bmjopen.bmj.com€¦ · for peer review only references 1. idf diabetes atlas - 8th edition 2017...

BMJ Open is committed to open peer review. As part of this commitment we make the peer review history of every article we publish publicly available. When an article is published we post the peer reviewers’ comments and the authors’ responses online. We also post the versions of the paper that were used during peer review. These are the versions that the peer review comments apply to. The versions of the paper that follow are the versions that were submitted during the peer review process. They are not the versions of record or the final published versions. They should not be cited or distributed as the published version of this manuscript. BMJ Open is an open access journal and the full, final, typeset and author-corrected version of record of the manuscript is available on our site with no access controls, subscription charges or pay-per-view fees (http://bmjopen.bmj.com). If you have any questions on BMJ Open’s open peer review process please email

[email protected]

on Novem

ber 13, 2020 by guest. Protected by copyright.

http://bmjopen.bm

j.com/

BM

J Open: first published as 10.1136/bm

jopen-2018-027199 on 3 April 2019. D

ownloaded from

http://bmjopen.bmj.com/

[email protected]


For peer review only

Impact of coronary artery disease severity and age on risk of cardiovascular outcome in diabetes patients: a

nationwide observational study

Journal: BMJ Open

Manuscript ID bmjopen-2018-027199

Article Type: Research

Date Submitted by the Author: 10-Oct-2018

Complete List of Authors: Jernberg, Tomas; Danderyd University Hospital, Karolinska Institutet Lindholm, Daniel ; Uppsala Clinical Research Center, Svennblad, Bodil; Uppsala Clinical Research Center Hasvold, Lars Pål; AstraZeneca Nordic, Medical department Bodegård, Johan; AstraZeneca Nordic, Medical department Andersson, Karolina; AstraZeneca R&D Thursesson, Marcus; Statisticon, Erlinge, David; Lunds Universitet, Clinical science Janzon, Magnus; Linkopings universitet, Cardiology

Keywords: Coronary heart disease < CARDIOLOGY, General diabetes < DIABETES & ENDOCRINOLOGY, Cardiac Epidemiology < CARDIOLOGY

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml

BMJ Open on N

ovember 13, 2020 by guest. P

rotected by copyright.http://bm

jopen.bmj.com

/B

MJ O

pen: first published as 10.1136/bmjopen-2018-027199 on 3 A

pril 2019. Dow

nloaded from



Impact of coronary artery disease severity and age on risk of cardiovascular outcome in

diabetes patients: a nationwide observational study

Tomas Jernberga)

, Daniel Lindholmb,c)

, Pål Hasvoldd)

, Bodil Svennbladc), Johan Bodegård

d), Karolina

Andersson Sundelle)

, Marcus Thuressonf), David Erlinge

g), Magnus Janzon

h)

a) Department of clinical sciences, Danderyd University Hospital, Karolinska Institutet, Stockholm,

Sweden

b) Department of Medical Sciences, Cardiology, Uppsala University, Uppsala, Sweden

c) Uppsala Clinical Research Center, Uppsala, Sweden

d) AstraZeneca Nordic-Baltic, Södertälje, Sweden

e) AstraZeneca R&D, Gothenburg, Sweden

f) Statisticon AB, Uppsala, Sweden

g) Lund University, Lund, Sweden

h) Department of Cardiology and Department of Medical and Health Sciences, Linköping University,

Linköping, Sweden

Corresponding author:

Pål Hasvold, Medical department, AstraZeneca Nordic-Baltic, Fredrik Selmers vei 6

Box 6050 Etterstad, 0601 Oslo, Norway, mail: [email protected]

Word count: 5549

Page 1 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



ABSTRACT

Background: The aim was to compare short-term cardiovascular outcome in type 2 diabetes (T2D)

patients without coronary artery disease (CAD), with CAD but no prior myocardial infarction (MI), and

those with prior MI; and assess the impact on risk of age when initiating first-time glucose-lowering

drug (GLD).

Methods: This was a cohort study linking morbidity, mortality, and medication data from Swedish

national registries. Predicted cumulative incidence for the CV outcome (MI, stroke and CV-mortality)

was estimated. Furthermore, a Cox model was developed where age at GLD start and CV risk was

modeled.

Results: The study included 260 070 first-time users of GLD T2D patients during 2007-2016. Of these,

221 226 (85%) had no CAD, 16 294 (6%) had stable CAD - prior MI, and 22 550 (9%) had CAD + MI.

T2D patients without CAD had a lower risk of CV outcome compared with the CAD populations (-/+

prior MI), (3-year incidence 4.78% vs. 5.85% and 8.04%). The difference in CV outcome was primarily

driven by a relative greater MI risk among the CAD patients. For T2D patients without CAD an almost

linear association between age at start of GLD and relative risk was observed, whereas in CAD

patients, the younger (< 60 years) patients had a relative greater risk compared with older patients.

Conclusions: T2D patients without CAD had a lower risk of the CV outcome compared to the T2D

populations with CAD, primarily driven by a greater risk of MI. For T2D patients without CAD an

almost linear association between age at start of GLD and relative risk was observed, whereas in CAD

patients, the younger patients had a relative greater risk compared with older patients. Our findings

suggest that intense risk prevention should be the key strategy in the management of T2D patients,

especially for younger patients.

Page 2 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



Strengths and limitations of this study

Strengths

• The study was conducted in a nationwide national cohort including all patients who were

dispensed a first-time GLD during the observational period and thus limit the potential

problems with selection bias.

Limitations

• Access to clinical data describing the extent and severity of diabetes (blood glucose, HbA1C,

weight, smoking pattern and kidney function) which might have an impact on the risk was

not available.

• The study is reliant on ICD-10 codes for morbidity data and therefore, the possibility of

coding errors cannot be ruled out.

• Another limitation of our study was the lack of available data on socio-economic status which

is known to affect risk in T2D patients.

Page 3 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



BACKGROUND

In recent years, there has been an increase in the incidence of diagnosed type 2 diabetes (T2D) in

both developed and developing societies, with approximately 425 million people (8–9%) living with

diabetes worldwide in 2017 (1).

T2D is commonly associated with macrovascular complications, often resulting in early manifestation

of coronary artery disease (CAD) and increased risk of cerebrovascular disease (2). Diabetes is

associated with a substantial increase in risk of major cardiovascular (CV) events in patients both

with and without established cardiovascular disease (CVD) (3, 4). This to such extent that CVD-related

death is the most common cause of mortality among diabetes patients (5).

Patients with T2D without a previous myocardial infarction (MI) have as high a risk of MI as patients

without diabetes who have had a previous MI (6). Thus, when it comes to prevention strategies,

diabetes is considered equivalent to CAD, with a high or very high risk depending on whether target

organ damage is present (7, 8). The risk of CV events increases with a confirmed diagnosis of

atherosclerosis where a history of MI is associated with an even higher risk of further cardiac events

(9).

There has been an improvement in post-MI survival in Western countries, leading to an overall

growth of the population with a history of CAD (10). Combined with the increased incidence of T2D

patients, it is likely that the T2D patient population with a history of CAD will increase in the coming

decades and thus, increased knowledge of the short-term cardiovascular event pattern is important.

So far, there are no studies comparing the short-term prognostic impact of a history of clinical stable

CAD with that of an atherothrombotic disease demonstrated as previous MI in diabetic patients.

Moreover, the consequences of age when initiating glucose lowering drug (GLD) in relation to short-

term CVD risk have not been well described either. These are all important considerations when

targeting patients for intensified secondary preventive measures.

The aim of the present study was to compare short-term (3 year) cardiovascular outcome in T2D

patients without CAD, with CAD but no prior MI, and those with prior MI; as well as to assess the

impact of age when initiating first-time GLD. For these purposes, we used a highly representative

nationwide sample of all T2D patients initiating first-time GLD in Sweden over 7 years.

Page 4 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



METHOD

Study design

Data for this observational study was retrieved by linking data from Swedish mandatory nationwide

registers: the Swedish National Patient Register (NPR; including inpatient admission and discharge

dates, and main and secondary diagnoses according to the International Statistical Classification of

Diseases and Related Health Problems, 10th Revision [ICD-10]), the Swedish Prescribed Drug Register

(SPDR) (11), and the Swedish Cause of Death Register (12). The Swedish NPR covers more than 99%

of all somatic (including surgery) and psychiatric hospital admissions and discharges (13). The SPDR

contains data on all prescribed medications dispensed from pharmacies in Sweden. Linkage of

patient-level data was performed by the Swedish National Board of Health and Welfare utilizing the

unique personal identification (ID) numbers, mandatory for every citizen in Sweden, and thereafter

replaced by a study ID for further data processing. The study was approved by the Stockholm

regional ethics committee (registration number 2013/2206-31).

Study population

The study population included all patients with T2D initiating use of GLD (ATC code A10B) from

January 1 2007-December 31 2016. The index date was defined as the date of the first filled

prescription of a GLD during the observation period. To be defined as a first-time user the patient

should not have filled any prescriptions for GLD prior to the index date. Patient characteristics at

baseline were established using hospitalization and drug utilization data from national registers from

1987 onwards. Patients dispensed GLD before the study period were excluded.

Three study populations were defined based on patients’ clinical status when their first GLD was

dispensed (see Supplementary data for ICD-10 codes).

1. T2D patients without CAD: T2D patients without any previous diagnosis of CAD (defined as a

history of myocardial infarction (MI), unstable angina, or stable angina pectoris)

2. T2D patients with CAD without prior MI: T2D patients with previous diagnosis of CAD without

MI (defined as a history of stable or unstable angina pectoris, but no myocardial infarction)

3. T2D patients with CAD with prior MI: defined as T2D patients with a history of MI.

The three study populations were stratified into the following age categories (based on age when

dispensed their first GLD): < 55 years, 55-64 years, 65-74 years, 75-84 years, >85 years.

Page 5 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



Outcomes

The primary CV outcome was a composite of hospitalization with main diagnosis of non-fatal MI (ICD-

10: I21), non-fatal stroke (ICD-10: I61-I64) or CV death (death with ICD-10 codes I00–I99 as a primary

diagnosis).

Sensitivity analysis

In order to test the impact of prior stroke, outcomes in T2D patients without CAD, MI or stroke, were

compared with T2D patients with CAD without prior MI or stroke, and T2D patients with CAD with MI

without stroke.

Statistical analyses

Baseline characteristics are presented as mean and standard deviation for continuous variables and

absolute and relative frequencies for categorical variables. Each patient was followed from date of

index date to date of death, or end of study observational period. Comparison between groups with

respect to time to event outcomes were analyzed using Cox proportional hazards models adjusted

for age, sex, diabetes duration, atrial fibrillation, and heart failure. The results are illustrated using

predicted cumulative incidence plots (based on the Cox models), as well as in unadjusted Kaplan-

Meier plots.

In order to explore the change in relative risk related to age, a Cox model was developed where age

was modeled using a restricted cubic spline with 5 knots. The results are illustrated as the log of the

hazard ratio over time with the mean age of the total cohort as the reference.

Results are presented as hazard ratios (HRs) and 95% confidence intervals (CIs). Statistical analyses

were performed using SAS version 9.3 (SAS Institute Inc., Cary, NC, USA) and R version 3.5.0.

Page 6 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



RESULTS

Overall, 260 070 T2D patients were first-time users of a GLD during the observation period and could

be included. Of these, 221 226 (85%) were included in T2D population without CAD, 16 294 (6%) in

the T2D CAD population without MI, and 22 550 (9%) in the T2D CAD population with a history of MI.

Mean follow-up (FU) was 4.5 years with a maximum of 9.0 years, compromising a total of 1 179 802

patient-years of FU.

T2D patients without CAD were younger (61.4 years), more often female (45%), and had a lower

incidence of stroke (5%), atrial fibrillation (6%) and heart failure (3%) compared to the two T2D CAD

populations (Table 1). The CAD population without a history of MI had a mean age of 70.9 vs. 70.3

years in CAD patients with a history of MI. The two CAD populations included 40% vs. 29% women,

10% vs. 11% stroke, 22% vs. 22% atrial fibrillation, and 19% vs. 28% heart failure, respectively. There

were only minor differences in GLD therapy among the three study populations, with the majority of

patients treated with metformin (>76%), sulfonylurea (>6%) or insulin (>11%). More patients in the

two CAD populations were treated with statins (68% vs. 26%), anti-platelets (71% vs. 17%) and anti-

hypertensives (92% vs. 56%) than the T2D patients without CAD.

T2D patients without CAD had a lower risk of the CV outcome compared to the T2D populations with

CAD (3 year adjusted cumulative incidence for a 63-years-old patient (mean age of the study

population) 4.78% vs. 5.85% and 8.04%) (Figure 1 and Supplementary data, Table 1a). The greater

risk seen for T2D CAD patients with no prior MI vs patients without CAD was primarily driven by MI (3

year adjusted cumulative incidence 1.66% vs. 3.09% vs (Figure 1 and Supplementary data, Table 1b).

The results of the sensitivity analysis with exclusion of patients with a prior history of stroke showed

a consistent pattern to the main results, T2D patients without CAD or stroke, had a lower risk of CV

outcome compared to T2D CAD patients without MI and stroke and T2D patients with MI without

stroke (3 year adjusted cumulative incidence for a 62-year-old patient (mean age of the sensitivity

analysis population) 4.22% vs, 5.28% and 7.80%). Also, in this population the difference in risk was

primarily driven by MI (Supplementary data, Figure 1)

The baseline characteristics for patients were stratified by age at first GLD dispense irrespective of

CAD status during the observation period (Table 2). The proportion of women was lower, in the

younger categories (<55 years), 42% compared to 61% among patients older than 85 years. The

proportion of patients with cardiovascular comorbidities was greater in older patients, the

proportion of patients with previous MI in > 85 years was 19%, compared to 3% in patients < 55

years, and the corresponding numbers for heart failure were 26% and 1%, respectively. The

increased cardiovascular burden among the older patients were also reflected in the proportion of

Page 7 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



patients treated with secondary preventive drugs; the proportion of patients on statin therapy

increased from 16% to 26% and for anti-platelets from 8% to 53% for patients < 55 years to > 85

years. A larger proportion of the older patients were treated with insulin and sulfonylurea, whereas

younger patients predominately were treated with metformin.

The short-term (3 year after index) risk of CV outcome for T2D patients differed among the three

study populations in relation to age at first GLD dispense. Patients without CAD showed an almost

linear association between age and relative risk of CV outcome respectively (Figures 2 and 3).

Presence of CAD was associated with a relatively higher increase in relative risk of CV outcome in

younger (< 60 years) patients, despite having less observed cardiovascular comorbidities at baseline

(Table 2 and Figure 2 and 3). In patients with CAD, with or without previous MI, the relative CV and

MI risk did not increase with age in patients younger than 65 years. In patients older 65 years, there

was an increased relative risk of CV outcome and risk of MI with increasing age (Figure 2 and 3).

A relative greater risk of CV outcome was seen among the younger T2D CAD patients, with a 3-year

cumulative incidence of primary outcome in patients with a history of MI < 55 years 10.07% (8.65-

11.47%) vs. patients 65-74 years 14.52% (13.66-15.37%) (Supplementary data, Figure 2). In all age

categories, the MI risk was the main risk contributor, both for T2D patients without CAD or MI and

for the two CAD populations.

Page 8 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



DISCUSSION

In the present study, we examined a highly representative sample of all T2D patients initiating first-

time glucose lowering treatment in a whole country for an 8-year period. There were two key

findings: First, T2D patients without CAD had a lower risk of the CV outcome compared to the T2D

populations with CAD. The difference in CV outcome was primarily driven by a greater risk of MI in

T2D patients with CAD. Second, for T2D patients without CAD an almost linear association between

age at start of GLD therapy and relative risk of CV outcome and MI was observed, whereas in CAD

patients, the younger (< 60 years) patients had a relative greater increase in short-term CV risk

compared with older patients, despite having less cardiovascular comorbidities at baseline. Also, in

T2D patients with CAD, there was no increase in relative CV risk with increasing age until the age of

65, with a linear increase in risk thereafter.

