To the best of our knowledge, our study is the first prospective cohort study to reveal the association between the NHHR and the risk of all-cause and cardiovascular mortality in participants with diabetes or prediabetes. Our findings indicate a U-shaped correlation between the NHHR and all-cause mortality, as well as an L-shaped correlation with cardiovascular mortality among participants with diabetes or prediabetes. The results of the threshold effect analysis reveal that the NHHR exhibits inflection points at 2.71 and 2.85 for the risks of all-cause and cardiovascular mortality, respectively. Our study underscores the significance of the NHHR as a valuable clinical prognostic indicator for both all-cause and cardiovascular mortality in patients with diabetes or prediabetes. Furthermore, it holds immense importance for disease risk stratification and prognosis. The subgroup analysis demonstrated ethnic-specific differences in the relationship between the NHHR and cardiovascular mortality.
The NHHR is a novel lipid ratio indicator that is cost-effective and easily obtainable, and previous studies have revealed its clinical predictive value in various diseases. A longitudinal cohort study conducted by Sheng et al. [14] involving 15,464 participants demonstrated that the NHHR is a superior indicator for predicting the risk of diabetes compared to other conventional lipid markers, with an inflection point at approximately 2.74. This finding essentially aligns with the results of our study. You et al. [25] reported a correlation between the NHHR and coronary artery disease, with an elevated NHHR being associated with an increased risk of acute coronary syndrome. Similarly, Mao et al. [17] demonstrated that the NHHR serves as an independent predictor for severe coronary artery disease and major adverse cardiovascular events and is associated with the prognosis of patients with non-ST-segment elevation myocardial infarction. These studies indirectly support the conclusions of our research. Additionally, other studies have revealed a relationship between the NHHR and depression as well as suicidal ideation in adults, shedding light on the connection between lipid metabolism and mental health [23, 26].
Lipid markers provide crucial information about lipid metabolism and an individual’s health status and can be used to predict the risk of certain diseases [27]. Although the biological mechanisms linking the NHHR to mortality remain unclear, several studies have suggested a potential association between the NHHR and atherosclerosis [14, 28]. Arteriosclerosis in patients with diabetes is associated with several pathological mechanisms, including diabetes-related dyslipidemia, hyperglycemia, and insulin resistance, all of which play important roles in the development of arteriosclerosis and are associated with adverse CVD outcomes [29, 30]. In 2013, the American Heart Association/American College of Cardiology reported that elevated non-HDL-C levels were the underlying cause of arteriosclerosis and the core driver of ASCVD [31]. Non-HDL-C comprises LDL-C and all other components contributing to the development of arteriosclerosis, making it a superior predictor for ASCVD [32–34]. Conversely, HDL-C may exert anti-inflammatory, antioxidant, and anti-atherogenic effects, displaying a negative correlation with the incidence of ASCVD[35, 36]. The NHHR incorporates all lipid-related information relevant to both atherogenic and anti-atherogenic processes, offering a more comprehensive depiction of their balance. An increased NHHR may contribute to vascular endothelial layer impairment and the accumulation of atherosclerotic plaques. The ultimate outcome of this process can be thrombotic events triggered by plaque erosion or rupture, ultimately leading to acute cardiovascular events and, consequently, patient mortality.
Our study revealed a negative correlation between the NHHR and age at baseline, while a positive correlation was observed between the NHHR and BMI. This finding is partially supported by Zhang et al.[37], who found a negative correlation between non-HDL-C levels and age in participants aged over 57 years. Our study also revealed a U-shaped correlation between the NHHR and all-cause mortality in participants with diabetes or prediabetes. Additionally, the NHHR exhibited an L-shaped association with cardiovascular mortality. We further developed a segmented Cox proportional hazards model, and the results revealed that when the baseline NHHR was below the inflection point, the NHHR was negatively associated with both all-cause and cardiovascular mortality. For every unit increase in the NHHR, the all-cause and cardiovascular mortality decreased by 22% and 20%, respectively. Conversely, when the NHHR exceeded the inflection point, there was a positive correlation between the NHHR and all-cause mortality. For every unit increase in the NHHR, the all-cause mortality increased by 7%.
Our results are partially supported by multiple studies. A meta-analysis[38] showed that compared to coronary heart disease patients with low baseline levels of non-HDL-C, those with high levels of non-HDL-C experienced a 24% increase in mortality. This could be attributed to the accelerated development of atherosclerosis associated with extremely high non-HDL-C levels, leading to an elevated risk of mortality. However, extremely low levels of non-HDL-C can also lead to an increased risk of mortality. Several studies[39–41] have shown a U-shaped relationship between non-HDL-C and all-cause and cardiovascular mortality in populations of hypertensive participants, those with chronic kidney disease, and males not receiving statin therapy. These findings indicate the necessity of maintaining non-HDL-C levels within a reasonable range. Furthermore, two prospective cohort studies have shown a U-shaped relationship between HDL-C levels and all-cause mortality, indicating that both excessively high and excessively low HDL-C levels can increase the risk of death[42]. Another study from the CANHEART cohort supported this notion, demonstrating that participants with very low HDL-C levels (≤ 30 mg/dl) have higher rates of cardiovascular and non-cardiovascular mortality compared to those with intermediate levels of HDL-C[43].
Notably, our study demonstrates that both excessively high and excessively low NHHRs are associated with an increased risk of mortality. We also found that excessively low NHHRs can lead to a significant increase in the risk of mortality. Possible mechanisms for this observation are as follows: (1) patients with lower levels of TC may experience poorer health conditions[44] or a decrease in cholesterol levels due to weakness and illness[45, 46]; (2) according to lipid calculation formulas, there is typically an inverse relationship between HDL-C and non-HDL-C, with low non-HDL-C levels being equivalent to high HDL-C levels. Furthermore, studies have shown that high HDL-C levels can increase the risk of mortality in patients[39, 42]. Therefore, the specific mechanisms underlying the association between the NHHR and the risk of diabetes-related mortality remain to be further investigated.
We further investigated the correlation between the NHHR and mortality in participants with diabetes and those with prediabetes, respectively. The results revealed a U-shaped relationship between the NHHR and all-cause mortality in participants with diabetes, while a similar L-shaped correlation was observed with cardiovascular mortality. In participants with prediabetes, the NHHR was still associated with all-cause mortality in a U-shaped manner but showed a linear correlation with cardiovascular mortality. The aforementioned discrepancies may be attributed to the relatively small number of participants experiencing cardiovascular mortality in the study. In the subgroup analysis, we found an interaction effect between the NHHR and race in terms of cardiovascular mortality, indicating that racial factors could potentially influence the association between NHHR and cardiovascular mortality. The sensitivity analysis results indicated that the relationships between the NHHR and both all-cause and cardiovascular mortality remained relatively stable when participants with self-reported cancer at baseline, those who died within the first 2 years of follow-up, and those with extreme NHHRs were excluded.
There are several limitations in this study. First, as an observational study, we cannot establish a causal relationship between the NHHR and mortality. Second, this study only assessed the prognostic value of the baseline NHHR and did not investigate the association between changes in the NHHR over time and mortality. Third, we did not adjust for diabetes classification, disease duration, or medication use, which could lead to potential biases. Fourth, we did not control for the influences of factors such as diet, season, or the use of lipid-lowering medications on the NHHR or lipid levels. Fifth, the study population consisted primarily of the general population in the US, so caution should be exercised when extrapolating the findings to other ethnicities. Finally, the proportion of patients with CVD outcomes in the study population was relatively small, which may have limited the statistical power to detect differences between groups.