Risk factor control in patients with Type 2 diabetes and coronary heart disease: findings from the Swedish National Diabetes Register (NDR)

Authors

  • S. Gudbjörnsdottir,

    1. Department of Medicine, Sahlgrenska University Hospital, Göteborg University, Göteborg, *Department of Public Health and Caring Sciences, Family Medicine and Clinical Epidemiology and †Department of Public Health and Caring Sciences, Geriatrics, Uppsala University, Uppsala, and ‡Department of Clinical Sciences, Lund University, University Hospital, Malmö, Sweden
    Search for more papers by this author
  • K. Eeg-Olofsson,

    1. Department of Medicine, Sahlgrenska University Hospital, Göteborg University, Göteborg, *Department of Public Health and Caring Sciences, Family Medicine and Clinical Epidemiology and †Department of Public Health and Caring Sciences, Geriatrics, Uppsala University, Uppsala, and ‡Department of Clinical Sciences, Lund University, University Hospital, Malmö, Sweden
    Search for more papers by this author
  • J. Cederholm,

    1. Department of Medicine, Sahlgrenska University Hospital, Göteborg University, Göteborg, *Department of Public Health and Caring Sciences, Family Medicine and Clinical Epidemiology and †Department of Public Health and Caring Sciences, Geriatrics, Uppsala University, Uppsala, and ‡Department of Clinical Sciences, Lund University, University Hospital, Malmö, Sweden
    Search for more papers by this author
  • * B. Zethelius,

    1. Department of Medicine, Sahlgrenska University Hospital, Göteborg University, Göteborg, *Department of Public Health and Caring Sciences, Family Medicine and Clinical Epidemiology and †Department of Public Health and Caring Sciences, Geriatrics, Uppsala University, Uppsala, and ‡Department of Clinical Sciences, Lund University, University Hospital, Malmö, Sweden
    Search for more papers by this author
  • B. Eliasson,

    1. Department of Medicine, Sahlgrenska University Hospital, Göteborg University, Göteborg, *Department of Public Health and Caring Sciences, Family Medicine and Clinical Epidemiology and †Department of Public Health and Caring Sciences, Geriatrics, Uppsala University, Uppsala, and ‡Department of Clinical Sciences, Lund University, University Hospital, Malmö, Sweden
    Search for more papers by this author
  • P. M. Nilsson,

    1. Department of Medicine, Sahlgrenska University Hospital, Göteborg University, Göteborg, *Department of Public Health and Caring Sciences, Family Medicine and Clinical Epidemiology and †Department of Public Health and Caring Sciences, Geriatrics, Uppsala University, Uppsala, and ‡Department of Clinical Sciences, Lund University, University Hospital, Malmö, Sweden
    Search for more papers by this author
  • on behalf of the Swedish National Diabetes Register (NDR)

    1. Department of Medicine, Sahlgrenska University Hospital, Göteborg University, Göteborg, *Department of Public Health and Caring Sciences, Family Medicine and Clinical Epidemiology and †Department of Public Health and Caring Sciences, Geriatrics, Uppsala University, Uppsala, and ‡Department of Clinical Sciences, Lund University, University Hospital, Malmö, Sweden
    Search for more papers by this author

: Peter M. Nilsson, MD, PhD, Department of Clinical Sciences, Lund University, University Hospital, S-205 02 Malmö, Sweden. E-mail: Peter.Nilsson@med.lu.se

Abstract

Aims  Patients with Type 2 diabetes and coronary heart disease (CHD) are infrequently treated to risk factor targets in current guidelines. We aimed to examine risk factor management and control levels in a large sample of patients with Type 2 diabetes with CHD.

Methods  This was an observational study of 1612 patients with first incidence of CHD before 2002, and of 4570 patients with first incidence of CHD before 2005, from the Swedish National Diabetes Register (NDR).