Previously it has been shown that a T2D population compared with the general population in Sweden

has significant increased risk for cardiovascular events like myocardial infarction, heart failure atrial

fibrillation and all-cause death (14). Over time from 1998 to 2013, the incidence of hospitalization for

cardiovascular disease and cardiovascular mortality has almost decreased by half in patients with

T2D but remained considerably higher than in matched controls without T2D (15).

Our findings highlight that there is a marked difference in CV and MI risk in different T2D populations

related to the presence and severity of CAD disease. Even the T2D patients without prior MI have a

risk that is comparable to a post myocardial infarction population (10). Bearing in mind that we focus

on difference in short-term risk in this paper (3 year after initiation of GLD treatment for T2D), and

still see large differences in risk between the different study populations, it is inevitable that T2D

patients with a history of CAD should be carefully monitored and managed with a long-term

perspective. That is, by effective prevention programs and aggressive drug therapy after being

diagnosed with T2D, particularly in those considered to be at high risk of ischemic events. A recent

study also from the Swedish National Diabetes Registry showed that patients with T2DM who had

appropriate risk factor control had little or no excess risk of CV events as compared with the general

population (16).

Only a limited number of studies have examined the effect of age at T2D diagnosis/start of first-time

GLD on CV risk. A recent study from Australia showed a that a younger age at T2D diagnosis was

Page 9 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



associated with a higher risk of all-cause and CVD related death (17). Hence, the duration of T2D was

shown to be an important factor for life-time risk, as also reported previously in several other

publications (14). That said, there is still a scarcity of data describing CV risk in T2D age-stratified

populations, comparable to our findings, focusing on the relatively short-term CV risk after T2D

diagnosis, and not the full life-time risk in relation to age at T2D diagnosis.

For patients without a history of CAD at T2D diagnosis, an almost linear association between age and

CV and MI risk was observed. In contrast, a relatively higher increase in risk for CV outcome and MI

events was observed in younger patients with established CAD when being diagnosed with T2D,

compared to older patients with CAD despite having less cardiovascular comorbidities. Furthermore,

a low proportion of the patients were treated with statins and anti-platelets, ranging from 16%

statins and 8% anti-platelets for patients below 55 years, to 45% statins and 43% anti-platelets for

patients 75-84 years.

Recent data from Sweden complementing our data, examined clinical characteristics in age stratified

T2D patients, and showed that patients who develop T2D earlier in life are more frequently obese,

have a more adverse lipid profile, higher HbA1c levels, and a faster deterioration in glycaemic control

compared with individuals who develop diabetes later in life (18). This more severe metabolic

dysregulation could be associated with accelerated atherosclerosis. This should be amenable to

primary prevention both by lifestyle changes and medical treatment. However, they also found that a

low proportion of these young patients received blood pressure lowering drugs, statins and anti-

platelet drugs (18).

The combination of these metabolic risk factors, T2D diagnosis and presences of CAD in young age

predicts high risk of major adverse cardiovascular events, especially a risk for of MI. These findings

further highlight the importance of providing a close monitoring of younger diabetes patients with

established CAD. That said, there may be a possible uncertainty regarding clinical responsibility for

drug treatment initiation between specialist and primary care for these patients, which may hamper

a thorough T2D patient management.

The strengths of our study are that it was conducted in a nationwide national cohort including all

patients who were dispensed a first-time GLD during the observational period and thus limit the

potential problems with selection bias. The study however also has limitations. Firstly, we did not

Page 10 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



have access to clinical data describing the extent and severity of diabetes (blood glucose, HbA1C,

weight, smoking pattern and kidney function) which might have an impact on the risk. However,

complementary data on these clinical variables from Sweden were recently published, showing that

an unfavorable metabolic profile of the younger patients could be a part of the explanation for our

findings (18). Second, our study is reliant on ICD-10 codes for morbidity data and therefore, the

possibility of coding errors cannot be ruled out. However, previous data show that coding is correct

in 98% of Swedish NPR entries (13). Another limitation of our study was the lack of available data on

socio-economic status which is known to affect risk in T2D patients (19).

Page 11 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



CONCLUSION

In conclusion, T2D patients without CAD had a lower risk of the CV outcome compared to the T2D

populations with CAD. The difference in CV outcome was primarily driven by a relative greater MI risk

among the T2D CAD patients. Younger (< 60 years) T2D CAD patients had a relatively higher risk of CV

outcome, with MI as the main risk driver, compared to the older T2D CAD patients and T2D patients

without a history of CAD. Risk factors other than age and conventional cardiovascular comorbidities

seemed to be more important in T2D CAD patients below the age of 60 years at diagnosis, compared

to older T2D CAD patients.

Our findings suggest that intense risk prevention should be the key strategy in the management of

T2D patients, especially for younger patients, including both encouragement for positive lifestyle

changes and prescription of secondary preventive drug therapy with antiplatelet therapy and statins.

Ideally, to reduce CV outcome and progression of T2D, younger T2D patients with CAD should be

offered participation in guideline-recommended risk reduction programs.

Page 12 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



LIST OF ABBREVIATIONS

T2D: type 2 diabetes

CAD: coronary artery disease

MI: myocardial infarction

GLD: glucose-lowering drug

CV: cardiovascular

CVD: cardiovascular disease

NPR: Swedish National Patient Register

SPDR: Swedish Prescribed Drug Register

ICD-10: International Statistical Classification of Diseases and Related Health Problems, 10th Revision

ID: personal identification

HR: hazard ratio

CI: confidence interval

FU: follow-up

Page 13 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



DECLARATIONS

Ethics approval and consent to participate

The study was approved by the Stockholm regional ethics committee (registration number

2013/2206-31). The linkage of registers data was approved and performed by the Swedish National

Board of Health and Welfare. Patients do not need to give consent for use of public register data in

Sweden.

Consent for publication

All authors read and approved the final manuscript. All authors gave consent to publish these data.

Availability of data and material

The dataset supporting the conclusions of this article can be available upon request.

Competing interests

PH, JB and KAS are employed by AstraZeneca.

MT is employed at Statisticon for which AstraZeneca is a client.

TJ, DL, BS, DE and MJ report no conflict of interest relevant to this article.

Funding

The study was sponsored by AstraZeneca.

Authors' contributions

Data collection was performed by JB. Statistical analysis was conducted by TJ, PH and MT.

Analysis, interpretation and drafting of the manuscript was conducted by TJ and PH and in

cooperation with the other authors. All authors approved the manuscript before submission.

Page 14 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



Acknowledgments

The authors would like to thank Urban Olsson, Statisticon AB, for data management.

Page 15 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



REFERENCES

1. IDF DIABETES ATLAS - 8TH EDITION 2017 [Available from:

http://diabetesatlas.org/resources/2017-atlas.html.

2. Vazzana N, Ranalli P, Cuccurullo C, Davi G. Diabetes mellitus and thrombosis. Thromb Res.

2012;129(3):371-7.

3. Preis SR, Pencina MJ, Hwang SJ, D'Agostino RB, Sr., Savage PJ, Levy D, et al. Trends in

cardiovascular disease risk factors in individuals with and without diabetes mellitus in the

Framingham Heart Study. Circulation. 2009;120(3):212-20.

4. Bhatt DL, Eagle KA, Ohman EM, Hirsch AT, Goto S, Mahoney EM, et al. Comparative

determinants of 4-year cardiovascular event rates in stable outpatients at risk of or with

atherothrombosis. JAMA. 2010;304(12):1350-7.

5. Roger VL, Go AS, Lloyd-Jones DM, Adams RJ, Berry JD, Brown TM, et al. Heart disease and

stroke statistics--2011 update: a report from the American Heart Association. Circulation.

2011;123(4):e18-e209.

6. Haffner SM, Lehto S, Ronnemaa T, Pyorala K, Laakso M. 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. 1998;339(4):229-34.

7. Inzucchi SE, Bergenstal RM, Buse JB, Diamant M, Ferrannini E, Nauck M, et al. Management

of hyperglycemia in type 2 diabetes, 2015: a patient-centered approach: update to a position

statement of the American Diabetes Association and the European Association for the Study of

Diabetes. Diabetes Care. 2015;38(1):140-9.

8. International Diabetes Federation Guideline Development G. Global guideline for type 2

diabetes. Diabetes Res Clin Pract. 2014;104(1):1-52.

9. Giorda CB, Avogaro A, Maggini M, Lombardo F, Mannucci E, Turco S, et al. Recurrence of

cardiovascular events in patients with type 2 diabetes: epidemiology and risk factors. Diabetes Care.

2008;31(11):2154-9.

10. Jernberg T, Hasvold P, Henriksson M, Hjelm H, Thuresson M, Janzon M. Cardiovascular risk in

post-myocardial infarction patients: nationwide real world data demonstrate the importance of a

long-term perspective. Eur Heart J. 2015;36(19):1163-70.

11. Wallerstedt SM, Wettermark B, Hoffmann M. The First Decade with the Swedish Prescribed

Drug Register - A Systematic Review of the Output in the Scientific Literature. Basic Clin Pharmacol

Toxicol. 2016;119(5):464-9.

12. Cause of death Sweden 2017 [Available from:

http://www.socialstyrelsen.se/register/dodsorsaksregistret.

13. Ludvigsson JF, Andersson E, Ekbom A, Feychting M, Kim JL, Reuterwall C, et al. External

review and validation of the Swedish national inpatient register. BMC Public Health. 2011;11:450.

14. Norhammar A, Bodegard J, Nystrom T, Thuresson M, Eriksson JW, Nathanson D. Incidence,

prevalence and mortality of type 2 diabetes requiring glucose-lowering treatment, and associated

risks of cardiovascular complications: a nationwide study in Sweden, 2006-2013. Diabetologia.

2016;59(8):1692-701.

15. Rawshani A, Rawshani A, Franzen S, Eliasson B, Svensson AM, Miftaraj M, et al. Mortality and

Cardiovascular Disease in Type 1 and Type 2 Diabetes. N Engl J Med. 2017;376(15):1407-18.

16. Rawshani A, Rawshani A, Franzen S, Sattar N, Eliasson B, Svensson AM, et al. Risk Factors,

Mortality, and Cardiovascular Outcomes in Patients with Type 2 Diabetes. N Engl J Med.

2018;379(7):633-44.

17. Huo L, Magliano DJ, Ranciere F, Harding JL, Nanayakkara N, Shaw JE, et al. Impact of age at

diagnosis and duration of type 2 diabetes on mortality in Australia 1997-2011. Diabetologia. 2018.

18. Steinarsson AO, Rawshani A, Gudbjornsdottir S, Franzen S, Svensson AM, Sattar N. Short-

term progression of cardiometabolic risk factors in relation to age at type 2 diabetes diagnosis: a

longitudinal observational study of 100,606 individuals from the Swedish National Diabetes Register.

Diabetologia. 2018;61(3):599-606.

Page 16 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



19. Agardh E, Allebeck P, Hallqvist J, Moradi T, Sidorchuk A. Type 2 diabetes incidence and socio-

economic position: a systematic review and meta-analysis. Int J Epidemiol. 2011;40(3):804-18.

Page 17 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



Table 1. Baseline demographic and clinical characteristics for T2D patients without CAD, T2D CAD

patients without MI, and in T2D CAD patients with a history of MI

No CAD

n=221 226

CAD without MI

n=16 294

CAD with MI

n=22 550

Total

n=260 070

Age (years, mean, SD) 61.4 (13.8) 70.9 (10.5) 70.3 (11.1) 62.8 (13.8)

Age group

<55, n (%) 64 987 (29.4) 1005 (6.2) 1879 (8.3) 67 871 (26.1)

55-64, n (%) 60 785 (27.5) 3480 (21.4) 4944 (21.9) 69 209 (26.6)

65-74, n (%) 57 619 (26.0) 5671 (34.8) 7484 (33.2) 70 774 (27.2)

75-84, n (%) 29 105 (13.2) 4467 (27.4) 5763 (25.6) 39 335 (15.1)

85+, n (%) 8730 (3.9) 1671 (10.3) 2480 (11.0) 12 881 (5.0)

Female sex, n (%) 99389 (44.9) 6472 (39.7) 6632 (29.4) 112493 (43.3)

Myocardial infarction, n (%) 0 (0.0) 0 (0.0) 22 550 (100.0) 22 550 (8.7)

Unstable angina, n (%) 0 (0.0) 3286 (20.2) 5085 (22.5) 8371 (3.2)

Angina pectoris, n (%) 0 (0.0) 8014 (49.2) 6545 (29.0) 14 559 (5.6)

Stroke, n (%) 11 080 (5.0) 1549 (9.5) 2504 (11.1) 15 133 (5.8)

Ischemic stroke, n (%) 9730 (4.4) 1434 (8.8) 2324 (10.3) 13 488 (5.2)

Atrial fibrillation, n (%) 14 070 (6.4) 3530 (21.7) 4854 (21.5) 22 454 (8.6)

Heart failure, n (%) 7612 (3.4) 3025 (18.6) 6386 (28.3) 17 023 (6.5)

Prescribed drugs at first dispensing of glucose lowering drug

Anti-platelets, n (%) 37 214 (16.8) 11 670 (71.6) 18 081 (80.2) 66 965 (25.7)

- Clopidogrel, n (%) 1781 (0.8) 1312 (8.1) 3716 (16.5) 6809 (2.6)

- Low-dose ASA, n (%) 35 797 (16.2) 11 248 (69.0) 17 407 (77.2) 64 452 (24.8)

Anti-coagulants, n (%) 10 636 (4.8) 2376 (14.6) 2945 (13.1) 15 957 (6.1)

Statins, n (%) 58 288 (26.3) 11 146 (68.4) 17 160 (76.1) 86

594 (33.3)

Anti-hypertensives, n (%) 122 861 (55.5) 14 962 (91.8) 20 805 (92.3) 158 628 (61.0)

- Beta-blockers, n (%) 61 174 (27.7) 11 774 (72.3) 18 098 (80.3) 91 046 (35.0)

- ACEIs, n (%) 47 838 (21.6) 5681 (34.9) 10 674 (47.3) 64 193 (24.7)

- ARBs, n (%) 36 103 (16.3) 4263 (26.2) 5597 (24.8) 45 963 (17.7)

- Ca-blockers, n (%) 43 338 (19.6) 5629 (34.5) 6412 (28.4) 55 379 (21.3)

- Diuretics, n (%) 52 610 (23.8) 7228 (44.4) 9860 (43.7) 69 698 (26.8)

Glucose lowering drugs

- Insulin, n (%) 25 181 (11.4) 1917 (11.8) 3384 (15.0) 30 482 (11.7)

- Metformin, n (%) 185 387 (83.8) 12 995 (79.8) 17 211 (76.3) 215 593 (82.9)

- SU, n (%) 13 049 (5.9) 1311 (8.0) 1899 (8.4) 16 259 (6.3)

- DPP-4is, n (%) 1784 (0.8) 181 (1.1) 327 (1.5) 2292 (0.9)

- Metiglinides, n (%) 2488 (1.1) 274 (1.7) 430 (1.9) 3192 (1.2)

SD, standard deviation; ASA, acetylsalicylic acid; ACEI, angiotensin-converting enzyme inhibitor; ARB,

angiotensin receptor blocker; Ca-blockers, calcium-channel blocker; SU, sulfonylurea; DPP-4is,

pipeptidyl peptidase-4 inhibitors

Page 18 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



Table 2. Baseline demographic and clinical characteristics for the different diabetes age categories at

index date

<55

n=67 871

55-64

n=69 209

65-74

n=70 774

75-84

n=39 335

85+

n=12 881

Total

n=260 070

Age (years, mean, SD) 45.0 (7.7) 59.9 (2.9) 69.1 (2.8) 78.9 (2.8) 88.4 (3.1) 62.8 (13.8)

Female sex, n (%) 28 783

(42.4)

26 249

(37.9)

29 606

(41.8)

