Limitations of splenic angioembolization in treating blunt splenic injury

Robert Cooney, James Ku, Robert Cherry, George O. Maish, Daniel Carney, Leslie B. Scorza, J. Stanley Smith

Research output: Contribution to journalArticlepeer-review

54 Scopus citations


Background: When angiography is performed in all hemodynamically normal patients with splenic injury, only 30% require embolization. This study examines the use of selective splenic angioembolization (SAE) as part of a management algorithm for adult splenic injury. Methods: Criteria for selective SAE were added to our adult splenic injury protocol in July 1999. SAE was performed in hemodynamically stable patients if computed tomographic (CT) scan revealed injury to the hilum or vascular blush and when nonoperative patients had a gradual decrease in hematocrit. Patients were grouped by management strategy: nonoperative; operative; or SAE. Demographics, injury severity, and outcomes of the different groups were compared. Medical records, CT scans, and registry data were reviewed for all SAE cases, deaths, and treatment failures. Data are means ± SE. p < 0.05 versus nonoperative management by analysis of variance. Results: From July 1999 to August 2003, 194 adults were treated for splenic injury. Nine patients underwent SAE, six for CT findings (1° vascular blush) and three for decreasing hematocrit. Three patients failed SAE (33%), one for bleeding and two for delayed splenic infarction. Eleven patients failed nonoperative therapy (8%); splenorrhaphy was performed hi three and splenectomy in eight. Operative patients were more seriously injured and had higher Injury Severity Scores and mortality; splenectomy (39 of 48) was more commonly performed than splenorrhaphy (9 of 48) in this group. Conclusion: Use of a splenic injury algorithm is associated with a high success rate for nonoperative management of splenic trauma. Using selective criteria, only 5% of patients were treated with SAE. SAE salvaged six patients with high-grade splenic injury or decreasing hematocrit but had a 33% failure rate. Failure of nonoperative management was most commonly caused by errors in judgment, primarily recognition of "high-risk" injury patterns on CT scan or attempting nonoperative management in anticoagulated or coagulopathic patients.

Original languageEnglish (US)
Pages (from-to)926-932
Number of pages7
JournalJournal of Trauma - Injury, Infection and Critical Care
Issue number4
StatePublished - Oct 2005

All Science Journal Classification (ASJC) codes

  • Surgery
  • Critical Care and Intensive Care Medicine


Dive into the research topics of 'Limitations of splenic angioembolization in treating blunt splenic injury'. Together they form a unique fingerprint.

Cite this