Results  In patients with CHD 1–2 years before follow-up, the achievement of cardiovascular risk factor targets (follow-up 2002/follow-up 2005) was: HbA1c < 7%, 47%/54% (P < 0.01); blood pressure ≤ 130/80 mmHg, 31%/40% (P < 0.001); total cholesterol < 4.5 mmol/l, 47%/60% (P < 0.001); and low-density lipoprotein-cholesterol < 2.5 mmol/l, 49%/65% (P < 0.001). Use of medication: antihypertensives, 90%/94% (P < 0.01); lipid-lowering drugs, 75%/86% (P < 0.001); and aspirin, 85%/89% (P < 0.05). A high prevalence of adverse lifestyle characteristics prevailed (2002/2005): overweight [body mass index (BMI) ≥ 25 kg/m2], 86%/85%; obesity (BMI ≥ 30 kg/m2), 41%/42%; smokers in age group < 65 years, 16–23%/18–19%; as well as waist circumference ≥ 102 cm (men) or ≥ 88 cm (women), 68% in 2005.

Conclusions  Patients with a combination of Type 2 diabetes and CHD showed an increased use of lipid-lowering drugs over time, corresponding to improving blood lipid levels. A discrepancy existed between the prevalent use of antihypertensive drugs and the low proportion reaching blood pressure targets. Regretfully, a high prevalence of adverse lifestyle characteristics prevailed. Evidence-based therapy with professional lifestyle intervention and drugs seems urgent for improved quality of secondary prevention in these patients.

Abbreviations
BMI

body mass index

BP

blood pressure

CHD

coronary heart disease

DCCT

Diabetes Control and Complications Trial

HDL

high-density lipoprotein

LDL

low-density lipoprotein

MI

myocardial infarction

NDR

National Diabetes Register

PHC

primary healthcare

RIKS-HIA

Register of Information and Knowledge about Swedish Heart Intensive Care Admission

Introduction

The risk of coronary heart disease (CHD) is substantially increased in patients with Type 2 diabetes [1,2]. Furthermore, mortality 1 year after CHD is higher in patients with vs. without diabetes [3]. Glycaemic control can reduce the risk of microvascular end-points [4], and myocardial infarction (MI) if metformin is used [5]. The importance of treatment with antihypertensive, lipid-lowering and antiplatelet drugs for primary prevention of cardiovascular disease has also been documented [6–8]. Recently the benefit of statin use in patients with diabetes was high-lighted in a meta-analysis by the Cholesterol Treatment Trialists’ Collaborators [9].

The Swedish National Diabetes Register (NDR) was initiated in 1996 as a tool for quality assurance in diabetes care, with reports of patient data from hospital out-patient clinics and primary healthcare (PHC) centres nationwide. A more detailed description of the NDR and Swedish diabetes care has been published [10–15]. It offers a unique possibility to survey the treatment and risk factor control in patients with diabetes based on data from everyday clinical practice.

The aim of this study was to examine risk factor management in patients with Type 2 diabetes with CHD, and to compare risk factor levels in patients from the NDR followed up in 2002 and 2005.

Patients and methods

Reporting to the NDR is based on information collected during patient visits at hospital out-patient clinics and PHC centres all over Sweden. This is carried out by trained nurses or physicians via the internet (http://www.ndr.nu), by transferral of data from clinical records databases or by use of a printed form. All information is subsequently stored in a central database, and participating centres receive an annual report with local results compared with national for feedback. Ninety-five percent of all hospital diabetes out-patient clinics and 75% of all PHC centres in Sweden participated in the NDR during 2005. The mean numbers of reported patients were 402 and 144 per participating unit, respectively. Reporting to the NDR is not mandatory, but hospital out-patient clinics and PHC centres nationwide are encouraged to participate. All patients gave informed consent before agreeing to be included. The Regional Ethics Committee at the University of Gothenburg approved the study. Reports concerning the national data have been previously published by the Working Group of the NDR, describing trends in risk factor control, also including a more detailed description of the Swedish healthcare system for patients with diabetes [10–15].

