Article, Hematology

Management of spontaneous soft-tissue hemorrhage secondary to anticoagulant therapy: A cohort study

a b s t r a c t

Study objective: The optimal management of patients receiving heparin, warfarin or direct anti-coagulant therapy who experience spontaneous, severe, life-threatening soft-tissue hemorrhage (SSTH) is unclear. The purpose of this study is to investigate efficacy and safety of the interventional protocol implemented in our department. Methods: In this retrospective cohort study, we analyzed data from 80 consecutive patients with SSTH secondary to anticoagulation therapy diagnosed by the appropriate computed tomography scan. All patients received a structured clinical pathway, including aggressive resuscitation, reversal of coagulopathy when indicated, Inter- ventional Radiology procedures by transcatheter embolization (TE), clinical observation and repeated laboratory controls.

Results: We enrolled 80 patients from 2013 to 2017. Angiography was performed in 60 patients (75%). It revealed the bleeding site in 46 cases, and a TE was performed in all. The rates of technical success of TE, primary Clinical success and Bleeding control were 98% (45/46), 91% (73/80) and 89% (71/80) respectively. In 5 patients (6%) the control of the bleeding was obtained with a second TE. Short-term and 30-day mortality was 5% (4 patients) and 11% (9 patients), respectively. No adverse events were observed.

Conclusion: A structured clinical pathway, including TE seems to be an effective and safe method to manage the patients with SSTH due to anticoagulant treatment.

(C) 2018

Introduction

Background

Spontaneous soft-tissue hemorrhage (SSTH) is defined by the occur- rence of extravasation of blood in soft tissue, specifically in a muscle, most often in the abdominal waist not associated with trauma, surgery or interventional procedure [1-3]. Diagnosis of soft-tissue hemorrhage is often delayed because of the lack of specific presenting symptoms and SSTH can suddenly deteriorate and become a life-threatening con- dition [4], with high mortality for shock or complications, such as com- partment syndrome. The development of SSTB is significantly associated with anticoagulant treatment, especially in the elderly [5,6]. The majority of patients need intensive-unit-level of care and aggressive support, blood transfusions and reversal of anticoagulation [7,8].

? We disclose any grants or other financial support for this article.

* Corresponding author at: Department of Internal Medicine, Ospedali Riuniti di Ancona, Via Conca n.71, 60126 Ancona, Italy.

E-mail address: [email protected] (V.G. Menditto).

Multidetector computed tomography angiography (MDCTA) can show active bleeding requiring immediate treatment [9,10]. Patients for whom conservative management was unsuccessful, with the need of re- peated transfusions or in cases of hemodynamic instability, are then usually referred for angiography followed by transcatheter emboliza- tion (TE) [11-14]. Surgery is rarely indicated [15].

Importance

Management of life-threatening anticoagulation-related SSTH re- mains controversial because there are currently no specific guidelines. Moreover, the evidence is limited to retrospective studies, such as case reports [16-20] and small case series, and large prospective studies are lacking. Several studies have reported the benefit of embolization in se- vere cases [21-28], but they analyzed only the patients who underwent endovascular treatment and not those who did not, and the clinical set- ting for these studies was isolated to the intensive care units [29]. Rele- vant questions remain unanswered: can a Diagnostic protocol based on MDCTA screen patients with anticoagulation-related SSTH for the best treatment? Can specific therapeutic protocol impact hard outcomes,

https://doi.org/10.1016/j.ajem.2018.03.061

0735-6757/(C) 2018

such as short-term mortality, of patients with anticoagulation-related SSTH?

Goals of this investigation

The primary goal of this study was to report the safety and efficacy of a diagnostic and therapeutic protocol in our semi-intensive unit for the management of anticoagulation-related SSTH.

The clinical outcomes evaluated were the short-term and 30-day mortality, the bleeding control rate, the primary clinical success rate and the technical and primary success rate of TE.

Materials and methods

Study design and setting

We performed this retrospective cohort study at Academic Hospital in Ancona, Italy. Because the study was observational, local ethics com- mittee exempted it from formal review. We followed the “Strengthen- ing the Reporting of Observational Studies in Epidemiology” (STROBE) recommendations for reporting this observational study [30].

