top of page
15-1-4-3 figure 1.jpg
15-1-4-3 figure 2.jpg
15-1-4-3 figure 3.jpg
15-1-4-4 figure 4.jpg
15-1-4-4 figure 5.jpg
15-1-4-4 figure 6.jpg
Emergent Endovascular Management of Acute Arterial Bleeding after Initial Unsuccessful Surgical Control

Peter Chacha Magabe, Nelson Kimani

School of Medicine, University of Nairobi

Correspondence to: Dr Peter Magabe.



Background: Hemorrhage and the complications thereof have traditionally been managed surgically but over the last two decades endovascular techniques have been increasingly used. These include coiling, liquid and particle embolization among others. These minimally invasive techniques now available locally are safe and cost effective. They are however yet to be fully adopted in acute emergency settings.

Methods: Fifty-two patients with acute bleeding from known causes and whose first surgical intervention was not successful in controlling the bleeding were referred for emergent endovascular embolization.

 Results: The causes of the bleeding included post traumatic bleeds from intra-abdominal organ injury, iatrogenic pseudoaneurysms, massive hemoptysis among others. There was 100% technical and clinical success rate in hemorrhage control. None of the patients required a repeat endovascular procedure for bleeding control and no major complications were reported.

Conclusion: Endovascular embolization was successful in controlling bleeding in all of the patients. It is effective, safe and available as a first line choice. It should be considered in all patients presenting with severe hemorrhage from known causes in the initial presentation.

Keywords: Embolization, Interventional radiology, Vascular

Ann Afr Surg. 2018; 15(1):14-19


© 2018 Author. This work is licensed under the Creative Commons Attribution 4.0 International License

15-1-4-2 table 1.jpg
15-1-4-2 table 2.jpg


Patients presenting with bleeding and associated complications of hypovolemia and shock continue to account for a major proportion of emergencies handled in emergency departments of most hospitals. These range from gastrointestinal and bronchial bleeds to bleeding from traumatic injuries of major intra-abdominal organs. Whereas these are managed appropriately within the various medical specialties, little attention has been paid to the novel endovascular options available for management of these patients in the region. This is largely because these services and trained personnel were not previously available. Endovascular treatment options are currently available are minimally invasive, cost effective, life saving and most instances a better option to major surgical interventions.


All patients referred for embolization for acute arterial bleeds from a known cause and in whom initial surgical control was not successful were prospectively enrolled into the study over a four and half year period. Knowledge of availability of embolization services in the hospital by the referring physician was important as all referrals were from physicians who knew that the service was available. Patients who underwent embolization at the initial suggestion of the attending Interventional radiologist and without an initial surgical intervention for bleeding control were excluded. The surgical control interventions included explorative laparotomies, therapeutic endoscopies and colonoscopies for gastrointestinal bleeds, therapeutic bronchoscopies among many others. Patients were received in the Interventional angiography unit having undergone initial resuscitation with IV fluids and in some cases blood transfusion by the primary physicians. This was after the initial surgical procedure had failed in controlling the bleeding or when the patient rebled a few hours to 72 hours after the surgery. In two patients rebleeds occurred a week after the initial surgical control procedure. Each patient then underwent a percutaneous femoral puncture to obtain a femoral arterial access and a 5F or 6F vascular sheath subsequently inserted. Right femoral arterial access was used in all patients, followed by 5F vascular sheath in forty patients and 6F vascular sheath in twelve patients. The various arterial territories with suspected source of bleeding were accessed with 4F, 5F C2, Rim or Simmons 1 guiding catheters, threaded over a .035 hydrophillic guiding wire. In fifteen of the cases with predominantly mesenteric and bronchial artery bleeds, a progreat micro catheter (0.027D, Terumo Japan) was utilized inside the guiding catheter to gain distal access in small segmental arteries not otherwise accessible. Obtained arterial vascular images were analyzed for angiographic evidence of bleeding which included findings of pseudoaneuysms, abrupt vessel cut off, arteriovenous fistulas (AVF) and contrast extravasation. Embolization was done with coils, gel foam particles, PVA particles, histacryl/lipidiol mixture or absolute alcohol. A mixture of more than one embolic agent was used if necessary. Final post embolization DSA angiograms were carried out to confirm obliteration of the bleeding vessel and cessation of bleeding from the vascular territory.



Fifty two patients underwent emergent endovascular embolization after initial surgical intervention had failed to stop the bleeding. There was on average at least two repeat surgical interventions for bleeding control, with the highest repeats being for gastro intestinal bleeds. Patient ages ranged from 4-92 years with a mean age of 43 years. Of the thirty four patients who had initial hypovolemic shock at presentation, 90% underwent successful resuscitation with IV fluids and blood transfusions. Five patients (10%) continued resuscitation in the interventional suite and with the help of an anesthetist (Table 1). Gastro-intestinal and genito-urinary bleeds had the highest number with a total 48% of all the embolization procedures (Table 2). Angiographic evidence of cause bleeding was demonstrated in all cases, with pseudo aneurysm being the commonest cause (Table 3). Embolization was performed with coils in 16 patients, gel foam particles in 18, PVA particles in 6, histacryl/lipidiol mixture in 6 and absolute alcohol in 6 (Figures1-6). A combination of more than one embolic agent was used in twenty four patients predominantly coils/gelfoam and coils/liquid agent combination. No major complications were encountered. In five patients hematoma formation at femoral puncture site was controlled by prolonged focal pressure application lasting up to thirty minutes. No repeat interventions for bleeding control were required in subsequent follow up in all the patients.