The present study concerns Type 2 diabetes patients ages 18–79 years from the NDR with data available for the following analyses: first, 1612 female and male patients with first incidence of CHD in 1995–2001 and followed up in 2002; and second, 4570 female and male patients with first incidence of CHD 1998–2004 and followed up in 2005. Twenty-three percent of all patients in the second sample with CHD 3–7 years previously were also included in the first sample. About 80% of patients in this study were treated at PHC centres. The definition of Type 2 diabetes applied was treatment with diet only, or treatment with oral hypoglycaemic agents or with insulin only, or insulin combined with oral agents, with onset age of diabetes ≥ 40 years.

Registered variables were age, sex, onset age of diabetes, type of hypoglycaemic treatment, HbA1c, body mass index (BMI), waist circumference, physical leisure time activity, blood pressure (BP), blood lipids, smoking, microalbuminuria, and use of antihypertensive drugs, lipid-lowering drugs and aspirin (acetylsalicylic acid). BMI (kg/m2) was calculated as weight/height2. Waist circumference (cm) was measured horizontally at the level of the umbilicus, at the end of expiration when standing. Regular physical leisure time activity was defined as regular exercise three times per week or more. A smoker was defined as a patient regularly smoking one or more cigarettes daily, or who had stopped smoking within the past 3 months. The Swedish standard recommendation for BP recording is the mean value of two readings (Korotkoff 1–5) in the supine position using a cuff of appropriate size, and was further endorsed by the NDR according to national guidelines (National Board on Health and Welfare, Stockholm 1999).

Laboratory analyses were carried out at local laboratories. HbA1c has been quality assured since 1996 in Sweden. All Swedish diabetes out-patient clinics and primary care centres use methods regularly calibrated to the high-performance liquid chromatography Mono-S method. In this study, measured HbA1c values were converted to the Diabetes Control and Complications Trial (DCCT) standard values using the formula: HbA1c (DCCT) = 0.923 × HbA1c (MonoS) + 1.345 (R2 = 0.998) [16]. Low-density lipoprotein (LDL)-cholesterol values were calculated using Friedewald's formula: LDL-cholesterol = total cholesterol – high-density lipoprotein-cholesterol –(0.45 × triglycerides), if triglycerides < 4.0 mmol/l. Microalbuminuria was defined as urine albumin excretion 20–200 µg/min in two out of three consecutive tests.

Current treatment goals for cardiovascular risk factors have been applied, as recommended by European guidelines [17] and the American Diabetes Association [18]: HbA1c < 7.0% (DCCT standard); BMI < 25 kg/m2; waist < 94 cm (men) or < 80 cm (women); BP < 130/80 mmHg; total cholesterol < 4.5 mmol/l, LDL-cholesterol < 2.5 or ≤ 1.8 mmol/l, triglycerides < 1.7 mmol/l, and high-density lipoprotein (HDL)-cholesterol > 1.0 mmol/l (men) or > 1.2 mmol/l (women) [17,18].

Definition of coronary heart disease events

A first incident non-fatal CHD event, occurring between 1995 and 2001 in patients followed up in 2002 and occurring between 1998 and 2004 in patients followed-up in 2005, was defined as first incident non-fatal MI (ICD10 code I21), unstable angina pectoris (ICD10 code I20.0) or percutaneous transluminal coronary intervention and/or coronary artery by-pass grafting, whichever came first. The CHD events were retrieved by data linkage with the national Swedish Hospital Discharge Register (National Board of Health and Welfare, Stockholm).

Statistical methods

Results are presented as mean values ± one standard deviation (sd) or frequencies in Tables 1–3. Significance levels were adjusted for differences in age and sex by multivariate and logistic regression. Multivariate regression was used in Table 4 to analyse the association between time (years) from CHD event to follow-up in 2005 as dependent variable, and risk factor levels at follow-up in 2005 as independent variables. A P-value < 0.05 was considered statistically significant. All statistical analyses were performed using JMP version 5.1 (SAS Institute Inc., Cary, NC, USA).

