Low vitamin D and risk of bacterial pneumonias: Mendelian randomisation studies in two population-based cohorts.
Level 3 - non-randomized controlled study
Prospective cohort study and Mendelian randomisation analysis
PubMed 33109689 · doi:10.1136/thoraxjnl-2020-215288
What was done
Researchers evaluated the association between plasma 25-hydroxyvitamin D levels and the risk of hospital-diagnosed bacterial pneumonia using observational and Mendelian randomisation designs. They studied 116,335 randomly chosen white Danish participants aged 20 to 100 years from the Copenhagen City Heart Study and the Copenhagen General Population Study followed from 1981 through 2018 (up to 38 years). Measured plasma 25-hydroxyvitamin D was available for 35,833 participants. Genetic analyses used instrumental variants around CYP2R1, DHCR7, GEMIN2, and HAL. Negative control outcomes (urinary tract infections, skin infections, sepsis, and gastroenteritis) were assessed to detect non-specific confounding.
What was found
During follow-up, 6,342 bacterial pneumonias occurred in observational analyses and 13,916 in genetic analyses. In multivariable-adjusted observational analyses, individuals with 25-hydroxyvitamin D <25 nmol/L had a higher risk of bacterial pneumonia compared to those with ≥25 nmol/L (HR 1.27, 95% CI: 1.16 to 1.40). In genetic analyses, each 10 nmol/L lower genetically predicted plasma 25-hydroxyvitamin D was associated with increased risk of bacterial pneumonia: Wald's ratio OR 1.12 (95% CI: 1.02 to 1.23), inverse-variance weighted OR 1.12 (95% CI: 1.04 to 1.20), MR-Egger OR 1.63 (95% CI: 0.96 to 2.78), and weighted median OR 1.15 (95% CI: 1.05 to 1.26). Associations were strongest for CYP2R1 variants. Neither observational nor genetic analyses showed associations with negative control infection outcomes.
Why it matters
These findings provide genetic evidence supporting a potential causal relationship between lifelong lower vitamin D concentrations and increased susceptibility to bacterial pneumonia, rather than the link being solely an artifact of reverse causation or lifestyle confounding.
Limits
The study population was entirely composed of white Danish individuals, limiting generalisability to other ancestral or geographic groups. Measured vitamin D levels were available in only a subset (35,833 of 116,335) of participants, the MR-Egger estimate had wide confidence intervals crossing the null, and outcomes were limited to hospital-diagnosed bacterial pneumonias, missing mild or outpatient-treated infections.
Cited by
- supports Mendelian randomization meta-analyses demonstrate that genetic single nucleotide polymorphisms causing lower circulating 25-hydroxyvitamin D levels are associated with significantly increased respiratory tract infection mortality, cancer mortality, and all-cause mortality, but not cardiovascular mortality.