PHILADELPHIA – A protocol that employed CT angiography instead of or along with nuclear bleeding scans before visceral arteriography in patients with lower gastrointestinal hemorrhage helped localize bleeds more accurately and reduced the need for subsequent imaging studies, compared with VA alone, a single-center study found.
Ryan McVay/Thinkstock
CT angiography before visceral arteriography helped localize bleeds more accurately and reduced the need for subsequent imaging studies than VA alone.
“CTA rather than a nuclear bleeding scan [NBS] prior to visceral arteriography is associated with fewer imaging studies but better localization of bleeding; if patients do receive more contrast, we did not observe any change in the renal function as a result,” Dr. Christina Jacovides of the University of Pennsylvania, Philadelphia, reported at the annual meeting of the American Association for the Surgery of Trauma.
The study evaluated 161 patients over an 8-year observation period: 78 from 2005 to 2009, before the protocol was implemented; and 83 under the protocol.
“We saw that this protocol effectively changed the pre-VA approach from bleeding scans to CTA,” Dr. Jacovides said.
Lower gastrointestinal hemorrhage carries “substantial” morbidity, in the words of the study authors, so developing tools that can locate the bleeding and target treatment is critical in reducing deaths and complications, Dr. Jacovides said. “Bleeding scans have been demonstrated to be highly sensitive but have had poor localization, whereas CTA has also been demonstrated to be highly sensitive and provides localization similar to that seen with VA,” she said.
In the study, obtaining visceral arteriography without any imaging beforehand identified bleeding in 62% of cases, versus 94% with pre-VA imaging consisting of either CTA, NBS, or both. CTA only before VA resulted in more imaging studies than VA alone (2.5 vs. 1.3), but was also more than twice as likely to find bleeding (92% vs. 43%). Bleed localization rates were similar between VA only and pre-VA CTA.
When compared to NBS, CTA only before VA resulted in fewer imaging studies (2.1 vs. 2.5 for NBS) and similar rates of finding bleeding (around 94% for both), but significantly higher rates of bleed localization on VA (45.7% for CTA vs. 26.4% for NBS), according to the study findings. Embolization rates among the different protocols also varied, from 23% for NBS to 40% for CTA only before VA. The study did not evaluate costs.
How the university’s protocol affected renal function drew the attention of discussant Dr. Leslie Kobayashi,of the University of California, San Diego, who noted that creatinine levels were actually higher in the VA-only group. “This would suggest to me that patient factors such as severity of hemorrhage, presence of shock, and location of comorbidities are most associated with increases in creatinine rather than the actual contrast bolus or contrast dose,” she said. However, the study did not include a multivariate analysis to determine if CT contrast was associated with a rise in creatinine, Dr. Jacovides said.
Dr. Hasan Alam of the University of Michigan Health System, Ann Arbor, acknowledged that CTA before VA may be viable in hemodynamically stable patients, but not so for those with more critical injuries. “The problem is sometimes logistics, because you have to go to two different places for the CTA and VA, so a time factor is involved,” he said. “So if you have a sick patient who is robustly or briskly bleeding, who’s hypotensive, for those 5%-10% of patients, that causes a delay.”
Dr. Jacovides noted that at least among the patients who received pre-VA imaging, the protocol at the University of Pennsylvania actually got those patients to treatment quicker. “We did find that among patients who underwent CTA only instead of NBS, the time from the first scan to first VA was significantly reduced; it was about 1,200 minutes on bleeding scans and about 530 minutes with CTA on average,” she said.
Dr. Jacovides reported having no relevant financial disclosures. She is a surgical intern at Thomas Jefferson University Hospital, Philadelphia, and completed this research as a medical student at the University of Pennsylvania, where the senior authors are in the division of traumatology, surgical critical care, and emergency surgery.