Table 1.  Clinical characteristics and risk factors in 2002 and 2005, in patients with Type 2 diabetes and previous coronary heart disease (CHD)
 CHD 1–2 years previouslyCHD 3–7 years previously
2002200520022005
  1. Mean ± sd values and frequencies (%) are given.

  2. BP, blood pressure.

  3. Significance levels, after adjustment for age and sex, comparing patients in 2002 and 2005: *P < 0.001;P < 0.01;P < 0.05.

Numbers, n525141410873156
Years from CHD to follow-up1.5 ± 0.51.5 ± 0.54.9 ± 1.44.9 ± 1.4
Age, years67.0 ± 8.167.2 ± 8.267.8 ± 7.767.7 ± 7.7
Male gender, %68.867.769.170.6
Diabetes duration, years9.0 ± 7.38.9 ± 7.59.7 ± 7.39.2 ± 7.2
Diet/oral agents/insulin, %19/54/4819/60/4418/55/50*20/60/43
If obese (BMI ≥ 30), %16/61/5415/56/4913/58/5516/56/47
HbA1c, %7.26 ± 1.227.11 ± 1.147.42 ± 1.20*7.16 ± 1.13
HbA1c < 8.0, %79.281.674.7*80.9
HbA1c < 7.0, %47.154.440.9*51.6
HbA1c≤ 6.5, %27.832.722.7*29.2
BMI, kg/m229.7 ± 4.829.6 ± 4.829.7 ± 4.829.8 ± 4.8
BMI < 30 kg/m2, %58.958.057.356.0
BMI < 25 kg/m2, %13.515.214.614.0
Waist circumference, cm104.3 ± 12.3104.5 ± 12.6
 (n = 485) (n = 1189)
Waist < 102 cm men, < 88 cm women31.831.6
Waist < 94 cm men, < 80 cm women9.911.0
Regular physical exercise46.544.7
≥ 3 times per week, % (n = 1083) (n = 2449)
Antihypertensive drugs, %90.394.488.4*94.0
Systolic BP, mmHg140.0 ± 17.8137.8 ± 18.6140.5 ± 18.6*138.0 ± 17.8
Diastolic BP, mmHg75.9 ± 9.575.1 ± 9.676.3 ± 9.6*75.1 ± 9.6
Pulse pressure, mmHg64.1 + 16.262.7 + 16.664.2 + 16.262.9 + 16.0
Diastolic BP ≤ 80 mmHg, %78.382.074.9*80.4
Systolic BP ≤ 130 mmHg, %33.5*42.133.8*40.1
BP ≤ 130/80 mmHg, %30.5*39.631.7*37.4
BP < 130/80 mmHg, %17.523.818.021.7
Non-smokers, %88.488.286.887.9
men aged < 65 years, %83.981.080.780.6
women aged < 65 years, %76.981.570.775.6
Aspirin use, %85.189.382.6*88.1
Microalbuminuria, %13.119.818.321.4
(n = 381)(n = 895)(n = 791)(n = 2037)
Table 2.  Proportion of treatment with lipid-lowering drugs and blood lipid levels in 2002 and 2005 in patients with Type 2 diabetes and previous coronary heart disease (CHD)
 CHD 1–2 years previouslyCHD 3–7 years previously
2002200520022005
  1. Mean + sd values and frequencies (%) are given.

  2. TC, total cholesterol; LDL-C, low-density lipoprotein-cholesterol; TG, triglycerides; HDL-C, high-density lipoprotein-cholesterol; M, male; F, female.

  3. Significance levels, after adjustment for age and sex, comparing patients in 2002 and 2005: *P < 0.001;P < 0.01;P < 0.05.

