P0918 - Evaluating Abdominal CT Angiography Use in Suspected GI Bleeding: A Retrospective Study to Inform Future Practice at a Tertiary Care Hospital in NYC

Sharmitha Yerneni, MD, Stephanie Socias, MD, Eugene Hrabarchuk, MD, John B. Doyle, MD, Rachel Eklund, MD, Nelson Bean, MD, MPH, Ava Anklesaria, MD
Columbia University Irving Medical Center, New York, NY

Introduction: The ACG recommends abdominal CTA as the initial diagnostic test for suspected LGIB in patients with hemodynamic instability and ongoing blood loss. We postulated that most CTAs ordered for GI bleeding at our institution do not meet these criteria. Given the associated risks of radiation and costs to the healthcare system, evaluating the appropriateness of CTA ordering is critical to improving test utilization.

Methods: This retrospective study analyzed all abdominal CTAs performed over a 1-year period at a single tertiary care center for the indication of GI bleeding. Patients were identified by manually reviewing the stated indication of each scan and extracting only those ordered for suspected GIB. Relevant records were then reviewed for clinical context and subsequent interventions.

To determine if scans met ACG ordering guidelines of hemodynamic instability and ongoing blood loss, we utilized the shock index (SI), a validated metric derived from heart rate/systolic BP, and chart notes. A SI > 0.9 and documentation supporting ongoing blood loss deemed a scan consistent with ACG guidelines.

Results: 158 CTAs were ordered out of concern for GI bleeding: 78 (50%) for LGIBs (hematochezia or BRBPR), 59 (37%) for UGIBs (melena, coffee grounds, hematemesis), and 21 (13%) for other/non-specific GI bleeding (iron deficiency). Of these 158 scans, 28 (17.7%) demonstrated active extravasation and 14 (8.8%) subsequently underwent IR embolization. LGIBs accounted for 17 of the 28 positive scans and 6 of the 14 embolizations.

The degree of hemodynamic instability correlated with the type of bleeding in that UGIBs had a higher average SI than LGIBs and those who underwent IR embolization had the highest of all. Most strikingly, only 22 (28%) of the 78 scans ordered for LGIB had a SI >0.9 and thus fell within ACG guidelines.

Discussion: The utility of abdominal CTAs in identifying actionable lower GI bleeds is predicated on its use in the appropriate clinical context: ongoing, hemodynamically significant overt bleeding. This study showed that only a fraction (28%) of CTAs ordered for LGIBs were ordered accordingly.

These results support what we anecdotally suspected: CTAs for LGIB are often ordered outside of ACG guidelines and represent a targetable area for improvement. Reducing unnecessary scans would not only minimize patient exposure to radiation and contrast but yield significant cost savings for the healthcare system while preserving radiologists’ time for needed studies.

Disclosures:
Sharmitha Yerneni indicated no relevant financial relationships.
Stephanie Socias indicated no relevant financial relationships.
Eugene Hrabarchuk indicated no relevant financial relationships.
John Doyle indicated no relevant financial relationships.
Rachel Eklund indicated no relevant financial relationships.
Nelson Bean indicated no relevant financial relationships.
Ava Anklesaria indicated no relevant financial relationships.

Sharmitha Yerneni, MD, Stephanie Socias, MD, Eugene Hrabarchuk, MD, John B. Doyle, MD, Rachel Eklund, MD, Nelson Bean, MD, MPH, Ava Anklesaria, MD. P0918 - Evaluating Abdominal CT Angiography Use in Suspected GI Bleeding: A Retrospective Study to Inform Future Practice at a Tertiary Care Hospital in NYC, ACG 2025 Annual Scientific Meeting Abstracts. Phoenix, AZ: American College of Gastroenterology.