International Journal of Cardiology 180 (2015) 264–269
Contents lists available at ScienceDirect
International Journal of Cardiology journal homepage: www.elsevier.com/locate/ijcard
Diabetes and poor glycaemic control in rural patients with coronary artery disease☆ Catherine Ngan a,⁎,1, Ya-Chu May Tsai a,1, Dharshan Palasubramaniam a,1, Amy Wilson-O'Brien a,1, Jamie Layland a,1, Robert Whitbourn b,1, Andrew Wilson a,1 a b
The University of Melbourne, Department of Medicine, St Vincent's Hospital, Melbourne, Australia St Vincent's Hospital, Melbourne, Australia
a r t i c l e
i n f o
Article history: Received 3 August 2014 Received in revised form 18 November 2014 Accepted 22 November 2014 Available online 26 November 2014 Keywords: Cardiovascular risk factors Diabetes Rural medicine Epidemiology
a b s t r a c t Background: The burden of cardiovascular disease is higher in rural populations. Existing data on rural cardiovascular health is mainly based on community surveys. Regional differences are not well addressed. This study aims to identify regional inequalities in cardiovascular risk factors (CVRFs) in Australian patients with suspected coronary artery disease. Methods and results: 538 subjects (72% male; mean age 63 years) were recruited from a single cardiac catheter laboratory over a 24-month period. Subjects were stratified into Remoteness Areas (RAs) according to the Australian Standard Geographical Classification (RA1 corresponds to Major Cities, RA2 to Inner Regional Areas, RA3 to Outer Regional Areas). Body-mass index, blood pressure, hypertension, dyslipidaemia, diabetes and smoking history were recorded. A blood sample taken before the angiogram was analysed for lipids and fasting blood glucose (FBG). Distribution of the study population across RA1, RA2 and RA3 was 34.8%, 46.1% and 19.1%. Only FBG (p = 0.019) and diagnosed diabetes (p = 0.009) were significantly different i.e. higher in RA1. Of those without known diabetes, RA3 had the highest prevalence of dysglycaemia (p = 0.023) with two-thirds having either pre-diabetes or undiagnosed diabetes. Logistic regression showed that age and RA3 were the only statistically significant predictors of elevated FBG. Conclusion: CAD patients from remote Australia had higher rates of pre-diabetes, undiagnosed diabetes and poorer glycaemic control. Analysis of the main CVRFs revealed a regional inequality in the recognition and management of diabetes alone. Attention to this gap in rural and urban healthcare is crucial to future cardiovascular health outcomes in Australia. © 2014 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Cardiovascular disease (CVD) is the leading cause of noncommunicable deaths in the world, accounting for almost 40% of deaths under 70 [1]. With population ageing being a significant global trend, particularly in developed countries, CVD mortality and morbidity are projected to rise substantially. In developed countries such as the United States, Canada, UK and Australia, CVD is the first or second leading cause of mortality, with coronary artery disease being the chief specific cause of death in these countries [2–5]. In addition to its impact on national mortality and morbidity, CVD also imposes significant financial and societal cost from the magnitude of healthcare expenditure, loss of ☆ Funding sources: University of Melbourne Faculty of Medicine, Dentistry and Health Sciences Equipment Grant (2010). ⁎ Corresponding author at: The University of Melbourne, Department of Medicine, St Vincent's Hospital, Melbourne, Australia. E-mail address: [email protected] (C. Ngan). 1 This author takes responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation.
http://dx.doi.org/10.1016/j.ijcard.2014.11.120 0167-5273/© 2014 Elsevier Ireland Ltd. All rights reserved.
productivity due to work absenteeism, early retirement and premature mortality [6]. It follows that managing the burden of CVD is a health priority in the modern society. It is well accepted that the progression of ischemic heart disease is linked to a history of modifiable and inherited characteristics, and thus addressing modifiable cardiovascular risk factors (CVRFs) are mainstays for management of general cardiovascular health [7,8]. Knowledge of the geographic prevalence of these risk factors is essential to the formation of population specific health policies and targeted delivery of health services. In particular, the countries mentioned above face the challenge of addressing the health concerns of rural and urban communities; 20–30% of US, Canadian and Australian populations live outside of Major Cities, where there are unique geographical, lifestyle and health characteristics. The impact of CVD is particularly high in these rural populations, where in Australia the disease burden can be up to 15% greater [9]. Rates of cardiovascular hospitalisations and deaths increase with remoteness, significantly contributing to the overall poorer health outcomes in regional and remote communities [10–13]. There is also a
C. Ngan et al. / International Journal of Cardiology 180 (2015) 264–269
greater prevalence of modifiable CVRFs in populations outside of the Major Cities, with higher rates of hypertension, obesity, smoking and risky alcohol consumption [11,13]. Since much of the burden of CVD can be decreased by managing CVRFs, greater knowledge of the specific differences in risk factor awareness and management across rural populations is vital to improving health outcomes. Awareness of regional health has been a focus of Australian health policy for many years, and the existing literature on this topic is based on comprehensive risk factor surveys of regional and national health [11,14,15]. Local studies on CVRFs have recruited from electoral rolls or used Census data, where rurality has not necessarily correlated with cardiovascular mortality or morbidity [16,17], or the findings have been gender specific [18]. However, the usefulness of this data is limited by the lack of blood tests in determining the presence of a risk factor, relying instead on patient recall and potentially reflecting patient awareness rather than true prevalence. Conditions that rely on diagnostic blood tests, such as diabetes, are thus likely under-reported. Another feature of these population surveys is that there is little focus on groups most at risk. In contrast, this study examines the regional prevalence and management of CVRFs in patients with suspected coronary artery disease, in a well-characterised Australian population. The investigation of reported and biochemical risk factors along with angiographic evidence offers a fresh insight into cardiovascular healthcare in a group of patients for whom risk factor management is a priority. 2. Methods 2.1. Ethics approval This study was approved by the St Vincent's Hospital Research Governance Unit, in accordance with National Health and Medical Research Council guidelines. All patients were invited to participate before their procedure and written consent was obtained. 2.2. Study population Study subjects were recruited from St Vincent's Hospital Melbourne, which is a major tertiary public teaching hospital. Patients presenting to the St Vincent's Cardiac Catheterisation Laboratory between May 2009 and May 2011 for coronary angiography and/or percutaneous coronary intervention (PCI) were eligible for this study. Subjects were recruited as part of the Biomarkers of Atherosclerosis, Vascular and Endothelial Dysfunction in Heart Disease Study (BRAVEHEART), and thus the following exclusion criteria were applied: patients with chronic or acute infections, systemic inflammatory conditions, recent or untreated malignancies and serum creatinine levels greater than 160 μmol/l. 2.3. Patient data Cardiovascular histories were obtained through interview and review of medical records. The use of hypoglycaemic, lipid lowering and antihypertensive medication was recorded. Blood pressure (BP) on admission and body-mass index (BMI) were recorded. BMI was calculated as body weight in kilograms divided by the square of the height in metres. Hypertension was defined according to the 2007 Cardiac Society of Australia and New Zealand (CSANZ) Guidelines as BP N140/90 mm Hg [19]. Obesity was defined according to the World Health Organisation (WHO) Guidelines as BMI ≥ 30 kg/m2 [20]. Severity of CAD was angiographically assessed by a cardiologist. Vessel disease was defined as a greater than 50% stenosis of a major coronary artery or history of stenting of that artery. The location of disease and the number of diseased vessels were recorded. 2.4. The ASGC-RA classification This study uses the Australian Standard Geographical Classification-Remoteness Areas (ASGC-RA) system, which was devised by the Australian Bureau of Statistics to organise data into broad geographical categories based on population size and distance from an urban centre [21]. The Remoteness Area (RA) boundaries are updated after each Census, and this report is based on the 2006 Census, which at the time was the most recent. To classify the study population, the postcode of the subject's current address was obtained from the hospital registry. All subjects were classified into three RAs: Major Cities, Inner Regional Areas and Outer Regional Areas. 2.5. Biochemical blood analysis Arterial blood samples were collected via femoral or radial arterial access at the beginning of angiography before the use of iodinated contrast agent. Samples were analysed through St Vincent's Pathology for lipid profile, fasting blood glucose (FBG) and glycated haemoglobin (HbA1c). Normal FBG and HbA1c levels were defined according to the
265
2004 American Diabetes Association (ADA) Guidelines [22]: normal FBG b5.6 mmol/l, impaired FBG 5.6 to 6.9 mmol/l, diabetic FBG ≥7 mmol/l and normal HbA1c b6.5%. Normal lipid levels were defined according to the 2007 CSANZ Guidelines [19]: total cholesterol (TC) b4 mmol/l, low-density lipoprotein cholesterol (LDL-C) b2.0 mmol/l, high-density lipoprotein cholesterol (HDL-C) N1.0 mmol/l and triglycerides (TG) b1.5 mmol/l.