Lipid-lowering drugs, %74.9*85.972.7*86.3
(n = 525)(n = 1414)(n = 1087)(n = 3156)
TC, mmol/l4.68 ± 1.05*4.34 ± 0.994.75 ± 1.04*4.40 ± 0.93
(n = 455)(n = 1236)(n = 979)(n = 2769)
TC < 5.0 mmol/l, %64.4*76.162.3*76.4
TC < 4.5  mmol/l, %47.3*59.841.3*55.4
LDL-C, mmol/l2.61 ± 0.86*2.29 ± 0.822.69 ± 0.91*2.33 ± 0.77
(n = 384)(n = 1029)(n = 819)(n = 2310)
LDL-C < 3.0 mmol/l, %71.9*81.768.4*83.6
LDL-C < 2.5 mmol/l, %48.7*65.445.1*63.3
LDL-C ≤ 1.8 mmol/l, %16.4*29.613.4*24.8
TG, mmol/l2.01 ± 1.101.91 ± 1.182.01 ± 1.161.97 ± 1.19
(n = 425)(n = 1136)(n = 916)(n = 2542)
TG < 1.7 mmol/l, %44.5†52.845.749.0
HDL-C, mmol/l1.16 ± 0.32*1.24 ± 0.391.18 ± 0.41*1.22 ± 0.35
(n = 410)(n = 1108)(n = 880)(n = 2482)
HDL-C > 1.0 M, > 1.2 F, %50.759.153.1*60.8
TG/HDL-C1.961.721.981.81
Table 3.  Clinical characteristics and risk factors in 2005 in Type 2 diabetic patients with previous coronary heart disease (CHD), by gender
 CHD 1–2 years previouslyCHD 3–7 years previously
MenWomenMenWomen
  1. Mean ± sd values and frequencies (%) are given.

  2. BP, blood pressure; TC, total cholesterol; LDL-C, low-density lipoprotein-cholesterol; TG, triglycerides; HDL-C, high-density lipoprotein-cholesterol.

  3. Significance levels comparing women and men, after adjustment for age: *P < 0.001;P < 0.01;P < 0.05.

Numbers, n9564582229927
Diabetes duration, years8.8 ± 7.49.2 ± 7.79.0 ± 7.19.8 ± 7.6
Diet/oral/insulin, %20/58/4317/62/4720/61/4119/58/47
HbA1c, %7.09 ± 1.167.16 ± 1.107.14 ± 1.257.23 ± 1.15
BMI, kg/m229.2 ± 4.530.4 ± 5.3*29.5 ± 4.430.6 ± 5.6*
Waist circumference, cm105.5 ± 12.1101.7 ± 12.5105.8 ± 11.8100.8 ± 14.0*
(n = 334)(n = 151)(n = 872)(n = 317)
Regular physical49.639.848.735.0*
exercise, %(n = 736)(n = 347)(n = 1735)(n = 714)
BP-lowering drugs, %93.496.593.595.2
Systolic BP, mmHg137.3 ± 18.7138.7 ± 18.3137.7 ± 17.3138.8 ± 19.0
Diastolic BP, mmHg75.7 ± 9.573.7 ± 9.575.9 ± 9.473.2 ± 9.8*
Non-smokers, %87.689.587.888.1
Age < 65 years, %81.081.580.675.6
Aspirin, %88.790.687.888.6
Lipid-lowering drugs, %85.886.086.186.8
TC, mmol/l4.23 ± 0.964.58 ± 1.0*4.32 ± 0.924.60 ± 0.93*
(n = 852)(n = 384)(n = 1967)(n = 802)
TG, mmol/l1.87 ± 1.181.98 ± 1.171.93 ± 1.232.08 ± 1.08*
(n = 782)(n = 354)(n = 1801)(n = 741)
HDL-C, mmol/l1.19 ± 0.381.35 ± 0.39*1.18 ± 0.331.31 ± 0.37*
(n = 764)(n = 344)(n = 1757)(n = 725)
LDL-C, mmol/l2.24 ± 0.802.40 ± 0.852.32 ± 0.762.38 ± 0.81
(n = 711)(n = 318)(n = 1639)(n = 671)
Microalbuminuria, %21.715.723.316.9*
(n = 608)(n = 287)(n = 1417)(n = 620)
Table 4.  Multiple regression analysis in 3447 patients with Type 2 diabetes describing the association between time (median 4 years, range 1–7 years) from first incidence of coronary heart disease (CHD) to follow-up in 2005 as dependent variable, and risk factor levels in 2005 as independent variables
PredictorsYears from CHD event to 2005 
Estimate ± se*I ratio
  • *

    Predictors were adjusted for each other and for age, sex and diabetes duration.

