AR and psychiatric disorders may share many physiological pathways, including inflammation and sleep disturbances. Inflammation plays an important role in psychiatric disorders beyond depression, deemed as a breakthrough finding in the last two decades.21 However, in the aspect of the neurophysiological mechanism, the possibility that psychiatric disorders may be related to nasal inflammatory processes involving the intranasal and the other pathways have never been considered. The contributions of nurture (i.e. the environment) and nature (i.e. genetics) have been long touted for their aetiological importance in both in AR and psychiatric disorders progression and clinical onset. Disappointingly, knowledge gains in these areas, while individually successful, have to a large extent occurred in isolation from each other.
In the present MR study, we systematically assessed the associations of AR with a broad range of psychiatric disorders. Our study demonstrated that genetically predicted AR was associated with an increased risk of bipolar disorder and MDD. We did not find any evidence of causal associations between AR and the other psychiatric diseases. Bipolar disorder and ADHD may be a protective factor for AR but suffers from low power, we found no evidence of a causal association between the other psychiatric diseases and AR.
Bipolar disorder has a high heritability (approximately 70%). There seems to be some common genetic risk with conditions such as Alzheimer’s disease, diabetes and coronary heart disease.22 Hypertension (20.5%), asthma (12.5%) and hypothyroidism (8.1%) were the top medical comorbidities found in bipolar patients.23 Specifically, asthma patients had higher rates of bipolar disorder than controls, and BD patients had higher rates of asthma than controls.24 A retrospective cohort study using the Taiwan National Health Insurance Research Database reported an increased risk (HR = 4.62, 95% CI 3.17–6.75) of developing BD in later life among those with AR.25 This bidirectional MR study determined a negative association between bipolar disorder and AR but had inadequate power, it showed that bipolar disorder may be a possible protective factor for AR. Several evidence has shown that patients with AR may develop bipolar disorder, Manalai et al26 reported that pollen-specific immunoglobulin E positivity was associated with worsening depression scores in bipolar disorder patients during high pollen season. Chen et al27 reported a significantly increased incidence of bipolar disorder in cases of AR (0.77 vs. 0.18 per 1000 person-years) and an increased risk (HR = 4.62, 95% CI 3.17–6.75) of developing bipolar disorder in later life among those with AR. At present, there is no clinical and experimental data to support this conclusion. Therefore, our results should encourage further examine causality, epidemiology, mechanisms, and treatment potential for bipolar disorder on AR.
The cytokines including interleukin (IL)-1β, IL-6 and tumor-necrosis factor-α (TNF-α) play a role in the pathophysiology of both AR and bipolar disorders.28 However, a recent MR study showed that 20 immunological proteins/traits (pro-inflammatory cytokines: IL-6, TNF-α, IL-12, IL-16, IL-17, IL-18; anti-inflammatory cytokines: IL-1 receptor antagonist, IL-10, IL-13; chemokines: IL-8, monocyte chemo-attractant protein-1; lymphoid growth-factors: soluble (s) IL-2Rα, IL-4, IL-7, IL-9; myeloid growth-factor: IL-5; acute phase protein: C-Reactive Protein; immune cells: neutrophils, lymphocytes; neurotrophic factor: brain derived neurotrophic factor) were not associated with bipolar disorder.28 There is no doubt at all that the strongest and most consistent positive association were observed between AR and bipolar disorder in this MR study, further studies are warranted to explore these associations in a causal fashion.
AR is more prevalent in patients with MDD and those having family history of depressive disorder.29 The patients with AR had a 1.7-fold higher chance of being diagnosed with depression than did the controls.30 A meta-analysis showed that patients with AR had 1.6 times more chances to present depression compared to healthy controls.31 Thus, a close association between AR and MDD has been repeatedly identified by both observational and experimental studies. The present MR study provided new evidence that genetic predisposition to AR was associated with an increased risk of MDD. There is a major need of experimental and longitudinal investigations, as well as clinical trials to further our understanding of best practices in patients with comorbid AR and MDD and to find new therapeutic and preventative interventions.
ADHD is a severely impairing neurodevelopmental disorder affecting up to 7.1% in children and adolescents.32 Several epidemiological studies have reported that children with ADHD have a high risk of developing allergic diseases, but the explanation for the allergic disease manifestations remained ambiguous.32 A recently systematic review and meta-analysis confirmed that children with ADHD experienced 59% greater odds of having AR relative to the children without ADHD. The low statistical power most likely resulted from the statistical heterogeneity and the methodological diversity of the included studies, with a further potential confounder in factors that were not measured (e.g., climate, pollution or microbial agents).33 Some observation studies have reported no evidence of a link between allergy and ADHD.34 In recently MR study showed that there was little evidence for inferring a causal effect of ADHD on AR.35 We also confirmed there was no evidence for potential causal effects of ADHD, bipolar disorder, or schizophrenia on AR using the IVW method. Our MR-PRESSO analysis indicated the potential of outlier (rs112984125) affecting ADHD with AR (ukb-a-447) as outcomes (p = 0.0065). After the outlier correction, MR-PRESSO outlier test further supported the significant association with OR = 0.984, 95%CI: 0.973–0.995, P = 0.006. The results revealed that ADHD was a possible protective factor for AR. At present, there is no clinical and experimental data to support this conclusion. Therefore, our results should encourage further examine causality, mechanisms, and treatment potential for ADHD on AR.
In this study, we used a bidirectional design to assess the association of AR (Exposures) with psychiatric disorders (Outcomes) as well as to test whether psychiatric disorders (Exposures) causes AR(Outcomes). There are several strengths in this study. First, cause-effect relationship could not be possibly established by observational study, including meta-analysis from well-designed studies. MR analysis can work as an analogue to randomized controlled trial (RCT), likelihood of confounding effects and maximal avoidance of reverse causality. To our knowledge, this is the first study that investigated the potential bidirectional associations between the genetic liability for allergic rhinitis and psychiatric disorders using bidirectional MR. Second, population stratification bias was reduced in this study because SNP selection and summary-level data of psychiatric disorders were merely based on individuals of European descent and population structure was adjusted for in the GWASs. Moreover, we used all GWAS studies related to AR or psychiatric disorders for MR analysis, effectively avoiding selection bias. Among them, different AR related exposures were all significative risk factors for bipolar disorder, greatly increasing the reliability of this study.
Several limitations shall be noted in interpreting our study. First, we are unable to test for a comprehensive causal association between AR and psychiatric disorders, 4 of 11 psychiatric disorders summary statistics which we used as exposures were only available for significantly associated SNPs, the other psychiatric disorders including anorexia nervosa, anxiety disorder, ASD, insomnia complaints, OCD, panic disorder and PTSD were excluded from further analysis due to insufficient number of SNPs. Second, given bipolar I disorder and bipolar II disorder shared distinct characteristics, we were unable to examine the association between AR and different bipolar disorder phenotypes.36 Consequently, if our included genetic variants did not represent the risk of all subtypes of bipolar disorder, our obtained effect estimate was difficult to interpret. Third, bidirectional analyses could potentially detect dynastic effects. When findings suggested a potentially causal effect, possible bidirectional effects were further tested. There was no bidirectional causal relationship between AR and psychiatric disorders. Nevertheless, our study alone did not provide definitive evidence to absence of a causal association. Alternatively, there is possibility that the effect is nonlinear. Fourth, the summary level data that we used did not allow for stratified analyses by covariates of interest, such as age, sex, body mass index, smoking, and use of nasal hormone therapy. It is worth noting that AR is mainly determined by physical health and other environmental factors. Hence, it is inappropriate to deduce that genetic effects are independent of environmental factors.