parameter. The segments measured in both CAVI and baPWV include elastic and muscular vessels (aorta and femoral-tibial arteries), but those measured in k-CAVI include only elastic vessels, other than the tibial arteries (aorta to knee). As shown in Figure 4, ΔCAVI or Δk-CAVI during general anesthesia was not seen as parallel shifts with ΔSBP. In particular, the correlation coefficient between ΔSBP and Δk-CAVI in the elderly group was much lower than that between ΔSBP and ΔCAVI, despite showing no significant correlations. This may be because the segments measured in k-CAVI exclude the muscular tibial arteries, and thus k-CAVI may also be useful as an index of vessel stiffness.

Arterial distensibility includes organic arterial sclerosis and a functional factor, which is controlled for BP (cardiac output) and constriction of vascular smooth muscle. Thus, CAVI and k-CAVI are slightly affected by changes in autonomic nervous function and muscular factors in peripheral vessels. This suggests that CAVI can be readily applied to the clinical evaluation of both organic arteriosclerosis and functional stiffness, including the vasoconstriction of smooth muscle.

There have been several reports on increases in arterial distensibility parameters being strongly correlated with aging. As vascular elasticity is greatly influenced by aging, it is necessary to remove the influence of physiological aging to evaluate the risk of blood vessel distensibility in patients with arteriosclerotic disease. In this study, both CAVI and baPWV were correlated with age, and thus distensibility parameters were evaluated separately in the elderly and middle-aged groups.

Study limitations

Our study had several limitations. First, it was difficult to clearly confirm the utility of CAVI in this study because of the small number of subjects. Further studies with larger numbers of subjects, particularly arteriosclerosis patients showing high CAVI values, are necessary. Second, the data in this study lacked objectivity; thus, comparative studies using objective measurements (for example, adrenalin concentration and HR variability (HRV) at the time of the CAVI measurement) are needed.

Conclusion

Our findings suggest that propofol-induced general anesthesia induces no changes in CAVI, but leads to decreases in baPWV, due to the different characteristics of the arterial distensibility parameters. CAVI is a simple quantitative method for reliably assessing arterial stiffness and is less influenced by short-term changes in BP occurring under general anesthesia. CAVI may thus be more closely linked with true arterial stiffness than baPWV.