  • Significance levels:P < 0.01;P < 0.01;§P < 0.05.

Total cholesterol0.161 ± 0.0423.9
Smoking0.171 ± 0.0543.2
Body mass index0.015 ± 0.0072.0§
HbA1c0.045 ± 0.0331.4
Diastolic blood pressure0.005 ± 0.0041.3
Triglycerides–0.014 ± 0.035–0.4
Systolic blood pressure–0.004 ± 0.002–1.8
High-density lipoprotein-cholesterol–0.256 ± 0.106–2.4§

Results

Risk factor levels in 2002 and 2005

Patients with Type 2 diabetes and first incidence of CHD were followed up in 2002 and in 2005, as shown in Table 1. Comparing patients with CHD 1–2 years before follow-up, it was found that the achievement of treatment targets for HbA1c, BP and blood lipids was considerably improved in 2005 compared with 2002. The proportion of patients with HbA1c < 7.0% increased from 47 to 54%, and the proportion of those with BP ≤ 130/80 mmHg increased from 31 to 40%. The achievement of blood lipid level targets was even better (Table 2): total cholesterol < 4.5 mmol/l increased from 47 to 60%, LDL-cholesterol < 2.5 mmol/l from 49 to 56%, triglycerides < 1.7 mmol/l from 45 to 52%, and HDL-cholesterol > 1.0 mmol/l (men) or 1.3 mmol/l (women) from 47 to 56%. This was accompanied by an increased use of oral hypoglycaemic agents, and a considerable increase in lipid-lowering drugs (mainly statins) from 75 to 86%, whereas the proportion of antihypertensive drugs prescribed remained high, from 90 to 94%.

The same picture was seen when comparing patients with CHD 3–7 years before follow-up in 2002 and 2005, with improved achievement in HbA1c < 7.0 (41–52%), total cholesterol < 4.5 mmol/l (41–55%) and LDL-cholesterol < 2.5 mmol/l (50–58%). Use of antihypertensive drugs and lipid-lowering drugs also increased, from 88 to 94% and from 73 to 86%, respectively.

In contrast, as regards lifestyle characteristics in all patients with previous CHD, a high mean BMI around 30 kg/m2 showed no decline from 2002 to 2005. The frequency of obesity (BMI ≥ 30 kg/m2) was around 43% in 2005. Very few patients (approximately10%) reached the current target for waist circumference, and only 32% a more conservative target. Furthermore, the proportion of smokers in all patients with CHD (13%) showed no decline from 2002 to 2005. In a subgroup of patients aged < 65 years, smokers were even more frequent, especially in women: 19% of women with CHD 1–2 years previously, 24% in women with CHD 3–7 years previously, and 19% of men were still smoking in 2005. Physical exercise was performed regularly three times per week or more in about half of all patients.

The prevalence of microalbuminuria increased from 2002 (13–18% for patients with CHD 1–2 years or 3–7 years previously) to 2005 (20–21%). The use of aspirin increased somewhat from 2002 (83–85%) to 2005 (88–89%).

Gender differences

Comparing women and men with CHD at follow-up in 2005 (Table 3), no significant differences was seen regarding levels of mean HbA1c and systolic BP, or use of antihypertensive drugs, lipid-lowering drugs and aspirin. Women exercised less regularly, and had higher mean BMI, total cholesterol, LDL-cholesterol and triglycerides, although also a higher mean HDL-cholesterol. Men had a higher frequency of microalbuminuria, and a higher mean waist circumference.

