European Heart Journal

European Heart Journal Advance Access originally published online on December 21, 2007
European Heart Journal 2008 29(2):166-176; doi:10.1093/eurheartj/ehm518

This Article Abstract FREE Full Text (PDF) All Versions of this Article: 29/2/166    most recent ehm518v1 E-letters: Submit a response Alert me when this article is cited Alert me when E-letters are posted Alert me if a correction is posted Services Email this article to a friend Related articles in EHJ Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Archive Download to citation manager Search for citing articles in: ISI Web of Science (7) Request Permissions Disclaimer Google Scholar Articles by Mellbin, L. G. Search for Related Content PubMed PubMed Citation Articles by Mellbin, L. G. Social Bookmarking          What's this?

Published on behalf of the European Society of Cardiology. All rights reserved. © The Author 2007. For permissions please email: journals.permissions@oxfordjournals.org

The impact of glucose lowering treatment on long-term prognosis in patients with type 2 diabetes and myocardial infarction: a report from the DIGAMI 2 trial

Linda G. Mellbin^1,*, Klas Malmberg¹, Anna Norhammar¹, Hans Wedel², Lars Rydén for the DIGAMI 2 Investigators¹

¹ Cardiology Unit, Department of Medicine, Karolinska Institutet, Stockholm, Sweden
² Nordic School of Public Health, Göteborg, Sweden

Received 24 June 2007; revised 11 October 2007; accepted 18 October 2007; online publish-ahead-of-print 21 December 2007.

^* Corresponding author. Tel: +46 8 51772171, Fax: +46 8 344964, Email: linda.mellbin{at}karolinska.se

See page 141 for the editorial comment on this article (doi:10.1093/eurheartj/ehm595)

	Abstract

Top Abstract Introduction Methods Results Discussion Conclusions Funding Authors' contribution Acknowledgements References

 
Aims: To explore the impact of glucose lowering treatment on prognosis in diabetic patients with myocardial infarction.

Methods and results: 1181 type 2 diabetic patients (mean age 68 years; 67% males) discharged after myocardial infarction were followed (median of 2.1 years). At discharge, 436 patients (37%) had oral glucose lowering agents whereof 268 sulphonylureas and 200 metformin, while 690 patients (58%) were on insulin. The impact of treatment was analysed by an updated Cox proportional hazards regression model, correcting for confounders. Cardiovascular mortality was not influenced by metformin [Hazard ratio (HR) 0.93, 95% CI 0.60–1.43; P = 0.73], sulphonylureas (HR 1.15, 95% CI 0.80–1.64; P = 0.45), or insulin (HR 1.05, 95% CI 0.75–1.46; P = 0.77). The risk for non-fatal myocardial infarction and stroke increased significantly in patients on insulin (HR 1.73, 95% CI 1.26–2.37; P = 0.0007), whereas this risk was lower among those on metformin (HR 0.63, CI 0.42–0.95; P = 0.03) and unchanged with sulphonylureas (HR 0.81, 95% CI 0.57–1.14; P = 0.23). This finding remained analysing only patients with newly instituted insulin and those randomly allocated to newly instituted insulin.

Conclusion: Controlling for confounders including glycemic control, there was no significant difference in mortality between sulphonylureas, metformin, and insulin. In this post hoc analysis, the risk of non-fatal myocardial infarction and stroke increased significantly by insulin treatment while metformin was protective. It is emphasized that this observation is done in an epidemiological analysis and should encourage to further confirmation in randomized trials.

Key Words: Myocardial infarction • Diabetes mellitus • Sulphonylurea • Metformin • Insulin • Cardiovascular events

	Introduction

Top Abstract Introduction Methods Results Discussion Conclusions Funding Authors' contribution Acknowledgements References

 
The proportion of type 2 diabetes is as high as 20–30% among patients admitted for acute myocardial infarction.^1–2 Due to better management the overall prognosis has improved following acute coronary events but patients with diabetes continue to have a considerably more dismal prognosis than their non-diabetic counterparts.^3–4 Importantly, this increased risk is not only seen in older patients. In the Swedish Register of Information and Knowledge about Swedish Heart Intensive Care Admission (RIKS-HIA) the relative risk for 1 year mortality among diabetic patients were 1.66, 1.42, and 1.34 for the age groups <65 years, 65–75 years, and >75 years compared with those without diabetes.² This excess, indeed an almost doubled mortality, has remained over the years indicating that benefits achieved in the non-diabetic patients with myocardial infarction have not to the same extent included those with diabetes.

Factors of possible importance for the poor prognosis may relate to diabetes specific reasons, including diffuse coronary atherosclerosis, diabetic cardiomyopathy, autonomic neuropathy, metabolic factors, and an increased thrombogenicity.⁵ Another factor may be a suboptimal use of evidence based management strategies although several treatment modalities, such as beta-blockade, and the use of statins are as beneficial in diabetic as in non-diabetic patients.^1,3,6

Uncertainties of the optimal glucose lowering treatment and deficiencies in reaching glucose treatment targets may also contribute to the unfavourable prognosis. It is firmly established that hyperglycemia is an important predictor for long-term outcome after acute coronary events.^7–10 Thus, meticulous glucose control is important, but how this should be accomplished is not well established. In the first DIGAMI study, improved glucose control based on insulin caused a considerable decrease in long-term mortality.⁸ These findings were, however, not verified in the second DIGAMI trial. The most likely explanation was the lack of difference in glucose control between the three study arms.⁹ Indeed, a non-significant trend towards an increased morbidity in the form of non-fatal myocardial infarctions and stroke was observed among patients treated with insulin. As recently reviewed, oral glucose lowering with sulphonylureas have been questioned in patients with coronary artery disease (CAD) due to potentially harmful metabolic effects on ischaemic myocardium.¹¹ These concerns have, however, not been clinically verified.¹² In the United Kingdom Prospective Diabetes Study (UKPDS) metformin appeared to have survival benefits, but this was in a general diabetic population rather than in subjects with established CAD.¹³ Furthermore, Nissen and Wolski¹⁴ highlighted the need for long-term studies with glucose lowering agents, not only focusing on glycemic control but also on hard outcomes. Thus, it seems of great interest to further study the prognostic implications of glucose lowering drugs in patients with various CAD manifestations.

