Background
Lower extremity amputations, including excision of toes (usually referred to as minor amputation) and amputations above the ankle (usually referred to as major amputation), are significant complications of diabetes-related peripheral neuropathy and peripheral artery disease (PAD) [
1]. Individuals that undergo a major amputation have reduced quality of life and approximately three fold higher risk of mortality than people who have not undergone a major amputation [
2]. There is considerable variation in amputation rates in different geographical locations [
1]. The reported cumulative age-adjusted incidence of first lower extremity amputations in the United Kingdom ranges from 5 to 176 per 100,000 across different centres [
1].
Recent studies have suggested a decline in major amputation rates despite a rise in the prevalence of diabetes [
1,
3]. A report from Queensland Australia has suggested a significant 45 % decrease in diabetes-related major amputations between 2005 and 2010 [
4]. There has been limited study of trends in major amputation rates in high-risk sub-populations, such as Aboriginal and Torres Strait Islander people. Census data suggest that Aboriginal and Torres Strait Islander people represent approximately 3 % of the total Australian population [
5]. Aboriginal and Torres Strait Islander people more frequently have risk factors for major amputation, such as PAD [
6], diabetes, end-stage renal disease [
7] and smoking [
8]. A study from Western Australia reported that Aboriginal and Torres Strait Islander people with diabetes were approximately 40 times more likely to undergo a major amputation compared to non-Indigenous people with diabetes during the years 2000 to 2008 [
9]. Similarly, in a centre in Cairns, Australia, where approximately 10 % of the population are Aboriginal or Torres Strait Islander people [
10], 52 % of major amputations between 1998 and 2008 occurred in this group [
11]. Collectively, these data highlight a high incidence of major amputation in Aboriginal and Torres Strait Islander people [
12]. There has been no previous study of trends in major amputation rates over time in this population or comparison of clinical outcomes between Aboriginal and Torres Strait Islander people and non-Indigenous people having major amputations. Such data are important in order to assess whether current services for the high-risk foot are achieving their goals. The current study was focused in North Queensland, Australia where the prevalence of diabetes-related foot disease is reported to be high [
11,
13]. This study aimed to assess the annual incidence of and trends over time in major amputation for Aboriginal and Torres Strait Islander and non-Indigenous people in North Queensland. The study also aimed to identify significant risk factors for major amputation and compare the survival of Aboriginal and Torres Strait Islander and non-Indigenous people after major amputation.
Discussion
Recent systematic reviews have reported a decline in the incidence of major amputation in various countries and attributed this to improved multi-disciplinary care of the high-risk foot [
1,
3] This finding is consistent with recent analyses of data from the Australian population [
4,
20,
22,
23]. The age- and sex-adjusted incidence rates of major amputations in Queensland were reported to decrease by 26 % in the general population and 45 % in people with diabetes between 2005 to 2010 [
4]. Similarly, Kurowski and colleagues reported a 6 % annual decline in major amputation rates in people that had diabetes in Western Australia from 2000 to 2010 [
23]. The Fremantle Diabetes Study (FDS) reported a 72 % reduced risk of lower limb amputation in two cohorts with type II diabetes from the same Australian community over a 15-year period [
22]. In the current study the estimated annual incidence of major amputation did not significantly change during the period from 2000 to 2015 and was higher than reported Australian national incidence rates for the available time period [
20]. Reasons for this contrasting trend have not been identified in this study. It may reflect a growing burden of risk factors for major amputations in the North Queensland population [
11,
13]. It may also be related to poor implementation of optimal medical and foot care management [
24,
25].
Direct comparisons of major amputation rates between other studies are difficult since there is considerable variation in the methods used in previous investigations [
1,
20]. A number of previous investigations have reported the cumulative incidence of major amputation or compared the amputation rates at two time points to illustrate the change in major amputations rates over time [
1]. These studies have used a fixed at-risk population denominator throughout the study period which was obtained from census data [
1,
9,
26]. Interpretation of these prior analyses are difficult as they did not account for population growth and did not capture trends in amputation incidence over time [
1]. In the current study sensitivity analyses were performed to allow for various possible rates of population growth. When using a fixed at-risk population size a significant but small increase in major amputation incidence over time was observed. When an annual population growth rate of 2 % was assumed, which is representative of the average per annum population growth rate for Townsville [
19], no significant change in amputation rate over time was observed.