2.6. Statistical analysis Statistical analyses were undertaken using SPSS 17 statistical software. Continuous variables were examined for normality of distribution with the Kolmogorov–Smirnov test. Variables not normally distributed were appropriately log-transformed for parametric analyses. Differences in physical measurements, reported risk factors, biochemical risk factors and severity of CAD across the RA groups were determined by one-way analysis of variance (ANOVA) with post-hoc Tukey analysis for continuous variables, and chi-square analysis for discrete variables. Logistic regression analysis was used to assess independence of correlations, and to identify the variables most significantly associated with elevated FBG. The variables included in the model were age, gender, hypertension, dyslipidaemia, smoking history, BMI, use of hypoglycaemic medication, the presence of CAD and RA. Significant values were defined as p b 0.05.
3. Results 3.1. Characteristics at baseline 538 subjects (72% male; mean age 63 years) were included in the analyses. All subjects were Victorian residents and were classified by their postcode into Major Cities, Inner Regional Areas and Outer Regional Areas as defined by the ASGC-RA system [21]. There were 187 subjects from Major Cities (75% male, mean age 61 years), 248 from Inner Regional Areas (70% male, mean age 63 years) and 103 from Outer Regional Areas (72% male, mean age 64 years) (Table 1, Fig. 2).
Table 1 Baseline characteristics. Total number in population Average age (years) Gender Remoteness Area (number of patients) Major Cities Inner Regional Outer Regional Reported CVRF (%) Hypertension Dyslipidaemia History of smoking Diabetes mellitus Anthropometric measurements BMI (kg/m2) Systolic BP (mm Hg) Diastolic BP (mm Hg) Biochemical risk factors FBG (mmol/l) HbA1c (%) TC (mmol/l) TG (mmol/l) HDL-C (mmol/l) LDL-C (mmol/l) Medication Reported diabetes (n = 146) Hypoglycaemic medication Reported hypertension (n = 434) Antihypertensives Reported dyslipidaemia (n = 465) Lipid lowering medication Angiography Coronary artery disease 1 vessel disease 2 vessel disease 3 vessel disease
538 63 (0.5) 72% male 187 248 103 81% 87% 71% 27% 29.6 (0.2) 135 (1) 82 (1) 6.4 (0.1) 6.1 (0.1) 4.3 (0.0) 1.6 (0.0) 1.1 (0.0) 2.5 (0.0)
67% 81% 81% 70% 27% 23% 21%
Data expressed as mean (standard error of mean) where appropriate. CVRF represents cardiovascular risk factor; BP, blood pressure; BMI, body-mass index; FBG, fasting blood glucose; HbA1c, glycated haemoglobin; TC, total cholesterol; TG, triglycerides; HDL-C, high-density lipoprotein cholesterol; and LDL-C, low-density lipoprotein cholesterol.
266
C. Ngan et al. / International Journal of Cardiology 180 (2015) 264–269
Fig. 1. Remoteness Areas of Victoria. Image adapted from [21]
3.2. Characteristics by Remoteness Area Risk factors, BMI and BP were compared between RA. Chi-square analysis of these risk factors revealed a significant difference between the RAs only for reported diabetes mellitus (p = 0.009). Reported diabetes was most prevalent in Major Cities and least prevalent in Inner Regional Areas. BMI and BP were similar across the RAs, as confirmed by ANOVA (Table 2). Average lipid, FBG and HbA1c results were compared between RA. ANOVA tests revealed significant differences between RA only for FBG
(p = 0.019) and HbA1c (p = 0.002). The average FBG in each RA as well as the overall population were above the cut-points for normal levels (FBG b 5.6 mmol/l) [22]. Both FBG and HbA1c levels were highest in subjects from Major Cities, with an average FBG of 6.7 mmol/l, and HbA1c of 6.4% (Table 3). 3.3. Use of medication The use of hypoglycaemic, lipid lowering and antihypertensive medication was analysed in subjects who reported having the relevant risk factor. Analysis between the regions revealed no significant differences, however, the use of hypoglycaemic medication followed a downward trend with increasing rurality (Table 4). Table 2 Distribution of reported risk factors and anthropometric measurements by Remoteness Area.
Reported CVRF (%) Hypertension Dyslipidaemia History of smoking Diabetes mellitus Anthropometric measurements BMI (kg/m2) Systolic BP (mm Hg) Diastolic BP (mm Hg) Fig. 2. Distribution of study population by Remoteness Area.
Major City
Inner Regional
Outer Regional
p-Value
79% 85.0% 71.0% 34.9%
84% 87.9% 70.9% 21.9%
77% 86.3% 73.0% 25.5%
0.204 0.680 0.920 0.009⁎
29.3 (0.4) 134 (2) 81 (1)
29.6 (0.4) 135 (2) 82 (1)
29.7 (0.6) 137 (3) 81 (1)
0.838 0.391 0.677
Data expressed as mean (standard error of mean) where appropriate. ⁎ p b 0.05.
C. Ngan et al. / International Journal of Cardiology 180 (2015) 264–269 Table 3 Biochemical risk factors by Remoteness Area.
FBG (mmol/l) HbA1c (%) TC (mmol/l) TG (mmol/l) HDL-C (mmol/l) LDL-C (mmol/l)
267
Table 5 Stages of dysglycaemia by Remoteness Area.
Major City
Inner Regional
Outer Regional
p-Value
6.7 (0.2) 6.4 (0.1) 4.3 (0.1) 1.6 (0.1) 1.1 (0.0) 2.5 (0.1)
6.2 (0.1) 6.0 (0.1) 4.3 (0.1) 1.6 (0.1) 1.1 (0.0) 2.5 (0.1)
6.4 (0.2) 6.0 (0.1) 4.2 (0.1) 1.6 (0.1) 1.1 (0.0) 2.5 (0.1)
0.019⁎ 0.003⁎ 0.699 0.779 0.169 0.953
Data expressed as mean (standard error of mean). FBG represents fasting blood glucose; HbA1c, glycated haemoglobin; TC, total cholesterol; TG, triglycerides; HDL-C, high-density lipoprotein cholesterol; and LDL-C, low-density lipoprotein cholesterol. ⁎ p b 0.05.
History of diabetes (N = 146) Normal FBG: b5.6 mmol/l Impaired FBG: 5.6–6.9 mmol/l Diabetic FBG: 7.0+ mmol/l No history of diabetes (N = 392) Normal FBG: b5.6 mmol/l Impaired FBG: 5.6–6.9 mmol/l Diabetic FBG: 7.0+ mmol/l Mean HbA1c (%) History of diabetes No history of diabetes
Major City
Inner Regional
Outer Regional
p-Value
8% 18% 75%
13% 23% 59%
12% 36% 52%
0.745 0.129 0.093
52% 43% 5%
52% 44% 4%
33% 59% 8%
0.023⁎ 0.052 0.691
7.6 (0.2) 5.7 (0.1)
7.4 (0.2) 5.6 (0.0)
7.1 (0.3) 5.7 (0.1)
0.295 0.366
Data expressed as mean (standard error of mean). FBG represents fasting blood glucose; HbA1c, glycated haemoglobin. ⁎ p b 0.05.
