Geography or pathology? Regional variation in atrial septal defect closure rates and techniques

Abstract

Since the introduction of percutaneous closure in the United States, rates of secundum atrial septal defect and patent foramen ovale closures have increased substantially. Whether or not closure rates are uniform or vary due to differences in regional practice patterns is unknown. We sought to investigate this by comparing regional rates of closure across Florida. We identified all atrial septal defect closures from 2001 to 2006 in the Florida State Inpatient Database. Using small area analysis, zip codes were assigned to Hospital Referral Regions based on where patients were most likely to go for closure. We obtained population-normalised rates of overall, percutaneous, and surgical closure. Of 1830 atrial septal defect and patent foramen ovale closures from 2001 to 2006, 751 were surgical and 1004 were percutaneous. The statewide closure rate was 1.91 per 100,000 people per year; regional rates varied 3.8-fold from 0.78 to 2.94 per 100,000 people per year. Percutaneous rates varied sevenfold from 0.25 to 1.75 per 100,000 people per year, while surgical rates varied 2.71-fold from 0.53 to 1.44 per 100,000 people per year. Despite a consistent prevalence of atrial septal defects, and patent foramens ovale, rates of repair vary across regions, suggesting that closure is driven by provider practice patterns rather than patient pathology. Efforts should be directed towards increasing consensus regarding the appropriate, evidence-based indications for closure so as to avoid the costs and potential negative sequelae of over- or undertreatment.