The ankle–brachial index (ABI), which is a ratio of systolic blood pressure (SBP) of the arm to that of the leg, provides a simple, convenient, and non-invasive measure of lower-extremity vascular function. Reproducibility of the ABI is considered acceptable, with subsequent examinations on the same patient yielding a consistent result (Vowden and Vowden, 2001). According to the literature, normal values, independent of age and gender, should be above 0.9 (Rose, 2000; Brooks et al, 2001; McDermott et al, 2003). Results below this indicate varying levels of peripheral arterial disease. An ABI over 1.3 indicates arterial wall calcification and is unreliable in determining lower-limb perfusion (McDermott et al, 1998).
Effects of ageing and gender on arterial function
Ageing modifies vessel structure, particularly in elastic arteries; SBP increases because the vessel wall becomes less elastic and does not conform to blood pressure changes as readily (Vuong and Berry, 2002). Atrophy of large peripheral arterial smooth muscle cells also occurs in the ageing process, as does atherosclerosis, which thickens the vascular wall, reduces lumen size and therefore increases SBP (Vuong and Berry, 2002). In men, basal blood flow to the lower limb progressively declines throughout adulthood (Dinenno et al, 2001; Tanaka et al, 2001). Gender differences have been noted for brachial artery compliance, which has been found to be significantly higher in older women than older men (van der Heijden-Spek et al, 2000). These physiological findings certainly warrant an investigation of whether they influence the ABI.
Lower cut-offs for the ABI indicating normal blood pressure ranges have been reported from 0.8 to 1.1 (Hiatt et al, 1995; Moffatt and O’Hare, 1995; Rose, 2000; Brooks et al, 2001; Vowden and Vowden, 2001; McDermott et al, 2003), and angiographic evidence of peripheral vascular disease with an ABI of less than 0.90 has been stated to be 95 % sensitive but between 50 % and 99 % specific for angiographically significant peripheral arterial disease (Applegate, 1993; Olin, 1998). In addition, cuff size, gender, daily stresses and age have been reported to influence the ABI measure (Walsh and Bower, 1993; Anderson, 2002; Austin et al, 2003). It is therefore vital that research be undertaken to identify any differences in ABI results due to physiological differences between age groups and between genders, in order to improve evidence-based practice and patient outcomes through earlier diagnosis.
People aged <30 years or >40 years were recruited as part of the Charles Sturt University Allied Health Clinic diabetes complications screening programme. Exclusion criteria were the presence of cardiovascular disease, peripheral vascular disease, diabetes, kidney disease, smoking or excessive alcohol consumption. Participants’ data were also excluded from the analysis if the participants were found to be taking any medications for hypertension, kidney function, autonomic nervous system neuropathy or cardiac pathology. Sixty-one participants (26 males and 35 females) were included in the data analysis.
To avoid the effect of outliers on statistical analysis, the data were ranked and the inter-quartile range was multiplied by 1.5 to determine the normal range. Values outside this range were excluded for purposes of this analysis as they were deemed to indicate a high probability of peripheral arterial disease or vessel calcification (Fowkes et al, 1988).
Procedures and instrumentation
The ABI was determined using an examination plinth with participants in a supine position. After a minimum of 5 minutes, SBP measurements were taken. Examination rooms were all of a comfortable temperature, with minimal noise during ABI measurement. Using a pressure cuff, a sphygmomanometer and a Doppler ultrasound unit (Hadeco ES-1000 SPII, Hayashi Denki Co., Kawasaki, Japan) with an 8 MHz probe, SBP measurements were obtained from each brachial artery and each dorsalis pedis artery unless the dorsalis pedis artery showed no clear pulse, in which case the posterior tibial artery was used. The ankle SBP reading was then divided by the highest brachial SBP reading to give a numerical value: the ABI (Brooks et al, 2001).
Participants were divided into gender and age groups of <30 years and >40 years for the final data analysis.
Descriptive statistics and parametric statistical tests were used to observe characteristics of central tendency and variability within the data-set as appropriate for normative data collection.
To identify any statistically significant age and gender differences in the ABI and SBP, a standard two-sample t-test and fitted analysis of variance (ANOVA) were undertaken. The SBP and ABI values were the dependent variables and age and gender the independent variables to determine whether age and gender can predict a significant difference in SPB and ABI. Statistical significance was identified using α=0.05.
