Introduction: Chronic back pain (BP; back pain lasting more than three months) is a leading cause of disability worldwide, with prevalence exceeding 30% and multifactorial aetiology. Epidemiological observation suggests that previous infection with human cytomegalovirus (CMV) may be associated with BP. CMV infection is commonly acquired in childhood and 50% of adults have anti-CMV IgG. The aim of this study was to investigate a potential causal relationship between CMV infection and BP.
Methods: The UK Biobank (UKBB) was used to retrieve data for 9,689 participants having CMV status available. CMV seropositivity and BP status overlapped in 5,140 participants. A logistic regression model adjusting for age, sex, body mass index (BMI), ethnicity and socio-economic status was used to examine the association between CMV and BP. A genome-wide association study (GWAS), followed by a one-sample Mendelian randomization (MR) based on independent genetic variants predicting CMV positivity, was conducted in the Northern European samples using individual-level data for CMV and BP. To validate the causal association, the MR was conducted using the CMV polygenic risk score (PRS) generated and optimised using PRSice2 software. We used CMV PRS as an instrumental variable for CMV and fitted MR models using the two-stage least squares method.
Results: There was a significant association between CMV seropositivity and CBP (OR adj = 1.30; 95% CI: 1.144–1.478, p-value = 0.001). The GWAS of CMV using 8,883 Northern European samples followed by clumping revealed 86 SNPs with a p-value < 2x10-4 for one-sample MR. These SNPs were used as proxies of genetically predicted categories of CMV infection. The predicted categories were significantly associated with BP (OR = 1.150; 95% CI: 1.005–1.317, p-value = 0.043). Also, the causal association was confirmed by using a PRS calculated using 49,404 SNPS as an instrumental variable in the MR analysis (OR = 1.299; 95% CI: 1.141–1.479, p-value = 0.001).
Discussion: Our results provide evidence for a causal relationship between CMV infection and CBP. Further studies including a two-sample MR approach and tissue virology are being explored to understand further the role of common infection in BP and may lead to patient stratification and a personalised approach to treatment.
Acknowledgement: UKBB data were obtained under the project #18219.