Years passed since CHD event vs. risk factor control

Multiple regression analysis was used to investigate the association between time (years) from first incidence of CHD to follow-up as dependent variable and risk factor status in 2005 as independent variables (see Table 4). Median (range) years was 4 (1–7). It was found that the number of years from CHD to follow-up was significantly associated with levels of total cholesterol, smoking and BMI (positively) and HDL-cholesterol (negatively). Thus, levels of these risk factors deteriorated with longer time from CHD to follow-up. Levels of HbA1c and BP, however, did not deteriorate with time.

Discussion

The implementation of evidence-based therapies [17,18] and optimized risk factor control is especially important in patients with the combination of diabetes and CHD, as potential health benefits are proportional to risk factor level achievement and the proportion of patients treated with drugs such as statins [9].

This observational study has demonstrated encouragingly that the current treatment targets for total cholesterol and LDL-cholesterol were achieved by around 60% of patients with the combination of Type 2 diabetes and CHD in 2005, whereas the current targets for BP control and glycaemic control (HbA1c) were reached by about 40–50% of subjects. Furthermore, all these achievements increased significantly during the study period, most markedly for blood lipid levels. However, a high prevalence of adverse lifestyle characteristics was also found; mean BMI and waist circumference were high. Obesity was present in > 40% (2005), and a high waist circumference above target level in 70%. As many as 20–25% of patients aged < 65 years continued to smoke in 2005, and only around 50% exercised regularly three times per week or more.

The increase in use of lipid-lowering drugs (almost exclusively statins) was substantial in patients with CHD. This was accompanied by the achievement of less stringent targets for total cholesterol and LDL-cholesterol in as many as 80% of patients with CHD (2005). Moreover, the use of lipid-lowering drugs also contributed to current triglyceride and HDL-cholesterol targets attained by as many as 50–60%, although statins do not have their main effect on these blood lipid variables. However, multivariate regression showed that total cholesterol and HDL-cholesterol values tended to deteriorate with longer time passed from CHD event to follow-up, and the mean ratio of triglycerides/HDL-cholesterol still remained relatively high, 1.7–1.8. Both triglycerides and HDL-cholesterol have been verified as risk factors for cardiovascular disease, and are also indicators of dyslipidaemia associated with insulin resistance [19,20].

As many as four-fifths of patients with CHD reached a less stringent HbA1c target < 8.0% in 2005, and the increasing achievement from 2002 to 2005 was accompanied by an increasing trend for use of oral hypoglycaemic agents. However, only one-half of subjects reached the current HbA1c target of < 7.0%. Strongly intensified lifestyle measures and possible use of weight-reducing therapies seem necessary in order to obtain higher achievement of the current HbA1c target due to the high proportion of obesity. Regular exercise of 30 min twice per week has been shown to reduce cardiovascular risk by 30% in patients with hypertension [21].

Although almost all patients with CHD were treated with antihypertensive drugs in 2005, only 40% reached the BP target of ≤ 130/80 mmHg. Even fewer (25%) reached the current target of < 130/80 mmHg. It was clearly demonstrated that the systolic BP target was difficult to reach, whereas diastolic BP ≤ 80 mmHg was obtained by as many as 80%. More evidence-based use of antihypertensive drugs seems necessary, also suitable to decrease a high systolic BP level. This necessity was also underlined by the finding that one-fifth had microalbuminuria, a marker of greatly increased cardiovascular morbidity and mortality in patients with Type 2 diabetes [13,22].

