THE ROLE OF EMERGENT AVAILABILITY OF INTERVENTIONAL RADIOLOGY AND OUTCOMES OF GASTROINTESTINAL BLEEDING AMONG THE ELDERLY.
Kevin B. Ricci*1, Angela M. Ingraham2, Amy P. Rushing1, Vijaya T. Daniel3, Adrian Diaz1, Anghela Z. Paredes1, Victor Heh1, M. Didem Ayturk4, Holly E. Baselice1, Wendelyn M. Oslock1, Scott Strassels1, Heena P. Santry1
1Surgery, The Ohio State University Wexner Medical Center, Columbus, OH; 2Surgery, University of Wisconsin, Madison, WI; 3Surgery, University of Massachusetts, Worcester, MA; 4Quantitative Health Sciences, University of Massachusetts, Worcester, MA
Introduction: With the advent of Interventional Radiology (IR), surgery has increasingly become an option of last resort option for patients with massive gastrointestinal bleeding (GIB). We examined the role of IR availability within 1 hour on outcomes for GIB among Medicare Beneficiaries.
Methods: In 2015, we surveyed 2,811 US hospitals regarding EGS practices including availability of IR services within 1 hour (always, often, sometimes, rarely, never). 1,690 US hospitals completed surveys (60% response). Survey responses were linked to 2015 Medicare inpatient claims data using American Hospital Association unique identifiers. We identified all Medicare beneficiaries age ≥65 hospitalized emergently with a primary diagnosis of GIB (both upper and lower). Patients transferred in from another acute care hospital were excluded. We determined rates of transfer, unadjusted odds, and adjusted (demographic characteristics, co-morbidities, prior interventions, hospital clustering) odds of transfer to another acute care hospital, angioembolization, operation and mortality based upon IR availability at the treating hospital. Patients who were transferred for further care were excluded from the IR, OR and mortality analyses.
Results: We identified 54,406 individual GIB patients at 1461 unique hospitals. The majority of patients were white (81.0%), female (54.0%), 75-84 years old (37.3%), with Elixhauser Index = 3 (72.1%). Bleeding locations were 28.4% UGI, 0.4% small bowel, and 71.2% LGI. Overall transfer rates were 1.7% and highest proportion of transfers was from hospitals with no IR availability. Table 1 shows transfer, angioembolization, operation, and death rates by IR availability. Overall, angioembolization, operation, and mortality rates were 0.6%, 2.0% and 2.4%, respectively. Angioembolization rates were highest among hospitals with greater IR availability. Operative rates were highest where IR was "˜rarely' available. Table 2 shows odds of transfer, angioembolization, operation, and death by IR availability. Odds of transfer increased consistently as IR availability decreased. Odds of operation were highest when IR was "˜rarely' available. IR availability did not predict mortality.
Discussion: As IR availability decreases, patients are more likely to be transferred, presumably to a more resourced hospital. Patients who are not transferred appear to be at highest risk for a possibly preventable operation if the treating hospital rarely mobilizes IR within an hour. Presenting to a hospital lacking IR entirely may confer a protective effect by serving as a clear indication for transfer. Providers at hospitals with some IR presence, however, may over-estimate access angioembolization and retain patients for whom surgery, and any adverse post-operative outcomes, might have been avoided at a hospital with better IR availability.
Table 1. Hospital-level rates of transfer, angioembolization, operation and mortality based on IR availability for patients (N = 54,406) admitted with GIB.
| IR availability | Transfer Rate+ | Angioembolization Rate*+ | Operation Rate*+ | Mortality Rate* |
| Always | 0.7 | 0.9 | 2.0 | 2.6 |
| Often | 1.1 | 0.9 | 1.9 | 2.7 |
| Sometimes | 1.7 | 0.5 | 1.8 | 2.4 |
| Rarely | 2.5 | 0.4 | 2.7 | 2.7 |
| Never | 4.9 | -- | 1.8 | 2.5 |
*Excludes patients (N = 936) who were transferred to another acute care hospital
+Chi2 test association with P-value <0.05
Table 2. Patient-level unadjusted and adjusted odds of transfer, IR procedure, operation and mortality based on IR availability
| IR Availability | Transfer | Angioembolization* | Operation* | Mortality* | ||||
| OR | aOR1 | OR | aOR1 | OR | aOR2 | OR | aOR3 | |
| Always (ref) | -- | -- | -- | -- | -- | -- | -- | -- |
| Often | 1.28 (1.04-1.59) | 1.30 (1.05-1.60) | 1.04(0.82-1.33) | 1.04 (0.81-1.33) | 0.93 (0.79-1.11) | 0.93 (0.79-1.11) | 1.05 (0.91-1.21) | 1.05 (0.91-1.21) |
| Sometimes | 2.14 (1.75-2.63) | 2.18 (1.78-2.67) | 0.54 (0.38-0.77) | 0.54 (0.38-0.77) | 0.88 (0.73-1.07) | 0.91 (0.75-1.10) | 0.94 (0.80-1.12) | 0.95 (0.81-1.13) |
| Rarely | 3.00 (2.39-3.77) | 3.03 (2.41-3.82) | 0.47 (0.28-0.80) | 0.47 (0.28-080) | 1.33 (1.07-1.67) | 1.37 (1.10-1.71) | 1.07 (0.87-1.32) | 1.09 (0.88 -1.35) |
| Never | 6.07 (5.17-7.13) | 6.15 (5.24-7.23) | -- | -- | 0.90 (0.74-1.09) | 0.99 (0.82-1.20) | 0.98 (0.83-1.15) | 1.01 (0.86-1.20) |
*Excludes patients (N = 936) who were transferred to another acute care hospital
1. Adjusted for patient characteristics
2. Adjusted for patient characteristics + IR procedure
3. Adjusted for patient characteristics + IR procedure + Operation
Back to 2019 Abstracts