3.4. Dysglycaemia The population was divided into those with and without a past diagnosis of diabetes to account for subjects whose glycaemic status may be controlled by lifestyle or medical therapy. The population of each RA was further divided into stages of dysglycaemia as defined by the ADA [22] (Table 5). Among those with a diagnosis of diabetes, subjects from Major Cities had the poorest glycaemic control, with 75% having an FBG ≥ 7.0 mmol/l. Conversely, diabetic subjects from Outer Regional Areas had better glycaemic control, with 52% having elevated FBGs (p = 0.093) (Fig. 3). Among those without a diagnosis of diabetes, two-thirds of subjects from Outer Regional Areas had fasting sugars in the impaired and diabetic range so ≥5.6 mmol/l (p = 0.023). Of this group, 59% had impaired fasting sugars and 8% had an FBG ≥7.0 mmol/l (Fig. 4). Average HbA1c was also examined, where normal levels were defined as b6.5% according to ADA guidelines [22]. Subjects with a diagnosis of diabetes had elevated average HbA1c, with the highest average in Major Cities and the lowest in Outer Regional Areas. ANOVA tests showed that these differences were not statistically significant. Logistic regression analysis was performed to assess the independence of variables associated with elevated FBG, defined as N5.6 mmol/l. The variables included in the model were age, gender, hypertension, dyslipidaemia, smoking history, BMI, use of hypoglycaemic medication, the presence of CAD and RA. Of these factors, only age, medication and RA made a unique statistically significant contribution to the model (p b 0.05). After adjusting for all the above factors, the odds of having elevated FBG in Outer Regional Areas compared to Major Cities was 2.3 (95% CI 1.2–4.4) (Table 6).
3.5. The severity of coronary artery disease The presence of coronary artery disease and the number of diseased coronary vessels were compared between each RA. Chi-square analysis showed no significant differences in the prevalence of disease or the number of diseased vessels (Table 7).
4. Discussion This Australian study differs from contemporary risk factor surveys because of its focus on patients for whom CVRF management is a priority. Recruitment from a catheterisation laboratory provided this study with individual biochemical blood analysis and angiographic evidence of coronary artery disease, which enabled all statistical analysis and observations to be accurately corrected for severity of CVRFs and coronary disease. The study population was recruited from the state of Victoria, where 25% of the Australian population resides [23]. Of the Victorian population, 75% live in Major Cities, 20% live in Inner Regional Areas and 5% live in Outer Regional and Remote Areas [24] (Fig. 1). Although subjects were recruited from a tertiary metropolitan hospital, all three major RAs were well represented in the study population: 35% were from Major Cities, 46% from Inner Regional Areas and 19% from Outer Regional Areas. The key finding from our regional analysis was that after correcting for age, gender, other risk factors and severity of CAD, there were clear regional differences in the prevalence and management of diabetes alone. Baseline analysis of this study population revealed a remarkable similarity in risk factor profiles, which is in contrast to state and national surveys reporting a higher prevalence of CVD and CVRFs in regional populations [10,25]. Our investigation of high-risk patients revealed similarities in reported risk factors, blood lipid profiles and severity of coronary disease between the three RAs. The use of antihypertensive and lipid lowering medication was also similarly consistent, and this was reflected in the relatively well-controlled BP and lipid profile of each RA. However, one risk factor showed significant regional variation: the prevalence, awareness and management of hyperglycaemia. Hyperglycaemia is characteristic of diabetes mellitus, where the secretion or action of insulin is chronically impaired and the tissue
Table 4 Use of medication for reported risk factors by Remoteness Area.
Reported diabetes (n = 146) Hypoglycaemic medication Reported hypertension (n = 434) Antihypertensives Reported dyslipidaemia (n = 465) Lipid lowering medication
Major City
Inner Regional
Outer Regional
p-Value
73%
66%
54%
0.234
79%
81%
82%
0.814
80%
87%
85%
0.171
Fig. 3. Stages of dysglycaemia by Remoteness Area in patients with a diagnosis of diabetes.
268
C. Ngan et al. / International Journal of Cardiology 180 (2015) 264–269 Table 7 Severity of coronary artery disease.
Coronary artery disease Number of diseased vessels
Fig. 4. Stages of dysglycaemia by Remoteness Area in patients without a diagnosis of diabetes.
uptake of blood glucose is thus compromised [22]. The term hyperglycaemia also incorporates an intermediate stage of ‘pre-diabetes’, where FBG is above normal levels but still below diabetic cut-points [22]. Both diabetes and pre-diabetes are major risk factors of CVD development and premature mortality [26,27], and are independent predictors of CVD mortality [26]. Furthermore, diabetes is an equivalent risk factor for future myocardial infarction as having a history of a previous infarct [28]. Of relevance to this study population, diabetes is also a risk factor for restenosis after angioplasty or stenting [2]. Such evidence provides an important rationale for aggressive management and monitoring of patients with high blood glucose levels. Yet in this population of subjects with suspected coronary artery disease, there was significant under-recognition and suboptimal management of diabetes. Independent of coronary disease severity, the prevalence of diabetes was highest in subjects from Major Cities at 35%, followed by Outer Regional Areas at 26% and Inner Regional Areas at 22% (p = 0.009). Fasting sugars were also highest in Major Cities at 6.7 mmol/l, followed by Outer Regional Areas at 6.4 mmol/l and then Inner Regional Areas at 6.2 mmol/l (p = 0.019). HbA1c levels were similarly elevated in Major Cities at 6.4% (p = 0.003). The use of hypoglycaemic medication was again highest in Major Cities at 25%. Thus the first impression of this study cohort is that Major Cities have more diabetes, and accordingly have higher fasting sugars, higher HbA1c levels and greater use of hypoglycaemic medication. However, these results do not account for the influence of lifestyle or medical therapy on FBGs. To gain a more accurate representation of
Table 6 Logistic regression applied to entire cohort. B
RA 1 RA 2 RA 3 Age Gender Smoker Dyslipidaemia Hypertension BMI Use of hypoglycaemics CAD Constant
S.E.
Wald
df
Sig.
2 1 1 1 1 1 1 1 1 1
.009⁎ .775 .011⁎ b.001⁎ .041⁎ .974 .900 .212 .096 b.001⁎
1 1
.498 .002
−.07 .84 .04 .52 .01 −.04 −.36 .04 3.46
.25 .33 .01 .25 .25 .33 .29 .02 .73
9.51 .08 6.48 13.47 4.17 .00 .02 1.56 2.77 22.26
−.18 −3.24
.26 1.05
.46 9.48
Exp(B)
95% C.I. for EXP(B) Lower
Upper
.93 2.31 1.04 1.67 1.01 .96 .70 1.04 31.89
.57 1.21 1.02 1.02 .62 .51 .40 .99 7.57
1.52 4.39 1.06 2.74 1.65 1.81 1.23 1.08 134.37
.84 .04
.51
1.39
RA represents Remoteness Area; BMI, body-mass index; and CAD, coronary artery disease. ⁎ p b 0.05.
1 2 3
Major City
Inner Regional
Outer Regional
p-Value
66% 24% 20% 23%
73% 28% 25% 19%
72% 28% 22% 23%
0.347 0.561 0.378 0.619
glycaemic control, the study population was divided into those with a history of diabetes and those without, and these two groups were divided into normal, impaired and diabetic FBG. This subgroup analysis was revealing in the population without a diagnosis of diabetes; twothirds of apparently non-diabetic patients from Outer Regional Areas had elevated fasting glucose levels (p = 0.023), with the highest prevalence of pre-diabetes at 59%, and also the highest prevalence of undiagnosed diabetes at 8%. Once again, these results were independent of disease severity. This is a significant and worrying regional discrepancy, which points to poorer disease recognition in areas furthest from city centres. To assess the independence of variables associated with an elevated FBG, logistic regression analysis was performed. After correcting for potential confounding factors including age, gender, hypertension, dyslipidaemia, smoking history, BMI and ischemic heart disease as determined by angiography, the odds of having elevated fasting sugars in Outer Regional Areas compared to Major Cities was 2.3. In other words, subjects furthest from city centres had more than double the risk of having diabetes or pre-diabetes, independent of coronary disease severity. However, not all was worse in subjects from Outer Regional Areas. Among those with known diabetes, subjects from Major Cities had the poorest glycaemic control, with three-quarters having an FBG ≥ 7.0 mmol/l. Conversely, diabetic subjects from Outer Regional Areas had better glycaemic control, with about half having normal fasting sugars (p = 0.093). It would appear that although subjects from Outer Regional Areas were more likely to have high FBGs, once diagnosed, these subjects had better glycaemic management. The use of hypoglycaemic medication was also analysed among those with a diagnosis of diabetes. 73% of subjects from Major Cities were on hypoglycaemic agents, compared to 54% in Outer Regional Areas. This difference, however, was not statistically significant. From these observations on reported risk factors, blood tests and use of medication, regional inconsistencies were apparent in the prevalence and management of diabetes alone. The high proportion of unrecognised hyperglycaemia raises concerns about awareness and management of diabetes in both regional and metropolitan areas. This is particularly true for subjects from Outer Regional Victoria, of whom two-thirds of those without a history of diabetes had unrecognised hyperglycaemia. The relevance of this finding extends beyond cardiovascular health; diabetic patients have a high burden of disease, and can go on to develop peripheral vascular disease, cerebrovascular disease, retinopathy, neuropathy, and renal disease, to name but a few complications [4,16–18]. 4.1. Strengths and limitations A major strength of this study was the inclusion of biochemical profiles in a specific population of patients with known or suspected coronary artery disease, of which 70% had significant vessel disease. Whereas most risk factor studies survey broadly to gain a general overview of the population, this study recruited from a pool of patients for whom CVRF management is a priority. Additionally, the recruitment of patients presenting to a catheterisation laboratory provided this study not only with fasted blood samples for detection of dyslipidaemia or hyperglycaemia, but also angiographic evidence of disease, all of which are unique features of the data collection.