Some previous surveys have analysed risk factor management in diabetic patients with coronary artery disease [23–25]. The EUROASPIRE II Study was carried out on 1086 male and female Type 2 diabetic patients (mean age 63 years) with CHD 1–2 years before follow-up in 1999–2000 [26]. Considerably more patients with CHD 1–2 years before 2005 in this study reached current targets for total cholesterol and LDL-cholesterol (60 and 65%, respectively) than in EUROASPIRE II (29 and 25%, respectively). Accordingly, 86% used lipid-lowering drugs in the Swedish study, but only 62% in EUROASPIRE II. However, the achievement of BP < 130/80 mmHg was similar in both studies (24% vs. 23%), in spite of a high use of antihypertensive drugs in both studies (94% vs. 92%, respectively). Furthermore, many patients were still smoking in both studies (15% vs. 17%). The recent Euro Heart Survey on Diabetes and the Heart examined 1524 diabetic patients with CHD from 25 countries in 2003 and found that only 30% achieved the BP target of < 140/90 mmHg [25]. Lipid-lowering drugs were given to 66%, whereas only 45% reached total cholesterol < 5 mmol/l and LDL-cholesterol < 3 mmol/l. In comparison, at follow-up in 2005 in this study 85% had lipid-lowering drugs and around 80% achieved these two lipid targets.

In Swedish patients with diabetes and MI, the national Register of Information and Knowledge about Swedish Heart Intensive Care Admission (RIKS-HIA) register-based study has demonstrated a clear decrease in 1-year post-MI total mortality from 29.7 to 19.7% during the period 1995–2002, although it is still higher than in patients without diabetes (16.6 to 12.1%) [27,28]. The relative 1-year mortality risk in diabetic patients, compared with non-diabetic patients, encouragingly decreased significantly from 1.44 (1.36–1.52) in 1995–1998 to 1.31 (1.24–1.38) in 1999–2002. The RIKS-HIA study found improved pre-infarction management of cardiovascular risk factors together with increased in-hospital use of anticoagulants, statins, β-blockers, ACE-inhibitors and early re-vascularization to be likely contributors, although factors such as improved glycaemic control during and after the MI were not analysed. For comparison in our study, 80% of patients with CHD reached the HbA1c target of < 8.0%, whereas half of them reached the target of < 7.0%.

Limitations of the study

The coverage rate of NDR is still not optimal, but an increasing proportion of all patients with diabetes in Sweden are now being registered on an annual basis. We believe that data regarding onset age of diabetes, continuous risk factor variables and use of drugs were reliable in this study. Smoking might be somewhat biased due to under-reporting by patients or examiners. We also believe data concerning end-point events to be reliable, as all events were diagnosed during treatment at hospitals reporting to the national Mortality Register and the Hospital Discharge Register, and CHD defined by combining data from these registers is an efficient validated alternative to revised hospital discharge notes and death certificates [29]. The definition of Type 2 diabetes used here could still include some patients with possible late autoimmune disease of the adult, because no antibody (glutamic acid decarboxylase, islet cell antibody) level information was available.

The strength of this study was supported by the fact that the numbers of patients and events were relatively high, and that patients in the NDR were collected from the general Swedish diabetes population at PHC centres and hospital diabetes clinics nationwide. Similar trends of improved control influencing prognosis have recently also been reported from another related Swedish register study, RIKS-HIA [28].

In conclusion, the increasing and high proportion of lipid-lowering drugs prescribed in patients with CHD has been accompanied by improved mean blood lipid levels. Yet considerable problems remain in achieving current treatment goals for BP and HbA1c. The high prevalence of adverse lifestyle characteristics, such as obesity, high waist circumference and smoking, with their tendency to aggravate with longer time passed since the CHD event, contributes to this situation. Intensified use of structured programmes, evidence-based drug medication and professional lifestyle intervention support, similar to what was applied in the Steno-2 trial regarding primary prevention of cardiovascular disease [30], seems urgent in these patients with a high risk for future cardiovascular events including re-infarction.

Competing interests

Nothing to declare.

Acknowledgements

We thank the regional NDR coordinators, all participating nurses, physicians and other staff members who have contributed to this study. We also thank the patients with diabetes who, both individually and collectively, through their patient organization, the Swedish Diabetes Federation, support the NDR. The Swedish Board of Health and Welfare and the Swedish Society for Diabetology jointly fund the NDR.

Ancillary