19 959

(50.7)

7896

(61.3)

11 2493

(43.3)

Myocardial infarction, n (%) 1879 (2.8) 4944 (7.1) 7484

(10.6)

5763

(14.7)

2480

(19.3)

22 550

(8.7)

Years since last MI

(mean, SD) -2.2 (2.9) -3.1 (3.4) -3.5 (3.7) -3.1 (3.5) -2.7 (3.3) -3.1 (3.5)

Angina pectoris, n (%) 686 (1.0) 2669 (3.9) 4799 (6.8) 4370

(11.1)

2035

(15.8)

14 559

(5.6)

Stroke, n (%) 911 (1.3) 2602 (3.8) 4706 (6.6) 4594

(11.7)

2320

(18.0)

15 133

(5.8)

Heart failure, n (%) 974 (1.4) 2450 (3.5) 4780 (6.8) 5488

(14.0)

3331

(25.9)

17 023

(6.5)

Atrial fibrillation, n (%) 816 (1.2) 2961 (4.3) 7162

(10.1)

7661

(19.5)

3854

(29.9)

22 454

(8.6)

Major bleedings, n (%) 923 (1.4) 1614 (2.3) 2264 (3.2) 1924 (4.9) 1002

(7.8) 7727 (3.0)

Chronic renal dysfunction,

n (%) 539 (0.8) 537 (0.8) 515 (0.7) 256 (0.7) 64 (0.5) 1911 (0.7)

Chronic obstructive

pulmonary disease, n (%) 520 (0.8) 1743 (2.5) 3381 (4.8) 2476 (6.3) 701 (5.4) 8821 (3.4)

Malign cancer, n (%) 2117 (3.1) 5550 (8.0) 11 119

(15.7)

8556

(21.8)

3114

(24.2)

30 456

(11.7)

Presecribed drugs at first dispensing of glucose lowering drug

Anti-platelets, n (%) 5060 (7.5) 14 830

(21.4)

23 256

(32.9)

17 037

(43.3)

6782

(52.7)

66 965

(25.7)

- Low-dose ASA, n (%) 4885 (7.2) 14 332

(20.7)

22 393

(31.6)

16 323

(41.5)

6519

(50.6)

64 452

(24.8)

Anticoagulants, n (%) 739 (1.1) 2184 (3.2) 5464 (7.7) 5680

(14.4)

1890

(14.7)

15 957

(6.1)

Statins, n (%) 10 513

(15.5)

23 785

(34.4)

31 437

(44.4)

17 528

(44.6)

3331

(25.9)

86 594

(33.3)

Anti-hypertensives, n (%) 21 840

(32.2)

41 789

(60.4)

51 740

(73.1)

32 144

(81.7)

11 115

(86.3)

158 628

(61.0)

Page 19 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



<55

n=67 871

55-64

n=69 209

65-74

n=70 774

75-84

n=39 335

85+

n=12 881

Total

n=260 070

- Beta-blockers, n (%) 10 460

(15.4)

22 918

(33.1)

30 686

(43.4)

20 197

(51.3)

6785

(52.7)

91 046

(35.0)

- ACEIs, n (%) 9547 (14.1) 17 434

(25.2)

21 055

(29.7)

12 430

(31.6)

3727

(28.9)

64 193

(24.7)

- ARBs, n (%) 5907 (8.7) 13 136

(19.0)

16 180

(22.9)

8680

(22.1)

2060

(16.0)

45 963

(17.7)

- Ca-blockers, n (%) 6448 (9.5) 14 246

(20.6)

19 171

(27.1)

11 925

(30.3)

3589

(27.9)

55 379

(21.3)

- Diuretics, n (%) 6971 (10.3) 14 717

(21.3)

21 819

(30.8)

17 970

(45.7)

8221

(63.8)

69 698

(26.8)


- Insulin, n (%) 7541 (11.1) 6649 (9.6) 7063

(10.0)

5623

(14.3)

3606

(28.0)

30482

(11.7)

- Metformin, n (%) 59 342

(87.4)

60 819

(87.9)

60 507

(85.5)

28 845

(73.3)

6080

(47.2)

215 593

(82.9)

- SU, n (%) 2449 (3.6) 2813 (4.1) 3774 (5.3) 4473

(11.4)

2750

(21.3)

16 259

(6.3)

- DPP-4ies, n (%) 531 (0.8) 535 (0.8) 572 (0.8) 471 (1.2) 183 (1.4) 2292 (0.9)

- Metiglinides, n (%) 580 (0.9) 594 (0.9) 739 (1.0) 817 (2.1) 462 (3.6) 3192 (1.2)

SD, standard deviation; ASA, acetylsalicylic acid; ACEI, angiotensin-converting enzyme inhibitor; ARB,

angiotensin receptor blocker; Ca-blockers, calcium-channel blocker; SU, sulfonylurea; DPP-4is,

pipeptidyl peptidase-4 inhibitors

Page 20 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



Figure titles and legends

Figure 1. Adjusted probability plots* for time to the first occurrence of the CV composite outcome,

and the components myocardial infarction, stroke and cardiovascular death separately among T2D

patients without CAD, T2D CAD patients without MI, and in T2D CAD patients with a history of MI

*) Predicted for the “average” 63-year-old patient (mean age of the study population, and “mean” of

the following risk factors: sex, diabetes duration, atrial fibrillation, and heart failure

Figure 2. Spline plots for risk of composite CV outcome by age and CAD severity. Reference is mean

age (63 years) in the T2D no CAD population

No CAD: T2D patients without CAD: T2D patients without any previous diagnosis of CAD

CAD wo MI: T2D patients with CAD without prior MI: T2D patients with previous diagnosis of CAD

without MI

CAD with MI: T2D patients with CAD with prior MI: defined as T2D patients with a history of MI, and

unstable angina or angina pectoris

Figure 3. Spline plots for MI risk by age and CAD severity. Reference is mean age (63 years) of the

T2D no CAD population



without MI



Page 21 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



Adjusted probability plots* for time to the first occurrence of the CV composite outcome, and the components myocardial infarction, stroke and cardiovascular death separately among T2D patients without

CAD, T2D CAD patients without MI, and in T2D CAD patients with a history of MI

*) Predicted for the “average” 63-year-old patient (mean age of the study population, and “mean” of the following risk factors: sex, diabetes duration, atrial fibrillation, and heart failure

177x152mm (250 x 250 DPI)

Page 22 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



Figure 2. Spline plots for risk of composite CV outcome by age and CAD severity. Reference is mean age (63 years) in the T2D no CAD population

No CAD: T2D patients without CAD: T2D patients without any previous diagnosis of CAD CAD wo MI: T2D patients with CAD without prior MI: T2D patients with previous diagnosis of CAD without MI

CAD with MI: T2D patients with CAD with prior MI: defined as T2D patients with a history of MI, and unstable angina or angina pectoris

177x152mm (250 x 250 DPI)

Page 23 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



Figure 3. Spline plots for MI risk by age and CAD severity. Reference is mean age (63 years) of the T2D no CAD population

No CAD: T2D patients without CAD: T2D patients without any previous diagnosis of CAD CAD wo MI: T2D patients with CAD without prior MI: T2D patients with previous diagnosis of CAD without MI

CAD with MI: T2D patients with CAD with prior MI: defined as T2D patients with a history of MI, and unstable angina or angina pectoris

177x152mm (250 x 250 DPI)

Page 24 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



Supplementary data. Table 1 A

Cumulative incidence, CV composite outcome (myocardial infarction, stroke or cardiovascular death)

0.5 years 1 years 1.5 years 2 years 2.5 years 3 years

No CAD 1.09 (1.05-1.12) 1.84 (1.79-1.89) 2.55 (2.49-2.62) 3.29 (3.21-3.36) 4.01 (3.92-4.10) 4.78 (4.68-4.88)

CAD wo MI 1.33 (1.26-1.41) 2.26 (2.14-2.37) 3.13 (2.98-3.29) 4.03 (3.83-4.23) 4.91 (4.67-5.15) 5.85 (5.57-6.13)

CAD with MI 1.85 (1.76-1.93) 3.13 (2.99-3.26) 4.33 (4.16-4.51) 5.56 (5.34-5.78) 6.76 (6.50-7.02) 8.04 (7.73-8.34)

Adjusted for age, gender, prior heart failure, prior atrial fibrillation and prior stroke



without MI



Supplementary data. Table 1 B

Cumulative incidence, myocardial infarction


No CAD 0.34 (0.32-0.36) 0.61 (0.58-0.64) 0.87 (0.83-0.91) 1.13 (1.09-1.18) 1.40 (1.35-1.45) 1.66 (1.60-1.72)

CAD wo MI 0.64 (0.58-0.70) 1.14 (1.04-1.24) 1.63 (1.49-1.77) 2.12 (1.94-2.29) 2.61 (2.40-2.82) 3.09 (2.84-3.34)

CAD with MI 1.14 (1.05-1.22) 2.03 (1.89-2.16) 2.89 (2.71-3.06) 3.74 (3.52-3.97) 4.61 (4.33-4.87) 5.44 (5.13-5.76)




without MI



Supplementary data. Table 1 C

Cumulative incidence, stroke


No CAD 0.51 (0.48-0.54) 0.83 (0.80-0.87) 1.12 (1.08-1.17) 1.42 (1.37-1.47) 1.71 (1.65-1.76) 2.03 (1.97-2.10)

CAD wo MI 0.55 (0.50-0.60) 0.89 (0.81-0.97) 1.20 (1.10-1.30) 1.52 (1.39-1.65) 1.82 (1.67-1.98) 2.17 (1.99-2.35)

CAD with MI 0.57 (0.53-0.62) 0.93 (0.86-1.01) 1.26 (1.16-1.35) 1.59 (1.48-1.71) 1.91 (1.77-2.05) 2.28 (2.11-2.44)




without MI



Page 25 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



Supplementary data. Table 1 D

Cumulative incidence, cardiovascular death


No CAD 0.22 (0.20-0.23) 0.38 (0.36-0.40) 0.56 (0.53-0.58) 0.74 (0.70-0.77) 0.93 (0.89-0.97) 1.14 (1.09-1.19)

CAD wo MI 0.23 (0.21-0.26) 0.42 (0.38-0.45) 0.61 (0.56-0.65) 0.80 (0.74-0.87) 1.01 (0.93-1.09) 1.24 (1.15-1.34)

CAD with MI 0.32 (0.30-0.35) 0.57 (0.53-0.61) 0.83 (0.77-0.89) 1.10 (1.03-1.17) 1.39 (1.30-1.48) 1.70 (1.59-1.81)




without MI



Page 26 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



Supplementary data, Figure 1. Kaplan–Meier estimate of the risk of the composite CV outcome

during the first three years after initiating glucose lowering drug among T2D patients without CAD,

T2D CAD patients without MI, and in T2D CAD patients with a history of MI

Page 27 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



Supplementary data, Figure 2. Adjusted probability plots* (for a 62-year-old patient) of time to the

first occurrence of composite CV outcome, and the components myocardial infarction, stroke and

cardiovascular death separately among T2D patients without CAD, T2D CAD patients without MI, and

in T2D CAD patients with a history of MI

*) Predicted for the “average” 62-year-old patient (mean age of the sensitivity study population, and

“mean” of the following risk factors: sex, diabetes duration, atrial fibrillation, and heart failure

Page 28 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



Supplementary data. Figure 2. CV outcome risk and MI risk in different age categories (age when

initiating first time glucose lowering drug)

Page 29 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



1

STROBE Statement—Checklist of items that should be included in reports of cohort studies

Impact of coronary artery disease severity and age on risk of cardiovascular outcome in diabetes

patients: a nationwide observational study

Item

No Recommendation

Title and abstract 1 (a) Indicate the study’s design with a commonly used term in the title or the abstract

(b) Provide in the abstract an informative and balanced summary of what was done

and what was found

Page 1 and 2

Introduction

Background/rationale 2 Explain the scientific background and rationale for the investigation being reported

Page 4

Objectives 3 State specific objectives, including any prespecified hypotheses

Page 4

Methods

Study design 4 Present key elements of study design early in the paper

Page 5

Setting 5 Describe the setting, locations, and relevant dates, including periods of recruitment,

exposure, follow-up, and data collection

Page 5

Participants 6 (a) Give the eligibility criteria, and the sources and methods of selection of

participants. Describe methods of follow-up

Page 5

(b) For matched studies, give matching criteria and number of exposed and

unexposed

Page NA

Variables 7 Clearly define all outcomes, exposures, predictors, potential confounders, and effect

modifiers. Give diagnostic criteria, if applicable

Page 5

Data sources/

measurement

8* For each variable of interest, give sources of data and details of methods of

assessment (measurement). Describe comparability of assessment methods if there is

more than one group

Page 5

Bias 9 Describe any efforts to address potential sources of bias

Not applicabøe

Study size 10 Explain how the study size was arrived at

Not relevant

Quantitative variables 11 Explain how quantitative variables were handled in the analyses. If applicable,

describe which groupings were chosen and why

Not relevant

Statistical methods 12 (a) Describe all statistical methods, including those used to control for confounding

Page 6

(b) Describe any methods used to examine subgroups and interactions

Page 6

(c) Explain how missing data were addressed

Page 6 and 11

(d) If applicable, explain how loss to follow-up was addressed

Not applicable

Page 30 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



2

(e) Describe any sensitivity analyses

Not don in this study. Do in the “sister” clinical publications

Results

Participants 13* (a) Report numbers of individuals at each stage of study—eg numbers potentially

eligible, examined for eligibility, confirmed eligible, included in the study,

completing follow-up, and analysed

Page 7-8

(b) Give reasons for non-participation at each stage

(c) Consider use of a flow diagram

Not included

Descriptive data 14* (a) Give characteristics of study participants (eg demographic, clinical, social) and

information on exposures and potential confounders

Page 7

(b) Indicate number of participants with missing data for each variable of interest

Not applicable

(c) Summarise follow-up time (eg, average and total amount)

Outcome data 15* Report numbers of outcome events or summary measures over time

Page 7

Main results 16 (a) Give unadjusted estimates and, if applicable, confounder-adjusted estimates and

their precision (eg, 95% confidence interval). Make clear which confounders were

adjusted for and why they were included

(b) Report category boundaries when continuous variables were categorized

(c) If relevant, consider translating estimates of relative risk into absolute risk for a

meaningful time period

Page 8

Other analyses 17 Report other analyses done—eg analyses of subgroups and interactions, and

sensitivity analyses

See supplementary data

Discussion

Key results 18 Summarise key results with reference to study objectives

Page 9

Limitations 19 Discuss limitations of the study, taking into account sources of potential bias or

imprecision. Discuss both direction and magnitude of any potential bias

Page 11

Interpretation 20 Give a cautious overall interpretation of results considering objectives, limitations,

multiplicity of analyses, results from similar studies, and other relevant evidence

Page 12

Generalisability 21 Discuss the generalisability (external validity) of the study results

Page 12

Other information

Funding 22 Give the source of funding and the role of the funders for the present study and, if

applicable, for the original study on which the present article is based

Page 14

*Give information separately for exposed and unexposed groups.

Note: An Explanation and Elaboration article discusses each checklist item and gives methodological background and

published examples of transparent reporting. The STROBE checklist is best used in conjunction with this article (freely

Page 31 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



3

available on the Web sites of PLoS Medicine at http://www.plosmedicine.org/, Annals of Internal Medicine at

http://www.annals.org/, and Epidemiology at http://www.epidem.com/). Information on the STROBE Initiative is

available at http://www.strobe-statement.org.

