2584. Triple-Phase CTA For Melena Without Bright Red Blood Per Rectum: Diagnosis and Prediction of Clinical Outcomes
Authors * Denotes Presenting Author
  1. Hadiseh Kavandi *; Beth Israel Deaconess Medical Center
  2. Francesca Rigiroli; Beth Israel Deaconess Medical Center
  3. Youssef Masmoudi; Beth Israel Deaconess Medical Center
  4. Timothy Sotman; Beth Israel Deaconess Medical Center
  5. Alexander Brook; Beth Israel Deaconess Medical Center
  6. Olga Brook; Beth Israel Deaconess Medical Center
Melena (black stool) usually associated with upper GI bleeding, though up to 30% ascribed to lower GI slow bleeding. The goal of this study was to determine yield of triple phase CTA in diagnosis and predicting clinical outcomes in patients with melena without bright red blood per rectum.

Materials and Methods:
In this retrospective IRB-approved, HIPAA-compliant study, consecutive patients with melena without bright red blood per rectum between 1/1/2015 and 3/30/2021 that underwent triple-phase mesenteric CTA in a large tertiary center were included. Outcomes assessed included RBC transfusion, endoscopic treatment, endovascular interventions, surgical management, readmissions, and death.

111 patients with melena (age 68.7 years, 48/111 (43.24%) females) were included. There was active extravasation on CTA in 15/111 (14%) with 14/15 (93%) detected on arterial phase and venous phases and in 1/15 (7%) on venous phase only. Extravasation was seen from duodenum in 7/15 (47%) patients, stomach in 1/15 (7%), small bowel in 2/15 (14%) and colon in 3/15 (21%) and rectum 2/15 (14%). 31/111 (28%) patients underwent 34 IR procedures, a median of 0.5 days after CTA (IQR 0.1 – 2.8 days); with arterial access in 23/111 (20.7%) and variceal access in 8/111 (7.2%). There were 7/34 (20%) diagnostic only procedures, 10/34 (30%) prophylactic embolizations (no extravasation on angio) and 17/34 (50%) therapeutic embolization (extravasation on angio). Only 10/34 (29.4%) of patients that had IR procedures had active extravasation. Among them, 6/10 (60%) were therapeutic embolizations. 3/111 (2.7%) patients underwent surgery. 67/111 (60%) of patients underwent subsequent endoscopy. 9/15 (60%) patients with active extravasation on CT underwent IR intervention vs. 14/96 (14.5%) without (p<0.0001). A strong association was found between active extravasation on CT and RBC transfusion (p=0.0005), but not with surgery. Active extravasation on CT did not predict readmission and death. No significant association was noted between BMI, hemodynamic status, hematocrit level, comorbidities, anticoagulant or anti-platelet medications with the presence of extravasation on CT. Patients with active extravasation were older 75.8 years vs. 67.5 years without extravasation, p=0.04; had slightly higher INR of 1.6 vs. 1.4 in patients without extravasation, p=0.04; and had a lower platelet count of 127 vs. 194 in patients without extravasation, p=0.01.

Active extravasation is detected in 14% of patients with melena without BRBPR on triple-phase mesenteric CTA, vast majority seen both on arterial and venous phases mostly originating from upper GI source. Higher age, higher INR, and lower platelet count were associated with active extravasation on CT. The presence of active extravasation was strongly associated with IR intervention and the need for RBC transfusion, but not surgery, readmission, and death.