PPI-stroke link traces in part to confounders

  • Gastroenterology

  • Jenny Blair, MD
  • Clinical Essentials
L'accesso ai contenuti di questo sito è riservato agli operatori del settore sanitario italiano L'accesso ai contenuti di questo sito è riservato agli operatori del settore sanitario italiano

Takeaway

  • Residual confounders might explain much, though not all, of the link between proton pump inhibitor (PPI) use and ischemic stroke.

Why this matters

  • 2 recent studies reported links between PPI use and incident stroke.
  • However, those studies’ designs may have overestimated the size of this link.

Key results

  • 6.5% of women and 16.1% of men reported regular PPI use; users had higher rates of multiple chronic diseases.
  • 2599 incident strokes occurred in study period.
  • Age-adjusted analysis for PPI users vs nonusers:
    • Total and hemorrhagic stroke rates similar (HRs, 1.11 [95% CI, 1.00-1.24] and 1.00 [95% CI, 0.72-1.38], respectively);
    • Ischemic stroke rate higher among users (HR, 1.25; 95% CI, 1.08-1.46).
  • Ischemic stroke analysis adjusted for multiple confounders: HR, 1.18; 95% CI, 1.02-1.37.
  • Ischemic stroke analysis adjusted further for several PPI indications, including history of peptic ulcer disease or gastroesophageal reflux disease (GERD): HR, 1.08; 95% CI, 0.91-1.27.

Study design

  • Analysis of PPI use during Nurses’ Health Study since 2000 (n=68,514 women) and Health Professionals Follow-up Study since 2004 (n=28,989 men) (both prospective cohort studies).
  • Primary outcome: first incident stroke.
  • Funding: NIH.

Limitations

  • Researchers included probable and chart-confirmed strokes.
  • Dosage information not available.
  • Analyses did not adjust for metabolic syndrome.