Epidemiology of Cholesteatoma in the UK Biobank

Objectives: To identify factors associated with cholesteatoma in a large UK cohort. Although some risk factors are frequently reported (male sex, history of chronic otitis media), other associations require further evidence (deprivation, smoking). Design and Setting: Briefly, 1140 cholesteatoma cases from UK BioBank were compared to 4551 non-cholesteatoma middle ear disease and 493 832 ear disease-free controls. Adjusted odds ratios were calculated for demographic factors including age, sex, ethnicity, deprivation and smoking status with logistic regressions. Odds ratios for overlapping ICD-10 codes are also calculated. Results: Cholesteatoma was significantly associated with sex (Adjusted odds ratio (AOR) for males = 1.33, 95%CI = [1.179–1.491]), age (AOR = 1.02, 95%CI = [1.011–1.026]) and deprivation (AOR = 1.08, 95%CI = [1.059–1.097]) compared to ear disease-free controls (p < 0.001). Age and deprivation distributions for cholesteatoma and non-cholesteatoma ear disease were similar. Although there was no significant association with smoking status, cholesteatoma was significantly associated with the ICD-10 code mental and behavioural disorders due to tobacco use (OR = 2.34, p < 0.001, 95%CI = [1.942, 2.813]). Cholesteatoma was also strongly associated with a wide range of inflammatory middle ear conditions and chronic sinus inflammation, suggesting an increased susceptibility to inflammation of the upper airways. Conclusion: This study shows a large overlap between cholesteatoma and non-cholesteatoma ear disease in terms of numbers and demographics, with sex being a key factor distinguishing between the two, suggesting that there are both common and distinct associated factors.