Page 32 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from


For peer review onlyImpact of ischemic heart disease severity and age on risk of cardiovascular outcome in diabetes patients: a nationwide

observational study

Journal: BMJ Open

Manuscript ID bmjopen-2018-027199.R1


Date Submitted by the Author: 11-Feb-2019

Complete List of Authors: Jernberg, Tomas; Danderyd University Hospital, Karolinska InstitutetLindholm, Daniel ; Uppsala Clinical Research Center, Hasvold, Lars Pål; AstraZeneca Nordic, Medical departmentSvennblad, Bodil; Uppsala Clinical Research CenterBodegård, Johan; AstraZeneca Nordic, Medical departmentAndersson, Karolina; AstraZeneca R&DThuresson, Marcus; Statisticon, Erlinge, David; Lunds Universitet, Clinical scienceJanzon, Magnus; Linkopings universitet, Cardiology

<b>Primary Subject Heading</b>: Cardiovascular medicine

Secondary Subject Heading: Cardiovascular medicine



BMJ Open on N



jopen.bmj.com

/B

MJ O


pril 2019. Dow

nloaded from



1

Impact of ischemic heart disease severity and age on risk of cardiovascular outcome in

diabetes patients: a nationwide observational study

Tomas Jernberga), Daniel Lindholmb,c), Pål Hasvoldd), Bodil Svennbladc), Johan Bodegårdd), Karolina

Andersson Sundelle), Marcus Thuressonf), David Erlingeg), Magnus Janzonh)

a) Department of clinical sciences, Danderyd University Hospital, Karolinska Institutet, Stockholm, Sweden

b) Department of Medical Sciences, Cardiology, Uppsala University, Uppsala, Swedenc) Uppsala Clinical Research Center, Uppsala, Swedend) AstraZeneca Nordic-Baltic, Södertälje, Swedene) AstraZeneca R&D, Gothenburg, Swedenf) Statisticon AB, Uppsala, Swedeng) Lund University, Lund, Swedenh) Department of Cardiology and Department of Medical and Health Sciences, Linköping University,

Linköping, Sweden




Word count: 4119

Page 1 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



2

ABSTRACT

Background: The aim was to compare short-term cardiovascular outcome in type 2 diabetes (T2D)

patients without ischemic heart disease (IHD), with IHD but no prior myocardial infarction (MI), and

those with prior MI; and assess the impact on risk of age when initiating first-time glucose-lowering

drug (GLD).

Methods: This was a cohort study linking morbidity, mortality, and medication data from Swedish

national registries. Predicted cumulative incidence for the CV outcome (MI, stroke and CV-mortality)

was estimated. Furthermore, a Cox model was developed where age at GLD start and CV risk was

modeled.

Results: The study included 260 070 first-time users of GLD T2D patients during 2007-2016. Of these,

221 226 (85%) had no IHD, 16 294 (6%) had stable IHD - prior MI, and 22 550 (9%) had IHD + MI. T2D

patients without IHD had a lower risk of CV outcome compared with the IHD populations (-/+ prior

MI), (3-year incidence 4.78% vs. 5.85% and 8.04%). The difference in CV outcome was primarily

driven by a relative greater MI risk among the IHD patients. For T2D patients without IHD an almost

linear association between age at start of GLD and relative risk was observed, whereas in IHD

patients, the younger (< 60 years) patients had a relative greater risk compared with older patients.

Conclusions: T2D patients without IHD had a lower risk of the CV outcome compared to the T2D

populations with IHD, primarily driven by a greater risk of MI. For T2D patients without IHD an almost





Page 2 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



3


Strengths

The study was conducted in a nationwide national cohort including all patients who collected a first-time GLD at the pharmacy during the observational period, limiting the potential problems with selection bias.

Limitations

Access to clinical data describing the T2D duration (prior to GLD therapy), extent and severity of T2D (blood glucose, HbA1C, weight, smoking pattern and kidney function) which have an impact on the risk was not available.

The study is reliant on ICD-10 codes for morbidity data and therefore, the possibility of coding errors cannot be ruled out.

Another limitation of our study was the lack of available data on socio-economic status which is known to affect risk in T2D patients.

Page 3 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



4

BACKGROUND





of ischemic heart disease (IHD) and increased risk of cerebrovascular disease (2). Diabetes is


with and without established cardiovascular disease (CVD) (3-7). This to such extent that CVD-related


There are several theories for why T2D patients have increased CV risk, a common view is that

increased hyperglycemic stress may worsen the prognosis in T2D patients and that T2D patients may

be exposed to plaque instability due to the pro-inflammatory/oxidative properties of their plaque (9,

10). This may result in a more severe IHD among T2D patients, as it is well-known that T2D patients

often have coronary multivessel disease and often more severe CV outcome than other IHD patients

(11).


growth of the population with a history of IHD (12). Combined with the increased incidence of T2D, it

is likely that the T2D patient population with a history of IHD will increase in the coming decades and

thus, increased knowledge of the short-term cardiovascular event pattern is important.


IHD with that of an atherothrombotic disease demonstrated as previous MI in diabetic patients.




The primary objective of the present study was to compare short-term (3 year) cardiovascular

outcome in T2D patients without IHD, with IHD but no prior MI, and those with prior MI. Secondary

objectives were to assess the impact of age when initiating first-time GLD. For these purposes, we

used a highly representative nationwide sample of all T2D patients initiating first-time GLD in Sweden

over 7 years.

Page 4 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



5

METHOD

Study design





(SPDR) (13), and the Swedish Cause of Death Register (14). The Swedish NPR has had mandatory

registration since 1984 and covers more than 99% of all somatic (including surgery) and psychiatric

hospital admissions and discharges (15). The SPDR contains data on all prescribed medications

collected from pharmacies in Sweden since 2005. Linkage of patient-level data was performed by the

Swedish National Board of Health and Welfare utilizing the unique personal identification (ID)

numbers, mandatory for every citizen in Sweden, and thereafter replaced by a study ID for further

data processing. The study was approved by the Stockholm regional ethics committee (registration

number 2013/2206-31).

Study population


January 1 2007-December 31 2016. The index date was defined as the date of the first collected

prescription of a GLD from the pharmacy during the observation period. To be defined as a first-time

user the patient should not have collected any prescriptions for GLD prior to the index date. Patient

characteristics at baseline were established using hospitalization (ICD-10 diagnoses codes) and drug

utilization data from national registers from 1987 onwards. Patients with collection of GLD from the

pharmacy before the study period were excluded.

Three study populations were defined based on patients’ clinical characteristics when collecting their

first-time GLD was from the pharmacy (baseline/index).

1. T2D patients without IHD: first-time GLD treated T2D patients without any previous diagnosis

of IHD (defined as a history of myocardial infarction (MI), unstable angina, or stable angina

pectoris) (ICD-10: I20-25).

2. T2D patients with IHD without prior MI: first-time GLD treated T2D patients with previous

diagnosis of IHD without MI (defined as a history of stable or unstable angina pectoris (ICD-

10: I25), but no myocardial infarction).

Page 5 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



6

3. T2D patients with IHD with prior MI: defined as first-time GLD treated T2D patients with a

history of MI (ICD-10: I21–24).

For the secondary objective, the three study populations were further stratified into the following

age categories (based on age at first GLD collection GLD): < 55 years, 55-64 years, 65-74 years, 75-84

years, >85 years.

Outcomes



diagnosis).

Sensitivity and additional analyses

In order to test the impact of prior stroke, outcomes in T2D patients without IHD, MI or stroke, were

compared with T2D patients with IHD without prior MI or stroke, and T2D patients with IHD with MI

without stroke.

All-cause mortality in the three study populations was described.

Patient and public involvement

This was a cohort study using nationwide register data. No patients were involved in the design of

the study. The presented results will hopefully lead to an increased awareness of cardiovascular risk

for various T2D populations and thus lead to improved management of patients with T2D.






for age, sex, T2D duration, atrial fibrillation, and heart failure. The results are illustrated using

Page 6 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



7


Meier plots.






Page 7 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



8

RESULTS


be included. Of these, 221 226 (85%) were included in T2D population without IHD, 16 294 (6%) in

the T2D IHD population without MI, and 22 550 (9%) in the T2D IHD population with a history of MI.



T2D patients without IHD were younger (61.4 years), more often female (45%), and had a lower

incidence of stroke (5%), atrial fibrillation (6%) and heart failure (3%) compared to the two T2D IHD

populations (Table 1). The IHD population without a history of MI had a mean age of 70.9 vs. 70.3

years in IHD patients with a history of MI. The two IHD populations included 40% vs. 29% women,


were minor differences in GLD therapy among the three study populations, with the majority of

patients treated with metformin (>76%), sulfonylurea (>6%) or insulin (>11%). A greater proportion

of the T2D IHD population with a history of MI were treated with insulin (15.0%). More patients in

the two IHD populations were treated with statins (68% vs. 26%), anti-platelets (71% vs. 17%) and

anti-hypertensives (92% vs. 56%) than the T2D patients without IHD.

The cumulative rate of the primary composite CV outcome (MI, stroke, or cardiovascular death) was

5.69% in T2D population without IHD, 13.08% in the T2D IHD population without MI, 18.83% in the

T2D IHD population with a history of MI during the 3 years follow-up (Supplementary data, Figure 1).

T2D patients without IHD had a lower risk of the CV outcome compared to the T2D populations with

IHD (3 year adjusted cumulative incidence for a 63-years-old patient (mean age of the study


risk seen for T2D IHD patients with no prior MI vs patients without IHD was primarily driven by MI (3


The risk for stroke and cardiovascular death followed the same pattern as did the primary CV

outcome and MI (Supplementary data, Table 1c and 1d). The 3-years cumulative mortality rate was

8.08%, 14.77% and 18.68% in the three study groups (Supplementary data, Figure 2).


a consistent pattern to the main results, T2D patients without IHD or stroke, had a lower risk of CV

outcome compared to T2D IHD patients without MI and stroke and T2D patients with MI without



primarily driven by MI.

Page 8 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



9

The baseline characteristics for patients were stratified by age at first pharmacy GLD collection

irrespective of IHD status during the observation period (Table 2). The proportion of women was

lower, in the younger categories (<55 years), 42% compared to 61% among patients older than 85

years. The proportion of patients with cardiovascular comorbidities was greater in older patients, the









study populations in relation to age when collecting first GLD from pharmacy. Patients without IHD

showed an almost linear association between age and relative risk of CV outcome respectively

(Figures 2 and 3). Presence of IHD was associated with a relatively higher increase in relative risk of

CV outcome in younger (< 60 years) patients Figure 2 and 3). In patients with IHD, with or without

previous MI, the relative CV and MI risk did not increase with age in patients younger than 65 years.

In patients older 65 years, there was an increased relative risk of CV outcome and risk of MI with

increasing age (Figure 2 and 3). In all age categories, the MI risk was the main risk contributor, both

for T2D patients without IHD or MI and for the two IHD populations.

Page 9 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



10

DISCUSSION



findings: First, T2D patients without IHD had a lower risk of the CV outcome compared to the T2D

populations with IHD. The difference in CV outcome was primarily driven by a greater risk of MI in

T2D patients with IHD. Second, for T2D patients without IHD an almost linear association between

age at start of GLD therapy and relative risk of CV outcome and MI was observed, whereas in IHD


compared with older patients. Also, in T2D patients with IHD, there was no increase in relative CV

risk with increasing age until the age of 65, with a linear increase in risk thereafter.


has significant increased risk for cardiovascular events like myocardial infarction, heart failure, atrial





related to the presence and severity of IHD disease. Even the T2D patients without prior MI have a




patients with a history of IHD should be carefully monitored and managed with a long-term


diagnosed with T2D, particularly in those considered to be at high risk of ischemic events. The

importance of risk factor control has recently been shown in a study from the Swedish National

Diabetes Registry where patients with T2DM who had appropriate risk factor control had little or no

excess risk of CV events as compared with the general population (18). The data included in our study

is recent, but as management of T2D is rapidly progressing, the observed GLD therapy, with a high

proportion of patients treated with insulin. T2D drug therapy with the novel drug options, including

glucagon-like peptide 1 (GLP1) analogues and sodium-glucose cotransporter 2 (SGLT2) inhibitors,

have in addition to improving glucose control, been shown to be associated with lower rates of

cardiovascular events and mortality compared to older glucose lowering therapies (19-22).

Page 10 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



11

Only a limited number of studies have examined the association between age at start of first-time

GLD and CV risk. A recent study from Australia showed a that a younger age at T2D diagnosis was




populations, comparable to our findings, focusing on the relatively short-term CV risk after start of

GLD therapy, and not the full life-time risk in relation to age at start of GLD therapy. Our findings are

based on time from first collection of GLD from the pharmacy and not the total duration of T2D prior

to index (start of GLD therapy). We can thus only speculate that the IHD populations in general are

older, have a higher morbidity burden ,and consequently would have seen more frequently by health

care professionals and thus might have a relative earlier detection of their T2D than the non-IHD

patients (24). The Swedish National Diabetes Registry (NDR) has reported that time from T2D

diagnosis to GLD therapy initiation has been shorten during 2002 to 2011 (25).

For patients without a history of IHD at start of GLD therapy, an almost linear association between

age and CV and MI risk was observed. In contrast, a relatively higher increase in risk for CV outcome

and MI events was observed in younger patients with established IHD when starting with GLD,

compared to older patients with IHD. Furthermore, a low proportion of the patients were treated

with statins and anti-platelets, ranging from 16% statins and 8% anti-platelets for patients below 55

years, to 45% statins and 43% anti-platelets for patients 75-84 years.

Recent data from Sweden complementing our data examined clinical characteristics in age stratified



compared with individuals who develop diabetes later in life (26). This more severe metabolic

dysregulation could be associated with accelerated atherosclerosis. This should be amenable to

primary prevention both by lifestyle changes and medical treatment. However, they also found that a

low proportion of these young patients received blood pressure lowering drugs, statins and anti-

platelet drugs (26). This is in line with our findings that younger patients in general where not

receiving adequate treatment with blood pressure lowering drugs, statins and antiplatelet drugs.

Page 11 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



12

The combination of these metabolic risk factors, start of GLD therapy and presences of IHD in young

age predicts high risk of major adverse cardiovascular events, especially risk for MI. These findings

further highlight the importance of providing a close monitoring of younger T2D patients with

established IHD. That said, there may be a possible uncertainty regarding clinical responsibility for

drug treatment initiation between specialist and primary care for these patients, which may hamper

a thorough T2D patient management.


patients collecting a first-time GLD from pharmacy during the observational period, which limits the


have access to clinical data describing the duration of T2D prior to GLD start, extent and severity of

T2D (blood glucose, HbA1C, weight, smoking pattern and kidney function) which have an impact on

the risk (27, 28). However, complementary data on these clinical variables from Sweden were

recently published, showing that an unfavorable metabolic profile of the younger patients could be a

part of the explanation for our findings (26). Second, our study is reliant on ICD-10 codes for

morbidity data and therefore, the possibility of coding errors cannot be ruled out. However, previous

data show that coding is correct in 98% of Swedish NPR entries (15). Another limitation of our study

was the lack of available data on socio-economic status which is known to affect risk in T2D and CAD

patients (29).

Page 12 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



13

CONCLUSION

In conclusion, T2D patients without IHD had a lower risk of the CV outcome compared to the T2D

populations with IHD. The difference in CV outcome was primarily driven by a relative greater MI risk

among the T2D IHD patients. Younger (< 60 years) T2D IHD patients had a relatively higher risk of CV

outcome, with MI as the main risk driver, compared to the older T2D IHD patients and T2D patients

without a history of IHD. Risk factors other than age and conventional cardiovascular comorbidities

seemed to be more important in T2D IHD patients below the age of 60 years at diagnosis, compared

to older T2D IHD patients.




Ideally, to reduce CV outcome and progression of T2D, younger T2D patients with IHD should be


Page 13 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



14



IHD: ischemic heart disease



CV: cardiovascular






HR: hazard ratio


FU: follow-up

GLP1: glucagon-like peptide 1

SGLT2: sodium-glucose cotransporter 2 inhibitors

Page 14 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



15

DECLARATIONS





Sweden.





Competing interests

PH, DL, JB and KAS are employed by AstraZeneca.


TJ, BS, DE and MJ report no conflict of interest relevant to this article.

DL was at the time this research was performed employed by Uppsala University, but has since been

employed by AstraZeneca.

