Department of Clinical Sciences, Lund University, Jan Waldenströms gata 35, 91-12-027, SE-214 28, Malmö, Sweden.
Department of Emergency and Internal Medicine, Skåne University Hospital, Malmö, Sweden.
BMC Med. 2022 Apr 21;20(1):122. doi: 10.1186/s12916-022-02326-z.
The global burden of cardiovascular disease and type 2 diabetes could be decreased by improving dietary factors, but identification of groups suitable for interventional approaches can be difficult. Reporting of dietary intake is prone to errors, and measuring of metabolites has shown promise in determining habitual dietary intake. Our aim is to create a metabolic signature that is associated with healthy eating and test if it associates with type 2 diabetes and coronary artery disease risk.
Using plasma metabolite data consisting of 111 metabolites, partial least square (PLS) regression was used to identify a metabolic signature associated with a health conscious food pattern in the Malmö Offspring Study (MOS, n = 1538). The metabolic signature's association with dietary intake was validated in the Malmö Diet and Cancer study (MDC, n = 2521). The associations between the diet-associated metabolic signature and incident type 2 diabetes and coronary artery disease (CAD) were tested using Cox regression in MDC and logistic regression in Malmö Preventive Project (MPP, n = 1083). Modelling was conducted unadjusted (model 1), adjusted for potential confounders (model 2) and additionally for potential mediators (model 3).
The metabolic signature was associated with lower risk for type 2 diabetes in both MDC (hazard ratio: 0.58, 95% CI 0.52-0.66, per 1 SD increment of the metabolic signature) and MPP (odds ratio: 0.54, 95% CI 0.44-0.65 per 1 SD increment of the metabolic signature) in model 2. The results were attenuated but remained significant in model 3 in both MDC (hazard ratio 0.73, 95% CI 0.63-0.83) and MPP (odds ratio 0.70, 95% CI 0.55-0.88). The diet-associated metabolic signature was also inversely associated with lower risk of CAD in both MDC and MPP in model 1, but the association was non-significant in model 3.
In this proof-of-concept study, we identified a healthy diet-associated metabolic signature, which was inversely associated with future risk for type 2 diabetes and coronary artery disease in two different cohorts. The association with diabetes was independent of traditional risk factors and might illustrate an effect of health conscious dietary intake on cardiometabolic health.
通过改善饮食因素，全球心血管疾病和 2 型糖尿病的负担可以减轻，但确定适合干预措施的人群可能具有挑战性。饮食摄入的报告容易出现错误，而代谢物的测量已显示出确定习惯性饮食摄入的潜力。我们的目的是创建与健康饮食相关的代谢特征，并检验其与 2 型糖尿病和冠状动脉疾病风险的相关性。
使用包含 111 种代谢物的血浆代谢物数据，使用偏最小二乘（PLS）回归确定与马尔默后代研究（MOS，n=1538）中健康饮食模式相关的代谢特征。在马尔默饮食与癌症研究（MDC，n=2521）中验证代谢特征与饮食摄入的相关性。使用 Cox 回归在 MDC 中测试与饮食相关的代谢特征与 2 型糖尿病和冠状动脉疾病（CAD）事件之间的关联，并在马尔默预防项目（MPP，n=1083）中使用逻辑回归进行测试。在未调整模型（模型 1）、调整潜在混杂因素模型（模型 2）和进一步调整潜在中介因素模型（模型 3）中进行建模。
在调整潜在混杂因素后（模型 2），该代谢特征与 MDC（风险比：0.58，95%置信区间 0.52-0.66，代谢特征每增加 1 个标准差）和 MPP（比值比：0.54，95%置信区间 0.44-0.65，代谢特征每增加 1 个标准差）中的 2 型糖尿病风险降低相关。在模型 3 中，结果减弱但仍有统计学意义，在 MDC（风险比 0.73，95%置信区间 0.63-0.83）和 MPP（比值比 0.70，95%置信区间 0.55-0.88）中也是如此。在模型 1 中，该饮食相关的代谢特征也与 MDC 和 MPP 中 CAD 风险降低呈负相关，但在模型 3 中无统计学意义。
在这项概念验证研究中，我们确定了一个与健康饮食相关的代谢特征，该特征与两个不同队列的未来 2 型糖尿病和冠状动脉疾病风险呈负相关。与糖尿病的相关性独立于传统风险因素，可能表明健康饮食对心血管代谢健康的影响。