C. Ngan et al. / International Journal of Cardiology 180 (2015) 264–269
Investigation of potential factors behind the hyperglycaemia observed in regional populations was beyond the scope of this study, but it is known that access to healthcare services, income, education and employment are all factors that contribute to poorer rural health. However, it seems unlikely that such broad characteristics could selectively influence one risk factor out of all the major CVRFs. Further investigation into prescribing practices, patient compliance and lifestyle management of diabetic patients would provide deeper insight into this regional inequality. There is also the potential for referral bias in this study population, where patients presenting to this referral centre are more aware of their cardiovascular health than the general population. Nonetheless, the study findings remain pertinent — despite known ischemic heart disease and perhaps greater awareness of cardiovascular health, there was still far more dysglycaemia in those without a diagnosis of diabetes in the Outer Regional cohort. 5. Conclusions In conclusion, analysis of reported CVRFs, blood tests and use of medication in a geographically diverse population of subjects with suspected coronary artery disease showed an isolated regional inequality in the prevalence and management of hyperglycaemia alone. Subjects from Major Cities were more likely to be diagnosed with diabetes but have poorer glycaemic control. Conversely, subjects from rural Australia were less likely to be diagnosed, but once diagnosed had better glycaemic control. In terms of clinical practice, this study demonstrates that presentation for a coronary angiogram or PCI is an opportunity for a full physical and biochemical CVRF assessment. Despite patient awareness of heart disease, diabetes is particularly under-managed across urban and rural regions, and all patients would benefit from education about cardiovascular health and glycaemic control in the setting of an angiogram. Although detection of diabetes is lower in patients from rural areas, once diagnosed, there appears to be better glycaemic control, and it is in these patients that opportunistic screening and education may be of most benefit. Conflict of interest None. Acknowledgements The authors thank the Department of Medicine at St Vincent's Hospital Melbourne, and the staff at the St Vincent's catheter laboratory for their assistance and support. References [1] A. Alwan, Global Status Report on Noncommunicable Diseases 2010, World Health Organisation, Italy, 2011.
269
[2] A.S. Go, D. Mozaffarian, V.L. Roger, E.J. Benjamin, J.D. Berry, M.J. Blaha, et al., Heart disease and stroke statistics — 2014 update: a report from the American Heart Association, Circulation 129 (2014) e28–e292. [3] Australian Institute of Health and Welfare, Australia's Health 2012. Australia's Health Series No. 13. Cat. No. AUS 156, AIHW, Canberra, 2012. [4] Statistics Canada, Leading Causes of Death by Sex (Both Sexes), 2014. [5] Office for National Statistics, Leading Causes of Death in England and Wales, 2009, 2011. [6] National Heart Foundation Australia, Access economics, The Shifting Burden of Cardiovascular Disease in Australia. Sydney (Australia)2005. [7] W. Peter, E. Jonathan, M. Richard, A.G. Shaper, Effect of potentially modifiable risk factors associated with myocardial infarction in 52 countries (the INTERHEART study): case–control study, Lancet 365 (2005) 117. [8] J.F. Price, P.I. Mowbray, A.J. Lee, A. Rumley, G.D. Lowe, F.G. Fowkes, Relationship between smoking and cardiovascular risk factors in the development of peripheral arterial disease and coronary artery disease: Edinburgh Artery Study, Eur. Heart J. 20 (1999) 344–353. [9] S. Begg, T. Vos, B. Barker, C. Stevenson, L. Stanley, A.D. Lopez, The Burden of Disease and Injury in Australia 2003. PHE 82, AIHW, Canberra, 2007. [10] Australian Institute of Health and Welfare, Rural, regional and remote health — indicators of health, AIHW Cat. No. PHE 59, AIHW, Canberra, 2005. (Rural Health Series no. 5). [11] Australian Institute of Health and Welfare, Cardiovascular medicines and primary health care: a regional analysis, Cardiovascular Disease Series No. 32. Cat. No. 48, AIHW, Canberra, 2010. [12] P.D. Patterson, C.G. Moore, J.C. Probst, J.A. Shinogle, Obesity and physical inactivity in rural America, J. Rural Health 20 (2004) 151–159. [13] M.S. Eberhardt, E.R. Pamuk, The importance of place of residence: examining health in rural and nonrural areas, Am. J. Public Health 94 (2004) 1682–1686. [14] VicGov, Victorian Population Health Survey 2008, Department of Health, Melbourne, Victoria, 2008. [15] Australian Institute of Health and Welfare, Cardiovascular disease: Australian facts 2011, Cardiovascular Disease Series. Cat. No. CVD 53, AIHW, Canberra, 2011. [16] P.T.A. Tideman, E. Janus, B. Philpot, R. Clark, E. Peach, T. Laatikainen, E. Vartiainen, R. Tirimacco, A. Montgomerie, J. Grant, V. Versace, J.A. Dunbar, A comparison of Australian rural and metropolitan cardiovascular risk and mortality: the Greater Green Triangle and North West Adelaide population surveys, BMJ 3 (8) (23 2013) e003203. [17] N.P.B. Davis-Lameloise, E.D. Janus, V.L. Versace, T. Laatikainen, E.A. Vartiainen, J.A. Dunbar, Occupational differences, cardiovascular risk factors and lifestyle habits in South Eastern rural Australia, BMC Public Health 13 (2013) 1090. [18] S.W.A. Jordan, A. Dobson, Management of heart conditions in older rural and urban Australian women, Int. Med. J. 41 (10) (2011) 722–729. [19] National Heart Foundation of Australia and the Cardiac Society of Australia and New Zealand, Reducing Risk in Heart Disease 2007, 2008. [20] WHO, Obesity and Overweight, 2011. [21] Australian Bureau of Statistics, Census paper 03/01, ASGC Remoteness Classification: Purpose and Use2003. [22] American Diabetes Association, Diagnosis and classification of diabetes mellitus, Diabetes Care 27 (2004) S5–S10. [23] Australian Bureau of Statistics, 3101.0 — Australian Demographic Statistics, Sep 2013, 2014. [24] Australian Bureau of Statistics, 4102.0 — Australian Social Trends, 2008. [25] Australian Institute of Health and Welfare, Australia's health 2010, Australia's Health Series No. 12. Cat. No. AUS 122, AIHW, Canberra, 2010. [26] E.L.M. Barr, P.Z. Zimmet, T.A. Welborn, D. Jolley, D.J. Magliano, D.W. Dunstan, et al., Risk of cardiovascular and all-cause mortality in individuals with diabetes mellitus, impaired fasting glucose, and impaired glucose tolerance: the Australian Diabetes, Obesity, and Lifestyle Study (AusDiab), Circulation 116 (2007) 151–157. [27] T.S. Church, A.M. Thompson, P.T. Katzmarzyk, X. Sui, N. Johannsen, C.P. Earnest, et al., Metabolic syndrome and diabetes, alone and in combination, as predictors of cardiovascular disease mortality among men, Diabetes Care 32 (2009) 1289–1294. [28] S.M. Haffner, S. Lehto, T. Ronnemaa, K. Pyorala, M. Laakso, Mortality from coronary heart disease in subjects with type 2 diabetes and in nondiabetic subjects with and without prior myocardial infarction, N. Engl. J. Med. 339 (1998) 229–234.
