Blunt Splenic Injury: A Brief Update on the Role of Angioembolization

Several recent clinical experiences with trauma patients with blunt splenic injury (BSI) prompted an interest in reviewing the guidelines for nonoperative management (NOM) of BSI. The 2012 Practice Management Guideline from the Eastern Association for the Surgery of Trauma (EAST) provides evidence based recommendations for the selective nonoperative management of blunt splenic injury (https://www.east.org/resources/treatment-guidelines/blunt-splenic-injury,-selective-nonoperative-management-of, [1]).

To summarize, NOM is the appropriate treatment in patients with BSI who are hemodynamically stable, regardless of grade of injury, patient age, or the presence of associated injuries. However, it should only be pursued in a clinical setting with adequate patient monitoring, readily accessible serial evaluations (physical examinations, laboratory tests, CT imaging), and an operating room available for emergent laparotomy.

The diagnostic modality of choice for BSI is contrast enhanced CT.  For reference, the American Association for the Surgery of Trauma (AAST) spleen injury scale:

Spleen injury scale (1994 revision)
Grade*	Injury type	Description of injury	ICD-9	AIS-90
I	Hematoma	Subcapsular, <10% surface area	865-01	2
			865.11
	Laceration	Capsular tear, <1cm	865.02	2
		parenchymal depth	865.12	2
II	Hematoma	Subcapsular, 10%-50% surface area	865.01
		intraparenchymal, <5 cm in diameter	865.11
				2
	Laceration	Capsular tear, 1-3cm parenchymal depth that does not	865.02
		involve a trabecular vessel	865.12
				3
III	Hematoma	Subcapsular, >50% surface area or expanding; ruptured
		subcapsular or parecymal hematoma; intraparenchymal
		hematoma > 5 cm or expanding
	Laceration	>3 cm parenchymal depth or involving trabecular vessels	865.03	3
			865.13
IV	Laceration	Laceration involving segmental or hilar vessels producing
		major devascularization (>25% of spleen)		4
V	Laceration	Completely shattered spleen	865.04	5
	Vascular	Hilar vascular injury with devascularizes spleen	865.14	5
*Advance one grade for multiple injuries up to grade III.

(http://www.aast.org/library/traumatools/injuryscoringscales.aspx#spleen)

Angiography and embolization are important adjuncts to NOM for BSI. The EAST guidelines state, “Angiography should be considered for patients with American Association for the Surgery of Trauma (AAST) grade of greater than III injuries, presence of a contrast blush, moderate hemoperitoneum, or evidence of ongoing splenic bleeding." [1]

 In light of a recent case in which NOM failed and our patient required a splenectomy, I undertook a brief review of the literature published since the 2012 EAST guidelines to determine if more specific criteria or protocols for angiography in BSI are currently utilized at trauma centers, and if they are associated with improved success of NOM. The following is a brief summary of three pertinent articles.

In a single institution study from Wake Forest University, a protocol was developed requiring referral of all BSI grade III to V, without indication for immediate operation, for angiography and embolization [2]. This prospective study was part of a performance improvement project, and the outcomes were compared with historic controls in which referral for angiography was based on presence of contrast blush on CT and surgeon preference.  In the protocol period, all hemodynamically stable patients with grade III or greater BSI underwent angiography regardless of presence or absence of blush on CT (those with lower grade injuries and contrast blush on CT also underwent angiography but were not included in the study population). The nature of the embolization was at the discretion of the angiographer and protocol compliance was defined as a patient undergoing both angiography and embolization of any type.

The historic control group (early) included 153 patients admitted with grade III to V BSI from 2007-2009 and the protocol group (late) included 168 patients with similar injuries treated from 2010-2012. The groups were well matched for age, gender, and splenic injury grade (Table 1 below).

From Miller et al. 2014

As expected, the use of angiography in NOM patients increased from 26% in the early group to 94% in the late group (p < 0.0001). Of those who underwent angiography, 71% in the early group and 86% in the late group underwent embolization. Notably, only 50% of those in the late group referred for angiography based on grade of splenic injury had contrast blush on CT. The NOM failure rate was 15% in the early group, compared with 5% in the late group managed according to protocol (p = 0.04).  Of the patients in the early group who failed NOM, none (n = 12) underwent angiography and all but one (n = 11) did not have contrast blush on CT. Of the patients in the late group who failed NOM (n = 9), only 4 had both angiography and embolization and the remaining 5 were considered protocol deviations due to angiography without embolization. Of those in whom the protocol was not followed (n= 16), the NOM failure rate was 31%.

This study demonstrated that using the criteria of grade III to V splenic injury to refer patients with BSI for angioembolization decreased the failure rate of NOM from 15% to 5%. Interestingly, only 50% of patients in the late group referred for angiography based on grade of injury had contrast blush on CT, suggesting that use of contrast blush alone as a criteria for angiography (as done in the early group) is inadequate. Further, patients in the late group who underwent angiography but were not embolized had a high NOM failure rate, emphasizing the central role of embolization in successful NOM of BSI.

Another single institution study reported on the utility of an early repeat imaging protocol to determine the need for angioembolization in patients undergoing NOM of BSI [3]. Prior to 2000, NOM patients had repeat CT scans at approximately 7 days to determine success or failure of NOM.  Since 2000, when splenic arterial embolization (SAE) became available at this institution, any splenic pseudoaneurysm (SPA) or arterial extravasation (AE) seen on initial CT has been treated with SAE. In addition, a protocol was instituted requiring a repeat CT scan at 48 hours for all patients undergoing NOM of BSI regardless of SAE at initial presentation. This approach was based on concern for delayed splenic rupture secondary to SPA. If SPA or AE is detected on the 48 hour CT scan, SAE is then performed.