Data collection

Two blinded investigators reviewed the electronic medical records of the wound information system (File Maker Pro 4.0) using manual re- search for the Inclusion/exclusion criteria of the study. For eligible pa- tients we recorded: a) demographics; b) indication for anticoagulation; c) International normalized ratio when appropri- ate; d) indication for the procedure including signs and anatomic local- ization of active bleeding at MDCTA; e) angiographic findings, embolic material used and the complications of TE such as ischemic or infectious complications in the embolized territories, non-target embolization, or puncture site hemorrhage; f) the Transfusion requirements before and after the embolization procedure during the observation period.

Identification and selection of study population

Two blinded investigators analyzed all consecutive patients admit- ted (n = 4030) at our semi-intensive unit (“Medicina d’Urgenza”) from January 2013 to December 2016. To be sure that all eligible pa- tients were included, other two blinded investigators screened the elec- tronic medical record of the hospital (Sistema Informatico Ospedaliero, Fly Tecnologie e Servizi, Perugia, Italy) using the appropriate Interna- tional Classification of Diseases codes and they found no missed sub- jects. SSTH was defined as an occurrence by appropriate MDCTA of a hematoma (Anterior abdominal wall, iliopsoas compartment, gluteal muscle and thighs) not associated with trauma, surgery, interventional procedure within 7 days or specific organic disease [2].

We selected patients according to the following inclusion and exclu-

sion criteria.

Inclusion criteria: age, older than 14 years; SSTH; period from symp- toms to appropriate CT scan b48 h and from CT scan to the admission in our semi-intensive unit b48 h; anticoagulation therapy (unfractioned heparin, Low Molecular Weight Heparin, fondaparinux, warfarin, dabigatran, rivaroxaban, epixaban, edoxaban) for at least 1 week.

Exclusion criteria: concomitant gastrointestinal, intracranial, or intraparenchymal visceral bleeding; incomplete clinical data available.

We defined life threatening patients as those who presented with at least one of these characteristics: hypotension at admission; shock

index N0.9 [31], Hb b8 g/dl at admission, >=3RBC units before admission, INR >=3 on Warfarin therapy [32,33].

Protocol of diagnostic and therapeutic management

We treated all patients with intravenous fluid resuscitation therapy and, if Hb b8 g/dl with RBC transfusion. If the patient is anticoagulated with warfarin, a reversal of coagulopathy was immediately performed with 10 mg intravenous vitamin K plus 50 units/kg 4-factor prothrom- bin complex concentrate [34]; eventual Antiplatelet therapy was stopped as well.

After admission, patients were screened for angiography by a multi- disciplinary discussion involving the internist in charge of the patient and the radiologist. Renal failure (clearance b 30 mL/min) at diagnosis has not been considered an absolute contraindication for the adminis- tration of contrast media. Hemodynamic unstable patients were imme- diately referred for embolization. Stable patients with active contrast extravasation or contrast blush (CB), defined as a focal area of high den- sity within the hematoma on MDTCA scan, were diagnosed with ‘active bleeding’. These patients underwent also a TE as common practice at our institution. Stable patients without CB on MDTCA scan that were re- fractory to conservative management and needing repeated transfu- sions were candidate to embolization. All the patients were monitored in our semi-intensive unit until recovery with clinical and laboratory controls, every 6 h for the first 12 h after TE, then every 24 h until to the achievement of persistent hemodynamic stability.

Angiography and transcatheter embolization

We performed angiographic evaluation and subsequent TE via fem- oral access under fluoroscopic guidance (Philips, Allura Xper R7 or Axiom Artis Zee, Siemens Healthcare) in the angiography suite. In cases of retroperitoneal bleeding, aortography with a catheter 4F UF Cordis was performed, followed by selective lumbar artery catheteriza- tion using 4F Simmons Cordis or Terumo or 4F Cobra Cordis or Terumo. In cases of anterior abdominal wall, gluteal, or thigh bleeding, a contra- lateral femoral approach was preferred, and a selective angiographic evaluation of the internal iliac, common femoral, or deep femoral artery was performed using a 4F Ber Cordis. In all cases, superselective cathe- terization of the bleeding vessel was performed using a microcatheter

2.7F Progreat Terumo. The following embolic materials were used: microcoil and sterile compressed sponge.

