T. Shiraev MBBS (Hons), BSc (Hons); M. G. Condous MBBS, FRACS.
VASCULAR & VASCULAR ACCESS
Incidence and outcomes of ruptured abdominal aortic aneurysms in rural and urban Australia
Article first published online: 30 JAN 2013
© 2013 The Authors. ANZ Journal of Surgery © 2013 Royal Australasian College of Surgeons
ANZ Journal of Surgery
Volume 83, Issue 11, pages 838–843, November 2013
How to Cite
Shiraev, T. and Condous, M. G. (2013), Incidence and outcomes of ruptured abdominal aortic aneurysms in rural and urban Australia. ANZ Journal of Surgery, 83: 838–843. doi: 10.1111/ans.12080
The corresponding author carried out this research as a medical student. He graduated in 2012 and is currently a JMO at St George Hospital, Kogarah, Australia.
- Issue published online: 29 OCT 2013
- Article first published online: 30 JAN 2013
- Manuscript Accepted: 7 JAN 2013
- AAA rupture;
Rural and regional populations suffer higher rates of preventable disease and all-cause mortality than urban areas, with rural areas of the USA experiencing double the rate of ruptured abdominal aortic aneurysms (AAAs). We investigated the incidence and outcomes of ruptured AAAs in an Australian rural and regional setting, and compared these with those of an urban population.
We undertook a retrospective analysis of all patients suffering AAA rupture in New South Wales (NSW) from 2009/2010 to 2010/2011. Variables included rates of rupture, mortality and intensive care admission. Urban and rural–regional areas were stratified according to NSW Health Local Health Districts, and comparisons between the two groups were performed.
Ruptured AAAs had an incidence of 4.1/100 000, with males twice as likely to suffer AAA rupture (P = 0.009), but females 88% more likely to die from rupture (P = 0.001). There was no significant difference between AAA rupture rates (5.0 versus 3.4 per 100 000; P = 0.054) nor case-fatality rates (41.22% versus 40.94%; P = 0.087) in rural–regional and urban populations. Patients in urban areas had a longer hospital stay (5 days versus 1 day, P = 0.001), were more likely to be admitted to intensive care unit (29.4% versus 19.9%, P = 0.001) and were less likely to be transferred to another hospital (19% versus 32%, P = 0.001).
AAA ruptures remain a significant public health burden. Rural and regional areas suffer disproportionately, which may be improved by implementation of AAA screening and funding for rural and regional hospitals to sustain adequate surgical and intensive care facilities.
The abdominal aorta is 1.5–2.5 cm in diameter depending on gender and body habitus, and is defined as aneurysmal once it reaches 3 cm. By this definition, 5% of men and 1.7% of women aged 65 years and older have an abdominal aortic aneurysm (AAA), with prevalence increasing by 6% per decade thereafter. Atherosclerosis is the most frequent cause of AAAs (via thinning of the arterial wall secondary to destruction of the media), and thus, risk factors include smoking, hyperlipidemia, family history and ageing. AAAs may be detected incidentally, or present with pain in the abdomen, lower back, iliac fossa or groin, as an expansile mass, undergoing thrombotic occlusion, or compressing surrounding structures. The most dire consequence of an AAA is rupture, which presents as cardiovascular collapse. Rupture risk is directly proportional to AAA diameter, from 9% per year when 5.5–5.9 cm to 33% per year when over 7 cm. If rupture occurs, 50% of patients die before reaching hospital, 12% die before and 20% die after surgery, resulting in ruptured AAAs having a mortality rate of 80–90%.[10-12] In 2011, there were 1881 elective AAA repairs in Australia (562 open and 1319 endoluminal), compared with 262 ruptured AAA repairs (210 open and 52 endoluminal). AAA rupture is responsible for the death of 1.3% of men and 0.8% of women over the age of 65, and although incidence of AAA rupture has decreased significantly in recent decades,[14, 15] it remains a significant cause of morbidity and mortality.
The health of rural and regional populations in Australia is significantly worse than urban populations, with higher rates of preventative disease and all-cause mortality. The USA similarly suffers marked geographic variation in access to and quality of health care received, with rural hospitals experiencing twice the ruptured AAA presentations, with subsequently increased mortality. Furthermore, rural hospitals are more likely to transfer ruptured AAA patients to larger centres for management, with the resultant delay in repair thus increasing mortality. This is of great concern to rural communities, as AAA rupture not only causes mortality in 1% of the elderly population, but an increase in odds ratio of 2.5 in rural communities makes AAA rupture a significant public health issue, with disproportionate burden of disease when compared to urban communities. A comparison of the incidence of ruptured AAAs in rural and urban populations has not been performed in Australia.