The early cohort (prior to 2000) included 83 patients and the present cohort (since 2000) included 475 patients, with a significantly higher proportion of the total BSI population selected for NOM in the early vs. present cohort (77% vs. 53%, p < 0.01). The groups were well matched for age, Injury Severity Score, and distribution of splenic injury grade. The NOM failure rate in the early cohort was 12% (n = 10), compared with 0.6% (n = 3) in the present cohort (p < 0.01). A decrease in length of stay was also noted in the present cohort compared to the early cohort (6 vs. 8 days, p < 0.001). No significant differences between cohorts were seen in transfusion requirements or mortality. Of the patients in the present cohort, 5% underwent SAE after initial imaging and an additional 6% underwent SAE after repeat imaging at 48 hours. Delayed development of SPA and/or AE was increasingly likely with higher grade of splenic injury (Figure 3 below).

While the reduction in failure of NOM for SBI is significant when comparing the early and present cohorts, both the treatment modality and the repeat imaging protocol changed in the interim. Prior to 2000, angiography and embolization were not available, so even if a repeat CT scan had been done at 48 hours, the treatment for SPA or AE would have been operative. The more useful data from this study is that in the present cohort, about 6% of SPAs and/or AEs were not present on initial CT scan but seen at 48 hours, indicating a delayed presentation. It would be helpful to have clinical data from this 6% of patients to determine whether the SPA and/or AE was clinically significant (i.e. decreased hemoglobin/hematocrit, unstable vital signs, presence of peritonitis) since the embolization was performed based on imaging results alone.

The third article is a retrospective review of hemodynamically stable patients with BSI who underwent selective angioembolization (AE) based on risk factors associated with failure of NOM [4]. In this protocol, AE was performed for contrast blush on initial CT, grade IV to V injuries on initial CT, and/or decreasing hemoglobin after admission for NOM. Failure of NOM occurred if a patient required operative management at any time after attempted NOM with or without AE. Of the 539 patients admitted for NOM, 104 (19%) underwent AE and 435 (81%) did not. The indication for AE was contrast blush in 74%, grade IV to V injury in 14%, and decreasing hemoglobin in 11%. All AE for decreasing hemoglobin were in low grade injuries and performed based on the rate of decreasing hemoglobin within the first 48 hours of admission; all were hemodynamically stable and had successful splenic salvage after AE.

The NOM failure rate was 3.8% (n = 4) in the AE group and 4.4% (n = 19) in the no-AE group for a total cohort NOM failure rate of 4.3%. There was a significant decrease in failure of NOM with the addition of AE for grade IV (23% vs. 3%, p = 0.04) and grade V (63% vs. 9%, p = 0.03) splenic injuries (Table 2 below).  Multiple logistic regression analysis demonstrated that grade IV to V injuries and presence of contrast blush were independent risk factors for failure of NOM (p < 0.05).

The authors highlight the protocol factors which contributed to success:  “1. Hemodynamically unstable patients belong in the OR, 2. Selective use of AE for patients at high risk for failure (contrast blush on initial CT, high grade injuries [IV to V] on initial CT, and/or decreasing hemoglobin during NOM observation) rather than universally for all patients with BST, and 3. Delayed AE can be safely performed to salvage hemodynamically stable patients with decreasing hemoglobin levels.” 

Based on the results of these three studies, the use of angiography is associated with improved outcomes and the success of NOM in a particular subset of patients. Miller et al. highlighted the importance of using splenic injury grade as criteria for angioembolization and demonstrated a NOM failure rate of only 5% when AE was performed in all patients with grade III or greater injuries. The study by Leeper at al. emphasized using imaging criteria of both splenic pseudoaneurysm and arterial extravasation as indications for angioembolization regardless of grade of injury and demonstrated a NOM failure rate of less than 1%. Their results also demonstrated the utility of repeat imaging to identify delayed findings that may lead to splenic rupture, although the cost-benefit ratio of such a protocol must be determined. The selective protocol for angioembolization implemented by Bhullar et al. had the most convincing success when results demonstrated identical NOM failure rates in AE and no-AE cohorts when AE was performed for those at highest risk for NOM failure based on imaging findings, high splenic injury grade, or ongoing bleeding. These three studies independently support the EAST guidelines for the use of angiography in BSI (grade III or greater, presence of a contrast blush, moderate hemoperitoneum, or evidence of ongoing splenic bleeding) and help clarify the indications for angiography in patients undergoing NOM.

References:

1. Stassen NA, Bhullar I, Cheng JD, et al. Selective nonoperative management of blunt splenic injury: An Eastern Association for the Surgery of Trauma practice management guideline.  J Trauma Acute Care Surg 2012; 73 (5 Suppl 4): S294-S300.

2. Miller PR, Chang MC, Hoth JJ et al. Prospective trial of angiography and embolization for all grade III to V blunt splenic injuries: Nonoperative management success rate is significantly improved. J Am Coll Surg 2014; 218: 644-651.

3. Leeper WR, Leeper TJ, Ouellette D et al. Delayed hemorrhagic complications in the early nonoperative management of blunt splenic trauma: Early screening leads to a decrease in failure rate. J Trauma Acute Care Surg 2014; 76: 1349-1353.

4. Bhullar IS, Frykberg ER, Siragusa D et al. Selective angiographic embolization of blunt splenic traumatic injuries in adults decreases failure rate of nonoperative management. J Trauma 2012; 72:1127-1134.

from Dr. Megan Miller