Chronic obstructive pulmonary disease (COPD) is a heterogeneous disease characterized by chronic respiratory symptoms (dyspnea, cough, phlegm, acute exacerbation), which are caused by airway (bronchitis, bronchiolitis) and/or alveolar abnormalities (emphysema), causing continuous progressive exacerbation of restricted airflow (1). The pathogenesis of COPD primarily involves chronic airway inflammation, and oxidative and antioxidant imbalance. It is estimated that there are 300 to 400 million patients with COPD worldwide (2), making it the third leading cause of death globally (3), with approximately 3 million people dying from COPD each year. By 2050, the number of COPD patients is projected to reach 600 million (4). Given the significant health risks associated with COPD, exploring potential strategies to reduce mortality is essential to extend patients’ lives and alleviate the global burden of disease.

Dietary patterns have been proven to be related to many diseases. Current studies indicate that inflammatory diets increase all cause mortality including the risk of death from cancer and cardiovascular diseases (CVD) (5). On the other hand, diets with higher antioxidant capacity are associated with lower all-cause, cardiovascular, and respiratory disease mortality (6, 7).

The dietary inflammation index (DII) is an indicator based on the population’s intake of dietary nutrients. It is used to assess the inflammatory potential of diets (8), and is most widely used in epidemiological studies in order to explore the association between dietary inflammation potential and diseases (9–12). The composite dietary antioxidant index (CDAI) was developed to assess the antioxidant capacity of dietary intake (13).

Many studies have indicated that dietary factors are related to the onset and progression of COPD (14, 15). Two prospective cohort studies showed a positive relationship between DII and all-cause mortality in COPD patients or those with sarcopenia in the United States (16, 17). Another cross-sectional study found that CDAI is linearly negatively correlated with COPD risk (18). Moreover, cross-sectional and prospective cohort studies have confirmed that co-exposure to DII and CDAI has an impact on lung function [forced expiratory volume in 1 s (FEV1), forced vital capacity (FVC), preserved ratio impair spirometry (PRISm)] and all-cause death in patients with asthma (19, 20). At present, there has been no research conducted on the impact of both a pro-inflammatory diet and an antioxidant diet on COPD mortality. We aim to determine the effect of co-exposure to DII and CDAI on all-cause mortality in COPD patients aged 40 and above by utilizing the data from the National Health and Nutrition Examination Survey (NHANES) database from 2007 to 2018.

Eventually, a total of 1,457 patients with COPD were included. Males accounted for 51.58% of all the citizen representatives. Table 1 described the clinical characteristics of the participants, categorized by survival outcomes. Deceased patients had higher DII scores and lower CDAI scores.

Adjusting for all potential confounding factors, weighted multivariable Cox regression analysis revealed that both the crude model and the adjusted model indicated a positive association between DII and all-cause mortality in COPD, while CDAI was negatively associated with all-cause mortality (Table 3).

Compared to the lowest tertile of DII, the weighted multivariate HR (95% CI) of all-cause death was 1.71(1.18, 2.48) in Q2 and 1.36 (1.00, 1.85) in Q3 (P for trend = 0.02). In addition, using the Q1 level of CDAI as a reference, the HR (95% CI) of Q2 and Q3 were 1.30 (0.97, 1.75), 0.63 (0.43, 0.92) respectively (p = 0.01) (Table 3).

The threshold for CDAI associated with all-cause mortality is −0.12. When CDAI > −0.12, for every unit increase in CDAI, the risk of all-cause mortality decreased by 9% [HR (95% CI): 0.91 (0.87, 0.96); p = 0.001]; When CDAI < −0.12, it was not associated with all-cause mortality [HR (95% CI): 1.07 (0.99, 1.15); p = 0.075].

To our knowledge, this survey was the first to explore the impact of DII and CDAI on all-cause mortality in COPD patients. After adjusting for potential confounding factors, there was a nonlinear association between DII or CDAI and all-cause deaths, with inflection points of 2.32 and − 0.12, respectively. The results showed that higher DII and lower CDAI increased all-cause deaths from COPD. In addition, these relationships were consistent across many subgroups. Pro-inflammatory diets or antioxidant diets have a more pronounced effect on death in non-smokers.

Chronic airway inflammation is one of the main characteristics of COPD. Airway inflammation is associated with the progression and mortality of COPD disease (22, 23). Data from Pan MH et al. paper stated that dietary factors can alter the expression of inflammatory genes in the human body (24). Epidemiological studies have found that a higher DII score was positively correlated with the risk of COPD or early COPD and a decline in lung function (25, 26). We further analyzed the impact of pro-inflammatory diets on all-cause mortality in patients with COPD, which was consistent with the research of Tian et al. (17). These findings support that dietary modification can be an intervention to improve outcomes in patients with COPD.