This study seeks to be the first to compare the rates of AAA rupture in a rural–regional and urban setting in Australia. We also aim to provide direction for future service planning in rural Australia by demonstrating the disparity in health between rural and urban populations in the vascular domain, which is essential considering the public health burden due to AAA rupture.
A retrospective analysis was conducted on demographic and clinical data from all public hospitals in New South Wales, via records from the Admitted Patients Data Collection (a NSW Health Department database). International Classification of Diseases (10th Revision, Australian Modification) code I71.3 (abdominal aortic aneurysm, ruptured) was used to identify patients who suffered AAA rupture between 2009/2010 and 2010/2011. Local Health Districts (LHD) were used to geographically stratify patient populations, with Central Coast, Illawarra, Nepean-Blue Mountains, Northern Sydney, South Eastern Sydney, South Western Sydney, Sydney and Western Sydney considered urban; and Far West, Hunter-New England, Mid North Coast, Murrumbidgee, Northern NSW, Southern NSW and Western NSW considered rural and regional. St Vincent's Hospital Network was considered to be part of the South Eastern Sydney LHD. Total population in each LHD was obtained from Health Statistics New South Wales.
Rates of AAA rupture and death from ruptured AAA were age-corrected using the December 2010 Australian population data as the denominator, based on inter-census estimates carried out by the Australian Bureau of Statistics. Case-fatality rate was calculated as the number of deaths due to ruptured AAA divided by the total ruptured AAA presentations in that geographic area. Patients transferred from rural to urban hospitals were not considered as urban admissions in order to avoid counting rural patients twice.
Comparisons of AAA rupture rate between the two groups (rural and urban) were conducted via independent sample Mann–Whitney U-tests for continuous variables, with Pearson's chi-squared and binomial testing for categorical variables. Data were analysed via SPSS Statistics Standard Edition 20 software (IBM, New York, NY, USA). All tests were two-sided, and a P-value of less than 0.05 was considered statistically significant. P-values less than 0.001 were rounded up to 0.001.
There were a total of 532 abdominal aortic aneurysm ruptures during the study period: 271 in 2009/2010 and 261 in 2010/2011. Median age of patients suffering ruptured AAA was 80 years (range 28–101). Rupture incidence for the study population was 4.1/100 000, with males suffering a significantly higher rupture rate than females (6.01 versus 2.23 per 100 000 population, P = 0.001). Rupture rate increased with age from 1 to 62 per 100 000 population in the 55–59 and 85 plus age groups, respectively (Fig. 1). Median length of stay in hospital was 3 days, with 44% of patients staying 1 day (of whom 68.2% died and 30.1% were transferred to another hospital). In addition, 43.0% of ruptured AAA patients died, 25.4% were discharged home, 24.1% were transferred to another hospital, 0.9% was transferred to a nursing home, 0.6% was discharged to palliative care and 1 person self-discharged against medical advice. The number of ruptured AAAs per 100 000 population in each LHD is shown in Figure 2.
Ruptured AAAs were responsible for 1.7 deaths per 100 000 population. Mortality increased with age, from 0.1/100 000 to 39.1/100 000 at 55–59 and 85 plus years, respectively. Men were significantly more likely to die from AAA rupture than women on a population level (2.58 versus 1.46 per 100 000, P = 0.001) (Fig. 3), but women were 88% more likely to die once rupture occurred (64% versus 34% case fatality, P = 0.009). Patients that died had a median age of 84 years, which was significantly higher than patients merely suffering AAA rupture (P = 0.001). Case-fatality rate by LHD is shown in Figure 4. Median length of stay was 1 day, with 71.2% of patients that died doing so within the first 24 h, 6.1% in the second 24 h, 7.4% on day 3 to day 4, 9.0% on day 5 to day 12 and 6.1% on day 13 and later. Furthermore, 82.1% of patients that died were not admitted to an intensive care unit (ICU) and 16.6% of all patients that died received an autopsy.
Comparisons between characteristics of patients suffering AAA rupture in rural–regional and urban areas are found in Table 1. There was a 45% higher rupture rate in regional and rural areas, although this did not reach significance (P = 0.054). Similarly, while there were 48% more deaths due to AAA ruptures in rural and regional areas, this was not statistically significant (P = 0.054). Patients had a significantly longer length of stay (P = 0.001) and were almost 50% more likely to be admitted to ICU (P = 0.014) in the urban setting. Significantly more patients from rural and regional areas were transferred to another hospital than in urban areas (32% versus 19%, P = 0.001).
|Urban||Rural and regional|
|Total number of AAA ruptures||326||206|
|Median age (years)||80||80|
|Sex distribution (% of total admissions)|
|AAA ruptures per 100 000 population||3.43 (1.23)||4.96 (1.33)|
|Case-fatality rate (%)||40.94||41.22|
|Deaths per 100 000 population||1.40 (0.50)||2.07 (0.73)|
|Median length of stay (days)||5||1*|
|ICU admission (% of total AAA rupture patients)||29.4||19.9*|
The current study demonstrated no significant difference between rates of abdominal aortic aneurysm rupture in rural and urban populations, nor a difference in mortality rate in these patients. However, patients with ruptured AAAs in urban areas had a markedly longer hospital stay, were more likely to be admitted to ICU and were less likely to be transferred to another hospital. There were significant differences in gender presentations and outcomes with ruptured AAAs.