End-points definition

‘Technical success’ was defined as the complete Arterial occlusion of the target bleeding arteries [35,36]. ‘Primary clinical success’ was de- fined as no need for repeated transfusion after the admission (b3 RBC units per 48 h) [35,36]. ‘Bleeding control’ was defined as hemodynamic stability persistent from admission to the discharge [35,36]. ‘Short term mortality’ was defined as death occurred from the admission to the dis- charge and the ’30 day-mortality’ as the death occurred within 30 days from the admission.

Outcome measures

Clinical outcomes were retrospectively assigned based on the analy- sis of the electronic hospital record (File Maker Pro 4.0).

The end points of our study were defined as follows: 1. death (short term mortality); 2. need of surgery because of compartment syndrome;

achievement of bleeding control, primary clinical success and/or tech- nical success as defined above. 30 day-mortality was retrospectively evaluated by telephone calls to the patients, their relatives or their fam- ily physicians. Data were analyzed from admission for one month or until the patient’s death.

2.8. Primary data analysis

We analyzed our data descriptively with SPSS (version 13; SPSS Inc., Chicago, IL).

Results

Patient flow is show in the Fig. 1. 4030 patients were evaluated; 168 had soft tissue hematoma as first or secondary diagnosis. Of these 168 patients, 80 were excluded because of traumatic or post-surgical/ procedural soft tissue hematoma, 2 patients because they weren’t on anticoagulation therapy and 6 patients because of lack of complete data. 80 (48%) were eligible for the study and their baseline characteris- tics are shown in Table 1.

Median time from onset of symptoms to the admission was one day (range 0-7). In 37% of patients, SSTH was located in the rectus sheath or in the retroperitoneal space. 68 patients (85%) had a life-threatening condition. Of these 68 patients, 23 were on therapy with warfarin and 52% had an INR N3 (12/23). Additionally, 15 patients (19%) were receiv- ing both anticoagulation and Antiplatelet agents.

At admission all patients (80/80) received IV fluid resuscitation ther- apy and 69 (86%) underwent RBC transfusion. Moreover, all patients re- ceiving warfarin underwent reversal of coagulopathy within 6 h from the onset of symptoms. MDCTA was performed whiting 2 h from admis- sion in all patients. 60 (75%) patients underwent urgent arteriography due to active bleeding on MDCTA. 13 of these patients were hypotensive at admission. Active bleeding was confirmed by angiography in 46 of 60 patients (76%) after successful catheterization of the target artery and TE was performed within the First two days from admission. The rate of technical success of TE was 98% (45/46). No complications of TE were observed. 16 (75%) out of 20 patients who didn’t undergo

arteriography, had potential life threatening bleeding at the time of first observation, but they showed a rapid good response to medical treatment.

Short-term and 30-day mortality was 5% (4 patients) and 11% (9 pa- tients), respectively. Two patients died early because of hemorrhagic shock and in one case cardiac arrest occurred during arteriography; the other two patients died because of septic shock from infection of the hematoma after TE. Of the five patients who had delayed death, 2 underwent arteriography, but not TE, the other 3 patients died as a con- sequence of hospital related infection (Fig. 1). Primary clinical success and bleeding control was achieved in 91% (73/80) and 89% (71/80) re- spectively (Table 2). In 5 patients (6%) the control of the bleeding re- quired a second TE (Fig. 1). One patient was admitted in intensive care unit because of acute respiratory distress, and he recovered. Two other patients had surgery to decompress a gluteal and a well chest compartment syndrome respectively. We found a sustained increase in hemoglobin from a mean value of 7.79 g/dl before admission to

9.49 g/dl at discharge/transfer (p = 0.002; Table 2). Moreover, there was a decrease in transfusion requirements (a mean of 2.8 and 1.9 re- spectively before admission and at discharge/transfer, p b 0.001; Table 2). We found three cases of hospital related infections. Table 3 shows how artery was involved and which embolic materials were used in the technical successfully performed TE.

