The association of asthma, nasal allergies, and positive skin prick tests with obesity, leptin, and adiponectin.

BACKGROUND

Cross-sectional and longitudinal reports show that obese adults have more asthma than non-obese adults. A proposed mechanism is via effects of adipokines (leptin and adiponectin) on the immune system.

OBJECTIVE

We wished to measure the associations of asthma and other atopic diseases with serum adipokine levels and to find whether the associations with asthma were strong enough to rule out the possibility that they are secondary to the association of fatness measures with asthma.

METHODS

The Global Asthma and Allergy Network of Excellence (GA(2) LEN) clinical follow-up survey is a clinical survey, embedded in a larger multi-centre cross-sectional postal survey, involving, with a case/control design, enrichment of the sample with subjects with asthma and chronic rhinosinusitis (CRS). We recorded serum leptin or adiponectin in 845 men and 1110 women in 15 centres and also anthropometric measures of fatness including body mass index and waist/hip ratio, current asthma, and specific skin prick and IgE sensitisation. We used inverse sampling-probability-weighted rank and regression statistics to measure population associations of disease outcomes with adipokines in males and females, adjusting for confounders (area, age, smoking history, and number of elder siblings) and also mutually adjusting associations with adipokines and fatness measures.

RESULTS

One thousand nine hundred and fifty-five subjects aged 16-77 years had information on leptin or adiponectin levels. Leptin and leptin/adiponectin ratio were positively associated with the level of asthma, especially in females (Somers' D of leptin by asthma score, 0.20; 95% CI, 0.08-0.30; P = 0.00079). These associations were attenuated after adjusting for confounders and became non-significant after additionally adjusting for fatness measures and multiple comparisons.

CONCLUSIONS

Asthma levels are positively associated with serum leptin. However, we cannot rule out the possibility that this association is secondary to associations of both with fatness measures.