This epidemiological report from DIGAMI 2 contains a detailed analysis of the impact of glucose lowering therapies on long-term prognosis in diabetic patients surviving the hospital phase of acute myocardial infarctions.

	Methods

Top Abstract Introduction Methods Results Discussion Conclusions Funding Authors' contribution Acknowledgements References

 
Patients
DIGAMI 2, a prospective, randomized trial compared three different management strategies in patients with type 2 diabetes and suspect acute myocardial infarction. An extensive description of the study design has been presented elsewhere.⁹ In brief, 1253 patients with established type 2 diabetes or an admission blood glucose >11.0 mmol/L, admitted to participating coronary care units, were eligible for inclusion according to the following criteria: suspect acute myocardial infarction due to symptoms (chest pain >15 min during the preceding 24 h) and/or recent ECG signs (new Q-waves and/or ST-segment deviations in 2 leads). They were randomized to one of three study arms receiving: (i) a 24 h insulin-glucose infusion followed by subcutaneous insulin-based long-term glucose control (n = 474); (ii) the same initial treatment followed by standard glucose control (n = 473); or (iii) glucose lowering treatment according to local practice (n = 306). The objective was to compare total mortality and morbidity between these management strategies. There was no significant difference in the primary endpoint, total mortality.⁹ Increasing plasma glucose was, however, an independent predictor of fatal outcome. The present patient material consists of the 1181 subjects who were discharged alive from the index hospitalization.

Treatment
Glucose-lowering treatment
Group 1 and 2 were initially treated with a glucose–insulin infusion aiming at blood glucose between 7–10 mmol/L as fast as possible and to be continued until stable normoglycemia and at least for 24 h. In group 1, the infusion was followed by multidose, subcutaneous insulin, short- and intermediate long-acting, targeting a fasting blood glucose between 5-7 mmol/L, and non-fasting <10 mmol/L. Besides during the insulin–glucose infusion in group 2, there was no specified treatment goals in this group and group 3 with the choice of glucose lowering treatment left to the discretion of the responsible physician.

Concomitant treatment
The protocol stated that the use of concomitant treatment should be as uniform as possible and according to evidence-based international guidelines for acute myocardial infarction.^15,16 Particular emphasis was put on prescribing aspirin, thrombolytic agents, beta-blockers, lipid-lowering drugs, Angiotensin-converting enzyme (ACE)-inhibitors, and revascularization procedures if appropriate and not contraindicated.

Laboratory investigations
Blood glucose was obtained at each follow-up visit to be reported as locally analysed whole blood glucose in mmol/L. Likewise blood lipids, serum creatinine, and electrolytes were analysed at the local laboratories. Glycated haemoglobin A1c (HbA1c) was analysed at a central core laboratory (Department of Laboratory Medicine, Malmö Hospital, Sweden) by high-performance liquid chromatography on capillary blood applied on filter paper with an upper normal limit of 5.3%. (Boehringer Mannheim Scandinavian AB, Bromma, Sweden)¹⁷

Follow-up
Out-patient visits were scheduled after 3, 6, 9, and 12 months and thereafter every sixth month. All patients were followed for a minimum of 6 months, whereas the maximum time of follow-up was 3 years. At each follow-up, a case record form was completed comprising information on fasting glucose, HbA1c, a detailed report on ongoing glucose lowering and other therapy and adverse events. Events including mortality, non-fatal infarctions, stroke, and coronary interventions were reported by means of special event forms.

Definitions
Myocardial infarction was diagnosed according to the joint ESC and ACC recommendations.¹⁸ A re-infarction was defined as an event >72 h from the index infarction.

Stroke was defined as unequivocal signs of focal or global neurological deficit of sudden onset and vascular origin, lasting >24 h.

Deaths were verified with death certificates and autopsy reports when available. Sudden cardiovascular deaths were those that occurred within 24 h following onset of symptoms and without any other obvious reason for the fatal outcome. Deaths were labelled as cardiovascular or non-cardiovascular.

An independent committee composed of three experienced cardiologists, blinded for study group allocation, adjudicated all these events by means of case record forms, and relevant extracts from hospital records including laboratory values and electrocardiograms.

Updated blood glucose was defined as the average value of all fasting glucose values. The first was recorded at the day of hospital discharge and subsequent values at each visit until that preceding an event or at the end of follow-up.

The use of drugs was updated at each visit until the last preceding an event or end of follow-up. This was done to ascertain that drugs added or excluded during the time of follow-up were taken into account when looking at the impact of therapy.

Local ethics review boards approved the protocol. Written informed consent was obtained from all patients prior to enrolment.