Limitations

First, our study was conducted in a single third-level hospital, and this limits our results’ reproducibility. However, we believe that our hospital is representative of the setting in which these patients would be treated. Second, the retrospective study design raises the risk of selection- or information-bias. Third, we evaluated the re-introduction of anti-thrombotic therapy only in a sub-group of patients so, even if

1

4

Early survivors n=42 (93%) 30days survivors n=39 (87%)

1 UTI admission (2%) ; 1 surgery (2%)

Early survivors n=14 (93%) 30days survivors n=14 (93%) 0 UTI admission (0%)

Early survivors n=20 (100%) 30days survivors n=18 (90%)

0 UTI admission; 1 surgery (2%)

additional embolization

Embolization n=45 (75%)

No embolization n=15 (25%)

Arteriography n=60 (75%)

No arteriography n=20 (25%)

Eligible n=80 (48%)

Excluded because of uncompleted data n=6 (3%)

Excluded because spontaneous hematoma but not anticoagulation n= 2 (1%)

Excluded because of early iatrogenic cause (+/- anticoagulation) n= 80 (48%)

Junuary 2013-August 2017:

assessed for eligibility (soft tissue hematoma) n=168

Fig. 1. Selection and study group with the overall prevalence of outcomes.

Table 1

Characteristics of the study population.

Table 3

Characteristics of 45 technical successfully performed transcatheter embolization.

Characteristics Value

Total population 80

Age median (range, SD), y 79 (36-102, 11.1)

Female (%) 46 (57)

Platelet count at admission mean (range, SD) x 109/l 223 (35-987, 195) Antiplatelet therapy (%) 15 (19%)

Indication for anticoagulant therapy (%)

Rhythmic cardiopathy 44 (55)

Venous thromboembolic disease 13 (16)

Prophylaxis 17 (21)

Prosthetic valve 6 (8)

Anticoagulant therapy (%)

Enoxaparin 31 (38)

Enoxaparin prophylactic dose 6 (8)

Warfarin +- enoxaparina 23 (28)

Direct oral anticoagulantsb 2 (4)

Fondaparinux 10 (12)

Fondaparinux prophylactic dose 7 (8)

UFH 1 (2)

Hematoma location (%)

Rectum 29 (37)

RPH 17 (21)

Thigh/gluteal 15 (19)

Iliopsoas 10 (12)

Chest wall 9 (11)

Life threatening patientsc (%)

Total 68 (85)

Hb b8 g/dl at admission 51 (64)

>=3 RBC units before admission 40 (50)

Shock Index N0,9 23 (29)

Hypotensiond at admission 13 (16)

INR >=3 on warfarin therapy (%) 12 (15)

Compartment syndrome (%) 3 (4)

Minimal Hb before admission median (range, SD) g/dl 7.8 (4.7-11.1, 1.3) Active bleeding (CB) on CT scan (%) 60 (75)

Active bleeding site on arteriography (%) 46 (57)

UFH: unfractionated heparin. RPH: Retroperitoneal hematoma. Hb: hemoglobin. SD: stan- dard deviation. RBC: red blood cell. INR: international normalized ratio. CB: contrast blush. CT: computed tomography.

a 4 patients on therapy with warfarin and enoxaparin.

b One patient on therapy with rivaroxaban and one with apixaban.

c Life threatening condition was defined as presence of at least one of these character- istics: hypotension at admission; shock index N0.9, Hb b8 g/dl at admission, >=3RBC units before admission, INR >=3 on warfarin therapy.

d Hypotension = PAs b90 mm Hg.

it was randomly sorted, the numbers are too small to speculate about any role of re-introduction of anticoagulants on mortality. However, the hard outcomes, as short and intermediate mortality, were analyzed in the whole population. Fourth, during data collection, we didn’t for- mally evaluate patients’ comorbidities, a potential confounding factor for death. However, the inclusion criteria didn’t impose restrictions, and the study population fairly represents the common practice.

Table 2

Treatment and outcomes of the study population.

Characteristics

Value

Total population

80

RBC units after admission (mean)

1,2 (1.6)

Angiography (%)

60 (75)

Attempted TE (%)

46 (57)

Technical success of TE (%)

45/46 (98%)

Hb at discharge median (range, SD) g/dl

9.4 (6.6-12.4, 1.2)

Primary clinical success (%)

73 (91)

Bleeding control (%)

71 (89)

Second TE (%)

5 (6)

Short term mortality

4 (5)

30 days mortality

9 (11)

Length of stay days, mean (SD)

5.3 (3.6)

RBC: red blood cell. TE: transcatheter embolization. SD: standard deviation. Hb: hemoglobin.

