Mixed results with angiography for splenic injuries

Major finding: The likelihood of delayed splenectomy was 33% higher at centers without angiography and 49% higher at low-angiography centers, compared with high-angiography centers, but the differences were not statistically significant.

Data source: Retrospective analysis of data from the National Trauma Data Bank on 6,870 patients treated for blunt splenic injury at 267 hospitals.

Disclosures: Dr. Zarzaur reported having no financial disclosures.

View on the News

Database can’t answer the question

Almost 20 years after the initial description and 30 years after we began using splenic angiography in the management of blunt splenic injury, why is it that we simply can’t settle this question? When is splenic angiography and/or catheter therapy useful in high-grade injuries?

In this study, the authors have reviewed the National Trauma Data Bank (NTDB) and have demonstrated that, at hospitals that use angiography more frequently than other hospitals, the rate of delayed splenectomy in high-grade splenic injury (defined as grades 3-5) is not different. They suggest that angiography should be selective, with particular attention to screening for vascular abnormalities.

Dr. Thomas M. Scalea

The authors collected information on both angiography and angiography with embolization. However, in the manuscript and the talk, they only refer to high-angiography centers. This seems to me to be a fundamental problem. This is particularly true because recent data from Jacksonville suggest that embolization of truly high-grade injury such as grade 4 injuries (even in the absence of blush) improves the salvage rate of nonoperative management.

In addition, the authors selected 20% as their cutoff for high- and low-angiography centers because 20% represented the 90th percentile of centers with regard to angiography use. I think that’s French for "it made the data analyzable." However, 20% is relatively low. If the authors wish to look at nonselective angiography they should look at us. We’re maximalists. We do angiography on 100% of patients with grades 3, 4 and 5 injuries. That’s nonselective use of angiography.

The intelligent use of this technique requires interpretation of CT and then the details of the patient presentation. For instance, a grade 4 splenic injury with or without blush but no hemoperitoneum and a totally stable patient is, in my mind, amenable to catheter therapy. Another patient with a grade 4 injury and reactive extravasation outside of the spleen and a huge peritoneum is probably best served by operative exploration. Both are grade 4 injuries, but the patients are fundamentally different.

The authors’ conclusions suggest that there is a relationship between splenic vascular injury identified on CT and success or failure of nonoperative management. This is clearly the authors’ prejudice, as they have published these findings a number of times. We’ve known for years, thanks to work from the authors’ institution, that expectant management of a patient with blush on CT fails 70% of the time and a significant number of the blushes are seen on day 3 but not day 1. Since the authors have absolutely no information on the presence or absence of blush in this data set, I fail to see how that can be one of their conclusions.

How, then, can we make sense of this? I believe the answer is in the manuscript’s last paragraph, which begins, "Another limitation of this study stems from the limitations of the NTDB." There’s little doubt that the NTDB can record an accurate snapshot of practice in the United States, but in my mind, it lacks the specificity to really answer the question, when is splenic angiography useful in high-grade injuries?

The authors have no information on presence of absence of blush, hemodynamic status other than at admission, blood transfusion rate, or technique of embolization, and they recognize that some of the data may not be accurate. I just don’t believe that the NTDB can actually answer this question.

In the end, rules are rarely helpful in the care of patients. Intelligent application of innovative techniques cannot solely be governed by rules. Perhaps the take-home message here is that the use of angiography and embolization to treat higher-grade splenic injuries is perhaps not something that everybody should be using. It may be that this technique is best preserved for high-volume centers with a real interest and a real expertise in this subject.

Dr. Thomas M. Scalea is a professor of surgery at the University of Maryland, Baltimore. These are excerpts of his remarks as a discussant of the study at the meeting. He reported having no financial disclosures.



SAN FRANCISCO – Trauma centers that nonselectively performed angiography on patients with high-grade blunt splenic injury did not significantly reduce the likelihood of delayed splenectomy in a retrospective analysis of data on 6,870 patients treated at 267 hospitals.