Statistics
The Cox proportional hazards model was the basis for the main analyses. In order to check the proportional hazards model, the proportionality was assessed by considering the interactions of the treatment indicators and time. In all adjusted multivariable models, predicting endpoints, the following covariates, recorded at the time for hospital admission, remained significant: age, smoking habits, previous myocardial infarction, previous congestive heart failure, creatinin at randomization, sex, percutaneous transluminal coronary angioplasty (PTCA) or coronary artery bypass grafting (CABG) before randomization, and also during the hospitalization or blood glucose at randomization. The same set of variables was used for all endpoints. Thus, no model building procedures were used. Instead, covariates known to be predictive from the literature and the DIGAMI 2 study were applied. The first models used mainly baseline variables as covariates. No interaction terms was used due to limited power. Only insulin and blood glucose were used as time dependent variables. Blood glucose was preferred as a measure of glycemic control rather than HbA1c by several reasons as further outlined in the discussion. Initial variables for prediction of the use of insulin by a propensity score in a logistic regression were: current smoker, diabetes duration, myocardial infarction, heart failure, hypertension, PTCA, CABG, HbA1c, blood glucose, potassium, cholesterol, triglycerides, beta-blockers, aspirin, ACE-inhibitors, lipid lowering, diuretics, and time in hospital. In the final propensity score model, by using stepwise logistic regression, the following variables of those listed were selected: diabetes duration, heart failure, hypertension, blood glucose, and lipid lowering. The propensity score consisted of a linear combination of these variables. No interactions were allowed in this score. The c-statistics in the final model was 0.747. The c-statistics did not increase if all variables and interactions between variables were included. Due to missing values, a total of 1119 of 1181 patients (insulin n = 656; non-insulin n = 463) were included in the sensitivity analysis. The distribution of propensity score linear term for insulin patients was mean 0.87 (SD 1.18) compared with mean –0.04 (SD 0.78) for non-insulin patients. For all updated and adjusted endpoint analyses, the inclusion of the linear term of propensity score did not change the main results of this study.

Time in study was measured from the time of discharge from hospital. In the ‘insulin at discharge’ models, insulin status was characterized at the moment of discharge from hospital and kept as this value during follow-up. In the updated models insulin status varied and given an actual value each day (updated by day). Subsequently time-dependent Cox proportional hazards model was used for updated variables.¹⁹ Adjustments were done for prognostic variables at discharge and in addition updated blood glucose and updated insulin treatment (see Definitions) in order to account for variability in glucose control and the use insulin.

For illustration purposes, we have chosen to present Kaplan–Meier curves estimated from Cox models and adjusted for variables at discharge using insulin/no insulin as strata. This information is given with the time dependent adjusted Hazard ratio (HR) as a complement. Two-tailed statistical tests were used at a 5% significance level. SAS version 8:02 was used for all statistical analyses.

	Results

Top Abstract Introduction Methods Results Discussion Conclusions Funding Authors' contribution Acknowledgements References

 
Characteristics at baseline
The total DIGAMI 2 cohort consisted of 1253 patients with type 2 diabetes of whom 1181 (94%), the present population, were discharged alive. The clinical characteristics, biochemical data, and treatment at hospital discharge are presented in Table 1 dividing the patients by insulin treatment. The median study duration was 2.1 (inter-quartile range 1.03–3.00) years, and no patient was lost to follow-up. The median time of follow-up was 2.1 years (mean 2.0 ± 0.9) for patients on insulin and 2.5 years (mean 2.1 ± 0.9) for the remaining patients. Newly detected diabetes, defined as of duration <1 year, was seen in 251 of the 1181 patients (21%). At hospital discharge, 1011 patients (86%) fulfilled the diagnosis of myocardial infarction whereof 47% had Q-wave and 53% non-Q-wave infarctions. Of the remaining hospital survivors, 111 had CAD (unstable angina pectoris n = 58) and 59 miscellaneous diagnoses.

View this table: [in this window] [in a new window]   Table 1 Clinical characteristics and biochemical data at the time for hospital admission and treatment at the time for hospital discharge in all 1181 patients divided by insulin therapy

 
Treatment
The distribution of patients on various oral glucose lowering drugs and on insulin is presented in Table 2. At the time for discharge, 436 patients were treated with oral glucose lowering agents, 268 (61%) with sulphonylureas, 200 (46%) metformin, and 9 (2%) acarbos. At discharge, 690 patients (58%) were on insulin while 176 (15%) did not receive any pharmacological glucose lowering treatment. A total of 173 (15%) patients were on a combination of various glucose lowering agents. The proportion of patients on insulin remained stable over the time of follow-up (after 12 months = 58%; 24 months = 60%; 36 months = 63%).

View this table: [in this window] [in a new window]   Table 2 Glucose lowering treatment at the time for hospital admission and discharge

 
Mortality and morbidity
During follow-up, 206 of the 1181 patients died while 162 and 54 had non-fatal myocardial infarction or stroke respectively. The impact of glucose lowering drugs on cardiovascular events was analysed in three steps. The first compared all different glucose lowering therapies (insulin, oral, or no pharmacological) with each other as regards cardiovascular events. Following adjustment none of the oral glucose-lowering treatments were associated with improved survival during the study period (Figure 1). However, the likelihood to develop a non-fatal myocardial reinfarction or stroke was significantly lower in patients on metformin (P = 0.0269) while sulphonylureas did not relate to morbidity.

View larger version (21K): [in this window] [in a new window] [Download PowerPoint slide]   Figure 1 Effect of different updated glucose lowering treatments on mortality and morbidity. CV, cardiovascular

 
The effects of insulin on mortality and cardiovascular events are presented in Table 3 and Figure 2A and B. Insulin was not associated with all cause (adjusted HR 1.19, 95% CI 0.88–1.60; P = 0.2575) or cardiovascular mortality (adjusted HR 1.14, 95% CI 0.82–1.60; P = 0.4284) but with an increased proportion of non-fatal reinfarction or stroke (adjusted HR 1.71, 95% CI 1.25–2.35; P = 0.0008). The statistical models were also tested for consistency by adding propensity score with hypertension, ACE-inhibitors, lipid lowering drugs, diuretics, diabetes duration and time at hospital as covariates. The results were not influenced by any of the added corrections.