Characteristics Value

Total population 45

Artery involved? (%)

Inferior epigastric

19 (43)

Lumbar

9 (20)

Branches of femoral

8 (19)

Iliolumbar

3 (6)

Inferior gluteal

3 (6)

Intercostal

2 (4)

Humeral

1 (2)

Embolic materials (%)

Amagnetic spiral alone

35 (78)

Amagnetic spiral + fibrin clot

10 (22)

Discussion

SSTH is less common than traumatic STH. However, in recent years, its incidence has greatly increased because of the broader use of antico- agulant therapies in Elderly people [37,38]. Indeed, in our study popula- tion, we have also observed a progressive increase in the incidence of SSTH, with doubling of the cases observed from twelve cases in 2013 to twenty-four cases in 2016, as described elsewhere [38].

To date, few treatment algorithms for managing patients with SSTH have been reported [7,8]. Moreover, management of life-threatening SSTH remains controversial, lacking large studies [11-14,16,17,22,39], and the majority of patients received intensive-care-unit-level of care [29]. To our knowledge, our series is the largest so far reported. The al- gorithm proposed in this study was applied in semi-intensive setting and included not only the patients who underwent TE, but also those who didn’t. We found that 98% (166/168) of screened patients were anticoagulated. This value appears a little bit higher than reported in lit- erature (86-89%) [13,36]. Cherry and Mueller [4] and Llitjos et al. [29] found that respectively 69% of patients with rectus sheath hematoma and 72% of patients with in ilio-psoas hematoma were on anticoagulation therapy. However, in our cohort, patients with recent abdominal surgery or trauma have been excluded, so reducing inci- dence of SSTH in non-Anticoagulated patients. Furthermore, as majority of our included subjects were transferred from other hospitals, it is pos- sible that non-anticoagulated patients with minimal, not severe, SSTH are underrepresented in our sample. Our patients were at least as severe as the patients presented in literature [2,29,40]; they were old (median age 79 years) with Multiple comorbidities, all (not some, as in the ma- jority of previous reports) receiving anticoagulant therapy and 85% of our study population had life-threatening condition. Also, the most fre- quent sites of hematoma, such as rectus sheath, retroperitoneal, thigh/ gluteal and iliopsoas, were comparable to those from other series [2,35,40].

Zissin et al. [9], Farrelly et al. [13], and Guzzardi et al. [38] suggested

that algorithm integrating MDCTA findings seems to be useful for selecting patients and directing TE. We confirmed that MDTCA was ef- fective to detect contrast extravasation in 80% of the cases and selective catheterizations were started with the target artery defined with the MDCTA in 75% of performed angiography. However, the present data suggest that the impact of MDCTA on patient outcomes still needs fur- ther prospective evaluation. In fact, 30-day mortality in our patients who hadn’t active bleeding on MDCTA was up to 10%.

The role of the TE in patients with SSTH management remains un- clear and there are no validated guidelines. In our experience, TE achieved technical success in 98% of cases, in agreement with all pub- lished studies [12,13,35,36]. Despite this high rate of technical success, we observed a postoperative mortality of 7% with 4/4 bleeding-related death and a 30 day mortality of 11% with 3/5 bleeding-related death. These data probably reflect the consequences of a hemodynamically sig- nificant bleed in an elderly population with multiple comorbidities. It is

therefore imperative to develop more effective treatment strategies for SSHT management. Other studies [3,12,13,28,35,36,38,40] have shown higher short term and 30-day mortality respectively from 6% to 12% and from 10% to 32%, despite a primary clinical success consistent with our 91%. Taking into account that not all patients were taking anti- coagulants in those series, our protocol seems to have an impact on prognosis of patients with SSTH, even if applied in a semi-intensive set- ting. The structured nature of the proposed protocols and the well- established expertise of all professionals involved could be the corner- stones to build dedicated multidisciplinary clinical pathways for the care of patients with life threatening SSTH.

Finally, in our experience we used a magnetic spiral and fibrin clot to close bleeding sources, as we are confident that these materials can be easily handled. Other reports, however, have widely described the use of absorbable gel [41], which could lead to permanent vessel occlusion in small vessels, or, more recently, N-butyl-2-cyanoacrylate, which could make difficult the catheter removal [28].

In conclusion, our data support a structured clinical pathway, includ- ing TE, to manage the patients with SSTH secondary to anticoagulant treatment. This protocol could contribute to lower the high mortality of this severe condition.