On an individual patient level, however, use of angiography was associated with a reduced risk of delayed splenectomy (more than 6 hours after admission) after researchers controlled for the influence of multiple other factors, Dr. Ben L. Zarzaur and his associates reported at the annual meeting of the American Association for the Surgery of Trauma.

Dr. Ben L. Zarzaur

These somewhat conflicting findings suggest that "nonselective protocol-driven use of angiography at the hospital in the setting of high-grade blunt splenic injury does not benefit in terms of splenic salvage. Angiography use should be tailored to the individual patient," said Dr. Zarzaur of the University of Tennessee, Memphis.

"Attention should be paid to overall injury severity and splenic injury severity" because more severe injuries were associated with delayed splenectomy in the study, he said, adding, "Particular attention should be considered for screening for splenic vascular abnormalities."

The investigators used data from the National Trauma Data Bank (NTDB) on adults treated for high-grade blunt splenic injury at Level I or II trauma centers that admitted at least 10 such patients in 2007-2010, with high-grade injury defined as Abbreviated Injury Scale grade 3 or higher. They stratified hospital angiography use as none, low (in less than 20% of patients with high-grade blunt splenic injury), or high (in 20% or more of these patients).

Approximately 30% of patients at high-angiography centers underwent urgent splenectomy, compared with 33%-36% at hospitals with no or low-angiography use, a difference that was statistically significant. While the likelihood of a delayed splenectomy was 33% higher at low-angiography hospitals and 49% higher at hospitals without angiography, compared with high-angiography hospitals, these differences were not significant, Dr. Zarzaur reported.

The investigators used the classification of hospitals – no-, low-, or high-angiography use – to represent the three schools of thought that have developed over the past few decades regarding angiography for patients with blunt splenic injury who do not undergo immediate urgent splenectomy. The minimalist school of thought recommends using observation, not angiography for blunt splenic injury. The maximalist school of thought favors protocol-driven use of angiography for patients with certain grades of spleen injury. In between, physicians who favor a selective strategy use CT or clinical criteria or both to try and identify patients at high risk for delayed splenectomy, and reserve the risks of angiography for those patients, he said.

They chose a cutoff of 20% angiography use in patients with high-grade blunt splenic injury to discriminate between low- and high-angiography use because that represented the 90th percentile for all trauma centers in the study.

Nine percent of patients were treated at hospitals that did not use angiography for blunt splenic injury, 66% at low-angiography hospitals, and 25% at high-angiography hospitals.

Patients with grade 5 blunt splenic injury were more than twice as likely to need delayed splenectomy, compared with patients with grade 3 or 4 injury. Higher overall Injury Severity Scores (10 or higher) also doubled the risk for delayed splenectomy.

Patients with grades 4 or 5 blunt splenic injury were significantly more likely to undergo angiography at high-angiography centers than at low-angiography centers. High-angiography centers were more likely than were low-angiography centers to remove spleens with grade 5 injury after angiography, though this difference did not reach statistical significance.

Continuing controversy around the use of angiography for blunt splenic injury is illustrated by a 2011 survey of members of the American Association for the Surgery of Trauma. Members favored observation, not angiography, for grades 1 and 2 spleen injuries but showed no consensus on higher-grade injuries (J. Trauma 2011;70:1026-31).

A recent study of 1,275 patients treated for blunt splenic injury at four trauma centers that showed a significantly better chance of saving the spleen at hospitals with higher use of splenic artery embolization, especially in patients with higher-grade splenic injury (J. Trauma Acute Care Surg. 2013;75:69-74).

The current study excluded patients who died on arrival at the hospital, patients who were admitted more than 24 hours after injury, and patients who underwent splenectomy within 6 hours of admission (early splenectomy).

Dr. Zarzaur reported having no financial disclosures.

On Twitter @sherryboschert

Next Article:

   Comments ()