Primary biliary cirrhosis does not increase the risk of UTIs following diagnosis compared to other chronic liver diseases?

Authors

  • Fumi K. Varyani,

    1. Department of Gastroenterology, Lincoln County Hospital, Lincoln, UK
    Search for more papers by this author
  • Joe West,

    1. Division of Epidemiology and Public Health, University of Nottingham, Nottingham, UK
    2. Nottingham Digestive Diseases Centre Biomedical Research Unit, Nottingham, UK
    Search for more papers by this author
  • Timothy R. Card

    Corresponding author
    1. Division of Epidemiology and Public Health, University of Nottingham, Nottingham, UK
    2. Nottingham Digestive Diseases Centre Biomedical Research Unit, Nottingham, UK
    • Department of Gastroenterology, Lincoln County Hospital, Lincoln, UK
    Search for more papers by this author

Correspondence

Timothy Card, Division of Epidemiology and Public Health, Clinical Sciences Building Phase 2, Nottingham City Hospital, Hucknall Rd, NG5 1PB, Nottingham, UK

Tel: (44) 115 8231346

Fax: (44) 115 8231337

e-mail: Tim.Card@nottingham.ac.uk

Abstract

Background

Urinary Tract Infections (UTIs) occur more frequently in patients with Primary Biliary Cirrhosis (PBC). Previous studies have compared UTI occurrence in PBC and general population controls, however, it remains unclear if UTI is a feature of all chronic liver diseases (CLD)s, or is specific to PBC, or if this is a cause or consequence of PBC.

Aims

We aimed to determine if UTIs are more common after a diagnosis of PBC compared to general population and CLD controls.

Methods

A cohort study was conducted using the General Practice Research Database. We selected all cases of PBC plus 10 age- and sex-matched general population controls, and an unmatched group with other CLDs. We formed a Cox–proportional hazard model of time to first UTI following diagnosis.

Results

Two hundred and forty-eight (24.6%) of PBC cases had a UTI event compared with 2127 (21.1%) of matched and 2131 (11.7%) of the unmatched CLD controls. Comparing PBC with matched controls showed an approximately 30% increased risk of UTI [hazard ratio (HR) 1.33 confidence interval (CI) 1.17–1.52]. Adjusting for diabetes, smoking and previous UTI reduced this (HR 1.25 CI 1.09–1.42). The Hazard Ratio comparing PBC with unmatched CLD controls was 2.00 (CI 1.76–2.28), but this became non-significant when adjusting for age, sex, diabetes, smoking and previous UTI 0.98 (0.86–1.12).

Conclusions

There is increased risk of UTI in PBC patients compared to general population controls, but not compared to CLD controls suggesting that this association is not specific to PBC after diagnosis.

Ancillary