References

  1. Titone C, Lipsius M, Krakauer JS. “Spontaneous” hematoma of the rectus abdominis muscle: critical review of 50 cases with emphasis on early diagnosis and treatment. Surgery 1972;72(4):568-72.
  2. Dohan A, Darnige L, Sapoval M, et al. Spontaneous soft tissue hematomas. Diagn Interv Imaging 2015;96(7-8):789-96.
  3. Sunga K, Bellolio F, Gilmore R, et al. Spontaneous retroperitoneal hematoma: etiol- ogy, characteristics, management, and outcome. J Emerg Med 2012;43:e157-1.
  4. Cherry WB, Mueller PS. Rectus sheath hematoma: review of 126 cases at a single in- stitution. Medicine (Baltimore) 2006;85(2):105-10.
  5. Hodin E, Dass T. Spontaneous Retroperitoneal hemorrhage complicating anticoagu- lant therapy. Ann Surg 1969;170:848-51.
  6. Zidane M, Schram MT, Planken EW, et al. Frequency of major hemorrhage in pa- tients treated with unfractionated Intravenous heparin for deep venous thrombosis or pulmonary embolism: a study in routine clinical practice. Arch Intern Med 2000; 160(15):2369-73.
  7. Berna JD, Zuazu I, Madrigal M, et al. Conservative treatment of large rectus sheath hematoma in patients undergoing anticoagulant therapy. Abdom Imaging 2000;25 (3):230-4.
  8. Ekici Y, Aydogan C, Yagmurdur MC, et al. Treatment approaches for spontaneous ret- roperitoneal bleeding. Int Surg 2009;94:257-62.
  9. Zissin R, Ellis M, Gayer C. The CT findings of abdominal anticoagulant-related hema- tomas. Semin Ultrasound CT MR 2006;27:117-25.
  10. Wang ZW, Xue HD, Li XG, et al. Life-threatening spontaneous retroperitoneal haem- orrhage: role of multidetector CT-angiography for the emergency management. Chin Med Sci J 2016;31(1):43-8.
  11. Sharafuddin MJ, Andresen KJ, Sun S, et al. Spontaneous extraperitoneal hemorrhage with Hemodynamic collapse in patients undergoing anticoagulation: management with selective arterial embolization. J Vasc Interv Radiol 2001;12(10):1231-4.
  12. Isokangas JM, Perala JM. endovascular embolization of spontaneous retroperitoneal hemorrhage secondary to anticoagulant treatment. Cardiovasc Intervent Radiol 2004;27(6):607-11.
  13. Farrelly C, Fidelman N, Durack JC, et al. transcatheter arterial embolization of spon- taneous life-threatening extraperitoneal hemorrhage. J Vasc Interv Radiol 2011;22 (10):1396-402.
  14. Wick MC, Klocker J, Grundtman C, et al. Transcatheter embolization for the manage- ment of acute active inferior epiGastric artery hemorrhages. J Endovasc Ther 2013;20 (4):561-7.
  15. Pode D, Caine M. Spontaneous retroperitoneal hemorrhage. J Urol 1992;147(2): 311-8.
  16. Dabney A, Bastani B. Enoxaparin-associated severe retroperitoneal bleeding and ab- dominal compartment syndrome: a report of two cases. Intensive Care Med 2001;27 (12):1954-7.
  17. Vaya A, Mira Y, Aznar J, et al. Enoxaparin-related fatal spontaneous retroperitoneal hematoma in the elderly. Thromb Res 2003;110:69-71.
  18. Anton E, Marti J. Enoxaparin-associated spontaneous thigh haematoma. Age Ageing 2004;33(6):641-2.
  19. Surani S, Estement B, Manchandan S, et al. Spontaneous extraperitoneal lumbar ar- tery hemorrhage. J Emerg Med 2011;40(6):e111-.
  20. Tsai JL, Yang PJ, Lin HY, et al. Spontaneous iliopsoas hematoma. J Emerg Med 2016; 51(3):e53-.
  21. Katsumori T, Nakajima K. A case of spontaneous hemorrhage of the abdominal wall caused by rupture of a deep iliac Circumflex artery treated by transcatheter arterial embolization. Eur Radiol 1998;8(4):550-2.
  22. Qanadli SD, El Hajjam M, Mignon F, et al. Life-threatening spontaneous psoas haematoma treated by transcatheter arterial embolization. Eur Radiol 1999;9(6): 1231-4.
