Renal Sub-capsular Hemorrhage: Unique Imaging Findings and Role of Interventional Management

Introduction

Distal renal artery perforation during renal arterial stenting with a guidewire is an uncommon but potentially fatal complication.¹ Although guidewire perforation leading to subcapsular or perirenal hematoma may be successfully managed conservatively or by embolotherapy, surgical intervention is occasionally required.^2,3 Spontaneous renal subcapsular hemorrhage (SRSCH) can occur in patients with an underlying renal mass or in those with a bleeding diathesis or on anticoagulant medication.

Case Report

Case 1. A 65-year-old woman with difficult to control hypertension and right renal artery stenosis (revealed by computed tomography) was referred for elective renal artery stenting. The serum creatinine was 197µmol/l, and international normalized ratio (INR) was 1.2. Following selective cannulation of the right renal artery with a 5-Fr Simmons 1 catheter (Cordis, Miami Lakes, Florida), a 7-Fr Brite tip sheath (Cordis) was advanced through the stenosis over a 0.035” Rosen guidewire (Cook, Bloomington, Indiana). A 6mm x 14mm stent (Genesis, Cordis) was deployed over 0.018' (V-18, Boston Scientific, Natick Massachusetts) guidewire. Check angiography demonstrated adequate restoration of the renal arterial lumen. Intra-arterial administration of 3000U of Heparin was administered during the procedure.

The procedure was uneventful, and following puncture site closure with a 6-Fr Angioseal (St Jude Medical), the patient was immediately commenced on an intravenous heparin infusion at the rate of 750U/hr. At two hours post procedure, the patient developed a tender, right-flank swelling and hypotension associated with a drop in the blood hematocrit (Hb from 102g/l to 83g/l). Retroperitoneal hemorrhage related to guidewire perforation was suspected. The angiographic images were reviewed but none of the images demonstrated guidewire extension beyond the lateral margin of the right kidney. CT of the abdomen demonstrated a large subcapsular hematoma with extension into perinephric and pararenal spaces. On the arterial phase images, active extravasation of contrast from the posterior aspect of the midpole cortex of the right kidney was evident. The site of extravasation corresponded closely to the site of 0.018” guidewire placement at the time of renal arterial stent placement. Smaller areas of active contrast extravasation were also identified elsewhere over the renal cortical surface. Catheter angiography confirmed the CT findings of multiple areas of contrast extravasation, the most active corresponding with midpole bleeding site identified on the arterial phase.

Embolization of the bleeding artery in the midpole was performed using PVA particles. Check angiography confirmed cessation of extravasation from the embolized branch although there were noted to be a number of small, persistent areas of contrast extravasation in the periphery of the mid- and lower-pole regions. The feeding vessels to the bleeding sites were considered too small and ill defined to embolize. Reversal of anticoagulation was chosen in favor of more proximal embolization in order to preserve as much functional renal tissue as possible. Over the next 12 hours, the patient received 3 units of blood with no proportionate increase in the hematocrit (79 to 83g/l). A repeat angiogram was performed with embolization of two further actively bleeding lower pole cortical vessels (MWCE-18-1.0-0-Hilal, Cook). Following the repeat embolization, the patient became hemodynamically stable and further recovery was uneventful.

Case 2. A 70-year-old patient on warfarin therapy for cardiac arrhythmias presented to the Emergency Department with a history of acute left flank pain. The patient’s INR was 7.3 at the time of presentation. Although hemodynamically stable at the time of presentation, a significant fall in the patient’s blood pressure occurred over the next hour (BP 92/62 mm Hg). He was transfused with 2 units of FFP and an urgent abdominal CT scan ordered. This demonstrated a large left subcapsular hematoma extending into the perinephric and paranephric spaces. No underlying renal mass was identified. Arterial phase images demonstrated multiple sites of active contrast extravasation from the renal cortical surface. A subsequent catheter angiogram confirmed the presence of multiple sites of contrast extravasation. The most prominent bleeding site, situated in the upper pole of the kidney, was embolized with coils (MWCE-18-1.0-0-HILAL, Tornado Embolization Coils, Cook). Following the embolization procedure and normalization of his INR, the patient recovered uneventfully.