Contents lists available at ScienceDirect
International Journal of Cardiology journal homepage: www.elsevier.com/locate/ijcard
Diabetes and poor glycaemic control in rural patients with coronary artery disease☆ Catherine Ngan a,⁎,1, Ya-Chu May Tsai a,1, Dharshan Palasubramaniam a,1, Amy Wilson-O'Brien a,1, Jamie Layland a,1, Robert Whitbourn b,1, Andrew Wilson a,1 a b
The University of Melbourne, Department of Medicine, St Vincent's Hospital, Melbourne, Australia St Vincent's Hospital, Melbourne, Australia
a r t i c l e
i n f o
Article history: Received 3 August 2014 Received in revised form 18 November 2014 Accepted 22 November 2014 Available online 26 November 2014 Keywords: Cardiovascular risk factors Diabetes Rural medicine Epidemiology
a b s t r a c t Background: The burden of cardiovascular disease is higher in rural populations. Existing data on rural cardiovascular health is mainly based on community surveys. Regional differences are not well addressed. This study aims to identify regional inequalities in cardiovascular risk factors (CVRFs) in Australian patients with suspected coronary artery disease. Methods and results: 538 subjects (72% male; mean age 63 years) were recruited from a single cardiac catheter laboratory over a 24-month period. Subjects were stratified into Remoteness Areas (RAs) according to the Australian Standard Geographical Classification (RA1 corresponds to Major Cities, RA2 to Inner Regional Areas, RA3 to Outer Regional Areas). Body-mass index, blood pressure, hypertension, dyslipidaemia, diabetes and smoking history were recorded. A blood sample taken before the angiogram was analysed for lipids and fasting blood glucose (FBG). Distribution of the study population across RA1, RA2 and RA3 was 34.8%, 46.1% and 19.1%. Only FBG (p = 0.019) and diagnosed diabetes (p = 0.009) were significantly different i.e. higher in RA1. Of those without known diabetes, RA3 had the highest prevalence of dysglycaemia (p = 0.023) with two-thirds having either pre-diabetes or undiagnosed diabetes. Logistic regression showed that age and RA3 were the only statistically significant predictors of elevated FBG. Conclusion: CAD patients from remote Australia had higher rates of pre-diabetes, undiagnosed diabetes and poorer glycaemic control. Analysis of the main CVRFs revealed a regional inequality in the recognition and management of diabetes alone. Attention to this gap in rural and urban healthcare is crucial to future cardiovascular health outcomes in Australia. © 2014 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Cardiovascular disease (CVD) is the leading cause of noncommunicable deaths in the world, accounting for almost 40% of deaths under 70 [1]. With population ageing being a significant global trend, particularly in developed countries, CVD mortality and morbidity are projected to rise substantially. In developed countries such as the United States, Canada, UK and Australia, CVD is the first or second leading cause of mortality, with coronary artery disease being the chief specific cause of death in these countries [2–5]. In addition to its impact on national mortality and morbidity, CVD also imposes significant financial and societal cost from the magnitude of healthcare expenditure, loss of ☆ Funding sources: University of Melbourne Faculty of Medicine, Dentistry and Health Sciences Equipment Grant (2010). ⁎ Corresponding author at: The University of Melbourne, Department of Medicine, St Vincent's Hospital, Melbourne, Australia. E-mail address: [email protected] (C. Ngan). 1 This author takes responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation.
http://dx.doi.org/10.1016/j.ijcard.2014.11.120 0167-5273/© 2014 Elsevier Ireland Ltd. All rights reserved.
productivity due to work absenteeism, early retirement and premature mortality [6]. It follows that managing the burden of CVD is a health priority in the modern society. It is well accepted that the progression of ischemic heart disease is linked to a history of modifiable and inherited characteristics, and thus addressing modifiable cardiovascular risk factors (CVRFs) are mainstays for management of general cardiovascular health [7,8]. Knowledge of the geographic prevalence of these risk factors is essential to the formation of population specific health policies and targeted delivery of health services. In particular, the countries mentioned above face the challenge of addressing the health concerns of rural and urban communities; 20–30% of US, Canadian and Australian populations live outside of Major Cities, where there are unique geographical, lifestyle and health characteristics. The impact of CVD is particularly high in these rural populations, where in Australia the disease burden can be up to 15% greater [9]. Rates of cardiovascular hospitalisations and deaths increase with remoteness, significantly contributing to the overall poorer health outcomes in regional and remote communities [10–13]. There is also a
C. Ngan et al. / International Journal of Cardiology 180 (2015) 264–269
greater prevalence of modifiable CVRFs in populations outside of the Major Cities, with higher rates of hypertension, obesity, smoking and risky alcohol consumption [11,13]. Since much of the burden of CVD can be decreased by managing CVRFs, greater knowledge of the specific differences in risk factor awareness and management across rural populations is vital to improving health outcomes. Awareness of regional health has been a focus of Australian health policy for many years, and the existing literature on this topic is based on comprehensive risk factor surveys of regional and national health [11,14,15]. Local studies on CVRFs have recruited from electoral rolls or used Census data, where rurality has not necessarily correlated with cardiovascular mortality or morbidity [16,17], or the findings have been gender specific [18]. However, the usefulness of this data is limited by the lack of blood tests in determining the presence of a risk factor, relying instead on patient recall and potentially reflecting patient awareness rather than true prevalence. Conditions that rely on diagnostic blood tests, such as diabetes, are thus likely under-reported. Another feature of these population surveys is that there is little focus on groups most at risk. In contrast, this study examines the regional prevalence and management of CVRFs in patients with suspected coronary artery disease, in a well-characterised Australian population. The investigation of reported and biochemical risk factors along with angiographic evidence offers a fresh insight into cardiovascular healthcare in a group of patients for whom risk factor management is a priority. 2. Methods 2.1. Ethics approval This study was approved by the St Vincent's Hospital Research Governance Unit, in accordance with National Health and Medical Research Council guidelines. All patients were invited to participate before their procedure and written consent was obtained. 2.2. Study population Study subjects were recruited from St Vincent's Hospital Melbourne, which is a major tertiary public teaching hospital. Patients presenting to the St Vincent's Cardiac Catheterisation Laboratory between May 2009 and May 2011 for coronary angiography and/or percutaneous coronary intervention (PCI) were eligible for this study. Subjects were recruited as part of the Biomarkers of Atherosclerosis, Vascular and Endothelial Dysfunction in Heart Disease Study (BRAVEHEART), and thus the following exclusion criteria were applied: patients with chronic or acute infections, systemic inflammatory conditions, recent or untreated malignancies and serum creatinine levels greater than 160 μmol/l. 2.3. Patient data Cardiovascular histories were obtained through interview and review of medical records. The use of hypoglycaemic, lipid lowering and antihypertensive medication was recorded. Blood pressure (BP) on admission and body-mass index (BMI) were recorded. BMI was calculated as body weight in kilograms divided by the square of the height in metres. Hypertension was defined according to the 2007 Cardiac Society of Australia and New Zealand (CSANZ) Guidelines as BP N140/90 mm Hg [19]. Obesity was defined according to the World Health Organisation (WHO) Guidelines as BMI ≥ 30 kg/m2 [20]. Severity of CAD was angiographically assessed by a cardiologist. Vessel disease was defined as a greater than 50% stenosis of a major coronary artery or history of stenting of that artery. The location of disease and the number of diseased vessels were recorded. 2.4. The ASGC-RA classification This study uses the Australian Standard Geographical Classification-Remoteness Areas (ASGC-RA) system, which was devised by the Australian Bureau of Statistics to organise data into broad geographical categories based on population size and distance from an urban centre [21]. The Remoteness Area (RA) boundaries are updated after each Census, and this report is based on the 2006 Census, which at the time was the most recent. To classify the study population, the postcode of the subject's current address was obtained from the hospital registry. All subjects were classified into three RAs: Major Cities, Inner Regional Areas and Outer Regional Areas. 2.5. Biochemical blood analysis Arterial blood samples were collected via femoral or radial arterial access at the beginning of angiography before the use of iodinated contrast agent. Samples were analysed through St Vincent's Pathology for lipid profile, fasting blood glucose (FBG) and glycated haemoglobin (HbA1c). Normal FBG and HbA1c levels were defined according to the
265
2004 American Diabetes Association (ADA) Guidelines [22]: normal FBG b5.6 mmol/l, impaired FBG 5.6 to 6.9 mmol/l, diabetic FBG ≥7 mmol/l and normal HbA1c b6.5%. Normal lipid levels were defined according to the 2007 CSANZ Guidelines [19]: total cholesterol (TC) b4 mmol/l, low-density lipoprotein cholesterol (LDL-C) b2.0 mmol/l, high-density lipoprotein cholesterol (HDL-C) N1.0 mmol/l and triglycerides (TG) b1.5 mmol/l.