Funding




Analysis, interpretation and drafting of the manuscript was conducted by TJ and PH in cooperation

with DL, BS, JB, KAS, MT, DE and MJ. All authors approved the manuscript before submission.

Page 15 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



16

Acknowledgments


Page 16 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



17

REFERENCES

1. IDF DIABETES ATLAS - 8TH EDITION 2017 [Available from: http://diabetesatlas.org/resources/2017-atlas.html.2. Vazzana N, Ranalli P, Cuccurullo C, Davi G. Diabetes mellitus and thrombosis. Thromb Res. 2012;129(3):371-7.3. Preis SR, Pencina MJ, Hwang SJ, D'Agostino RB, Sr., Savage PJ, Levy D, et al. Trends in cardiovascular disease risk factors in individuals with and without diabetes mellitus in the Framingham Heart Study. Circulation. 2009;120(3):212-20.4. Bhatt DL, Eagle KA, Ohman EM, Hirsch AT, Goto S, Mahoney EM, et al. Comparative determinants of 4-year cardiovascular event rates in stable outpatients at risk of or with atherothrombosis. JAMA. 2010;304(12):1350-7.5. Inzucchi SE, Bergenstal RM, Buse JB, Diamant M, Ferrannini E, Nauck M, et al. Management of hyperglycemia in type 2 diabetes, 2015: a patient-centered approach: update to a position statement of the American Diabetes Association and the European Association for the Study of Diabetes. Diabetes Care. 2015;38(1):140-9.6. International Diabetes Federation Guideline Development G. Global guideline for type 2 diabetes. Diabetes Res Clin Pract. 2014;104(1):1-52.7. Giorda CB, Avogaro A, Maggini M, Lombardo F, Mannucci E, Turco S, et al. Recurrence of cardiovascular events in patients with type 2 diabetes: epidemiology and risk factors. Diabetes Care. 2008;31(11):2154-9.8. Roger VL, Go AS, Lloyd-Jones DM, Adams RJ, Berry JD, Brown TM, et al. Heart disease and stroke statistics--2011 update: a report from the American Heart Association. Circulation. 2011;123(4):e18-e209.9. Sardu C, Barbieri M, Balestrieri ML, Siniscalchi M, Paolisso P, Calabro P, et al. Thrombus aspiration in hyperglycemic ST-elevation myocardial infarction (STEMI) patients: clinical outcomes at 1-year follow-up. Cardiovasc Diabetol. 2018;17(1):152.10. Marfella R, Rizzo MR, Siniscalchi M, Paolisso P, Barbieri M, Sardu C, et al. Peri-procedural tight glycemic control during early percutaneous coronary intervention up-regulates endothelial progenitor cell level and differentiation during acute ST-elevation myocardial infarction: effects on myocardial salvage. Int J Cardiol. 2013;168(4):3954-62.11. Marfella R, Sardu C, Balestrieri ML, Siniscalchi M, Minicucci F, Signoriello G, et al. Effects of incretin treatment on cardiovascular outcomes in diabetic STEMI-patients with culprit obstructive and multivessel non obstructive-coronary-stenosis. Diabetol Metab Syndr. 2018;10:1.12. Jernberg T, Hasvold P, Henriksson M, Hjelm H, Thuresson M, Janzon M. Cardiovascular risk in post-myocardial infarction patients: nationwide real world data demonstrate the importance of a long-term perspective. Eur Heart J. 2015;36(19):1163-70.13. Wallerstedt SM, Wettermark B, Hoffmann M. The First Decade with the Swedish Prescribed Drug Register - A Systematic Review of the Output in the Scientific Literature. Basic Clin Pharmacol Toxicol. 2016;119(5):464-9.14. Bourdin A, Molinari N, Vachier I, Pahus L, Suehs C, Chanez P. Mortality: a neglected outcome in OCS-treated severe asthma. Eur Respir J. 2017;50(5).15. Ludvigsson JF, Andersson E, Ekbom A, Feychting M, Kim JL, Reuterwall C, et al. External review and validation of the Swedish national inpatient register. BMC Public Health. 2011;11:450.16. Norhammar A, Bodegard J, Nystrom T, Thuresson M, Eriksson JW, Nathanson D. Incidence, prevalence and mortality of type 2 diabetes requiring glucose-lowering treatment, and associated risks of cardiovascular complications: a nationwide study in Sweden, 2006-2013. Diabetologia. 2016;59(8):1692-701.17. Rawshani A, Rawshani A, Franzen S, Eliasson B, Svensson AM, Miftaraj M, et al. Mortality and Cardiovascular Disease in Type 1 and Type 2 Diabetes. N Engl J Med. 2017;376(15):1407-18.

Page 17 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from

http://diabetesatlas.org/resources/2017-atlas.html



18

18. Rawshani A, Rawshani A, Franzen S, Sattar N, Eliasson B, Svensson AM, et al. Risk Factors, Mortality, and Cardiovascular Outcomes in Patients with Type 2 Diabetes. N Engl J Med. 2018;379(7):633-44.19. Bethel MA, Patel RA, Merrill P, Lokhnygina Y, Buse JB, Mentz RJ, et al. Cardiovascular outcomes with glucagon-like peptide-1 receptor agonists in patients with type 2 diabetes: a meta-analysis. Lancet Diabetes Endocrinol. 2018;6(2):105-13.20. Marso SP, Daniels GH, Brown-Frandsen K, Kristensen P, Mann JF, Nauck MA, et al. Liraglutide and Cardiovascular Outcomes in Type 2 Diabetes. N Engl J Med. 2016;375(4):311-22.21. Zinman B, Wanner C, Lachin JM, Fitchett D, Bluhmki E, Hantel S, et al. Empagliflozin, Cardiovascular Outcomes, and Mortality in Type 2 Diabetes. N Engl J Med. 2015;373(22):2117-28.22. Marso SP, Bain SC, Consoli A, Eliaschewitz FG, Jodar E, Leiter LA, et al. Semaglutide and Cardiovascular Outcomes in Patients with Type 2 Diabetes. N Engl J Med. 2016;375(19):1834-44.23. Huo L, Magliano DJ, Ranciere F, Harding JL, Nanayakkara N, Shaw JE, et al. Impact of age at diagnosis and duration of type 2 diabetes on mortality in Australia 1997-2011. Diabetologia. 2018.24. Hasvold LP, Bodegard J, Thuresson M, Stalhammar J, Hammar N, Sundstrom J, et al. Diabetes and CVD risk during angiotensin-converting enzyme inhibitor or angiotensin II receptor blocker treatment in hypertension: a study of 15,990 patients. J Hum Hypertens. 2014;28(11):663-9.25. Guðbjörnsdóttir S EB, Cederholm J, Zethelius B, Svensson A, Samuelsson P. Swedish National Diabetes Register

Annual report 2013 2013 [cited 2019 Feb 05]. Available from: https://www.ndr.nu/pdfs/Annual_Report_NDR_2013.pdf.26. Steinarsson AO, Rawshani A, Gudbjornsdottir S, Franzen S, Svensson AM, Sattar N. Short-term progression of cardiometabolic risk factors in relation to age at type 2 diabetes diagnosis: a longitudinal observational study of 100,606 individuals from the Swedish National Diabetes Register. Diabetologia. 2018;61(3):599-606.27. Gambardella J, Sardu C, Sacra C, Del Giudice C, Santulli G. Quit smoking to outsmart atherogenesis: Molecular mechanisms underlying clinical evidence. Atherosclerosis. 2017;257:242-5.28. Sardu C, Pieretti G, D'Onofrio N, Ciccarelli F, Paolisso P, Passavanti MB, et al. Inflammatory Cytokines and SIRT1 Levels in Subcutaneous Abdominal Fat: Relationship With Cardiac Performance in Overweight Pre-diabetics Patients. Front Physiol. 2018;9:1030.29. Agardh E, Allebeck P, Hallqvist J, Moradi T, Sidorchuk A. Type 2 diabetes incidence and socio-economic position: a systematic review and meta-analysis. Int J Epidemiol. 2011;40(3):804-18.

Page 18 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from

https://www.ndr.nu/pdfs/Annual_Report_NDR_2013.pdf



19

Table 1. Baseline demographic and clinical characteristics for first time GLD treated T2D patients without IHD, T2D IHD patients without MI, and in T2D IHD patients with a history of MI

No IHDn=221 226

IHD without MIn=16 294

IHD with MIn=22 550

Totaln=260 070

Age (years, mean, SD) 61.4 (13.8) 70.9 (10.5) 70.3 (11.1) 62.8 (13.8)Age group <55, n (%) 64 987 (29.4) 1005 (6.2) 1879 (8.3) 67 871 (26.1)

55-64, n (%) 60 785 (27.5) 3480 (21.4) 4944 (21.9) 69 209 (26.6)

65-74, n (%) 57 619 (26.0) 5671 (34.8) 7484 (33.2) 70 774 (27.2)

75-84, n (%) 29 105 (13.2) 4467 (27.4) 5763 (25.6) 39 335 (15.1) 85+, n (%) 8730 (3.9) 1671 (10.3) 2480 (11.0) 12 881 (5.0)Female sex, n (%) 99389 (44.9) 6472 (39.7) 6632 (29.4) 112493 (43.3)Myocardial infarction, n (%) 0 (0.0) 0 (0.0) 22 550 (100.0) 22 550 (8.7)Unstable angina, n (%) 0 (0.0) 3286 (20.2) 5085 (22.5) 8371 (3.2)Angina pectoris, n (%) 0 (0.0) 8014 (49.2) 6545 (29.0) 14 559 (5.6)Stroke, n (%) 11 080 (5.0) 1549 (9.5) 2504 (11.1) 15 133 (5.8)Ischemic stroke, n (%) 9730 (4.4) 1434 (8.8) 2324 (10.3) 13 488 (5.2)Atrial fibrillation, n (%) 14 070 (6.4) 3530 (21.7) 4854 (21.5) 22 454 (8.6)Heart failure, n (%) 7612 (3.4) 3025 (18.6) 6386 (28.3) 17 023 (6.5)

Prescribed drugs at first collection of glucose lowering drugAnti-platelets, n (%) 37 214 (16.8) 11 670 (71.6) 18 081 (80.2) 66 965 (25.7)- Clopidogrel, n (%) 1781 (0.8) 1312 (8.1) 3716 (16.5) 6809 (2.6)- Low-dose ASA, n (%) 35 797 (16.2) 11 248 (69.0) 17 407 (77.2) 64 452 (24.8)Anti-coagulants, n (%) 10 636 (4.8) 2376 (14.6) 2945 (13.1) 15 957 (6.1)

Statins, n (%) 58 288 (26.3) 11 146 (68.4) 17 160 (76.1)86

594 (33.3)Anti-hypertensives, n (%) 122 861 (55.5) 14 962 (91.8) 20 805 (92.3) 158 628 (61.0)- Beta-blockers, n (%) 61 174 (27.7) 11 774 (72.3) 18 098 (80.3) 91 046 (35.0)- ACEIs, n (%) 47 838 (21.6) 5681 (34.9) 10 674 (47.3) 64 193 (24.7)- ARBs, n (%) 36 103 (16.3) 4263 (26.2) 5597 (24.8) 45 963 (17.7)- Ca-blockers, n (%) 43 338 (19.6) 5629 (34.5) 6412 (28.4) 55 379 (21.3)- Diuretics, n (%) 52 610 (23.8) 7228 (44.4) 9860 (43.7) 69 698 (26.8)Glucose lowering drugs- Insulin, n (%) 25 181 (11.4) 1917 (11.8) 3384 (15.0) 30 482 (11.7)- Metformin, n (%) 185 387 (83.8) 12 995 (79.8) 17 211 (76.3) 215 593 (82.9)- SU, n (%) 13 049 (5.9) 1311 (8.0) 1899 (8.4) 16 259 (6.3)- DPP-4is, n (%) 1784 (0.8) 181 (1.1) 327 (1.5) 2292 (0.9)- Metiglinides, n (%) 2488 (1.1) 274 (1.7) 430 (1.9) 3192 (1.2)

GLD, glucose lowering drug; SD, standard deviation; ASA, acetylsalicylic acid; ACEI, angiotensin-converting enzyme inhibitor; ARB, angiotensin receptor blocker; Ca-blockers, calcium-channel blocker; SU, sulfonylurea; DPP-4is, pipeptidyl peptidase-4 inhibitors

Page 19 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



20

Table 2. Baseline demographic and clinical characteristics for the different diabetes age categories at index date

<55n=67 871

55-64 n=69 209

65-74 n=70 774

75-84 n=39 335

85+ n=12 881

Total n=260 070

Age (years, mean, SD) 45.0 (7.7) 59.9 (2.9) 69.1 (2.8) 78.9 (2.8)88.4 (3.1)

62.8 (13.8)

Female sex, n (%)28 783 (42.4)

26 249 (37.9)

29 606 (41.8)

19 959 (50.7)

7896 (61.3)

11 2493 (43.3)

Myocardial infarction, n (%) 1879 (2.8) 4944 (7.1)7484 (10.6)

5763 (14.7)

2480 (19.3)

22 550 (8.7)

Years since last MI

(mean, SD)-2.2 (2.9) -3.1 (3.4) -3.5 (3.7) -3.1 (3.5) -2.7 (3.3) -3.1 (3.5)

Angina pectoris, n (%) 686 (1.0) 2669 (3.9) 4799 (6.8)4370 (11.1)

2035 (15.8)

14 559 (5.6)

Stroke, n (%) 911 (1.3) 2602 (3.8) 4706 (6.6)4594 (11.7)

2320 (18.0)

15 133 (5.8)

Heart failure, n (%) 974 (1.4) 2450 (3.5) 4780 (6.8)5488 (14.0)

3331 (25.9)

17 023 (6.5)

Atrial fibrillation, n (%) 816 (1.2) 2961 (4.3)7162 (10.1)

7661 (19.5)

3854 (29.9)

22 454 (8.6)

Major bleedings, n (%) 923 (1.4) 1614 (2.3) 2264 (3.2) 1924 (4.9)1002 (7.8)

7727 (3.0)


n (%)539 (0.8) 537 (0.8) 515 (0.7) 256 (0.7) 64 (0.5) 1911 (0.7)

Chronic obstructive pulmonary disease, n (%)

520 (0.8) 1743 (2.5) 3381 (4.8) 2476 (6.3) 701 (5.4) 8821 (3.4)

Malign cancer, n (%) 2117 (3.1) 5550 (8.0)11 119 (15.7)

8556 (21.8)

3114 (24.2)

30 456 (11.7)

Dispensed drugs when collecting first glucose lowering drug at pharmacy

Anti-platelets, n (%) 5060 (7.5)14 830 (21.4)

23 256 (32.9)

17 037 (43.3)

6782 (52.7)

66 965 (25.7)

- Low-dose ASA, n (%) 4885 (7.2)14 332 (20.7)

22 393 (31.6)

16 323 (41.5)

6519 (50.6)

64 452 (24.8)

Anticoagulants, n (%) 739 (1.1) 2184 (3.2) 5464 (7.7)5680 (14.4)

1890 (14.7)

15 957 (6.1)

Statins, n (%)10 513 (15.5)

23 785 (34.4)

31 437 (44.4)

17 528 (44.6)

3331 (25.9)

86 594 (33.3)

Anti-hypertensives, n (%)21 840 (32.2)

41 789 (60.4)

51 740 (73.1)

32 144 (81.7)

11 115 (86.3)

158 628 (61.0)

Page 20 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



21

<55n=67 871

55-64 n=69 209

65-74 n=70 774

75-84 n=39 335

85+ n=12 881

Total n=260 070

- Beta-blockers, n (%)10 460 (15.4)

22 918 (33.1)

30 686 (43.4)

20 197 (51.3)

6785 (52.7)

91 046 (35.0)

- ACEIs, n (%) 9547 (14.1)17 434 (25.2)

21 055 (29.7)

12 430 (31.6)

3727 (28.9)

64 193 (24.7)

