Mendelian randomization clarifies the effects of alcohol on telomeres
The results of observational studies on the relationship between alcohol consumption and telomere size have been contradictory. In the largest study of this type to date, involving 4,567 participants, there was no correlation between alcohol consumption and baseline or longitudinal changes in telomere size [9]. A second analysis of two American cohorts with a total of 2623 participants also found null results [10]. A few smaller studies (sample sizes ranged from 255 to 1800) found associations between heavy drinking and AUD. Telomere length was reported to be shorter in AUD participants compared to controls [11]. In a longitudinal study of Helsinki-based businessmen, higher alcohol consumption in midlife was linked to shorter telomeres as people aged [8]. In a Korean study, drinking 30 grams of alcohol per day was linked to shorter telomeres among older participants [12]. The associations were stronger among those who experienced the alcohol flush, which raises the intriguing possibility of acetaldehyde being the toxic breakdown product of ethanol. A recent review of 27 research studies found that 10 showed significant correlations between alcohol consumption and telomere size [13]. These studies used cross-sectional as well as longitudinal designs. Most of the participants were Europeans whose ages ranged from third to seventh decades. In the majority of studies, alcohol consumption was positively associated with LTL. The heterogeneity of studies regarding the methods used to measure telomeres and categorize alcohol consumption hindered meta analysis and aggregated data.
MR aims to identify potential causal determinants for an outcome. It estimates the relationship between genetically predicted exposure levels and a desired outcome. The aim is to make residual confounding and reverse causality less problematic than with most other methods for analyzing observational data. With MR, genetic proxies can be used to study the effects of genetically-predicted variability in alcohol consumption or AUD risk. No MR study has been conducted to date on alcohol and telomeres.
We performed a large-scale observational study on two alcohol phenotypes: alcohol consumption, AUD and leucocyte. Then, we performed linear MR analysis to examine the evidence of a causal relationship between alcohol consumption/AUD/LTL. We compared the estimates generated by our genetic and observational methods. Their separate analysis is motivated by the genetic distinction between alcohol use traits. While moderately genetically related to AUD, other traits such as the number of drinks consumed per week or AUDIT C (Alcohol Use disorders Identification Test Consumption – a 3-item screening tool) have distinct patterns in genetic correlation [13]. As there have been many speculations about possible J-shaped relationships between drinking alcohol and health outcomes [15], a nonlinear MR was performed to examine the shape and relationship between alcohol consumption, and telomere size.
Source:
https://www.nature.com/articles/s41380-022-01690-9
