FORT LAUDERDALE, FLA. — A steep rise in secundum atrial septal defect repair in U.S. hospitals is being driven by a similarly dramatic shift toward percutaneous closure, mostly in older patients.
“The perceived decreased morbidity of percutaneous closure, coupled with a lower threshold for repair, may be driving increased utilization,” according to Dr. Tara Karamlou of the Oregon Health and Science University, Portland.
“However, what is surprising is that the technology was adopted so rapidly nationwide and in an older population in which one could question the appropriateness of device closures,” she said. “My suspicion is that these patients are coming in for coronary disease, getting a screening echocardiogram or cardiac catheterization that incidentally finds a small atrial septal defect, and that prompts device closure.”
Dr. Karamlou and her colleagues investigated nationwide trends in percutaneous closure (PC) use compared to surgical closure (SC) and their outcomes over 17 years. Using the Nationwide Inpatient Sample and ICD-9 procedure and diagnosis codes, they identified all atrial septal defect (ASD) closures performed from 1988 to 2005.
Over this period, the authors identified more than 15,000 secundum ASD closures, yielding a national estimate of 80,000 closures. Of these, 5,500 (national estimate 27,500) were percutaneous, and more than 10,000 (national estimate 54,000) were surgical. The remaining closures were of unspecified or undetermined type, Dr. Karamlou said at the annual meeting of the Society of Thoracic Surgeons.
Per capita, ASD closures increased dramatically from 1.08 per 10,000 population in 1988 to 2.59 per 10,000 in 2005. Surgical closures increased over the period from 0.86 to 1.07 per 100,000, whereas PC increased from 0.04 to 1.43 per 100,000, Dr. Karamlou reported.
Age was an important and statistically significant determinant of repair type, with younger patients preferentially treated with surgical repair. “Importantly, there was a shift in the demographics of the patient population over time, favoring ASD closure in older patients,” the authors said. This shift was noted in both closure types, but was more pronounced for PC, they said, adding that for PC, the mean age had increased from 12 years in 1988 to 42 years in 2005. The mean age for surgical closure increased from 26.5 to 27 years.
“This logarithmic increase in ASD closures was very surprising and unexpected,” Dr. Karamlou said in an interview. “Percutaneous closure was introduced in the early 1980s, and there was a significant lag before it took off. Then, around 2001, we started seeing a yearly doubling that gained momentum in 2004 and 2005.”
“The prolific increase in ASD closure rates therefore is due almost exclusively to treatment of older patients, aged 40–49,” Dr. Karamlou said.
The investigators identified 23 (national estimate 119) in-hospital deaths among PC patients and 84 (national estimate 472) deaths among patients undergoing surgical closure. Over the study period, crude mortality rates remained virtually the same for the two closure approaches.
The fact that mean hospital length of stay was twice as long for surgical closure (6 days vs. 3 days) must be viewed in light of the fact that PC patients were older and had fewer noncardiac anomalies than did those undergoing surgical closure, Dr. Karamlou cautioned.
That the data show a decrease in the ages of patients undergoing device closures in large centers suggests that large centers “are using device closures more appropriately,” Dr. Karamlou said in an interview. “But in general, PC closure devices are being used in an older patient population that one could argue really don't need to have their defects closed, and in the past, when surgery was the only option, these patients certainly would not have undergone closure.”
Dr. Karamlou concluded that, in the absence of meaningful benchmarks, prospective studies comparing outcomes, criteria, and cost for SC vs. PC are needed to determine whether increased atrial septal defect closure rates are justified.
“Assuming that the overall incidence of ASD is constant over time, the dramatic increase in ASD closure per capita suggests that increased utilization underlies this phenomenon, rather than an appropriate response to increased disease prevalence within the national population.”