Impact of interfacility transfer of ruptured abdominal aortic aneurysm patients.

Objective: The interfacility transfer (IT) of patients with a ruptured abdominal aortic aneurysm (rAAA) occurs not infrequently to allow for a higher level of care. In the present study, we evaluated, using a contemporary administrative database, the effects of IT on mortality after rAAA repair.

Methods: The Healthcare Cost and Utilization Project Database for New York (2016) and New Jersey, Maryland, and Florida (2016-2017) was queried using the International Classification of Diseases, 10th edition, to identify patients who had undergone open or endovascular repair of AAAs. The hospitals were categorized into quartiles (Qs) per overall volume. The mortality rates for IT vs nontransferred (NT) rAAA patients stratified by treatment modality (open aneurysm repair of an rAAA [rOAR] vs endovascular aneurysm repair of an rAAA [rEVAR]) were compared. A Cox proportional hazard model was used to estimate the hazard ratios (HRs) for mortality.

Results: A total of 1476 patients had presented with a rAAA, of whom 673 (45.7%) were not treated. Of the remaining 803 patients, 226 (28.1%) were transferred, of whom 50 (22.1%) had died without repair after IT. The remaining 753 patients (IT, n = 176; NT, n = 576) had undergone rEVAR (n = 492) or rOAR (n = 261). The baseline characteristics were similar between the IT and NT patients, except for a greater proportion of black patients (P = .03), lower income families (P = .049), and rOAR (45.5% vs 31.4%; P = .001) for the IT patients. The overall mortality rates were similar between the NT (30.2%) and IT (27.3%) groups (P = .46). The subgroup analysis revealed that the operative mortality rates after rEVAR were similar between the NT and IT patients, without significant differences among the hospital quartiles. After rOAR, however, the operative mortality rates were lower for the IT patients, largely owing to improved outcomes in the Q4 hospitals (Q4 vs Q1-Q3, P = .001). Cox regression analysis demonstrated that age (HR, 1.03; 95% confidence interval, 1.00-1.06; P = .02) and treatment at a low-volume hospital (Q1-Q3; HR, 1.89; 95% confidence interval, 1.02-3.51; P = .04) were predictors of mortality. The total charges were similar (IT, $286,727; vs NT, $265,717; P = .38).

Conclusions: The results from the present study have shown that <30% of rAAA patients deemed a candidate for repair will be transferred. We found that IT did not affect the mortality rates after rEVAR, irrespective of the hospital volume. For rOAR candidates, however, regionalization of care with prompt transfer to a high-volume center could improve the survival benefits without increased healthcare costs.