View this table: [in this window] [in a new window]   Table 3 The effect of insulin treatment on time to first event

 

View larger version (11K): [in this window] [in a new window] [Download PowerPoint slide]   Figure 2 Kaplan–Meier estimates of all cause mortality (A) and non-fatal reinfarction or stroke (B) in patients with and without insulin treatment at hospital discharge

 
Thirty-two per cent of the patients (n = 375) discharged alive were already on insulin at hospital admission. To look at the effect of newly instituted insulin a second analysis was made including all patients (n = 317), who were prescribed new insulin during the initial hospitalization (n = 245 from study group1 and n = 72 from study groups 2 and 3). They were compared with all patients (n = 489), who never had been on insulin and who were discharged without such therapy. Updated information on insulin was used in the analysis adjusted for baseline variables and updated glucose. The observation that insulin related to an increased risk for non-fatal myocardial infarction or stroke compared with non-insulin based treatment remained (adjusted HR 1.95, 95% CI 1.35–2.82; P = 0.0004; Table 3 and Figure 3A and B) in this subgroup. Newly instituted insulin did not relate to all cause (adjusted HR 1.09, 95% CI 0.74–1.61; P = 0.6631) or cardiovascular (adjusted HR 1.11, 95% CI 0.72–1.72; P = 0.6237) mortality.

View larger version (10K): [in this window] [in a new window] [Download PowerPoint slide]   Figure 3 Kaplan–Meier estimates of all cause mortality (A) and non-fatal reinfarction or stroke (B) in patients with and without newly instituted insulin treatment at hospital discharge

 
Finally, a third analysis compared all patients that were randomized to and discharged with new insulin according to the study protocol (study group 1; n = 245) with patients (n = 422) in groups 2 and 3 without insulin at admission and discharge i.e. without chronic insulin therapy. Clinical and biochemical characteristics for these two groups are presented in Table 4. As can be seen there were few differences between the groups compared with what was seen when looking at all patients with and without insulin therapy (Table 1). As outlined in Table 3, the significantly increased risk for myocardial infarction or stroke remained even in this cohort (HR 2.07, 95% CI 1.37–3.14; P = 0.0006). When the propensity score was added into the models of Table 3, the results did not change.

View this table: [in this window] [in a new window]   Table 4 Clinical characteristics and biochemical data at the time for hospital admission and treatment at the time for hospital discharge in patients randomized to and discharged with new insulin and in patients without insulin at hospital admission and discharge i.e. without any insulin treatment (see text for further explanation)

 

	Discussion

Top Abstract Introduction Methods Results Discussion Conclusions Funding Authors' contribution Acknowledgements References

 
The important message from this epidemiological analysis of the DIGAMI 2 trial is that the agent used for long-term glucose control may play a considerable role for the development of future non-fatal cardiovascular events. Chronic insulin was associated with an increased number while metformin was beneficial and sulphonylureas neutral in this respect.

The present observations are based on a prospectively composed and regularly followed group of diabetic patients. The study design allowed successive updating of fasting glucose and treatment modalities. The total number of patients (n = 1181) and events (n = 422) was large, and all events were adjudicated by an independent committee. Updated blood glucose was preferred as a measure of glycemic control partly because a total of 253 HbA1c analyses were reported as missing while the information on blood glucose was complete. Glucose was an independent and strong predictor of mortality, indeed stronger than HbA1c in the DIGAMI 2 trial.⁹ Furthermore, the half-life of HbA1c of 3 months makes this variable a more blunt expression of glucose control than fasting glucose in a study with limited follow-up.

Moreover, updated glucose should reflect the glycemic control more accurately than the use of occasional glucose levels e.g. at the time for hospital discharge or closest to an event. Previous reports on diabetic post-infarction patients are often based on subgroup analysis of clinical trials or registry studies, which by necessity introduce uncertainties in the definition of the diabetic state and glucose lowering treatment as well as long-term glucose control.^12,20–23

The vast majority of patients on oral glucose lowering treatment in DIGAMI 2 got sulphonylureas or metformin. Sulphonylureas did not relate to mortality and morbidity. Since the original report by the University Group Diabetes Program,²⁴ the use of sulphonylureas has been questioned and claimed to increase cardiovascular mortality, especially in patients at a high cardiovascular risk.^11,12,20 The proposed explanation is a combination of harmful effects on vascular tone and interference with ischaemic preconditioning.²⁵ Riveline et al.²⁶ made a systematic literature review of experimental, epidemiological and clinical studies. The authors concluded that the cardiac effects of sulphonylureas differ between the first and second generation. The only presently available, prospective, clinical study, UKPDS,²⁷ did not report on any deleterious effects of glibenclamide compared with chlorpropamide and insulin. The present findings support the view that sulphonylureas may be used without harm.

Glucose lowering by means of metformin did not influence mortality but was related to a lower proportion of non-fatal cardiovascular events. The latter finding is in agreement with UKPDS¹³ but in this study metformin did also have a mortality lowering effect. The discrepancy is most likely explained by differences in the studied populations and time of follow-up. UKPDS comprised patients with newly detected diabetes without previous cardiovascular disease, whereas DIGAMI 2 recruited diabetic patients with acute myocardial infarction reasonably at a substantially higher risk. It is not unlikely that the lower incidence of non-fatal cardiovascular events in metformin treated patients in DIGAMI 2 may reduce mortality during extended follow-up.

