Most halogenated persistent organic pollutants (POP) are lipophilic. They sequester in lipid stores such as adipose tissue and breast milk. Studies have demonstrated that serum or plasma POP levels can vary widely depending on the blood lipid content.1 If data are lipid adjusted, POP concentrations among matrices (e.g., adipose tissue, breast milk, serum, or plasma) in the same person are roughly equivalent.2,3
For many decades, simple equations have been used to estimate the lipid content of a specimen and using that, to correct blood concentrations based on volume or weight (e.g., µg/L) of POP on a lipid basis (µg/g lipid). These equations use one of the following:
- Gravimetrically determined total lipids,
- Total lipids estimated with a “long equation” using free and total cholesterol, phospholipids, and triglycerides,1 or
- Total lipids estimated with a “short equation” using only total cholesterol and triglycerides.4,5
Alternatively, it is also common practice to adjust for lipids by including separate uncorrected toxicant and lipids variable(s) in the statistical model evaluating the relationship between the toxicant exposure and health outcome.
More recently, this practice has been questioned because of potential interactions of blood lipids with the outcomes being studied or because the total amount circulating in the blood, irrespective of lipid content, may be important for understanding the health impact.6-8
Directed acyclic graphs (DAGs) have been used to detail scenarios where lipid adjustment resulted in biased estimates of the relationship between the exposure and outcome of interest. In the literature, there has also been discussion of whether lipids are a true confounder that would require adjustment, or if they act as a concomitant variable, that is, a variable that is not a confounder but will improve precision of the estimate of interest if included in the analysis.9 If it is hypothesized that lipids are a concomitant variable in relation to a specific exposure–disease relationship, an alternative approach that utilizes a Box–Cox transformation and a simple Bayesian hierarchical model is suggested for optimal lipid adjustment.9
While lipid adjustment is still commonly and routinely used, its usefulness and most appropriate method in each data set should be evaluated independently. To enable comparison with other studies, we also recommend reporting lipid-adjusted concentrations in addition to whole weight concentrations even if lipid-adjusted values are not used in statistical models.
NOTES:
- Additional sample volumes are required for lipid measurements.
- A few POP, namely perfluorinated alkyl substances including perfluorooctane sulfonate (PFOS) and perfluorooctanoic acid (PFOA), are not lipophilic but adhere to albumin and therefore should not be adjusted for lipids.
- Phillips DL, Pirkle JL, Burse VW, et al. Chlorinated hydrocarbon levels in human serum: effects of fasting and feeding. Archives of Environmental Contamination and Toxicology. 1989;18(4):495-500.
- Patterson DG Jr, Needham LL, Pirkle JL, et al. Correlation between serum and adipose tissue levels of 2,3,7,8-tetrachlorodibenzo-p-dioxin in 50 persons from Missouri. Archives of Environmental Contamination and Toxicology. 1988;17(2):139-143.
- Ploteau S, Antignac JP, Volteau C, et al. Distribution of persistent organic pollutants in serum, omental, and parietal adipose tissue of French women with deep infiltrating endometriosis and circulating versus stored ratio as new marker of exposure. Environment International. 2016;97:125-136.
- Akins JR, Waldrep K, Bernert JT Jr. The estimation of total serum lipids by a completely enzymatic "summation" method. Clinica Chimica Acta. 1989;184(3):219-226.
- Bernert JT, Turner WE, Patterson DG Jr, et al. Calculation of serum "total lipid" concentrations for the adjustment of persistent organohalogen toxicant measurements in human samples. Chemosphere. 2007;68(5):824-831.
- Schisterman EF, Whitcomb BW, Louis GM, et al. Lipid adjustment in the analysis of environmental contaminants and human health risks. Environmental Health Perspectives. 2005;113(7):853-857.
- O'Brien KM, Upson K, Buckley JP. Lipid and creatinine adjustment to evaluate health effects of environmental exposures. Current Environmental Health Reports. 2017;4(1):44-50.
- O'Brien KM, Upson K, Cook NR, et al. Environmental chemicals in urine and blood: improving methods for creatinine and lipid adjustment. Environmental Health Perspectives. 2016;124(2):220-227.
- Li D, Longnecker MP, Dunson DB. Lipid adjustment for chemical exposures: accounting for concomitant variables. Epidemiology. 2013;24(6):921-928.