Modeling U-shaped dose-response curves for manganese using categorical regression

Area of Science:

• Nutritional Science
• Toxicology
• Biostatistics

Background:

• Manganese is an essential nutrient with a narrow optimal intake range.
• Both insufficient and excessive manganese intake can lead to adverse health effects.
• A U-shaped exposure-response relationship characterizes manganese's impact on health.

Purpose of the Study:

• To estimate the optimal dietary intake of manganese.
• To balance the risks associated with manganese deficiency and excess intake.
• To model the U-shaped exposure-response relationship for manganese.

Main Methods:

• Conducted a systematic literature review (1930-2013) on manganese deficiency and excess.
• Created a database of oral manganese toxicity from diverse studies.
• Standardized toxicological outcomes on an 18-point severity scale and used logistic regression for U-shaped curve modeling.

Main Results:

• Analyzed data from 6113 rats, establishing a comprehensive manganese toxicity database.
• The nadir of the U-shaped curve, representing optimal intake, was estimated at 2.70 mg/kg body weight/day (95% CI: 2.51-3.02).
• Both deficient and excess manganese intake were associated with a 90% probability of adverse events.

Conclusions:

• The developed manganese database aids in estimating optimal intake by integrating adverse effect data.
• Further human studies are crucial for validating and applying these findings.
• Interspecies differences in manganese sensitivity necessitate adjustments for translating rat data to humans.