- ARBs, n (%) 5907 (8.7)13 136 (19.0)

16 180 (22.9)

8680 (22.1)

2060 (16.0)

45 963 (17.7)

- Ca-blockers, n (%) 6448 (9.5)14 246 (20.6)

19 171 (27.1)

11 925 (30.3)

3589 (27.9)

55 379 (21.3)

- Diuretics, n (%) 6971 (10.3)14 717 (21.3)

21 819 (30.8)

17 970 (45.7)

8221 (63.8)

69 698 (26.8)


- Insulin, n (%) 7541 (11.1) 6649 (9.6)7063 (10.0)

5623 (14.3)

3606 (28.0)

30482 (11.7)

- Metformin, n (%)59 342 (87.4)

60 819 (87.9)

60 507 (85.5)

28 845 (73.3)

6080 (47.2)

215 593 (82.9)

- SU, n (%) 2449 (3.6) 2813 (4.1) 3774 (5.3)4473 (11.4)

2750 (21.3)

16 259 (6.3)

- DPP-4ies, n (%) 531 (0.8) 535 (0.8) 572 (0.8) 471 (1.2) 183 (1.4) 2292 (0.9)

- Metiglinides, n (%) 580 (0.9) 594 (0.9) 739 (1.0) 817 (2.1) 462 (3.6) 3192 (1.2)

SD, standard deviation; ASA, acetylsalicylic acid; ACEI, angiotensin-converting enzyme inhibitor; ARB, angiotensin receptor blocker; Ca-blockers, calcium-channel blocker; SU, sulfonylurea; DPP-4is, pipeptidyl peptidase-4 inhibitors

Page 21 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



22


Figure 1. Adjusted probability plots* for time to the first occurrence of the composite CV composite outcome, and the components myocardial infarction, stroke and cardiovascular death separately among T2D patients without IHD, T2D IHD patients without MI, and in T2D IHD patients with a history of MI

*) Predicted for the “average” 63-year-old patient (mean age of the study population, and “mean” of the following risk factors: sex, diabetes duration, atrial fibrillation, and heart failure)

Figure 2. Spline plots for risk of composite CV outcome by age and IHD severity. Reference is mean age (63 years) in the T2D no IHD population

No IHD: T2D patients without IHD: T2D patients without any previous diagnosis of IHD

IHD wo MI: T2D patients with IHD without prior MI: T2D patients with previous diagnosis of IHD without MI

IHD with MI: T2D patients with IHD with prior MI: defined as T2D patients with a history of MI, and unstable angina or angina pectoris

Figure 3. Spline plots for MI risk by age and IHD severity. Reference is mean age (63 years) of the T2D no IHD population




Page 22 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



Adjusted probability plots* for time to the first occurrence of the composite CV composite outcome, and the components myocardial infarction, stroke and cardiovascular death separately among T2D patients without

IHD, T2D IHD patients without MI, and in T2D IHD patients with a history of MI


104x90mm (300 x 300 DPI)

Page 23 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



Spline plots for risk of composite CV outcome by age and IHD severity. Reference is mean age (63 years) in the T2D no IHD population

No IHD: T2D patients without IHD: T2D patients without any previous diagnosis of IHD IHD wo MI: T2D patients with IHD without prior MI: T2D patients with previous diagnosis of IHD without MI IHD with MI: T2D patients with IHD with prior MI: defined as T2D patients with a history of MI, and unstable

angina or angina pectoris

104x89mm (300 x 300 DPI)

Page 24 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



Spline plots for MI risk by age and IHD severity. Reference is mean age (63 years) of the T2D no IHD population



104x90mm (300 x 300 DPI)

Page 25 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



Supplementary data.

Table 1 A



No IHD 1.09 (1.05-1.12) 1.84 (1.79-1.89) 2.55 (2.49-2.62) 3.29 (3.21-3.36) 4.01 (3.92-4.10) 4.78 (4.68-4.88)

IHD wo MI 1.33 (1.26-1.41) 2.26 (2.14-2.37) 3.13 (2.98-3.29) 4.03 (3.83-4.23) 4.91 (4.67-5.15) 5.85 (5.57-6.13)

IHD with MI 1.85 (1.76-1.93) 3.13 (2.99-3.26) 4.33 (4.16-4.51) 5.56 (5.34-5.78) 6.76 (6.50-7.02) 8.04 (7.73-8.34)



IHD wo MI: T2D patients with IHD without prior MI: T2D patients with previous diagnosis of IHD

without MI

IHD with MI: T2D patients with IHD with prior MI: defined as T2D patients with a history of MI, and


Table 1 B



No IHD 0.34 (0.32-0.36) 0.61 (0.58-0.64) 0.87 (0.83-0.91) 1.13 (1.09-1.18) 1.40 (1.35-1.45) 1.66 (1.60-1.72)

IHD wo MI 0.64 (0.58-0.70) 1.14 (1.04-1.24) 1.63 (1.49-1.77) 2.12 (1.94-2.29) 2.61 (2.40-2.82) 3.09 (2.84-3.34)

IHD with MI 1.14 (1.05-1.22) 2.03 (1.89-2.16) 2.89 (2.71-3.06) 3.74 (3.52-3.97) 4.61 (4.33-4.87) 5.44 (5.13-5.76)




without MI



Table 1 C



No IHD 0.51 (0.48-0.54) 0.83 (0.80-0.87) 1.12 (1.08-1.17) 1.42 (1.37-1.47) 1.71 (1.65-1.76) 2.03 (1.97-2.10)

IHD wo MI 0.55 (0.50-0.60) 0.89 (0.81-0.97) 1.20 (1.10-1.30) 1.52 (1.39-1.65) 1.82 (1.67-1.98) 2.17 (1.99-2.35)

IHD with MI 0.57 (0.53-0.62) 0.93 (0.86-1.01) 1.26 (1.16-1.35) 1.59 (1.48-1.71) 1.91 (1.77-2.05) 2.28 (2.11-2.44)




without MI

Page 26 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from





Table 1 D



No IHD 0.22 (0.20-0.23) 0.38 (0.36-0.40) 0.56 (0.53-0.58) 0.74 (0.70-0.77) 0.93 (0.89-0.97) 1.14 (1.09-1.19)

IHD wo MI 0.23 (0.21-0.26) 0.42 (0.38-0.45) 0.61 (0.56-0.65) 0.80 (0.74-0.87) 1.01 (0.93-1.09) 1.24 (1.15-1.34)

IHD with MI 0.32 (0.30-0.35) 0.57 (0.53-0.61) 0.83 (0.77-0.89) 1.10 (1.03-1.17) 1.39 (1.30-1.48) 1.70 (1.59-1.81)




without MI



Page 27 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



Figure 1. Kaplan–Meier estimate of the risk of the composite CV outcome (CV-death/MI/stroke)

during the first three years after collecting first-time GDL from pharmacy among T2D patients

without IHD, T2D IHD patients without MI, and in T2D IHD patients with a history of MI

Page 28 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



Figure 2. Kaplan–Meier estimate of the risk of all-cause mortality during the first three years after

collecting first-time GDL from pharmacy among T2D patients without IHD, T2D IHD patients without

MI, and in T2D IHD patients with a history of MI

Page 29 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



1




Item

No Recommendation



and what was found

Page 1 and 2

Introduction


Page 4


Page 4

Methods


Page 5



Page 5



Page 5


unexposed

Page NA



Page 5

Data sources/

measurement



more than one group

Page 5


Not applicabøe


Not relevant



Not relevant


Page 6


Page 6


Page 6 and 11


Not applicable

Page 30 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



2



Results




Page 7-8



Not included



Page 7


Not applicable



Page 7







Page 8




Discussion


Page 9



Page 11



Page 12


Page 12

Other information



Page 14




Page 31 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



3




Page 32 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from


For peer review onlyImpact of ischemic heart disease severity and age on risk of

cardiovascular outcome in diabetes patients in Sweden: a nationwide observational study

Journal: BMJ Open

Manuscript ID bmjopen-2018-027199.R2


Date Submitted by the Author: 25-Feb-2019

Complete List of Authors: Jernberg, Tomas; Danderyd University Hospital, Karolinska InstitutetLindholm, Daniel ; Uppsala Clinical Research Center, Hasvold, Lars Pål; AstraZeneca Nordic, Medical departmentSvennblad, Bodil; Uppsala Clinical Research CenterBodegård, Johan; AstraZeneca Nordic, Medical departmentAndersson, Karolina; AstraZeneca R&DThuresson, Marcus; Statisticon, Erlinge, David; Lunds Universitet, Clinical scienceJanzon, Magnus; Linkopings universitet, Cardiology

<b>Primary Subject Heading</b>: Cardiovascular medicine

Secondary Subject Heading: Cardiovascular medicine



BMJ Open on N



jopen.bmj.com

/B

MJ O


pril 2019. Dow

nloaded from



1

Impact of ischemic heart disease severity and age on risk of cardiovascular outcome in

diabetes patients in Sweden: a nationwide observational study

Tomas Jernberga), Daniel Lindholmb,c), Pål Hasvoldd), Bodil Svennbladc), Johan Bodegårdd), Karolina

Andersson Sundelle), Marcus Thuressonf), David Erlingeg), Magnus Janzonh)

a) Department of clinical sciences, Danderyd University Hospital, Karolinska Institutet, Stockholm, Sweden

b) Department of Medical Sciences, Cardiology, Uppsala University, Uppsala, Swedenc) Uppsala Clinical Research Center, Uppsala, Swedend) AstraZeneca Nordic-Baltic, Södertälje, Swedene) AstraZeneca R&D, Gothenburg, Swedenf) Statisticon AB, Uppsala, Swedeng) Lund University, Lund, Swedenh) Department of Cardiology and Department of Medical and Health Sciences, Linköping University,

Linköping, Sweden




Word count: 4016

Page 1 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



2

ABSTRACT

Objectives To compare short-term cardiovascular outcome in type 2 diabetes (T2D) patients without

ischemic heart disease (IHD), with IHD but no prior myocardial infarction (MI), and those with prior

MI; and assess the impact on risk of age when initiating first-time glucose-lowering drug (GLD).

Design Cohort study linking morbidity, mortality, and medication data from Swedish national

registries.

Setting First-time users of GLD during 2007-2016.

Outcomes Predicted cumulative incidence for the CV outcome (MI, stroke and CV-mortality) was

estimated. A Cox model was developed where age at GLD start and CV risk was modelled.

Results 260 070 first-time GLD users were included, 221 226 (85%) had no IHD, 16 294 (6%) had

stable IHD - prior MI, and 22 550 (9%) had IHD + MI. T2D patients without IHD had a lower risk of CV

outcome compared with the IHD populations (-/+ prior MI), (3-year incidence 4.78% vs. 5.85% and

8.04%). The difference in CV outcome was primarily driven by a relative greater MI risk among the

IHD patients. For T2D patients without IHD an almost linear association between age at start of GLD

and relative risk was observed, whereas in IHD patients, the younger (< 60 years) patients had a

relative greater risk compared with older patients.

Conclusions T2D patients without IHD had a lower risk of the CV outcome compared to the T2D

populations with IHD, primarily driven by a greater risk of MI. For T2D patients without IHD an almost





Page 2 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



3


Strengths

The study was conducted in a nationwide national cohort including all patients who collected

a first-time GLD at the pharmacy during the observational period, limiting the potential

problems with selection bias.

Limitations

Access to clinical data describing the T2D duration (prior to GLD therapy), extent and severity

of T2D (blood glucose, HbA1C, weight, smoking pattern and kidney function) which have an

impact on the risk was not available.

The study is reliant on ICD-10 codes for morbidity data and therefore, the possibility of

coding errors cannot be ruled out.

Another limitation of our study was the lack of available data on socio-economic status which

is known to affect risk in T2D patients.

Page 3 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



4

BACKGROUND





of ischemic heart disease (IHD) and increased risk of cerebrovascular disease (2). Diabetes is


with and without established cardiovascular disease (CVD) (3-7). This to such extent that CVD-related


There are several theories for why T2D patients have increased CV risk, a common view is that

increased hyperglycemic stress may worsen the prognosis in T2D patients and that T2D patients may

be exposed to plaque instability due to the pro-inflammatory/oxidative properties of their plaque (9,

10). This may result in a more severe IHD among T2D patients, as it is well-known that T2D patients

often have coronary multivessel disease and often more severe CV outcome than other IHD patients

(11).


growth of the population with a history of IHD (12). Combined with the increased incidence of T2D, it

is likely that the T2D patient population with a history of IHD will increase in the coming decades and

thus, increased knowledge of the short-term cardiovascular event pattern is important.


IHD with that of an atherothrombotic disease demonstrated as previous MI in diabetic patients.




The primary objective of the present study was to compare short-term (3 year) cardiovascular

outcome in T2D patients without IHD, with IHD but no prior MI, and those with prior MI. Secondary

objectives were to assess the impact of age when initiating first-time GLD. For these purposes, we

used a highly representative nationwide sample of all T2D patients initiating first-time GLD in Sweden

over 7 years.

Page 4 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



5

METHOD

Study design





(SPDR) (13), and the Swedish Cause of Death Register (14). The Swedish NPR has had mandatory

registration since 1984 and covers more than 99% of all somatic (including surgery) and psychiatric

hospital admissions and discharges (15). The SPDR contains data on all prescribed medications

collected from pharmacies in Sweden since 2005. Linkage of patient-level data was performed by the

Swedish National Board of Health and Welfare utilizing the unique personal identification (ID)

numbers, mandatory for every citizen in Sweden, and thereafter replaced by a study ID for further

data processing. The study was approved by the Stockholm regional ethics committee (registration

number 2013/2206-31).

Study population


January 1 2007-December 31 2016. The index date was defined as the date of the first collected

prescription of a GLD from the pharmacy during the observation period. To be defined as a first-time

user the patient should not have collected any prescriptions for GLD prior to the index date. Patient

characteristics at baseline were established using hospitalization (ICD-10 diagnoses codes) and drug

utilization data from national registers from 1987 onwards. Patients with collection of GLD from the

pharmacy before the study period were excluded.

Three study populations were defined based on patients’ clinical characteristics when collecting their

first-time GLD was from the pharmacy (baseline/index).

1. T2D patients without IHD: first-time GLD treated T2D patients without any previous diagnosis

of IHD (defined as a history of myocardial infarction (MI), unstable angina, or stable angina

pectoris) (ICD-10: I20-25).

2. T2D patients with IHD without prior MI: first-time GLD treated T2D patients with previous

diagnosis of IHD without MI (defined as a history of stable or unstable angina pectoris (ICD-

10: I25), but no myocardial infarction).

Page 5 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



6

3. T2D patients with IHD with prior MI: defined as first-time GLD treated T2D patients with a

history of MI (ICD-10: I21–24).

For the secondary objective, the three study populations were further stratified into the following

age categories (based on age at first GLD collection GLD): < 55 years, 55-64 years, 65-74 years, 75-84

years, >85 years.

Outcomes



diagnosis).

Sensitivity and additional analyses

In order to test the impact of prior stroke, outcomes in T2D patients without IHD, MI or stroke, were

compared with T2D patients with IHD without prior MI or stroke, and T2D patients with IHD with MI

without stroke.

All-cause mortality in the three study populations was described.

Patient and public involvement

This was a cohort study using nationwide register data. No patients were involved in the design of

the study. The presented results will hopefully lead to an increased awareness of cardiovascular risk

for various T2D populations and thus lead to improved management of patients with T2D.






for age, sex, T2D duration, atrial fibrillation, and heart failure. The results are illustrated using

Page 6 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



7


Meier plots.






Page 7 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



8

RESULTS


be included. Of these, 221 226 (85%) were included in T2D population without IHD, 16 294 (6%) in

the T2D IHD population without MI, and 22 550 (9%) in the T2D IHD population with a history of MI.