2.6. Statistical analysis Statistical analyses were undertaken using SPSS 17 statistical software. Continuous variables were examined for normality of distribution with the Kolmogorov–Smirnov test. Variables not normally distributed were appropriately log-transformed for parametric analyses. Differences in physical measurements, reported risk factors, biochemical risk factors and severity of CAD across the RA groups were determined by one-way analysis of variance (ANOVA) with post-hoc Tukey analysis for continuous variables, and chi-square analysis for discrete variables. Logistic regression analysis was used to assess independence of correlations, and to identify the variables most significantly associated with elevated FBG. The variables included in the model were age, gender, hypertension, dyslipidaemia, smoking history, BMI, use of hypoglycaemic medication, the presence of CAD and RA. Significant values were defined as p b 0.05.
3. Results 3.1. Characteristics at baseline 538 subjects (72% male; mean age 63 years) were included in the analyses. All subjects were Victorian residents and were classified by their postcode into Major Cities, Inner Regional Areas and Outer Regional Areas as defined by the ASGC-RA system [21]. There were 187 subjects from Major Cities (75% male, mean age 61 years), 248 from Inner Regional Areas (70% male, mean age 63 years) and 103 from Outer Regional Areas (72% male, mean age 64 years) (Table 1, Fig. 2).
Table 1 Baseline characteristics. Total number in population Average age (years) Gender Remoteness Area (number of patients) Major Cities Inner Regional Outer Regional Reported CVRF (%) Hypertension Dyslipidaemia History of smoking Diabetes mellitus Anthropometric measurements BMI (kg/m2) Systolic BP (mm Hg) Diastolic BP (mm Hg) Biochemical risk factors FBG (mmol/l) HbA1c (%) TC (mmol/l) TG (mmol/l) HDL-C (mmol/l) LDL-C (mmol/l) Medication Reported diabetes (n = 146) Hypoglycaemic medication Reported hypertension (n = 434) Antihypertensives Reported dyslipidaemia (n = 465) Lipid lowering medication Angiography Coronary artery disease 1 vessel disease 2 vessel disease 3 vessel disease
538 63 (0.5) 72% male 187 248 103 81% 87% 71% 27% 29.6 (0.2) 135 (1) 82 (1) 6.4 (0.1) 6.1 (0.1) 4.3 (0.0) 1.6 (0.0) 1.1 (0.0) 2.5 (0.0)
67% 81% 81% 70% 27% 23% 21%
Data expressed as mean (standard error of mean) where appropriate. CVRF represents cardiovascular risk factor; BP, blood pressure; BMI, body-mass index; FBG, fasting blood glucose; HbA1c, glycated haemoglobin; TC, total cholesterol; TG, triglycerides; HDL-C, high-density lipoprotein cholesterol; and LDL-C, low-density lipoprotein cholesterol.
266
C. Ngan et al. / International Journal of Cardiology 180 (2015) 264–269
Fig. 1. Remoteness Areas of Victoria. Image adapted from [21]
3.2. Characteristics by Remoteness Area Risk factors, BMI and BP were compared between RA. Chi-square analysis of these risk factors revealed a significant difference between the RAs only for reported diabetes mellitus (p = 0.009). Reported diabetes was most prevalent in Major Cities and least prevalent in Inner Regional Areas. BMI and BP were similar across the RAs, as confirmed by ANOVA (Table 2). Average lipid, FBG and HbA1c results were compared between RA. ANOVA tests revealed significant differences between RA only for FBG
(p = 0.019) and HbA1c (p = 0.002). The average FBG in each RA as well as the overall population were above the cut-points for normal levels (FBG b 5.6 mmol/l) [22]. Both FBG and HbA1c levels were highest in subjects from Major Cities, with an average FBG of 6.7 mmol/l, and HbA1c of 6.4% (Table 3). 3.3. Use of medication The use of hypoglycaemic, lipid lowering and antihypertensive medication was analysed in subjects who reported having the relevant risk factor. Analysis between the regions revealed no significant differences, however, the use of hypoglycaemic medication followed a downward trend with increasing rurality (Table 4). Table 2 Distribution of reported risk factors and anthropometric measurements by Remoteness Area.
Reported CVRF (%) Hypertension Dyslipidaemia History of smoking Diabetes mellitus Anthropometric measurements BMI (kg/m2) Systolic BP (mm Hg) Diastolic BP (mm Hg) Fig. 2. Distribution of study population by Remoteness Area.
Major City
Inner Regional
Outer Regional
p-Value
79% 85.0% 71.0% 34.9%
84% 87.9% 70.9% 21.9%
77% 86.3% 73.0% 25.5%
0.204 0.680 0.920 0.009⁎
29.3 (0.4) 134 (2) 81 (1)
29.6 (0.4) 135 (2) 82 (1)
29.7 (0.6) 137 (3) 81 (1)
0.838 0.391 0.677
Data expressed as mean (standard error of mean) where appropriate. ⁎ p b 0.05.
C. Ngan et al. / International Journal of Cardiology 180 (2015) 264–269 Table 3 Biochemical risk factors by Remoteness Area.
FBG (mmol/l) HbA1c (%) TC (mmol/l) TG (mmol/l) HDL-C (mmol/l) LDL-C (mmol/l)
267
Table 5 Stages of dysglycaemia by Remoteness Area.
Major City
Inner Regional
Outer Regional
p-Value
6.7 (0.2) 6.4 (0.1) 4.3 (0.1) 1.6 (0.1) 1.1 (0.0) 2.5 (0.1)
6.2 (0.1) 6.0 (0.1) 4.3 (0.1) 1.6 (0.1) 1.1 (0.0) 2.5 (0.1)
6.4 (0.2) 6.0 (0.1) 4.2 (0.1) 1.6 (0.1) 1.1 (0.0) 2.5 (0.1)
0.019⁎ 0.003⁎ 0.699 0.779 0.169 0.953
Data expressed as mean (standard error of mean). FBG represents fasting blood glucose; HbA1c, glycated haemoglobin; TC, total cholesterol; TG, triglycerides; HDL-C, high-density lipoprotein cholesterol; and LDL-C, low-density lipoprotein cholesterol. ⁎ p b 0.05.
History of diabetes (N = 146) Normal FBG: b5.6 mmol/l Impaired FBG: 5.6–6.9 mmol/l Diabetic FBG: 7.0+ mmol/l No history of diabetes (N = 392) Normal FBG: b5.6 mmol/l Impaired FBG: 5.6–6.9 mmol/l Diabetic FBG: 7.0+ mmol/l Mean HbA1c (%) History of diabetes No history of diabetes
Major City
Inner Regional
Outer Regional
p-Value
8% 18% 75%
13% 23% 59%
12% 36% 52%
0.745 0.129 0.093
52% 43% 5%
52% 44% 4%
33% 59% 8%
0.023⁎ 0.052 0.691
7.6 (0.2) 5.7 (0.1)
7.4 (0.2) 5.6 (0.0)
7.1 (0.3) 5.7 (0.1)
0.295 0.366
Data expressed as mean (standard error of mean). FBG represents fasting blood glucose; HbA1c, glycated haemoglobin. ⁎ p b 0.05.