The most surprising finding was the increase in non-fatal cardiovascular events seen with the use of insulin compared with oral glucose lowering agents or life-style measures only. It should be underlined that this observation was made following adjustment for updated glucose i.e. cannot be explained by a less efficient glycemic control by insulin. It is unlikely that the findings are explained by insulin serving as a marker for a more serious clinical condition since the untoward effect remained following adjustment for a number of variables related to the condition and treatment of the patients including a propensity score. Furthermore the three steps of analyses were completely consistent. A potentially harmful effect of insulin on cardiovascular mortality and morbidity has been discussed based on registry studies or subgroup analysis of a clinical trial. Nichols et al.²⁸ and Smooke et al.²² reported on an increased incidence of heart failure and an enhanced mortality in heart failure among diabetic patients on insulin compared with patients on non-insulin based therapy, respectively. The SYMPHONY investigators²¹ noted that insulin providing (insulin or sulphonylureas) compared with insulin sensitizing (biguanides and/or thiazolidinedione) post-myocardial infarction treatment in diabetic patients related to a significantly higher event rate. Johnsen et al.¹² performed a registry based review of glucose lowering drugs given to patients hospitalized for a first myocardial infarction who were compared with population based controls. Patients prescribed insulin had a higher rate of myocardial infarction than those on oral glucose lowering drugs. These findings have, however, not been proven in prospective, well-defined patient populations. This makes the present observation that the composite endpoint of mortality and non-fatal myocardial infarction, and stroke was significantly more frequent among patients on chronic insulin of particular interest. This increase was largely driven by a higher number of non-fatal cardiovascular events although a non-significant trend towards a higher mortality was observed. It may be speculated that the higher number of non-fatal cardiovascular events would enhance the mortality difference by time. The separate analysis on patients with newly instituted insulin did not change the pattern. Thus, the dismal impact cannot be explained by longer periods on insulin. The final analysis, including only patients randomly allocated to insulin, was made to as far as possible exclude a prescription bias in the direction that insulin more often was administered to patients with advanced disease. The outcome confirmed and further strengthened the observation made in the total patient cohort. The present finding contrasts the results of DIGAMI 1 in which insulin based glucose control improved long-term mortality.⁸ An important difference between the two DIGAMI trials is that the first succeeded in getting a significantly better glucose control in the insulin-based treatment arm, whereas glucose control was similar with oral glucose lowering agents and insulin in the latter. This fits with observations made by Van den Berghe et al.^29,30 that meticulous metabolic control, rather than insulin per se, has a positive effect on mortality and morbidity at least in a short-term perspective. The interpretation would be that, if insulin is used for glucose control, it is important that the treatment target is normoglycemia in order to balance the favourable impact of glucose control against a potentially harmful effect of insulin in itself.

Limitations
The present epidemiological analysis does only permit speculations on the reasons for the harmful effects of insulin. Deleterious effects have mostly been related to endogenous hyperinsulinaemia although some reports addressed exogenous insulin. Negative effects on endothelial function, sympathetic nervous system activation, smooth muscle cell hypertrophy, impaired fibrinolysis, and induction of platelet dysfunction have been discussed as potential pathophysiological factors.^31–37 Moreover, a pro-atherogenic effect of insulin may be mediated by growth factor like activity.³⁸

Although based on a well-defined patient material, it should be acknowledged that data presented are based on an epidemiological analysis. Despite adjustments, including the use of a propensity score, there still is a possibility of confounding by indication due to unobserved covariates. Bias may therefore remain in the treatment estimation. It is, however, unlikely that the main finding, an untoward impact of insulin, is an expression of such confounding. Newly instituted insulin was mainly prescribed according to the randomization of patients to study group 1, and the observation of a trend towards increased mortality and a higher number of non-fatal cardiovascular events remained in the separate analysis of this group. Another limitation is the relatively short period of follow-up. Finally, the possibility for competing risk has to be discussed. Since there were no significant mortality differences between the various glucose lowering treatments the exposure for non-fatal events should not have been much different between treatment with metformin, sulphonylureas, and insulin thereby reducing competing risk as an explanation to the present results.

	Conclusions

Top Abstract Introduction Methods Results Discussion Conclusions Funding Authors' contribution Acknowledgements References

 
Controlling for confounders including updated blood glucose, there was no significant difference in mortality between sulphonylureas, metformin, and insulin. In this post hoc analysis, the risk of non-fatal myocardial infarction or stroke was higher in patients on insulin treatment while metformin seemed to be protective. Whether this difference will translate into increased mortality can only be answered by more extensive periods of follow-up. The present observations make it important to further study the impact of insulin in a randomized study design and to search for other pharmacological glucose lowering agents. It also underlines the safety of the oral glucose lowering agents, sulphonylureas and metformin, in post-myocardial infarction patients with diabetes mellitus.

	Funding

Top Abstract Introduction Methods Results Discussion Conclusions Funding Authors' contribution Acknowledgements References

 
DIGAMI 2 was supported by the Swedish Heart-Lung Foundation, AFA Insurance and by unconditional research grants from Aventis Sweden and Novo Nordic Denmark.

	Authors' contribution

Top Abstract Introduction Methods Results Discussion Conclusions Funding Authors' contribution Acknowledgements References

 
L.G.M. took a main responsibility for bringing the present data together, wrote the original manuscript and finalised the present version. She participated in the statistical considerations.

L.R. and K.M. designed DIGAMI 2 and as chairman of the steering committee and principal investigator they also were responsible for its conductance. They were part of data collection and came up with the idea to this particular analysis and data interpretation and writing.

A.N. was part of the study group and collected data. She has contributed to the interpretation of the present data.

H.W. has as statistician been involved with DIGAMI 2 since its start. He was responsible for the present statistical calculations and the use of statistical methods. He took part in the writing process.

	Acknowledgements

Top Abstract Introduction Methods Results Discussion Conclusions Funding Authors' contribution Acknowledgements References

 
The authors are grateful to Aldina Pivodic M.Sc. and Mattias Molin B.Sc., Statistical Consulting Group, Gothenburg, for excellent support with data handling.