Our study demonstrated AAA rupture rates of 6.0 and 2.2 per 100 000 for men and women, respectively, which compare favourably with the rates in the United Kingdom (14 and 2 per 100 000) and Sweden (18 and 4 per 100 000), but remain higher than those in New Zealand (4 and 1 per 100 000, respectively). However, we demonstrated mortality rates from ruptured AAA of 2.6 and 1.5 per 100 000 population for men and women, respectively, which is similar to rates in these countries, suggesting that Australian outcomes from ruptured AAAs are comparable internationally despite higher rupture rates, likely due to lower case-fatality rates. Additionally, the current study demonstrated an Australian case-fatality rate which has halved since in the early 1990s, suggesting a marked improvement in ruptured AAA incidence outcomes (also observed internationally), possibly due to reduced smoking prevalence, increased elective AAA repairs and improvements in anaesthetic and intensive care support.
Although it has been previously demonstrated that Australian men are six times more likely to suffer an AAA, the current study found that rate of AAA rupture was only two-and-a-half times higher, and interestingly, case fatality was almost half that of women. The reasons for the gender disparity in rates of AAA rupture and death are unclear, but may be due to the fact that using a 5.5-cm threshold for intervention is not appropriate in women given the initial size of their aorta, and repair at 4.5–5 cm should be considered instead. Regardless of the cause, greater awareness of AAAs in women is certainly necessary, considering their disproportionate mortality.[15, 25]
AAA rupture and death rates were 40–50% higher in rural and regional areas, and this approached but did not reach statistical significance. The lack of significance with a discrepancy of this size is possibly a type II error (a false negative) due to the high standard deviation resulting from the relatively small sample size (eight urban and seven rural populations). Further studies could rectify this by obtaining and pooling data from all Australian states and territories. The lack of significance may also be due to wide variability in rural–regional and urban rupture and death rates, possibly related to inconsistency in distances of LHD hospitals from the main urban centre of Sydney (e.g. Wyong Hospital of the Central Coast LHD is less than 100 km from Sydney, while Milton-Ulladulla Hospital of the Illawarra LHD is 220 km from Sydney). Patients in urban areas had a fivefold higher median length of stay in hospital, possibly due to the significantly higher rate of admission to ICU. Additionally, 70% more patients in rural and regional areas were transferred to another hospital, suggesting that the hospitals these patients present to are not adequately equipped for emergency vascular surgical procedures so are transported elsewhere and undergo care in hospitals far from their place of residence. It is also likely that the most complex patients are transferred from regional to urban hospitals, reducing median length of stay in rural areas while increasing it in urban areas. Outcomes of elective AAA repair in the USA are worse in rural than urban hospitals, with repair volume of less than 15 per year increasing mortality rate by an odds ratio of 1.3–1.7. Additionally, a US population study found that shifting AAA repairs to high-volume centres (not merely from rural to urban, but even from urban to specialist-urban centres) reduced mortality by 23%, and there is currently argument as to whether AAA repairs should be performed at high-volume centres only.[28-30] Furthermore, conversion of endovascular to open surgery, mortality and length of stay have been shown to decrease with increasing incidence of AAAs in a region, suggesting that it may be the available facilities, as well as the experience of the hospital staff (and importantly the vascular surgeon), that determine outcome in AAA repairs. This is supported by numerous studies demonstrating volume-dependent improvements in outcome in cardiac, gastrointestinal and oncological procedures as well as various vascular surgeries, especially AAA repairs.[33-36] Specifically, a recent study involving over 110 000 patients demonstrated that composite surgeon volume (i.e. the overall volume of vascular surgeries) rather than volume of open AAA repairs specifically predicted in-hospital mortality rates after open AAA repair, as surgeons with the highest composite volumes had half the mortality rate of those with the lowest composite volumes. Notably, it is surgeon case-volume rather than institution case-volume that determines in-hospital mortality in open AAA repairs. The current study demonstrated that AAAs in Australia do not have the same disparity in rupture rate and outcomes as others in the international setting, with a similar mortality rate in both rural and urban populations, which is likely due to composite surgeon volume in rural areas. The higher rates of preventative disease and all-cause mortality in rural areas in Australia is most likely due to the reduced availability of preventative services, including general practitioners and allied health, as well as limited access to specialist services. In the year 2000, the Federal budget made numerous financial commitments to improving the state of rural health, although significant health disparities remain. While ruptured AAA numbers and mortality rates are not areas in which a marked discrepancy exists, access to hospitals with adequate surgical facilities and ICUs must be improved for rural and regional populations.