3.4. Dysglycaemia The population was divided into those with and without a past diagnosis of diabetes to account for subjects whose glycaemic status may be controlled by lifestyle or medical therapy. The population of each RA was further divided into stages of dysglycaemia as defined by the ADA [22] (Table 5). Among those with a diagnosis of diabetes, subjects from Major Cities had the poorest glycaemic control, with 75% having an FBG ≥ 7.0 mmol/l. Conversely, diabetic subjects from Outer Regional Areas had better glycaemic control, with 52% having elevated FBGs (p = 0.093) (Fig. 3). Among those without a diagnosis of diabetes, two-thirds of subjects from Outer Regional Areas had fasting sugars in the impaired and diabetic range so ≥5.6 mmol/l (p = 0.023). Of this group, 59% had impaired fasting sugars and 8% had an FBG ≥7.0 mmol/l (Fig. 4). Average HbA1c was also examined, where normal levels were defined as b6.5% according to ADA guidelines [22]. Subjects with a diagnosis of diabetes had elevated average HbA1c, with the highest average in Major Cities and the lowest in Outer Regional Areas. ANOVA tests showed that these differences were not statistically significant. Logistic regression analysis was performed to assess the independence of variables associated with elevated FBG, defined as N5.6 mmol/l. The variables included in the model were age, gender, hypertension, dyslipidaemia, smoking history, BMI, use of hypoglycaemic medication, the presence of CAD and RA. Of these factors, only age, medication and RA made a unique statistically significant contribution to the model (p b 0.05). After adjusting for all the above factors, the odds of having elevated FBG in Outer Regional Areas compared to Major Cities was 2.3 (95% CI 1.2–4.4) (Table 6).
3.5. The severity of coronary artery disease The presence of coronary artery disease and the number of diseased coronary vessels were compared between each RA. Chi-square analysis showed no significant differences in the prevalence of disease or the number of diseased vessels (Table 7).
4. Discussion This Australian study differs from contemporary risk factor surveys because of its focus on patients for whom CVRF management is a priority. Recruitment from a catheterisation laboratory provided this study with individual biochemical blood analysis and angiographic evidence of coronary artery disease, which enabled all statistical analysis and observations to be accurately corrected for severity of CVRFs and coronary disease. The study population was recruited from the state of Victoria, where 25% of the Australian population resides [23]. Of the Victorian population, 75% live in Major Cities, 20% live in Inner Regional Areas and 5% live in Outer Regional and Remote Areas [24] (Fig. 1). Although subjects were recruited from a tertiary metropolitan hospital, all three major RAs were well represented in the study population: 35% were from Major Cities, 46% from Inner Regional Areas and 19% from Outer Regional Areas. The key finding from our regional analysis was that after correcting for age, gender, other risk factors and severity of CAD, there were clear regional differences in the prevalence and management of diabetes alone. Baseline analysis of this study population revealed a remarkable similarity in risk factor profiles, which is in contrast to state and national surveys reporting a higher prevalence of CVD and CVRFs in regional populations [10,25]. Our investigation of high-risk patients revealed similarities in reported risk factors, blood lipid profiles and severity of coronary disease between the three RAs. The use of antihypertensive and lipid lowering medication was also similarly consistent, and this was reflected in the relatively well-controlled BP and lipid profile of each RA. However, one risk factor showed significant regional variation: the prevalence, awareness and management of hyperglycaemia. Hyperglycaemia is characteristic of diabetes mellitus, where the secretion or action of insulin is chronically impaired and the tissue
Table 4 Use of medication for reported risk factors by Remoteness Area.
Reported diabetes (n = 146) Hypoglycaemic medication Reported hypertension (n = 434) Antihypertensives Reported dyslipidaemia (n = 465) Lipid lowering medication
Major City
Inner Regional
Outer Regional
p-Value
73%
66%
54%
0.234
79%
81%
82%
0.814
80%
87%
85%
0.171
Fig. 3. Stages of dysglycaemia by Remoteness Area in patients with a diagnosis of diabetes.
268
C. Ngan et al. / International Journal of Cardiology 180 (2015) 264–269 Table 7 Severity of coronary artery disease.
Coronary artery disease Number of diseased vessels
Fig. 4. Stages of dysglycaemia by Remoteness Area in patients without a diagnosis of diabetes.
uptake of blood glucose is thus compromised [22]. The term hyperglycaemia also incorporates an intermediate stage of ‘pre-diabetes’, where FBG is above normal levels but still below diabetic cut-points [22]. Both diabetes and pre-diabetes are major risk factors of CVD development and premature mortality [26,27], and are independent predictors of CVD mortality [26]. Furthermore, diabetes is an equivalent risk factor for future myocardial infarction as having a history of a previous infarct [28]. Of relevance to this study population, diabetes is also a risk factor for restenosis after angioplasty or stenting [2]. Such evidence provides an important rationale for aggressive management and monitoring of patients with high blood glucose levels. Yet in this population of subjects with suspected coronary artery disease, there was significant under-recognition and suboptimal management of diabetes. Independent of coronary disease severity, the prevalence of diabetes was highest in subjects from Major Cities at 35%, followed by Outer Regional Areas at 26% and Inner Regional Areas at 22% (p = 0.009). Fasting sugars were also highest in Major Cities at 6.7 mmol/l, followed by Outer Regional Areas at 6.4 mmol/l and then Inner Regional Areas at 6.2 mmol/l (p = 0.019). HbA1c levels were similarly elevated in Major Cities at 6.4% (p = 0.003). The use of hypoglycaemic medication was again highest in Major Cities at 25%. Thus the first impression of this study cohort is that Major Cities have more diabetes, and accordingly have higher fasting sugars, higher HbA1c levels and greater use of hypoglycaemic medication. However, these results do not account for the influence of lifestyle or medical therapy on FBGs. To gain a more accurate representation of
Table 6 Logistic regression applied to entire cohort. B
RA 1 RA 2 RA 3 Age Gender Smoker Dyslipidaemia Hypertension BMI Use of hypoglycaemics CAD Constant
S.E.
Wald
df
Sig.
2 1 1 1 1 1 1 1 1 1
.009⁎ .775 .011⁎ b.001⁎ .041⁎ .974 .900 .212 .096 b.001⁎
1 1
.498 .002
−.07 .84 .04 .52 .01 −.04 −.36 .04 3.46
.25 .33 .01 .25 .25 .33 .29 .02 .73
9.51 .08 6.48 13.47 4.17 .00 .02 1.56 2.77 22.26
−.18 −3.24
.26 1.05
.46 9.48
Exp(B)
95% C.I. for EXP(B) Lower
Upper
.93 2.31 1.04 1.67 1.01 .96 .70 1.04 31.89
.57 1.21 1.02 1.02 .62 .51 .40 .99 7.57
1.52 4.39 1.06 2.74 1.65 1.81 1.23 1.08 134.37
.84 .04
.51
1.39
RA represents Remoteness Area; BMI, body-mass index; and CAD, coronary artery disease. ⁎ p b 0.05.
1 2 3
Major City
Inner Regional
Outer Regional
p-Value
66% 24% 20% 23%
73% 28% 25% 19%
72% 28% 22% 23%
0.347 0.561 0.378 0.619
glycaemic control, the study population was divided into those with a history of diabetes and those without, and these two groups were divided into normal, impaired and diabetic FBG. This subgroup analysis was revealing in the population without a diagnosis of diabetes; twothirds of apparently non-diabetic patients from Outer Regional Areas had elevated fasting glucose levels (p = 0.023), with the highest prevalence of pre-diabetes at 59%, and also the highest prevalence of undiagnosed diabetes at 8%. Once again, these results were independent of disease severity. This is a significant and worrying regional discrepancy, which points to poorer disease recognition in areas furthest from city centres. To assess the independence of variables associated with an elevated FBG, logistic regression analysis was performed. After correcting for potential confounding factors including age, gender, hypertension, dyslipidaemia, smoking history, BMI and ischemic heart disease as determined by angiography, the odds of having elevated fasting sugars in Outer Regional Areas compared to Major Cities was 2.3. In other words, subjects furthest from city centres had more than double the risk of having diabetes or pre-diabetes, independent of coronary disease severity. However, not all was worse in subjects from Outer Regional Areas. Among those with known diabetes, subjects from Major Cities had the poorest glycaemic control, with three-quarters having an FBG ≥ 7.0 mmol/l. Conversely, diabetic subjects from Outer Regional Areas had better glycaemic control, with about half having normal fasting sugars (p = 0.093). It would appear that although subjects from Outer Regional Areas were more likely to have high FBGs, once diagnosed, these subjects had better glycaemic management. The use of hypoglycaemic medication was also analysed among those with a diagnosis of diabetes. 73% of subjects from Major Cities were on hypoglycaemic agents, compared to 54% in Outer Regional Areas. This difference, however, was not statistically significant. From these observations on reported risk factors, blood tests and use of medication, regional inconsistencies were apparent in the prevalence and management of diabetes alone. The high proportion of unrecognised hyperglycaemia raises concerns about awareness and management of diabetes in both regional and metropolitan areas. This is particularly true for subjects from Outer Regional Victoria, of whom two-thirds of those without a history of diabetes had unrecognised hyperglycaemia. The relevance of this finding extends beyond cardiovascular health; diabetic patients have a high burden of disease, and can go on to develop peripheral vascular disease, cerebrovascular disease, retinopathy, neuropathy, and renal disease, to name but a few complications [4,16–18]. 4.1. Strengths and limitations A major strength of this study was the inclusion of biochemical profiles in a specific population of patients with known or suspected coronary artery disease, of which 70% had significant vessel disease. Whereas most risk factor studies survey broadly to gain a general overview of the population, this study recruited from a pool of patients for whom CVRF management is a priority. Additionally, the recruitment of patients presenting to a catheterisation laboratory provided this study not only with fasted blood samples for detection of dyslipidaemia or hyperglycaemia, but also angiographic evidence of disease, all of which are unique features of the data collection.