The mechanism by which DII affects all-cause mortality in COPD patients might involve elevating inflammatory factors. A large number of studies have confirmed that DII is positively associated with inflammatory markers (e.g., CRP, IL-6, TNF-α, IL-1, 2, IFN-γ) of a variety of chronic diseases (27–32). Consumption of vegetables, fruits, and legumes rich in fiber and vitamins is associated with lower markers of inflammation (33–35). This suggests that dietary factors may affect the prognosis of COPD by affecting the level of inflammatory markers in the body. It is suggested that anti-inflammatory diets may help reduce inflammatory responses in the body. We should pay attention to the impact of unhealthy eating habits on public health.

Oxidative stress is the primary driving mechanism in the development of COPD (36, 37). Smoke, air pollution, and lung inflammation generate reactive oxygen species, leading to an increase in oxidative stress in COPD patients. This oxidative stress further amplifies chronic inflammation and accelerates lung aging (36). Many studies have shown that an antioxidant-rich diet can lower the risk of COPD, improve lung function, and improve inflammatory status (38–43). Two cross-sectional studies have revealed that CDAI can reduce the risk of chronic respiratory diseases and COPD (18, 44). A cross-sectional study from the United States has indicated that CDAI is positively correlated with FEV1, FVC, and PRISm (19). However, previous research has not investigated the impact of CDAI on all-cause mortality in COPD. This study found that CDAI can reduce all-cause mortality in COPD, with a turning point at −0.12. When CDAI > −0.12, with an increase in CDAI, the all-cause mortality in COPD patients decreased. Moreover, this negative correlation still existed across most subgroups. This indicates that increasing antioxidant-rich diets may be an effective strategy to improve the prognosis of COPD.

The pro-inflammatory diet mainly includes alcohol, processed meat, red meat (such as pigs, cattle, and lamb), refined carbohydrates (such as cakes), fried foods, sugary beverages, and saturated fatty acids (45). Vegetables, fruits, dietary fiber, legumes, whole grains, tea, coffee, and foods rich in unsaturated fatty acids (such as nuts, fish, melon seeds, and olives) are called anti-inflammatory foods (46). Diet has the potential to regulate inflammation and influence the outcomes of infections and chronic diseases (47). The dietary pyramid of patients with COPD pointed out that it is beneficial to adjust their lifestyle by increasing the intake of a large number of fruits and vegetables, dietary fiber, and antioxidants to improve FEV1 and reduce oxidative stress (48). It is recommended that patients with COPD should reduce the intake of pro-inflammatory foods and increase the intake of foods with specific antioxidant functions.

This study has some advantages and disadvantages. It is the first to analyze the impact of exposure factors DII and CDAI on all-cause mortality in COPD, using complex sampling weighting analysis to represent the general COPD patients in the United States, and adjusting for as many potential confounding effects as possible. Of course, the research has its limitations. First of all, from 2013 to 2018, the lack of pulmonary function data in the database led to the diagnosis of COPD relying on questionnaires, which may have resulted in either overestimation or underestimation. This method does not fully reflect the true situation of COPD. If this part of the population has spirometry data, it will improve the robustness of our research results. Secondly, dietary recall data cannot represent long-term dietary habits, which limits our analysis of the relationship between changing dietary factors and all-cause death. In addition to further strengthening large-scale prospective studies in the future, intervention trial studies should also be gradually carried out to provide stronger evidence for clarifying the relationship of DII and CDAI with COPD.

Adopting a healthier diet as part of a healthy lifestyle to protect lung function and prevent or improve COPD has great potential to improve public health (49). In addition to encouraging avoidance or quitting smoking, our results revealed a maximum pro-inflammatory inflection point of the diet is 2.32, while the ideal CDAI level of the antioxidant diet is above −0.12. We recommend eating more antioxidant foods and less inflammatory foods under the premise of balanced nutrition, following current nutritional guidelines such as the COPD diet pyramid.

These findings indicate that inflammatory diets can increase the all-cause mortality risk in COPD patients, while antioxidant diets can decrease all-cause mortality in COPD. It suggests that healthy dietary patterns can be used as interventions and management measures for COPD. Of course, further confirmation of the relationship between the two would require additional cohort studies and research into potential mechanisms.