Conflict of interest: None of the authors have any financial interests, besides research funding as outlined above, or relationships and affiliations related to the relevance to the subject of this manuscript. K.M. is professor of cardiology at Karolinska Institutet, Stockholm 25% of his time and employed by AstraZeneca to 75%. This company has not been involved in the trial and has no interest in this manuscript.

	References

Top Abstract Introduction Methods Results Discussion Conclusions Funding Authors' contribution Acknowledgements References

 

1. Norhammar A, Lindbäck J, Rydén L, Wallentin L, Stenestrand U, for the Register of Information Knowledge about Swedish Heart Intensive Care Admission (RIKS-HIA). Improved but still high short and long-term mortality after myocardial infarction in patients with diabetes mellitus. A time trend report from the Swedish Register of Information and Knowledge about Swedish Heart Intensive Care Admission. Heart (2007) Jan 19; [Epub ahead of print].
2. Bartnik M, Rydén L, Ferrari R, Malmberg K, Pyörälä K, Simoons M, Standl E, Soler-Soler J, Öhrik J, on behalf of the Euro Heart Survey Investigators. The prevalence of abnormal glucose regulation in patients with coronary artery disease across Europe. The Euro Heart Survey on diabetes and the heart. Eur Heart J (2004) 25:1880–1890.[Abstract/Free Full Text]
3. Gitt AK, Schiele R, Wienbergen H, Zeymer U, Schneider S, Gottwik MG, Senges J, for the MITRA study group. Intensive treatment of coronary artery disease in diabetic patients in clinical practice: results of the MITRA study. Acta Diabetol (2003) 40:S343–S347.[CrossRef][Web of Science][Medline]
4. Hasdai D, Behar S, Wallentin L, Danchin N, Gitt AK, Boersma E, Fioretti PM, Simoons ML, Battler A. A prospective survey of the characteristics, treatments and outcome of patients with acute coronary syndromes in Europe and the Mediterranean basin; the Euro Heart Survey of Acute Coronary Syndromes. Eur Heart J (2002) 23:1190–1201.[Abstract/Free Full Text]
5. Rydén L, Malmberg K. Treatment of coronary heart disease in patients with diabetes. In: The Diabetic Coronary Patient—Stanley WC, Rydén L, eds. (2003) London: Science Press Ltd.
6. EUROASPIRE II Study Group. Lifestyle and risk factor management and use of drug therapies in coronary patients from 15 countries. Principal results from EUROASPIRE II Euro Heart Survey Programme. Eur Heart J (2001) 22:554–572.[Abstract/Free Full Text]
7. Norhammar A, Malmberg K, Diderholm E, Lagerqvist B, Lindahl B, Rydén L, Wallentin L. Diabetes mellitus: the major risk factor in unstable coronary artery disease even after consideration of the extent of coronary artery disease and benefits of revascularization. J Am Coll Cardiol (2004) 43:585–591.[Abstract/Free Full Text]
8. Malmberg K, Norhammar A, Wedel H, Rydén L. Glycometabolic state at admission: important risk marker of mortality in conventionally treated patients with diabetes mellitus and acute myocardial infarction. Long-term results from the diabetes and insulin–glucose infusion in acute myocardial infarction (DIGAMI) study. Circulation (1999) 99:2626–2632.[Abstract/Free Full Text]
9. Malmberg K, Rydén L, Wedel H, Birkeland K, Bootsma A, Dickstein K, Efendic S, Fisher M, Hamsten A, Herlitz J, Hildebrandt P, MacLeold K, Laakso M, Torp-Pederson C, Waldenström A, for the DIGAMI 2 Investigators. Intense metabolic control by means of insulin in patients with diabetes mellitus and acute myocardial infarction (DIGAMI 2): effects on mortality and morbidity. Eur Heart J (2005) 26:650–661.[Abstract/Free Full Text]
10. Bhadriraju S, Ray KK, DeFranco AC, Barber K, Bhadriraju P, Murphy SA, Morrow DA, McCabe CH, Gibson CM, Cannon CP, Braunwald E. Association between blood glucose and long-term mortality in patients with acute coronary syndromes in the OPUS-TIMI 16 trial. Am J Cardiol (2006) 97:1573–1577.[CrossRef][Web of Science][Medline]
11. Thisted H, Johnsen SP, Rungby J. Sulfonylureas and the risk of myocardial infarction. Metabolism (2006) 55(Suppl. 1):S16–S19.[CrossRef][Web of Science][Medline]
12. Johnsen SP, Monster TB, Olsen ML, Thisted H, McLaughlin JK, Sorenson HT, Lervang HH, Rungby J. Risk and short-term prognosis of myocardial infarction among users of antidiabetic drugs. Am J Therap (2006) 13:134–140.[CrossRef]
13. UK Prospective Diabetes Study (UKPDS) Group. Effect of intensive blood-glucose control with metformin on complications in overweight patients with type 2 diabetes (UKPDS 34). Lancet (1998) 352:854–865.[CrossRef][Web of Science][Medline]
14. Nissen S, Wolski K. Effect of Rosiglitazone on myocardial infraction and death from cardiovascular causes. N Engl J Med (2007) 356:2457–2471.[Abstract/Free Full Text]
15. The Task Force on the Management of Acute Myocardial Infarction of the European Society of Cardiology. Acute myocardial infarction: pre-hospital and in-hospital management. Eur Heart J (1996) 17:43–63.[Free Full Text]
16. Van de Werf F, Ardissino D, Betriu A, Cokkinos DV, Falk E, Fox KA, Julian D, Lengyel M, Neumann FJ, Ruzyllo W, Thygesen C, Underwood SR, Vahanian A, Verheugt FW, Wijns W, Task Force on the Management of Acute Myocardial Infarction of the European Society of Cardiology. Management of acute myocardial infarction in patients presenting with ST-segment elevation. The Task Force on the Management of Acute Myocardial Infarction of the European Society of Cardiology. Eur Heart J (2003) 24:28–66.[Free Full Text]