Because of the high prevalence of AAAs and their considerable morbidity and mortality, a recent Cochrane review demonstrated that population screening for AAA via ultrasound in men aged 65–79 years reduced mortality by 48% and was cost-effective, and as such, both England and Scotland have commenced population-based screening programmes,[39, 40] with similar recommendations for the US population. There is every reason to implement a similar programme in Australia, especially as AAA ruptures and subsequent mortality contribute a substantial public health burden. In fact, such a recommendation had been made at a national government level in 2008, but little has been done. Furthermore, screening for AAAs may be even more important in rural and regional areas due to the discrepancies discussed earlier.
The main limitations of the current study are its retrospective nature and reliance on hospital admission data. There is also the potential for AAA rupture to be incorrectly diagnosed as another cause of death and thus be missed by the data collection methods of the current study, especially as death certificates are accurate in as few as 50% of cases. In fact, 84% of patients in the aforementioned study did not undergo autopsy, thus increasing the risk of another disease process being classified as a ruptured AAA (and vice versa).
The poor outcomes of rural and regional compared to urban AAA ruptures suggests that the prevention and management of AAAs in Australia needs to be altered on two fronts. Firstly, population-based screening should be implemented, as has been carried out successfully in numerous international settings.[39, 40] Secondly, rural and regional vascular facilities and ICUs must be made more widely accessible in order to better deal with the significant disease burden presented by ruptured AAAs in the rural setting.
The authors would like to thank John Agland for his kind assistance.
- 2Robbins and Cotran Pathologic Basis of Disease, 7th edn. Philadelphia, PA, USA: Elsevier, 2005., , .
- 3Melbourne Vascular Surgery Quality Initiative. Vascular surgery in Victorian public hospitals 2007/2008. Report to the Public. 2008, p. 28.
- 7Principles and Practice of Surgery. Sydney: Elsevier, 2007., , , .
- 8Isolated right testicular pain for six days: an unusual presentation of occult abdominal aortic aneurysm leak. Ann. R. Coll. Surg. Engl. 2011; 93: e1–2., , , .
- 13Australian and New Zealand Society for Vascular Surgery. Australasian Vascular Audit Report 2011. 2011. Edition, [Cited 7 Jan 2013.] Available from URL: http://www.anzsvs.org.au/national-audit/
- 14Reduction in ruptured aortic aneurysms is not due to increases in endovascular repairs. Am. Surg. 2011; 77: 1395–1398., , , , .
- 16Australian Human Rights Comission (AHRC). The State of Rural Health. 2011. 1st edn. [Cited 3 Apr 2012.] Available from URL: http://www.hreoc.gov.au/human_rights/rural_health/state_of_rural_health.html
- 19Health Statistics New South Wales. Estimated residential population growth by Local Health District, NSW. 2011. 1st edn. [Cited 9 Jul 2012.] Available from URL: http://www.healthstats.nsw.gov.au/Indicator/dem_pop_lhnmap
- 20Australian Bureau of Statistics. Australian Demographic Statistics Data Cubes, December 2010. 2010. 1st edn. [Cited 11 Jul 2012.] Available from URL: http://www.abs.gov.au/AUSSTATS/abs@.nsf/DetailsPage/3101.0Dec%202010?OpenDocument
- 31Geographic location does not limit innovative technology: broad spread assimilation of endovascular aortic aneurysm repair (EVAR). J. Am. Coll. Surg. 2011; 213: S112., , .
- 39Scottish Abdominal Aortic Aneurysm Screening Programme. CMO and Public Health Directorate HISD, 11th edn. 2010..
- 40NHS Abdominal Aortic Aneurysm Screening Programme. The newsletter of the NHS abdominal aortic aneurysm screening programme. NAAASP Insight 2009; 1: 1.
- 42Department of Health and Ageing. Horizon Scanning Technology Prioritising Summary: Targeted Screening for Abdominal Aortic Aneurysm. 2008. 1st edn. [Cited 28 Jun 2012.] Available from URL: http://www.horizonscanning.gov.au/internet/horizon/publishing.nsf/Content/BB580B674729F620CA2575AD0080F351/$File/_Aug_2008_Targetted%20screening%20for%20abdominal%20aortic%20aneurysm.pdf