Association between Diet Quality and Risk of Type 2 Diabetes Mellitus in Patients with Coronary Heart Disease: Findings from the CORDIOPREV Study

Abstract

The incidence of type 2 diabetes mellitus (T2DM) is growing in Western countries. Nutritional interventions that promote high-quality dietary patterns could help reverse this trend. We aimed to evaluate whether changes in Nutrient-Rich Food Index 9.3 (NRF9.3) were related to the risk of developing T2DM in patients with coronary heart disease (CHD). The study was carried out in the context of two healthy dietary interventions (a Mediterranean and a low-fat diet). For this purpose, we evaluated all the patients in the CORDIOPREV study without T2DM at baseline. Data were obtained during the first 5 years of dietary intervention. The score was calculated using the Food Frequency Questionnaires at baseline and after 1 year of intervention. After 5 years of follow-up, 106 patients developed T2DM (incident-T2DM), while 316 subjects did not (non-T2DM). Total NRF9.3 score and changes during the first year of intervention were compared between incident-T2DM and non-T2DM. Incident-T2DM showed less improvement in NRF9.3 than non-T2DM (p = 0.010). In the multi-adjusted Cox proportional hazard study, patients with greater improvement in NRF9.3 had over 50% less risk of developing T2DM compared with the lowest tertile (HR 2.10, 95%, CI = 1.12-3.56). In conclusion, improved diet quality in terms of nutrient density after the dietary intervention was associated with a lower risk of T2DM in patients with CHD.

Keywords: NRF9.3; coronary heart disease; diet quality; nutrient density.

Figure 1

(A) Effect of the dietary intervention on NRF9.3 score and diabetes status. Data are presented as Δchanges of NRF9.3 produced between post- and pre-intervention ± SEM. Variables were compared using the analysis of variance (univariate ANOVA) adjusted by age, sex, statin treatment, smoking habits, BMI, LDL, HDL, TG, HOMA-IR, ISI, DI and IGI. Incident-T2DM (n = 106) and Non-T2DM (n = 316). Differences were considered to be significant when p < 0.05. * Significant differences between incident-T2DM and Non-T2DM. (B) Effect of the dietary intervention on AHEI-2010 score and diabetes status. Data are presented as Δchanges of AHEI-2010 produced between post- and pre-intervention ± SEM. Variables were compared using the analysis of variance (univariate ANOVA) adjusted by age, sex, statin treatment, smoking habits, BMI, LDL, HDL, TG, HOMA-IR, ISI, DI and IGI. Incident-T2DM (n = 106) and Non-T2DM (n = 316). Differences were considered to be significant when p < 0.05. (C) Effect of the dietary intervention on NRF9.3 score according to randomized diet group. Data are presented as Δchanges of NRF9.3 produced between post- and pre-intervention ± SEM. Variables were compared using the analysis of variance (univariate ANOVA) adjusted by age, sex, statin treatment, smoking habits, BMI, LDL, HDL, TG, HOMA-IR, ISI, DI and IGI. Low-fat diet (n = 190) and Mediterranean diet (n = 232). Differences were considered to be significant when p < 0.05. Abbreviation: NRF9.3, Nutrient-Rich Food index 9.3AHEI-2010, Alternative Healthy Eating Index-2010; BMI, Body mass index; LDL, Low-density lipoprotein; HDL, high-density lipoprotein; TG, tryglicerides; HOMA-IR, homeostatic model assessment; ISI, insulin sensitivity index; DI, disposition index; IGI, insulinogenic index.

Figure 2

Probability of T2DM development by COX analysis according to the tertiles of ΔNRF9.3. (A) unadjusted model. (B) adjusted model controlled for sex, statins, age, BMI, LDL, smoking habits, HDL, TG. C fully adjusted model controlled for sex, statins, age, BMI, LDL, smoking habits, HDL, TG., HOMA-IR, ISI, DI and IGI. Reference was the Tertile 1 (lowest). The hazard ratio (HR) between groups was calculated. Abbreviation: BMI, Body mass index; LDL, Low-density lipoprotein; HDL, high-density lipoprotein; TG, Tryglicerides, HOMA-IR, homeostatic model assessment; ISI, insulin sensitivity index; DI, disposition index; IGI, insulinogenic index. * p < 0.05.