Click to view table 1

Click to view table 2

Click to view figure 1 a,b,c


Click to view figure 2 a,b,c

Click to view figure 3 a,b,c


Click to view figure 4 a,b

Click to view figure 5 a,b,c


Click to view figure 6 a,b,c




Endovascular embolization for bleeding control has been widely used in the last few years and gaining ground in the region. Any bleeding artery can be accessed and occluded by endovascular techniques, making this the first choice method for bleeding control, unless an additional curative surgical procedure is required (1, 2). Because of lack of appropriate facilities and trained personnel in the past, this option was largely limited and patients continued to undergo traditional surgical approaches as demonstrated by the fifty two cases. These patients were referred to embolization service when the bleeding could not be stopped or when surgery was likely to result in loss of an organ (e.g. splenectomy). Right femoral arterial puncture is the preferred route by many interventionists and was the exclusive mode in the study. Other vascular access routes can be radial, brachial and even direct carotid or subclavian arterial punctures in certain cases (). The bleeding artery is usually localized at DSA angiography with characteristic appearance of a pseudo aneurysm with obvious contrast leakage, an associated arteriovenous fistula or angiographic blush (3).Most traumatic cases were associated with a pseudo aneurysm which leaked due to the false sac causing significant hemorrhage. A few of the patients had subtle angiographic signs of bleeding like abrupt arterial cut off in a case of splenic laceration, and irregular tuff of arterioles and capillaries secondary to neovascularization in a case of severe hemoptysis from pulmonary tuberculosis. Embolization for gastro intestinal bleeding and others follow a similar approach (4). Choice of embolic material is dependent on arterial territory being embolized, size of target vessel, primary pathology causing the bleeding, among other factors (). Coils are preferable in cases of pseudoaneuysms since the main arterial branch can be preserved. Particle embolization with gel foam or PVA particles may be indicated in cases of multiple feeder arteries to a hypervascular bleeding tumor. Histacryl and onyx, both liquid embolic agents are indicated in cases where there is need for higher rate of complete hemostasis (5, 6). In some cases there is use of more than one embolic agent (7). The embolization procedure is stopped once bleeding control is confirmed angiographically.Repeat embolization can be done if rebleeding occurs. This is uncommon but has been observed in arterial bleeds from pelvic fractures (8).Multidisciplinary management of primary cause of the bleed continues thereafter after endovascular control of the bleeding is achieved.



Endovascular management was successful in controlling bleeding in all of the patients after initial unsuccessful surgical intervention. It is effective, safe, affordable and available as a first line choice. It should be considered initially in all patients presenting with arterial bleeding.



  1. Pierre E, Rafael D, David C. ET Embolization for Multicompartmental Bleeding in Patients in Hemodynamically Unstable Condition: Prognostic Factors and Outcomes. J Vasc Interv Radiol. Vol. 23, Issue 6, p751–60.

  2. Lars-Gunnar E, Mikael L, Magnus S, et al. Transcatheter Arterial Embolization versus Surgery in the Treatment of Upper Gastrointestinal Bleeding after Therapeutic Endoscopy Failure. J Vasc Interv Radiol., Vol. 19, Issue 10, p1413–8.

  3. Siddharth A. P, Michael A.G, James S. N Et al Effectiveness of Coil Embolization in Angiographically Detectable versus Non-detectable Sources of Upper Gastrointestinal Hemorrhage J Vasc Interv Radiol. Vol. 20, Issue 4, p461–6.

  4. S.P. K, A. E, B. Somarouthu Et al. Wicky Angiographic Intervention in Patients with Gastrointestinal Bleeding. J Vasc Interv Radiol. Vol. 24, Issue 4, p157

  5. Shinsaku Y, Takashi I, Toshio K Et al. Transcatheter Arterial Embolization of Acute Arterial Bleeding in the Upper and Lower Gastrointestinal Tract with N-Butyl-2-Cyanoacrylate. J Vasc Interv Radiol. Vol. 24, Issue 3, p422–31.

  6. Jose Garbo. Selective Arterial Embolization with Ethylene–Vinyl Alcohol Copolymer for Control of Massive Lower Gastrointestinal Bleeding: Feasibility and Initial Experience J Vasc Interv Radiol. Vol. 25, Issue 12, p2006.

  7. T Koichiro Y, Atsuhiro N, Naoshi T Et al. Transcatheter Arterial Embolization of Ruptured Pseudoaneurysms with Coils and n-Butyl Cyanoacrylate J Vasc Interv Radiol. Vol. 11, Issue 1, p66–72.

  8. Fang JF1, Shih LY, Wong YC, et al. Repeat Transcatheter Arterial Embolization for the Management of Pelvic arterial Hemorrhage.J Trauma. 2009; 66(2):429-35.


Read More >>>
bottom of page