C. Ngan et al. / International Journal of Cardiology 180 (2015) 264–269
Investigation of potential factors behind the hyperglycaemia observed in regional populations was beyond the scope of this study, but it is known that access to healthcare services, income, education and employment are all factors that contribute to poorer rural health. However, it seems unlikely that such broad characteristics could selectively influence one risk factor out of all the major CVRFs. Further investigation into prescribing practices, patient compliance and lifestyle management of diabetic patients would provide deeper insight into this regional inequality. There is also the potential for referral bias in this study population, where patients presenting to this referral centre are more aware of their cardiovascular health than the general population. Nonetheless, the study findings remain pertinent — despite known ischemic heart disease and perhaps greater awareness of cardiovascular health, there was still far more dysglycaemia in those without a diagnosis of diabetes in the Outer Regional cohort. 5. Conclusions In conclusion, analysis of reported CVRFs, blood tests and use of medication in a geographically diverse population of subjects with suspected coronary artery disease showed an isolated regional inequality in the prevalence and management of hyperglycaemia alone. Subjects from Major Cities were more likely to be diagnosed with diabetes but have poorer glycaemic control. Conversely, subjects from rural Australia were less likely to be diagnosed, but once diagnosed had better glycaemic control. In terms of clinical practice, this study demonstrates that presentation for a coronary angiogram or PCI is an opportunity for a full physical and biochemical CVRF assessment. Despite patient awareness of heart disease, diabetes is particularly under-managed across urban and rural regions, and all patients would benefit from education about cardiovascular health and glycaemic control in the setting of an angiogram. Although detection of diabetes is lower in patients from rural areas, once diagnosed, there appears to be better glycaemic control, and it is in these patients that opportunistic screening and education may be of most benefit. Conflict of interest None. Acknowledgements The authors thank the Department of Medicine at St Vincent's Hospital Melbourne, and the staff at the St Vincent's catheter laboratory for their assistance and support. References [1] A. Alwan, Global Status Report on Noncommunicable Diseases 2010, World Health Organisation, Italy, 2011.
269
[2] A.S. Go, D. Mozaffarian, V.L. Roger, E.J. Benjamin, J.D. Berry, M.J. Blaha, et al., Heart disease and stroke statistics — 2014 update: a report from the American Heart Association, Circulation 129 (2014) e28–e292. [3] Australian Institute of Health and Welfare, Australia's Health 2012. Australia's Health Series No. 13. Cat. No. AUS 156, AIHW, Canberra, 2012. [4] Statistics Canada, Leading Causes of Death by Sex (Both Sexes), 2014. [5] Office for National Statistics, Leading Causes of Death in England and Wales, 2009, 2011. [6] National Heart Foundation Australia, Access economics, The Shifting Burden of Cardiovascular Disease in Australia. Sydney (Australia)2005. [7] W. Peter, E. Jonathan, M. Richard, A.G. Shaper, Effect of potentially modifiable risk factors associated with myocardial infarction in 52 countries (the INTERHEART study): case–control study, Lancet 365 (2005) 117. [8] J.F. Price, P.I. Mowbray, A.J. Lee, A. Rumley, G.D. Lowe, F.G. Fowkes, Relationship between smoking and cardiovascular risk factors in the development of peripheral arterial disease and coronary artery disease: Edinburgh Artery Study, Eur. Heart J. 20 (1999) 344–353. [9] S. Begg, T. Vos, B. Barker, C. Stevenson, L. Stanley, A.D. Lopez, The Burden of Disease and Injury in Australia 2003. PHE 82, AIHW, Canberra, 2007. [10] Australian Institute of Health and Welfare, Rural, regional and remote health — indicators of health, AIHW Cat. No. PHE 59, AIHW, Canberra, 2005. (Rural Health Series no. 5). [11] Australian Institute of Health and Welfare, Cardiovascular medicines and primary health care: a regional analysis, Cardiovascular Disease Series No. 32. Cat. No. 48, AIHW, Canberra, 2010. [12] P.D. Patterson, C.G. Moore, J.C. Probst, J.A. Shinogle, Obesity and physical inactivity in rural America, J. Rural Health 20 (2004) 151–159. [13] M.S. Eberhardt, E.R. Pamuk, The importance of place of residence: examining health in rural and nonrural areas, Am. J. Public Health 94 (2004) 1682–1686. [14] VicGov, Victorian Population Health Survey 2008, Department of Health, Melbourne, Victoria, 2008. [15] Australian Institute of Health and Welfare, Cardiovascular disease: Australian facts 2011, Cardiovascular Disease Series. Cat. No. CVD 53, AIHW, Canberra, 2011. [16] P.T.A. Tideman, E. Janus, B. Philpot, R. Clark, E. Peach, T. Laatikainen, E. Vartiainen, R. Tirimacco, A. Montgomerie, J. Grant, V. Versace, J.A. Dunbar, A comparison of Australian rural and metropolitan cardiovascular risk and mortality: the Greater Green Triangle and North West Adelaide population surveys, BMJ 3 (8) (23 2013) e003203. [17] N.P.B. Davis-Lameloise, E.D. Janus, V.L. Versace, T. Laatikainen, E.A. Vartiainen, J.A. Dunbar, Occupational differences, cardiovascular risk factors and lifestyle habits in South Eastern rural Australia, BMC Public Health 13 (2013) 1090. [18] S.W.A. Jordan, A. Dobson, Management of heart conditions in older rural and urban Australian women, Int. Med. J. 41 (10) (2011) 722–729. [19] National Heart Foundation of Australia and the Cardiac Society of Australia and New Zealand, Reducing Risk in Heart Disease 2007, 2008. [20] WHO, Obesity and Overweight, 2011. [21] Australian Bureau of Statistics, Census paper 03/01, ASGC Remoteness Classification: Purpose and Use2003. [22] American Diabetes Association, Diagnosis and classification of diabetes mellitus, Diabetes Care 27 (2004) S5–S10. [23] Australian Bureau of Statistics, 3101.0 — Australian Demographic Statistics, Sep 2013, 2014. [24] Australian Bureau of Statistics, 4102.0 — Australian Social Trends, 2008. [25] Australian Institute of Health and Welfare, Australia's health 2010, Australia's Health Series No. 12. Cat. No. AUS 122, AIHW, Canberra, 2010. [26] E.L.M. Barr, P.Z. Zimmet, T.A. Welborn, D. Jolley, D.J. Magliano, D.W. Dunstan, et al., Risk of cardiovascular and all-cause mortality in individuals with diabetes mellitus, impaired fasting glucose, and impaired glucose tolerance: the Australian Diabetes, Obesity, and Lifestyle Study (AusDiab), Circulation 116 (2007) 151–157. [27] T.S. Church, A.M. Thompson, P.T. Katzmarzyk, X. Sui, N. Johannsen, C.P. Earnest, et al., Metabolic syndrome and diabetes, alone and in combination, as predictors of cardiovascular disease mortality among men, Diabetes Care 32 (2009) 1289–1294. [28] S.M. Haffner, S. Lehto, T. Ronnemaa, K. Pyorala, M. Laakso, Mortality from coronary heart disease in subjects with type 2 diabetes and in nondiabetic subjects with and without prior myocardial infarction, N. Engl. J. Med. 339 (1998) 229–234.