The mean and standard deviation values for the two age groups are shown in Table 1. In all of the parameters the older age group had higher scores. The age differences for SBP values all reached a high level of significance. When this was translated into the ABI, although the older age group had a higher mean for both the left and the right reading, no such significance was reached.
Table 2 shows the results for the gender comparison. Our results indicate that a statistically significant difference exists between the brachial SBPs of males and females. No significant difference was noted for the ankle SBPs or for the ABIs.
Table 3 shows results by age group and gender. Comparison between genders within each age group indicated that in the <30 years group both brachial SBPs were significantly different (P<0.05), whereas in the >40 years group only the right brachial SBP was significantly different (P<0.05). Comparison between age groups revealed significant differences (P<0.05) for the brachial and ankle SPBs for each gender. No significant differences were noted for ABI measures when controlling for age or gender.
Identifying age and gender trends with respect to the ABI has implications for the evidence-based practice of patient management. The literature has already highlighted physiological variations in SBP between genders and age ranges (Vuong and Berry, 2002). It is because of these differences, and the importance of using scientific evidence to base clinical decisions on, that further research into the ABI was warranted.
Age and ABI
Previous research on lower-limb blood flow has demonstrated a difference in leg blood flow between younger and older males (Dinenno et al, 2001). A variation in blood vessel morphology with age has also been shown in the brachial artery (van der Heijden-Spek et al, 2000). A statistically significant difference was observed in our study between age groups for the brachial and ankle SBPs. Significant differences in SBP were also found when comparing the two age groups within each gender. However, significance in SBP differences did not equate to significance when converted to the ABI. Despite this, the results for the two age groups shown in Table 1 indicate a trend in the ABI, we feel, with the older age group having higher ABI scores.
Of interest are the results of the range variation in ABI results between age groups. Here the younger age group had a left ABI range of 0.78–1.24 (mean ± standard deviation, 1.02 ± 0.1) and a right ABI range of 0.73–1.2 (1.01 ± 0.1), while the older age group had a left ABI range of 0.92–1.33 (1.04 ± 0.1) and a right ABI range of 0.86–1.17 (1.02 ± 0.1). These results can be used as an indicator of normal range when testing patients in either of these two age groups. The range of ABI values suggests that a lower cut-off for normal values may be applicable, compared with the current accepted cut-off of 0.9. However, note that the upper limit remained fairly close to the current suggested upper limit of 1.3, whereas some authors (e.g. Resnick et al, 2004) have suggested an upper cut-off of 1.4.
Gender and ABI
Statistical tests comparing ABI scores between genders found no significant differences. The findings of this research support those of other studies (e.g. van der Heijden-Spek et al, 2000) that found a significant difference in the brachial artery SBP between genders. Results from the ankle SBP were not significantly different between genders. A difference in ABI was noted between age groups and gender in this study. The fitted ANOVA model indicated that both age and gender were able to predict significant differences for brachial SPB measures (P<0.01), but only age was able to predict significant differences for ankle SBP measures (P<0.01). The model was not applicable for identifying significant differences for the ABI values. The respective left and right ABI ranges were found to be 0.78–1.33 and 0.73–1.24 for females and 0.81–1.19 and 0.85–1.2 for males. These ranges again suggest that a lower normal ABI cut-off can be considered. While the ABI differences between genders, or age groups, did not reach statistical significance, we feel they do indicate that consideration of age and gender may be warranted for diagnosis of peripheral vascular disease using ABI.
Gender and age have been shown to influence normal function of blood flow by leading to an increase in vessel wall hardening and other structural changes. Some of these physiological changes are reflected by changes in the SBPs recorded from either the arm or ankle. However, age and gender differences in the ABI were not statistically significant in our study. We observed that the mean ABI for the younger age group (<30 years) was lower in both the female and male groups and that the mean male ABI was lower compared with the mean female ABI. We recommend that future research involves a larger cohort and uses 10-year increments for age to assess incremental changes in ABI. This may provide data similar to those reported for age and gender influence on blood pressure when assessing hypertension, where absolute values in blood pressure for different age groups or between genders are not significantly different but reflect a trend in the population.