The Lancet 2007; 370:667-675

Incidence of new-onset diabetes and impaired fasting glucose in
patients with recent myocardial infarction and the effect of clinical
and lifestyle risk factors

Dr Dariush Mozaffarian MD  a ,   RosaMaria Marfisi MS b,   Giacomo
Levantesi MD b,   Maria G Silletta MS b,   Luigi Tavazzi MD c,
Gianni Tognoni MD b,   Franco Valagussa MD ‡ d   and   Roberto
Marchioli MD b
‡F Valagussa (the GISSI-Prevenzione Chairman) died in December 2006

Background
Individuals with diabetes are at higher risk of myocardial infarction
than non-diabetics. However, much less is known about the incidence
of, and risk factors for, development of diabetes and impaired fasting
glucose in patients who have had a myocardial infarction. We set out
to estimate this incidence and investigate whether lifestyle factors
such as dietary habits might alter this risk.

Methods
We used prospectively obtained data for 8291 Italian patients with a
myocardial infarction within the previous 3 months, who were free of
diabetes (determined by medication use, a physician-reported
diagnosis, or fasting glucose ≥7 mmol/L) at baseline. Incidence of new-
onset diabetes (new diabetes medication or fasting glucose ≥7 mmol/L)
and impaired fasting glucose (fasting glucose ≥6·1 mmol/L and <7 mmol/
L) were assessed at follow-up at 0·5, 1·0, 1·5, 2·5, and 3·5 years.
Baseline data for body-mass index (BMI), other risk factors, dietary
habits, and medications were updated during follow-up. A Mediterranean
diet score was assigned according to consumption of cooked and raw
vegetables, fruit, fish, and olive oil. Associations of demographic,
clinical, and lifestyle risk-factors with incidence of diabetes and
impaired fasting glucose were assessed with multivariable Cox
proportional hazards.

Findings
During 26 795 person-years (mean follow-up 3·2 years [SD 0·9]), 998
individuals (12%) developed new-onset diabetes (incidence 37 cases per
1000 person-years). Of the 7533 without impaired fasting glucose at
baseline, 2514 (33%) developed new-onset impaired fasting glucose or
diabetes (incidence 123 cases per 1000 person-years), rising to 3859
(62%) of 6229 with the lower cutoff for impaired fasting glucose of
5·6 mmol/L (incidence 321 cases per 1000 person-years). Independent
risk factors for new-onset diabetes or impaired fasting glucose
included older age, hypertension, use of beta-blockers, lipid-lowering
medications (protective), and diuretic use. Independent lifestyle risk-
factors included higher BMI, greater BMI gain during follow-up,
current smoking, a lower Mediterranean dietary score, and wine
consumption of more than 1 L/day. Data for physical activity were
unavailable, but inability to perform exercise testing was associated
with higher incidence of diabetes and impaired fasting glucose.

Interpretation
Compared with population-based cohorts, patients with a recent
myocardial infarction had a higher annual incidence rate of impaired
fasting glucose (1·8 vs 27·5% in our study) and diabetes (0·8–1·6%
compared with 3·7%) in this study. Thus, our results indicate that
myocardial infarction could be a prediabetes risk equivalent. Smoking
cessation, prevention of weight gain, and consumption of typical
Mediterranean foods might lower this risk, which emphasises the need
for guidance on diet and other lifestyle factors for patients who have
had a myocardial infarction.

Affiliations

a. Departments of Medicine, Epidemiology, and Nutrition, Harvard
Medical School and Harvard School of Public Health, Boston, MA, USA
b. Consorzio Mario Negri Sud, Santa Maria Imbaro, Chieti, Italy
c. IRCCS Policlinico San Matteo, Pavia, Italy
d. Ospedale San Gerardo, Monza, Italy

Correspondence to Dr Dariush Mozaffarian, Departments of Medicine,
Epidemiology, and Nutrition, Harvard Medical School and Harvard School
of Public Health, 665 Huntington Ave, Boston, MA 02115