  23. Pathi R, Voyvodic F, Thompson WR. Spontaneous extraperitoneal haemorrhage: computed tomography diagnosis and treatment by selective arterial embolization. Australas Radiol 2004;48(2):123-8.
  24. Zissin R, Gayer G, Kots E, et al. Transcatheter arterial embolization of concurrent spontaneous hematomas of the rectus sheath and psoas muscle in patients undergo- ing anticoagulation. Cardiovasc Intervent Radiol 2004;27(6):659-62.
  25. Zissin R, Gayer G, Kots E, et al. Transcatheter arterial embolisation in anticoagulant- related haematoma-a current therapeutic option: a report of four patients and re- view of the literature. Int J Clin Pract 2007;61(8):1321-7.
  26. Wick MC, Klocker J, Grundtman C, et al. Transcatheter embolization for the manage- ment of acute active Inferior epigastric artery hemorrhages. J Endovasc Ther 2013;20 (4):561-7.
  27. Kolber MK, Shukla PA, Kumar A, et al. Ethylene vinyl alcohol copolymer (onyx) em- bolization for acute hemorrhage: a systematic review of peripheral applications. J Vasc Interv Radiol 2015;26(6):809-15.
  28. Ozyer U. Transcatheter arterial embolization with N-butyl-2-cyanoacrylate in the management of spontaneous hematomas. Cardiovasc Intervent Radiol 2017;40(1): 41-9.
  29. Llitjos JF, Daviaud F, Grimaldi D, et al. Ilio-psoas hematoma in the intensive care unit: a multicentric study. Ann Intensive Care 2016;6(1):8-14.
  30. von Elm E, Altman DG, Egger M, et al. The Strengthening the Reporting of Observa- tional Studies in Epidemiology (STROBE) statement: guidelines for reporting obser- vational studies. Lancet 2007;370:1453-7.
  31. Olaussen A, Blackburn T, Mitra B, et al. Review article: shock index for prediction of critical bleeding post-trauma: a systematic review. Emerg Med Australas 2014;26 (3):223-38.
  32. Ageno W, Gallus AS, Wittkowsky A, et al. Oral anticoagulant therapy: Antithrom- botic Therapy and Prevention of Thrombosis, 9th ed.: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines. Chest 2012;141:e44S-88S.
  33. Rubboli A, Becattini C, Verheugt FW. Incidence, clinical impact and risk of bleeding during Oral anticoagulation therapy. World J Cardiol 2011;3(11):351-8.
  34. Tornkvist M, Smith JG, Labaf A. Current evidence of oral Anticoagulant reversal: a systematic review. Thromb Res 2017;162:22-31.
  35. Dohan A, Sapoval M, Chousterman BG, et al. Spontaneous soft-tissue hemorrhage in anticoagulated patients: safety and efficacy of embolization. AJR Am J Roentgenol 2015;204(6):1303-10.
  36. Popov M, Sotiriadis C, Gay F, et al. Spontaneous Intramuscular hematomas of the ab- domen and pelvis: a new multilevel algorithm to direct transarterial embolization and patient management. Cardiovasc Intervent Radiol 2017;40(4):537-45.
  37. Makris M, Van Veen JJ, Tait CR, et al. Guideline on the management of bleeding in patients on antithrombotic agents. Br J Haematol 2013;160:35-46.
  38. Guzzardi G, Fossaceca R, Cerini P, et al. Endovascular treatment of spontaneous extraperitoneal haemorrhage: immediate and long-term clinical efficiency. Radiol Med 2014;119(2):121-7.
  39. Choo IW, Sproat IA, Cho KJ. Transcatheter embolization of the marginal artery of Drummond as treatment for life-threatening retroperitoneal hemorrhage compli- cating heparin therapy. Cardiovasc Intervent Radiol 1994;17(3):161-3.
  40. Maleux G, Van Sonhoven F, Hofkens P-J, et al. Oft tissue bleeding associated with an- tithrombotic treatment: technical and clinical outcomes after transcatheter emboli- zation. J Vasc Interv Radiol 2012;23:910-6.
  41. Levy JM, Gordon HW, Pitha NR, et al. Gelfoam embolization for control of bleeding from rectus sheath hematoma. AJR Am J Roentgenol 1980;135(6):1283-4.