17. Jeppsson J-O, Jerntorp P, Almer L-O, Persson R, Ekberg G, Sundkvist G. Capillary blood on filter paper for determination of HbA_1c by ion exchange chromatography. Diabetes Care (1996) 19:42–145.[Medline]
18. Myocardial infarction redefined—A consensus document of The Joint European Society of Cardiology/American College of Cardiology Committee for the Redefinition of Myocardial Infarction. Eur Heart J (2000) 21:1502–1516.[Abstract/Free Full Text]
19. Kalbfleisch J D, Prentice R L. The Statistical Analysis of Failure Time Data (1980) New York: John Wiley & Sons.
20. Simpson SH, Majumdar SR, Tsuyuki RT, Eurich DT, Johnson JA. Dose-response relation between sulfonylurea drugs and mortality in type 2 diabetes mellitus: a population-based cohort study. CMAJ (2006) 174:169–174.[Abstract/Free Full Text]
21. McGuire DK, Newby LK. K, Bhapkar MV, Moliterno DJ, Hochman JS, Klein WW, Weaver DD, Pfisterer M, Corbalán R, Dellborg M, Granger CB, Van de Werf F, Topol EJ, Califf RM, for the SYMPHONY, 2nd SYMPHONY Investigators. Association of diabetes mellitus and glycemic control strategies with clinical outcomes after acute coronary syndromes. Am Heart J (2004) 147:246–252.[CrossRef][Web of Science][Medline]
22. Smooke S, Horwich TB, Fonarow GC. Insulin-treated diabetes is associated with a marked increase in mortality in patients with advanced heart failure. Am Heart J (2005) 149:168–174.[CrossRef][Web of Science][Medline]
23. Anselmino M, Bartnik M, Malmberg K, Rydén L, on behalf of the Euro Heart Survey Investigators. Management of coronary artery disease in patients with and without diabetes mellitus. Acute management reasonable but secondary prevention unacceptably poor—a report from the Euro Heart Survey on diabetes and the heart. Eur J Cardiovasc Prev Rehabil (2007) 14:28–36.[CrossRef][Web of Science][Medline]
24. Meinert CL, Knatterud GL, Prout TE, Klimt CR, University Group Diabetes Program. A study of the effects of hypoglycemic agents on vascular complications in patients with adult-onset diabetes II. Mortality results. Diabetes (1970) 19:789–830.[Web of Science][Medline]
25. Wikström G, Malmberg K, Rydén L. Improved knowledge of antidiabetic treatment—a necessity for the modern cardiologist. Eur Heart J (1999) 20:403–405.[Web of Science][Medline]
26. Riveline JP, Danchin N, Ledru F, Varroud-Vial M, Charpentier G. Sulfonylureas and cardiovascular effects: from experimental data to clinical use. Available data in humans and clinical applications. Diabetes Metab (2003) 29:207–222.[Web of Science][Medline]
27. UK Prospective Diabetes Study (UKPDS) Group. Intensive blood-glucose control with sulphonylureas or insulin compared with conventional treatment and risk of complications in patients with type 2 diabetes (UKPDS 33). Lancet (1998) 352:837–853.[CrossRef][Web of Science][Medline]
28. Nichols GA, Koro CE, Guillon CE, Ephross SA, Brown JB. The incidence of congestive heart failure associated with antidiabetic therapies. Diabetes Metab Res Rev (2005) 21:51–57.[CrossRef][Web of Science][Medline]
29. Van den Berghe G, Wouters PJ, Weekers F, Verwaest C, Bruyninckx F, Schetz M, Vlasselaers D, Ferdinande P, Lauwers P, Bouillon R. Intensive insulin therapy in critically ill patients. N Engl J Med (2001) 345:1359–1367.[Abstract/Free Full Text]
30. Van den Berghe G, Wouters P, Bouillon R, Weekers F, Verwaest C, Schetz M, Vlasselaers D, Ferdinande P, Lauwers P. Outcome benefit of intensive insulin therapy in the critically ill: Insulin dose versus glycemic control. Crit Care Med (2003) 31:359–366.[CrossRef][Web of Science][Medline]
31. Angiolillo DJ, Bernardo E, Ramírez C, Costa M, Sabaté M, Jimenez-Quevedo P, Hernández R, Moreno R, Escaned J, Alfonso F, Bañuelos C, Bass TA, Macaya C, Fernandez-Ortiz A. Insulin therapy is associated with platelet dysfunction in patients with type 2 diabetes mellitus on dual oral antiplatelet treatment. J Am Coll Cardiol (2006) 48:298–304.[Abstract/Free Full Text]
32. King GL, Brownlee M. The cellular and molecular mechanisms of diabetic complications. Endocrinol Meta Clin North Am (1996) 25:256–270.
33. Reaven GM, Lithell H, Landsberg L. Hypertension and associated metabolic abnormalities—the role of insulin resistance and the sympathoadrenal system. N Engl J Med (1996) 334:374–381.[Free Full Text]
34. Sobel BE. Effects of glycemic control and other determinants on vascular disease in type 2 diabetes. Am J Med (2002) 113:12S–22S.[CrossRef][Web of Science][Medline]
35. Kim J, Montagnani M, Koh KK, Quon MJ. Reciprocal relationship between insulin resistance and endothelial dysfunction. Molecular and pathophysiological mechanisms. Circulation (2006) 113:1888–1904.[Abstract/Free Full Text]
36. Arcaro G, Cretti A, Balzano S, Lechi A, Muggeo M, Bonora E, Bonadonna RC. Insulin causes endothelial dysfunction in humans. Sites and Mechanisms. Circulation (2002) 105:576–582.[Abstract/Free Full Text]
37. Kragelund C, Snorgaard O, Køber L, Bengtsson B, Ottesen M, Højberg S, Olesen C, Kjaergaard JJ, Carlsen J, Torp-Petersen C, on behalf of the TRACE study group. Hyperinsulinemia is associated with increased long-term mortality following acute myocardial infarction in non-diabetic patients. Eur Heart J (2004) 25:1891–1897.[Abstract/Free Full Text]
38. Ruiz-Torres A, Lozano R, Melón J, Carraro R. On how insulin may influence ageing and become atherogenic throughout the insulin-like growth factor-1 receptor pathway: in vitro studies with human vascular smooth muscle cells. Gerontology (2005) 51:225–230.[CrossRef][Web of Science][Medline]