T2D patients without IHD were younger (61.4 years), more often female (45%), and had a lower

incidence of stroke (5%), atrial fibrillation (6%) and heart failure (3%) compared to the two T2D IHD

populations (Table 1). The IHD population without a history of MI had a mean age of 70.9 vs. 70.3

years in IHD patients with a history of MI. The two IHD populations included 40% vs. 29% women,


were minor differences in GLD therapy among the three study populations, with the majority of

patients treated with metformin (>76%), sulfonylurea (>6%) or insulin (>11%). A greater proportion

of the T2D IHD population with a history of MI were treated with insulin (15.0%). More patients in

the two IHD populations were treated with statins (68% vs. 26%), anti-platelets (71% vs. 17%) and

anti-hypertensives (92% vs. 56%) than the T2D patients without IHD.

The cumulative rate of the primary composite CV outcome (MI, stroke, or cardiovascular death) was

5.69% in T2D population without IHD, 13.08% in the T2D IHD population without MI, 18.83% in the

T2D IHD population with a history of MI during the 3 years follow-up (Supplementary data, Figure 1).

T2D patients without IHD had a lower risk of the CV outcome compared to the T2D populations with

IHD (3 year adjusted cumulative incidence for a 63-years-old patient (mean age of the study


risk seen for T2D IHD patients with no prior MI vs patients without IHD was primarily driven by MI (3


The risk for stroke and cardiovascular death followed the same pattern as did the primary CV

outcome and MI (Supplementary data, Table 1c and 1d). The 3-years cumulative mortality rate was

8.08%, 14.77% and 18.68% in the three study groups (Supplementary data, Figure 2).


a consistent pattern to the main results, T2D patients without IHD or stroke, had a lower risk of CV

outcome compared to T2D IHD patients without MI and stroke and T2D patients with MI without



primarily driven by MI.

Page 8 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



9

The baseline characteristics for patients were stratified by age at first pharmacy GLD collection

irrespective of IHD status during the observation period (Table 2). The proportion of women was

lower, in the younger categories (<55 years), 42% compared to 61% among patients older than 85

years. The proportion of patients with cardiovascular comorbidities was greater in older patients, the









study populations in relation to age when collecting first GLD from pharmacy. Patients without IHD

showed an almost linear association between age and relative risk of CV outcome respectively

(Figures 2 and 3). Presence of IHD was associated with a relatively higher increase in relative risk of

CV outcome in younger (< 60 years) patients Figure 2 and 3). In patients with IHD, with or without

previous MI, the relative CV and MI risk did not increase with age in patients younger than 65 years.

In patients older 65 years, there was an increased relative risk of CV outcome and risk of MI with

increasing age (Figure 2 and 3). In all age categories, the MI risk was the main risk contributor, both

for T2D patients without IHD or MI and for the two IHD populations.

Page 9 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



10

DISCUSSION



findings: First, T2D patients without IHD had a lower risk of the CV outcome compared to the T2D

populations with IHD. The difference in CV outcome was primarily driven by a greater risk of MI in

T2D patients with IHD. Second, for T2D patients without IHD an almost linear association between

age at start of GLD therapy and relative risk of CV outcome and MI was observed, whereas in IHD


compared with older patients. Also, in T2D patients with IHD, there was no increase in relative CV

risk with increasing age until the age of 65, with a linear increase in risk thereafter.


has significant increased risk for cardiovascular events like myocardial infarction, heart failure, atrial





related to the presence and severity of IHD disease. Even the T2D patients without prior MI have a




patients with a history of IHD should be carefully monitored and managed with a long-term


diagnosed with T2D, particularly in those considered to be at high risk of ischemic events. The

importance of risk factor control has recently been shown in a study from the Swedish National

Diabetes Registry where patients with T2DM who had appropriate risk factor control had little or no

excess risk of CV events as compared with the general population (18).

The data included in our study is recent, but as management of T2D is rapidly progressing, the

observed GLD therapy, with a high proportion of patients treated with insulin. T2D drug therapy with

the novel drug options, including glucagon-like peptide 1 (GLP1) analogues and sodium-glucose

cotransporter 2 (SGLT2) inhibitors, have in addition to improving glucose control, been shown to be

Page 10 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



11

associated with lower rates of cardiovascular events and mortality compared to older glucose

lowering therapies (19-22).

Only a limited number of studies have examined the association between age at start of first-time

GLD and CV risk. A recent study from Australia showed a that a younger age at T2D diagnosis was




populations, comparable to our findings, focusing on the relatively short-term CV risk after start of

GLD therapy, and not the full life-time risk in relation to age at start of GLD therapy. Our findings are

based on time from first collection of GLD from the pharmacy and not the total duration of T2D prior

to index (start of GLD therapy). We can thus only speculate that the IHD populations in general are

older, have a higher morbidity burden ,and consequently would have seen more frequently by health

care professionals and thus might have a relative earlier detection of their T2D than the non-IHD

patients (24). The Swedish National Diabetes Registry (NDR) has reported that time from T2D

diagnosis to GLD therapy initiation has been shorten during 2002 to 2011 (25).

For patients without a history of IHD at start of GLD therapy, an almost linear association between

age and CV and MI risk was observed. In contrast, a relatively higher increase in risk for CV outcome

and MI events was observed in younger patients with established IHD when starting with GLD,

compared to older patients with IHD. Furthermore, a low proportion of the patients were treated

with statins and anti-platelets, ranging from 16% statins and 8% anti-platelets for patients below 55

years, to 45% statins and 43% anti-platelets for patients 75-84 years.

Recent data from Sweden complementing our data examined clinical characteristics in age stratified



compared with individuals who develop diabetes later in life (26). They also found that a low

proportion of these young patients received blood pressure lowering drugs, statins and anti-platelet

drugs (26). This more severe metabolic dysregulation, and inadequately provided drug therapy seen,

may be associated with accelerated atherosclerosis, supported by the recent findings of a correlation

between younger age at T2D diagnosis, increasing number of variables not within target ranges, and

a higher relative risk of cardiovascular disease outcomes (18). The commonly seen low socio-

economic status among younger T2D patients might be a contributing factor to the noted increased

Page 11 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



12

risk, as low socio-economic status in itself increases risk among T2D patients (27, 28). This should be

amenable to primary prevention both by lifestyle changes and medical treatment.

The combination of metabolic risk factors, start of GLD therapy and presences of IHD in young age

predicts high risk of major adverse cardiovascular events, especially risk for MI. These findings

further highlight the importance of providing a closer monitoring of younger T2D patients with

established IHD. This is also further highlighted by our findings, where younger T2D patients in

general are not receiving adequate treatment with blood pressure lowering drugs, statins and

antiplatelet therapy. That said, there may be a possible uncertainty regarding clinical responsibility

for drug treatment initiation between specialist and primary care for these patients, which may

hamper a thorough T2D patient management.


patients collecting a first-time GLD from pharmacy during the observational period, which limits the


have access to clinical data describing the duration of T2D prior to GLD start, extent and severity of

T2D (blood glucose, HbA1C, weight, smoking pattern and kidney function) which have an impact on

the risk (29, 30). However, complementary data on these clinical variables from Sweden were

recently published, showing that an unfavorable metabolic profile of the younger patients could be a

part of the explanation for our findings (26). Second, our study is reliant on ICD-10 codes for

morbidity data and therefore, the possibility of coding errors cannot be ruled out. However, previous

data show that coding is correct in 98% of Swedish NPR entries (15). Another limitation of our study

was the lack of available data on socio-economic status which is known to affect risk in T2D and CAD

patients (31).

Page 12 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



13

CONCLUSION

In conclusion, T2D patients without IHD had a lower risk of the CV outcome compared to the T2D

populations with IHD. The difference in CV outcome was primarily driven by a relative greater MI risk

among the T2D IHD patients. Younger (< 60 years) T2D IHD patients had a relatively higher risk of CV

outcome, with MI as the main risk driver, compared to the older T2D IHD patients and T2D patients

without a history of IHD. Risk factors other than age and conventional cardiovascular comorbidities

seemed to be more important in T2D IHD patients below the age of 60 years at diagnosis, compared

to older T2D IHD patients.




Ideally, to reduce CV outcome and progression of T2D, younger T2D patients with IHD should be


Page 13 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



14



IHD: ischemic heart disease



CV: cardiovascular






HR: hazard ratio


FU: follow-up

GLP1: glucagon-like peptide 1

SGLT2: sodium-glucose cotransporter 2 inhibitors

Page 14 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



15

DECLARATIONS





Sweden.





Competing interests

PH, DL, JB and KAS are employed by AstraZeneca.


TJ, BS, DE and MJ report no conflict of interest relevant to this article.

DL was at the time this research was performed employed by Uppsala University, but has since been

employed by AstraZeneca.

Funding




Analysis, interpretation and drafting of the manuscript was conducted by TJ and PH in cooperation

with DL, BS, JB, KAS, MT, DE and MJ. All authors approved the manuscript before submission.

Page 15 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



16

Acknowledgments


Page 16 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



17

REFERENCES

1. IDF DIABETES ATLAS - 8TH EDITION 2017 [Available from: http://diabetesatlas.org/resources/2017-atlas.html.2. Vazzana N, Ranalli P, Cuccurullo C, Davi G. Diabetes mellitus and thrombosis. Thromb Res. 2012;129(3):371-7.3. Preis SR, Pencina MJ, Hwang SJ, D'Agostino RB, Sr., Savage PJ, Levy D, et al. Trends in cardiovascular disease risk factors in individuals with and without diabetes mellitus in the Framingham Heart Study. Circulation. 2009;120(3):212-20.4. Bhatt DL, Eagle KA, Ohman EM, Hirsch AT, Goto S, Mahoney EM, et al. Comparative determinants of 4-year cardiovascular event rates in stable outpatients at risk of or with atherothrombosis. JAMA. 2010;304(12):1350-7.5. Inzucchi SE, Bergenstal RM, Buse JB, Diamant M, Ferrannini E, Nauck M, et al. Management of hyperglycemia in type 2 diabetes, 2015: a patient-centered approach: update to a position statement of the American Diabetes Association and the European Association for the Study of Diabetes. Diabetes Care. 2015;38(1):140-9.6. International Diabetes Federation Guideline Development G. Global guideline for type 2 diabetes. Diabetes Res Clin Pract. 2014;104(1):1-52.7. Giorda CB, Avogaro A, Maggini M, Lombardo F, Mannucci E, Turco S, et al. Recurrence of cardiovascular events in patients with type 2 diabetes: epidemiology and risk factors. Diabetes Care. 2008;31(11):2154-9.8. Roger VL, Go AS, Lloyd-Jones DM, Adams RJ, Berry JD, Brown TM, et al. Heart disease and stroke statistics--2011 update: a report from the American Heart Association. Circulation. 2011;123(4):e18-e209.9. Sardu C, Barbieri M, Balestrieri ML, Siniscalchi M, Paolisso P, Calabro P, et al. Thrombus aspiration in hyperglycemic ST-elevation myocardial infarction (STEMI) patients: clinical outcomes at 1-year follow-up. Cardiovasc Diabetol. 2018;17(1):152.10. Marfella R, Rizzo MR, Siniscalchi M, Paolisso P, Barbieri M, Sardu C, et al. Peri-procedural tight glycemic control during early percutaneous coronary intervention up-regulates endothelial progenitor cell level and differentiation during acute ST-elevation myocardial infarction: effects on myocardial salvage. Int J Cardiol. 2013;168(4):3954-62.11. Marfella R, Sardu C, Balestrieri ML, Siniscalchi M, Minicucci F, Signoriello G, et al. Effects of incretin treatment on cardiovascular outcomes in diabetic STEMI-patients with culprit obstructive and multivessel non obstructive-coronary-stenosis. Diabetol Metab Syndr. 2018;10:1.12. Jernberg T, Hasvold P, Henriksson M, Hjelm H, Thuresson M, Janzon M. Cardiovascular risk in post-myocardial infarction patients: nationwide real world data demonstrate the importance of a long-term perspective. Eur Heart J. 2015;36(19):1163-70.13. Wallerstedt SM, Wettermark B, Hoffmann M. The First Decade with the Swedish Prescribed Drug Register - A Systematic Review of the Output in the Scientific Literature. Basic Clin Pharmacol Toxicol. 2016;119(5):464-9.14. Bourdin A, Molinari N, Vachier I, Pahus L, Suehs C, Chanez P. Mortality: a neglected outcome in OCS-treated severe asthma. Eur Respir J. 2017;50(5).15. Ludvigsson JF, Andersson E, Ekbom A, Feychting M, Kim JL, Reuterwall C, et al. External review and validation of the Swedish national inpatient register. BMC Public Health. 2011;11:450.16. Norhammar A, Bodegard J, Nystrom T, Thuresson M, Eriksson JW, Nathanson D. Incidence, prevalence and mortality of type 2 diabetes requiring glucose-lowering treatment, and associated risks of cardiovascular complications: a nationwide study in Sweden, 2006-2013. Diabetologia. 2016;59(8):1692-701.17. Rawshani A, Rawshani A, Franzen S, Eliasson B, Svensson AM, Miftaraj M, et al. Mortality and Cardiovascular Disease in Type 1 and Type 2 Diabetes. N Engl J Med. 2017;376(15):1407-18.

Page 17 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from

http://diabetesatlas.org/resources/2017-atlas.html



18

18. Rawshani A, Rawshani A, Franzen S, Sattar N, Eliasson B, Svensson AM, et al. Risk Factors, Mortality, and Cardiovascular Outcomes in Patients with Type 2 Diabetes. N Engl J Med. 2018;379(7):633-44.19. Bethel MA, Patel RA, Merrill P, Lokhnygina Y, Buse JB, Mentz RJ, et al. Cardiovascular outcomes with glucagon-like peptide-1 receptor agonists in patients with type 2 diabetes: a meta-analysis. Lancet Diabetes Endocrinol. 2018;6(2):105-13.20. Marso SP, Daniels GH, Brown-Frandsen K, Kristensen P, Mann JF, Nauck MA, et al. Liraglutide and Cardiovascular Outcomes in Type 2 Diabetes. N Engl J Med. 2016;375(4):311-22.21. Zinman B, Wanner C, Lachin JM, Fitchett D, Bluhmki E, Hantel S, et al. Empagliflozin, Cardiovascular Outcomes, and Mortality in Type 2 Diabetes. N Engl J Med. 2015;373(22):2117-28.22. Marso SP, Bain SC, Consoli A, Eliaschewitz FG, Jodar E, Leiter LA, et al. Semaglutide and Cardiovascular Outcomes in Patients with Type 2 Diabetes. N Engl J Med. 2016;375(19):1834-44.23. Huo L, Magliano DJ, Ranciere F, Harding JL, Nanayakkara N, Shaw JE, et al. Impact of age at diagnosis and duration of type 2 diabetes on mortality in Australia 1997-2011. Diabetologia. 2018.24. Hasvold LP, Bodegard J, Thuresson M, Stalhammar J, Hammar N, Sundstrom J, et al. Diabetes and CVD risk during angiotensin-converting enzyme inhibitor or angiotensin II receptor blocker treatment in hypertension: a study of 15,990 patients. J Hum Hypertens. 2014;28(11):663-9.25. Guðbjörnsdóttir S EB, Cederholm J, Zethelius B, Svensson A, Samuelsson P. Swedish National Diabetes Register

Annual report 2013 2013 [cited 2019 Feb 05]. Available from: https://www.ndr.nu/pdfs/Annual_Report_NDR_2013.pdf.26. Steinarsson AO, Rawshani A, Gudbjornsdottir S, Franzen S, Svensson AM, Sattar N. Short-term progression of cardiometabolic risk factors in relation to age at type 2 diabetes diagnosis: a longitudinal observational study of 100,606 individuals from the Swedish National Diabetes Register. Diabetologia. 2018;61(3):599-606.27. Rawshani A, Svensson AM, Zethelius B, Eliasson B, Rosengren A, Gudbjornsdottir S. Association Between Socioeconomic Status and Mortality, Cardiovascular Disease, and Cancer in Patients With Type 2 Diabetes. JAMA Intern Med. 2016;176(8):1146-54.28. Connolly V, Unwin N, Sherriff P, Bilous R, Kelly W. Diabetes prevalence and socioeconomic status: a population based study showing increased prevalence of type 2 diabetes mellitus in deprived areas. J Epidemiol Community Health. 2000;54(3):173-7.29. Gambardella J, Sardu C, Sacra C, Del Giudice C, Santulli G. Quit smoking to outsmart atherogenesis: Molecular mechanisms underlying clinical evidence. Atherosclerosis. 2017;257:242-5.30. Sardu C, Pieretti G, D'Onofrio N, Ciccarelli F, Paolisso P, Passavanti MB, et al. Inflammatory Cytokines and SIRT1 Levels in Subcutaneous Abdominal Fat: Relationship With Cardiac Performance in Overweight Pre-diabetics Patients. Front Physiol. 2018;9:1030.31. Agardh E, Allebeck P, Hallqvist J, Moradi T, Sidorchuk A. Type 2 diabetes incidence and socio-economic position: a systematic review and meta-analysis. Int J Epidemiol. 2011;40(3):804-18.

