Evidence based diabetology : strategies to prevent macrovascular disease and to reduce mortality

Diabetes mellitus (DM) is the most common metabolic disease worldwide and the
number of newly diagnosed cases is increasing. DM is strongly associated with a
number of devastating chronic late complications, including retinopathy, nephropathy
and neuropathy (microvascular complications), as well as cardiac, cerebrovascular
and peripheral vascular disease (macrovascular complications). Despite lots of
progress in therapeutic possibilities during the last decades, mortality risk due to
macrovascular complications is still increased in patients with DM when compared to
non diabetic individuals. It has also been suggested that specific risk factors
influence macrovascular risk differentially in persons with and without DM. In
addition, there is still uncertainty whether the effectiveness of certain treatment forms
differs between patients with and without DM, and in the patients with DM, between
type 1 and type 2. The focus of my thesis was on prevention and therapy of
macrovascular disease in patients with type 1 and type 2 DM, as well as on
comparisons with patients without DM. Three studies (Studies A-C) investigated the
effectiveness of specific treatment forms on macrovascular disease by means of
systematic reviews and meta-analyses, whereas two studies (Studies D and E)
evaluated novel risk indicators using survival analysis based on data from the 'Swiss
Cohort of the World Health Organisation (WHO) Multinational Study of Vascular
Disease in Diabetes'.
The aim of Study A (published in Am Heart J 2006 Jul;152(1):27-38) was to assess
the effect of improved glycaemic control on cardiac, cerebrovascular and peripheral
vascular complications in type 1 and type 2 DM. Outcomes included the incidence
rate ratios (IRRs) for any macrovascular event, cardiac events, stroke, and peripheral
arterial disease. Results showed a 62% (95% Confidence Interval (CI) 44-74%) and
19% (95% CI 9-27%) reduction in macrovascular risk for improved glycaemic control
in type 1 and type 2 DM, respectively. In type 1 DM the effect was mainly based on a
reduction of cardiac and peripheral vascular events. In type 2 DM it was due to
reductions in stroke and peripheral vascular events. The effects appeared to be
particularly important in younger patients with shorter duration of DM.
Study B (published in Curr Med Res Opin 2006 Mar;22(3):617-23) examined the
effectiveness of fibrates (peroxisome proliferator activated receptor α-agonists) in the
prevention of coronary heart disease in type 2 DM. The primary outcome of this
meta-analysis was the IRR for coronary heart disease (CHD) events (a combination
of non fatal myocardial infarction and death due to CHD). Secondary endpoints were
death due to CHD, fatal and non fatal myocardial infarction, and fatal and non fatal
stroke. The results of Study B showed a 16% risk reduction for CHD events (95% CI
4-26%) in patients with type 2 DM when treated with fibrates compared to placebo.
For the secondary endpoints a tendency towards reduction in risk was found,
although this did not reach conventional levels of statistical significance.
Coronary stenting is established as a treatment of coronary heart disease. The aim of
Study C (published in Heart 2006 May;92(5):650-7) was to indirectly compare the
effects of polymer based sirolimus versus paclitaxel eluting coronary stents and to
examine whether they are equally effective in the prevention of restenosis in patients
with and without DM. The indirect comparisons were performed by calculating the
ratio of incidence rate ratios (RIRR) of studies comparing sirolimus eluting stents
versus conventional bare metal stents and studies comparing paclitaxel eluting
versus bare metal stents. The overall study population and patients with and without
DM were analysed separately. Outcomes included in-stent- and in-segment
restenosis, target lesion revascularisation, and major adverse cardiac events. The
results of this study showed that rates of revascularisation procedures are reduced
by sirolimus as well as paclitaxel eluting stents when compared to bare metal stents
independent of the study population. However, in persons without DM a superiority of
the sirolimus eluting stent to the paclitaxel eluting stent was found for all endpoints
under investigation. In contrast, for persons with DM no statistically significant
differences between the two drug eluting stents were found. A meta-regression
analysis confirmed a difference between individuals with and without DM.
Study D (published in Diabetologia 2006, DOI 10.1007/s00125-006-0483-1)
evaluated the long-term association of two parameters with mortality, namely QT
interval and resting heart rate (rHR) in patients with type 1 and type 2 DM. Based on
the 23-year follow up of the 'Swiss Cohort of the WHO Multinational Study of
Vascular Disease in Diabetes', the prognostic values of these two risk factors were
examined on all-cause, cardiovascular and cardiac mortality and mortality due to
ischaemic heart disease using a Cox proportional hazards model. Results showed an
association of prolonged QT interval (corrected for heart rate, QTc) with an increased
mortality risk due to all causes, as well as cardiovascular and cardiac disease in type
1 DM, whereas no association was found for rHR. In contrast, in patients with type 2
DM elevated rHR but not QTc was associated with an increased risk of all-cause
mortality as well as death due to cardiovascular, cardiac and ischaemic heart
disease.
Study E (published in J Intern Med 2006 Sep;260(3):272-80) was based on an 15-
year follow up of the 'Swiss Cohort of the WHO Multinational Study of Vascular
Disease in Diabetes'. This study evaluated the long-term association of
apolipoprotein B (apo B) with mortality risk in patients with type 1 DM. Compared to
Study D, follow up was shorter due to the fact that apo B was only measured later in
the course of the study. Analyses were performed for all-cause and cardiac mortality
and mortality due to ischaemic heart disease, using a parametric proportional
hazards model based on the Weibull distribution. Apo B was found to be positively
related to all-cause and cardiac mortality, and mortality due to ischaemic heart
disease. An apo B >0.96 g/L translated into a doubling of overall mortality, and a
sevenfold increase of mortality due to cardiac disease or ischaemic heart disease.
In conclusion, this thesis showed that:
• The incidence of macrovascular events is reduced by improved glycaemic
control, both in type 1 and type 2 DM. Although effects on specific
manifestations of macrovascular disease are different between the two types
of DM, in absolute terms benefits are comparable.
• CHD events are substantially reduced in patients with type 2 DM when treated
with fibrates. Nevertheless, their exact role in lipid lowering treatment needs to
be investigated further.
• Compared to bare metal stents, sirolimus as well as paclitaxel eluting stents
are effective in reducing the rates of revascularisation procedures. Based on
indirect evidence, stents eluting sirolimus appear to be superior to paclitaxel
eluting stents in patients without DM but not in patients with DM.
• Prolongation of QTc is related to an increased mortality risk in patients with
type 1 DM, whereas elevated rHR is associated with higher mortality risk in
patients with type 2 DM.
• Increased apo B levels are consistently associated with mortality risk in type 1
DM.
In addition to these conclusions, two general statements can be made:
• The effectiveness of therapeutic interventions may be different in persons with
compared to persons without DM.
• Within the group of patients with DM, the effectiveness of specific interventions
may vary between type 1 and type 2 DM.