Invasive procedures and risk of brain abscess:a nationwide, population-based case-control study

Objectives: It is unknown whether invasive procedures are associated with brain abscess. Methods: Nationwide, population-based, matched case–control study of patients with culture verified brain abscess in Denmark from 1989 to 2016. Exposure was invasive procedures 0–6 months before study inclusion. Results: We identified 435 patients and 3909 controls. The level of comorbidity was higher among patients with brain abscess than among controls. A total of 48 cases (11%) had one or more invasive procedures 0–6 months before study inclusion (adjusted odds ratios (aOR) of 3.6 (95% confidence interval (CI): 2.5–5.1), a population attributable fractions of 8% (95% CI: 7–9)). In primary care, ear, nose and throat (ENT) procedures were associated with brain abscess (aOR of 4.0 (95% CI: 2.0–8.0)), but gastrointestinal endoscopies were not (aOR of 1.0 (95% CI: 0.3–3.2)). No bronchoscopies were performed in primary care. In the hospital-based setting, ENT procedures, bronchoscopies and gastrointestinal endoscopies were associated with an increased risk of brain abscess (aOR of 14.5 (95% CI: 4.8–43.8), 20.3 (95% CI: 3.8–110.1) and 3.4 (95% CI: 2.0–5.6), respectively). Conclusions: The association between invasive procedures and brain abscess was more pronounced in the hospital-based setting than in primary care.