Page 18 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from

https://www.ndr.nu/pdfs/Annual_Report_NDR_2013.pdf



19

Table 1. Baseline demographic and clinical characteristics for first time GLD treated T2D patients without IHD, T2D IHD patients without MI, and in T2D IHD patients with a history of MI

No IHDn=221 226

IHD without MIn=16 294

IHD with MIn=22 550

Totaln=260 070

Age (years, mean, SD) 61.4 (13.8) 70.9 (10.5) 70.3 (11.1) 62.8 (13.8)Age group <55, n (%) 64 987 (29.4) 1005 (6.2) 1879 (8.3) 67 871 (26.1)

55-64, n (%) 60 785 (27.5) 3480 (21.4) 4944 (21.9) 69 209 (26.6)

65-74, n (%) 57 619 (26.0) 5671 (34.8) 7484 (33.2) 70 774 (27.2)

75-84, n (%) 29 105 (13.2) 4467 (27.4) 5763 (25.6) 39 335 (15.1) 85+, n (%) 8730 (3.9) 1671 (10.3) 2480 (11.0) 12 881 (5.0)Female sex, n (%) 99389 (44.9) 6472 (39.7) 6632 (29.4) 112493 (43.3)Myocardial infarction, n (%) 0 (0.0) 0 (0.0) 22 550 (100.0) 22 550 (8.7)Unstable angina, n (%) 0 (0.0) 3286 (20.2) 5085 (22.5) 8371 (3.2)Angina pectoris, n (%) 0 (0.0) 8014 (49.2) 6545 (29.0) 14 559 (5.6)Stroke, n (%) 11 080 (5.0) 1549 (9.5) 2504 (11.1) 15 133 (5.8)Ischemic stroke, n (%) 9730 (4.4) 1434 (8.8) 2324 (10.3) 13 488 (5.2)Atrial fibrillation, n (%) 14 070 (6.4) 3530 (21.7) 4854 (21.5) 22 454 (8.6)Heart failure, n (%) 7612 (3.4) 3025 (18.6) 6386 (28.3) 17 023 (6.5)

Prescribed drugs at first collection of glucose lowering drugAnti-platelets, n (%) 37 214 (16.8) 11 670 (71.6) 18 081 (80.2) 66 965 (25.7)- Clopidogrel, n (%) 1781 (0.8) 1312 (8.1) 3716 (16.5) 6809 (2.6)- Low-dose ASA, n (%) 35 797 (16.2) 11 248 (69.0) 17 407 (77.2) 64 452 (24.8)Anti-coagulants, n (%) 10 636 (4.8) 2376 (14.6) 2945 (13.1) 15 957 (6.1)

Statins, n (%) 58 288 (26.3) 11 146 (68.4) 17 160 (76.1)86

594 (33.3)Anti-hypertensives, n (%) 122 861 (55.5) 14 962 (91.8) 20 805 (92.3) 158 628 (61.0)- Beta-blockers, n (%) 61 174 (27.7) 11 774 (72.3) 18 098 (80.3) 91 046 (35.0)- ACEIs, n (%) 47 838 (21.6) 5681 (34.9) 10 674 (47.3) 64 193 (24.7)- ARBs, n (%) 36 103 (16.3) 4263 (26.2) 5597 (24.8) 45 963 (17.7)- Ca-blockers, n (%) 43 338 (19.6) 5629 (34.5) 6412 (28.4) 55 379 (21.3)- Diuretics, n (%) 52 610 (23.8) 7228 (44.4) 9860 (43.7) 69 698 (26.8)Glucose lowering drugs- Insulin, n (%) 25 181 (11.4) 1917 (11.8) 3384 (15.0) 30 482 (11.7)- Metformin, n (%) 185 387 (83.8) 12 995 (79.8) 17 211 (76.3) 215 593 (82.9)- SU, n (%) 13 049 (5.9) 1311 (8.0) 1899 (8.4) 16 259 (6.3)- DPP-4is, n (%) 1784 (0.8) 181 (1.1) 327 (1.5) 2292 (0.9)- Metiglinides, n (%) 2488 (1.1) 274 (1.7) 430 (1.9) 3192 (1.2)

GLD, glucose lowering drug; SD, standard deviation; ASA, acetylsalicylic acid; ACEI, angiotensin-converting enzyme inhibitor; ARB, angiotensin receptor blocker; Ca-blockers, calcium-channel blocker; SU, sulfonylurea; DPP-4is, pipeptidyl peptidase-4 inhibitors

Page 19 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



20

Table 2. Baseline demographic and clinical characteristics for the different diabetes age categories at index date

<55n=67 871

55-64 n=69 209

65-74 n=70 774

75-84 n=39 335

85+ n=12 881

Total n=260 070

Age (years, mean, SD) 45.0 (7.7) 59.9 (2.9) 69.1 (2.8) 78.9 (2.8)88.4 (3.1)

62.8 (13.8)

Female sex, n (%)28 783 (42.4)

26 249 (37.9)

29 606 (41.8)

19 959 (50.7)

7896 (61.3)

11 2493 (43.3)

Myocardial infarction, n (%) 1879 (2.8) 4944 (7.1)7484 (10.6)

5763 (14.7)

2480 (19.3)

22 550 (8.7)

Years since last MI

(mean, SD)-2.2 (2.9) -3.1 (3.4) -3.5 (3.7) -3.1 (3.5) -2.7 (3.3) -3.1 (3.5)

Angina pectoris, n (%) 686 (1.0) 2669 (3.9) 4799 (6.8)4370 (11.1)

2035 (15.8)

14 559 (5.6)

Stroke, n (%) 911 (1.3) 2602 (3.8) 4706 (6.6)4594 (11.7)

2320 (18.0)

15 133 (5.8)

Heart failure, n (%) 974 (1.4) 2450 (3.5) 4780 (6.8)5488 (14.0)

3331 (25.9)

17 023 (6.5)

Atrial fibrillation, n (%) 816 (1.2) 2961 (4.3)7162 (10.1)

7661 (19.5)

3854 (29.9)

22 454 (8.6)

Major bleedings, n (%) 923 (1.4) 1614 (2.3) 2264 (3.2) 1924 (4.9)1002 (7.8)

7727 (3.0)


n (%)539 (0.8) 537 (0.8) 515 (0.7) 256 (0.7) 64 (0.5) 1911 (0.7)

Chronic obstructive pulmonary disease, n (%)

520 (0.8) 1743 (2.5) 3381 (4.8) 2476 (6.3) 701 (5.4) 8821 (3.4)

Malign cancer, n (%) 2117 (3.1) 5550 (8.0)11 119 (15.7)

8556 (21.8)

3114 (24.2)

30 456 (11.7)

Dispensed drugs when collecting first glucose lowering drug at pharmacy

Anti-platelets, n (%) 5060 (7.5)14 830 (21.4)

23 256 (32.9)

17 037 (43.3)

6782 (52.7)

66 965 (25.7)

- Low-dose ASA, n (%) 4885 (7.2)14 332 (20.7)

22 393 (31.6)

16 323 (41.5)

6519 (50.6)

64 452 (24.8)

Anticoagulants, n (%) 739 (1.1) 2184 (3.2) 5464 (7.7)5680 (14.4)

1890 (14.7)

15 957 (6.1)

Statins, n (%)10 513 (15.5)

23 785 (34.4)

31 437 (44.4)

17 528 (44.6)

3331 (25.9)

86 594 (33.3)

Anti-hypertensives, n (%)21 840 (32.2)

41 789 (60.4)

51 740 (73.1)

32 144 (81.7)

11 115 (86.3)

158 628 (61.0)

Page 20 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



21

<55n=67 871

55-64 n=69 209

65-74 n=70 774

75-84 n=39 335

85+ n=12 881

Total n=260 070

- Beta-blockers, n (%)10 460 (15.4)

22 918 (33.1)

30 686 (43.4)

20 197 (51.3)

6785 (52.7)

91 046 (35.0)

- ACEIs, n (%) 9547 (14.1)17 434 (25.2)

21 055 (29.7)

12 430 (31.6)

3727 (28.9)

64 193 (24.7)

- ARBs, n (%) 5907 (8.7)13 136 (19.0)

16 180 (22.9)

8680 (22.1)

2060 (16.0)

45 963 (17.7)

- Ca-blockers, n (%) 6448 (9.5)14 246 (20.6)

19 171 (27.1)

11 925 (30.3)

3589 (27.9)

55 379 (21.3)

- Diuretics, n (%) 6971 (10.3)14 717 (21.3)

21 819 (30.8)

17 970 (45.7)

8221 (63.8)

69 698 (26.8)


- Insulin, n (%) 7541 (11.1) 6649 (9.6)7063 (10.0)

5623 (14.3)

3606 (28.0)

30482 (11.7)

- Metformin, n (%)59 342 (87.4)

60 819 (87.9)

60 507 (85.5)

28 845 (73.3)

6080 (47.2)

215 593 (82.9)

- SU, n (%) 2449 (3.6) 2813 (4.1) 3774 (5.3)4473 (11.4)

2750 (21.3)

16 259 (6.3)

- DPP-4ies, n (%) 531 (0.8) 535 (0.8) 572 (0.8) 471 (1.2) 183 (1.4) 2292 (0.9)

- Metiglinides, n (%) 580 (0.9) 594 (0.9) 739 (1.0) 817 (2.1) 462 (3.6) 3192 (1.2)

SD, standard deviation; ASA, acetylsalicylic acid; ACEI, angiotensin-converting enzyme inhibitor; ARB, angiotensin receptor blocker; Ca-blockers, calcium-channel blocker; SU, sulfonylurea; DPP-4is, pipeptidyl peptidase-4 inhibitors

Page 21 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



22


Figure 1. Adjusted probability plots* for time to the first occurrence of the composite CV composite outcome, and the components myocardial infarction, stroke and cardiovascular death separately among T2D patients without IHD, T2D IHD patients without MI, and in T2D IHD patients with a history of MI


Figure 2. Spline plots for risk of composite CV outcome by age and IHD severity. Reference is mean age (63 years) in the T2D no IHD population




Figure 3. Spline plots for MI risk by age and IHD severity. Reference is mean age (63 years) of the T2D no IHD population




Page 22 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



Adjusted probability plots* for time to the first occurrence of the composite CV composite outcome, and the components myocardial infarction, stroke and cardiovascular death separately among T2D patients without

IHD, T2D IHD patients without MI, and in T2D IHD patients with a history of MI


104x90mm (300 x 300 DPI)

Page 23 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



Spline plots for risk of composite CV outcome by age and IHD severity. Reference is mean age (63 years) in the T2D no IHD population



104x89mm (300 x 300 DPI)

Page 24 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



Spline plots for MI risk by age and IHD severity. Reference is mean age (63 years) of the T2D no IHD population



104x90mm (300 x 300 DPI)

Page 25 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



Supplementary data.

Table 1 A



No IHD 1.09 (1.05-1.12) 1.84 (1.79-1.89) 2.55 (2.49-2.62) 3.29 (3.21-3.36) 4.01 (3.92-4.10) 4.78 (4.68-4.88)

IHD wo MI 1.33 (1.26-1.41) 2.26 (2.14-2.37) 3.13 (2.98-3.29) 4.03 (3.83-4.23) 4.91 (4.67-5.15) 5.85 (5.57-6.13)

IHD with MI 1.85 (1.76-1.93) 3.13 (2.99-3.26) 4.33 (4.16-4.51) 5.56 (5.34-5.78) 6.76 (6.50-7.02) 8.04 (7.73-8.34)




without MI



Table 1 B



No IHD 0.34 (0.32-0.36) 0.61 (0.58-0.64) 0.87 (0.83-0.91) 1.13 (1.09-1.18) 1.40 (1.35-1.45) 1.66 (1.60-1.72)

IHD wo MI 0.64 (0.58-0.70) 1.14 (1.04-1.24) 1.63 (1.49-1.77) 2.12 (1.94-2.29) 2.61 (2.40-2.82) 3.09 (2.84-3.34)

IHD with MI 1.14 (1.05-1.22) 2.03 (1.89-2.16) 2.89 (2.71-3.06) 3.74 (3.52-3.97) 4.61 (4.33-4.87) 5.44 (5.13-5.76)




without MI



Table 1 C



No IHD 0.51 (0.48-0.54) 0.83 (0.80-0.87) 1.12 (1.08-1.17) 1.42 (1.37-1.47) 1.71 (1.65-1.76) 2.03 (1.97-2.10)

IHD wo MI 0.55 (0.50-0.60) 0.89 (0.81-0.97) 1.20 (1.10-1.30) 1.52 (1.39-1.65) 1.82 (1.67-1.98) 2.17 (1.99-2.35)

IHD with MI 0.57 (0.53-0.62) 0.93 (0.86-1.01) 1.26 (1.16-1.35) 1.59 (1.48-1.71) 1.91 (1.77-2.05) 2.28 (2.11-2.44)




without MI

Page 26 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from





Table 1 D



No IHD 0.22 (0.20-0.23) 0.38 (0.36-0.40) 0.56 (0.53-0.58) 0.74 (0.70-0.77) 0.93 (0.89-0.97) 1.14 (1.09-1.19)

IHD wo MI 0.23 (0.21-0.26) 0.42 (0.38-0.45) 0.61 (0.56-0.65) 0.80 (0.74-0.87) 1.01 (0.93-1.09) 1.24 (1.15-1.34)

IHD with MI 0.32 (0.30-0.35) 0.57 (0.53-0.61) 0.83 (0.77-0.89) 1.10 (1.03-1.17) 1.39 (1.30-1.48) 1.70 (1.59-1.81)




without MI



Page 27 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



Figure 1. Kaplan–Meier estimate of the risk of the composite CV outcome (CV-death/MI/stroke)

during the first three years after collecting first-time GDL from pharmacy among T2D patients

without IHD, T2D IHD patients without MI, and in T2D IHD patients with a history of MI

Page 28 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



Figure 2. Kaplan–Meier estimate of the risk of all-cause mortality during the first three years after

collecting first-time GDL from pharmacy among T2D patients without IHD, T2D IHD patients without

MI, and in T2D IHD patients with a history of MI

Page 29 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



1




Item

No Recommendation



and what was found

Page 1 and 2

Introduction


Page 4


Page 4

Methods


Page 5



Page 5



Page 5


unexposed

Page NA



Page 5

Data sources/

measurement



more than one group

Page 5


Not applicabøe


Not relevant



Not relevant


Page 6


Page 6


Page 6 and 11


Not applicable

Page 30 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



2



Results




Page 7-8



Not included



Page 7


Not applicable



Page 7







Page 8




Discussion


Page 9



Page 11



Page 12


Page 12

Other information



Page 14




Page 31 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from



3




Page 32 of 32


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on Novem


http://bmjopen.bm

j.com/

BM


jopen-2018-027199 on 3 April 2019. D

ownloaded from


bmjopen.bmj.com€¦ · for peer review only references 1. idf diabetes atlas - 8th edition 2017...

Documents