CiteULike    Connotea    Del.icio.us    What's this?

Related articles in EHJ:

Glucose-lowering therapy after myocardial infarction: more questions than answers

Peter W. Radke and Heribert Schunkert
EHJ 2008 29: 141-143. [Extract] [FREE Full Text]  

This article has been cited by other articles:

				J. Wilding Results of the LEAD-2 study in relation to NICE and ADA/EASD guidelines for the treatment of type 2 diabetes in the UK The British Journal of Diabetes & Vascular Disease, July 1, 2009; 9(4): 177 - 185. [Abstract] [PDF]

				B. Hemmingsen, S. S Lund, J. Wetterslev, and A. Vaag Oral hypoglycaemic agents, insulin resistance and cardiovascular disease in patients with type 2 diabetes Eur. J. Endocrinol., July 1, 2009; 161(1): 1 - 9. [Abstract] [Full Text] [PDF]

				L G Mellbin, K Malmberg, A Waldenstrom, H Wedel, L Ryden, and for the DIGAMI 2 investigators Prognostic implications of hypoglycaemic episodes during hospitalisation for myocardial infarction in patients with type 2 diabetes: a report from the DIGAMI 2 trial Heart, May 1, 2009; 95(9): 721 - 727. [Abstract] [Full Text] [PDF]

				S. Palomba, A. Falbo, F. Zullo, and F. Orio Jr. Evidence-Based and Potential Benefits of Metformin in the Polycystic Ovary Syndrome: A Comprehensive Review Endocr. Rev., February 1, 2009; 30(1): 1 - 50. [Abstract] [Full Text] [PDF]

				F. Cosentino, L. Rydén, P. Francia, and L. G. Mellbin CHAPTER 14 Diabetes Mellitus and Metabolic Syndrome ESC Textbook of Cardiovascular Medicine, January 1, 2009; 2(1): med-9780199566990-chapter - med-9780199566990-chapter. [Abstract] [Full Text] [PDF]

				R. G. Garcia and P. Lopez-Jaramillo Cardiovascular prevention in high-risk patients with type 2 diabetes mellitus: when to start it? Eur. Heart J., August 2, 2008; 29(16): 2058 - 2058. [Full Text] [PDF]

				L. G. Mellbin, L. Ryden, K. Malmberg, A. Norhammar, and H. Wedel Cardiovascular prevention in high-risk patients with type 2 diabetes mellitus: when to start it?: reply Eur. Heart J., August 2, 2008; 29(16): 2058 - 2059. [Full Text] [PDF]

				U. N. Das Glucose, insulin, and coronary heart disease Eur. Heart J., April 2, 2008; 29(8): 1075 - 1076. [Full Text] [PDF]

				P. W. Radke and H. Schunkert Glucose-lowering therapy after myocardial infarction: more questions than answers Eur. Heart J., January 2, 2008; 29(2): 141 - 143. [Full Text] [PDF]

				M. Anselmino, J. Ohrvik, K. Malmberg, E. Standl, L. Ryden, and on behalf of the Euro Heart Survey Investigators Glucose lowering treatment in patients with coronary artery disease is prognostically important not only in established but also in newly detected diabetes mellitus: a report from the Euro Heart Survey on Diabetes and the Heart Eur. Heart J., January 2, 2008; 29(2): 177 - 184. [Abstract] [Full Text] [PDF]

This Article Abstract FREE Full Text (PDF) All Versions of this Article: 29/2/166    most recent ehm518v1 E-letters: Submit a response Alert me when this article is cited Alert me when E-letters are posted Alert me if a correction is posted Services Email this article to a friend Related articles in EHJ Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Archive Download to citation manager Search for citing articles in: ISI Web of Science (7) Request Permissions Disclaimer Google Scholar Articles by Mellbin, L. G. Search for Related Content PubMed PubMed Citation Articles by Mellbin, L. G. Social Bookmarking          What's this?

Online ISSN 1522-9645 - Print ISSN 0195-668x

Copyright © 2009 European Society of Cardiology

Oxford Journals Oxford University Press

• Site Map
• Privacy Policy
• Frequently Asked Questions

Other Oxford University Press sites: Oxford University Press Oxford Journals China Oxford Journals Japan American National Biography Booksellers' Information Service Children's Fiction and Poetry Children's Reference Corporate & Special Sales Dictionaries Dictionary of National Biography Digital Reference English Language Teaching Higher Education Textbooks Humanities International Education Unit Law Medicine Music Online Products Oxford English Dictionary Reference Rights and Permissions Science School Books Social Sciences Very Short Introductions World's Classics