UROGENITAL TRAUMA

doi: 10.1590/S1677-553820090006000017

Radiographic predictors of need for angiographic embolization after traumatic renal injury
Nuss GR, Morey AF, Jenkins AC, Pruitt JH, Dugi DD 3rd, Morse B, Shariat SF
Department of Urology, University of Texas Southwestern Medical Centre, Dallas, Texas, USA
J Trauma. 2009; 67: 578-82; discussion 582.

• Background: Although the American Association of the Surgery for Trauma Organ Injury Scale is the gold standard for staging renal trauma, it does not address characteristics of perirenal hematomas that may indicate significant hemorrhage. Angiographic embolization has become well established as an effective method for achieving hemostasis. We evaluated two novel radiographic indicators--perirenal hematoma size and intravascular contrast extravasation (ICE)--to test their association with subsequent angiographic embolization.
  Methods: Among 194 patients with renal trauma between 1999 and 2004, 52 having a grade 3 (n = 33) or grade 4 (n = 19) renal laceration were identified. Computed tomography scans were reviewed by a staff radiologist and urologist blinded to outcomes. ICE was defined as contrast within the perirenal hematoma during the portal venous phase having signal density matching contrast in the renal artery. Hematoma size was determined in four ways: hematoma area (HA), hematoma to kidney area ratio (HKR), difference between hematoma and kidney area (HKD), and perirenal hematoma rim distance (PRD).
  Results: Of the 52 patients, 8 had ICE and 4 of these (50%) required embolization, whereas none of the 42 (0%) patients without ICE needed embolization (p = 0.001). Likewise, all four measures of perirenal hematoma size assessed were significantly greater in patients receiving embolization [HA (128.3 vs. 75.4 cm, p = 0.009), HKR (2.75 vs. 1.65, p = 0.008), HKD (76.5 vs. 30.2 cm, p = 0.006), and PRD (4.0 vs. 2.5 cm, p = 0.041)].
  Conclusion: Perirenal hematoma size and ICE are readily detectible radiographic features and are associated with the need for angiographic embolization.

• Editorial Comment
  The concept on controlling arterial renal bleeding with transcatheter embolization is a concept that has been around since the 1970s. However, over the years the technical skills of interventional radiologists, imaging equipment and their ability to perform superselective embolization of even the smallest vessels has gradually improved. In general, renal injuries that do not involve avulsion of the renal hilum can typically be managed nonoperatively. The assumption is that nearly 85-90% of Grade 3 and 4 renal injuries will have venous bleeds (or very small arterial bleeds) that will fill the confined space of Gerota’s fascia and tamponade. When the bleeds are arterial, and if they involve larger segmental vessels, the bleeding will not tamponade and thus require subsequent embolization. The key is how do we predict who will not stop bleeding and need either a surgical exploration or transcatheter embolization? Can we tell by signs on the CT?
  The San Francisco General and the Parkland Hospital Groups advocate an early and aggressive policy of embolization for renal injuries. As an out growth of the success of selective embolization has had with blunt splenic injuries over the years, where signs like the presence of a vascular blush suggested a significant bleed – embolization can equally be used for the blunt kidney injury. In general I feel we under utilize angiography and selective embolization for our major blunt renal trauma patients.
  Embolization coils that are in common use today are made of platinum and highly radio-opaque. Platinum is a softer metal so the coils can be tighter packed and cause less vessel wall injury. Each coil also has multiple Dacron side fibers which markedly increase thrombogenicity. The coils are 0.018”, 0.035” and 0.038” in diameter and thus can typically be deployed via a 5Fr Angiocath.
  The San Francisco General Group addresses more the patient who you admit and follow conservatively and who do you decide to send to interventional radiology in a delayed fashion. Most delayed renal bleeding occurs after 5 or more days as it takes time for the tamponaded hematoma to start to lyse and then release some of the tamponade effect. When the delayed bleed is from an arterial injury it typically is a pseudo-aneurysm. Breyer et al. concluded that potential criteria for angiography are persistent bleeding from segmental artery with or without parenchymal laceration, an unstable patient with Grade 3 or 4 injury, persistent and significant gross hematuria, or a rapidly declining HCT – needing > 2 u pRBC transfused in 24 hours. In our experience at Barnes Hospital experience, we use > 3 u RBC/ 24 hr transfused period to decide on further renal intervention. However, these are not hard fast rules just guidelines for intervention.
  They further state that Grade 5 renal injuries, main renal artery or vein avulsion injuries - which by definition are life-threatening injuries should not be managed by an attempt at embolization but rather prompt surgery. The Parkland Group addresses more the acute CT findings and need for embolization. Just as with blunt splenic injuries, the presence of intravascular contrast extravasation (“a contrast blush”) on the arterial phase of a CT scan and the presence of a large perirenal hematoma (> 4 cm from renal capsule to hematoma edge) greatly predicted the likelihood for persistent bleeding and thus a significant vascular injury. The Parkland group argues that the AAST injury grading scale for major renal injuries should be subclassified into Grade 4, A = low risk and G4 B = High Risk, where a blush and large hematoma were present. I think that such a suggestion is a good one and would help to better guide therapy. Overall, I think we need to be more vigilant about identifying early signs of significant arterial renal injuries and early and quickly sending these patients to the interventional radiologist for super selective embolization.

Dr. Steven B. Brandes
Associate Professor, Division of Urologic Surgery
Washington University in St. Louis
St. Louis, Missouri, USA
E-mail: brandess@wudosis.wustl.edu