Association between dietary cholesterol, serum lipids, and cancer-specific mortality in U.S. adults.

e22609 Background: The relationship between dietary cholesterol and its impact on serum lipids and health outcomes remains a topic of debate. This study aims to investigate whether dietary cholesterol intake is associated with lipid abnormalities and cancer-specific mortality in a nationally representative U.S. adult population. Methods: We included adults > 20 years from the 1999-2018 National Health and Nutrition Examination Surveys, excluding those consuming < 500 or > 4000kcal/day, or taking anti-lipid agents. Dietary cholesterol intake was categorized into ≤100, 101-200, 201-300, 301-400, >400mg/day based on two 24-hour dietary recalls. Outcomes were abnormal serum lipids (triglycerides >150, HDL <40 for men, <50 for women, and LDL >160mg/dl) and cancer-specific determined via linkage to the 2019 National Death Index. Odds of abnormal lipids were estimated using logistic regression, adjusting for demographics, comorbidities, & dietary intake. Cancer mortality risk was estimated using Cox regression in adults without cancer. Analyses accounted for survey design & weights for nationally representative estimates. Results: Our cohort included 1,909 cancer survivors (mean age 62y, 61% female, 84% White) and 23,238 adults without cancer (mean age 47y, 51% female, 70% White). Adults consuming > 400mg/day of dietary cholesterol vs. < 100mg/day were more likely to have higher protein intake (men: 117 vs. 50g/day, women: 93 vs. 43g/day), higher % saturated fat (men: 12% vs. 8.1%, women: 13% vs. 9.0%), higher total caloric intake (men: 2,708 kcal vs. 1,619kcal, women: 2,186 vs. 1,322kcal) and similar fiber intake (men: 18.7g vs. 20.0g, women: 16.1g vs. 14.7g; p = 0.15). In adults without cancer, higher dietary cholesterol was associated with high triglycerides (adjusted OR [95% CI], 400mg/day vs. <100mg/day: 1.22 [1.01, 1.49]; p-trend = 0.019) but not with low HDL (1.03 [0.84, 1.25]) or high LDL (0.89 [0.63, 1.24]). These associations were similar in cancer survivors. Over a median follow-up of 8 years (IQR, 4-11), there were a total of 564 cancer-specific deaths. Protein intake > 0.8g/kg/day modified the association between dietary cholesterol and cancer-specific mortality (p-interaction = 0.01). In those consuming protein > 0.8g/kg/day, there was no association between dietary cholesterol and cancer mortality (adjusted HR [95% CI]; > 400mg, 0.97 [0.48, 1.96]; 301-400mg, 1.13 [0.54, 2.36]; 201-300mg, 0.65 [0.32, 1.30]; 101-200mg, 0.95 [0.45, 2.02]); however, in those consuming < 0.8g/kg/day, dietary cholesterol did increase risk of cancer mortality ( > 400mg, 1.18 [0.53, 2.65]; 301-400mg, 2.32 [1.21, 4.44]; 201-300mg, 2.02 [1.18, 3.48]; 101-200mg, 1.80 [1.11, 2.91]). Conclusions: Dietary cholesterol >400mg/day was associated with higher triglyceride levels but not HDL or LDL in both adults with & without cancer. Dietary cholesterol only increased cancer mortality risk in those with protein intake < 0.8g/kg/day. Further studies on sources of cholesterol and impact on health outcomes are needed.