SAN FRANCISCO – One-third of perioperative deaths and complications after elective endovascular repair of abdominal aortic aneurysms occur post discharge, results from a large analysis showed.
"Improved predischarge surveillance and close postdischarge follow-up of identified high-risk patients may further improve 30-day outcomes after EVAR," Dr. Prateek K. Gupta said at the Society for Vascular Surgery annual meeting.
Dr. Prateek Gupta
Outcome improvement in the field of aortic surgery, specifically endovascular repair of abdominal aortic aneurysms, "has received much attention," said Dr. Gupta of the department of surgery at the University of Wisconsin Hospital and Clinics, Madison. "With EVAR, the index hospital stay after aortic surgery has decreased significantly, leaving a need for better understanding of postdischarge outcomes, which is necessary to improve quality and reduce readmission rates with implementation of targeted outpatient interventions."
In an effort to examine postdischarge 30-day outcomes after elective EVAR, Dr. Gupta and his associates identified 11,229 patients from the American College of Surgeons National Surgical Quality Improvement Program (NSQIP) database who underwent an elective EVAR for AAA between 2005 and 2010. The primary outcome of interest was postdischarge mortality, while the secondary outcome was postdischarge overall morbidity. The researchers performed univariate and multiple logistic regression analysis to assess factors associated with the primary and secondary study outcomes.
Of the 11,229 patient 83% were male and their mean age was 75 years. Dr. Gupta reported that 117 patients died within 30 days of EVAR, for a rate of 1%. Of these deaths, 31% occurred after hospital discharge, and the median time to death was 9 days. At the same time, 1,204 patients experienced complications within 30 days of EVAR, for a rate of 11%. Of these, 500 (40%) occurred post discharge, and the median time for a complication to occur was 3 days.
Only 20% of patients (7/36) who died post discharge experienced an in-hospital complication. Compared with patients who did not develop a postdischarge complication, those who had more than a sixfold likelihood of reoperation (20.4% vs. 3.1%, respectively; P less than .0001) and death (3.0% vs. 0.2%; P less than .0001) within 30 days of surgery.
Multivariable analysis revealed the following factors that were independently and significantly associated with postdischarge mortality: preoperative heart failure (adjusted odds ratio, 4.7), admission from a skilled nursing facility (AOR, 2.2), increase in age per year (AOR, 1.09), postdischarge renal failure requiring dialysis (AOR, 72.5), postdischarge cardiac arrest/MI (AOR, 46.6), and postdischarge pneumonia (AOR, 26.5).
Dr. Gupta reported that the 30-day postdischarge rate among patients admitted from a nursing facility or acute care was 2.5%. "In contrast to patients who survived after EVAR, patients who died post discharge were more likely to have been admitted from a nursing facility or acute care (13.9% vs. 1.8%; P less than .0001)," he said.
The 30-day post-discharge mortality was highest among patients who had postdischarge renal failure (27%),postdischarge MI (19%), and postdischarge pneumonia (15%).
The researchers also found that patients with a history of peripheral artery disease (PAD) had a significantly higher post-discharge complication rate after EVAR (7.1% vs. 4.3%; P = .001). This also correlated with a higher wound infection rate (3.2% vs. 1.7%; P = .01). A previous cardiac surgery also predisposed patients toward a higher overall postdischarge complication rate (5.3% vs. 4.2%; P = .007).
"Usually, patients undergoing EVAR are followed up at 2 weeks for wound evaluation, or at 1 month with a CT scan," Dr. Gupta said. "In the present study, the median occurrence for most of the postdischarge complications was within the first 10 days after surgery. The interquartile range was 11-22 days for the diagnosis of a wound infection after EVAR. These data suggest that earlier follow-up of high-risk patients may help identify and possibly prevent some of these complications and subsequently decrease readmissions. A standardized protocol for triage and surveillance of high-risk patients post EVAR is needed."
Limitations of the study include that fact that causality could not be determined because it was a retrospective analysis. "In addition, the timing of the operation is not specified in NSQIP," so it could either be a predischarge event or it could have occurred on readmission, Dr. Gupta said. "Data on readmission is not available from the 2005-2010 data sets."
Dr. Gupta said that he had no relevant financial disclosures to make.