Discussion

Guidewire perforation of a distal renal artery branch is a rare but potentially serious complication of renal artery stent placement.³ A fatal case of guidewire perforation following renal arterial stenting has been reported by Axelrod et al.¹ As renal arterial stenting is usually performed using single plane (anterior or anterior oblique) imaging, it may not be possible to know the exact location of the guidewire tip along the plane of the x-ray beam. In our first patient, the site of most active contrast extravasation on arterial phase CT was from the posterior renal cortex, an area that was not seen in profile during the stenting procedure. A 0.018 guidewire, particularly one with a hydrophilic tip as used in our procedure, is easily passed into cortical vessels. Anticoagulation, routinely used in our institution during renal arterial stenting, further increases the risk of hemorrhagic complications in the event of guidewire perforation of the kidney. Additionally, restoration of blood flow into the kidney and exposing it to systemic arterial pressure by successful renal arterial stenting is another factor, which may increase the risk of hemorrhage following guidewire perforation. Renal subcapsular hemorrhage, either spontaneous in the anticoagulated patient or resulting from guidewire injury, can be managed conservatively in hemodynamically-stable patients.^3,4 Serial CT or US can be used in these cases to assess for resolution.⁴ Fine needle aspiration was used in addition to imaging to exclude an underlying neoplasm in one reported case of SRSCH, which was followed to complete resolution.⁵ Anticoagulation with markedly elevated INR was the sole identified risk factor for the development of subcapsular hemorrhage in our second patient. In view of the markedly elevated INR, hemodynamic instability and active extravasation visible on arterial phase CT, it was elected to perform immediate embolization in addition to reversal of anticoagulation in this individual. Rupture of an underlying tumor is an established cause of SRSCH.⁶

Extracorporeal shock wave lithotripsy has also been reported in the literature as a rare cause of iatrogenic SRSCH. In the reported case of post ESWL subcapsular renal hemorrhage, selective embolization of the renal artery was required following unsuccessful conservative management.⁷ Following perforation of the renal cortex, the blood dissects between the renal parenchyma and renal capsule, collecting in the subcapsular space. The stripping of the renal capsule from the underlying kidney results in the tearing of adjacent cortical arteries, which further increases bleeding. The tamponading effect of a contained subcapsular hematoma, which acts to suppress further bleeding, is rendered less effective if capsular perforation permits leaking into the perinephric and paranephric spaces. In the report by Axelrod, the radiological findings were correlated with postmortem gross examination of the kidney, which showed that the capsule had separated from the renal cortex. The point of perforation by the guidewire was localized, and multiple torn capsular branches were identified.¹

The imaging findings in cases of guidewire perforation of a distal renal artery branch with subcapsular bleeding are unique. The primary site of capsular perforation may be identified as a prominent site of contrast extravasation and possibly corresponds to the site of maximum thickness of the subcapsular hematoma. A characteristic “watering-can” appearance has been described on conventional angiography due to the contrast extravasation from multiple torn cortical vessels.¹ This results when the capsule is stripped off the renal cortex and multiple branches from the renal surface start bleeding into the sub-capsular space. This same appearance was clearly demonstrated on the arterial phase CT scans on both of our patients.

Superselective segmental renal artery catheterization and embolization is a safe and efficient method for the treatment of patients with severe renal hemorrhage.⁸ In both of our cases, our aim was to embolize the primary bleeding site at the time of initial angiography. On both occasions, small residual foci of contrast extravasation were identifiable on post-procedure angiography. These would correlate with secondary bleeding sites from torn cortical vessels following capsular separation. In our first patient, a second embolization performed because of continued bleeding was successful in stabilizing her hemodynamic status. In the second patient with spontaneous bleeding, a single embolization of the most prominent site of extravasation was adequate in achieving hemodynamic stability despite small residual areas of contrast extravasation persisting on the completion angiography. There was no change in the renal function in both patients.

Renal subcapsular hemorrhage may be a life-threatening problem that requires early detection and immediate angiographic intervention. CT with arterial phase scans greatly assists in detecting active hemorrhage and planning the appropriate intervention. The derived CT angiographic images can closely mimic the “watering-can” appearance of multiple bleeding sites described angiographically and may indicate that embolization of the primary bleeding cortical artery alone may be insufficient to control hemorrhage. To our knowledge, this is the first time this CT angiographic appearance has been described in the literature. Whilst the initial aim in the interventional treatment of SRSCH should be to “save the kidney” with a limited embolization, the CTA imaging findings of multiple bleeding sites may pre-empt a more aggressive approach to embolization in order to “save the life”.