The clinical characteristics of people undergoing major amputation in this study were comparable to other Australian reports [
9,
11]. Aboriginal and Torres Strait Islander people who underwent a major amputation were more likely to be of younger age, have diabetes and be female than non-Indigenous patients, consistent with some previous reports from regional populations in Australia and internationally [
11,
27]. They were also more likely to have a diagnosis of ESRD, which is consistent with a previous study in North Queensland [
28]. After adjusting for age and sex differences, Aboriginal and Torres Strait Islander people were approximately twice as likely, as non-indigenous people, to die during follow-up. After adjusting for differences in the prevalence of IHD, diabetes, ESRD and hypertension this increased risk of mortality disappeared. Individuals with ESRD had approximately 2-fold greater risk of mortality. The findings suggest that the increased mortality rate in Aboriginal and Torres Strait Islander people was likely because of their high rate of co-morbidities, particularly ESRD [
28]. This observation has also been identified in other populations [
29].
This study had several limitations including its retrospective design and relatively small sample size. The study used participants’ self-identifying as Aboriginal people and Torres Strait Islanders at the time of admission. This was not cross-referenced with any other records and it is therefore possible that the number of Aboriginal people and Torres Strait Islanders may have been under or over estimated. Patient data were obtained retrospectively from hospital records and data on smoking were not recorded due to inconsistent reporting of smoking behaviour in hospital charts. This is a recognised limitation of data from hospital records and has been highlighted in previous datasets [
30]. Higher quality data is required to better assess the prevalence of risk factors for lower limb amputations in both Aboriginal people and Torres Strait Islanders and non-Indigenous Australians. Comparisons of major amputation rates with other populations was difficult due to the marked heterogeneity in reporting of these data in previous studies [
1,
3]. Furthermore, comparisons with national major amputation rates were limited to a 4 year period [
20]. Importantly, estimates of the number of Aboriginal people and Torres Strait Islanders within the population and the prevalence of diabetes within the THHS catchment area were based upon data collected at single time points. We were unable to include ESRD in the quasipoisson regression analysis as state-wide ESRD prevalence data stratified by each of the other covariates were not available to estimate the at-risk population. Notably, Aboriginal and Torres Strait islander status was not reported in 12,800 individuals from the census and this is a recognised limitation of census data [
5]. Sensitivity analyses conducted suggested that the main findings presented in this paper were robust to the assumptions about the size of the background at-risk population. Data were not collected on the in-patient management of the included people, such as the surgical treatment of PAD. A previous study performed in the USA reported that African American patients with PAD had fewer revascularisation attempts and worse limb salvage rates than non-African American patients [
31]. The higher major amputation rates in Aboriginal and Torres Strait Islander people could reflect less suitability for surgical treatment of limb ischemia but this was not investigated in this study [
11,
13]. Diabetes and ESRD are risk factors for infra-popliteal artery disease, which has been associated with a higher risk of major amputation than more proximal athero-thrombosis [
20]. In the current study, Aboriginal and Torres Strait Islander people had a greater prevalence of diabetes and ESRD and it is possible that they more frequently had infra-popliteal artery disease than non-Indigenous patients [
13], although this was not assessed.
In conclusion, this study suggests that the annual incidence of major amputation in North Queensland is high and is not decreasing over time, which is in contrast with reports of reducing amputation rates in other regions within Australia [
4,
22,
23] and internationally [
1,
3]. Aboriginal and Torres Strait Islander people more frequently underwent major amputation compared to non-Indigenous people, highlighting a major health gap for this population. Furthermore, Aboriginal and Torres Strait Islander people undergoing major amputations were at a greater risk of subsequent mortality associated with high rates of co-morbidities, such as ESRD, in this population. Dissemination of effective preventative treatments is needed to reduce the high incidence of major amputation in North Queensland.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.